Upload
eliavecellio
View
221
Download
0
Embed Size (px)
Citation preview
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 150
Motorcycle rider training for the prevention of road traffic
crashes (Review)
Kardamanidis K Martiniuk A Ivers RQ Stevenson MR Thistlethwaite K
This is a reprint of a Cochrane review prepared and maintained by The Cochrane Collaboration and published in The Cochrane Library 2010 Issue 10
httpwwwthecochranelibrarycom
Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 250
T A B L E O F C O N T E N T S
1HEADER 1ABSTRACT
2PLAIN LANGUAGE SUMMARY
2BACKGROUND
3OBJECTIVES
3METHODS
11RESULTS
20DISCUSSION
23AUTHORSrsquo CONCLUSIONS
24ACKNOWLEDGEMENTS
24REFERENCES
29CHARACTERISTICS OF STUDIES
46DATA AND ANALYSES
46APPENDICES
46HISTORY
46CONTRIBUTIONS OF AUTHORS
47DECLARATIONS OF INTEREST
47SOURCES OF SUPPORT
47DIFFERENCES BETWEEN PROTOCOL AND REVIEW
47INDEX TERMS
iMotorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 350
[Intervention Review]
Motorcycle rider training for the prevention of road trafficcrashes
Katina Kardamanidis1 Alexandra Martiniuk 1 Rebecca Q Ivers1 Mark R Stevenson1 Katrina Thistlethwaite1
1The George Institute for Global Health Sydney Australia
Contact address Rebecca Q Ivers The George Institute for Global Health PO Box M201 Missenden Road Sydney NSW 2050
Australia riversgeorgeorgau
Editorial group Cochrane Injuries GroupPublication status and date New published in Issue 10 2010
Review content assessed as up-to-date 8 August 2010
Citation Kardamanidis K Martiniuk A Ivers RQ Stevenson MR Thistlethwaite K Motorcycle rider training for the
prevention of road traffic crashes Cochrane Database of Systematic Reviews 2010 Issue 10 Art No CD005240 DOI
10100214651858CD005240pub2
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
A B S T R A C T
Background
Riding a motorcycle (a two-wheeled vehicle that is powered by a motor and has no pedals) is associated with a high risk of fatal crashes
particularly in new riders Motorcycle rider training has therefore been suggested as an important means of reducing the number of
crashes and the severity of injuries
Objectives
To quantify the effectiveness of pre- and post-licence motorcycle rider training on the reduction of traffic offences traffic crash
involvement injuries and deaths of motorcycle riders
Search strategy
We searched the Cochrane Injuries Group Specialised Register CENTRAL (The Cochrane Library 2008 Issue 3) TRANSPORT
MEDLINE EMBASE CINAHL WHOLIS (World Health Organization Library Information System) PsycInfo LILACS (LatinAmerican and Caribbean Health Sciences) ISI Web of Science Social Sciences Citation Index (SSCI) ERIC ZETOC and SIGLE
Database searches covered all available dates up to October 2008 We also checked reference lists of relevant papers and contacted study
authors in an effort to identify published unpublished and ongoing trials related to motorcycle rider training
Selection criteria
We included all relevant intervention studies such as randomised and non-randomised controlled trials interrupted time-series and
observational studies such as cohort and case-control studies
Data collection and analysis
Two review authors independently analysed data about the study population study design and methods interventions and outcome
measures as well as data quality from each included study and compared the findings We resolved differences by discussion with a
third review author
1Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 450
Main results
We reviewed 23 studies three randomised trials two non-randomised trials 14 cohort studies and four case-control studies Five
examined mandatory pre-licence training 14 assessed non-mandatory training three of the case-control studies assessed lsquoanyrsquo type of
rider training and one case-control study assessed mandatory pre-licence training and non-mandatory training The types of assessed
rider training varied in duration and content
Most studies suffered from serious methodological weaknesses Most studies were non-randomised and controlled poorly for con-
founders Most studies also suffered from detection bias due to the poor use of outcome measurement tools such as the sole reliance
upon police records or self-reported data Small sample sizes and short follow-up time after training were also common
Authorsrsquo conclusions
Due to the poor quality of studies identified we were unable to draw any conclusions about the effectiveness of rider training on crash
injury or offence rates The findings suggest that mandatory pre-licence training may be an impediment to completing a motorcycle
licensing process possibly indirectly reducing crashes through a reduction in exposure It is not clear if training (or what type) reducesthe risk of crashes injuries or offences in motorcyclists and a best rider training practice can therefore not be recommended As some
type of rider training is likely to be necessary to teach motorcyclists to ride a motorcycle safely rigorous research is needed
P L A I N L A N G U A G E S U M M A R Y
Motorcycle rider training for preventing road traffic crashes
Riders of motorcycles (a two-wheeled vehicle that is powered by a motor and has no pedals - Oxford English Dictionary Online)
especially novice riders have an increased risk of being involved in fatal crashes compared to other road users Motorcycle rider training
could be an important way of reducing the number of crashes and the severity of injuries
The authors of this review examined all research studies that report an evaluation of the effectiveness of motorcycle rider coursesin reducing the number of traffic offences motorcycle rider crashes injuries and deaths This review included 23 research studies
including three randomised trials two non-randomised trials 14 cohort studies and four case-control studies The types of rider training
that were evaluated varied in content and duration The findings suggest that mandatory pre-licence training may present a barrier to
completing a motorcycle licensing process thus possibly indirectly reducing crash injury death and offence rates through a reduction
in exposure to riding a motorcycle However on the basis of the existing evidence it is not clear if (or what type of) training reduces
the risk of crashes injuries deaths or offences in motorcyclists and the selection of the best rider training practice can therefore not be
recommended
It is likely that some type of rider training is necessary to teach motorcyclists basic motorcycle handling techniques and to ride a
motorcycle safely It is therefore important that further research work be conducted to rigorously evaluate motorcycle rider training
courses particularly in low income countries where the main burden of motorcycle injuries and deaths occur
B A C K G R O U N D
Each year anestimated 12million peopleare killedin road crashes
worldwide and an additional 50 million people are injured Of
these deaths 85 occur in low-income and middle-income coun-
tries It has been predicted that between the year 2000 and 2020
without increased efforts the number of road traffic deaths in
high-income countries will decrease by approximately 30 yet
in low- and middle-income countries the number is expected to
increase by over 80 (Kopits 2003 Peden 2004) Road traffic
injuries are likely to become the third leading cause of a global
burden of disease and injury by 2020 if further action is not taken
(Peden 2004)
Riding a motorcycle (a two-wheeled vehicle that is powered by a
motor and has no pedals - Oxford English Dictionary Online) is
associated with a higher risk of fatalities compared with driving a
car (Leung 1983 Bjornstig 1985 Radin Umar 1995 Tsai 1995
FORS 1999 Yuan 2000 Johnston 2008) In Australia motorcy-
2Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 550
clesaccounted for45 of allAustralian passenger vehicle registra-
tions and 09 of vehicle kilometres travelled in 2007 Howevermotorcycle riders accounted for approximately 15 of all road
crash deaths and an even higher proportion of serious injuries
Per distance travelled the Australian rate of motorcyclist deaths
was approximately 30 times the rate for car occupants The cor-
responding rate for a serious injury was approximately 41 times
higher Similar elevated rates were also found in other developed
countries (Johnston 2008) In low and middle-income countries
the motorcycle is one of the most frequently used means of trans-
port and consequently motorcyclists constitute a large proportion
of road crash victims (Mohan 2002)
The lack of experience of many motorcycle riders has been cited
as one potential cause of motorcycle crashes (Winn 1987 Mullin1997) Training of motorcycle riders has therefore been suggested
as an important countermeasure for reducing both the number of
crashes and the severity of injury (Vis 1995 Noordzij 2001)
Pre-licence motorcycle rider trainingcourses have been made com-
pulsory in several countries (Mullin 1997) Pre-licence training
courses are aimed at providing motorcycle riders with the skills
and knowledge needed to improve control of the motorcycle and
to identify and avoid critical situations in traffic (Elliot 2003)
Many pre-licence motorcycle rider training courses are aimed at
novice riders and in many jurisdictions in high-income countries
such as Australia the USA and Canada the completion of such
courses is a mandatory requirement of licensing
Post-licence rider training is not linked with licensing and such
training is generally not compulsory It usually focuses on riding
skills and other topics that have not been covered or dealt with
sufficiently in pre-licence rider training Post-licence rider train-
ing is used to reinforce fundamental skills and knowledge or to
provide some training for motorcyclists who are self-taught and
have never received formal training (Anderson 1980)
Estimates of the number of motorcyclists attending post-licence
rider training are not readily available but existing data suggest
such courses are not well attended although attendance varies by
country In the USA 24 of the total number of motorcyclists
trained in 1988 and 1989 attended an experienced rider course
(Weaver 1990)
The aim of this systematic review is to quantify the effectiveness
of pre-and post-licence rider training for motorcycle riders in re-
ducing traffic offences road traffic crash involvement injury and
death
O B J E C T I V E S
To quantify the effectiveness of pre- and post-licence motorcycle
rider training on the reduction of traffic offences traffic crash
involvement injuries and deaths of motorcycle riders
M E T H O D S
Criteria for considering studies for this review
Types of studies
All relevant intervention studies such as randomised and non-ran-
domised controlled trials and interrupted time series studies and
observational studies including cohort and case-control studies
Types of participants
Motorcyclists and riders of mopeds or motor scooters of all ages
Types of interventions
Motorcycle ridertraining (pre-or post-licence mandatory or non-
mandatory) versus no training
One form of pre-licence rider training versus another form (eg
on-road off-road theory or a combination)
One form of post-licence rider training versus another form (eg
on-road off-road theory or a combination)
Types of outcome measures
Offences crashes injuries and death
Search methods for identification of studies
We did not restrict the searches by date language or publication
status
Electronic searches
We searched the following electronic databases
bull Cochrane Injuries Group Specialised Register (searched 22
Sept 2008)
bull CENTRAL (The Cochrane Library 2008 Issue 3)bull TRANSPORT 1988 to 200706 (latest date available)
bull ATRI Australian Transport Index (to Sept 2008)
bull MEDLINE (Ovid SP) 1950 to Sept 2008
bull EMBASE (Ovid SP) 1980 to Sept 2008
bull CINAHL (EBSCO) 1982 to Sept 2008
bull WHOLIS (World Health Organization Library
Information System) (to Sept 2008)
bull PsycINFO (Ovid SP) 1806 to Sept 2008
bull LILACS (BIREME) (Latin American and Caribbean
Health Sciences) (1982 to Sept 2008)
bull ISI Web of Science Social Sciences Citation Index (SSCI)
1970 to Sept 2008
3Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 650
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 750
Table 1 Risk of selection bias (Continued)
Osga 1980 Differences at baseline reported on sex and age
Confounding factors were not dealt with in univariate analysis Riding exposure was accounted for in the
cumulative survival analysis
Selection intervention group were all training participants on the register and the control group were
respondents to a diverse range of recruitment strategies
Satten 1980 Differences at baseline reported on age sex helmet use riding experience and exposure and previous
training
Confounding factors were not dealt with in design or analysis
Selection intervention group was selected from a list of training graduates and the control group was
partly selected from a state register of motorcyclists
Jonah 1982 Differences at baseline reported on age sex riding experience time licensed previous training and alcohol
use
Confounding factors were dealt with through multivariate analysis
Selection intervention group was selected from a list of training graduates and the control group was
randomly selected from the Ontario motorcycle operator file
Mortimer 1984 Differences reported at baseline on age licence status riding experience years licensed and helmet use
Confounding factors age years licensed and rider experience were dealt with through stratification and
riding exposure through standardisation
Selection intervention group was selected from a list of training enrolees and the control group was
selected from motorcycle dealerships during three successive Saturdays
Cooper 1988 Differences reported at baseline on age gender licensure and highway experience
Confounding factors age gender riding exposure rider characteristics rider attitudes and behaviour were
accounted for in a multivariate analysis
Selection there was no information on how the intervention and control groups were selected and from
where
Mortimer 1988 Differences reported at baseline on years licensed riding experience and helmet use
Confounding factors age years licensed and rider experience were dealt with through stratification and
standardisation of riding exposure A regression analysis was also conducted but poorly reported and not
usable
Selection intervention group was selected from a list of training enrolees and the control group was
selected from motorcycle dealerships
Adams 1985 Differences at baseline reported on previous crash rates
Confounding factors not controlled for in analysis
Selection the intervention group was a non-random sample of those offered training within the Australia
Post motorcycle delivery riders the control group were the remaining Post riders
Leung 1987 Differences at baseline reported on age and sex
Confounding factors age and sex controlled for in analysis by stratification
Selection those exposed to training were MRC graduates those unexposed were taken from the licensing
records
5Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 850
Table 1 Risk of selection bias (Continued)
McDavid 1989 Differences at baseline other than matched characteristics were not reported
Confounding factors age ownership of car licence years licensed previous crashesand offences accounted
for through matching Other confounding effects not controlled
Selection intervention group was selected from a list of training graduates and the control group was
selected from Bureau of Motor Vehicle records
Kloeden 1994 Differences at baseline reported were in geographical area of residence and in riding exposure
Confounding factors were accounted for through exclusion of participants from rural areas
Other confounding factors were not reported
Selection the Ridersafe course was phased in based on postcode of riderrsquos home address analyses were
based on a database of those who obtained motorcycle rider learnerrsquos permits of which about 2000 were
Ridersafe riders and about 5000 were control riders
Billheimer 1998 Differences at baseline reported on age and previous riding experience
Other potential confounding factors assessed were riding exposure primary purpose of riding and helmet
use
Confounding factors accounted for by matching (age sex primary purpose of riding) standardisation
(riding exposure) and stratification (previous riding experience)
Selection the intervention group were those who had taken the rider training programme the control
group were California motorcycle riders matched on the above variables but who had not taken a riding
course
Perrino 2002 Differences at baseline reported were a higher number of females in the trained group and a higher
proportion of experienced riders in the untrained group
Confounding factors were not accounted for in the analysis
Selection the intervention group were those who had done the Motorcycle Riding Course the control
group had not done the course Both groups were identified through a register of licensed motorcyclists
Kraus 1975 Population based case-control study
Cases all motorcycle riders with a confirmed medically treated motorcycle injury (including fatalities)
from all hospitals in study base and peripheral area hospitals and all deaths due to motorcycle crashes in
the study base All serious injuries (leading to hospitalisation) and fatalities likely to be included
Community controls a random sample of registered motorcycle owners in the study base
Potential confounders assessed were age sex weight height riding experience motorcycle use charac-
teristics of motorcycle and previous crashes and offences
Confounding factors were accounted for through exclusion of females and stratification by age and
motorcycle useSelection cases and controls drawn from the same population
Hurt 1981 Population based case-control study
Cases and controls not drawn from the same population
Cases motorcycle riders who had a motorcycle crash Not all reported cases included
Community controls motorcycle riders who passed the crash site at the same time of day and day of week
that the crash occurred but 2 years later
Potential confounders assessed were age sex time type of motorcycle roadway alignment helmet use
weather conditions road surface conditions
Confounding factors were accounted for through matching for time of day and day of week that crash
occurred and stratification
Selection cases were recruited from police reports not all cases were included controls were matched
6Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 950
Table 1 Risk of selection bias (Continued)
community controls from randomly selected sites matched for same time of dayday of week as the crash
occurred
Haworth 1997 Population based case-control study
Cases and controls drawn from the same population
Cases motorcycle riders or pillion riders with an injury from a motorcycle crash from hospital and police
records Not clear whether all cases were included
Community controls motorcycle rider or pillion who passed the crash site at the same time of day and
day of week that the crash occurred
Potential confounders assessed were rider factors age marital status licence status riding experience
and exposure alcohol and drugs use of protective gear including helmet use pillion factors motorcycle
factors trip factorsConfounding factors were accounted for through matching for time of day and day of week that crash
occurred and stratification for age blood alcohol level and licence status
Selectioncases were riders or pillions involved in on-road crash recruited from hospitals and police records
not sure if all cases were included controls were matched community controls who passed the crash site
at the same time of dayday of week as the crash
Savolainen 2007 Differences at baseline reported on sex safety helmet use riding ability riding experience and licence
status
Adjusted for these confounders through statistical modelling although exact process unclear
Selection intervention group from list of trainees control group from Indiana Bureau of Motor Vehicles
database Trainees and the control group were invited to participate in surveys
Davis 1997 Differences at baseline not reported but likely No adjustment for any confounders at baseline or during analysis
Selection intervention group from list of riders involved in crashes who were also on a list of training
graduates control group from the same list of riders involved in crashes who were not also on a list of
training graduates
Waller 1992 Differences at baseline not reported but likely
Adjusted for confounders through exclusion of riders younger than 18 years riders without a motorcycle
operator driver license riders with location of residence outside of the Texas metropolitan areas of Austin
through matching by age sex and location of residence through stratification by age and sex and through
standardisation (number of crashes by person-days at risk)
Selection intervention group from the Motorcycle Operator Training class roster (number of trained
riders that the sample was selected from is unclear) control group through a random number generator
generating driver license numbers until an equal sample size was selected as the intervention group
MAIDS 2004 Population based case-control study
Unclear whether cases and controls not drawn from the same population
Cases motorcycle riders who had a motorcycle crash Unclear whether all cases included
Community controls motorcycle riders recruited at petrol stations Unclear whether and how matched
to cases
No potential confounders were reported
There was no report of whether and how potentially confounding factors were accounted for in the study
7Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 1050
Table 2 Risk of allocation bias
Anderson 1980 Allocation concealment unclear
Hall 1988 Allocation concealment unclear
Swaddiwudhipong 1998 Allocation concealment unclear
Table 3 Risk of detection bias
Anderson 1980 Blinding of outcome assessors not reported
Driver records holding police recorded crashes and offences not validated
Hall 1988 Blinding of outcome assessors not reported
Driver records holding police recorded crashes and offences not validated
Swaddiwudhipong 1998 Blinding of outcome assessors not reported
Police and hospital records Post mortem examination for deaths not validated
Adams 1985 Blinding of outcome assessors not reported
Australia Post crash records not validated but argued that underreporting is unlikely
Leung 1987 Blinding of outcome assessors not reported
Driver records holding police recorded crashes and offences
Raymond 1979 Blinding of outcome assessors not reported
Self-report through personal interviews or questionnaires - both methods validated
Osga 1980 Blinding of outcome assessors not reported
Self-report through structured questionnaires validated by cross-checking with police
recorded crash records
Satten 1980 Blinding of outcome assessors not reported
Self-report through structured interviews or questionnaires - both methods were not
validated
Intervention group were given telephone interviews but the control group mailed ques-
tionnaires
Jonah 1982 Blinding of outcome assessors not reported
Self-report through structured interviews not validated Driver records holding police
recorded crashes and offences
Mortimer 1984 Blinding of outcome assessors not reported
Self-report through questionnaires not validated
Intervention group received mailed questionnaires but the control group received ques-
tionnaires personally at the motorcycle dealership
Cooper 1988 Blinding of outcome assessors not reported
Driver records holding police recorded crashes and offences
8Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 1150
Table 3 Risk of detection bias (Continued)
Mortimer 1988 Blinding of outcome assessors not reported
Self-report through questionnaires not validated
Intervention group received mailed questionnaires but the control group received ques-
tionnaires personally at the motorcycle dealership
McDavid 1989 Blinding of outcome assessors not reported
Driver records holding police recorded crashes and offences
Kloeden 1994 Blinding of outcome assessors not reported
Driver records holding police recorded crashes and offences
Billheimer 1998 Blinding of outcome assessors not reported
Police recorded crashes and offences
Perrino 2002 Blinding of outcome assessors not reported
Self-report through mailed out questionnaires police records not validated
Kraus 1975 Blinding of intervention assessors to case or control status not reported
Intervention measured police records self-report though questionnaires not validated
Length of follow up not specified
Hurt 1981 Blinding of intervention assessors to case or control status not reported
Intervention measured face-to-face in-depth interviews (ie no structured question-
naires) Validity and reliability of data ensured by reported high level of quality controlLength of follow up not specified
Haworth 1997 Blinding of intervention assessors to case or control status not reported
Intervention measured self-report either by personal or by telephone interview for
both cases and controls not validated
Length of follow up in the 2 years prior to the interview or more than 2 years ago
Savolainen 2007 Blinding of outcome assessors not reported
Source of outcome data not reported
Length of follow up up to 5 years
Davis 1997 Blinding of outcome assessors not reported
Source of outcome data police records
Length of follow up 1 year
Waller 1992 Blinding of outcome assessors not reported
Source of outcome data Department of Public Safety driver history file and official
accident reports
Length of follow up 3 years
MAIDS 2004 Blinding of intervention assessors to case or control status not reported
Intervention measured face-to-face in-depth interviews (ie no structured question-
naires) No report on whether and how validity and reliability of data was ensured
Length of follow up 2 years
9Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 1250
Table 4 Risk of attrition bias
Anderson 1980 Percentage of participants in analysis 99 forthe total group Unclear forintervention andcontrol groups
separately
Hall 1988 Percentage of participants in analysis 96-97 forall intervention groups and 98for the control group
Swaddiwudhipong 1998 Percentage of participating sub-districts in analysis 100 for both intervention and control sub-districts
Raymond 1979 Percentage of participants in analysis 74 for the intervention group and less than 57 for the control
group
Osga 1980 Percentage of participants in analysis 73 for the intervention group and 100 for the control group
Satten 1980 Percentage of participants in analysis is unknown Initial sample size is unclear
Jonah 1982 Percentage of participants in analysis 23 for the intervention group and 45 for the control group
Mortimer 1984 Percentage of participants in analysis 35 for the intervention group and 84 for the control group
Cooper 1988 Percentage of participants in analysis 98 for the total group Unclear for intervention and control
groups separately
Mortimer 1988 Percentage of participants in analysis 40 for the intervention group and 90 in the control group
Adams 1985 Percentage of participants in analysis 61 forthe total group Unclear forintervention andcontrol groups
separately
Leung 1987 Percentage of participants in analysis unclear because initial sample size was not reported
McDavid 1989 Percentage of participants in analysis 40 of final matched cohort
Kloeden 1994 Percentage of participants in analysis 39 for the intervention group and 6 for the control group
Billheimer 1998 Percentage of participants in analysis unclear percentage of the initial sample size
Perrino 2002 Percentage of participants in analysis unclear percentage of the initial sample size
Kraus 1975 Percentage of participants in analysis 58 of cases 59 of controls
Hurt 1981 Percentage of participants in analysis 18 of cases lt 29 of controls
Haworth 1997 Percentage of participants in analysis 70 of cases 47 of controls
Savolainen 2007 Percentage of participants in analysis 18 for the intervention group and 15 for the control group
Davis 1997 Percentage of participants in analysis 100 for the intervention group and unclear for the control group
10Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 1350
Table 4 Risk of attrition bias (Continued)
Waller 1992 Percentage of participants in analysis unclear percentage for the intervention group and 100 for the
control group
MAIDS 2004 Percentrage of participants in analysis unclear for both cases and controls
R E S U L T S
Description of studies
See Characteristicsof included studies Characteristics of excluded
studies Characteristics of ongoing studies
Our search resulted in 102 studies that seemed relevant of which
we excluded 79 studies because they did not meet the criteria for
considering studies for this review (see Characteristics of excluded
studies)
Basedon ourcriteriawe includeda total of 23 studies in this review
There were three randomised trials (Anderson 1980 Hall 1988
Swaddiwudhipong 1998) two non-randomised trials (Adams
1985 Leung 1987) 14 cohort studies (Raymond 1979 Osga
1980 Satten 1980 Jonah 1982 Mortimer 1984 Cooper 1988
Mortimer 1988 McDavid 1989 Waller 1992 Kloeden 1994
Davis 1997 Billheimer 1998 Perrino 2002 Savolainen 2007)
and four case-control studies (Kraus 1975 Hurt 1981 Haworth
1997 MAIDS 2004)
Over half of the studies were conducted 20 or more years
ago and the oldest was conducted 31 years ago Only three
studies were from this decade (Perrino 2002 MAIDS 2004
Savolainen 2007) Over 60 of the studies were conducted in
the USA (Kraus 1975 Anderson 1980 Osga 1980 Satten 1980
Hurt 1981 Mortimer 1984 Leung 1987 Hall 1988 Mortimer
1988 Waller 1992 Davis 1997 Billheimer 1998 Perrino 2002
Savolainen2007) three inAustralia (Adams 1985 Haworth 1997
Kloeden 1994) three in Canada (Jonah 1982 Cooper 1988
McDavid 1989) onefromthe United Kingdom (Raymond 1979)one from Europe (with sampling areas located in France Ger-
many Netherlands Spain and Italy) (MAIDS 2004) and one
in Thailand (Swaddiwudhipong 1998) Only seven studies were
published in peer reviewed journals (Kraus 1975 Jonah 1982
Mortimer 1984 Cooper 1988 Mortimer 1988 McDavid 1989
Swaddiwudhipong 1998) The other studies were reports forhigh-
way safety and road traffic organisations
The studies by Anderson 1980 Hall 1988 Kloeden 1994
Billheimer 1998 (the basic California Motorcyclist Safety Program
component of the study) and Perrino 2002 examined manda-
tory pre-licence training Cooper 1988 McDavid 1989 and
Swaddiwudhipong 1998 assessed non-mandatory training that
when passed resulted in a licence Raymond 1979 Osga 1980
Satten 1980 Jonah 1982 Mortimer 1984 Adams 1985 Leung
1987 Mortimer 1988 Waller 1992 Davis 1997 Billheimer 1998
(the Experienced Rider Course component of the study) andSavolainen 2007 assessed non-mandatory training that could be
taken by motorcycle riders with and without a licence The case-
control studies (Kraus 1975 Hurt 1981 Haworth 1997) assessed
lsquoanyrsquo type of rider training and therefore could not be grouped by
pre- or post-licence rider training The MAIDS 2004 case-control
study did make a distinction between participants having received
compulsory pre-licence training or additional (non-compulsory)
training We did not find a study that compared one form of pre-
or post-licence training (eg on-road off-road theory or a com-
bination) versus another form of training All studies compared
eithermandatory or non-mandatory formal training versus no for-
mal training with the exception of Haworth 1997 who compared
different levels of training courses (beginner intermediate and ad-vanced)
We have presented a more detailed description of the individual
studies in Characteristics of included studies We have provided
a description below of the types of motorcycle rider training that
were evaluated in the included studies
Types of motorcycle rider training evaluated in the
studies
The types of training were varied Duration of training ranged
from two or three hours to 24 hours administered over two or
three days Most training consisted of a theoretical and a practical
component usually off-road The content ranged from provid-ing riders with manuals lectures discussions showing films and
administering knowledge and skill tests Little information was
provided about the qualifications of trainers who conducted the
training
Six studies (Osga 1980 Satten 1980 Mortimer 1984 Mortimer
1988 Waller 1992 Savolainen 2007) assessed the effectiveness of
the USA-based Motorcycle Safety Foundationrsquos Motorcycle Rider
Course (MSF MRC) This course was released in 1976 It com-
prises eight hours of theory (lectures discussions films and multi-
plechoice tests) a textbook and 12 hours of off-road practice The
Motorcycle Safety Foundation is a national not-for-profit organ-
11Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 1450
isation sponsored by manufacturers and distributors of motorcy-
cles Many USA states have adopted the MSF Basic Rider Courseand Experienced Rider Course (wwwmsf-usaorg) since its release
in 1976
The Motorcycle Operators Skill Test was also developed
by the USA Motorcycle Safety Foundation in coopera-
tion with others (httpwwwwebbikeworldcommotorcycle-
trainingmotorcycle-test) The MOST is a licensing test consist-
ing of a manual knowledge and skill test (comprising eight exer-
cises of increasing difficulty and skill requirements) and a mini-
skill test conducted with a simulator Some studies compared this
alternative licensing test to the standard licensing test However
as the MOST is not a type of rider training we have included only
the two studies that assessed the MOST in combination with one
or more training components in this review Both Anderson 1980and Hall 1988 assessed the MOST in combination with a train-
ing component of three hoursrsquo duration consisting of instructions
about the skills needed to pass the MOST (30 minutes of the-
ory and two and a half hours on the motorcycle) Hall 1988 also
assessed the MOST in combination with the Motorcycle Safety
Foundationrsquos Motorcycle Rider Course as described above
Other USA studies evaluated the Wisconsin motorcycle rider
course a 22-hour off-street programme that teachesthe basic rules
of safe motorcycling (Leung 1987) the USA Maryland Motorcy-
cle Rider Course (MRC) a mandatory (for those younger than 18
years)pre-licence 20-hour coursecomprising eight hours of theory
and12 hours of practice and created by the Maryland Motor Vehi-
cle Administration in 1985 (Perrino 2002) the California Motor-cyclist Safety Program (CMSP) a mandatory (for those younger
than 18 years) pre-licence 16-hour basic course (Billheimer 1998)
and Experienced Rider Course (ERC) of eight hoursrsquo duration
(Billheimer 1998) The CMSP still offers these courses ( http
wwwca-msporg) but they are now MSF courses as described
above Davis 1997 did not give any information about the assessed
rider training
The Canadian studies evaluated the British Colombia Safety
Councilrsquos 37-hour motorcycle safety programme (no further in-
formation on the content was provided) (McDavid 1989) and
the Canada Safety Council Motorcycle Training Program (MTP)
comprising four hours of theory and 16 hours of practice (six
hours on-road) over twoweekends (Jonah 1982) The MTPcoursewas made available in 1974 and is still being offered in Canada
(httpwwwsafety-councilorgtrainingmtpmtphtm) Cooper
1988 did not give any information about the assessed rider train-
ing However the study was conducted in British Colombia in
Canada in 1988 the same time as the McDavid 1989 study was
conducted
In the United Kingdom Raymond 1979 assessed the non-manda-
tory RACACU (Royal Automobile ClubAuto-Cycle Union)
training scheme The training programme consisted of either 24
one-hour sessions or 13 two-hour sessions Twelve hours of the
training focused on theory about basic operational procedures
traffic laws and highway codes The practical session focused on
basic operational skills riding on a training area and riding in traf-fic with an instructor A proficiency test was administered at the
end of the course
One study from rural Thailand (Swaddiwudhipong 1998) evalu-
ated a community-based rider education programme consisting of
theory on motorcycle injury risk helmet use traffic laws vehicle
regulations and traffic signs and a written and skill test leading to
a licence
TheAustralian studies evaluateda rider trainingcourse of 12 hours
over two consecutive days (four hours of theory and eight hours of
practice)administeredto AustraliaPosttelegram deliverers(Adams
1985) and a mandatory pre-licence Ridersafe course consisting
of two tiers the first tier being a training course on the basics
of motorcycle riding where a pass entitled the rider to obtain alearnerrsquos permit and the second tier being a skill test where a pass
entitled the riderto obtain a probationary motorcycle licence The
training course consisted of two four-hour training sessions on
separate days involving instruction videos and actual riding on
motorcycles The skill test consisted of a four-hour preparation
course of 74 minutes of lecturediscussion 38 minutes of video
and 75 minutes of on-course riding which was the testing phase
(Kloeden 1994) The Rider Safe course is still being provided by
Transport South Australia a state government organisation (http
wwwtransportsagovau )
The duration or content of rider training in the four case-control
studies (Kraus 1975 Hurt 1981 Haworth 1997 MAIDS 2004)
was not described
Risk of bias in included studies
We describe the methodological quality of the included studies
by study type below More detailed information can be found in
Table 1 Table 2 Table 3 and Table 4
When considering the findings from these studies it is important
to note the quality (or lack thereof) of reporting of the studies
Quality of reporting is the extent to which a report provides in-
formation about the design conduct and analysis of a study ( Juumlni2001) Huwiler-Muumlntener 2002 found that even though trials
may be well reported it does not mean that the methodological
quality is of equal standard and well-conducted trials may never-
theless be reported badly The reporting quality of most studies in
this review was poor Most studies were old and did not meet con-
temporary reporting standards This made data extraction com-
plicated It also made it difficult to contact authors for additional
information
Because of significant heterogeneity between studies as well as the
poor quality of most studiespooling of studyresults to calculatean
overall measure of effect was not appropriate Instead we provided
a description of our judgement of risk of bias in the text
12Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 1550
Risk of bias in the randomised trials
There were three randomised trials (Anderson 1980 Hall 1988Swaddiwudhipong 1998) Anderson 1980 and Hall 1988 had an
initial sample size of 41188 and 26464 motorcycle licence ap-
plicants respectively Both studies assessed the effectiveness of the
pre-licence MOST with a trainingcomponent Swaddiwudhipong
1998 evaluated the effectiveness of a community-based rider ed-
ucation programme in six districts
Allocation bias
Anderson 1980 reported using random birth date tables to allocate
participants to either the intervention or control group Hall 1988
reported the use of a computerised algorithm set in a central of-
fice but also used manual random assignment when the computer
went down Swaddiwudhipong 1998 reported lsquosimple random
samplingrsquo for the delivery of the intervention where the sampling
frame included nine districts Three districts were excluded since
they were too hard to access The remaining six districts were ran-
domly allocated to the intervention (three) and the control (three)
groups Systematic sampling wasthen used ordered by village size
for the villages within districts Allocation concealment was not
reported in any of the studies The descriptions of randomisation
in these studies suggest that there may have been limitations to
the process resulting in a lack of equivalence of the intervention
and control groups at baseline and therefore the possibility of
allocation bias Swaddiwudhipong 1998 reported that before theintervention the annual incidence rate of motorcycle-related in-
juries was slightly (but not statistically significant) higher in the
intervention areas than in the control areas Participants in the
intervention districts were also more likely to have a valid licence
and to wear a helmet more often than participants in the control
districts Anderson 1980 measured differences at baseline on the
participant and prior driving record variables and found these to
be negligible Hall 1988 did not report on baseline differences
between the intervention and control groups As the Anderson
1980 and Hall 1988 studies had sample sizes of over 40000 and
26000 respectively and used some form of randomisation it is
likely that any confounding factors at baseline would have been
equally distributed between comparison and intervention groups
Selection bias
After randomisation Anderson 1980 and Hall 1988 excluded
08 and 3 participants respectively because of assignment to
the wrong group administration of the wrong intervention or
other reasons The resulting lack of analysis by rsquointention-to-treatrsquo
can introduce biasby removing the effectivenessof randomisation
Nonetheless the percentage excluded was small and the number
of overall participants large so it is unlikely to have had a major
effect on the final results In the Swaddiwudhipong 1998 study
villages were systematically sampled into intervention or control
groups The intervention was implemented and outcomes of in-
jury and death were measured at a community level Interventionand control districts were randomly selected however there were
only three intervention and three control communities
Detection bias
None of the randomised trials reported blinding of outcome as-
sessors Data collection tools used by Anderson 1980 and Hall
1988 were driver records holding police recorded crashes and of-
fences Swaddiwudhipong 1998 used police and hospital records
at a district level and post-mortem examination for deaths None
reported on validity or reliability for any of the instruments
Attrition bias
Anderson 1980 and Hall 1988 had driver records available for
96 to 100 of participants They also collected self-reported
data but the response rate to the questionnaires was very low (low-
est response rate 42 and 30 respectively) Attrition bias in
Swaddiwudhipong 1998 was not an issue as outcome data were
available for all sub-districts included in the trial
The intervention reported in Anderson 1980 and Hall 1988 tri-
als had an unanticipated effect on the participants In both trials
an average of only 28 of licence applicants in the intervention
group eventually obtained their licence compared to 35 in the
control group This indicates that the licensing process in generalseems to have deterred applicants from obtaining a licence and
although no statistical tests of the difference were conducted this
deterrence appears even greater for those who had rider training
This is not true attrition bias as the studies did not experience
differential loss to follow up from outcome measurement Instead
differential attainment of the intermediary step of licensure pre-
cluded participants from being at risk for the outcomes of interest
they are not driving a motorcycle hence they are not at risk for
motorcycle-related crashes injury death or offences In short it is
not possible to assess whether a reduction in crashes in the inter-
vention group would be due to changes in riding habits as a result
of training or simply due to delayed or never attained licensure
Intervention integrity
Neither Anderson 1980 nor Hall 1988 reported on the consis-
tency in the way the intervention was administered the possible
lsquocontaminationrsquo of participants in the intervention group by other
types of formal or informal rider training or of participants in the
control group by the intervention Swaddiwudhipong 1998 did
not report on the number of villagers in each group that actually
attended the intervention However they did report on a number
of other potential training courses that were being implemented
at the same time
13Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 1650
Analysis
Anderson 1980 and Hall 1988 acknowledging the lsquodeterrent ef-fectrsquo described above conducted two separate analyses The first
analysis included the total study population of licence applicants
over two years from the time of their application (the population
denominator) The second analysis was performed on the licensed
(intervention and control) subgroups only two years from the
time of obtaining their licence (the licensed rider denominator)
As only a third were licensed and exposed to riding the motorcy-
cle it was impossible to determine whether reported differences
in crash rates across groups were due to the intervention or to
other lsquoconfoundingrsquo characteristics of those who were determined
in their efforts to obtain the licence When analysis was restricted
only to those who gained their licence the same difficulties re-
mained The benefits of the randomisation process were thus lost
due to this lack of licensure of riders during the process As there
were only three intervention and three control groups in the study
reported by Swaddiwudhipong 1998 analyses were hindered by
smallnumbers for outcome comparisons betweenthe intervention
and the control districts
Risk of bias in the cohort studies and non-randomised
trials
We discuss the methodological quality of cohort studies (seven
prospective studies (Raymond 1979 McDavid 1989 Waller 1992
Kloeden 1994 Billheimer 1998 Perrino 2002 Savolainen 2007)and seven retrospective studies (Osga 1980 Satten 1980 Jonah
1982 Mortimer 1984 Cooper 1988 Mortimer 1988 Davis
1997)) and non-randomised trials (Adams 1985 Leung 1987) in
the one section as the biases for the two study types are similar
The quality of these studies was in general poor
Selection bias
The confounding factors expected to have been addressed in the
studies were sex age riding experience prior driving records (of-
fences crashes) previous training and the following proxies for
risk taking behaviour helmet use speed and alcoholdrug use All
of these factors have been shown to be associated with an increasedrisk of motorcycle crash (Peden 2004 Zambon 2006) The pos-
sible source of bias in cohort studies and non-randomised trials
is that groups may differ on important characteristics at baseline
participants lsquoself-selectrsquo into either the rider training group by de-
ciding to undertake a rider training course or into the lsquocontrol
grouprsquo by deciding not to undertake a course Participants who
undertake a non-mandatory rider training course may differ from
those who choose not to on characteristics such as age sex rid-
ing experience and riding behaviour as well as other possible con-
founders For example riders who take a training course may be
more safety-conscious or on the contrary may do so because they
have previously experienced a crash or because they perceive that
they have poor riding skills Therefore ideally all participants in
the intervention group need to be included in the study as wellas a random sample of controls taken from the same population
as the intervention group If that is not possible intervention and
control groups either need to be carefully matched to make them
comparable or differences at baseline need to be adjusted for in
the analysis
Most of the studies using a non-randomised trial or cohort design
compared an intervention group to a control group that differed
on a number of important factors The studies in general failed to
adjust for important differences found at baseline and their results
cannot be meaningfully interpreted Ten studies (Raymond 1979
Osga 1980 Satten 1980 Jonah 1982 Mortimer 1984 Mortimer
1988 McDavid 1989 Waller 1992 Davis 1997 Savolainen 2007)
assessed intervention and control groups that were not drawnfrom the same population In most of these cases the intervention
group was selected from a list of training graduates who were usu-
ally young female and less experienced riders whereas controls
were recruited from a general population of motorcycle riders who
were usually older male and experienced riders For three stud-
ies it was unclear how they were selected (Leung 1987 Cooper
1988 Perrino 2002) Only Kloeden 1994 recruited groups from
the same population of motorcycle applicants and Adams 1985
from a group of professional riders Nevertheless due to non-ran-
dom selection of intervention and control groups differences at
baseline are likely even though study groups were from the same
population Three studies adjusted for some confounding fac-
tors found at baseline through matching (McDavid 1989 Waller1992 Billheimer 1998) However matching has limitations in
termsof controlling for multiple confounders especially those that
are unknown to influence the effect of rider training on the out-
comes One study matched the control and intervention groups
by excluding controls (approximately 3500 (66 of initial con-
trol group) that differed from the intervention group) (Kloeden
1994) Five studies (Osga 1980 Satten 1980 Adams 1985 Davis
1997 Perrino 2002) did not adjust for differences between study
groups at baseline either in the study design or analysis Results
from these studies are therefore not reliable The other six stud-
ies (Raymond 1979 Jonah 1982 Mortimer 1984 Cooper 1988
Mortimer 1988 Leung 1987 McDavid 1989 Billheimer 1998)
adjusted for some differences between groups at baseline by eitherbi-variate or multi-variate analyses or stratification and standardi-
sation of riding exposure (outcomemiles driven) or stratification
by age group The limitation of stratification is that it is difficult to
control for many variables simultaneously because a large number
of strata will need to be generated resulting in small numbers that
cannot be meaningfully analysed Savolainen 2007 developed sta-
tistical models to assess the probability of a crash after taking the
rider training course in an intervention and control group while
examining differences by sub-groups (eg age) Participation rates
were also very low in this study with only 5584000 (1395)
in the eligible intervention group and 5884000 (147) for the
14Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 1750
control group completing surveys
Detection bias
None of the studies reported blinding of outcome assessors to the
intervention or control status of the participants Seven studies
used driver records containing police recorded data as their main
measure of outcome (Leung 1987 Cooper 1988 McDavid 1989
Waller 1992 Kloeden 1994 Davis 1997 Billheimer 1998) An-
other four studies (Raymond 1979 Satten 1980 Mortimer 1984
Mortimer 1988) only used self-report Raymond 1979 was the
only study using self-report as the main outcome measurement in
which study instruments were validated Self-report might prevent
the underreporting that has been described for the use of police-
driver records (see rsquoRisk of biasrsquo in the randomised trials above)but has other risks of bias especially if questionnaires or inter-
view methods have not been validated and tested for reliability
Recall bias arises from mistakes in recollecting events both be-
cause of failures of memory and because of looking at things lsquowith
hindsightrsquo and possibly changed views (httpwwwcochraneorg
resourcesglossaryhtm) As many studies were retrospective risk
of recall bias is likely and may have increased with length of follow
up For studies that used self-reported outcomes only it is also
possible that social desirability response bias played a role where
participants in the training group may have felt it was lsquoexpectedrsquo
to report good driving histories Multiple methods to measure the
same outcome for both groups decreases the risk of detection bias
for example because self-reported answers can be cross checkedwith other sources of information such as driver records
In this regard three studies (Osga 1980 Jonah 1982 Perrino
2002) used both self-report and driver records Osga 1980 found
that for both intervention and control groups more crashes were
reported through self-report than on state crash records an aver-
age of 54 of self-reported crashes could be found back on state
crash records One industry-based study (Adams 1985) used oc-
cupational health records which were not validated but which are
unlikely to suffer from underreporting due to work cover claims
Savolainen 2007 used information on rider crashes for a five-year
period but did not specify from where this information was ob-
tained
Attrition bias
The majority of studies (Raymond 1979 Satten 1980 Adams
1985 Leung 1987 Kloeden 1994 Davis 1997 Billheimer 1998
Perrino 2002) inadequately reported on losses of participants to
the study We can therefore not be certain that there were no sys-
tematic differences in the characteristics between the intervention
and control group participants Savolainen 2007 reported the re-
sponse rates to their baseline survey but did not report on the per-
centage of outcome data that was available for their study sample
A similar deterrent effect of rider training as described for the ran-
domised trials (Anderson 1980 Hall 1988) occurred in two stud-
ies Jonah 1982 reported that 65(14992310) of the trained rid-
ers stopped riding compared to 47 (9492029) of the untrainedriders They excluded non-active and non-licensed riders before
the study commenced so the differential choice to continue riding
after training did not affect their analyses Leung 1987 reported
that the percentage of participants in the intervention group ob-
taining their licence was 50 (no number provided by authors)
lower than in the control group This may have biased the mea-
surement of outcomes because unlicensed participants will not be
exposed to riding the motorcycle and therefore not be at risk of
being involved in a crash or violation As explained above this
makes it difficult to assess whether the reduction in crashes in the
intervention group is due to a change in driving habits or simply
due to delayed licensure
Mortimer 1984 and Mortimer 1988 describe that about 44 and30 of course graduates respectively were excluded from analysis
because they had not ridden a motorcycle in the previous year
However because these studies were retrospective and the controls
were activeriders at the time of recruitment a decision on whether
the deterrent effect was differential between the two groups could
not be made Though not reported quantitatively Kloeden 1994
also describes the effectof rider training deterringindividuals from
becoming licensed and hence exposure to riding and risk of crash
death or offences involving a motorcycle Osga 1980 reported
that 27 of the trained riders did not ride after the course and
excluded these from the analysis
Use of different data collection strategies may have contributed to
differential loss to follow up in the cohort studies Satten 1980for example interviewed the intervention group over the tele-
phone but mailed questionnaires to the control group Mortimer
1984 and Mortimer 1988 distributed questionnaires in person to
the control group but mailed them to the intervention group
This may have played a role in the observed differential response
rates where more subjects responded in the personally approached
groups
Survival bias
For those studies that assessed offences crashes and injuries but
not death survival bias may have also played a role Motorcyclists
who were fatally or seriously injured in crashes during the follow-
up period after the intervention would have been less likely or
unable to respond to surveys This could bias the effect measures
towards or away from the null None of the studies wrote about
this issue or examined the characteristics of those lost to follow up
in order to address the possibility of survival bias
Intervention integrity
Although the interventions were intended for the entire lsquoexposedrsquo
groups most studies reported difficulty in tracking participants
whoreceivedthe interventionand those whodid not Adams 1985
15Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 1850
estimatedthat 62of thetotalstudysample received the interven-
tion The remaining studies did not report the percentage of thestudy population that received the intervention as was intended
or reported that this percentage was unknown (Raymond 1979
Kloeden 1994 Satten 1980 Jonah 1982 Mortimer 1984 Leung
1987 Cooper 1988 Mortimer 1988 Billheimer 1998 Perrino
2002 Savolainen 2007) Only Raymond 1979 reported the degree
to which their intervention was delivered as intended No studies
reported on possible lsquocontaminationrsquo of participants by other inter-
ventions than the one to which they were allocated Because Osga
1980 Satten 1980 Jonah 1982 Mortimer 1984 Cooper 1988
Mortimer 1988 Waller 1992 and Davis 1997 were retrospective
studies and participants were recruited from a list of graduates it
was known that they had completed and passed the course
Analysis
In some studies some participants who reported at baseline that
they were untrained decided to enroll in a rider training pro-
gramme after the baseline survey The non-randomised trial by
Adams 1985 experienced this move of subjects from the untrained
to the trained group but failed to conduct analyses by lsquointention-
to-treatrsquo This can bias the results in randomised and non-ran-
domised trials if participants who changed groups differ signifi-
cantly with regard to potentially confounding characteristics On
the other hand Perrino 2002 also experienced this move of sub-
jects from theuntrained to the trained groupmore than 100riders(5 of the total sample) originally stated that they were untrained
at baseline but then received training between the baseline and
follow-up surveys Yet since the Perrino 2002 study was a cohort
their analyses then proceeded by including those riders who were
lsquonewlyrsquo trained as exposed Riders who were lsquonewlyrsquo trained since
baseline but with a programme other than the one under study
were excluded from the analysis Nevertheless confounding fac-
tors were not accounted for in the analyses
McDavid 1989 matched exposed with unexposed subjects but
could not find suitable matches for the majority (60) of their
subjects They could therefore only conduct analysis on the result-
ing40 (139346) which is an inappropriate andinefficient anal-
ysis Some studies that reported on the issue of non-active licensedriders(Satten 1980 Jonah 1982 Mortimer 1984 Mortimer 1988
Kloeden 1994) excluded non-active riders from analysis intro-
ducing selection bias as it is likely that riders who continued to
ride differed with regards to potentially confounding character-
istics across intervention and control groups Kloeden 1994 for
example conducted several analyses to assess the effect of various
sub-sets of the control group Original analyses included all con-
trols and led to erroneous results since only a small proportion of
controls hadgone on to get a licence after their learnerrsquos permitand
were actually exposed to the possibility of experiencing a crash
Subsequent analyses compared only intervention and control sub-
jects who had a licence Waller 1992 reported crashes by person-
days of exposure but did not explain how they could be sure that
each study intervention and control participant was an active riderand actually exposed to the risk of experiencing a crash
Potential confounders that are unequally distributed by study arm
at baseline need to be dealt with in the analysis Satten 1980
Adams 1985 and Davis 1997 did not take into account any con-
founding factors either by design or in the analysis McDavid
1989 Waller 1992 and Billheimer 1998 accounted for potential
confounding factors at the design stage by matching the subjects
in the intervention and control groups Other studies (Raymond
1979 Mortimer 1984 Mortimer 1988 Perrino 2002 and also
Waller 1992) adjusted for one or two confounders by stratifica-
tionor standardisation (eg crashesmiles driven age group) Osga
1980 conducteda univariate analysis of crash rates in which no po-
tential confounders were accounted for but also calculated cumu-lative crash rates by mileage travelled Only Jonah 1982 Cooper
1988 and Savolainen 2007 conducted a multivariate analysis
Risk of bias in the case-control studies
A case-control design is suitable for the assessment of risk fac-
tors for outcomes such as motorcycle crashes injuries or deaths
but not appropriate for the evaluation of an intervention such
as motorcycle rider training It can potentially result in spurious
conclusions regarding the efficacy of the intervention because the
intervention has not been applied at random to an intervention
and control group Even with the control of possible confounders
using a case-control design to evaluate an intervention can lead tothese kinds of results since those who choose to undertake rider
training may differ on important prognostic factors that affect the
outcome compared to those who do not choose to participate in
rider training
Selection bias
A potential source of bias in case-control studies as with all other
study types where the intervention is not randomly allocated to
participants is that comparison groups are likely to differ on po-
tentially confounding characteristics This means that ideally all
cases need to be included in the study as well as a random sample
of controls taken from the population that gave rise to the cases If that is not possible cases and controls either need to be carefully
matched to make themcomparable or differences at baseline need
to be adjusted for in the analysis All four case-control studies were
population based Kraus 1975 included all fatalities and injured
cases leading to hospitalisation and appropriately recruited a ran-
dom sample of controls from the same population Hurt 1981 did
not manage to recruit all or a random sample of cases Only half
of the notifications of crashes to the police were received in time
by the researchers the research team for unreported reasons was
only able to respond to half of those again Data collection was
therefore completed for only about a fifth of all cases Neverthe-
less this was still a large sample of 900 cases Haworth 1997 using
16Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 1950
a similar study design was more likely to have included all cases
but this is not entirely clear from the study reported It is unclearfor example whether fatalities were included as they did not use
coronerrsquos reports for the recruitment of cases The MAIDS 2004
study was conducted in five European countries but it was unclear
how cases were recruited whether they were all cases during the
two years that the study lasted or a random sample of all cases
what the response rates were etc Controls were recruited from
randomly selected petrol stations but it was likewise not clear what
the eligibility criteria for inclusion in the study were how they
were recruited (eg randomly) and what the response rate was
Hurt 1981 and Haworth 1997 both matched controls to cases by
recruiting themat the sametime ofdayandsamedayof weekas the
crash occurred Hurt 1981 however recruited controls two years
later andit is likelythat there would have beendifferences betweenthe two comparison groups One important difference was that
during data collection for cases there was very little specialised
motorcycletrainingavailable in the studyregionwhereas two years
later when control data were collected there was some training
available
These four case-control studies also suffered from selection bias
due to potential differences between those that gave consent to
participate in the study and responded to interviews andquestion-
naires and those that did not None of the studies have assessed
such differences although Hurt 1981 did acknowledge that con-
trols giving consent to cooperate may have differed from those
motorcyclists who did not
Detection bias
Measures that can be taken to reduce detection bias in case-con-
trol studies are blinding of interviewers for case or control status
blinding of the participant for the study factor of interest validat-
ing measurement instruments and using these in an identical way
for cases and controls and cross checking self-reported informa-
tion with other more objective sources such as registers None of
the studies reported blinding of the intervention assessors to case
or control status but all used consistent measurement methods for
both case and control groups
All case-control studies used self-report as their main measure toassess whether cases and controls had received rider training and
if so the type of training The major potential source of bias in
case-control studies when self-report is used as a measure is recall
bias One way of reducing recall bias is by double-checking self-
reported data with more objective data Kraus 1975 also used po-
lice accident reports for information on rider characteristics but
such information was not available about rider training Haworth
1997 and MAIDS 2004 did not report on any measures to limit
recall bias Hurt 1981 reported further questioning of cases by
the interviewer when answers did not make sense or when par-
ticipants seemed to contradict themselves as part of the in-depth
unstructured interview technique This qualitative research tech-
nique can result in valuable information but as interviewers were
not blinded to case or control status of the participants it may have led to more thorough examination of cases compared to the
controls and thus differential detection bias
Intervention integrity
The level of detail about the rider training provided in the case-
control studies differed significantly but was minimal for each
Kraus 1975 simply asked whether lsquoanyrsquo motorcycle rider training
had lsquoeverrsquo been carried out Hurt 1981 asked participants whether
they hadbeen self taught informally taught or completed formally
taught rider training courses MAIDS 2004 asked whether partic-
ipants had had pre-license training additional (non-compulsory)
training other training or no training whereas Haworth 1997asked about the number of courses that a participant had com-
pleted whetherthey were basic intermediate or advanced courses
and how long ago the training had been done None of the case-
control studies reported collecting information about further de-
tails about any training programmes (eg type of training cur-
riculum duration of training)
Analysis
Kraus 1975 accounted for confounding factors through the exclu-
sion of females and stratification by age and frequency of motor-
cycle use Hurt 1981 matched cases and controls for time of day
and day of the week that the crash occurred and did not adjust forany confounding factors in the analysis Haworth 1997 matched
groups for time of day and day of the week that the crash occurred
and stratified by age blood alcohol level and licence status in the
analysis None of the studies used multivariate methods MAIDS
2004 did not provide any information about any participant char-
acteristics other than the study exposures of interest and did not
describe any methods to adjust for potential confounding factors
Effects of interventions
We have presented the results in a descriptive manner due to the
substantial heterogeneity in study types interventions and out-comes and the generally poor quality of the studies We have
grouped the interventions by mandatory and non-mandatory pre-
licence rider training and post-licence rider training and further
by study type and by outcome (crashes injurious and fatal crashes
and offences) We discuss results that are reported as statistically
significant (whether a P value was reported or not)
Pre-licence rider training - mandatory
Randomised trials
17Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2050
Crashes
Total group of participants
Anderson 1980 found a reduction in motorcycle crash rates in
the intervention group (MOST skill test with remedial training)
compared to the control group (who received the skill testwithout
a training component) The percent differenceranged from 143
six months after the programme (P lt 005) to 211 (P lt 0001)
after one year and 127 (P lt 001) after two years Hall 1988
found no statistically significant difference in crash rates between
intervention groups at any point in time
Subgroup of participants who obtained their licence only
Anderson 1980 found a reduction in motorcycle crash rates in the
intervention arm but this was statistically significant at one year
only (24 to 218 reduction P lt 005) Hall 1988 found no
statistically significant differences
Offences
Total group of participants
The percentage of offences in Anderson 1980 after six months
and one and two years was slightly higher in the intervention
arm compared to the control group but these differences were
not statistically significant Likewise the study by Hall 1988 did
not find a statistically significant difference in offences between
groups
Subgroup of participants who obtained their licence only
Hall 1988 found no significant difference in offence rates between
the intervention and control groups Anderson 1980 found thatthe percentage of offences increased in the intervention arms com-
pared to the control group This difference was statistically signif-
icant after two years (51 to 115 increase P lt 005)
Cohort studies
Crashes
Neitherof the cohort studies (Kloeden1994 and Billheimer1998)
found a statistically significant difference in crash rates between
the intervention and control groups
Offences
Billheimer 1998 (for the California Motorcyclist Safety Program
course) and Perrino 2002 did not find a statistically significant
difference in offence rates between the intervention and control
groups
Case-control studies
Crashes
MAIDS 2004 found that a slightly higher percentage of controls
(not involved in a crash) had had pre-license rider training than
cases (involved in a crash) This difference was not statistically significant (50 of controls vs 477 of cases OR 091 95 CI
076 to 110)
Pre-licence rider training - non-mandatory
Randomised trial
Injuries and deaths
Swaddiwudhipong 1998 found a significantly lower injury ratein the intervention villages (1051000 population 95 CI 92
to 120) compared to the control villages (1691000 population
95 CI 152 to 185) one year after the intervention This differ-
ence was not maintained two years after the intervention (127
95 CI 112 to 143 intervention 151 95 CI 135 to 167
control) There was no significant difference in death rates be-
tween groups
Cohort studies
Crashes
Cooper 1988 foundno statistically significant differencein crashes
between the intervention and control groups McDavid 1989
found a lower crash rate in the intervention group (14139) com-
pared to the control group (23139) but no test of statistical sig-
nificance was undertaken
Injuries and deaths
McDavid 1989 found that the intervention group had fewer in-
jurious crashes (814 crashes) compared to the control group (16
23 crashes) but no statistical significance test was reported
18Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2150
Offences
Cooper 1988 foundno statistically significantdifference in offencerates between the intervention and control groups McDavid 1989
foundno statistically significantdifference in offence ratesbetween
the intervention and control groups
Non-mandatory rider training
Non-randomised trials
Crashes
Neither Adams 1985 nor Leung 1987 found a statistically signif-icant effect of rider training on crash rates
Offences
Leung 1987 found a statistically significant lower offence rate in
the intervention group compared to the control group for both
males and females and for all age groups except for women 35
years or older (P values not reported)
Cohort studies
Crashes
Osga 1980 in the univariate analysis reported a lower percentage
of crashesin theintervention group (that continued to ride) (28
168595) than in the control group (34 112329) In the cu-
mulative crash rate analysis however Osga1980 found statistically
significant (at the P lt 001 level) higher crash rates in the interven-
tion group than in the control group at allmileage intervals (1000
to 29999 miles) travelled After riding more than 30000 miles
(when over 50 of the trained riders had been involved in a crash
compared to approximately 40 in untrained riders) this differ-
ence was no longer statistically significant Satten 1980 found a
lower percentage of crashes in the intervention group (13 969)
compared to the control group (25 2079) but the differencewas not assessed statistically Davis 1997 found a lower crash rate
in trained riders but it was unclear how this was calculated be-
cause the size of the control group was not reported Jonah 1982
and Billheimer 1998 (for the Experienced Rider Course) found no
statistically significant difference in crashes between the interven-
tion and control groups Although the difference between those
who completed the Experienced Rider Course and no course in
Billheimer 1998 neared significance at six months after the course
(P = 007) large P values were observed for one year after (P =
030) Raymond 1979 found a statistically significant higher crash
rate for the intervention group (1861000 miles ridden) com-
pared to the control group (1181000 miles ridden) in both males
and females (P lt 0001) especially during the first 5000 miles of
riding Waller 1992 found a statistically significant higher crashrate for the intervention group (27 crashes in three years or 070
crashes10000 person days) compared to the control group (5
crashes in three years or 013 crashes10000 person days) with a
crude relative risk of 538 (X 2=1513 plt005) Mortimer 1984
and Mortimer 1988 found no statistically significant difference in
crashes between the intervention and control groups
Savolainen 2007 found that the intervention group (those who
completed the Basic Rider Course) was 44 more likely to be
involved in crashes than the control group and those who com-
pleted the course multiple times were an additional 180 more
likely to be involved in a crash This finding was based on low
numbers and no P value or confidence level was given for either
result
Injuries and deaths
Satten 1980 also found a lower percentage of injurious crashes
(33 39) in the intervention group compared to the control
group(60 1220) although again no statistical significance test
was reported
Offences
Jonah 1982 did find a significantly lower self-reportedoffence rate
in the intervention group (15) compared to the control group(34 P lt 001) (no numbers given) Satten 1980 also found
a lower self-reported offence rate in the intervention group (1
71 versus 19100) but again this was not tested for statistical
significance Mortimer 1984 Mortimer1988 and Billheimer 1998
found no statistically significant differencein offence ratesbetween
the intervention and control groups
Case-control studies
CrashesMAIDS 2004 found that a slightly higher percentage of cases
(involved in a crash) had had additional non-mandatory rider
training than controls (not involved in a crash) This difference
was not statistically significant (17 of cases vs 12 of controls
OR 146 95 CI 064 to 337)
Rider training unspecified
Case-control studies
19Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2250
Injuries and deaths
Hurt 1981 found that the odds of injurious crashes were lower inthose who had received professional training compared to those
who had not Kraus 1975 found that in the younger age group
(less than 20 years) and the older age group (over 35 years) a larger
percentage of injured riders (12 and 21 respectively) reported
prior motorcycle rider training compared with the uninjured con-
trol group (6 and 5 respectively) (P lt 001) (no numbers
given) Haworth 1997 reported several different outcomes one
beingthat having completed at leastone trainingcoursecompared
to no training courses at all had no significant effect on the odds
of crashing (afteradjustment for age blood alcohol level or licence
status (learner permitprobationaryfull)) When comparing three
different levels of training courses (beginner intermediate and ad-
vanced) with no training no significant differences were found
in the odds of crashing However after adjustment for age and
blood alcohol level completing an advanced course as compared
to a beginners course was associated with a significant decrease in
the odds of severe motorcycle crashes resulting in serious injury or
death (OR 02 95 CI 01 to 08)
D I S C U S S I O N
The objective of this review was to quantify the effectiveness of
pre- andpost-licence motorcycle rider training on the reduction of
traffic crash involvement injuries deaths and offences committedby motorcycle riders Most included studies were cohort studies
There were a few randomised and non-randomised trials and case-
control studies
All studies compared either mandatory or non-mandatory formal
training with no formal training except for Haworth 1997rsquos case-
control study that compared levels of training (beginner interme-
diate and advanced) to each other
We found studies that evaluated the effectiveness of manda-
tory pre-licence training in reducing crash and offence rates
(Anderson 1980 Hall 1988 Kloeden 1994 Billheimer 1998
(the basic California Motorcyclist Safety Program component of
the study) Perrino 2002) non-mandatory training that when
passed resulted in a licence (Cooper 1988 McDavid 1989
Swaddiwudhipong 1998) non-mandatory training that could be
taken by motorcycle riders with or without a licence (Raymond
1979 Osga 1980 Satten 1980 Jonah 1982 Mortimer 1984
Adams 1985 Leung 1987 Mortimer 1988 Waller 1992 Davis
1997 Billheimer 1998 (the Experienced Rider Course compo-
nent of the study)) and unspecified rider training on crash injury
and death rates (Kraus 1975 Hurt 1981 Haworth 1997 MAIDS
2004)
The studies provided conflicting evidence regarding motorcycle
rider training which is likely due to the poor design quality and
reporting of the studies Most of the studies included in the review
were likely to be adversely affected by significant selection biasalong with substantial detection bias(measurement and recall bias)
which limits our ability to make conclusive comments
Mandatory pre-licence rider training
There were two RCTs three cohort studies and one case-control
study that examined pre-licence ridertraining The two RCTs eval-
uated the Motorcycle Skills Operating Test (MOST) (Anderson
1980 Hall 1988) Anderson 1980 found a statistically significant
reduction in crashes in the intervention group at six months one
and twoyearsfollow up However when only participantswho ob-
tainedtheirlicence were included in theanalysis thereduction wasstatistically significant only at one year follow up The other RCT
(Hall 1988) found no statistically significant differences in crash
rates at any point of time neither in the total population group
nor in the licensed subgroup Neither of the two cohort stud-
ies assessing crash rates found a statistically significant difference
in crash rates between intervention and control groups ( Kloeden
1994 Billheimer1998) andthe case-control study (MAIDS 2004)
also did not find a statistically significant difference in the odds of
experiencing a crash between trained and untrained riders With
regards to offences neither of the RCTs found a statistically sig-
nificant difference in offence rate between the intervention and
control groups in the analysis of the total group of participants
entering the trial However Anderson 1980 found a statistically significant increase in offence rates in the intervention group but
only after two years of follow up andonly in the analysis of partici-
pants who obtained their licence The two cohort studies assessing
offence rates found no statistically significant difference between
the groups (Billheimer 1998 Perrino 2002)
None of the five studies (RCTs or cohort) found a reduction in
crashes or offences following mandatory pre-licence rider training
except for Anderson 1980 who observed a statistically significant
reduction in crashes in the intervention group at one year in both
analyses It is not clear why both RCTs that evaluated the same
MOST and training intervention observed different results Both
studies enrolled large samples and observed a similar uptake of
licensure following the intervention It is possible that differencesmay have arisen due to different study samples (California ver-
sus New York) or that the intervention delivery differed between
studies
In both trials of the Motorcycle Skills Operating Test including
rider training only a third of motorcycle licence applicants ob-
tained their motorcycle licence It is possible that a licensing pro-
cess including a skills test may reduce crashes through decreased
exposure
Non-mandatory pre-licence training
20Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2350
Three studies assessed the effect of non-mandatory pre-licence
training The Cooper 1988 cohort study found no difference incrash or offence rates between the trained and untrained groups
We cannot however be confident about this result as the study
sufferedfrom selectionbiaswhichcouldhave biasedthe measure of
effect toward the null The RCT conducted by Swaddiwudhipong
1998 examined injury and death rates and found a statistically
significant lower injury rate in the intervention villages after one
year of follow up but no longer at two year follow up There
was no statistically significant difference in death rates between
intervention and control villages Our confidence in the results of
this study are hindered by the inadequacy of the randomisation
process that led to significant differences between the intervention
and control districts on two important variables The intervention
districts were more likely than the control districts to have a validlicence andto wear a helmetSincethese differences were nottaken
into account in later analyses they could explain the differences
observed in injury rates between districts Furthermore with only
six districts randomised the sample size was probably too small
to detect a difference between intervention and control groups
had there been one The McDavid 1989 cohort study found a
lower percentage of injuries in the intervention group although
a statistical test was not conducted McDavid 1989 drew their
intervention and control groups from different populations but
did adjust for confounders although they only analysed 40 of
the initial sample (139 trained and 139 untrained) Because of
these substantial weaknesses we are unable to draw conclusions
from this studyAn overall effect of pre-licence training on the reductionof crashes
injuries deaths and offences was not found although one should
be aware that the research of this issue to date has had significant
flaws However a mandatory skill test (such as the MOST) in
combination with and even without pre-licence training as part
of a licensing process seems to reduce the number of individuals
completing the licensing procedure and therefore indirectly re-
duces crash rates through a reduction in exposure to motorcycle
riding Making licensure a more difficult process may therefore be
seen as advantageous if it delays licensure and therefore reduces
exposure It is possible that the group who chose not to continue
riding a motorcycle would not have been the group to experience a
crash or commit an offence if they had continued to ride anyhowfor example because they are more safety conscious and this was
the reason for their giving up riding in the first place If this is the
case deterring individuals from completing their licence and rid-
ing may not necessarily decrease overall crashes deaths or offence
rates In addition other potentially adverse effects of complicat-
ing a licensing procedure need to be considered such as increases
in unlicensed riding In Thailand non-mandatory rider training
offered as a community intervention may have decreased injury
rates but this needs to be confirmed through further research
Non-mandatory training
The majority of the studies assessed non-mandatory training
(training independent of any licensing procedure) and the results
were varied The two non-randomised trials assessing crash rates found no statistically significant difference in crash rates between
the intervention and control groups (Adams 1985 Leung 1987)
Of the nine cohort studies Satten 1980 found a lower percentage
of crashesin the intervention group although it is unclear whether
this was a statistically significant result Davis 1997 also found a
lower percentage of crashesin theinterventiongroup although it is
unclear with which control group the intervention group wascom-
pared Raymond 1979 Osga 1980 Waller 1992 and Savolainen
2007 found a statistically significant higher crash rate in the in-
tervention group and the remaining cohort studies (Jonah 1982
Mortimer 1984 Mortimer 1988 Billheimer 1998) found no sta-
tistically significant difference in crash rates between intervention
and control groups No conclusions can be drawn from these stud-
ies with confidence The Satten 1980 and Davis 1997 studies did
not adjust for any confounding factors even though important
differences existed between the intervention and control groups
Although Raymond 1979 and Waller 1992 adjusted for a num-
ber of confounders at baseline (Waller 1992) and in the analy-
sis (Raymond 1979 Waller 1992) the intervention and control
groups were drawn from such different populations that it is pos-
sible that unmeasured confounders were responsible for the sta-
tistically significant increased risk of crashes observed in the inter-
vention group The observed difference could also be attributed totheobserved differential loss to followup (Raymond 1979) which
could bias the measure of effect towards or away from the null
The Osga 1980 study found a protective effect of rider training
in the univariate analysis but statistically significant higher crash
rates in the intervention group when accounting for riding expo-
sure Sex (a significantly higher proportion of females in the in-
tervention group) and other potentially confounding factors that
could explain thedifferencebetween the groups were howeverun-
accounted for Finally Savolainen 2007 did control for potential
confounding variables yet it is not clear whether variables were
controlled for simultaneously or one at a time One case-control
study (MAIDS 2004) found no statistically significant reduction
in the odds of experiencing a crash between trained and untrainedriders but the quality of this study was poor mainly because the
analysis did not account for any confounding factors
One cohort study (Satten 1980) assessed the effect of non-manda-
tory rider training on injurious crashes and found a lower percent-
age of injuries in the intervention group although a statistical test
was not conducted As mentioned above the Satten 1980 studyrsquos
validity is reduced due to its lack of adjustment for confounding
factors
The only non-randomised trial that assessed offence rates (Leung
1987) found a statistically significant lower offence rate in the in-
tervention group and of the five cohort studies that assessed of-
fence rates (Satten 1980 Jonah 1982 Mortimer 1984 Mortimer
21Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2450
1988) one study (Jonah 1982) also found a statistically significant
lower offence rate in the intervention group However we have lit-tle confidence in the results from Leung 1987 since potential con-
founding factors were not appropriately addressed Jonah 1982
did conduct a multivariate analysis but was hindered by an initial
low response rate The other four cohort studies did control for
several potential confounders and found no difference in offence
rates following non-mandatory post-licence rider training
The non-randomised trials and cohort studies evaluating non-
mandatory rider training were at high risk of substantial selection
bias due to the non-random allocation of participants as well
as low participation rates and attrition bias and could therefore
not provide conclusive evidence with regard to the effect of the
training It should be noted that despite the biases and other issues
with this group of studies most didnot detectmajor effects relatedto rider training in either direction This may suggest that any
potential impact of rider training will be modest
Studies examining unspecified modes of ridertraining
Three case-control studies assessed whether injured or deceased
motorcyclists were more or less likely to have had rider training
compared to control motorcyclists All had conflicting results
Hurt1981 foundthe odds of havingan injuriouscrash waslower in
those who had received rider training whereas Kraus 1975 found
that the odds of having an injurious crash were higher in youngerand older trained riders Haworth 1997 who compared advanced
versus beginner training found no effect on motorcycle injuries
or deaths although they did find that those who had completed
advanced training had reduced chances of crashing compared to
those who had done a beginnersrsquo training course Similar issues
with uncontrolled confounding and biases of selection as described
above arose with the three case-control studies Hurt 1981 suffers
from inadequate control of confounding via poor matching and
both the Kraus 1975 and the Haworth 1997 findings were only
controlled for limited factors by means of stratified analysis It is
likely that these limitations are the reason for the contradictory
directions of effect observed for the three case-control studies
Methodological weaknesses of the studies
In summary most studies investigating the effectiveness of mo-
torcycle rider training on reducing crashes injuries and death suf-
fered from serious methodological weaknesses
The majority of studies were non-randomised and were there-
fore limited mainly by selection bias Although the approaches of
matching stratification or regression analyses can minimise bias
(Fleiss 2004) they were poorly employed in the reviewed studies
The inadequate adjustment for confounders in particular adds
to our lack of confidence in the studiesrsquo results since it has been
demonstrated that individuals who choose to take motorcycle rider
training differ from those not taking rider training For instance
trained riders often differ from controls in age sex and experi-ence usually ride less and may have a more positive attitude to
safety as indicated by the reported more frequent wearing of seat
belts (Mortimer 1982)
The two large randomised trials reported that a large proportion
of their study group were never exposed to the risk of offences or
crashes because they never attained licensure This made it impos-
sible to assess whether a reduction in crashes in the intervention
group was a result of training or due to delayed or never attained
licensure
In some studies the sample sizes were too small to detect any sta-
tistically significant differences Follow-up time after training dif-
fered between studies and was an impediment to observing any
real effects of training if they existed As riding experience ac-cumulates over time any differences in outcomes between those
trained and untrained would diminish in studies with lengthy fol-
low-up times This dilution of any possible effect of rider training
may have also occurred in studies that mixed both novice and ex-
perienced riders in the study groups since it is hypothesised that
novice riders serve to benefit most from rider training This would
increase the difficulty in finding any effect of rider training par-
ticularly when sample sizes are small
Detection bias may also account for some of the differences ob-
served between trained and untrained riders Detection bias would
have occurred if the self-report hospital driver or police records
used to record the outcomes of crash injury death or offences
differed between the trained and untrained riders Research hasdemonstrated that not all crashes are reported to police for ex-
ample small crashes or off-road crashes (Kraus 1975) even serious
injury crashes have been found to be underreported (Anderson
1980) However this would only introduce detection bias if the
reporting of outcome differs for trained and untrained riders This
does appear possible in that responsible riders may be more likely
to enroll in a rider training course and also may be more likely
to report crashes to police than irresponsible riders meaning that
the number of crashes from riskier riders could be underrepre-
sented on records An analysis to investigate the effectiveness of
rider training may then show that rider training is not effective in
reducing crashes
Rather than using reported outcomes some studies used self-re-ported survey data although this technique suffers from non-
response biases In fact two studies reported differences in de-
mographic characteristics in response rates to questionnaires
(Anderson 1980 Hall 1988) Self-reported data also introduce
the risk of recall bias survival bias and social desirability response
bias Recall bias may have been equally distributed among trained
and untrained riders howeverit is hypothesised that trained riders
may have felt it was lsquoexpectedrsquo to report good driving histories
hence leading the rider training to appear effective
22Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2550
Limitations of this review
Limitations of this review were that we found very few recentstudies (the two most recent ones published in 2002 and 2007)
and that we found very few studies conducted in low-and middle
income countries even though our search strategy was set up to
identify such studies Anotherlimitation wasthat forthe reviewed
studies we could not assess all outcomes of rider training that are
potentially useful Some of the included studies assessed outcomes
such as use of safety clothing including helmets Such outcomes
are valuable since the use of motorcycle helmets has been demon-
strated to be effective in preventing head injuries (Liu 2004) De-
spite this the use of safety clothing was not included as an out-
come in this review because the studies measuring this outcome
were limited by poor data quality due to very low response rates
Another outcome of interest in assessing the effectiveness of rider
training is lsquotype of injuryrsquo As Haworth 2005 state in their review
of motorcycle licensing and training rider training could be con-
sidered effective if injuries become less severe because motorcycle
riders have learned how to avoid more serious injuries through
rider training The studies included in this review did not specify
type of injury and so this outcome could not be assessed Last
changes in knowledge and attitudes following rider training are
other potentially useful outcomes We have excluded such studies
(Osga 1979 Clayton 1990 Goldenbeld 2004) as these outcomes
are proxies rather than direct measures of the effectiveness of mo-
torcycle rider training We opted to report on the outcomes of
crashes injury death and offences alone since there is little ev-
idence about how changes in knowledge and attitudes relate to
these outcomes
A U T H O R S rsquo C O N C L U S I O N S
Implications for practice
Due to the poor quality of studies included in this review we were
unable to draw any conclusions about the effectiveness of rider
training on crash rates injurious or fatal or offence rates The
findings do suggest that mandatory pre-licence training may be
an impediment to completing a motorcycle licensing process thuspossibly indirectly reducing crash injury death and offence rates
through a reduction in exposure to riding a motorcycle On the
basis of the existing evidence it is not clear if and what type of
training reduces the risk of crashes injuries deaths or offences in
motorcyclists and the selection of the best rider training practice
can therefore not be recommended The suggestion that manda-
tory pre-licence training indirectly reduces exposure to riding the
motorcycle because trainees are impeded from completing the li-
censing process does not help those motorcyclists who persist in
obtaining their licence and riding the motorcycle Some type of
rider training is likely necessary to teach motorcyclists to ride a
motorcycle safely Good policy is based on sound evidence and
evaluation but such evidence for motorcycle rider training is seri-
ously lacking Policy makers are therefore encouraged to include arigorous evaluation component to any new or existing rider train-
ing programme Such an evaluation should ideally be randomized
and assess whether the training programme is effective and if so
which component(s) of the programme work best
Implications for research
Given theresults of this review which found littlereliable evidence
about the effectiveness of motorcycle rider training it is necessary
to build up an evidence base of the effectiveness of motorcycle
rider training in the reduction of crash injury death and offence
rates To this end well designed randomised controlled trials need
to be conducted that examine the impact of rider training oncrashes and offences A randomised controlled trial is the preferred
method as it would significantly reduce most biases described in
this review To minimise the impact of participants delaying or
never attaining their licence and therefore exposure as found in
the randomised trials in this review such a trial should include
riders who are already licensed and reliable measures of riding
exposure in a reasonably lengthy follow up period It is expected
that a randomised controlled trial of sufficient sample size would
observe equal balance on possible confounders by study arm how-
ever an assessment of potentially important confounders (such as
helmet wearing) by study group would enable control for these
variables in the analyses if necessary It is imperative that the effec-
tiveness of each component of the rider training (such as theorypractice knowledge skills and modification of behaviour) and the
synergy of such components are evaluated Outcomes should be
objective ideally those that are routinely collected rather than self-
reported outcomes Even though riding the motorcycle poses dif-
ferent challenges and incurs different risks lessons may be learned
from research into the effects of car driver training
Motorcycles are a common means of transport in low- and middle
income countries and usage is increasing As the burden of mo-
torcycle crash victims is mostly borne in such countries research
should be focused on the effectiveness of types of rider training on
the reduction of crashes injuries and fatalities in such countries
and be supported by higher income countries This review found
only one study conducted in a middle income country that sug-gested decreasing injury rates It would be beneficial to conduct
additional research to confirm this suggestion and to enhance our
knowledge of the role of rider training in low- and middle-income
settings
March 2006 saw the launch of the University of North Carolina
Highway Safety Research Centre partnership with the Motorcycle
Safety Foundation (MSF) and plans were revealed to conduct a
study of the potential benefits of motorcycle rider training The
study is funded by the MSF and the National Highway Traffic Sa-
fety Administration (NHTSA) Called rsquoThe Discovery Projectrsquo it
will compare data on crash rates from riders participating in a basic
23Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2650
motorcycle safety class with riders participating in a basic motor-
cycle safety class as well as a variety of continuing education classesThe collection of pilot data was due to start in March 2007 and
the data collection portion of the study is expected to take three
years (Highway Safety Research Center) It is vital that researchers
and policy makers collaborate to develop furtherhigh quality eval-
uations of motorcycle rider training programmes to help inform
development of new and effective training programmes
A C K N O W L E D G E M E N T S
Staff from The University of Sydney library for helping us locate
and retrieve full reports of many papers
R E F E R E N C E S
References to studies included in this review
Adams 1985 published data only
Adams A Collingwood V Job RFS Evaluation of the 1982 Australia
Post motorcycle rider training scheme NSW Australia Traffic
Accident Research Unit Traffic Authority NSW 1985 [ RN 85]
Anderson 1980 published data only
Anderson J Ford JL Peck RC Improved motorcyclist licensing and testing project Washington DC National Highway Traffic Safety
Administration (NHTSA) USA 19801ndash217 [
DOTndashHSndash805ndash598]
Billheimer 1998 published data only
Billheimer JW Evaluation of the California Motorcyclist Safety
Program Transportation Research Record 19981640100ndash9 [
TRB Report Number 980652]
Cooper 1988 published data only
Cooper PJ Rothe JP Motorcyclists who they are and why they do
what they do Transportation Research Record 1988116878ndash85
Davis 1997 published data only
David CF Evaluation of community traffic safety programs and
motorcycle operator training programs Connecticut Department of Transportation 199763
Hall 1988 published data only
Hall CJ Buchanan L Engle R Motorcycle rider education evaluationproject Washington DC USA National Highway Traffic Safety
Administration (NHTSA) US Department of Transportation
1988 [ DTNH 22ndash80ndashCndash05123]
Haworth 1997 published data only
Haworth N Smith R Brumen I Pronk N In Monash University
Accident Research Centre editor(s) Case-control study of motorcycle
crashes Victoria Australia Department of Transport and Regional
Development The Federal Office of Road Safety 19971ndash149 [
ISBN 0 642 25501 6 ]
Hurt 1981 published data only
Hurt H Ouellet J Thom D Motorcycle accident cause factors and
identification of countermeasures USA Department of
Transportation 1981 [ Technical Report DOT HS 806ndash862]
Jonah 1982 published data onlylowast Jonah BA Dawson N Bragg B Are formally trained motorcyclists
safer Accident Analysis amp Prevention 198214(4)247ndash55
Jonah BA Dawson NE Bragg BW Evaluation of the Canada
Safety Councilrsquos Motorcycle Training Program Proceedings of theInternational Motorcycle Safety Conference Washington DC
USA 198059ndash93
Kloeden 1994 published data only
Kloeden CN Moore VM McLean AJ Evaluation of the pre-license
training program for motorcyclists in South Australia Walkerville
South Australia Australia NHMRC Road Accident Research Unit
The University of Adelaide 19941ndash37 [ ISBN 0 7308 0607 3]
Kraus 1975 published data only
Kraus JF Riggins RS Franti CE Some epidemiological features of
motorcycle collision injuries American Journal of Epidemiology 1975102(1)74ndash98
Leung 1987 published data only
Leung KS Reding VA Evaluation of the Wisconsin motorcycle rider course Wisconsin Bureau of Policy Planning and Analysis
Wisconsin Department of Transportation 1987
MAIDS 2004 published and unpublished data
MAIDS In-depth investigations of accidents involving powered
two wheelers ACEM (Acemacemeu) 2004 Vol Final report
20179
McDavid 1989 published data only
McDavid JC Lohrmann BA Lohrmann G Does motorcycle
training reduce accidents Evidence from a longitudinal quasi-
experimental study Journal of Safety Research 198920(2)61ndash72
Mortimer 1984 published data only
Mortimer R Evaluation of the motorcycle rider course Accident
Analysis amp Prevention 198416(1)63ndash71
24Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2750
Mortimer 1988 published data only
Mortimer RG A further evaluation of the motorcycle rider courseJournal of Safety Research 198819(4)187ndash96
Osga 1980 published data only
Osga GA An investigation of the riding experiences of MSF rider
course participants Dissertation South Dakota USA The
University of South Dakota 19801ndash224
Perrino 2002 published data only
Perrino CS Ahmed A Callender A Rozier E Cantwell A Stewart
O The role of Marylandrsquos motorcycle rider course in promoting saver
behaviors and attitudes Baltimore USA National Transportation
Center Morgan State University 20021ndash52
Raymond 1979 published data only
Raymond S Tatum S An evaluation of the effectiveness of the RAC
ACU motorcycle training scheme - final report Road Safety ResearchUnit Dept of Civil Engineering University of Salford London
19791ndash131
Satten 1980 published data only
Satten RS Analysis and evaluation of the motorcycle rider courses
in thirteen Illinois counties International Motorcycle Safety
Conference Washington DC 1980145ndash93
Savolainen 2007 published and unpublished data
Savolainen Peter Mannering Fred Additional evidence on the
effectiveness of motorcycle training and motorcyclistsrsquo risk-taking
behaviour Transportation Research Board 2007 Vol March 20
19
Swaddiwudhipong 1998 published data only
Swaddiwudhipong W Boonmak C Nguntra P Mahasakpan P
Effect of motorcycle rider education on changes in risk behavioursand motorcycle-related injuries in rural Thailand Tropical Medicine
and International Health 19983(10)767ndash70
Waller 1992 published data only
Waller FL An analysis of the impact of the Texas Department of
Public Safety Motorcycle Operator Training Program upon
motocycle related traffic accidents Texas State University Public
Administration Program Applied Research Projects 199277
References to studies excluded from this review
Anderson 1980a published data only
Anderson LC Better biking a motorcycle rider improvement
program International Motorcycle Safety Conference Washington
DC USA Motorcycle Safety Foundation 1980231ndash53Aupetit 2008 published and unpublished data
Aupetit S Riff J Buttelli O Espie S Ergonomics research program
on motorcycle learning Contributions to the development of a
motorcycle simulator for training Transport Research Arena
Ljubljana 20089
Awane 1999 published data only
Awane T Integrating simulators in motorcycle safety education
Journal of International Association of Traffic and Safety Sciences 199923(1)26ndash35
Baldi 2005 published data only
Baldi S Baer JD Cook AL Identifying best practices states in
motorcycle rider education and licensing Journal of Safety Research200536(1)19ndash32
Batchler 1988 published data only
Batchler MW Evaluation of the pre-license motorcycle training course at the New South Wales traffic education centre Federal Office of
Road Safety (FORS) 19881ndash122 [ Report No CR 77 ]
Blanchard 2006 published data only
Blanchard HT Tabloski PA Motorcycle safety educating riders at
the teachable moment Mosby (CV) Company 200632330ndash2
Bonnett 2005 published data only
Bonnett J Mature age skills training for experienced motorcycle
riders Motorcycle ridersrsquo association of the Australian Capital
Territory 2005
Braver 2007 published data only
Braver ER et alPersuasion and licensure a randomized controlled
intervention trial to increase licensure rates among Maryland
motorcycle owners Traffic Injury Prevention 20078(1)39ndash46Clayton 1980 published data only
Clayton AB Sudlow DE An evaluation of the effectiveness of a one day training programme for learner motorcyclists Berkshire UK
Transport and Road Research Laboratory 19801ndash23 [ ISSN
0266ndash7045]lowast Clayton AB Sudlow DE An evaluation of the effectiveness of a one-day training programme for learner motorcyclists Berkshire UK
Road Safety Division Berkshire 19801ndash23 [ISSN 0266ndash7045]
Council 1978 published data only
Council FM Desper LP Dutt AK Driver education for motorcycle operation University of North Carolina Highway Safety Research
Center 19781ndash40
Crowther 2005 published data onlylowast Crowther G Engaging with motorcyclists UK police and theBikeSafe Road Safety Programme Published proceedings of the
Australasian Road Safety Research Policing Education Conference
Wellington New Zealand 2005 issue ISBN 0ndash473ndash10636ndash11ndash11
Daniello 2009 published data only
Daniello A Gabler HC Mehta YA Effectiveness of motorcycle
training and licensing Transportation Research Board 88th Annual
Meeting 200913
Dewanti 2007 published data only
Dewanti M Challenging efforts in promoting young motorcyclist
safety in Indonesia Road Safety on Four Continents 14th
International Conference Swedish National Road and Transport
Research Institute 2007
Du Plooy 1990 published data onlyDu Plooy DW Motorcycle training in South Africa The Human
Element International Motorcycle Safety Conference Orlando
Florida USA Motorcycle Safety Foundation 19904-1 to 4-9
Engel 1989 published data only
Engel U Krogsgaringrd Thomsen L [Faeligrdselsundervisning og
propaganda ndash en faeligrdselssikkerhedsfremmende foranstaltning]
Dansk Vejtidsskrift 198912295ndash7
Goldenbeld 2002 published data only
Goldenbeld C Houwing S de Craen S The development of
driving skills of young moped riders [De ontwikkeling van de
rijvaardigheid van jonge bromfietsers] SWOV Institute for Road
Safety Research (Stichting Wetenschappelijk Onderzoek
Verkeersveiligheid) Leidschendam the Netherlands 200263
25Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2850
Goldenbeld 2004 published data only
Goldenbeld C Twisk D de Craen S Short and long term effects of moped training a field experiment Transportation Research Part F 20047(1)1ndash16
Haworth 2000 published data only
Haworth N Smith R Kowadlo N Motorcycle training in Victoria
evaluation of rider training curriculum in Victoria Monash
University Accident Research Centre Reports 2000 Vol 165 issue
3131
Haworth 2005 published data only
Haworth N Mulvihill C Review of motorcycle licensing and
training Monash University Accident Research Centre 2005 [
Report no240]
Haworth 2006 published data only
Haworth N Mulvihill C Rowdon P Teaching old dogs new tricksTraining and older motorcyclists Journal of the Australasian College of Road Safety 200718(4)20ndash5lowast Haworth NL Mulvihill C Rowden PJ Teaching old dogs new
tricks Training and older motorcyclists Australasian Road Safety
Research Policing and Education Conference Gold Coast
Queensland 2006
Hill 1983 published data only
Hill D Motorcycle rider training testing and licensing Motor
Cycle Council of New South Wales Sydney 1983
Imatake 1980 published data only
Imatake Y On the activities in Japan of motorcycle safe riding
promotion committee (MSRPC) - especially the education of
unskilled riders of class 1 motorbikes International Motorcycle
Safety Conference Washington DC USA Motorcycle Safety
Foundation 1980255ndash68
Ingebritsen 1990 published and unpublished data
Ingebrigtsen S [Risikofaktorer ved ferdsel med moped og
motorsykkel] TOslashI-rapport 66 Oslo Transportoslashkonomisk
institutt 1990
Jain 2009 published and unpublished data
Jain A et alTwo wheeler accidents on Indian roads a study from
Mangalore India Elsevier 200916130ndash3
Jonah 1979 published data only
Jonah BA Dawson NE Validation of the motorcycle operator skill
test Accident Analysis amp Prevention 197911(3)163ndash71
Jonah 1981 published data onlyJonah BA Dawson NE Bragg WE Predicting accident
involvement with the Motorcycle Operator Skill Test Accident Analysis amp Prevention 198113(4)307ndash418
Kadar 1990 published data only
Kadar KJ Countermeasures motorcycle rider education delivery
system The Human Element International Motorcycle Safety
Conference Motorcycle Safety Foundation 19906-27 to 6-41
Kelsey 1986 published data only
Kelsey SL Liddicoat C Ratz M Licensing novice motorcyclists a
comparison of Californiarsquos standard test and the MOST II (Motorcycle
Operator Skill Test) administered at centralized testing offices Sacramento California USA California Department of Motor
Vehicles Research and Development Office 19861ndash74
Koch 1980 published data only
Koch H Advanced course for motorcycle riders InternationalMotorcycle Safety Conference Washington DC USA Motorcycle
Safety Foundation 1980269ndash81
Koch 1990 published data only
Koch H Brendicke R How to influence risk taking behaviour
within motorcycle rider training programs The Human Element
International Motorcycle Safety Conference Orlando Florida
USA Motorcycle Safety Foundation 1990Motorcycle Safety
Foundation 1990 p 9-1 to 9-25A
Labbett 2005 published data only
Labbett S Langham M Training can make the problem worse
Road Safety Congress 20051ndash12
Law 2005 published and unpublished data
Law TH Radin Umar RS Zulkaurnain S Kulanthayan S Impactof the effect of economic crisis and the targeted motorcycle safety
programme on motorcycle-related accidents injuries and fatalities
in Malaysia International Journal of Injury Control and Safety
Promotion 200512(1)9ndash21
Lowes 1990 published data only
Lowes BF In-traffic rider training The Human Element
International Motorcycle Safety Conference Orlando Florida
USA Motorcycle Safety Foundation 19907-46 to 7-57
Mahon 1976 published data only
Mahon J Teaching accident avoidance on the motorcycle Journal of Traffic Safety Education 1976231ndash29
Maxwell 1985 published data only
Maxwell GS Evaluation of a Program for Training MotorcycleRiders Evaluation and Program Planning 19858(4)339ndash47
Mayhew 1996 published data only
Mayhew DR Simpson SH Effectiveness and role of driver educationand training in a graduated licensing system Ottowa Canada The
Traffic Injury Research Foundation 19961ndash89 [ ISBN
0ndash920071ndash12ndash0]
McKnight 1987 published data only
McKnight AJ Milburn BG An evaluation of the modified
motorcycle operator skill test exercises NHTSA Report 1987 [
Report No DOT HS 807 212 ]
McPherson 1976 published data only
McPherson K McKnight AJ The development and evaluation of a
motorcycle skill test manual and knowledge test Washington DCUSA National Highway Safety Administration 1976
Mohan 2004 published data only
Mohan D Evidence-based interventions for road traffic injuries in
South Asia Journal of the College of Physicians and SurgeonsPakistan 200414(12)746ndash7
Morgan 1980 published data only
Morgan JK Advanced motorcycle training course content and
study evaluation International Motorcycle Safety Conference
Washington DC USA Motorcycle Safety Foundation 1980
283ndash312
Motoki 1990 published data only
Motoki M Yamazaki S A study on effective motorcycle rider
education The Human Element International Motorcycle Safety
26Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2950
Conference Orlando Florida USA Motorcycle Safety
Foundation 19909-26 to 9-57A
Nairn 1992 published data onlylowast Nairn RJ Motorcycle safety research literature review 1987 to
1991 Federal Office of Road Safety Australia 1993 Vol Report
CR 117
Newland 1980 published data only
Newland R Darby J Automotive practicespre motorcycle rider
program International Motorcycle Safety Conference Washington
DC USA Motorcycle Safety Foundation 1980127ndash9
Newland 1982 published data only
Newland R Motorcycle rider training and licensing requirements
Driver training steering a course for the future Melbourne
Victoria Transport Regulation Board 1982
Newland 1990 published data only
Newland R Motorcycle training and licensing in Victoria 1980-
1990 The Human Element International Motorcycle Safety
Conference Orlando Florida USA Motorcycle Safety
Foundation 1990
Ormston 2003 published data only
Ormston R Dudleston A Pearson S Stradling Evaluation of
Bikesafe Scotland Scottish Executive Social Research Edinburgh
2003
Osga 1979 published data only
Osga GS Ellingstad VS An evaluation of the effectiveness of the 1978
South Dakota motorcycle rider course program South Dakota USA
University of South Dakota 19791ndash89
Panou 2007 published data only
Panou M Bekiaris E A new concept on the integration of driving
simulators in driving training the train-all approach Road Safety
on Four Continents 14th International Conference Swedish
National Road and Transport Research Institute 2007
Peterson 1990 published data only
Peterson A Kadar K Status report 1980-1989 Minnesota rider
education statistics The Human Element International
Motorcycle Safety Conference Orlando Florida USA 199013-
113 to 13-129
Prem 1984 published data only
Prem H Good MC Motorcycle rider skills assessment ACT
Australia Office of Road Safety Department of Transport 1984
1ndash377 [ Report No CR 34 ]
Radin Umar 2006 published data only
Radin Umar RS Motorcycle safety programmes in Malaysia how
effective are they International Journal of Injury Control and Safety Promotion 200613(2)71ndash9
Rockwell 1990 published data only
Rockwell TH Kiger S Carnot M An evaluation of the Ohio
motorcyclist enrichment program Ohio Department of Highway
Safety 1990
Rothe 1987 published data only
Rothe JP A motorcycle riding course Motorcyclists Image and reality Canada Insurance Corporation of British Columbia 1987
59ndash83
Rowden 2007 published data only
Rowden P Watson B Haworth N What can riders tell us aboutmotorcycle training A view from the other side of the fence
Australasian Road Safety Research Policing and Education
Conference Melbourne Australia Melbourne 2007
RSC Victoria 1993 published data only
Road Safety Committee Inquiry into Motorcycle Safety in Victoria
Parliament of Victoria 1993
Russam 1979 published data only
Russam K Motorcycle training and licensing in Japan UK Transport
and Road Research Laboratory (TRRL) 1979 [ Report 916 ]
Saffron 1981 published data only
Saffron DG Driver and rider education training and licensing a
brief review Traffic Accident Research Unit Traffic Authority of
New South Wales NSW Australia Traffic Accident Research UnitTraffic Authority of New South Wales 1981
Schulz 1990 published data only
Schulz U Motorcycle accidents among learner riders Motorcycle
licensing Evaluation of the effectiveness of a Graduated Licensing Scheme Forshungshefte des Institut fur Zweiradsicherheit 1990
Sexton 2004 published data only (unpublished sought but not used)
Sexton B Baughan C Elliott M Maycock G The accident risk of
motorcyclists 2004TRL Report TRL 607(ISSN 09684107)58
Shephard 1986 published data only
Shepherd R Do Motorcycle Safety Foundation programs work
Driver 198619(9)16ndash8
Simpson 1990 published data onlySimpson HM Mayhew DR The promotion of motorcycle safety
training education and awareness Health Education Research
Theory and Practice 19905(2)257ndash64
Sokolov 1990 published data only
Sokolov O Safe motorcycle riding some aspects of motorcycle
riding in the USSR The Human Element International
Motorcycle Safety Conference Motorcycle Safety Foundation
19904ndash37 - 4-48
Steffens 1988 published data only
Steffens U Gawatz R Willmes G Effectiveness of mofa courses
[Wirksamkeit von Mofakursen] Unfall- und sicherheitsforschung
strasenverkehr 198867168
Sudlow 2003 published data only
Sudlow D Scoping study on motorcycle training Road Safety
Research Report London UK The British Institute of Traffic
Education Research Department for Transport 2003 Vol June
[ Report No 36]
Tasmania 1995 published data only
Mannering FL Grodsky LL Motorcycle rider training scheme
Traffic Education Research Centre Armidale Australia Driver
Training Australia Limited 1995 Vol April1ndash56
Tetard 1985 published data only
Tetard C Motorcycle ridersrsquo training to detect indices in an
intersection situation [Lrsquoentrainement a la prise drsquoinformation dans
lrsquoapprentissage de la conduite motocycliste] Le Travail Humain198548(2)147ndash59
27Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3050
Thackray 1980 published data only
Thackray RM Prescott JC Field test of the motorcycle ridercourse International Motorcycle Safety Conference Washington
DC USA Motorcycle Safety Foundation 1980195ndash7
Thompson 1994 published data only
Thompson M Evaluation of compulsory basic training for
motorcyclists Transport Research Laboratory Crowthorne
Berkshire UK 1994 [ Report 63]
Tobler 1990 published data only
Tobler U Individual driving instruction for motorcycle driving
learners the Swiss experiment of motorcycle driving school The
Human Element International Motorcycle Safety Conference
Orlando Florida USA Motorcycle Safety Foundation 1990
Transport 1989 published data only
Transport Communications Road Crash Statistics Australia Bureauof Transport and Communications Economics 198914ndash20
Tuan 2007 published data only
Tuan VA Shimizu T Towards development and evaluation of the
motorcycle drivers re-education program in Vietnam modeling of
motorcycle driverrsquos undesired behaviors 11th World Conference
on Transport Research 200725
Victor 1980 published data only
Victor DJ Traffic safety education for motorcyclists International
Motorcycle Safety Conference Washington DC USA Motorcycle
Safety Foundation 1980221ndash8
Waterer 1990 published data only
Waterer TH Compulsory training for motorcyclists from liberty
to compulsion The Human Element International Motorcycle
Safety Conference Orlando Florida USA Motorcycle Safety
Foundation 19904-49 to 4-57
Weller 1989 published data only
Weller RB Chandler EW Motorcycle safety and motorcycle
education past research and survey results Journal of Public Policy
amp Marketing 1989893ndash108
Wells 1981 published data only
Wells P Observations of some RACACU motorcycle training
schemes Transport and Road Research Laboratory Crowthorne
Berkshire UK 1981
Winn 1986 published data only
Winn G Two important trends in motorcycle safety regulations
rider education and conspicuity improvement American
Motorcyclist Association 1986147ndash52
Wisher 1988 published data only
Wisher G Reid J Motorcycle accidents in Norfolk a three-year study of motorcycle training and the Part 1 test Norfolk UK Road Safety
Division Highways Department Norfolk County Council 1988
Wood 1987 published data only
Wood HT Bowen R Evaluation of the revised learner permit scheme
July 1983 to December 1985 Victoria Australia Road Traffic
Authority 19871ndash27 [ ISBN 0 7306 0247 8]
Yamazaki 1990 published data only
Yamazaki S Nakanishi M Survey on effect of Honda rider
education The Human Element International Motorcycle Safety
Conference Orlando Florida USA 199012-167 to 12-187
References to ongoing studies
Burgess 2010 unpublished data only
Burgess C Broughton P Fylan F Stradling Interim evaluation of
the UKs National Ride program In L Dorn editor(s) Driver
behaviour and Training Vol IV Aldershot Ashgate Publishing
Limited 2010
Additional references
Anderson 1980
Anderson J Ford JL Peck RC Improved motorcyclist licensing and
testing project Volume I and II National Highway Traffic Safety
Administration US department of Transportation Washington
DC 1980
Bjornstig 1985
Bjornstig U Bylund P Lekander T Brorsson B Motorcycle fatalitiesin Sweden Acta Chirurgica Scandinavica 1985151(7)577ndash81
Clayton 1990
Clayton A Sudlow DE An evaluation of the effectiveness of a one-
day training program for learner motorcyclists The Human
Element International Motorcycle Safety Conference Orlando
Florida USA 199013-105 to 13-111
Cochrane Collaboration 2008
Higgins JP Green S Cochrane Handbook for Systematic Reviews of
Interventions 500 The Cochrane Collaboration 2008
Elliot 2003
Elliott MA Baughan CJT Broughton J Grayson G Knowles J et
alMotorcycle Safety a Scoping Study TRL [ Report No
0968ndash4107]Fleiss 2004
Fleiss JL Levin B Paik MC Statistical methods for rates and proportions 3rd Edition John Wiley amp Sons Inc 2004 [ ISBN
978ndash0ndash471ndash52629ndash2 ]
FORS 1999
FORS Federal Office of Road Safety Road risk for sober licensed
motorcyclists Vol Monograph 27 Federal Office of Road Safety
19991ndash2
Haworth 2005
Haworth N Mulvihill C Review of motorcycle licensing and training
Victoria Australia Monash University Accident Research Centre
20051ndash83 [ Report No 240]
Huwiler-Muumlntener 2002Huwiler-Muumlntener K Juumlni P Junker C Egger M Quality of
Reporting of Randomized Trials as a Measure of Methodologic
Quality JAMA 20022872801ndash4
Johnston 2008
Johnston P Brooks C Savage H Fatal and serious road crashes
involving motorcyclists Road Safety Research and Analysis Report
ATSB Monograph 20 Canberra Department of Infrastructure
Transport Regional Development and Local Government 0205
081ndash26
Juumlni 2001
Juumlni P Altman DG Egger M Systematic reviews in health care
Assessing the quality of controlled clinical trials BMJ 2001323
42ndash6
28Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3150
Kopits 2003
Kopits E Cropper M Traffic fatalities and economic growth TheWorld Bank Development Research Group Infrastructure and
Environment Policy research working paper 3035 2003
Leung 1983
Leung P Chau BTW Motorcycle accidents in Hong Kong pattern
of injury treatment causative factors and socio-economic
implications Injury 198315(1)23ndash9
Liu 2004
Liu B Ivers R Norton R Blows S Lo SK Helmets for preventing
injury in motorcycle riders Cochrane Database of Systematic Reviews 2004 Issue 2 [DOI 10100214651858CD004333pub3]
Mohan 2002
Mohan D Road safety in less-motorized environments future
concerns International Journal of Epidemiology 200231(3)527ndash32Mortimer 1982
Mortimer RG An evaluation of the motorcycle rider course
Proceedings of the 20th conference of the American Association for
Automotive Medicine Arlington Heights IL USA American
Association for Automotive Medicine 198277ndash96
Mullin 1997
Mullin BT Risk factors for motorcycle injury the role of age
gender experience training and alcohol PhD thesis Auckland
New Zealand University of Auckland 1997
Noordzij 2001
Noordzij PC Forke E Brendicke R Chinn BP Integration of needs
of moped and motorcycle riders into safety measures review and
statistical analysis in the framework of the European research
project PROMISING SWOV Institute for Road Safety Research
Leidschendam The Netherlands 2001 [ Report No Dndash2001ndash5]
Osga 1979
Osga G Ellingstad VS An evaluation of the effectiveness of the 1978 South Dakota Motorcycle Riders Course Program South Dakota
University of South Dakota Vermillion Human Factors Lab 1979
Oxford English Dictionary Online
httpoxforddictionariescom Oxford English Dictionary Online
Oxford University Press 2010
Peden 2004
Peden M Scurfield R Sleet D World Report on Road TrafficInjury Prevention Geneva World Health Organization Available
at httpwwwwhointviolence˙injury˙preventionpublications
road˙trafficworld˙reporten 2004Chapter 3
Radin Umar 1995
Radin Umar R Macakay G Hills B Preliminary analysis of
motorcycle accidents Short-term impacts of the running
headlights campaign and regulations in Malaysia J Traffic Med 199523(1)17ndash27
Tsai 1995
Tsai YJ Wang JD et alCase-control study of the effectiveness of
different types of helmets for the prevention of head injuries among
motorcycle riders in Taipei Taiwan American Journal of
Epidemiology 1995142(9)974ndash81
Vis 1995
Vis AA In-depth study of the hazards of motorcycling [De
onveiligheid van motorrijden nader bekeken] SWOV Institute for
Road Safety Research Leidschendam the Netherlands 1995 [
Report No Rndash95ndash69]
Weaver 1990
Weaver EA Motorcycle safety education in the United States
Motorcycle Safety Foundation Orlando Florida USA 1990
Winn 1987
Winn G Two important trends in motorcycle safety regulations
rider education and conspicuity improvement No more
information available Detroit MI US Society of Automotive
Engineers 1987147ndash52
Yuan 2000
Yuan W The effectiveness of the rsquoride-brightrsquo legislation for
motorcycles in Singapore Accident Analysis amp Prevention 200032
(4)559ndash63
Zambon 2006
Zambon F Hasselberg M Factors affecting the severity of injuries
among young motorcyclists a Swedish nationwide cohort study
Traffic Injury Prevention 20067(2)143ndash9lowast Indicates the major publication for the study
29Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3250
C H A R A C T E R I S T I C S O F S T U D I E S
Characteristics of included studies [ordered by study ID]
Adams 1985
Methods Non-randomised trial
Participants Initial sample size
Total of 162 Australia Post telegram delivery riders
Not clear how many in intervention and control groups
Analysis sample size
Intervention group 63Control group 35
Total 98 (61)
Interventions No formal training vs voluntary rider training course of 12 hours over 2 consecutive days (theory 4 hours practice
8 hours)
Intervention integrity
No information given
Outcomes Australia Post recorded crashes
Length of follow up 1 year
Notes Study conducted in NSW Australia
Anderson 1980
Methods Randomised trial
Participants Initial sample size
Total 41188 motorcycle licence applicants
Analysis sample size
Intervention group with skills training 13160
Intervention group without skills training 12634
Control group 15080
Total 40874 (99)
Interventions Standard licensing skill test vs the Motorcycle Operators Skill Test (MOST) with or without a training component
MOST is not to be confused with rider training because it is a licensing test The training component was for
participants who failed the skill test on the first attempt was of 3 hours duration and consisted of instructions about
the skills needed to pass the skill test The training consisted of 30 min theory and 2 frac12 hours practice on the
motorcycle
Intervention integrity
33 in MOST with skills training 31 in MOST without skills training and 43 in control group obtained their
motorcycle licence in the end
Outcomes Police recorded crashes and offences
Length of follow up 3 years
30Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3350
Anderson 1980 (Continued)
Notes Study conducted in California USA
Billheimer 1998
Methods Cohort study
Participants Initial sample size
Intervention group Unclear numbers of CMSP graduates and ERC graduates
Control group Over 16000 untrained motorcyclists
Analysis sample size
1139 matched CMSP-pairs and 1182 ERC-pairs
Interventions No formal training vs the mandatory (for those younger than 18 years) pre-licence 16-hours basic California
Motorcyclist Safety Program (CMSP) and the 8-hour Experienced Rider Course (ERC)
Intervention integrity
No information given
Outcomes Police recorded crashes and offences
Length of follow up 2 years
Notes Study conducted in California USA
Cooper 1988
Methods Cohort study
Participants Initial sample size
Intervention group 418 graduates and 57 who failed
Control group 402
Analysis sample size
Intervention group unclear
Control group unclear
Total 863 (984)
Interventions No formaltrainingvs non-mandatory pre-licence rider trainingThe content of thetrainingwas notfurther explainedIntervention integrity
Participants had all completed the rider training
Outcomes Police recorded crashes and offences
Length of follow up 7 years
Notes Study conducted in British Columbia Canada
31Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3450
Davis 1997
Methods Retrospective cohort study
Participants Initial sample size
Intervention group 9320 graduates
Control group unclear
Analysis sample size
Intervention group 9320 graduates
Control group unclear
Interventions No formal training vs the non-mandatory Connecticut Motorcycle Operator Training Program The content of
training was not further explained
Intervention integrity Participants had all completed the rider training
Outcomes Police recorded crashes severity of crashes
Length of follow up 1 year
Notes Study conducted in Connecticut USA
Hall 1988
Methods Randomised trial
Participants Initial sample sizeIntervention group A 6248
Intervention group B 6616
Intervention group C 6996
Control group 6604
Total 26464 motorcycle licence applicants
Analysis sample size
Intervention group A 6012 (96)
Intervention group B 6428 (97)
Intervention group C 6778 (97)
Control group 6502 (98)
Interventions Standard licensing skill test vs the Motorcycle Operator Skill Test (MOST II) without training (A) MOST II with
3 hours training (B) or MOST II with 20 hours training (C) The 3 hours training comprised instructions aboutthe skills needed to pass the skill test (30 min theory and 2 frac12 hours practice on the motorcycle) The 20 hours
training consisted of the Motorcycle Safety Foundationrsquos Motorcycle Rider Course (MRC) comprising of 8 hours
theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-road practice
Intervention integrity
25 and 26 obtained their motorcycle licence in the lsquoMOST 3-hour trainingrsquo and lsquoMOST 20-hour trainingrsquo
intervention arm respectively In the lsquoMOST testrsquo intervention arm 29 obtained their licence and in the control
group 35
Outcomes Police recorded crashes and offences
Length of follow up 3 years
32Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3550
Hall 1988 (Continued)
Notes Study conducted in New York USA
Haworth 1997
Methods Population based case-control study
Participants Cases Motorcycle riders or pillion riders involved in motorcycle crash (on road) Recruited from hospital and police
records Not clear whether all cases (especially fatalities) were included
Community controls Multiple matched sample of motorcycle rider or pillion who passed the crash site at the same
time of day and day of week that the crash occurred
Initial sample size
Cases 336
Controls 1195
Analysis sample size
Cases 234 (70)
Controls 561 (47)
Interventions Basic handling skills course beginner intermediate or advanced course measured through self-report not validated
Outcomes Injuries with an Injury Severity Score of 5 or greater or admission for at least four hours and fatalities from a
motorcycle crash (from hospital and police records)
Length of follow up two years or more prior to the interview
Notes Study conducted in Victoria Australia
Hurt 1981
Methods Population based case-control study
Participants Cases Motorcycle riders who had a motorcycle crash that was reported to police (traffic accident reports) Not all
cases included
Community controls Multiple matched sample of motorcycle riders who passed the crash site at the same time of
day and day of week as the crash occurred but 2 years later
Initial sample size
Cases 4500 notifications of crashes to police (only 50 received in time research team able to respond to 1126 for
unreported reasons data collection completed for 900)
Controls Number approached unreported 2310 consented and interviewed
Analysis sample size
Cases 808 (808900 data on rider training) - 18 of initial sample
Controls 670 (6702310 data on rider training) - lt 29 of initial sample
Interventions Training experience self taught friends-family school-club formal-professional others measured through self-
report validated
Outcomes Injuries and fatalities from a motorcycle crash (from hospital records and coronerrsquos office)
33Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3650
Hurt 1981 (Continued)
Notes Study conducted in Los Angeles USA
Jonah 1982
Methods Cohort study
Participants Initial sample size
Intervention group
2310 licensed MTP graduates who are active riders
Control group
2029 licensed motorcyclists who are active riders
Non-licensed and non-active riders excluded
Analysis sample size
Intervention group
811 interviews (35) 533 driver records (23)
Control group 1080 interviews (71) 905 driver records (45)
Interventions No training vs the non-mandatory Canada Safety Council Motorcycle Training Program (MTP) comprising of 4
hours theory and 16 hours practise (6 hrs on-road) over 2 weekends
Intervention integrity
Participants had all successfully completed the MTP
Outcomes Self-reported and police reported crashes and offencesLength of follow up 4 years
Notes Study conducted in Ontario Canada
Kloeden 1994
Methods Cohort study
Participants Initial sample size
Intervention group 2533 licence applicants
Control group Unclear
Sample of riders obtaining probationary licenceIntervention group 1625 (64)
Control group 5015 (unclear)
Non-active riders excluded from control group (~66)
Analysis sample size
Intervention group 631 (39)
Control group 318 (6)
Interventions No formal training vs the mandatory pre-licence Ridersafe course consisting of two levelsthe first is a training course
in the basics of motorcycle riding where a pass entitles the rider to obtain a learnerrsquos permit The second is a skill
test where a pass entitles the rider to obtain a probationary motorcycle licence The training course consists of two 4
hour training sessions on separate days involving instruction videos and actual riding on motorcycles The skill test
consists of a 4 hour preparation course of 74 minutes of lecturediscussion 38 minutes of video and 75 minutes of
34Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3750
Kloeden 1994 (Continued)
on-course riding which is the testing phase
Intervention integrity
Unclear what percentage in the Ridersafe group and the control group obtained their probationary licence halfway
and their full motorcycle licence in the end
Outcomes Police reported crashes
Length of follow up 1 year
Notes Study conducted in SA Australia
Kraus 1975
Methods Population based case-control study
Participants Cases All motorcycle riders with a confirmed medically treated motorcycle injury (including fatalities) from all
hospitals in study base and peripheral area hospitals and all deaths due to motorcycle crashes in the study base All
serious injuries (leading to hospitalisation) and fatalities likely to be included
Community controls A random sample of registered motorcycle owners in the same temporary and geographical
study base as the cases arose from
Initial sample size Cases 1273
Controls 738
Analysis sample size Cases 741 (58)
Controls 434 (59)
Interventions Motorcycle rider training (not specified) measured through self-report
Outcomes Injuries leading to hospitalisation fatalities
Length of follow up not specified
Notes Study conducted in Sacramento county USA
Leung 1987
Methods Non-randomised trial
Participants Initial sample sizeIntervention group unknown number of MRC graduates
Control group unknown number of licensed motorcyclists
Analysis sample size
Intervention group 2941
Control group 43094
Interventions No formal training vs non-mandatory Wisconsin motorcycle rider course a 22 hours off-street programme that
teaches the basic rules of safe motorcycling
Intervention integrity
50 in the MRC group and 100 in the control group obtained their motorcycle licence in the end
35Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3850
Leung 1987 (Continued)
Outcomes Police recorded crashes and offences
Length of follow up up to 3 years
Notes Study conducted in Wisconsin USA
MAIDS 2004
Methods Population based case-control study
Participants Cases Motorcycle riders who had a motorcycle crash Unclear whether all cases included
Community controls Motorcycle riders identified at petrol stations Unclear whether matched to cases
Initial sample size
Cases not reported
Controls not reported
Analysis sample size
Cases 921 (Italy 200 Netherlands 200 Spain 121 Germany 250 France 150)
Controls 923 (Italy 200 Netherlands 200 Spain123 Germany 250 France 150)
Interventions Pre-license rider training or additional training (non-compulsory)
Outcomes Crashes
Notes Study in five sampling areas located in five European countries (France Germany Netherlands Spain and Italy)
McDavid 1989
Methods Cohort study
Participants Sample size after first match
Intervention group 346 male training graduates
Control group 346 male motorcycle riders licensed in same year
Sample size after second match
Intervention group 139 (40)
Control group 139 (40)
Analysis sample size
Intervention group 139 (40)
Control group 139 (40)
Interventions No formal training vs the non-mandatory pre-licence British Columbia Safety Councilrsquos 37-hour motorcycle safety
programme
Intervention integrity
Participants had all completed and passed the training
Outcomes Police reported crashes injurious crashes
Length of follow up 1-6 years
36Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3950
McDavid 1989 (Continued)
Notes Study conducted in British Columbia Canada
Mortimer 1984
Methods Cohort study
Participants Initial sample size
Intervention group 608 MRC graduates
Control group 360 motorcycle riders
Response rates
Intervention group 360 (59)
Control group 324 (90)
Non-active riders excluded
Analysis sample size
Intervention group 213 (35)
Control group 303 (84)
Interventions No formal training vs the non-mandatory Motorcycle Safety Foundationrsquos Motorcycle Rider Course comprising of
8 hours theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-road practice
Intervention integrity
Participants had all completed and passed the MRC
Outcomes Self-reported crashes and offencesLength of follow up 1 year
Notes Study conducted in Illinois USA
Mortimer 1988
Methods Cohort study
Participants Initial sample size
Intervention group 2307 MRC graduates
Control group 556 active licensed motorcycle riders
Response ratesIntervention group 1298 (56)
Control group 500 (90)
Non-active riders excluded
Analysis sample size
Intervention group 913 (40)
Control group 500 (90)
Interventions No formal training vs the non-mandatory Motorcycle Safety Foundationrsquos Motorcycle Rider Course comprising of
8 hours theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-road practice
Intervention integrity
Participants had all completed and passed the MRC
37Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4050
Mortimer 1988 (Continued)
Outcomes Self reported crashes and offences
Length of follow up 1 year
Notes Study conducted in Illinois USA
Osga 1980
Methods Cohort study
Participants Initial sample sizeIntervention group 1163 MSF MRC participants invited 818 (74) responded
Control group 329 motorcycle riders responded to recruitment strategies
Non-active riders excluded from analysis
Analysis sample size
Intervention group 595 (73)
Control group 329 (100)
Interventions No formal training vs the non-mandatory Motorcycle Safety Foundationrsquos Motorcycle Rider Course (MRC) com-
prising of 8 hours theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-road
practice
Intervention integrity
Participants had all successfully completed the MRC
Outcomes Self-reported crashes police recorded crash reports
Length of follow up 6 months - 3 12 years
Notes Study conducted in South Dakota USA
Perrino 2002
Methods Cohort study
Participants Initial sample size
Intervention group 2144 (unclear) MRC graduates (motorcycle licence applicants)
Control group 2601 (unclear) regular licensing test graduates (motorcycle licence applicants)Analysis sample size
Intervention group 1070 (unclear)
Control group 867 (unclear)
Interventions No formal training vs the mandatory (for those younger than 18 years) pre-licence Maryland Motorcycle Rider
Course (MRC) consisting of 8 hours theory and 12 hours practice
Intervention integrity
No information given
Outcomes Self-reported and police recorded crashes and offences
Length of follow up 1 year
38Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4150
Perrino 2002 (Continued)
Notes Study conducted in Maryland USA
Raymond 1979
Methods Cohort study
Participants Initial sample size
Intervention group
258 volunteers mainly (but not all) novice riders
Control group
1104 provisional licence applicants
Analysis sample size
Intervention group 190 (74)
Control group 627 (57)
Interventions No formal training vs the non-mandatory RACACU (Royal Automobile ClubAuto-Cycle Union) training scheme
The training program consisted of either 24 one-hour sessions or 13 two-hour sessions Twelve hours of the training
focused on theory about basic operational procedures traffic laws and highway code The practical session focused
on basic operational skills riding on a training area and riding in traffic with an instructor A proficiency test was
administered at the end of the course
Intervention integrity
No information given
Outcomes Self-reported crashes (including minor non-injury non-damage and off-road crashes)
Length of follow up 1 year
Notes Study conducted in the UK
Satten 1980
Methods Cohort study
Participants Initial sample size
Intervention group interviewers kept interviewing until they had 100 riders selected from a population of 891
enrollees in the MSF MRC course (unknown how many were asked and how many refused to participate)Control group unknown number of registered motorcycle riders
Non-active riders excluded from analysis
Analysis sample size
Intervention group 71
Control group 100
Interventions No formal training vs the non-mandatory Motorcycle Safety Foundationrsquos (MSF) Motorcycle Rider Course (MRC)
comprising of 8 hours theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-
road practice
Intervention integrity
Participants had all successfully completed the MRC
39Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4250
Satten 1980 (Continued)
Outcomes Self-reported crashes injuries and offences
Length of follow up 3 years
Notes Study conducted in Illinois USA
Savolainen 2007
Methods Cohort study
Participants Initial sample size
Intervention group 4000 riders that had conducted the MSF MRC course (passed or failed)
Control group 4000 motorcycle owners or permit or endorsement holders from the Indiana Bureau of Motor
Vehicles database
Analysis sample size
Intervention group 739 (18)
Control group 588 (15)
Interventions Intervention integrity
No information given
Outcomes Probability of crash involvement
Length of follow up up to 5 years
Notes Study conducted in Indiana USA
Swaddiwudhipong 1998
Methods Randomised trial
Participants Initial sample size
Intervention group
all 16 villages of 3 sub-districts
(participants were practising motorcyclists licensed and unlicensed)
Control group
18 villages of 3 other sub-districtsAnalysis sample size
Intervention group 3 districts (100)
Control group 3 districts (100)
Interventions No rider education vs non-mandatory community based rider education consisting of theory on motorcycle injury
risk helmet use traffic laws vehicle regulations and traffic signs and a written and skill test leading to a licence
Intervention integrity
No information given
Outcomes Injuries fatalities
Length of follow up 2 years
40Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4350
Swaddiwudhipong 1998 (Continued)
Notes Study conducted in Tak Province rural Thailand
Waller 1992
Methods Cohort study
Participants Initial sample size
Intervention group 349 riders who succesfully completed the Department of Public Safety motorcycle operator
training in 1988
Control group 349 motorcyclists who never completed the DPS motorcycle operator training
Analysis sample size
Intervention group 349
Control group 349
Interventions No DPS motorcycle operator training vs non-mandatory DPS motorcycle operator training The DPS training is
the Motorcycle Safety Foundationrsquos Motorcycle Rider Course comprising eight hours of theory (lectures discussions
films and multiple choice tests) a textbook and 12 hours off-road practice
Intervention integrity
All 349 riders succesfully completed the rider training
Outcomes Police recorded crashes
Length of follow up 3 years
Notes Study conducted in
Characteristics of excluded studies [ordered by study ID]
Study Reason for exclusion
Anderson 1980a Study type review and process evaluation of a rider course
Aupetit 2008 Outcome proposal of a rider training curriculum that combines traditional training and a motorcycle riding
simulator
Awane 1999 Study type review on use of simulators in rider and driver training
Baldi 2005 Intervention not rider training but intervention integrity of rider training availabilty of rider education
implementation of graduated licensing system etc on a state-wide level
Batchler 1988 Study type description of a training course
Blanchard 2006 Study type not a study assessing the association between rider training and outcomes
Bonnett 2005 Outcome qualitative data regarding satisfaction with the training programme
41Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4450
(Continued)
Braver 2007 Outcome licensure rates
Clayton 1980 Outcome knowledge skills attitude
Council 1978 Outcome crash reduction in ANY vehicle not motorcycles exclusively
Crowther 2005 Study type qualitative study
Daniello 2009 Study type review of the effectiveness of motorcycle training and licensing
Dewanti 2007 Intervention emotional characteristics behaviour (ie not motocycle rider training)
Du Plooy 1990 Study type review of types of rider training courses in South Africa
Engel 1989 Study type uncontrolled before-after study
Goldenbeld 2002 Outcome skills
Goldenbeld 2004 Outcome knowledge and skills
Haworth 2000 Intervention balance between attitudinal andvehicle control skills acrosscurriculla of motorcycle rider training
courses
Haworth 2005 Study type review
Haworth 2006 Study type cross-sectional survey
Hill 1983 Study type review and description of training system
Imatake 1980 Study type course descriptionreview
Ingebritsen 1990 Intervention risk factors for motorcycle crashes
Jain 2009 Study type cross-sectional descriptive study of motorcycle crashes over a 5 year period
Jonah 1979 Study type cross-sectional survey Outcome scores on motorcycle operator skill test
Jonah 1981 Intervention evaluation of a motorcycle rider skill test not motorcycle rider training
Kadar 1990 Study type review of components of rider training courses
Kelsey 1986 Intervention evaluation of a motorcycle rider skill test not motorcycle rider training
Koch 1980 Study type reviewcourse description
Koch 1990 Study type description of course content
42Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4550
(Continued)
Labbett 2005 Study type essay about possible explanations for motorcycle deaths
Law 2005 Intervention targeted motorcycle safety programme not specifically rider training
Lowes 1990 Study type review and descriptions of rider training
Mahon 1976 Study type review of motorcycle rider training methods
Maxwell 1985 Outcomesrecommendations forprogramme operatorsand otherson howthe motorcycle trainingprogramme
works and how it can be improved
Mayhew 1996 Study type review
McKnight 1987 Study type evaluation of methodological and administrative aspects of a motorcycle operator skill test
McPherson 1976 Outcome knowledge and skills
Mohan 2004 Study type review of evidence of interventions to reduce road traffic injuries
Morgan 1980 Study type reviewcourse description
Motoki 1990 Study type proposal and development of rider training curriculum
Nairn 1992 Study type literature review
Newland 1980 Study type description of development of motorcycle safety course
Newland 1982 Study type descriptionreview of a rider training course
Newland 1990 Study type uncontrolled BeforeAfter study
Ormston 2003 Outcome changes in attitude and self reported behaviour
Osga 1979 Outcome knowledge attitudes and skills
Panou 2007 Intervention new simulation tool to teach driverrider
Peterson 1990 Study factor demographic details of course participants completing the course vs those not completing the
course
Prem 1984 Study type experimental
Study factor examines association between skills and performance in tests
Radin Umar 2006 Intervention motorcycle safety programme not specifically rider training
Rockwell 1990 Study type methodologically flawed study type
43Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4650
(Continued)
Rothe 1987 Study type qualitative study of rider training in general
Rowden 2007 Study type qualitative study
RSC Victoria 1993 Study type review of motorcycle rider training and education in Victoria
Russam 1979 Study type review of training programmes in Japan
Saffron 1981 Study type literature review
Schulz 1990 Study factor predictors of motorcycle crashes
Sexton 2004 Study type cross-sectional analytic analysis
Shephard 1986 Study factor reviewdescription of programme
Simpson 1990 Study type review of motorcycle rider trainingeducation
Sokolov 1990 Study factor review of motorcycle riding courses in USSR
Steffens 1988 Outcome self-reported behaviour experiences and researcher observed riding behaviour
Sudlow 2003 Study type review of motorcycle training in the UK
Tasmania 1995 Study type cross-sectional descriptiveprocess evaluation
Tetard 1985 Study type experimental
Outcome detection of indices
Thackray 1980 Outcome instructional effectiveness user acceptance administrative feasibility
Thompson 1994 Study type cross-sectional survey
Outcome attitudes and knowledge
Tobler 1990 Study type review of course curricula
Transport 1989 Intervention syllabus for rider training
Tuan 2007 Study type cross-sectional study (survey) which does not assess the effects of an intervention but examines
motocyclistsrsquo behaviour
Victor 1980 Study type review
Waterer 1990 Study type review of UK rider training issues
Weller 1989 Study type review of studies evaluating motorcycle safety and education measures
44Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4750
(Continued)
Wells 1981 Study type description of different methods of training
Winn 1986 Study type review of rsquohistoryrsquo of rider training
Wisher 1988 Study type cross-sectional survey
Wood 1987 Study type Interrupted Time Series but did not meet the criteria for an ITS as published by the Cochrane
Effective Practice and Organisation of Care Group
Yamazaki 1990 Study type cross-sectional survey
Outcome riding skills skills retention and attitude
Characteristics of ongoing studies [ordered by study ID]
Burgess 2010
Trial name or title UKs National Ride program
Methods Cohort study
Participants Motorcycle offenders
Interventions Police led diversion program
Outcomes Self reported attitudinal and behavioural measures
Starting date
Contact information
Notes Expected to be completed in early 2010
45Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4850
D A T A A N D A N A L Y S E S
This review has no analyses
A P P E N D I C E S
Appendix 1 Search strategy
We searched the following electronic databases
bull Cochrane Injuries Group Specialised Register (searched 22 Sept 2008)
bull CENTRAL (The Cochrane Library 2008 Issue 3)
bull TRANSPORT 1988 to 200706 (latest date available)
bull ATRI Australian Transport Index (to Sept 2008)
bull MEDLINE (Ovid SP) 1950 to Sept 2008
bull EMBASE (Ovid SP) 1980 to Sept 2008
bull CINAHL (EBSCO) 1982 to Sept 2008
bull WHOLIS (World Health Organization Library Information System) (to Sept 2008)
bull PsycINFO (Ovid SP) 1806 to Sept 2008
bull LILACS (BIREME) (Latin American and Caribbean Health Sciences) (1982 to Sept 2008)
bull ISI Web of Science Social Sciences Citation Index (SSCI) 1970 to Sept 2008
bull ERIC (searched 25 Sept 2008)
bull ZETOC (searched 25 Sept 2008)
bull SIGLE (searched 25 Sept 2008)
All of the databases were searched using the following text-word search strategy The strategy was adapted using either rsquorsquo or rsquo$rsquo as atruncation symbol
1 motor-cycl or motorcycl or motor-bik or motorbik or motor-scooter or motorscooter or scooter or moped or motocycl
or motocicl or moto
2 train or educat or instruct or program or course or qualif or learn or improv
3 1 and 2
H I S T O R Y
Protocol first published Issue 2 2005
Review first published Issue 10 2010
46Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4950
C O N T R I B U T I O N S O F A U T H O R S
Katina Kardamanidis was the first author of the protocol conducted the literature search extracted full studies assessed the studies
according to inclusion criteria assessed quality of the studies compared findings with the second author discussed disagreements with
the third author wrote the drafts and the final review
Rebecca Ivers was the second author of the protocol assessed the studies according to inclusion criteria assessed quality of the studies
compared findings with the first author edited the final draft and provided overall guidance for the process
Alexandra Martiniuk assessed quality of the studies compared findings with the first author discussed disagreements with the third
author and helped write the first draft and the final review
Mark Stevenson contributed to the protocol reviewed methods of studies and contributed to the final draft
Katrina Thistlethwaite conducted a second literature search extracted full studies and contributed to the final draft
D E C L A R A T I O N S O F I N T E R E S T
None known
S O U R C E S O F S U P P O R T
Internal sources
bull No sources of support supplied
External sources
bull National Health and Medical Research Council of Australia Australia
Salary support for Martiniuk Ivers and Stevenson
bull University of Sydney Sesqui Research Grant Australia
Direct research costs and salary support for Kardamanidis
D I F F E R E N C E S B E T W E E N P R O T O C O L A N D R E V I E W
There are no differences between the protocol and the review
I N D E X T E R M S
Medical Subject Headings (MeSH)
lowastMotorcycles Accident Prevention [lowastmethods] Accidents Traffic [lowastprevention amp control] Licensure Program Evaluation Wounds
and Injuries [mortality lowastprevention amp control]
47Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 5050
MeSH check words
Humans
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 250
T A B L E O F C O N T E N T S
1HEADER 1ABSTRACT
2PLAIN LANGUAGE SUMMARY
2BACKGROUND
3OBJECTIVES
3METHODS
11RESULTS
20DISCUSSION
23AUTHORSrsquo CONCLUSIONS
24ACKNOWLEDGEMENTS
24REFERENCES
29CHARACTERISTICS OF STUDIES
46DATA AND ANALYSES
46APPENDICES
46HISTORY
46CONTRIBUTIONS OF AUTHORS
47DECLARATIONS OF INTEREST
47SOURCES OF SUPPORT
47DIFFERENCES BETWEEN PROTOCOL AND REVIEW
47INDEX TERMS
iMotorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 350
[Intervention Review]
Motorcycle rider training for the prevention of road trafficcrashes
Katina Kardamanidis1 Alexandra Martiniuk 1 Rebecca Q Ivers1 Mark R Stevenson1 Katrina Thistlethwaite1
1The George Institute for Global Health Sydney Australia
Contact address Rebecca Q Ivers The George Institute for Global Health PO Box M201 Missenden Road Sydney NSW 2050
Australia riversgeorgeorgau
Editorial group Cochrane Injuries GroupPublication status and date New published in Issue 10 2010
Review content assessed as up-to-date 8 August 2010
Citation Kardamanidis K Martiniuk A Ivers RQ Stevenson MR Thistlethwaite K Motorcycle rider training for the
prevention of road traffic crashes Cochrane Database of Systematic Reviews 2010 Issue 10 Art No CD005240 DOI
10100214651858CD005240pub2
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
A B S T R A C T
Background
Riding a motorcycle (a two-wheeled vehicle that is powered by a motor and has no pedals) is associated with a high risk of fatal crashes
particularly in new riders Motorcycle rider training has therefore been suggested as an important means of reducing the number of
crashes and the severity of injuries
Objectives
To quantify the effectiveness of pre- and post-licence motorcycle rider training on the reduction of traffic offences traffic crash
involvement injuries and deaths of motorcycle riders
Search strategy
We searched the Cochrane Injuries Group Specialised Register CENTRAL (The Cochrane Library 2008 Issue 3) TRANSPORT
MEDLINE EMBASE CINAHL WHOLIS (World Health Organization Library Information System) PsycInfo LILACS (LatinAmerican and Caribbean Health Sciences) ISI Web of Science Social Sciences Citation Index (SSCI) ERIC ZETOC and SIGLE
Database searches covered all available dates up to October 2008 We also checked reference lists of relevant papers and contacted study
authors in an effort to identify published unpublished and ongoing trials related to motorcycle rider training
Selection criteria
We included all relevant intervention studies such as randomised and non-randomised controlled trials interrupted time-series and
observational studies such as cohort and case-control studies
Data collection and analysis
Two review authors independently analysed data about the study population study design and methods interventions and outcome
measures as well as data quality from each included study and compared the findings We resolved differences by discussion with a
third review author
1Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 450
Main results
We reviewed 23 studies three randomised trials two non-randomised trials 14 cohort studies and four case-control studies Five
examined mandatory pre-licence training 14 assessed non-mandatory training three of the case-control studies assessed lsquoanyrsquo type of
rider training and one case-control study assessed mandatory pre-licence training and non-mandatory training The types of assessed
rider training varied in duration and content
Most studies suffered from serious methodological weaknesses Most studies were non-randomised and controlled poorly for con-
founders Most studies also suffered from detection bias due to the poor use of outcome measurement tools such as the sole reliance
upon police records or self-reported data Small sample sizes and short follow-up time after training were also common
Authorsrsquo conclusions
Due to the poor quality of studies identified we were unable to draw any conclusions about the effectiveness of rider training on crash
injury or offence rates The findings suggest that mandatory pre-licence training may be an impediment to completing a motorcycle
licensing process possibly indirectly reducing crashes through a reduction in exposure It is not clear if training (or what type) reducesthe risk of crashes injuries or offences in motorcyclists and a best rider training practice can therefore not be recommended As some
type of rider training is likely to be necessary to teach motorcyclists to ride a motorcycle safely rigorous research is needed
P L A I N L A N G U A G E S U M M A R Y
Motorcycle rider training for preventing road traffic crashes
Riders of motorcycles (a two-wheeled vehicle that is powered by a motor and has no pedals - Oxford English Dictionary Online)
especially novice riders have an increased risk of being involved in fatal crashes compared to other road users Motorcycle rider training
could be an important way of reducing the number of crashes and the severity of injuries
The authors of this review examined all research studies that report an evaluation of the effectiveness of motorcycle rider coursesin reducing the number of traffic offences motorcycle rider crashes injuries and deaths This review included 23 research studies
including three randomised trials two non-randomised trials 14 cohort studies and four case-control studies The types of rider training
that were evaluated varied in content and duration The findings suggest that mandatory pre-licence training may present a barrier to
completing a motorcycle licensing process thus possibly indirectly reducing crash injury death and offence rates through a reduction
in exposure to riding a motorcycle However on the basis of the existing evidence it is not clear if (or what type of) training reduces
the risk of crashes injuries deaths or offences in motorcyclists and the selection of the best rider training practice can therefore not be
recommended
It is likely that some type of rider training is necessary to teach motorcyclists basic motorcycle handling techniques and to ride a
motorcycle safely It is therefore important that further research work be conducted to rigorously evaluate motorcycle rider training
courses particularly in low income countries where the main burden of motorcycle injuries and deaths occur
B A C K G R O U N D
Each year anestimated 12million peopleare killedin road crashes
worldwide and an additional 50 million people are injured Of
these deaths 85 occur in low-income and middle-income coun-
tries It has been predicted that between the year 2000 and 2020
without increased efforts the number of road traffic deaths in
high-income countries will decrease by approximately 30 yet
in low- and middle-income countries the number is expected to
increase by over 80 (Kopits 2003 Peden 2004) Road traffic
injuries are likely to become the third leading cause of a global
burden of disease and injury by 2020 if further action is not taken
(Peden 2004)
Riding a motorcycle (a two-wheeled vehicle that is powered by a
motor and has no pedals - Oxford English Dictionary Online) is
associated with a higher risk of fatalities compared with driving a
car (Leung 1983 Bjornstig 1985 Radin Umar 1995 Tsai 1995
FORS 1999 Yuan 2000 Johnston 2008) In Australia motorcy-
2Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 550
clesaccounted for45 of allAustralian passenger vehicle registra-
tions and 09 of vehicle kilometres travelled in 2007 Howevermotorcycle riders accounted for approximately 15 of all road
crash deaths and an even higher proportion of serious injuries
Per distance travelled the Australian rate of motorcyclist deaths
was approximately 30 times the rate for car occupants The cor-
responding rate for a serious injury was approximately 41 times
higher Similar elevated rates were also found in other developed
countries (Johnston 2008) In low and middle-income countries
the motorcycle is one of the most frequently used means of trans-
port and consequently motorcyclists constitute a large proportion
of road crash victims (Mohan 2002)
The lack of experience of many motorcycle riders has been cited
as one potential cause of motorcycle crashes (Winn 1987 Mullin1997) Training of motorcycle riders has therefore been suggested
as an important countermeasure for reducing both the number of
crashes and the severity of injury (Vis 1995 Noordzij 2001)
Pre-licence motorcycle rider trainingcourses have been made com-
pulsory in several countries (Mullin 1997) Pre-licence training
courses are aimed at providing motorcycle riders with the skills
and knowledge needed to improve control of the motorcycle and
to identify and avoid critical situations in traffic (Elliot 2003)
Many pre-licence motorcycle rider training courses are aimed at
novice riders and in many jurisdictions in high-income countries
such as Australia the USA and Canada the completion of such
courses is a mandatory requirement of licensing
Post-licence rider training is not linked with licensing and such
training is generally not compulsory It usually focuses on riding
skills and other topics that have not been covered or dealt with
sufficiently in pre-licence rider training Post-licence rider train-
ing is used to reinforce fundamental skills and knowledge or to
provide some training for motorcyclists who are self-taught and
have never received formal training (Anderson 1980)
Estimates of the number of motorcyclists attending post-licence
rider training are not readily available but existing data suggest
such courses are not well attended although attendance varies by
country In the USA 24 of the total number of motorcyclists
trained in 1988 and 1989 attended an experienced rider course
(Weaver 1990)
The aim of this systematic review is to quantify the effectiveness
of pre-and post-licence rider training for motorcycle riders in re-
ducing traffic offences road traffic crash involvement injury and
death
O B J E C T I V E S
To quantify the effectiveness of pre- and post-licence motorcycle
rider training on the reduction of traffic offences traffic crash
involvement injuries and deaths of motorcycle riders
M E T H O D S
Criteria for considering studies for this review
Types of studies
All relevant intervention studies such as randomised and non-ran-
domised controlled trials and interrupted time series studies and
observational studies including cohort and case-control studies
Types of participants
Motorcyclists and riders of mopeds or motor scooters of all ages
Types of interventions
Motorcycle ridertraining (pre-or post-licence mandatory or non-
mandatory) versus no training
One form of pre-licence rider training versus another form (eg
on-road off-road theory or a combination)
One form of post-licence rider training versus another form (eg
on-road off-road theory or a combination)
Types of outcome measures
Offences crashes injuries and death
Search methods for identification of studies
We did not restrict the searches by date language or publication
status
Electronic searches
We searched the following electronic databases
bull Cochrane Injuries Group Specialised Register (searched 22
Sept 2008)
bull CENTRAL (The Cochrane Library 2008 Issue 3)bull TRANSPORT 1988 to 200706 (latest date available)
bull ATRI Australian Transport Index (to Sept 2008)
bull MEDLINE (Ovid SP) 1950 to Sept 2008
bull EMBASE (Ovid SP) 1980 to Sept 2008
bull CINAHL (EBSCO) 1982 to Sept 2008
bull WHOLIS (World Health Organization Library
Information System) (to Sept 2008)
bull PsycINFO (Ovid SP) 1806 to Sept 2008
bull LILACS (BIREME) (Latin American and Caribbean
Health Sciences) (1982 to Sept 2008)
bull ISI Web of Science Social Sciences Citation Index (SSCI)
1970 to Sept 2008
3Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 650
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 750
Table 1 Risk of selection bias (Continued)
Osga 1980 Differences at baseline reported on sex and age
Confounding factors were not dealt with in univariate analysis Riding exposure was accounted for in the
cumulative survival analysis
Selection intervention group were all training participants on the register and the control group were
respondents to a diverse range of recruitment strategies
Satten 1980 Differences at baseline reported on age sex helmet use riding experience and exposure and previous
training
Confounding factors were not dealt with in design or analysis
Selection intervention group was selected from a list of training graduates and the control group was
partly selected from a state register of motorcyclists
Jonah 1982 Differences at baseline reported on age sex riding experience time licensed previous training and alcohol
use
Confounding factors were dealt with through multivariate analysis
Selection intervention group was selected from a list of training graduates and the control group was
randomly selected from the Ontario motorcycle operator file
Mortimer 1984 Differences reported at baseline on age licence status riding experience years licensed and helmet use
Confounding factors age years licensed and rider experience were dealt with through stratification and
riding exposure through standardisation
Selection intervention group was selected from a list of training enrolees and the control group was
selected from motorcycle dealerships during three successive Saturdays
Cooper 1988 Differences reported at baseline on age gender licensure and highway experience
Confounding factors age gender riding exposure rider characteristics rider attitudes and behaviour were
accounted for in a multivariate analysis
Selection there was no information on how the intervention and control groups were selected and from
where
Mortimer 1988 Differences reported at baseline on years licensed riding experience and helmet use
Confounding factors age years licensed and rider experience were dealt with through stratification and
standardisation of riding exposure A regression analysis was also conducted but poorly reported and not
usable
Selection intervention group was selected from a list of training enrolees and the control group was
selected from motorcycle dealerships
Adams 1985 Differences at baseline reported on previous crash rates
Confounding factors not controlled for in analysis
Selection the intervention group was a non-random sample of those offered training within the Australia
Post motorcycle delivery riders the control group were the remaining Post riders
Leung 1987 Differences at baseline reported on age and sex
Confounding factors age and sex controlled for in analysis by stratification
Selection those exposed to training were MRC graduates those unexposed were taken from the licensing
records
5Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 850
Table 1 Risk of selection bias (Continued)
McDavid 1989 Differences at baseline other than matched characteristics were not reported
Confounding factors age ownership of car licence years licensed previous crashesand offences accounted
for through matching Other confounding effects not controlled
Selection intervention group was selected from a list of training graduates and the control group was
selected from Bureau of Motor Vehicle records
Kloeden 1994 Differences at baseline reported were in geographical area of residence and in riding exposure
Confounding factors were accounted for through exclusion of participants from rural areas
Other confounding factors were not reported
Selection the Ridersafe course was phased in based on postcode of riderrsquos home address analyses were
based on a database of those who obtained motorcycle rider learnerrsquos permits of which about 2000 were
Ridersafe riders and about 5000 were control riders
Billheimer 1998 Differences at baseline reported on age and previous riding experience
Other potential confounding factors assessed were riding exposure primary purpose of riding and helmet
use
Confounding factors accounted for by matching (age sex primary purpose of riding) standardisation
(riding exposure) and stratification (previous riding experience)
Selection the intervention group were those who had taken the rider training programme the control
group were California motorcycle riders matched on the above variables but who had not taken a riding
course
Perrino 2002 Differences at baseline reported were a higher number of females in the trained group and a higher
proportion of experienced riders in the untrained group
Confounding factors were not accounted for in the analysis
Selection the intervention group were those who had done the Motorcycle Riding Course the control
group had not done the course Both groups were identified through a register of licensed motorcyclists
Kraus 1975 Population based case-control study
Cases all motorcycle riders with a confirmed medically treated motorcycle injury (including fatalities)
from all hospitals in study base and peripheral area hospitals and all deaths due to motorcycle crashes in
the study base All serious injuries (leading to hospitalisation) and fatalities likely to be included
Community controls a random sample of registered motorcycle owners in the study base
Potential confounders assessed were age sex weight height riding experience motorcycle use charac-
teristics of motorcycle and previous crashes and offences
Confounding factors were accounted for through exclusion of females and stratification by age and
motorcycle useSelection cases and controls drawn from the same population
Hurt 1981 Population based case-control study
Cases and controls not drawn from the same population
Cases motorcycle riders who had a motorcycle crash Not all reported cases included
Community controls motorcycle riders who passed the crash site at the same time of day and day of week
that the crash occurred but 2 years later
Potential confounders assessed were age sex time type of motorcycle roadway alignment helmet use
weather conditions road surface conditions
Confounding factors were accounted for through matching for time of day and day of week that crash
occurred and stratification
Selection cases were recruited from police reports not all cases were included controls were matched
6Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 950
Table 1 Risk of selection bias (Continued)
community controls from randomly selected sites matched for same time of dayday of week as the crash
occurred
Haworth 1997 Population based case-control study
Cases and controls drawn from the same population
Cases motorcycle riders or pillion riders with an injury from a motorcycle crash from hospital and police
records Not clear whether all cases were included
Community controls motorcycle rider or pillion who passed the crash site at the same time of day and
day of week that the crash occurred
Potential confounders assessed were rider factors age marital status licence status riding experience
and exposure alcohol and drugs use of protective gear including helmet use pillion factors motorcycle
factors trip factorsConfounding factors were accounted for through matching for time of day and day of week that crash
occurred and stratification for age blood alcohol level and licence status
Selectioncases were riders or pillions involved in on-road crash recruited from hospitals and police records
not sure if all cases were included controls were matched community controls who passed the crash site
at the same time of dayday of week as the crash
Savolainen 2007 Differences at baseline reported on sex safety helmet use riding ability riding experience and licence
status
Adjusted for these confounders through statistical modelling although exact process unclear
Selection intervention group from list of trainees control group from Indiana Bureau of Motor Vehicles
database Trainees and the control group were invited to participate in surveys
Davis 1997 Differences at baseline not reported but likely No adjustment for any confounders at baseline or during analysis
Selection intervention group from list of riders involved in crashes who were also on a list of training
graduates control group from the same list of riders involved in crashes who were not also on a list of
training graduates
Waller 1992 Differences at baseline not reported but likely
Adjusted for confounders through exclusion of riders younger than 18 years riders without a motorcycle
operator driver license riders with location of residence outside of the Texas metropolitan areas of Austin
through matching by age sex and location of residence through stratification by age and sex and through
standardisation (number of crashes by person-days at risk)
Selection intervention group from the Motorcycle Operator Training class roster (number of trained
riders that the sample was selected from is unclear) control group through a random number generator
generating driver license numbers until an equal sample size was selected as the intervention group
MAIDS 2004 Population based case-control study
Unclear whether cases and controls not drawn from the same population
Cases motorcycle riders who had a motorcycle crash Unclear whether all cases included
Community controls motorcycle riders recruited at petrol stations Unclear whether and how matched
to cases
No potential confounders were reported
There was no report of whether and how potentially confounding factors were accounted for in the study
7Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 1050
Table 2 Risk of allocation bias
Anderson 1980 Allocation concealment unclear
Hall 1988 Allocation concealment unclear
Swaddiwudhipong 1998 Allocation concealment unclear
Table 3 Risk of detection bias
Anderson 1980 Blinding of outcome assessors not reported
Driver records holding police recorded crashes and offences not validated
Hall 1988 Blinding of outcome assessors not reported
Driver records holding police recorded crashes and offences not validated
Swaddiwudhipong 1998 Blinding of outcome assessors not reported
Police and hospital records Post mortem examination for deaths not validated
Adams 1985 Blinding of outcome assessors not reported
Australia Post crash records not validated but argued that underreporting is unlikely
Leung 1987 Blinding of outcome assessors not reported
Driver records holding police recorded crashes and offences
Raymond 1979 Blinding of outcome assessors not reported
Self-report through personal interviews or questionnaires - both methods validated
Osga 1980 Blinding of outcome assessors not reported
Self-report through structured questionnaires validated by cross-checking with police
recorded crash records
Satten 1980 Blinding of outcome assessors not reported
Self-report through structured interviews or questionnaires - both methods were not
validated
Intervention group were given telephone interviews but the control group mailed ques-
tionnaires
Jonah 1982 Blinding of outcome assessors not reported
Self-report through structured interviews not validated Driver records holding police
recorded crashes and offences
Mortimer 1984 Blinding of outcome assessors not reported
Self-report through questionnaires not validated
Intervention group received mailed questionnaires but the control group received ques-
tionnaires personally at the motorcycle dealership
Cooper 1988 Blinding of outcome assessors not reported
Driver records holding police recorded crashes and offences
8Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 1150
Table 3 Risk of detection bias (Continued)
Mortimer 1988 Blinding of outcome assessors not reported
Self-report through questionnaires not validated
Intervention group received mailed questionnaires but the control group received ques-
tionnaires personally at the motorcycle dealership
McDavid 1989 Blinding of outcome assessors not reported
Driver records holding police recorded crashes and offences
Kloeden 1994 Blinding of outcome assessors not reported
Driver records holding police recorded crashes and offences
Billheimer 1998 Blinding of outcome assessors not reported
Police recorded crashes and offences
Perrino 2002 Blinding of outcome assessors not reported
Self-report through mailed out questionnaires police records not validated
Kraus 1975 Blinding of intervention assessors to case or control status not reported
Intervention measured police records self-report though questionnaires not validated
Length of follow up not specified
Hurt 1981 Blinding of intervention assessors to case or control status not reported
Intervention measured face-to-face in-depth interviews (ie no structured question-
naires) Validity and reliability of data ensured by reported high level of quality controlLength of follow up not specified
Haworth 1997 Blinding of intervention assessors to case or control status not reported
Intervention measured self-report either by personal or by telephone interview for
both cases and controls not validated
Length of follow up in the 2 years prior to the interview or more than 2 years ago
Savolainen 2007 Blinding of outcome assessors not reported
Source of outcome data not reported
Length of follow up up to 5 years
Davis 1997 Blinding of outcome assessors not reported
Source of outcome data police records
Length of follow up 1 year
Waller 1992 Blinding of outcome assessors not reported
Source of outcome data Department of Public Safety driver history file and official
accident reports
Length of follow up 3 years
MAIDS 2004 Blinding of intervention assessors to case or control status not reported
Intervention measured face-to-face in-depth interviews (ie no structured question-
naires) No report on whether and how validity and reliability of data was ensured
Length of follow up 2 years
9Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 1250
Table 4 Risk of attrition bias
Anderson 1980 Percentage of participants in analysis 99 forthe total group Unclear forintervention andcontrol groups
separately
Hall 1988 Percentage of participants in analysis 96-97 forall intervention groups and 98for the control group
Swaddiwudhipong 1998 Percentage of participating sub-districts in analysis 100 for both intervention and control sub-districts
Raymond 1979 Percentage of participants in analysis 74 for the intervention group and less than 57 for the control
group
Osga 1980 Percentage of participants in analysis 73 for the intervention group and 100 for the control group
Satten 1980 Percentage of participants in analysis is unknown Initial sample size is unclear
Jonah 1982 Percentage of participants in analysis 23 for the intervention group and 45 for the control group
Mortimer 1984 Percentage of participants in analysis 35 for the intervention group and 84 for the control group
Cooper 1988 Percentage of participants in analysis 98 for the total group Unclear for intervention and control
groups separately
Mortimer 1988 Percentage of participants in analysis 40 for the intervention group and 90 in the control group
Adams 1985 Percentage of participants in analysis 61 forthe total group Unclear forintervention andcontrol groups
separately
Leung 1987 Percentage of participants in analysis unclear because initial sample size was not reported
McDavid 1989 Percentage of participants in analysis 40 of final matched cohort
Kloeden 1994 Percentage of participants in analysis 39 for the intervention group and 6 for the control group
Billheimer 1998 Percentage of participants in analysis unclear percentage of the initial sample size
Perrino 2002 Percentage of participants in analysis unclear percentage of the initial sample size
Kraus 1975 Percentage of participants in analysis 58 of cases 59 of controls
Hurt 1981 Percentage of participants in analysis 18 of cases lt 29 of controls
Haworth 1997 Percentage of participants in analysis 70 of cases 47 of controls
Savolainen 2007 Percentage of participants in analysis 18 for the intervention group and 15 for the control group
Davis 1997 Percentage of participants in analysis 100 for the intervention group and unclear for the control group
10Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 1350
Table 4 Risk of attrition bias (Continued)
Waller 1992 Percentage of participants in analysis unclear percentage for the intervention group and 100 for the
control group
MAIDS 2004 Percentrage of participants in analysis unclear for both cases and controls
R E S U L T S
Description of studies
See Characteristicsof included studies Characteristics of excluded
studies Characteristics of ongoing studies
Our search resulted in 102 studies that seemed relevant of which
we excluded 79 studies because they did not meet the criteria for
considering studies for this review (see Characteristics of excluded
studies)
Basedon ourcriteriawe includeda total of 23 studies in this review
There were three randomised trials (Anderson 1980 Hall 1988
Swaddiwudhipong 1998) two non-randomised trials (Adams
1985 Leung 1987) 14 cohort studies (Raymond 1979 Osga
1980 Satten 1980 Jonah 1982 Mortimer 1984 Cooper 1988
Mortimer 1988 McDavid 1989 Waller 1992 Kloeden 1994
Davis 1997 Billheimer 1998 Perrino 2002 Savolainen 2007)
and four case-control studies (Kraus 1975 Hurt 1981 Haworth
1997 MAIDS 2004)
Over half of the studies were conducted 20 or more years
ago and the oldest was conducted 31 years ago Only three
studies were from this decade (Perrino 2002 MAIDS 2004
Savolainen 2007) Over 60 of the studies were conducted in
the USA (Kraus 1975 Anderson 1980 Osga 1980 Satten 1980
Hurt 1981 Mortimer 1984 Leung 1987 Hall 1988 Mortimer
1988 Waller 1992 Davis 1997 Billheimer 1998 Perrino 2002
Savolainen2007) three inAustralia (Adams 1985 Haworth 1997
Kloeden 1994) three in Canada (Jonah 1982 Cooper 1988
McDavid 1989) onefromthe United Kingdom (Raymond 1979)one from Europe (with sampling areas located in France Ger-
many Netherlands Spain and Italy) (MAIDS 2004) and one
in Thailand (Swaddiwudhipong 1998) Only seven studies were
published in peer reviewed journals (Kraus 1975 Jonah 1982
Mortimer 1984 Cooper 1988 Mortimer 1988 McDavid 1989
Swaddiwudhipong 1998) The other studies were reports forhigh-
way safety and road traffic organisations
The studies by Anderson 1980 Hall 1988 Kloeden 1994
Billheimer 1998 (the basic California Motorcyclist Safety Program
component of the study) and Perrino 2002 examined manda-
tory pre-licence training Cooper 1988 McDavid 1989 and
Swaddiwudhipong 1998 assessed non-mandatory training that
when passed resulted in a licence Raymond 1979 Osga 1980
Satten 1980 Jonah 1982 Mortimer 1984 Adams 1985 Leung
1987 Mortimer 1988 Waller 1992 Davis 1997 Billheimer 1998
(the Experienced Rider Course component of the study) andSavolainen 2007 assessed non-mandatory training that could be
taken by motorcycle riders with and without a licence The case-
control studies (Kraus 1975 Hurt 1981 Haworth 1997) assessed
lsquoanyrsquo type of rider training and therefore could not be grouped by
pre- or post-licence rider training The MAIDS 2004 case-control
study did make a distinction between participants having received
compulsory pre-licence training or additional (non-compulsory)
training We did not find a study that compared one form of pre-
or post-licence training (eg on-road off-road theory or a com-
bination) versus another form of training All studies compared
eithermandatory or non-mandatory formal training versus no for-
mal training with the exception of Haworth 1997 who compared
different levels of training courses (beginner intermediate and ad-vanced)
We have presented a more detailed description of the individual
studies in Characteristics of included studies We have provided
a description below of the types of motorcycle rider training that
were evaluated in the included studies
Types of motorcycle rider training evaluated in the
studies
The types of training were varied Duration of training ranged
from two or three hours to 24 hours administered over two or
three days Most training consisted of a theoretical and a practical
component usually off-road The content ranged from provid-ing riders with manuals lectures discussions showing films and
administering knowledge and skill tests Little information was
provided about the qualifications of trainers who conducted the
training
Six studies (Osga 1980 Satten 1980 Mortimer 1984 Mortimer
1988 Waller 1992 Savolainen 2007) assessed the effectiveness of
the USA-based Motorcycle Safety Foundationrsquos Motorcycle Rider
Course (MSF MRC) This course was released in 1976 It com-
prises eight hours of theory (lectures discussions films and multi-
plechoice tests) a textbook and 12 hours of off-road practice The
Motorcycle Safety Foundation is a national not-for-profit organ-
11Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 1450
isation sponsored by manufacturers and distributors of motorcy-
cles Many USA states have adopted the MSF Basic Rider Courseand Experienced Rider Course (wwwmsf-usaorg) since its release
in 1976
The Motorcycle Operators Skill Test was also developed
by the USA Motorcycle Safety Foundation in coopera-
tion with others (httpwwwwebbikeworldcommotorcycle-
trainingmotorcycle-test) The MOST is a licensing test consist-
ing of a manual knowledge and skill test (comprising eight exer-
cises of increasing difficulty and skill requirements) and a mini-
skill test conducted with a simulator Some studies compared this
alternative licensing test to the standard licensing test However
as the MOST is not a type of rider training we have included only
the two studies that assessed the MOST in combination with one
or more training components in this review Both Anderson 1980and Hall 1988 assessed the MOST in combination with a train-
ing component of three hoursrsquo duration consisting of instructions
about the skills needed to pass the MOST (30 minutes of the-
ory and two and a half hours on the motorcycle) Hall 1988 also
assessed the MOST in combination with the Motorcycle Safety
Foundationrsquos Motorcycle Rider Course as described above
Other USA studies evaluated the Wisconsin motorcycle rider
course a 22-hour off-street programme that teachesthe basic rules
of safe motorcycling (Leung 1987) the USA Maryland Motorcy-
cle Rider Course (MRC) a mandatory (for those younger than 18
years)pre-licence 20-hour coursecomprising eight hours of theory
and12 hours of practice and created by the Maryland Motor Vehi-
cle Administration in 1985 (Perrino 2002) the California Motor-cyclist Safety Program (CMSP) a mandatory (for those younger
than 18 years) pre-licence 16-hour basic course (Billheimer 1998)
and Experienced Rider Course (ERC) of eight hoursrsquo duration
(Billheimer 1998) The CMSP still offers these courses ( http
wwwca-msporg) but they are now MSF courses as described
above Davis 1997 did not give any information about the assessed
rider training
The Canadian studies evaluated the British Colombia Safety
Councilrsquos 37-hour motorcycle safety programme (no further in-
formation on the content was provided) (McDavid 1989) and
the Canada Safety Council Motorcycle Training Program (MTP)
comprising four hours of theory and 16 hours of practice (six
hours on-road) over twoweekends (Jonah 1982) The MTPcoursewas made available in 1974 and is still being offered in Canada
(httpwwwsafety-councilorgtrainingmtpmtphtm) Cooper
1988 did not give any information about the assessed rider train-
ing However the study was conducted in British Colombia in
Canada in 1988 the same time as the McDavid 1989 study was
conducted
In the United Kingdom Raymond 1979 assessed the non-manda-
tory RACACU (Royal Automobile ClubAuto-Cycle Union)
training scheme The training programme consisted of either 24
one-hour sessions or 13 two-hour sessions Twelve hours of the
training focused on theory about basic operational procedures
traffic laws and highway codes The practical session focused on
basic operational skills riding on a training area and riding in traf-fic with an instructor A proficiency test was administered at the
end of the course
One study from rural Thailand (Swaddiwudhipong 1998) evalu-
ated a community-based rider education programme consisting of
theory on motorcycle injury risk helmet use traffic laws vehicle
regulations and traffic signs and a written and skill test leading to
a licence
TheAustralian studies evaluateda rider trainingcourse of 12 hours
over two consecutive days (four hours of theory and eight hours of
practice)administeredto AustraliaPosttelegram deliverers(Adams
1985) and a mandatory pre-licence Ridersafe course consisting
of two tiers the first tier being a training course on the basics
of motorcycle riding where a pass entitled the rider to obtain alearnerrsquos permit and the second tier being a skill test where a pass
entitled the riderto obtain a probationary motorcycle licence The
training course consisted of two four-hour training sessions on
separate days involving instruction videos and actual riding on
motorcycles The skill test consisted of a four-hour preparation
course of 74 minutes of lecturediscussion 38 minutes of video
and 75 minutes of on-course riding which was the testing phase
(Kloeden 1994) The Rider Safe course is still being provided by
Transport South Australia a state government organisation (http
wwwtransportsagovau )
The duration or content of rider training in the four case-control
studies (Kraus 1975 Hurt 1981 Haworth 1997 MAIDS 2004)
was not described
Risk of bias in included studies
We describe the methodological quality of the included studies
by study type below More detailed information can be found in
Table 1 Table 2 Table 3 and Table 4
When considering the findings from these studies it is important
to note the quality (or lack thereof) of reporting of the studies
Quality of reporting is the extent to which a report provides in-
formation about the design conduct and analysis of a study ( Juumlni2001) Huwiler-Muumlntener 2002 found that even though trials
may be well reported it does not mean that the methodological
quality is of equal standard and well-conducted trials may never-
theless be reported badly The reporting quality of most studies in
this review was poor Most studies were old and did not meet con-
temporary reporting standards This made data extraction com-
plicated It also made it difficult to contact authors for additional
information
Because of significant heterogeneity between studies as well as the
poor quality of most studiespooling of studyresults to calculatean
overall measure of effect was not appropriate Instead we provided
a description of our judgement of risk of bias in the text
12Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 1550
Risk of bias in the randomised trials
There were three randomised trials (Anderson 1980 Hall 1988Swaddiwudhipong 1998) Anderson 1980 and Hall 1988 had an
initial sample size of 41188 and 26464 motorcycle licence ap-
plicants respectively Both studies assessed the effectiveness of the
pre-licence MOST with a trainingcomponent Swaddiwudhipong
1998 evaluated the effectiveness of a community-based rider ed-
ucation programme in six districts
Allocation bias
Anderson 1980 reported using random birth date tables to allocate
participants to either the intervention or control group Hall 1988
reported the use of a computerised algorithm set in a central of-
fice but also used manual random assignment when the computer
went down Swaddiwudhipong 1998 reported lsquosimple random
samplingrsquo for the delivery of the intervention where the sampling
frame included nine districts Three districts were excluded since
they were too hard to access The remaining six districts were ran-
domly allocated to the intervention (three) and the control (three)
groups Systematic sampling wasthen used ordered by village size
for the villages within districts Allocation concealment was not
reported in any of the studies The descriptions of randomisation
in these studies suggest that there may have been limitations to
the process resulting in a lack of equivalence of the intervention
and control groups at baseline and therefore the possibility of
allocation bias Swaddiwudhipong 1998 reported that before theintervention the annual incidence rate of motorcycle-related in-
juries was slightly (but not statistically significant) higher in the
intervention areas than in the control areas Participants in the
intervention districts were also more likely to have a valid licence
and to wear a helmet more often than participants in the control
districts Anderson 1980 measured differences at baseline on the
participant and prior driving record variables and found these to
be negligible Hall 1988 did not report on baseline differences
between the intervention and control groups As the Anderson
1980 and Hall 1988 studies had sample sizes of over 40000 and
26000 respectively and used some form of randomisation it is
likely that any confounding factors at baseline would have been
equally distributed between comparison and intervention groups
Selection bias
After randomisation Anderson 1980 and Hall 1988 excluded
08 and 3 participants respectively because of assignment to
the wrong group administration of the wrong intervention or
other reasons The resulting lack of analysis by rsquointention-to-treatrsquo
can introduce biasby removing the effectivenessof randomisation
Nonetheless the percentage excluded was small and the number
of overall participants large so it is unlikely to have had a major
effect on the final results In the Swaddiwudhipong 1998 study
villages were systematically sampled into intervention or control
groups The intervention was implemented and outcomes of in-
jury and death were measured at a community level Interventionand control districts were randomly selected however there were
only three intervention and three control communities
Detection bias
None of the randomised trials reported blinding of outcome as-
sessors Data collection tools used by Anderson 1980 and Hall
1988 were driver records holding police recorded crashes and of-
fences Swaddiwudhipong 1998 used police and hospital records
at a district level and post-mortem examination for deaths None
reported on validity or reliability for any of the instruments
Attrition bias
Anderson 1980 and Hall 1988 had driver records available for
96 to 100 of participants They also collected self-reported
data but the response rate to the questionnaires was very low (low-
est response rate 42 and 30 respectively) Attrition bias in
Swaddiwudhipong 1998 was not an issue as outcome data were
available for all sub-districts included in the trial
The intervention reported in Anderson 1980 and Hall 1988 tri-
als had an unanticipated effect on the participants In both trials
an average of only 28 of licence applicants in the intervention
group eventually obtained their licence compared to 35 in the
control group This indicates that the licensing process in generalseems to have deterred applicants from obtaining a licence and
although no statistical tests of the difference were conducted this
deterrence appears even greater for those who had rider training
This is not true attrition bias as the studies did not experience
differential loss to follow up from outcome measurement Instead
differential attainment of the intermediary step of licensure pre-
cluded participants from being at risk for the outcomes of interest
they are not driving a motorcycle hence they are not at risk for
motorcycle-related crashes injury death or offences In short it is
not possible to assess whether a reduction in crashes in the inter-
vention group would be due to changes in riding habits as a result
of training or simply due to delayed or never attained licensure
Intervention integrity
Neither Anderson 1980 nor Hall 1988 reported on the consis-
tency in the way the intervention was administered the possible
lsquocontaminationrsquo of participants in the intervention group by other
types of formal or informal rider training or of participants in the
control group by the intervention Swaddiwudhipong 1998 did
not report on the number of villagers in each group that actually
attended the intervention However they did report on a number
of other potential training courses that were being implemented
at the same time
13Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 1650
Analysis
Anderson 1980 and Hall 1988 acknowledging the lsquodeterrent ef-fectrsquo described above conducted two separate analyses The first
analysis included the total study population of licence applicants
over two years from the time of their application (the population
denominator) The second analysis was performed on the licensed
(intervention and control) subgroups only two years from the
time of obtaining their licence (the licensed rider denominator)
As only a third were licensed and exposed to riding the motorcy-
cle it was impossible to determine whether reported differences
in crash rates across groups were due to the intervention or to
other lsquoconfoundingrsquo characteristics of those who were determined
in their efforts to obtain the licence When analysis was restricted
only to those who gained their licence the same difficulties re-
mained The benefits of the randomisation process were thus lost
due to this lack of licensure of riders during the process As there
were only three intervention and three control groups in the study
reported by Swaddiwudhipong 1998 analyses were hindered by
smallnumbers for outcome comparisons betweenthe intervention
and the control districts
Risk of bias in the cohort studies and non-randomised
trials
We discuss the methodological quality of cohort studies (seven
prospective studies (Raymond 1979 McDavid 1989 Waller 1992
Kloeden 1994 Billheimer 1998 Perrino 2002 Savolainen 2007)and seven retrospective studies (Osga 1980 Satten 1980 Jonah
1982 Mortimer 1984 Cooper 1988 Mortimer 1988 Davis
1997)) and non-randomised trials (Adams 1985 Leung 1987) in
the one section as the biases for the two study types are similar
The quality of these studies was in general poor
Selection bias
The confounding factors expected to have been addressed in the
studies were sex age riding experience prior driving records (of-
fences crashes) previous training and the following proxies for
risk taking behaviour helmet use speed and alcoholdrug use All
of these factors have been shown to be associated with an increasedrisk of motorcycle crash (Peden 2004 Zambon 2006) The pos-
sible source of bias in cohort studies and non-randomised trials
is that groups may differ on important characteristics at baseline
participants lsquoself-selectrsquo into either the rider training group by de-
ciding to undertake a rider training course or into the lsquocontrol
grouprsquo by deciding not to undertake a course Participants who
undertake a non-mandatory rider training course may differ from
those who choose not to on characteristics such as age sex rid-
ing experience and riding behaviour as well as other possible con-
founders For example riders who take a training course may be
more safety-conscious or on the contrary may do so because they
have previously experienced a crash or because they perceive that
they have poor riding skills Therefore ideally all participants in
the intervention group need to be included in the study as wellas a random sample of controls taken from the same population
as the intervention group If that is not possible intervention and
control groups either need to be carefully matched to make them
comparable or differences at baseline need to be adjusted for in
the analysis
Most of the studies using a non-randomised trial or cohort design
compared an intervention group to a control group that differed
on a number of important factors The studies in general failed to
adjust for important differences found at baseline and their results
cannot be meaningfully interpreted Ten studies (Raymond 1979
Osga 1980 Satten 1980 Jonah 1982 Mortimer 1984 Mortimer
1988 McDavid 1989 Waller 1992 Davis 1997 Savolainen 2007)
assessed intervention and control groups that were not drawnfrom the same population In most of these cases the intervention
group was selected from a list of training graduates who were usu-
ally young female and less experienced riders whereas controls
were recruited from a general population of motorcycle riders who
were usually older male and experienced riders For three stud-
ies it was unclear how they were selected (Leung 1987 Cooper
1988 Perrino 2002) Only Kloeden 1994 recruited groups from
the same population of motorcycle applicants and Adams 1985
from a group of professional riders Nevertheless due to non-ran-
dom selection of intervention and control groups differences at
baseline are likely even though study groups were from the same
population Three studies adjusted for some confounding fac-
tors found at baseline through matching (McDavid 1989 Waller1992 Billheimer 1998) However matching has limitations in
termsof controlling for multiple confounders especially those that
are unknown to influence the effect of rider training on the out-
comes One study matched the control and intervention groups
by excluding controls (approximately 3500 (66 of initial con-
trol group) that differed from the intervention group) (Kloeden
1994) Five studies (Osga 1980 Satten 1980 Adams 1985 Davis
1997 Perrino 2002) did not adjust for differences between study
groups at baseline either in the study design or analysis Results
from these studies are therefore not reliable The other six stud-
ies (Raymond 1979 Jonah 1982 Mortimer 1984 Cooper 1988
Mortimer 1988 Leung 1987 McDavid 1989 Billheimer 1998)
adjusted for some differences between groups at baseline by eitherbi-variate or multi-variate analyses or stratification and standardi-
sation of riding exposure (outcomemiles driven) or stratification
by age group The limitation of stratification is that it is difficult to
control for many variables simultaneously because a large number
of strata will need to be generated resulting in small numbers that
cannot be meaningfully analysed Savolainen 2007 developed sta-
tistical models to assess the probability of a crash after taking the
rider training course in an intervention and control group while
examining differences by sub-groups (eg age) Participation rates
were also very low in this study with only 5584000 (1395)
in the eligible intervention group and 5884000 (147) for the
14Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 1750
control group completing surveys
Detection bias
None of the studies reported blinding of outcome assessors to the
intervention or control status of the participants Seven studies
used driver records containing police recorded data as their main
measure of outcome (Leung 1987 Cooper 1988 McDavid 1989
Waller 1992 Kloeden 1994 Davis 1997 Billheimer 1998) An-
other four studies (Raymond 1979 Satten 1980 Mortimer 1984
Mortimer 1988) only used self-report Raymond 1979 was the
only study using self-report as the main outcome measurement in
which study instruments were validated Self-report might prevent
the underreporting that has been described for the use of police-
driver records (see rsquoRisk of biasrsquo in the randomised trials above)but has other risks of bias especially if questionnaires or inter-
view methods have not been validated and tested for reliability
Recall bias arises from mistakes in recollecting events both be-
cause of failures of memory and because of looking at things lsquowith
hindsightrsquo and possibly changed views (httpwwwcochraneorg
resourcesglossaryhtm) As many studies were retrospective risk
of recall bias is likely and may have increased with length of follow
up For studies that used self-reported outcomes only it is also
possible that social desirability response bias played a role where
participants in the training group may have felt it was lsquoexpectedrsquo
to report good driving histories Multiple methods to measure the
same outcome for both groups decreases the risk of detection bias
for example because self-reported answers can be cross checkedwith other sources of information such as driver records
In this regard three studies (Osga 1980 Jonah 1982 Perrino
2002) used both self-report and driver records Osga 1980 found
that for both intervention and control groups more crashes were
reported through self-report than on state crash records an aver-
age of 54 of self-reported crashes could be found back on state
crash records One industry-based study (Adams 1985) used oc-
cupational health records which were not validated but which are
unlikely to suffer from underreporting due to work cover claims
Savolainen 2007 used information on rider crashes for a five-year
period but did not specify from where this information was ob-
tained
Attrition bias
The majority of studies (Raymond 1979 Satten 1980 Adams
1985 Leung 1987 Kloeden 1994 Davis 1997 Billheimer 1998
Perrino 2002) inadequately reported on losses of participants to
the study We can therefore not be certain that there were no sys-
tematic differences in the characteristics between the intervention
and control group participants Savolainen 2007 reported the re-
sponse rates to their baseline survey but did not report on the per-
centage of outcome data that was available for their study sample
A similar deterrent effect of rider training as described for the ran-
domised trials (Anderson 1980 Hall 1988) occurred in two stud-
ies Jonah 1982 reported that 65(14992310) of the trained rid-
ers stopped riding compared to 47 (9492029) of the untrainedriders They excluded non-active and non-licensed riders before
the study commenced so the differential choice to continue riding
after training did not affect their analyses Leung 1987 reported
that the percentage of participants in the intervention group ob-
taining their licence was 50 (no number provided by authors)
lower than in the control group This may have biased the mea-
surement of outcomes because unlicensed participants will not be
exposed to riding the motorcycle and therefore not be at risk of
being involved in a crash or violation As explained above this
makes it difficult to assess whether the reduction in crashes in the
intervention group is due to a change in driving habits or simply
due to delayed licensure
Mortimer 1984 and Mortimer 1988 describe that about 44 and30 of course graduates respectively were excluded from analysis
because they had not ridden a motorcycle in the previous year
However because these studies were retrospective and the controls
were activeriders at the time of recruitment a decision on whether
the deterrent effect was differential between the two groups could
not be made Though not reported quantitatively Kloeden 1994
also describes the effectof rider training deterringindividuals from
becoming licensed and hence exposure to riding and risk of crash
death or offences involving a motorcycle Osga 1980 reported
that 27 of the trained riders did not ride after the course and
excluded these from the analysis
Use of different data collection strategies may have contributed to
differential loss to follow up in the cohort studies Satten 1980for example interviewed the intervention group over the tele-
phone but mailed questionnaires to the control group Mortimer
1984 and Mortimer 1988 distributed questionnaires in person to
the control group but mailed them to the intervention group
This may have played a role in the observed differential response
rates where more subjects responded in the personally approached
groups
Survival bias
For those studies that assessed offences crashes and injuries but
not death survival bias may have also played a role Motorcyclists
who were fatally or seriously injured in crashes during the follow-
up period after the intervention would have been less likely or
unable to respond to surveys This could bias the effect measures
towards or away from the null None of the studies wrote about
this issue or examined the characteristics of those lost to follow up
in order to address the possibility of survival bias
Intervention integrity
Although the interventions were intended for the entire lsquoexposedrsquo
groups most studies reported difficulty in tracking participants
whoreceivedthe interventionand those whodid not Adams 1985
15Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 1850
estimatedthat 62of thetotalstudysample received the interven-
tion The remaining studies did not report the percentage of thestudy population that received the intervention as was intended
or reported that this percentage was unknown (Raymond 1979
Kloeden 1994 Satten 1980 Jonah 1982 Mortimer 1984 Leung
1987 Cooper 1988 Mortimer 1988 Billheimer 1998 Perrino
2002 Savolainen 2007) Only Raymond 1979 reported the degree
to which their intervention was delivered as intended No studies
reported on possible lsquocontaminationrsquo of participants by other inter-
ventions than the one to which they were allocated Because Osga
1980 Satten 1980 Jonah 1982 Mortimer 1984 Cooper 1988
Mortimer 1988 Waller 1992 and Davis 1997 were retrospective
studies and participants were recruited from a list of graduates it
was known that they had completed and passed the course
Analysis
In some studies some participants who reported at baseline that
they were untrained decided to enroll in a rider training pro-
gramme after the baseline survey The non-randomised trial by
Adams 1985 experienced this move of subjects from the untrained
to the trained group but failed to conduct analyses by lsquointention-
to-treatrsquo This can bias the results in randomised and non-ran-
domised trials if participants who changed groups differ signifi-
cantly with regard to potentially confounding characteristics On
the other hand Perrino 2002 also experienced this move of sub-
jects from theuntrained to the trained groupmore than 100riders(5 of the total sample) originally stated that they were untrained
at baseline but then received training between the baseline and
follow-up surveys Yet since the Perrino 2002 study was a cohort
their analyses then proceeded by including those riders who were
lsquonewlyrsquo trained as exposed Riders who were lsquonewlyrsquo trained since
baseline but with a programme other than the one under study
were excluded from the analysis Nevertheless confounding fac-
tors were not accounted for in the analyses
McDavid 1989 matched exposed with unexposed subjects but
could not find suitable matches for the majority (60) of their
subjects They could therefore only conduct analysis on the result-
ing40 (139346) which is an inappropriate andinefficient anal-
ysis Some studies that reported on the issue of non-active licensedriders(Satten 1980 Jonah 1982 Mortimer 1984 Mortimer 1988
Kloeden 1994) excluded non-active riders from analysis intro-
ducing selection bias as it is likely that riders who continued to
ride differed with regards to potentially confounding character-
istics across intervention and control groups Kloeden 1994 for
example conducted several analyses to assess the effect of various
sub-sets of the control group Original analyses included all con-
trols and led to erroneous results since only a small proportion of
controls hadgone on to get a licence after their learnerrsquos permitand
were actually exposed to the possibility of experiencing a crash
Subsequent analyses compared only intervention and control sub-
jects who had a licence Waller 1992 reported crashes by person-
days of exposure but did not explain how they could be sure that
each study intervention and control participant was an active riderand actually exposed to the risk of experiencing a crash
Potential confounders that are unequally distributed by study arm
at baseline need to be dealt with in the analysis Satten 1980
Adams 1985 and Davis 1997 did not take into account any con-
founding factors either by design or in the analysis McDavid
1989 Waller 1992 and Billheimer 1998 accounted for potential
confounding factors at the design stage by matching the subjects
in the intervention and control groups Other studies (Raymond
1979 Mortimer 1984 Mortimer 1988 Perrino 2002 and also
Waller 1992) adjusted for one or two confounders by stratifica-
tionor standardisation (eg crashesmiles driven age group) Osga
1980 conducteda univariate analysis of crash rates in which no po-
tential confounders were accounted for but also calculated cumu-lative crash rates by mileage travelled Only Jonah 1982 Cooper
1988 and Savolainen 2007 conducted a multivariate analysis
Risk of bias in the case-control studies
A case-control design is suitable for the assessment of risk fac-
tors for outcomes such as motorcycle crashes injuries or deaths
but not appropriate for the evaluation of an intervention such
as motorcycle rider training It can potentially result in spurious
conclusions regarding the efficacy of the intervention because the
intervention has not been applied at random to an intervention
and control group Even with the control of possible confounders
using a case-control design to evaluate an intervention can lead tothese kinds of results since those who choose to undertake rider
training may differ on important prognostic factors that affect the
outcome compared to those who do not choose to participate in
rider training
Selection bias
A potential source of bias in case-control studies as with all other
study types where the intervention is not randomly allocated to
participants is that comparison groups are likely to differ on po-
tentially confounding characteristics This means that ideally all
cases need to be included in the study as well as a random sample
of controls taken from the population that gave rise to the cases If that is not possible cases and controls either need to be carefully
matched to make themcomparable or differences at baseline need
to be adjusted for in the analysis All four case-control studies were
population based Kraus 1975 included all fatalities and injured
cases leading to hospitalisation and appropriately recruited a ran-
dom sample of controls from the same population Hurt 1981 did
not manage to recruit all or a random sample of cases Only half
of the notifications of crashes to the police were received in time
by the researchers the research team for unreported reasons was
only able to respond to half of those again Data collection was
therefore completed for only about a fifth of all cases Neverthe-
less this was still a large sample of 900 cases Haworth 1997 using
16Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 1950
a similar study design was more likely to have included all cases
but this is not entirely clear from the study reported It is unclearfor example whether fatalities were included as they did not use
coronerrsquos reports for the recruitment of cases The MAIDS 2004
study was conducted in five European countries but it was unclear
how cases were recruited whether they were all cases during the
two years that the study lasted or a random sample of all cases
what the response rates were etc Controls were recruited from
randomly selected petrol stations but it was likewise not clear what
the eligibility criteria for inclusion in the study were how they
were recruited (eg randomly) and what the response rate was
Hurt 1981 and Haworth 1997 both matched controls to cases by
recruiting themat the sametime ofdayandsamedayof weekas the
crash occurred Hurt 1981 however recruited controls two years
later andit is likelythat there would have beendifferences betweenthe two comparison groups One important difference was that
during data collection for cases there was very little specialised
motorcycletrainingavailable in the studyregionwhereas two years
later when control data were collected there was some training
available
These four case-control studies also suffered from selection bias
due to potential differences between those that gave consent to
participate in the study and responded to interviews andquestion-
naires and those that did not None of the studies have assessed
such differences although Hurt 1981 did acknowledge that con-
trols giving consent to cooperate may have differed from those
motorcyclists who did not
Detection bias
Measures that can be taken to reduce detection bias in case-con-
trol studies are blinding of interviewers for case or control status
blinding of the participant for the study factor of interest validat-
ing measurement instruments and using these in an identical way
for cases and controls and cross checking self-reported informa-
tion with other more objective sources such as registers None of
the studies reported blinding of the intervention assessors to case
or control status but all used consistent measurement methods for
both case and control groups
All case-control studies used self-report as their main measure toassess whether cases and controls had received rider training and
if so the type of training The major potential source of bias in
case-control studies when self-report is used as a measure is recall
bias One way of reducing recall bias is by double-checking self-
reported data with more objective data Kraus 1975 also used po-
lice accident reports for information on rider characteristics but
such information was not available about rider training Haworth
1997 and MAIDS 2004 did not report on any measures to limit
recall bias Hurt 1981 reported further questioning of cases by
the interviewer when answers did not make sense or when par-
ticipants seemed to contradict themselves as part of the in-depth
unstructured interview technique This qualitative research tech-
nique can result in valuable information but as interviewers were
not blinded to case or control status of the participants it may have led to more thorough examination of cases compared to the
controls and thus differential detection bias
Intervention integrity
The level of detail about the rider training provided in the case-
control studies differed significantly but was minimal for each
Kraus 1975 simply asked whether lsquoanyrsquo motorcycle rider training
had lsquoeverrsquo been carried out Hurt 1981 asked participants whether
they hadbeen self taught informally taught or completed formally
taught rider training courses MAIDS 2004 asked whether partic-
ipants had had pre-license training additional (non-compulsory)
training other training or no training whereas Haworth 1997asked about the number of courses that a participant had com-
pleted whetherthey were basic intermediate or advanced courses
and how long ago the training had been done None of the case-
control studies reported collecting information about further de-
tails about any training programmes (eg type of training cur-
riculum duration of training)
Analysis
Kraus 1975 accounted for confounding factors through the exclu-
sion of females and stratification by age and frequency of motor-
cycle use Hurt 1981 matched cases and controls for time of day
and day of the week that the crash occurred and did not adjust forany confounding factors in the analysis Haworth 1997 matched
groups for time of day and day of the week that the crash occurred
and stratified by age blood alcohol level and licence status in the
analysis None of the studies used multivariate methods MAIDS
2004 did not provide any information about any participant char-
acteristics other than the study exposures of interest and did not
describe any methods to adjust for potential confounding factors
Effects of interventions
We have presented the results in a descriptive manner due to the
substantial heterogeneity in study types interventions and out-comes and the generally poor quality of the studies We have
grouped the interventions by mandatory and non-mandatory pre-
licence rider training and post-licence rider training and further
by study type and by outcome (crashes injurious and fatal crashes
and offences) We discuss results that are reported as statistically
significant (whether a P value was reported or not)
Pre-licence rider training - mandatory
Randomised trials
17Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2050
Crashes
Total group of participants
Anderson 1980 found a reduction in motorcycle crash rates in
the intervention group (MOST skill test with remedial training)
compared to the control group (who received the skill testwithout
a training component) The percent differenceranged from 143
six months after the programme (P lt 005) to 211 (P lt 0001)
after one year and 127 (P lt 001) after two years Hall 1988
found no statistically significant difference in crash rates between
intervention groups at any point in time
Subgroup of participants who obtained their licence only
Anderson 1980 found a reduction in motorcycle crash rates in the
intervention arm but this was statistically significant at one year
only (24 to 218 reduction P lt 005) Hall 1988 found no
statistically significant differences
Offences
Total group of participants
The percentage of offences in Anderson 1980 after six months
and one and two years was slightly higher in the intervention
arm compared to the control group but these differences were
not statistically significant Likewise the study by Hall 1988 did
not find a statistically significant difference in offences between
groups
Subgroup of participants who obtained their licence only
Hall 1988 found no significant difference in offence rates between
the intervention and control groups Anderson 1980 found thatthe percentage of offences increased in the intervention arms com-
pared to the control group This difference was statistically signif-
icant after two years (51 to 115 increase P lt 005)
Cohort studies
Crashes
Neitherof the cohort studies (Kloeden1994 and Billheimer1998)
found a statistically significant difference in crash rates between
the intervention and control groups
Offences
Billheimer 1998 (for the California Motorcyclist Safety Program
course) and Perrino 2002 did not find a statistically significant
difference in offence rates between the intervention and control
groups
Case-control studies
Crashes
MAIDS 2004 found that a slightly higher percentage of controls
(not involved in a crash) had had pre-license rider training than
cases (involved in a crash) This difference was not statistically significant (50 of controls vs 477 of cases OR 091 95 CI
076 to 110)
Pre-licence rider training - non-mandatory
Randomised trial
Injuries and deaths
Swaddiwudhipong 1998 found a significantly lower injury ratein the intervention villages (1051000 population 95 CI 92
to 120) compared to the control villages (1691000 population
95 CI 152 to 185) one year after the intervention This differ-
ence was not maintained two years after the intervention (127
95 CI 112 to 143 intervention 151 95 CI 135 to 167
control) There was no significant difference in death rates be-
tween groups
Cohort studies
Crashes
Cooper 1988 foundno statistically significant differencein crashes
between the intervention and control groups McDavid 1989
found a lower crash rate in the intervention group (14139) com-
pared to the control group (23139) but no test of statistical sig-
nificance was undertaken
Injuries and deaths
McDavid 1989 found that the intervention group had fewer in-
jurious crashes (814 crashes) compared to the control group (16
23 crashes) but no statistical significance test was reported
18Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2150
Offences
Cooper 1988 foundno statistically significantdifference in offencerates between the intervention and control groups McDavid 1989
foundno statistically significantdifference in offence ratesbetween
the intervention and control groups
Non-mandatory rider training
Non-randomised trials
Crashes
Neither Adams 1985 nor Leung 1987 found a statistically signif-icant effect of rider training on crash rates
Offences
Leung 1987 found a statistically significant lower offence rate in
the intervention group compared to the control group for both
males and females and for all age groups except for women 35
years or older (P values not reported)
Cohort studies
Crashes
Osga 1980 in the univariate analysis reported a lower percentage
of crashesin theintervention group (that continued to ride) (28
168595) than in the control group (34 112329) In the cu-
mulative crash rate analysis however Osga1980 found statistically
significant (at the P lt 001 level) higher crash rates in the interven-
tion group than in the control group at allmileage intervals (1000
to 29999 miles) travelled After riding more than 30000 miles
(when over 50 of the trained riders had been involved in a crash
compared to approximately 40 in untrained riders) this differ-
ence was no longer statistically significant Satten 1980 found a
lower percentage of crashes in the intervention group (13 969)
compared to the control group (25 2079) but the differencewas not assessed statistically Davis 1997 found a lower crash rate
in trained riders but it was unclear how this was calculated be-
cause the size of the control group was not reported Jonah 1982
and Billheimer 1998 (for the Experienced Rider Course) found no
statistically significant difference in crashes between the interven-
tion and control groups Although the difference between those
who completed the Experienced Rider Course and no course in
Billheimer 1998 neared significance at six months after the course
(P = 007) large P values were observed for one year after (P =
030) Raymond 1979 found a statistically significant higher crash
rate for the intervention group (1861000 miles ridden) com-
pared to the control group (1181000 miles ridden) in both males
and females (P lt 0001) especially during the first 5000 miles of
riding Waller 1992 found a statistically significant higher crashrate for the intervention group (27 crashes in three years or 070
crashes10000 person days) compared to the control group (5
crashes in three years or 013 crashes10000 person days) with a
crude relative risk of 538 (X 2=1513 plt005) Mortimer 1984
and Mortimer 1988 found no statistically significant difference in
crashes between the intervention and control groups
Savolainen 2007 found that the intervention group (those who
completed the Basic Rider Course) was 44 more likely to be
involved in crashes than the control group and those who com-
pleted the course multiple times were an additional 180 more
likely to be involved in a crash This finding was based on low
numbers and no P value or confidence level was given for either
result
Injuries and deaths
Satten 1980 also found a lower percentage of injurious crashes
(33 39) in the intervention group compared to the control
group(60 1220) although again no statistical significance test
was reported
Offences
Jonah 1982 did find a significantly lower self-reportedoffence rate
in the intervention group (15) compared to the control group(34 P lt 001) (no numbers given) Satten 1980 also found
a lower self-reported offence rate in the intervention group (1
71 versus 19100) but again this was not tested for statistical
significance Mortimer 1984 Mortimer1988 and Billheimer 1998
found no statistically significant differencein offence ratesbetween
the intervention and control groups
Case-control studies
CrashesMAIDS 2004 found that a slightly higher percentage of cases
(involved in a crash) had had additional non-mandatory rider
training than controls (not involved in a crash) This difference
was not statistically significant (17 of cases vs 12 of controls
OR 146 95 CI 064 to 337)
Rider training unspecified
Case-control studies
19Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2250
Injuries and deaths
Hurt 1981 found that the odds of injurious crashes were lower inthose who had received professional training compared to those
who had not Kraus 1975 found that in the younger age group
(less than 20 years) and the older age group (over 35 years) a larger
percentage of injured riders (12 and 21 respectively) reported
prior motorcycle rider training compared with the uninjured con-
trol group (6 and 5 respectively) (P lt 001) (no numbers
given) Haworth 1997 reported several different outcomes one
beingthat having completed at leastone trainingcoursecompared
to no training courses at all had no significant effect on the odds
of crashing (afteradjustment for age blood alcohol level or licence
status (learner permitprobationaryfull)) When comparing three
different levels of training courses (beginner intermediate and ad-
vanced) with no training no significant differences were found
in the odds of crashing However after adjustment for age and
blood alcohol level completing an advanced course as compared
to a beginners course was associated with a significant decrease in
the odds of severe motorcycle crashes resulting in serious injury or
death (OR 02 95 CI 01 to 08)
D I S C U S S I O N
The objective of this review was to quantify the effectiveness of
pre- andpost-licence motorcycle rider training on the reduction of
traffic crash involvement injuries deaths and offences committedby motorcycle riders Most included studies were cohort studies
There were a few randomised and non-randomised trials and case-
control studies
All studies compared either mandatory or non-mandatory formal
training with no formal training except for Haworth 1997rsquos case-
control study that compared levels of training (beginner interme-
diate and advanced) to each other
We found studies that evaluated the effectiveness of manda-
tory pre-licence training in reducing crash and offence rates
(Anderson 1980 Hall 1988 Kloeden 1994 Billheimer 1998
(the basic California Motorcyclist Safety Program component of
the study) Perrino 2002) non-mandatory training that when
passed resulted in a licence (Cooper 1988 McDavid 1989
Swaddiwudhipong 1998) non-mandatory training that could be
taken by motorcycle riders with or without a licence (Raymond
1979 Osga 1980 Satten 1980 Jonah 1982 Mortimer 1984
Adams 1985 Leung 1987 Mortimer 1988 Waller 1992 Davis
1997 Billheimer 1998 (the Experienced Rider Course compo-
nent of the study)) and unspecified rider training on crash injury
and death rates (Kraus 1975 Hurt 1981 Haworth 1997 MAIDS
2004)
The studies provided conflicting evidence regarding motorcycle
rider training which is likely due to the poor design quality and
reporting of the studies Most of the studies included in the review
were likely to be adversely affected by significant selection biasalong with substantial detection bias(measurement and recall bias)
which limits our ability to make conclusive comments
Mandatory pre-licence rider training
There were two RCTs three cohort studies and one case-control
study that examined pre-licence ridertraining The two RCTs eval-
uated the Motorcycle Skills Operating Test (MOST) (Anderson
1980 Hall 1988) Anderson 1980 found a statistically significant
reduction in crashes in the intervention group at six months one
and twoyearsfollow up However when only participantswho ob-
tainedtheirlicence were included in theanalysis thereduction wasstatistically significant only at one year follow up The other RCT
(Hall 1988) found no statistically significant differences in crash
rates at any point of time neither in the total population group
nor in the licensed subgroup Neither of the two cohort stud-
ies assessing crash rates found a statistically significant difference
in crash rates between intervention and control groups ( Kloeden
1994 Billheimer1998) andthe case-control study (MAIDS 2004)
also did not find a statistically significant difference in the odds of
experiencing a crash between trained and untrained riders With
regards to offences neither of the RCTs found a statistically sig-
nificant difference in offence rate between the intervention and
control groups in the analysis of the total group of participants
entering the trial However Anderson 1980 found a statistically significant increase in offence rates in the intervention group but
only after two years of follow up andonly in the analysis of partici-
pants who obtained their licence The two cohort studies assessing
offence rates found no statistically significant difference between
the groups (Billheimer 1998 Perrino 2002)
None of the five studies (RCTs or cohort) found a reduction in
crashes or offences following mandatory pre-licence rider training
except for Anderson 1980 who observed a statistically significant
reduction in crashes in the intervention group at one year in both
analyses It is not clear why both RCTs that evaluated the same
MOST and training intervention observed different results Both
studies enrolled large samples and observed a similar uptake of
licensure following the intervention It is possible that differencesmay have arisen due to different study samples (California ver-
sus New York) or that the intervention delivery differed between
studies
In both trials of the Motorcycle Skills Operating Test including
rider training only a third of motorcycle licence applicants ob-
tained their motorcycle licence It is possible that a licensing pro-
cess including a skills test may reduce crashes through decreased
exposure
Non-mandatory pre-licence training
20Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2350
Three studies assessed the effect of non-mandatory pre-licence
training The Cooper 1988 cohort study found no difference incrash or offence rates between the trained and untrained groups
We cannot however be confident about this result as the study
sufferedfrom selectionbiaswhichcouldhave biasedthe measure of
effect toward the null The RCT conducted by Swaddiwudhipong
1998 examined injury and death rates and found a statistically
significant lower injury rate in the intervention villages after one
year of follow up but no longer at two year follow up There
was no statistically significant difference in death rates between
intervention and control villages Our confidence in the results of
this study are hindered by the inadequacy of the randomisation
process that led to significant differences between the intervention
and control districts on two important variables The intervention
districts were more likely than the control districts to have a validlicence andto wear a helmetSincethese differences were nottaken
into account in later analyses they could explain the differences
observed in injury rates between districts Furthermore with only
six districts randomised the sample size was probably too small
to detect a difference between intervention and control groups
had there been one The McDavid 1989 cohort study found a
lower percentage of injuries in the intervention group although
a statistical test was not conducted McDavid 1989 drew their
intervention and control groups from different populations but
did adjust for confounders although they only analysed 40 of
the initial sample (139 trained and 139 untrained) Because of
these substantial weaknesses we are unable to draw conclusions
from this studyAn overall effect of pre-licence training on the reductionof crashes
injuries deaths and offences was not found although one should
be aware that the research of this issue to date has had significant
flaws However a mandatory skill test (such as the MOST) in
combination with and even without pre-licence training as part
of a licensing process seems to reduce the number of individuals
completing the licensing procedure and therefore indirectly re-
duces crash rates through a reduction in exposure to motorcycle
riding Making licensure a more difficult process may therefore be
seen as advantageous if it delays licensure and therefore reduces
exposure It is possible that the group who chose not to continue
riding a motorcycle would not have been the group to experience a
crash or commit an offence if they had continued to ride anyhowfor example because they are more safety conscious and this was
the reason for their giving up riding in the first place If this is the
case deterring individuals from completing their licence and rid-
ing may not necessarily decrease overall crashes deaths or offence
rates In addition other potentially adverse effects of complicat-
ing a licensing procedure need to be considered such as increases
in unlicensed riding In Thailand non-mandatory rider training
offered as a community intervention may have decreased injury
rates but this needs to be confirmed through further research
Non-mandatory training
The majority of the studies assessed non-mandatory training
(training independent of any licensing procedure) and the results
were varied The two non-randomised trials assessing crash rates found no statistically significant difference in crash rates between
the intervention and control groups (Adams 1985 Leung 1987)
Of the nine cohort studies Satten 1980 found a lower percentage
of crashesin the intervention group although it is unclear whether
this was a statistically significant result Davis 1997 also found a
lower percentage of crashesin theinterventiongroup although it is
unclear with which control group the intervention group wascom-
pared Raymond 1979 Osga 1980 Waller 1992 and Savolainen
2007 found a statistically significant higher crash rate in the in-
tervention group and the remaining cohort studies (Jonah 1982
Mortimer 1984 Mortimer 1988 Billheimer 1998) found no sta-
tistically significant difference in crash rates between intervention
and control groups No conclusions can be drawn from these stud-
ies with confidence The Satten 1980 and Davis 1997 studies did
not adjust for any confounding factors even though important
differences existed between the intervention and control groups
Although Raymond 1979 and Waller 1992 adjusted for a num-
ber of confounders at baseline (Waller 1992) and in the analy-
sis (Raymond 1979 Waller 1992) the intervention and control
groups were drawn from such different populations that it is pos-
sible that unmeasured confounders were responsible for the sta-
tistically significant increased risk of crashes observed in the inter-
vention group The observed difference could also be attributed totheobserved differential loss to followup (Raymond 1979) which
could bias the measure of effect towards or away from the null
The Osga 1980 study found a protective effect of rider training
in the univariate analysis but statistically significant higher crash
rates in the intervention group when accounting for riding expo-
sure Sex (a significantly higher proportion of females in the in-
tervention group) and other potentially confounding factors that
could explain thedifferencebetween the groups were howeverun-
accounted for Finally Savolainen 2007 did control for potential
confounding variables yet it is not clear whether variables were
controlled for simultaneously or one at a time One case-control
study (MAIDS 2004) found no statistically significant reduction
in the odds of experiencing a crash between trained and untrainedriders but the quality of this study was poor mainly because the
analysis did not account for any confounding factors
One cohort study (Satten 1980) assessed the effect of non-manda-
tory rider training on injurious crashes and found a lower percent-
age of injuries in the intervention group although a statistical test
was not conducted As mentioned above the Satten 1980 studyrsquos
validity is reduced due to its lack of adjustment for confounding
factors
The only non-randomised trial that assessed offence rates (Leung
1987) found a statistically significant lower offence rate in the in-
tervention group and of the five cohort studies that assessed of-
fence rates (Satten 1980 Jonah 1982 Mortimer 1984 Mortimer
21Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2450
1988) one study (Jonah 1982) also found a statistically significant
lower offence rate in the intervention group However we have lit-tle confidence in the results from Leung 1987 since potential con-
founding factors were not appropriately addressed Jonah 1982
did conduct a multivariate analysis but was hindered by an initial
low response rate The other four cohort studies did control for
several potential confounders and found no difference in offence
rates following non-mandatory post-licence rider training
The non-randomised trials and cohort studies evaluating non-
mandatory rider training were at high risk of substantial selection
bias due to the non-random allocation of participants as well
as low participation rates and attrition bias and could therefore
not provide conclusive evidence with regard to the effect of the
training It should be noted that despite the biases and other issues
with this group of studies most didnot detectmajor effects relatedto rider training in either direction This may suggest that any
potential impact of rider training will be modest
Studies examining unspecified modes of ridertraining
Three case-control studies assessed whether injured or deceased
motorcyclists were more or less likely to have had rider training
compared to control motorcyclists All had conflicting results
Hurt1981 foundthe odds of havingan injuriouscrash waslower in
those who had received rider training whereas Kraus 1975 found
that the odds of having an injurious crash were higher in youngerand older trained riders Haworth 1997 who compared advanced
versus beginner training found no effect on motorcycle injuries
or deaths although they did find that those who had completed
advanced training had reduced chances of crashing compared to
those who had done a beginnersrsquo training course Similar issues
with uncontrolled confounding and biases of selection as described
above arose with the three case-control studies Hurt 1981 suffers
from inadequate control of confounding via poor matching and
both the Kraus 1975 and the Haworth 1997 findings were only
controlled for limited factors by means of stratified analysis It is
likely that these limitations are the reason for the contradictory
directions of effect observed for the three case-control studies
Methodological weaknesses of the studies
In summary most studies investigating the effectiveness of mo-
torcycle rider training on reducing crashes injuries and death suf-
fered from serious methodological weaknesses
The majority of studies were non-randomised and were there-
fore limited mainly by selection bias Although the approaches of
matching stratification or regression analyses can minimise bias
(Fleiss 2004) they were poorly employed in the reviewed studies
The inadequate adjustment for confounders in particular adds
to our lack of confidence in the studiesrsquo results since it has been
demonstrated that individuals who choose to take motorcycle rider
training differ from those not taking rider training For instance
trained riders often differ from controls in age sex and experi-ence usually ride less and may have a more positive attitude to
safety as indicated by the reported more frequent wearing of seat
belts (Mortimer 1982)
The two large randomised trials reported that a large proportion
of their study group were never exposed to the risk of offences or
crashes because they never attained licensure This made it impos-
sible to assess whether a reduction in crashes in the intervention
group was a result of training or due to delayed or never attained
licensure
In some studies the sample sizes were too small to detect any sta-
tistically significant differences Follow-up time after training dif-
fered between studies and was an impediment to observing any
real effects of training if they existed As riding experience ac-cumulates over time any differences in outcomes between those
trained and untrained would diminish in studies with lengthy fol-
low-up times This dilution of any possible effect of rider training
may have also occurred in studies that mixed both novice and ex-
perienced riders in the study groups since it is hypothesised that
novice riders serve to benefit most from rider training This would
increase the difficulty in finding any effect of rider training par-
ticularly when sample sizes are small
Detection bias may also account for some of the differences ob-
served between trained and untrained riders Detection bias would
have occurred if the self-report hospital driver or police records
used to record the outcomes of crash injury death or offences
differed between the trained and untrained riders Research hasdemonstrated that not all crashes are reported to police for ex-
ample small crashes or off-road crashes (Kraus 1975) even serious
injury crashes have been found to be underreported (Anderson
1980) However this would only introduce detection bias if the
reporting of outcome differs for trained and untrained riders This
does appear possible in that responsible riders may be more likely
to enroll in a rider training course and also may be more likely
to report crashes to police than irresponsible riders meaning that
the number of crashes from riskier riders could be underrepre-
sented on records An analysis to investigate the effectiveness of
rider training may then show that rider training is not effective in
reducing crashes
Rather than using reported outcomes some studies used self-re-ported survey data although this technique suffers from non-
response biases In fact two studies reported differences in de-
mographic characteristics in response rates to questionnaires
(Anderson 1980 Hall 1988) Self-reported data also introduce
the risk of recall bias survival bias and social desirability response
bias Recall bias may have been equally distributed among trained
and untrained riders howeverit is hypothesised that trained riders
may have felt it was lsquoexpectedrsquo to report good driving histories
hence leading the rider training to appear effective
22Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2550
Limitations of this review
Limitations of this review were that we found very few recentstudies (the two most recent ones published in 2002 and 2007)
and that we found very few studies conducted in low-and middle
income countries even though our search strategy was set up to
identify such studies Anotherlimitation wasthat forthe reviewed
studies we could not assess all outcomes of rider training that are
potentially useful Some of the included studies assessed outcomes
such as use of safety clothing including helmets Such outcomes
are valuable since the use of motorcycle helmets has been demon-
strated to be effective in preventing head injuries (Liu 2004) De-
spite this the use of safety clothing was not included as an out-
come in this review because the studies measuring this outcome
were limited by poor data quality due to very low response rates
Another outcome of interest in assessing the effectiveness of rider
training is lsquotype of injuryrsquo As Haworth 2005 state in their review
of motorcycle licensing and training rider training could be con-
sidered effective if injuries become less severe because motorcycle
riders have learned how to avoid more serious injuries through
rider training The studies included in this review did not specify
type of injury and so this outcome could not be assessed Last
changes in knowledge and attitudes following rider training are
other potentially useful outcomes We have excluded such studies
(Osga 1979 Clayton 1990 Goldenbeld 2004) as these outcomes
are proxies rather than direct measures of the effectiveness of mo-
torcycle rider training We opted to report on the outcomes of
crashes injury death and offences alone since there is little ev-
idence about how changes in knowledge and attitudes relate to
these outcomes
A U T H O R S rsquo C O N C L U S I O N S
Implications for practice
Due to the poor quality of studies included in this review we were
unable to draw any conclusions about the effectiveness of rider
training on crash rates injurious or fatal or offence rates The
findings do suggest that mandatory pre-licence training may be
an impediment to completing a motorcycle licensing process thuspossibly indirectly reducing crash injury death and offence rates
through a reduction in exposure to riding a motorcycle On the
basis of the existing evidence it is not clear if and what type of
training reduces the risk of crashes injuries deaths or offences in
motorcyclists and the selection of the best rider training practice
can therefore not be recommended The suggestion that manda-
tory pre-licence training indirectly reduces exposure to riding the
motorcycle because trainees are impeded from completing the li-
censing process does not help those motorcyclists who persist in
obtaining their licence and riding the motorcycle Some type of
rider training is likely necessary to teach motorcyclists to ride a
motorcycle safely Good policy is based on sound evidence and
evaluation but such evidence for motorcycle rider training is seri-
ously lacking Policy makers are therefore encouraged to include arigorous evaluation component to any new or existing rider train-
ing programme Such an evaluation should ideally be randomized
and assess whether the training programme is effective and if so
which component(s) of the programme work best
Implications for research
Given theresults of this review which found littlereliable evidence
about the effectiveness of motorcycle rider training it is necessary
to build up an evidence base of the effectiveness of motorcycle
rider training in the reduction of crash injury death and offence
rates To this end well designed randomised controlled trials need
to be conducted that examine the impact of rider training oncrashes and offences A randomised controlled trial is the preferred
method as it would significantly reduce most biases described in
this review To minimise the impact of participants delaying or
never attaining their licence and therefore exposure as found in
the randomised trials in this review such a trial should include
riders who are already licensed and reliable measures of riding
exposure in a reasonably lengthy follow up period It is expected
that a randomised controlled trial of sufficient sample size would
observe equal balance on possible confounders by study arm how-
ever an assessment of potentially important confounders (such as
helmet wearing) by study group would enable control for these
variables in the analyses if necessary It is imperative that the effec-
tiveness of each component of the rider training (such as theorypractice knowledge skills and modification of behaviour) and the
synergy of such components are evaluated Outcomes should be
objective ideally those that are routinely collected rather than self-
reported outcomes Even though riding the motorcycle poses dif-
ferent challenges and incurs different risks lessons may be learned
from research into the effects of car driver training
Motorcycles are a common means of transport in low- and middle
income countries and usage is increasing As the burden of mo-
torcycle crash victims is mostly borne in such countries research
should be focused on the effectiveness of types of rider training on
the reduction of crashes injuries and fatalities in such countries
and be supported by higher income countries This review found
only one study conducted in a middle income country that sug-gested decreasing injury rates It would be beneficial to conduct
additional research to confirm this suggestion and to enhance our
knowledge of the role of rider training in low- and middle-income
settings
March 2006 saw the launch of the University of North Carolina
Highway Safety Research Centre partnership with the Motorcycle
Safety Foundation (MSF) and plans were revealed to conduct a
study of the potential benefits of motorcycle rider training The
study is funded by the MSF and the National Highway Traffic Sa-
fety Administration (NHTSA) Called rsquoThe Discovery Projectrsquo it
will compare data on crash rates from riders participating in a basic
23Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2650
motorcycle safety class with riders participating in a basic motor-
cycle safety class as well as a variety of continuing education classesThe collection of pilot data was due to start in March 2007 and
the data collection portion of the study is expected to take three
years (Highway Safety Research Center) It is vital that researchers
and policy makers collaborate to develop furtherhigh quality eval-
uations of motorcycle rider training programmes to help inform
development of new and effective training programmes
A C K N O W L E D G E M E N T S
Staff from The University of Sydney library for helping us locate
and retrieve full reports of many papers
R E F E R E N C E S
References to studies included in this review
Adams 1985 published data only
Adams A Collingwood V Job RFS Evaluation of the 1982 Australia
Post motorcycle rider training scheme NSW Australia Traffic
Accident Research Unit Traffic Authority NSW 1985 [ RN 85]
Anderson 1980 published data only
Anderson J Ford JL Peck RC Improved motorcyclist licensing and testing project Washington DC National Highway Traffic Safety
Administration (NHTSA) USA 19801ndash217 [
DOTndashHSndash805ndash598]
Billheimer 1998 published data only
Billheimer JW Evaluation of the California Motorcyclist Safety
Program Transportation Research Record 19981640100ndash9 [
TRB Report Number 980652]
Cooper 1988 published data only
Cooper PJ Rothe JP Motorcyclists who they are and why they do
what they do Transportation Research Record 1988116878ndash85
Davis 1997 published data only
David CF Evaluation of community traffic safety programs and
motorcycle operator training programs Connecticut Department of Transportation 199763
Hall 1988 published data only
Hall CJ Buchanan L Engle R Motorcycle rider education evaluationproject Washington DC USA National Highway Traffic Safety
Administration (NHTSA) US Department of Transportation
1988 [ DTNH 22ndash80ndashCndash05123]
Haworth 1997 published data only
Haworth N Smith R Brumen I Pronk N In Monash University
Accident Research Centre editor(s) Case-control study of motorcycle
crashes Victoria Australia Department of Transport and Regional
Development The Federal Office of Road Safety 19971ndash149 [
ISBN 0 642 25501 6 ]
Hurt 1981 published data only
Hurt H Ouellet J Thom D Motorcycle accident cause factors and
identification of countermeasures USA Department of
Transportation 1981 [ Technical Report DOT HS 806ndash862]
Jonah 1982 published data onlylowast Jonah BA Dawson N Bragg B Are formally trained motorcyclists
safer Accident Analysis amp Prevention 198214(4)247ndash55
Jonah BA Dawson NE Bragg BW Evaluation of the Canada
Safety Councilrsquos Motorcycle Training Program Proceedings of theInternational Motorcycle Safety Conference Washington DC
USA 198059ndash93
Kloeden 1994 published data only
Kloeden CN Moore VM McLean AJ Evaluation of the pre-license
training program for motorcyclists in South Australia Walkerville
South Australia Australia NHMRC Road Accident Research Unit
The University of Adelaide 19941ndash37 [ ISBN 0 7308 0607 3]
Kraus 1975 published data only
Kraus JF Riggins RS Franti CE Some epidemiological features of
motorcycle collision injuries American Journal of Epidemiology 1975102(1)74ndash98
Leung 1987 published data only
Leung KS Reding VA Evaluation of the Wisconsin motorcycle rider course Wisconsin Bureau of Policy Planning and Analysis
Wisconsin Department of Transportation 1987
MAIDS 2004 published and unpublished data
MAIDS In-depth investigations of accidents involving powered
two wheelers ACEM (Acemacemeu) 2004 Vol Final report
20179
McDavid 1989 published data only
McDavid JC Lohrmann BA Lohrmann G Does motorcycle
training reduce accidents Evidence from a longitudinal quasi-
experimental study Journal of Safety Research 198920(2)61ndash72
Mortimer 1984 published data only
Mortimer R Evaluation of the motorcycle rider course Accident
Analysis amp Prevention 198416(1)63ndash71
24Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2750
Mortimer 1988 published data only
Mortimer RG A further evaluation of the motorcycle rider courseJournal of Safety Research 198819(4)187ndash96
Osga 1980 published data only
Osga GA An investigation of the riding experiences of MSF rider
course participants Dissertation South Dakota USA The
University of South Dakota 19801ndash224
Perrino 2002 published data only
Perrino CS Ahmed A Callender A Rozier E Cantwell A Stewart
O The role of Marylandrsquos motorcycle rider course in promoting saver
behaviors and attitudes Baltimore USA National Transportation
Center Morgan State University 20021ndash52
Raymond 1979 published data only
Raymond S Tatum S An evaluation of the effectiveness of the RAC
ACU motorcycle training scheme - final report Road Safety ResearchUnit Dept of Civil Engineering University of Salford London
19791ndash131
Satten 1980 published data only
Satten RS Analysis and evaluation of the motorcycle rider courses
in thirteen Illinois counties International Motorcycle Safety
Conference Washington DC 1980145ndash93
Savolainen 2007 published and unpublished data
Savolainen Peter Mannering Fred Additional evidence on the
effectiveness of motorcycle training and motorcyclistsrsquo risk-taking
behaviour Transportation Research Board 2007 Vol March 20
19
Swaddiwudhipong 1998 published data only
Swaddiwudhipong W Boonmak C Nguntra P Mahasakpan P
Effect of motorcycle rider education on changes in risk behavioursand motorcycle-related injuries in rural Thailand Tropical Medicine
and International Health 19983(10)767ndash70
Waller 1992 published data only
Waller FL An analysis of the impact of the Texas Department of
Public Safety Motorcycle Operator Training Program upon
motocycle related traffic accidents Texas State University Public
Administration Program Applied Research Projects 199277
References to studies excluded from this review
Anderson 1980a published data only
Anderson LC Better biking a motorcycle rider improvement
program International Motorcycle Safety Conference Washington
DC USA Motorcycle Safety Foundation 1980231ndash53Aupetit 2008 published and unpublished data
Aupetit S Riff J Buttelli O Espie S Ergonomics research program
on motorcycle learning Contributions to the development of a
motorcycle simulator for training Transport Research Arena
Ljubljana 20089
Awane 1999 published data only
Awane T Integrating simulators in motorcycle safety education
Journal of International Association of Traffic and Safety Sciences 199923(1)26ndash35
Baldi 2005 published data only
Baldi S Baer JD Cook AL Identifying best practices states in
motorcycle rider education and licensing Journal of Safety Research200536(1)19ndash32
Batchler 1988 published data only
Batchler MW Evaluation of the pre-license motorcycle training course at the New South Wales traffic education centre Federal Office of
Road Safety (FORS) 19881ndash122 [ Report No CR 77 ]
Blanchard 2006 published data only
Blanchard HT Tabloski PA Motorcycle safety educating riders at
the teachable moment Mosby (CV) Company 200632330ndash2
Bonnett 2005 published data only
Bonnett J Mature age skills training for experienced motorcycle
riders Motorcycle ridersrsquo association of the Australian Capital
Territory 2005
Braver 2007 published data only
Braver ER et alPersuasion and licensure a randomized controlled
intervention trial to increase licensure rates among Maryland
motorcycle owners Traffic Injury Prevention 20078(1)39ndash46Clayton 1980 published data only
Clayton AB Sudlow DE An evaluation of the effectiveness of a one day training programme for learner motorcyclists Berkshire UK
Transport and Road Research Laboratory 19801ndash23 [ ISSN
0266ndash7045]lowast Clayton AB Sudlow DE An evaluation of the effectiveness of a one-day training programme for learner motorcyclists Berkshire UK
Road Safety Division Berkshire 19801ndash23 [ISSN 0266ndash7045]
Council 1978 published data only
Council FM Desper LP Dutt AK Driver education for motorcycle operation University of North Carolina Highway Safety Research
Center 19781ndash40
Crowther 2005 published data onlylowast Crowther G Engaging with motorcyclists UK police and theBikeSafe Road Safety Programme Published proceedings of the
Australasian Road Safety Research Policing Education Conference
Wellington New Zealand 2005 issue ISBN 0ndash473ndash10636ndash11ndash11
Daniello 2009 published data only
Daniello A Gabler HC Mehta YA Effectiveness of motorcycle
training and licensing Transportation Research Board 88th Annual
Meeting 200913
Dewanti 2007 published data only
Dewanti M Challenging efforts in promoting young motorcyclist
safety in Indonesia Road Safety on Four Continents 14th
International Conference Swedish National Road and Transport
Research Institute 2007
Du Plooy 1990 published data onlyDu Plooy DW Motorcycle training in South Africa The Human
Element International Motorcycle Safety Conference Orlando
Florida USA Motorcycle Safety Foundation 19904-1 to 4-9
Engel 1989 published data only
Engel U Krogsgaringrd Thomsen L [Faeligrdselsundervisning og
propaganda ndash en faeligrdselssikkerhedsfremmende foranstaltning]
Dansk Vejtidsskrift 198912295ndash7
Goldenbeld 2002 published data only
Goldenbeld C Houwing S de Craen S The development of
driving skills of young moped riders [De ontwikkeling van de
rijvaardigheid van jonge bromfietsers] SWOV Institute for Road
Safety Research (Stichting Wetenschappelijk Onderzoek
Verkeersveiligheid) Leidschendam the Netherlands 200263
25Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2850
Goldenbeld 2004 published data only
Goldenbeld C Twisk D de Craen S Short and long term effects of moped training a field experiment Transportation Research Part F 20047(1)1ndash16
Haworth 2000 published data only
Haworth N Smith R Kowadlo N Motorcycle training in Victoria
evaluation of rider training curriculum in Victoria Monash
University Accident Research Centre Reports 2000 Vol 165 issue
3131
Haworth 2005 published data only
Haworth N Mulvihill C Review of motorcycle licensing and
training Monash University Accident Research Centre 2005 [
Report no240]
Haworth 2006 published data only
Haworth N Mulvihill C Rowdon P Teaching old dogs new tricksTraining and older motorcyclists Journal of the Australasian College of Road Safety 200718(4)20ndash5lowast Haworth NL Mulvihill C Rowden PJ Teaching old dogs new
tricks Training and older motorcyclists Australasian Road Safety
Research Policing and Education Conference Gold Coast
Queensland 2006
Hill 1983 published data only
Hill D Motorcycle rider training testing and licensing Motor
Cycle Council of New South Wales Sydney 1983
Imatake 1980 published data only
Imatake Y On the activities in Japan of motorcycle safe riding
promotion committee (MSRPC) - especially the education of
unskilled riders of class 1 motorbikes International Motorcycle
Safety Conference Washington DC USA Motorcycle Safety
Foundation 1980255ndash68
Ingebritsen 1990 published and unpublished data
Ingebrigtsen S [Risikofaktorer ved ferdsel med moped og
motorsykkel] TOslashI-rapport 66 Oslo Transportoslashkonomisk
institutt 1990
Jain 2009 published and unpublished data
Jain A et alTwo wheeler accidents on Indian roads a study from
Mangalore India Elsevier 200916130ndash3
Jonah 1979 published data only
Jonah BA Dawson NE Validation of the motorcycle operator skill
test Accident Analysis amp Prevention 197911(3)163ndash71
Jonah 1981 published data onlyJonah BA Dawson NE Bragg WE Predicting accident
involvement with the Motorcycle Operator Skill Test Accident Analysis amp Prevention 198113(4)307ndash418
Kadar 1990 published data only
Kadar KJ Countermeasures motorcycle rider education delivery
system The Human Element International Motorcycle Safety
Conference Motorcycle Safety Foundation 19906-27 to 6-41
Kelsey 1986 published data only
Kelsey SL Liddicoat C Ratz M Licensing novice motorcyclists a
comparison of Californiarsquos standard test and the MOST II (Motorcycle
Operator Skill Test) administered at centralized testing offices Sacramento California USA California Department of Motor
Vehicles Research and Development Office 19861ndash74
Koch 1980 published data only
Koch H Advanced course for motorcycle riders InternationalMotorcycle Safety Conference Washington DC USA Motorcycle
Safety Foundation 1980269ndash81
Koch 1990 published data only
Koch H Brendicke R How to influence risk taking behaviour
within motorcycle rider training programs The Human Element
International Motorcycle Safety Conference Orlando Florida
USA Motorcycle Safety Foundation 1990Motorcycle Safety
Foundation 1990 p 9-1 to 9-25A
Labbett 2005 published data only
Labbett S Langham M Training can make the problem worse
Road Safety Congress 20051ndash12
Law 2005 published and unpublished data
Law TH Radin Umar RS Zulkaurnain S Kulanthayan S Impactof the effect of economic crisis and the targeted motorcycle safety
programme on motorcycle-related accidents injuries and fatalities
in Malaysia International Journal of Injury Control and Safety
Promotion 200512(1)9ndash21
Lowes 1990 published data only
Lowes BF In-traffic rider training The Human Element
International Motorcycle Safety Conference Orlando Florida
USA Motorcycle Safety Foundation 19907-46 to 7-57
Mahon 1976 published data only
Mahon J Teaching accident avoidance on the motorcycle Journal of Traffic Safety Education 1976231ndash29
Maxwell 1985 published data only
Maxwell GS Evaluation of a Program for Training MotorcycleRiders Evaluation and Program Planning 19858(4)339ndash47
Mayhew 1996 published data only
Mayhew DR Simpson SH Effectiveness and role of driver educationand training in a graduated licensing system Ottowa Canada The
Traffic Injury Research Foundation 19961ndash89 [ ISBN
0ndash920071ndash12ndash0]
McKnight 1987 published data only
McKnight AJ Milburn BG An evaluation of the modified
motorcycle operator skill test exercises NHTSA Report 1987 [
Report No DOT HS 807 212 ]
McPherson 1976 published data only
McPherson K McKnight AJ The development and evaluation of a
motorcycle skill test manual and knowledge test Washington DCUSA National Highway Safety Administration 1976
Mohan 2004 published data only
Mohan D Evidence-based interventions for road traffic injuries in
South Asia Journal of the College of Physicians and SurgeonsPakistan 200414(12)746ndash7
Morgan 1980 published data only
Morgan JK Advanced motorcycle training course content and
study evaluation International Motorcycle Safety Conference
Washington DC USA Motorcycle Safety Foundation 1980
283ndash312
Motoki 1990 published data only
Motoki M Yamazaki S A study on effective motorcycle rider
education The Human Element International Motorcycle Safety
26Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2950
Conference Orlando Florida USA Motorcycle Safety
Foundation 19909-26 to 9-57A
Nairn 1992 published data onlylowast Nairn RJ Motorcycle safety research literature review 1987 to
1991 Federal Office of Road Safety Australia 1993 Vol Report
CR 117
Newland 1980 published data only
Newland R Darby J Automotive practicespre motorcycle rider
program International Motorcycle Safety Conference Washington
DC USA Motorcycle Safety Foundation 1980127ndash9
Newland 1982 published data only
Newland R Motorcycle rider training and licensing requirements
Driver training steering a course for the future Melbourne
Victoria Transport Regulation Board 1982
Newland 1990 published data only
Newland R Motorcycle training and licensing in Victoria 1980-
1990 The Human Element International Motorcycle Safety
Conference Orlando Florida USA Motorcycle Safety
Foundation 1990
Ormston 2003 published data only
Ormston R Dudleston A Pearson S Stradling Evaluation of
Bikesafe Scotland Scottish Executive Social Research Edinburgh
2003
Osga 1979 published data only
Osga GS Ellingstad VS An evaluation of the effectiveness of the 1978
South Dakota motorcycle rider course program South Dakota USA
University of South Dakota 19791ndash89
Panou 2007 published data only
Panou M Bekiaris E A new concept on the integration of driving
simulators in driving training the train-all approach Road Safety
on Four Continents 14th International Conference Swedish
National Road and Transport Research Institute 2007
Peterson 1990 published data only
Peterson A Kadar K Status report 1980-1989 Minnesota rider
education statistics The Human Element International
Motorcycle Safety Conference Orlando Florida USA 199013-
113 to 13-129
Prem 1984 published data only
Prem H Good MC Motorcycle rider skills assessment ACT
Australia Office of Road Safety Department of Transport 1984
1ndash377 [ Report No CR 34 ]
Radin Umar 2006 published data only
Radin Umar RS Motorcycle safety programmes in Malaysia how
effective are they International Journal of Injury Control and Safety Promotion 200613(2)71ndash9
Rockwell 1990 published data only
Rockwell TH Kiger S Carnot M An evaluation of the Ohio
motorcyclist enrichment program Ohio Department of Highway
Safety 1990
Rothe 1987 published data only
Rothe JP A motorcycle riding course Motorcyclists Image and reality Canada Insurance Corporation of British Columbia 1987
59ndash83
Rowden 2007 published data only
Rowden P Watson B Haworth N What can riders tell us aboutmotorcycle training A view from the other side of the fence
Australasian Road Safety Research Policing and Education
Conference Melbourne Australia Melbourne 2007
RSC Victoria 1993 published data only
Road Safety Committee Inquiry into Motorcycle Safety in Victoria
Parliament of Victoria 1993
Russam 1979 published data only
Russam K Motorcycle training and licensing in Japan UK Transport
and Road Research Laboratory (TRRL) 1979 [ Report 916 ]
Saffron 1981 published data only
Saffron DG Driver and rider education training and licensing a
brief review Traffic Accident Research Unit Traffic Authority of
New South Wales NSW Australia Traffic Accident Research UnitTraffic Authority of New South Wales 1981
Schulz 1990 published data only
Schulz U Motorcycle accidents among learner riders Motorcycle
licensing Evaluation of the effectiveness of a Graduated Licensing Scheme Forshungshefte des Institut fur Zweiradsicherheit 1990
Sexton 2004 published data only (unpublished sought but not used)
Sexton B Baughan C Elliott M Maycock G The accident risk of
motorcyclists 2004TRL Report TRL 607(ISSN 09684107)58
Shephard 1986 published data only
Shepherd R Do Motorcycle Safety Foundation programs work
Driver 198619(9)16ndash8
Simpson 1990 published data onlySimpson HM Mayhew DR The promotion of motorcycle safety
training education and awareness Health Education Research
Theory and Practice 19905(2)257ndash64
Sokolov 1990 published data only
Sokolov O Safe motorcycle riding some aspects of motorcycle
riding in the USSR The Human Element International
Motorcycle Safety Conference Motorcycle Safety Foundation
19904ndash37 - 4-48
Steffens 1988 published data only
Steffens U Gawatz R Willmes G Effectiveness of mofa courses
[Wirksamkeit von Mofakursen] Unfall- und sicherheitsforschung
strasenverkehr 198867168
Sudlow 2003 published data only
Sudlow D Scoping study on motorcycle training Road Safety
Research Report London UK The British Institute of Traffic
Education Research Department for Transport 2003 Vol June
[ Report No 36]
Tasmania 1995 published data only
Mannering FL Grodsky LL Motorcycle rider training scheme
Traffic Education Research Centre Armidale Australia Driver
Training Australia Limited 1995 Vol April1ndash56
Tetard 1985 published data only
Tetard C Motorcycle ridersrsquo training to detect indices in an
intersection situation [Lrsquoentrainement a la prise drsquoinformation dans
lrsquoapprentissage de la conduite motocycliste] Le Travail Humain198548(2)147ndash59
27Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3050
Thackray 1980 published data only
Thackray RM Prescott JC Field test of the motorcycle ridercourse International Motorcycle Safety Conference Washington
DC USA Motorcycle Safety Foundation 1980195ndash7
Thompson 1994 published data only
Thompson M Evaluation of compulsory basic training for
motorcyclists Transport Research Laboratory Crowthorne
Berkshire UK 1994 [ Report 63]
Tobler 1990 published data only
Tobler U Individual driving instruction for motorcycle driving
learners the Swiss experiment of motorcycle driving school The
Human Element International Motorcycle Safety Conference
Orlando Florida USA Motorcycle Safety Foundation 1990
Transport 1989 published data only
Transport Communications Road Crash Statistics Australia Bureauof Transport and Communications Economics 198914ndash20
Tuan 2007 published data only
Tuan VA Shimizu T Towards development and evaluation of the
motorcycle drivers re-education program in Vietnam modeling of
motorcycle driverrsquos undesired behaviors 11th World Conference
on Transport Research 200725
Victor 1980 published data only
Victor DJ Traffic safety education for motorcyclists International
Motorcycle Safety Conference Washington DC USA Motorcycle
Safety Foundation 1980221ndash8
Waterer 1990 published data only
Waterer TH Compulsory training for motorcyclists from liberty
to compulsion The Human Element International Motorcycle
Safety Conference Orlando Florida USA Motorcycle Safety
Foundation 19904-49 to 4-57
Weller 1989 published data only
Weller RB Chandler EW Motorcycle safety and motorcycle
education past research and survey results Journal of Public Policy
amp Marketing 1989893ndash108
Wells 1981 published data only
Wells P Observations of some RACACU motorcycle training
schemes Transport and Road Research Laboratory Crowthorne
Berkshire UK 1981
Winn 1986 published data only
Winn G Two important trends in motorcycle safety regulations
rider education and conspicuity improvement American
Motorcyclist Association 1986147ndash52
Wisher 1988 published data only
Wisher G Reid J Motorcycle accidents in Norfolk a three-year study of motorcycle training and the Part 1 test Norfolk UK Road Safety
Division Highways Department Norfolk County Council 1988
Wood 1987 published data only
Wood HT Bowen R Evaluation of the revised learner permit scheme
July 1983 to December 1985 Victoria Australia Road Traffic
Authority 19871ndash27 [ ISBN 0 7306 0247 8]
Yamazaki 1990 published data only
Yamazaki S Nakanishi M Survey on effect of Honda rider
education The Human Element International Motorcycle Safety
Conference Orlando Florida USA 199012-167 to 12-187
References to ongoing studies
Burgess 2010 unpublished data only
Burgess C Broughton P Fylan F Stradling Interim evaluation of
the UKs National Ride program In L Dorn editor(s) Driver
behaviour and Training Vol IV Aldershot Ashgate Publishing
Limited 2010
Additional references
Anderson 1980
Anderson J Ford JL Peck RC Improved motorcyclist licensing and
testing project Volume I and II National Highway Traffic Safety
Administration US department of Transportation Washington
DC 1980
Bjornstig 1985
Bjornstig U Bylund P Lekander T Brorsson B Motorcycle fatalitiesin Sweden Acta Chirurgica Scandinavica 1985151(7)577ndash81
Clayton 1990
Clayton A Sudlow DE An evaluation of the effectiveness of a one-
day training program for learner motorcyclists The Human
Element International Motorcycle Safety Conference Orlando
Florida USA 199013-105 to 13-111
Cochrane Collaboration 2008
Higgins JP Green S Cochrane Handbook for Systematic Reviews of
Interventions 500 The Cochrane Collaboration 2008
Elliot 2003
Elliott MA Baughan CJT Broughton J Grayson G Knowles J et
alMotorcycle Safety a Scoping Study TRL [ Report No
0968ndash4107]Fleiss 2004
Fleiss JL Levin B Paik MC Statistical methods for rates and proportions 3rd Edition John Wiley amp Sons Inc 2004 [ ISBN
978ndash0ndash471ndash52629ndash2 ]
FORS 1999
FORS Federal Office of Road Safety Road risk for sober licensed
motorcyclists Vol Monograph 27 Federal Office of Road Safety
19991ndash2
Haworth 2005
Haworth N Mulvihill C Review of motorcycle licensing and training
Victoria Australia Monash University Accident Research Centre
20051ndash83 [ Report No 240]
Huwiler-Muumlntener 2002Huwiler-Muumlntener K Juumlni P Junker C Egger M Quality of
Reporting of Randomized Trials as a Measure of Methodologic
Quality JAMA 20022872801ndash4
Johnston 2008
Johnston P Brooks C Savage H Fatal and serious road crashes
involving motorcyclists Road Safety Research and Analysis Report
ATSB Monograph 20 Canberra Department of Infrastructure
Transport Regional Development and Local Government 0205
081ndash26
Juumlni 2001
Juumlni P Altman DG Egger M Systematic reviews in health care
Assessing the quality of controlled clinical trials BMJ 2001323
42ndash6
28Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3150
Kopits 2003
Kopits E Cropper M Traffic fatalities and economic growth TheWorld Bank Development Research Group Infrastructure and
Environment Policy research working paper 3035 2003
Leung 1983
Leung P Chau BTW Motorcycle accidents in Hong Kong pattern
of injury treatment causative factors and socio-economic
implications Injury 198315(1)23ndash9
Liu 2004
Liu B Ivers R Norton R Blows S Lo SK Helmets for preventing
injury in motorcycle riders Cochrane Database of Systematic Reviews 2004 Issue 2 [DOI 10100214651858CD004333pub3]
Mohan 2002
Mohan D Road safety in less-motorized environments future
concerns International Journal of Epidemiology 200231(3)527ndash32Mortimer 1982
Mortimer RG An evaluation of the motorcycle rider course
Proceedings of the 20th conference of the American Association for
Automotive Medicine Arlington Heights IL USA American
Association for Automotive Medicine 198277ndash96
Mullin 1997
Mullin BT Risk factors for motorcycle injury the role of age
gender experience training and alcohol PhD thesis Auckland
New Zealand University of Auckland 1997
Noordzij 2001
Noordzij PC Forke E Brendicke R Chinn BP Integration of needs
of moped and motorcycle riders into safety measures review and
statistical analysis in the framework of the European research
project PROMISING SWOV Institute for Road Safety Research
Leidschendam The Netherlands 2001 [ Report No Dndash2001ndash5]
Osga 1979
Osga G Ellingstad VS An evaluation of the effectiveness of the 1978 South Dakota Motorcycle Riders Course Program South Dakota
University of South Dakota Vermillion Human Factors Lab 1979
Oxford English Dictionary Online
httpoxforddictionariescom Oxford English Dictionary Online
Oxford University Press 2010
Peden 2004
Peden M Scurfield R Sleet D World Report on Road TrafficInjury Prevention Geneva World Health Organization Available
at httpwwwwhointviolence˙injury˙preventionpublications
road˙trafficworld˙reporten 2004Chapter 3
Radin Umar 1995
Radin Umar R Macakay G Hills B Preliminary analysis of
motorcycle accidents Short-term impacts of the running
headlights campaign and regulations in Malaysia J Traffic Med 199523(1)17ndash27
Tsai 1995
Tsai YJ Wang JD et alCase-control study of the effectiveness of
different types of helmets for the prevention of head injuries among
motorcycle riders in Taipei Taiwan American Journal of
Epidemiology 1995142(9)974ndash81
Vis 1995
Vis AA In-depth study of the hazards of motorcycling [De
onveiligheid van motorrijden nader bekeken] SWOV Institute for
Road Safety Research Leidschendam the Netherlands 1995 [
Report No Rndash95ndash69]
Weaver 1990
Weaver EA Motorcycle safety education in the United States
Motorcycle Safety Foundation Orlando Florida USA 1990
Winn 1987
Winn G Two important trends in motorcycle safety regulations
rider education and conspicuity improvement No more
information available Detroit MI US Society of Automotive
Engineers 1987147ndash52
Yuan 2000
Yuan W The effectiveness of the rsquoride-brightrsquo legislation for
motorcycles in Singapore Accident Analysis amp Prevention 200032
(4)559ndash63
Zambon 2006
Zambon F Hasselberg M Factors affecting the severity of injuries
among young motorcyclists a Swedish nationwide cohort study
Traffic Injury Prevention 20067(2)143ndash9lowast Indicates the major publication for the study
29Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3250
C H A R A C T E R I S T I C S O F S T U D I E S
Characteristics of included studies [ordered by study ID]
Adams 1985
Methods Non-randomised trial
Participants Initial sample size
Total of 162 Australia Post telegram delivery riders
Not clear how many in intervention and control groups
Analysis sample size
Intervention group 63Control group 35
Total 98 (61)
Interventions No formal training vs voluntary rider training course of 12 hours over 2 consecutive days (theory 4 hours practice
8 hours)
Intervention integrity
No information given
Outcomes Australia Post recorded crashes
Length of follow up 1 year
Notes Study conducted in NSW Australia
Anderson 1980
Methods Randomised trial
Participants Initial sample size
Total 41188 motorcycle licence applicants
Analysis sample size
Intervention group with skills training 13160
Intervention group without skills training 12634
Control group 15080
Total 40874 (99)
Interventions Standard licensing skill test vs the Motorcycle Operators Skill Test (MOST) with or without a training component
MOST is not to be confused with rider training because it is a licensing test The training component was for
participants who failed the skill test on the first attempt was of 3 hours duration and consisted of instructions about
the skills needed to pass the skill test The training consisted of 30 min theory and 2 frac12 hours practice on the
motorcycle
Intervention integrity
33 in MOST with skills training 31 in MOST without skills training and 43 in control group obtained their
motorcycle licence in the end
Outcomes Police recorded crashes and offences
Length of follow up 3 years
30Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3350
Anderson 1980 (Continued)
Notes Study conducted in California USA
Billheimer 1998
Methods Cohort study
Participants Initial sample size
Intervention group Unclear numbers of CMSP graduates and ERC graduates
Control group Over 16000 untrained motorcyclists
Analysis sample size
1139 matched CMSP-pairs and 1182 ERC-pairs
Interventions No formal training vs the mandatory (for those younger than 18 years) pre-licence 16-hours basic California
Motorcyclist Safety Program (CMSP) and the 8-hour Experienced Rider Course (ERC)
Intervention integrity
No information given
Outcomes Police recorded crashes and offences
Length of follow up 2 years
Notes Study conducted in California USA
Cooper 1988
Methods Cohort study
Participants Initial sample size
Intervention group 418 graduates and 57 who failed
Control group 402
Analysis sample size
Intervention group unclear
Control group unclear
Total 863 (984)
Interventions No formaltrainingvs non-mandatory pre-licence rider trainingThe content of thetrainingwas notfurther explainedIntervention integrity
Participants had all completed the rider training
Outcomes Police recorded crashes and offences
Length of follow up 7 years
Notes Study conducted in British Columbia Canada
31Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3450
Davis 1997
Methods Retrospective cohort study
Participants Initial sample size
Intervention group 9320 graduates
Control group unclear
Analysis sample size
Intervention group 9320 graduates
Control group unclear
Interventions No formal training vs the non-mandatory Connecticut Motorcycle Operator Training Program The content of
training was not further explained
Intervention integrity Participants had all completed the rider training
Outcomes Police recorded crashes severity of crashes
Length of follow up 1 year
Notes Study conducted in Connecticut USA
Hall 1988
Methods Randomised trial
Participants Initial sample sizeIntervention group A 6248
Intervention group B 6616
Intervention group C 6996
Control group 6604
Total 26464 motorcycle licence applicants
Analysis sample size
Intervention group A 6012 (96)
Intervention group B 6428 (97)
Intervention group C 6778 (97)
Control group 6502 (98)
Interventions Standard licensing skill test vs the Motorcycle Operator Skill Test (MOST II) without training (A) MOST II with
3 hours training (B) or MOST II with 20 hours training (C) The 3 hours training comprised instructions aboutthe skills needed to pass the skill test (30 min theory and 2 frac12 hours practice on the motorcycle) The 20 hours
training consisted of the Motorcycle Safety Foundationrsquos Motorcycle Rider Course (MRC) comprising of 8 hours
theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-road practice
Intervention integrity
25 and 26 obtained their motorcycle licence in the lsquoMOST 3-hour trainingrsquo and lsquoMOST 20-hour trainingrsquo
intervention arm respectively In the lsquoMOST testrsquo intervention arm 29 obtained their licence and in the control
group 35
Outcomes Police recorded crashes and offences
Length of follow up 3 years
32Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3550
Hall 1988 (Continued)
Notes Study conducted in New York USA
Haworth 1997
Methods Population based case-control study
Participants Cases Motorcycle riders or pillion riders involved in motorcycle crash (on road) Recruited from hospital and police
records Not clear whether all cases (especially fatalities) were included
Community controls Multiple matched sample of motorcycle rider or pillion who passed the crash site at the same
time of day and day of week that the crash occurred
Initial sample size
Cases 336
Controls 1195
Analysis sample size
Cases 234 (70)
Controls 561 (47)
Interventions Basic handling skills course beginner intermediate or advanced course measured through self-report not validated
Outcomes Injuries with an Injury Severity Score of 5 or greater or admission for at least four hours and fatalities from a
motorcycle crash (from hospital and police records)
Length of follow up two years or more prior to the interview
Notes Study conducted in Victoria Australia
Hurt 1981
Methods Population based case-control study
Participants Cases Motorcycle riders who had a motorcycle crash that was reported to police (traffic accident reports) Not all
cases included
Community controls Multiple matched sample of motorcycle riders who passed the crash site at the same time of
day and day of week as the crash occurred but 2 years later
Initial sample size
Cases 4500 notifications of crashes to police (only 50 received in time research team able to respond to 1126 for
unreported reasons data collection completed for 900)
Controls Number approached unreported 2310 consented and interviewed
Analysis sample size
Cases 808 (808900 data on rider training) - 18 of initial sample
Controls 670 (6702310 data on rider training) - lt 29 of initial sample
Interventions Training experience self taught friends-family school-club formal-professional others measured through self-
report validated
Outcomes Injuries and fatalities from a motorcycle crash (from hospital records and coronerrsquos office)
33Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3650
Hurt 1981 (Continued)
Notes Study conducted in Los Angeles USA
Jonah 1982
Methods Cohort study
Participants Initial sample size
Intervention group
2310 licensed MTP graduates who are active riders
Control group
2029 licensed motorcyclists who are active riders
Non-licensed and non-active riders excluded
Analysis sample size
Intervention group
811 interviews (35) 533 driver records (23)
Control group 1080 interviews (71) 905 driver records (45)
Interventions No training vs the non-mandatory Canada Safety Council Motorcycle Training Program (MTP) comprising of 4
hours theory and 16 hours practise (6 hrs on-road) over 2 weekends
Intervention integrity
Participants had all successfully completed the MTP
Outcomes Self-reported and police reported crashes and offencesLength of follow up 4 years
Notes Study conducted in Ontario Canada
Kloeden 1994
Methods Cohort study
Participants Initial sample size
Intervention group 2533 licence applicants
Control group Unclear
Sample of riders obtaining probationary licenceIntervention group 1625 (64)
Control group 5015 (unclear)
Non-active riders excluded from control group (~66)
Analysis sample size
Intervention group 631 (39)
Control group 318 (6)
Interventions No formal training vs the mandatory pre-licence Ridersafe course consisting of two levelsthe first is a training course
in the basics of motorcycle riding where a pass entitles the rider to obtain a learnerrsquos permit The second is a skill
test where a pass entitles the rider to obtain a probationary motorcycle licence The training course consists of two 4
hour training sessions on separate days involving instruction videos and actual riding on motorcycles The skill test
consists of a 4 hour preparation course of 74 minutes of lecturediscussion 38 minutes of video and 75 minutes of
34Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3750
Kloeden 1994 (Continued)
on-course riding which is the testing phase
Intervention integrity
Unclear what percentage in the Ridersafe group and the control group obtained their probationary licence halfway
and their full motorcycle licence in the end
Outcomes Police reported crashes
Length of follow up 1 year
Notes Study conducted in SA Australia
Kraus 1975
Methods Population based case-control study
Participants Cases All motorcycle riders with a confirmed medically treated motorcycle injury (including fatalities) from all
hospitals in study base and peripheral area hospitals and all deaths due to motorcycle crashes in the study base All
serious injuries (leading to hospitalisation) and fatalities likely to be included
Community controls A random sample of registered motorcycle owners in the same temporary and geographical
study base as the cases arose from
Initial sample size Cases 1273
Controls 738
Analysis sample size Cases 741 (58)
Controls 434 (59)
Interventions Motorcycle rider training (not specified) measured through self-report
Outcomes Injuries leading to hospitalisation fatalities
Length of follow up not specified
Notes Study conducted in Sacramento county USA
Leung 1987
Methods Non-randomised trial
Participants Initial sample sizeIntervention group unknown number of MRC graduates
Control group unknown number of licensed motorcyclists
Analysis sample size
Intervention group 2941
Control group 43094
Interventions No formal training vs non-mandatory Wisconsin motorcycle rider course a 22 hours off-street programme that
teaches the basic rules of safe motorcycling
Intervention integrity
50 in the MRC group and 100 in the control group obtained their motorcycle licence in the end
35Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3850
Leung 1987 (Continued)
Outcomes Police recorded crashes and offences
Length of follow up up to 3 years
Notes Study conducted in Wisconsin USA
MAIDS 2004
Methods Population based case-control study
Participants Cases Motorcycle riders who had a motorcycle crash Unclear whether all cases included
Community controls Motorcycle riders identified at petrol stations Unclear whether matched to cases
Initial sample size
Cases not reported
Controls not reported
Analysis sample size
Cases 921 (Italy 200 Netherlands 200 Spain 121 Germany 250 France 150)
Controls 923 (Italy 200 Netherlands 200 Spain123 Germany 250 France 150)
Interventions Pre-license rider training or additional training (non-compulsory)
Outcomes Crashes
Notes Study in five sampling areas located in five European countries (France Germany Netherlands Spain and Italy)
McDavid 1989
Methods Cohort study
Participants Sample size after first match
Intervention group 346 male training graduates
Control group 346 male motorcycle riders licensed in same year
Sample size after second match
Intervention group 139 (40)
Control group 139 (40)
Analysis sample size
Intervention group 139 (40)
Control group 139 (40)
Interventions No formal training vs the non-mandatory pre-licence British Columbia Safety Councilrsquos 37-hour motorcycle safety
programme
Intervention integrity
Participants had all completed and passed the training
Outcomes Police reported crashes injurious crashes
Length of follow up 1-6 years
36Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3950
McDavid 1989 (Continued)
Notes Study conducted in British Columbia Canada
Mortimer 1984
Methods Cohort study
Participants Initial sample size
Intervention group 608 MRC graduates
Control group 360 motorcycle riders
Response rates
Intervention group 360 (59)
Control group 324 (90)
Non-active riders excluded
Analysis sample size
Intervention group 213 (35)
Control group 303 (84)
Interventions No formal training vs the non-mandatory Motorcycle Safety Foundationrsquos Motorcycle Rider Course comprising of
8 hours theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-road practice
Intervention integrity
Participants had all completed and passed the MRC
Outcomes Self-reported crashes and offencesLength of follow up 1 year
Notes Study conducted in Illinois USA
Mortimer 1988
Methods Cohort study
Participants Initial sample size
Intervention group 2307 MRC graduates
Control group 556 active licensed motorcycle riders
Response ratesIntervention group 1298 (56)
Control group 500 (90)
Non-active riders excluded
Analysis sample size
Intervention group 913 (40)
Control group 500 (90)
Interventions No formal training vs the non-mandatory Motorcycle Safety Foundationrsquos Motorcycle Rider Course comprising of
8 hours theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-road practice
Intervention integrity
Participants had all completed and passed the MRC
37Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4050
Mortimer 1988 (Continued)
Outcomes Self reported crashes and offences
Length of follow up 1 year
Notes Study conducted in Illinois USA
Osga 1980
Methods Cohort study
Participants Initial sample sizeIntervention group 1163 MSF MRC participants invited 818 (74) responded
Control group 329 motorcycle riders responded to recruitment strategies
Non-active riders excluded from analysis
Analysis sample size
Intervention group 595 (73)
Control group 329 (100)
Interventions No formal training vs the non-mandatory Motorcycle Safety Foundationrsquos Motorcycle Rider Course (MRC) com-
prising of 8 hours theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-road
practice
Intervention integrity
Participants had all successfully completed the MRC
Outcomes Self-reported crashes police recorded crash reports
Length of follow up 6 months - 3 12 years
Notes Study conducted in South Dakota USA
Perrino 2002
Methods Cohort study
Participants Initial sample size
Intervention group 2144 (unclear) MRC graduates (motorcycle licence applicants)
Control group 2601 (unclear) regular licensing test graduates (motorcycle licence applicants)Analysis sample size
Intervention group 1070 (unclear)
Control group 867 (unclear)
Interventions No formal training vs the mandatory (for those younger than 18 years) pre-licence Maryland Motorcycle Rider
Course (MRC) consisting of 8 hours theory and 12 hours practice
Intervention integrity
No information given
Outcomes Self-reported and police recorded crashes and offences
Length of follow up 1 year
38Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4150
Perrino 2002 (Continued)
Notes Study conducted in Maryland USA
Raymond 1979
Methods Cohort study
Participants Initial sample size
Intervention group
258 volunteers mainly (but not all) novice riders
Control group
1104 provisional licence applicants
Analysis sample size
Intervention group 190 (74)
Control group 627 (57)
Interventions No formal training vs the non-mandatory RACACU (Royal Automobile ClubAuto-Cycle Union) training scheme
The training program consisted of either 24 one-hour sessions or 13 two-hour sessions Twelve hours of the training
focused on theory about basic operational procedures traffic laws and highway code The practical session focused
on basic operational skills riding on a training area and riding in traffic with an instructor A proficiency test was
administered at the end of the course
Intervention integrity
No information given
Outcomes Self-reported crashes (including minor non-injury non-damage and off-road crashes)
Length of follow up 1 year
Notes Study conducted in the UK
Satten 1980
Methods Cohort study
Participants Initial sample size
Intervention group interviewers kept interviewing until they had 100 riders selected from a population of 891
enrollees in the MSF MRC course (unknown how many were asked and how many refused to participate)Control group unknown number of registered motorcycle riders
Non-active riders excluded from analysis
Analysis sample size
Intervention group 71
Control group 100
Interventions No formal training vs the non-mandatory Motorcycle Safety Foundationrsquos (MSF) Motorcycle Rider Course (MRC)
comprising of 8 hours theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-
road practice
Intervention integrity
Participants had all successfully completed the MRC
39Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4250
Satten 1980 (Continued)
Outcomes Self-reported crashes injuries and offences
Length of follow up 3 years
Notes Study conducted in Illinois USA
Savolainen 2007
Methods Cohort study
Participants Initial sample size
Intervention group 4000 riders that had conducted the MSF MRC course (passed or failed)
Control group 4000 motorcycle owners or permit or endorsement holders from the Indiana Bureau of Motor
Vehicles database
Analysis sample size
Intervention group 739 (18)
Control group 588 (15)
Interventions Intervention integrity
No information given
Outcomes Probability of crash involvement
Length of follow up up to 5 years
Notes Study conducted in Indiana USA
Swaddiwudhipong 1998
Methods Randomised trial
Participants Initial sample size
Intervention group
all 16 villages of 3 sub-districts
(participants were practising motorcyclists licensed and unlicensed)
Control group
18 villages of 3 other sub-districtsAnalysis sample size
Intervention group 3 districts (100)
Control group 3 districts (100)
Interventions No rider education vs non-mandatory community based rider education consisting of theory on motorcycle injury
risk helmet use traffic laws vehicle regulations and traffic signs and a written and skill test leading to a licence
Intervention integrity
No information given
Outcomes Injuries fatalities
Length of follow up 2 years
40Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4350
Swaddiwudhipong 1998 (Continued)
Notes Study conducted in Tak Province rural Thailand
Waller 1992
Methods Cohort study
Participants Initial sample size
Intervention group 349 riders who succesfully completed the Department of Public Safety motorcycle operator
training in 1988
Control group 349 motorcyclists who never completed the DPS motorcycle operator training
Analysis sample size
Intervention group 349
Control group 349
Interventions No DPS motorcycle operator training vs non-mandatory DPS motorcycle operator training The DPS training is
the Motorcycle Safety Foundationrsquos Motorcycle Rider Course comprising eight hours of theory (lectures discussions
films and multiple choice tests) a textbook and 12 hours off-road practice
Intervention integrity
All 349 riders succesfully completed the rider training
Outcomes Police recorded crashes
Length of follow up 3 years
Notes Study conducted in
Characteristics of excluded studies [ordered by study ID]
Study Reason for exclusion
Anderson 1980a Study type review and process evaluation of a rider course
Aupetit 2008 Outcome proposal of a rider training curriculum that combines traditional training and a motorcycle riding
simulator
Awane 1999 Study type review on use of simulators in rider and driver training
Baldi 2005 Intervention not rider training but intervention integrity of rider training availabilty of rider education
implementation of graduated licensing system etc on a state-wide level
Batchler 1988 Study type description of a training course
Blanchard 2006 Study type not a study assessing the association between rider training and outcomes
Bonnett 2005 Outcome qualitative data regarding satisfaction with the training programme
41Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4450
(Continued)
Braver 2007 Outcome licensure rates
Clayton 1980 Outcome knowledge skills attitude
Council 1978 Outcome crash reduction in ANY vehicle not motorcycles exclusively
Crowther 2005 Study type qualitative study
Daniello 2009 Study type review of the effectiveness of motorcycle training and licensing
Dewanti 2007 Intervention emotional characteristics behaviour (ie not motocycle rider training)
Du Plooy 1990 Study type review of types of rider training courses in South Africa
Engel 1989 Study type uncontrolled before-after study
Goldenbeld 2002 Outcome skills
Goldenbeld 2004 Outcome knowledge and skills
Haworth 2000 Intervention balance between attitudinal andvehicle control skills acrosscurriculla of motorcycle rider training
courses
Haworth 2005 Study type review
Haworth 2006 Study type cross-sectional survey
Hill 1983 Study type review and description of training system
Imatake 1980 Study type course descriptionreview
Ingebritsen 1990 Intervention risk factors for motorcycle crashes
Jain 2009 Study type cross-sectional descriptive study of motorcycle crashes over a 5 year period
Jonah 1979 Study type cross-sectional survey Outcome scores on motorcycle operator skill test
Jonah 1981 Intervention evaluation of a motorcycle rider skill test not motorcycle rider training
Kadar 1990 Study type review of components of rider training courses
Kelsey 1986 Intervention evaluation of a motorcycle rider skill test not motorcycle rider training
Koch 1980 Study type reviewcourse description
Koch 1990 Study type description of course content
42Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4550
(Continued)
Labbett 2005 Study type essay about possible explanations for motorcycle deaths
Law 2005 Intervention targeted motorcycle safety programme not specifically rider training
Lowes 1990 Study type review and descriptions of rider training
Mahon 1976 Study type review of motorcycle rider training methods
Maxwell 1985 Outcomesrecommendations forprogramme operatorsand otherson howthe motorcycle trainingprogramme
works and how it can be improved
Mayhew 1996 Study type review
McKnight 1987 Study type evaluation of methodological and administrative aspects of a motorcycle operator skill test
McPherson 1976 Outcome knowledge and skills
Mohan 2004 Study type review of evidence of interventions to reduce road traffic injuries
Morgan 1980 Study type reviewcourse description
Motoki 1990 Study type proposal and development of rider training curriculum
Nairn 1992 Study type literature review
Newland 1980 Study type description of development of motorcycle safety course
Newland 1982 Study type descriptionreview of a rider training course
Newland 1990 Study type uncontrolled BeforeAfter study
Ormston 2003 Outcome changes in attitude and self reported behaviour
Osga 1979 Outcome knowledge attitudes and skills
Panou 2007 Intervention new simulation tool to teach driverrider
Peterson 1990 Study factor demographic details of course participants completing the course vs those not completing the
course
Prem 1984 Study type experimental
Study factor examines association between skills and performance in tests
Radin Umar 2006 Intervention motorcycle safety programme not specifically rider training
Rockwell 1990 Study type methodologically flawed study type
43Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4650
(Continued)
Rothe 1987 Study type qualitative study of rider training in general
Rowden 2007 Study type qualitative study
RSC Victoria 1993 Study type review of motorcycle rider training and education in Victoria
Russam 1979 Study type review of training programmes in Japan
Saffron 1981 Study type literature review
Schulz 1990 Study factor predictors of motorcycle crashes
Sexton 2004 Study type cross-sectional analytic analysis
Shephard 1986 Study factor reviewdescription of programme
Simpson 1990 Study type review of motorcycle rider trainingeducation
Sokolov 1990 Study factor review of motorcycle riding courses in USSR
Steffens 1988 Outcome self-reported behaviour experiences and researcher observed riding behaviour
Sudlow 2003 Study type review of motorcycle training in the UK
Tasmania 1995 Study type cross-sectional descriptiveprocess evaluation
Tetard 1985 Study type experimental
Outcome detection of indices
Thackray 1980 Outcome instructional effectiveness user acceptance administrative feasibility
Thompson 1994 Study type cross-sectional survey
Outcome attitudes and knowledge
Tobler 1990 Study type review of course curricula
Transport 1989 Intervention syllabus for rider training
Tuan 2007 Study type cross-sectional study (survey) which does not assess the effects of an intervention but examines
motocyclistsrsquo behaviour
Victor 1980 Study type review
Waterer 1990 Study type review of UK rider training issues
Weller 1989 Study type review of studies evaluating motorcycle safety and education measures
44Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4750
(Continued)
Wells 1981 Study type description of different methods of training
Winn 1986 Study type review of rsquohistoryrsquo of rider training
Wisher 1988 Study type cross-sectional survey
Wood 1987 Study type Interrupted Time Series but did not meet the criteria for an ITS as published by the Cochrane
Effective Practice and Organisation of Care Group
Yamazaki 1990 Study type cross-sectional survey
Outcome riding skills skills retention and attitude
Characteristics of ongoing studies [ordered by study ID]
Burgess 2010
Trial name or title UKs National Ride program
Methods Cohort study
Participants Motorcycle offenders
Interventions Police led diversion program
Outcomes Self reported attitudinal and behavioural measures
Starting date
Contact information
Notes Expected to be completed in early 2010
45Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4850
D A T A A N D A N A L Y S E S
This review has no analyses
A P P E N D I C E S
Appendix 1 Search strategy
We searched the following electronic databases
bull Cochrane Injuries Group Specialised Register (searched 22 Sept 2008)
bull CENTRAL (The Cochrane Library 2008 Issue 3)
bull TRANSPORT 1988 to 200706 (latest date available)
bull ATRI Australian Transport Index (to Sept 2008)
bull MEDLINE (Ovid SP) 1950 to Sept 2008
bull EMBASE (Ovid SP) 1980 to Sept 2008
bull CINAHL (EBSCO) 1982 to Sept 2008
bull WHOLIS (World Health Organization Library Information System) (to Sept 2008)
bull PsycINFO (Ovid SP) 1806 to Sept 2008
bull LILACS (BIREME) (Latin American and Caribbean Health Sciences) (1982 to Sept 2008)
bull ISI Web of Science Social Sciences Citation Index (SSCI) 1970 to Sept 2008
bull ERIC (searched 25 Sept 2008)
bull ZETOC (searched 25 Sept 2008)
bull SIGLE (searched 25 Sept 2008)
All of the databases were searched using the following text-word search strategy The strategy was adapted using either rsquorsquo or rsquo$rsquo as atruncation symbol
1 motor-cycl or motorcycl or motor-bik or motorbik or motor-scooter or motorscooter or scooter or moped or motocycl
or motocicl or moto
2 train or educat or instruct or program or course or qualif or learn or improv
3 1 and 2
H I S T O R Y
Protocol first published Issue 2 2005
Review first published Issue 10 2010
46Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4950
C O N T R I B U T I O N S O F A U T H O R S
Katina Kardamanidis was the first author of the protocol conducted the literature search extracted full studies assessed the studies
according to inclusion criteria assessed quality of the studies compared findings with the second author discussed disagreements with
the third author wrote the drafts and the final review
Rebecca Ivers was the second author of the protocol assessed the studies according to inclusion criteria assessed quality of the studies
compared findings with the first author edited the final draft and provided overall guidance for the process
Alexandra Martiniuk assessed quality of the studies compared findings with the first author discussed disagreements with the third
author and helped write the first draft and the final review
Mark Stevenson contributed to the protocol reviewed methods of studies and contributed to the final draft
Katrina Thistlethwaite conducted a second literature search extracted full studies and contributed to the final draft
D E C L A R A T I O N S O F I N T E R E S T
None known
S O U R C E S O F S U P P O R T
Internal sources
bull No sources of support supplied
External sources
bull National Health and Medical Research Council of Australia Australia
Salary support for Martiniuk Ivers and Stevenson
bull University of Sydney Sesqui Research Grant Australia
Direct research costs and salary support for Kardamanidis
D I F F E R E N C E S B E T W E E N P R O T O C O L A N D R E V I E W
There are no differences between the protocol and the review
I N D E X T E R M S
Medical Subject Headings (MeSH)
lowastMotorcycles Accident Prevention [lowastmethods] Accidents Traffic [lowastprevention amp control] Licensure Program Evaluation Wounds
and Injuries [mortality lowastprevention amp control]
47Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 5050
MeSH check words
Humans
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 350
[Intervention Review]
Motorcycle rider training for the prevention of road trafficcrashes
Katina Kardamanidis1 Alexandra Martiniuk 1 Rebecca Q Ivers1 Mark R Stevenson1 Katrina Thistlethwaite1
1The George Institute for Global Health Sydney Australia
Contact address Rebecca Q Ivers The George Institute for Global Health PO Box M201 Missenden Road Sydney NSW 2050
Australia riversgeorgeorgau
Editorial group Cochrane Injuries GroupPublication status and date New published in Issue 10 2010
Review content assessed as up-to-date 8 August 2010
Citation Kardamanidis K Martiniuk A Ivers RQ Stevenson MR Thistlethwaite K Motorcycle rider training for the
prevention of road traffic crashes Cochrane Database of Systematic Reviews 2010 Issue 10 Art No CD005240 DOI
10100214651858CD005240pub2
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
A B S T R A C T
Background
Riding a motorcycle (a two-wheeled vehicle that is powered by a motor and has no pedals) is associated with a high risk of fatal crashes
particularly in new riders Motorcycle rider training has therefore been suggested as an important means of reducing the number of
crashes and the severity of injuries
Objectives
To quantify the effectiveness of pre- and post-licence motorcycle rider training on the reduction of traffic offences traffic crash
involvement injuries and deaths of motorcycle riders
Search strategy
We searched the Cochrane Injuries Group Specialised Register CENTRAL (The Cochrane Library 2008 Issue 3) TRANSPORT
MEDLINE EMBASE CINAHL WHOLIS (World Health Organization Library Information System) PsycInfo LILACS (LatinAmerican and Caribbean Health Sciences) ISI Web of Science Social Sciences Citation Index (SSCI) ERIC ZETOC and SIGLE
Database searches covered all available dates up to October 2008 We also checked reference lists of relevant papers and contacted study
authors in an effort to identify published unpublished and ongoing trials related to motorcycle rider training
Selection criteria
We included all relevant intervention studies such as randomised and non-randomised controlled trials interrupted time-series and
observational studies such as cohort and case-control studies
Data collection and analysis
Two review authors independently analysed data about the study population study design and methods interventions and outcome
measures as well as data quality from each included study and compared the findings We resolved differences by discussion with a
third review author
1Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 450
Main results
We reviewed 23 studies three randomised trials two non-randomised trials 14 cohort studies and four case-control studies Five
examined mandatory pre-licence training 14 assessed non-mandatory training three of the case-control studies assessed lsquoanyrsquo type of
rider training and one case-control study assessed mandatory pre-licence training and non-mandatory training The types of assessed
rider training varied in duration and content
Most studies suffered from serious methodological weaknesses Most studies were non-randomised and controlled poorly for con-
founders Most studies also suffered from detection bias due to the poor use of outcome measurement tools such as the sole reliance
upon police records or self-reported data Small sample sizes and short follow-up time after training were also common
Authorsrsquo conclusions
Due to the poor quality of studies identified we were unable to draw any conclusions about the effectiveness of rider training on crash
injury or offence rates The findings suggest that mandatory pre-licence training may be an impediment to completing a motorcycle
licensing process possibly indirectly reducing crashes through a reduction in exposure It is not clear if training (or what type) reducesthe risk of crashes injuries or offences in motorcyclists and a best rider training practice can therefore not be recommended As some
type of rider training is likely to be necessary to teach motorcyclists to ride a motorcycle safely rigorous research is needed
P L A I N L A N G U A G E S U M M A R Y
Motorcycle rider training for preventing road traffic crashes
Riders of motorcycles (a two-wheeled vehicle that is powered by a motor and has no pedals - Oxford English Dictionary Online)
especially novice riders have an increased risk of being involved in fatal crashes compared to other road users Motorcycle rider training
could be an important way of reducing the number of crashes and the severity of injuries
The authors of this review examined all research studies that report an evaluation of the effectiveness of motorcycle rider coursesin reducing the number of traffic offences motorcycle rider crashes injuries and deaths This review included 23 research studies
including three randomised trials two non-randomised trials 14 cohort studies and four case-control studies The types of rider training
that were evaluated varied in content and duration The findings suggest that mandatory pre-licence training may present a barrier to
completing a motorcycle licensing process thus possibly indirectly reducing crash injury death and offence rates through a reduction
in exposure to riding a motorcycle However on the basis of the existing evidence it is not clear if (or what type of) training reduces
the risk of crashes injuries deaths or offences in motorcyclists and the selection of the best rider training practice can therefore not be
recommended
It is likely that some type of rider training is necessary to teach motorcyclists basic motorcycle handling techniques and to ride a
motorcycle safely It is therefore important that further research work be conducted to rigorously evaluate motorcycle rider training
courses particularly in low income countries where the main burden of motorcycle injuries and deaths occur
B A C K G R O U N D
Each year anestimated 12million peopleare killedin road crashes
worldwide and an additional 50 million people are injured Of
these deaths 85 occur in low-income and middle-income coun-
tries It has been predicted that between the year 2000 and 2020
without increased efforts the number of road traffic deaths in
high-income countries will decrease by approximately 30 yet
in low- and middle-income countries the number is expected to
increase by over 80 (Kopits 2003 Peden 2004) Road traffic
injuries are likely to become the third leading cause of a global
burden of disease and injury by 2020 if further action is not taken
(Peden 2004)
Riding a motorcycle (a two-wheeled vehicle that is powered by a
motor and has no pedals - Oxford English Dictionary Online) is
associated with a higher risk of fatalities compared with driving a
car (Leung 1983 Bjornstig 1985 Radin Umar 1995 Tsai 1995
FORS 1999 Yuan 2000 Johnston 2008) In Australia motorcy-
2Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 550
clesaccounted for45 of allAustralian passenger vehicle registra-
tions and 09 of vehicle kilometres travelled in 2007 Howevermotorcycle riders accounted for approximately 15 of all road
crash deaths and an even higher proportion of serious injuries
Per distance travelled the Australian rate of motorcyclist deaths
was approximately 30 times the rate for car occupants The cor-
responding rate for a serious injury was approximately 41 times
higher Similar elevated rates were also found in other developed
countries (Johnston 2008) In low and middle-income countries
the motorcycle is one of the most frequently used means of trans-
port and consequently motorcyclists constitute a large proportion
of road crash victims (Mohan 2002)
The lack of experience of many motorcycle riders has been cited
as one potential cause of motorcycle crashes (Winn 1987 Mullin1997) Training of motorcycle riders has therefore been suggested
as an important countermeasure for reducing both the number of
crashes and the severity of injury (Vis 1995 Noordzij 2001)
Pre-licence motorcycle rider trainingcourses have been made com-
pulsory in several countries (Mullin 1997) Pre-licence training
courses are aimed at providing motorcycle riders with the skills
and knowledge needed to improve control of the motorcycle and
to identify and avoid critical situations in traffic (Elliot 2003)
Many pre-licence motorcycle rider training courses are aimed at
novice riders and in many jurisdictions in high-income countries
such as Australia the USA and Canada the completion of such
courses is a mandatory requirement of licensing
Post-licence rider training is not linked with licensing and such
training is generally not compulsory It usually focuses on riding
skills and other topics that have not been covered or dealt with
sufficiently in pre-licence rider training Post-licence rider train-
ing is used to reinforce fundamental skills and knowledge or to
provide some training for motorcyclists who are self-taught and
have never received formal training (Anderson 1980)
Estimates of the number of motorcyclists attending post-licence
rider training are not readily available but existing data suggest
such courses are not well attended although attendance varies by
country In the USA 24 of the total number of motorcyclists
trained in 1988 and 1989 attended an experienced rider course
(Weaver 1990)
The aim of this systematic review is to quantify the effectiveness
of pre-and post-licence rider training for motorcycle riders in re-
ducing traffic offences road traffic crash involvement injury and
death
O B J E C T I V E S
To quantify the effectiveness of pre- and post-licence motorcycle
rider training on the reduction of traffic offences traffic crash
involvement injuries and deaths of motorcycle riders
M E T H O D S
Criteria for considering studies for this review
Types of studies
All relevant intervention studies such as randomised and non-ran-
domised controlled trials and interrupted time series studies and
observational studies including cohort and case-control studies
Types of participants
Motorcyclists and riders of mopeds or motor scooters of all ages
Types of interventions
Motorcycle ridertraining (pre-or post-licence mandatory or non-
mandatory) versus no training
One form of pre-licence rider training versus another form (eg
on-road off-road theory or a combination)
One form of post-licence rider training versus another form (eg
on-road off-road theory or a combination)
Types of outcome measures
Offences crashes injuries and death
Search methods for identification of studies
We did not restrict the searches by date language or publication
status
Electronic searches
We searched the following electronic databases
bull Cochrane Injuries Group Specialised Register (searched 22
Sept 2008)
bull CENTRAL (The Cochrane Library 2008 Issue 3)bull TRANSPORT 1988 to 200706 (latest date available)
bull ATRI Australian Transport Index (to Sept 2008)
bull MEDLINE (Ovid SP) 1950 to Sept 2008
bull EMBASE (Ovid SP) 1980 to Sept 2008
bull CINAHL (EBSCO) 1982 to Sept 2008
bull WHOLIS (World Health Organization Library
Information System) (to Sept 2008)
bull PsycINFO (Ovid SP) 1806 to Sept 2008
bull LILACS (BIREME) (Latin American and Caribbean
Health Sciences) (1982 to Sept 2008)
bull ISI Web of Science Social Sciences Citation Index (SSCI)
1970 to Sept 2008
3Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 650
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 750
Table 1 Risk of selection bias (Continued)
Osga 1980 Differences at baseline reported on sex and age
Confounding factors were not dealt with in univariate analysis Riding exposure was accounted for in the
cumulative survival analysis
Selection intervention group were all training participants on the register and the control group were
respondents to a diverse range of recruitment strategies
Satten 1980 Differences at baseline reported on age sex helmet use riding experience and exposure and previous
training
Confounding factors were not dealt with in design or analysis
Selection intervention group was selected from a list of training graduates and the control group was
partly selected from a state register of motorcyclists
Jonah 1982 Differences at baseline reported on age sex riding experience time licensed previous training and alcohol
use
Confounding factors were dealt with through multivariate analysis
Selection intervention group was selected from a list of training graduates and the control group was
randomly selected from the Ontario motorcycle operator file
Mortimer 1984 Differences reported at baseline on age licence status riding experience years licensed and helmet use
Confounding factors age years licensed and rider experience were dealt with through stratification and
riding exposure through standardisation
Selection intervention group was selected from a list of training enrolees and the control group was
selected from motorcycle dealerships during three successive Saturdays
Cooper 1988 Differences reported at baseline on age gender licensure and highway experience
Confounding factors age gender riding exposure rider characteristics rider attitudes and behaviour were
accounted for in a multivariate analysis
Selection there was no information on how the intervention and control groups were selected and from
where
Mortimer 1988 Differences reported at baseline on years licensed riding experience and helmet use
Confounding factors age years licensed and rider experience were dealt with through stratification and
standardisation of riding exposure A regression analysis was also conducted but poorly reported and not
usable
Selection intervention group was selected from a list of training enrolees and the control group was
selected from motorcycle dealerships
Adams 1985 Differences at baseline reported on previous crash rates
Confounding factors not controlled for in analysis
Selection the intervention group was a non-random sample of those offered training within the Australia
Post motorcycle delivery riders the control group were the remaining Post riders
Leung 1987 Differences at baseline reported on age and sex
Confounding factors age and sex controlled for in analysis by stratification
Selection those exposed to training were MRC graduates those unexposed were taken from the licensing
records
5Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 850
Table 1 Risk of selection bias (Continued)
McDavid 1989 Differences at baseline other than matched characteristics were not reported
Confounding factors age ownership of car licence years licensed previous crashesand offences accounted
for through matching Other confounding effects not controlled
Selection intervention group was selected from a list of training graduates and the control group was
selected from Bureau of Motor Vehicle records
Kloeden 1994 Differences at baseline reported were in geographical area of residence and in riding exposure
Confounding factors were accounted for through exclusion of participants from rural areas
Other confounding factors were not reported
Selection the Ridersafe course was phased in based on postcode of riderrsquos home address analyses were
based on a database of those who obtained motorcycle rider learnerrsquos permits of which about 2000 were
Ridersafe riders and about 5000 were control riders
Billheimer 1998 Differences at baseline reported on age and previous riding experience
Other potential confounding factors assessed were riding exposure primary purpose of riding and helmet
use
Confounding factors accounted for by matching (age sex primary purpose of riding) standardisation
(riding exposure) and stratification (previous riding experience)
Selection the intervention group were those who had taken the rider training programme the control
group were California motorcycle riders matched on the above variables but who had not taken a riding
course
Perrino 2002 Differences at baseline reported were a higher number of females in the trained group and a higher
proportion of experienced riders in the untrained group
Confounding factors were not accounted for in the analysis
Selection the intervention group were those who had done the Motorcycle Riding Course the control
group had not done the course Both groups were identified through a register of licensed motorcyclists
Kraus 1975 Population based case-control study
Cases all motorcycle riders with a confirmed medically treated motorcycle injury (including fatalities)
from all hospitals in study base and peripheral area hospitals and all deaths due to motorcycle crashes in
the study base All serious injuries (leading to hospitalisation) and fatalities likely to be included
Community controls a random sample of registered motorcycle owners in the study base
Potential confounders assessed were age sex weight height riding experience motorcycle use charac-
teristics of motorcycle and previous crashes and offences
Confounding factors were accounted for through exclusion of females and stratification by age and
motorcycle useSelection cases and controls drawn from the same population
Hurt 1981 Population based case-control study
Cases and controls not drawn from the same population
Cases motorcycle riders who had a motorcycle crash Not all reported cases included
Community controls motorcycle riders who passed the crash site at the same time of day and day of week
that the crash occurred but 2 years later
Potential confounders assessed were age sex time type of motorcycle roadway alignment helmet use
weather conditions road surface conditions
Confounding factors were accounted for through matching for time of day and day of week that crash
occurred and stratification
Selection cases were recruited from police reports not all cases were included controls were matched
6Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 950
Table 1 Risk of selection bias (Continued)
community controls from randomly selected sites matched for same time of dayday of week as the crash
occurred
Haworth 1997 Population based case-control study
Cases and controls drawn from the same population
Cases motorcycle riders or pillion riders with an injury from a motorcycle crash from hospital and police
records Not clear whether all cases were included
Community controls motorcycle rider or pillion who passed the crash site at the same time of day and
day of week that the crash occurred
Potential confounders assessed were rider factors age marital status licence status riding experience
and exposure alcohol and drugs use of protective gear including helmet use pillion factors motorcycle
factors trip factorsConfounding factors were accounted for through matching for time of day and day of week that crash
occurred and stratification for age blood alcohol level and licence status
Selectioncases were riders or pillions involved in on-road crash recruited from hospitals and police records
not sure if all cases were included controls were matched community controls who passed the crash site
at the same time of dayday of week as the crash
Savolainen 2007 Differences at baseline reported on sex safety helmet use riding ability riding experience and licence
status
Adjusted for these confounders through statistical modelling although exact process unclear
Selection intervention group from list of trainees control group from Indiana Bureau of Motor Vehicles
database Trainees and the control group were invited to participate in surveys
Davis 1997 Differences at baseline not reported but likely No adjustment for any confounders at baseline or during analysis
Selection intervention group from list of riders involved in crashes who were also on a list of training
graduates control group from the same list of riders involved in crashes who were not also on a list of
training graduates
Waller 1992 Differences at baseline not reported but likely
Adjusted for confounders through exclusion of riders younger than 18 years riders without a motorcycle
operator driver license riders with location of residence outside of the Texas metropolitan areas of Austin
through matching by age sex and location of residence through stratification by age and sex and through
standardisation (number of crashes by person-days at risk)
Selection intervention group from the Motorcycle Operator Training class roster (number of trained
riders that the sample was selected from is unclear) control group through a random number generator
generating driver license numbers until an equal sample size was selected as the intervention group
MAIDS 2004 Population based case-control study
Unclear whether cases and controls not drawn from the same population
Cases motorcycle riders who had a motorcycle crash Unclear whether all cases included
Community controls motorcycle riders recruited at petrol stations Unclear whether and how matched
to cases
No potential confounders were reported
There was no report of whether and how potentially confounding factors were accounted for in the study
7Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 1050
Table 2 Risk of allocation bias
Anderson 1980 Allocation concealment unclear
Hall 1988 Allocation concealment unclear
Swaddiwudhipong 1998 Allocation concealment unclear
Table 3 Risk of detection bias
Anderson 1980 Blinding of outcome assessors not reported
Driver records holding police recorded crashes and offences not validated
Hall 1988 Blinding of outcome assessors not reported
Driver records holding police recorded crashes and offences not validated
Swaddiwudhipong 1998 Blinding of outcome assessors not reported
Police and hospital records Post mortem examination for deaths not validated
Adams 1985 Blinding of outcome assessors not reported
Australia Post crash records not validated but argued that underreporting is unlikely
Leung 1987 Blinding of outcome assessors not reported
Driver records holding police recorded crashes and offences
Raymond 1979 Blinding of outcome assessors not reported
Self-report through personal interviews or questionnaires - both methods validated
Osga 1980 Blinding of outcome assessors not reported
Self-report through structured questionnaires validated by cross-checking with police
recorded crash records
Satten 1980 Blinding of outcome assessors not reported
Self-report through structured interviews or questionnaires - both methods were not
validated
Intervention group were given telephone interviews but the control group mailed ques-
tionnaires
Jonah 1982 Blinding of outcome assessors not reported
Self-report through structured interviews not validated Driver records holding police
recorded crashes and offences
Mortimer 1984 Blinding of outcome assessors not reported
Self-report through questionnaires not validated
Intervention group received mailed questionnaires but the control group received ques-
tionnaires personally at the motorcycle dealership
Cooper 1988 Blinding of outcome assessors not reported
Driver records holding police recorded crashes and offences
8Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 1150
Table 3 Risk of detection bias (Continued)
Mortimer 1988 Blinding of outcome assessors not reported
Self-report through questionnaires not validated
Intervention group received mailed questionnaires but the control group received ques-
tionnaires personally at the motorcycle dealership
McDavid 1989 Blinding of outcome assessors not reported
Driver records holding police recorded crashes and offences
Kloeden 1994 Blinding of outcome assessors not reported
Driver records holding police recorded crashes and offences
Billheimer 1998 Blinding of outcome assessors not reported
Police recorded crashes and offences
Perrino 2002 Blinding of outcome assessors not reported
Self-report through mailed out questionnaires police records not validated
Kraus 1975 Blinding of intervention assessors to case or control status not reported
Intervention measured police records self-report though questionnaires not validated
Length of follow up not specified
Hurt 1981 Blinding of intervention assessors to case or control status not reported
Intervention measured face-to-face in-depth interviews (ie no structured question-
naires) Validity and reliability of data ensured by reported high level of quality controlLength of follow up not specified
Haworth 1997 Blinding of intervention assessors to case or control status not reported
Intervention measured self-report either by personal or by telephone interview for
both cases and controls not validated
Length of follow up in the 2 years prior to the interview or more than 2 years ago
Savolainen 2007 Blinding of outcome assessors not reported
Source of outcome data not reported
Length of follow up up to 5 years
Davis 1997 Blinding of outcome assessors not reported
Source of outcome data police records
Length of follow up 1 year
Waller 1992 Blinding of outcome assessors not reported
Source of outcome data Department of Public Safety driver history file and official
accident reports
Length of follow up 3 years
MAIDS 2004 Blinding of intervention assessors to case or control status not reported
Intervention measured face-to-face in-depth interviews (ie no structured question-
naires) No report on whether and how validity and reliability of data was ensured
Length of follow up 2 years
9Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 1250
Table 4 Risk of attrition bias
Anderson 1980 Percentage of participants in analysis 99 forthe total group Unclear forintervention andcontrol groups
separately
Hall 1988 Percentage of participants in analysis 96-97 forall intervention groups and 98for the control group
Swaddiwudhipong 1998 Percentage of participating sub-districts in analysis 100 for both intervention and control sub-districts
Raymond 1979 Percentage of participants in analysis 74 for the intervention group and less than 57 for the control
group
Osga 1980 Percentage of participants in analysis 73 for the intervention group and 100 for the control group
Satten 1980 Percentage of participants in analysis is unknown Initial sample size is unclear
Jonah 1982 Percentage of participants in analysis 23 for the intervention group and 45 for the control group
Mortimer 1984 Percentage of participants in analysis 35 for the intervention group and 84 for the control group
Cooper 1988 Percentage of participants in analysis 98 for the total group Unclear for intervention and control
groups separately
Mortimer 1988 Percentage of participants in analysis 40 for the intervention group and 90 in the control group
Adams 1985 Percentage of participants in analysis 61 forthe total group Unclear forintervention andcontrol groups
separately
Leung 1987 Percentage of participants in analysis unclear because initial sample size was not reported
McDavid 1989 Percentage of participants in analysis 40 of final matched cohort
Kloeden 1994 Percentage of participants in analysis 39 for the intervention group and 6 for the control group
Billheimer 1998 Percentage of participants in analysis unclear percentage of the initial sample size
Perrino 2002 Percentage of participants in analysis unclear percentage of the initial sample size
Kraus 1975 Percentage of participants in analysis 58 of cases 59 of controls
Hurt 1981 Percentage of participants in analysis 18 of cases lt 29 of controls
Haworth 1997 Percentage of participants in analysis 70 of cases 47 of controls
Savolainen 2007 Percentage of participants in analysis 18 for the intervention group and 15 for the control group
Davis 1997 Percentage of participants in analysis 100 for the intervention group and unclear for the control group
10Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 1350
Table 4 Risk of attrition bias (Continued)
Waller 1992 Percentage of participants in analysis unclear percentage for the intervention group and 100 for the
control group
MAIDS 2004 Percentrage of participants in analysis unclear for both cases and controls
R E S U L T S
Description of studies
See Characteristicsof included studies Characteristics of excluded
studies Characteristics of ongoing studies
Our search resulted in 102 studies that seemed relevant of which
we excluded 79 studies because they did not meet the criteria for
considering studies for this review (see Characteristics of excluded
studies)
Basedon ourcriteriawe includeda total of 23 studies in this review
There were three randomised trials (Anderson 1980 Hall 1988
Swaddiwudhipong 1998) two non-randomised trials (Adams
1985 Leung 1987) 14 cohort studies (Raymond 1979 Osga
1980 Satten 1980 Jonah 1982 Mortimer 1984 Cooper 1988
Mortimer 1988 McDavid 1989 Waller 1992 Kloeden 1994
Davis 1997 Billheimer 1998 Perrino 2002 Savolainen 2007)
and four case-control studies (Kraus 1975 Hurt 1981 Haworth
1997 MAIDS 2004)
Over half of the studies were conducted 20 or more years
ago and the oldest was conducted 31 years ago Only three
studies were from this decade (Perrino 2002 MAIDS 2004
Savolainen 2007) Over 60 of the studies were conducted in
the USA (Kraus 1975 Anderson 1980 Osga 1980 Satten 1980
Hurt 1981 Mortimer 1984 Leung 1987 Hall 1988 Mortimer
1988 Waller 1992 Davis 1997 Billheimer 1998 Perrino 2002
Savolainen2007) three inAustralia (Adams 1985 Haworth 1997
Kloeden 1994) three in Canada (Jonah 1982 Cooper 1988
McDavid 1989) onefromthe United Kingdom (Raymond 1979)one from Europe (with sampling areas located in France Ger-
many Netherlands Spain and Italy) (MAIDS 2004) and one
in Thailand (Swaddiwudhipong 1998) Only seven studies were
published in peer reviewed journals (Kraus 1975 Jonah 1982
Mortimer 1984 Cooper 1988 Mortimer 1988 McDavid 1989
Swaddiwudhipong 1998) The other studies were reports forhigh-
way safety and road traffic organisations
The studies by Anderson 1980 Hall 1988 Kloeden 1994
Billheimer 1998 (the basic California Motorcyclist Safety Program
component of the study) and Perrino 2002 examined manda-
tory pre-licence training Cooper 1988 McDavid 1989 and
Swaddiwudhipong 1998 assessed non-mandatory training that
when passed resulted in a licence Raymond 1979 Osga 1980
Satten 1980 Jonah 1982 Mortimer 1984 Adams 1985 Leung
1987 Mortimer 1988 Waller 1992 Davis 1997 Billheimer 1998
(the Experienced Rider Course component of the study) andSavolainen 2007 assessed non-mandatory training that could be
taken by motorcycle riders with and without a licence The case-
control studies (Kraus 1975 Hurt 1981 Haworth 1997) assessed
lsquoanyrsquo type of rider training and therefore could not be grouped by
pre- or post-licence rider training The MAIDS 2004 case-control
study did make a distinction between participants having received
compulsory pre-licence training or additional (non-compulsory)
training We did not find a study that compared one form of pre-
or post-licence training (eg on-road off-road theory or a com-
bination) versus another form of training All studies compared
eithermandatory or non-mandatory formal training versus no for-
mal training with the exception of Haworth 1997 who compared
different levels of training courses (beginner intermediate and ad-vanced)
We have presented a more detailed description of the individual
studies in Characteristics of included studies We have provided
a description below of the types of motorcycle rider training that
were evaluated in the included studies
Types of motorcycle rider training evaluated in the
studies
The types of training were varied Duration of training ranged
from two or three hours to 24 hours administered over two or
three days Most training consisted of a theoretical and a practical
component usually off-road The content ranged from provid-ing riders with manuals lectures discussions showing films and
administering knowledge and skill tests Little information was
provided about the qualifications of trainers who conducted the
training
Six studies (Osga 1980 Satten 1980 Mortimer 1984 Mortimer
1988 Waller 1992 Savolainen 2007) assessed the effectiveness of
the USA-based Motorcycle Safety Foundationrsquos Motorcycle Rider
Course (MSF MRC) This course was released in 1976 It com-
prises eight hours of theory (lectures discussions films and multi-
plechoice tests) a textbook and 12 hours of off-road practice The
Motorcycle Safety Foundation is a national not-for-profit organ-
11Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 1450
isation sponsored by manufacturers and distributors of motorcy-
cles Many USA states have adopted the MSF Basic Rider Courseand Experienced Rider Course (wwwmsf-usaorg) since its release
in 1976
The Motorcycle Operators Skill Test was also developed
by the USA Motorcycle Safety Foundation in coopera-
tion with others (httpwwwwebbikeworldcommotorcycle-
trainingmotorcycle-test) The MOST is a licensing test consist-
ing of a manual knowledge and skill test (comprising eight exer-
cises of increasing difficulty and skill requirements) and a mini-
skill test conducted with a simulator Some studies compared this
alternative licensing test to the standard licensing test However
as the MOST is not a type of rider training we have included only
the two studies that assessed the MOST in combination with one
or more training components in this review Both Anderson 1980and Hall 1988 assessed the MOST in combination with a train-
ing component of three hoursrsquo duration consisting of instructions
about the skills needed to pass the MOST (30 minutes of the-
ory and two and a half hours on the motorcycle) Hall 1988 also
assessed the MOST in combination with the Motorcycle Safety
Foundationrsquos Motorcycle Rider Course as described above
Other USA studies evaluated the Wisconsin motorcycle rider
course a 22-hour off-street programme that teachesthe basic rules
of safe motorcycling (Leung 1987) the USA Maryland Motorcy-
cle Rider Course (MRC) a mandatory (for those younger than 18
years)pre-licence 20-hour coursecomprising eight hours of theory
and12 hours of practice and created by the Maryland Motor Vehi-
cle Administration in 1985 (Perrino 2002) the California Motor-cyclist Safety Program (CMSP) a mandatory (for those younger
than 18 years) pre-licence 16-hour basic course (Billheimer 1998)
and Experienced Rider Course (ERC) of eight hoursrsquo duration
(Billheimer 1998) The CMSP still offers these courses ( http
wwwca-msporg) but they are now MSF courses as described
above Davis 1997 did not give any information about the assessed
rider training
The Canadian studies evaluated the British Colombia Safety
Councilrsquos 37-hour motorcycle safety programme (no further in-
formation on the content was provided) (McDavid 1989) and
the Canada Safety Council Motorcycle Training Program (MTP)
comprising four hours of theory and 16 hours of practice (six
hours on-road) over twoweekends (Jonah 1982) The MTPcoursewas made available in 1974 and is still being offered in Canada
(httpwwwsafety-councilorgtrainingmtpmtphtm) Cooper
1988 did not give any information about the assessed rider train-
ing However the study was conducted in British Colombia in
Canada in 1988 the same time as the McDavid 1989 study was
conducted
In the United Kingdom Raymond 1979 assessed the non-manda-
tory RACACU (Royal Automobile ClubAuto-Cycle Union)
training scheme The training programme consisted of either 24
one-hour sessions or 13 two-hour sessions Twelve hours of the
training focused on theory about basic operational procedures
traffic laws and highway codes The practical session focused on
basic operational skills riding on a training area and riding in traf-fic with an instructor A proficiency test was administered at the
end of the course
One study from rural Thailand (Swaddiwudhipong 1998) evalu-
ated a community-based rider education programme consisting of
theory on motorcycle injury risk helmet use traffic laws vehicle
regulations and traffic signs and a written and skill test leading to
a licence
TheAustralian studies evaluateda rider trainingcourse of 12 hours
over two consecutive days (four hours of theory and eight hours of
practice)administeredto AustraliaPosttelegram deliverers(Adams
1985) and a mandatory pre-licence Ridersafe course consisting
of two tiers the first tier being a training course on the basics
of motorcycle riding where a pass entitled the rider to obtain alearnerrsquos permit and the second tier being a skill test where a pass
entitled the riderto obtain a probationary motorcycle licence The
training course consisted of two four-hour training sessions on
separate days involving instruction videos and actual riding on
motorcycles The skill test consisted of a four-hour preparation
course of 74 minutes of lecturediscussion 38 minutes of video
and 75 minutes of on-course riding which was the testing phase
(Kloeden 1994) The Rider Safe course is still being provided by
Transport South Australia a state government organisation (http
wwwtransportsagovau )
The duration or content of rider training in the four case-control
studies (Kraus 1975 Hurt 1981 Haworth 1997 MAIDS 2004)
was not described
Risk of bias in included studies
We describe the methodological quality of the included studies
by study type below More detailed information can be found in
Table 1 Table 2 Table 3 and Table 4
When considering the findings from these studies it is important
to note the quality (or lack thereof) of reporting of the studies
Quality of reporting is the extent to which a report provides in-
formation about the design conduct and analysis of a study ( Juumlni2001) Huwiler-Muumlntener 2002 found that even though trials
may be well reported it does not mean that the methodological
quality is of equal standard and well-conducted trials may never-
theless be reported badly The reporting quality of most studies in
this review was poor Most studies were old and did not meet con-
temporary reporting standards This made data extraction com-
plicated It also made it difficult to contact authors for additional
information
Because of significant heterogeneity between studies as well as the
poor quality of most studiespooling of studyresults to calculatean
overall measure of effect was not appropriate Instead we provided
a description of our judgement of risk of bias in the text
12Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 1550
Risk of bias in the randomised trials
There were three randomised trials (Anderson 1980 Hall 1988Swaddiwudhipong 1998) Anderson 1980 and Hall 1988 had an
initial sample size of 41188 and 26464 motorcycle licence ap-
plicants respectively Both studies assessed the effectiveness of the
pre-licence MOST with a trainingcomponent Swaddiwudhipong
1998 evaluated the effectiveness of a community-based rider ed-
ucation programme in six districts
Allocation bias
Anderson 1980 reported using random birth date tables to allocate
participants to either the intervention or control group Hall 1988
reported the use of a computerised algorithm set in a central of-
fice but also used manual random assignment when the computer
went down Swaddiwudhipong 1998 reported lsquosimple random
samplingrsquo for the delivery of the intervention where the sampling
frame included nine districts Three districts were excluded since
they were too hard to access The remaining six districts were ran-
domly allocated to the intervention (three) and the control (three)
groups Systematic sampling wasthen used ordered by village size
for the villages within districts Allocation concealment was not
reported in any of the studies The descriptions of randomisation
in these studies suggest that there may have been limitations to
the process resulting in a lack of equivalence of the intervention
and control groups at baseline and therefore the possibility of
allocation bias Swaddiwudhipong 1998 reported that before theintervention the annual incidence rate of motorcycle-related in-
juries was slightly (but not statistically significant) higher in the
intervention areas than in the control areas Participants in the
intervention districts were also more likely to have a valid licence
and to wear a helmet more often than participants in the control
districts Anderson 1980 measured differences at baseline on the
participant and prior driving record variables and found these to
be negligible Hall 1988 did not report on baseline differences
between the intervention and control groups As the Anderson
1980 and Hall 1988 studies had sample sizes of over 40000 and
26000 respectively and used some form of randomisation it is
likely that any confounding factors at baseline would have been
equally distributed between comparison and intervention groups
Selection bias
After randomisation Anderson 1980 and Hall 1988 excluded
08 and 3 participants respectively because of assignment to
the wrong group administration of the wrong intervention or
other reasons The resulting lack of analysis by rsquointention-to-treatrsquo
can introduce biasby removing the effectivenessof randomisation
Nonetheless the percentage excluded was small and the number
of overall participants large so it is unlikely to have had a major
effect on the final results In the Swaddiwudhipong 1998 study
villages were systematically sampled into intervention or control
groups The intervention was implemented and outcomes of in-
jury and death were measured at a community level Interventionand control districts were randomly selected however there were
only three intervention and three control communities
Detection bias
None of the randomised trials reported blinding of outcome as-
sessors Data collection tools used by Anderson 1980 and Hall
1988 were driver records holding police recorded crashes and of-
fences Swaddiwudhipong 1998 used police and hospital records
at a district level and post-mortem examination for deaths None
reported on validity or reliability for any of the instruments
Attrition bias
Anderson 1980 and Hall 1988 had driver records available for
96 to 100 of participants They also collected self-reported
data but the response rate to the questionnaires was very low (low-
est response rate 42 and 30 respectively) Attrition bias in
Swaddiwudhipong 1998 was not an issue as outcome data were
available for all sub-districts included in the trial
The intervention reported in Anderson 1980 and Hall 1988 tri-
als had an unanticipated effect on the participants In both trials
an average of only 28 of licence applicants in the intervention
group eventually obtained their licence compared to 35 in the
control group This indicates that the licensing process in generalseems to have deterred applicants from obtaining a licence and
although no statistical tests of the difference were conducted this
deterrence appears even greater for those who had rider training
This is not true attrition bias as the studies did not experience
differential loss to follow up from outcome measurement Instead
differential attainment of the intermediary step of licensure pre-
cluded participants from being at risk for the outcomes of interest
they are not driving a motorcycle hence they are not at risk for
motorcycle-related crashes injury death or offences In short it is
not possible to assess whether a reduction in crashes in the inter-
vention group would be due to changes in riding habits as a result
of training or simply due to delayed or never attained licensure
Intervention integrity
Neither Anderson 1980 nor Hall 1988 reported on the consis-
tency in the way the intervention was administered the possible
lsquocontaminationrsquo of participants in the intervention group by other
types of formal or informal rider training or of participants in the
control group by the intervention Swaddiwudhipong 1998 did
not report on the number of villagers in each group that actually
attended the intervention However they did report on a number
of other potential training courses that were being implemented
at the same time
13Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 1650
Analysis
Anderson 1980 and Hall 1988 acknowledging the lsquodeterrent ef-fectrsquo described above conducted two separate analyses The first
analysis included the total study population of licence applicants
over two years from the time of their application (the population
denominator) The second analysis was performed on the licensed
(intervention and control) subgroups only two years from the
time of obtaining their licence (the licensed rider denominator)
As only a third were licensed and exposed to riding the motorcy-
cle it was impossible to determine whether reported differences
in crash rates across groups were due to the intervention or to
other lsquoconfoundingrsquo characteristics of those who were determined
in their efforts to obtain the licence When analysis was restricted
only to those who gained their licence the same difficulties re-
mained The benefits of the randomisation process were thus lost
due to this lack of licensure of riders during the process As there
were only three intervention and three control groups in the study
reported by Swaddiwudhipong 1998 analyses were hindered by
smallnumbers for outcome comparisons betweenthe intervention
and the control districts
Risk of bias in the cohort studies and non-randomised
trials
We discuss the methodological quality of cohort studies (seven
prospective studies (Raymond 1979 McDavid 1989 Waller 1992
Kloeden 1994 Billheimer 1998 Perrino 2002 Savolainen 2007)and seven retrospective studies (Osga 1980 Satten 1980 Jonah
1982 Mortimer 1984 Cooper 1988 Mortimer 1988 Davis
1997)) and non-randomised trials (Adams 1985 Leung 1987) in
the one section as the biases for the two study types are similar
The quality of these studies was in general poor
Selection bias
The confounding factors expected to have been addressed in the
studies were sex age riding experience prior driving records (of-
fences crashes) previous training and the following proxies for
risk taking behaviour helmet use speed and alcoholdrug use All
of these factors have been shown to be associated with an increasedrisk of motorcycle crash (Peden 2004 Zambon 2006) The pos-
sible source of bias in cohort studies and non-randomised trials
is that groups may differ on important characteristics at baseline
participants lsquoself-selectrsquo into either the rider training group by de-
ciding to undertake a rider training course or into the lsquocontrol
grouprsquo by deciding not to undertake a course Participants who
undertake a non-mandatory rider training course may differ from
those who choose not to on characteristics such as age sex rid-
ing experience and riding behaviour as well as other possible con-
founders For example riders who take a training course may be
more safety-conscious or on the contrary may do so because they
have previously experienced a crash or because they perceive that
they have poor riding skills Therefore ideally all participants in
the intervention group need to be included in the study as wellas a random sample of controls taken from the same population
as the intervention group If that is not possible intervention and
control groups either need to be carefully matched to make them
comparable or differences at baseline need to be adjusted for in
the analysis
Most of the studies using a non-randomised trial or cohort design
compared an intervention group to a control group that differed
on a number of important factors The studies in general failed to
adjust for important differences found at baseline and their results
cannot be meaningfully interpreted Ten studies (Raymond 1979
Osga 1980 Satten 1980 Jonah 1982 Mortimer 1984 Mortimer
1988 McDavid 1989 Waller 1992 Davis 1997 Savolainen 2007)
assessed intervention and control groups that were not drawnfrom the same population In most of these cases the intervention
group was selected from a list of training graduates who were usu-
ally young female and less experienced riders whereas controls
were recruited from a general population of motorcycle riders who
were usually older male and experienced riders For three stud-
ies it was unclear how they were selected (Leung 1987 Cooper
1988 Perrino 2002) Only Kloeden 1994 recruited groups from
the same population of motorcycle applicants and Adams 1985
from a group of professional riders Nevertheless due to non-ran-
dom selection of intervention and control groups differences at
baseline are likely even though study groups were from the same
population Three studies adjusted for some confounding fac-
tors found at baseline through matching (McDavid 1989 Waller1992 Billheimer 1998) However matching has limitations in
termsof controlling for multiple confounders especially those that
are unknown to influence the effect of rider training on the out-
comes One study matched the control and intervention groups
by excluding controls (approximately 3500 (66 of initial con-
trol group) that differed from the intervention group) (Kloeden
1994) Five studies (Osga 1980 Satten 1980 Adams 1985 Davis
1997 Perrino 2002) did not adjust for differences between study
groups at baseline either in the study design or analysis Results
from these studies are therefore not reliable The other six stud-
ies (Raymond 1979 Jonah 1982 Mortimer 1984 Cooper 1988
Mortimer 1988 Leung 1987 McDavid 1989 Billheimer 1998)
adjusted for some differences between groups at baseline by eitherbi-variate or multi-variate analyses or stratification and standardi-
sation of riding exposure (outcomemiles driven) or stratification
by age group The limitation of stratification is that it is difficult to
control for many variables simultaneously because a large number
of strata will need to be generated resulting in small numbers that
cannot be meaningfully analysed Savolainen 2007 developed sta-
tistical models to assess the probability of a crash after taking the
rider training course in an intervention and control group while
examining differences by sub-groups (eg age) Participation rates
were also very low in this study with only 5584000 (1395)
in the eligible intervention group and 5884000 (147) for the
14Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 1750
control group completing surveys
Detection bias
None of the studies reported blinding of outcome assessors to the
intervention or control status of the participants Seven studies
used driver records containing police recorded data as their main
measure of outcome (Leung 1987 Cooper 1988 McDavid 1989
Waller 1992 Kloeden 1994 Davis 1997 Billheimer 1998) An-
other four studies (Raymond 1979 Satten 1980 Mortimer 1984
Mortimer 1988) only used self-report Raymond 1979 was the
only study using self-report as the main outcome measurement in
which study instruments were validated Self-report might prevent
the underreporting that has been described for the use of police-
driver records (see rsquoRisk of biasrsquo in the randomised trials above)but has other risks of bias especially if questionnaires or inter-
view methods have not been validated and tested for reliability
Recall bias arises from mistakes in recollecting events both be-
cause of failures of memory and because of looking at things lsquowith
hindsightrsquo and possibly changed views (httpwwwcochraneorg
resourcesglossaryhtm) As many studies were retrospective risk
of recall bias is likely and may have increased with length of follow
up For studies that used self-reported outcomes only it is also
possible that social desirability response bias played a role where
participants in the training group may have felt it was lsquoexpectedrsquo
to report good driving histories Multiple methods to measure the
same outcome for both groups decreases the risk of detection bias
for example because self-reported answers can be cross checkedwith other sources of information such as driver records
In this regard three studies (Osga 1980 Jonah 1982 Perrino
2002) used both self-report and driver records Osga 1980 found
that for both intervention and control groups more crashes were
reported through self-report than on state crash records an aver-
age of 54 of self-reported crashes could be found back on state
crash records One industry-based study (Adams 1985) used oc-
cupational health records which were not validated but which are
unlikely to suffer from underreporting due to work cover claims
Savolainen 2007 used information on rider crashes for a five-year
period but did not specify from where this information was ob-
tained
Attrition bias
The majority of studies (Raymond 1979 Satten 1980 Adams
1985 Leung 1987 Kloeden 1994 Davis 1997 Billheimer 1998
Perrino 2002) inadequately reported on losses of participants to
the study We can therefore not be certain that there were no sys-
tematic differences in the characteristics between the intervention
and control group participants Savolainen 2007 reported the re-
sponse rates to their baseline survey but did not report on the per-
centage of outcome data that was available for their study sample
A similar deterrent effect of rider training as described for the ran-
domised trials (Anderson 1980 Hall 1988) occurred in two stud-
ies Jonah 1982 reported that 65(14992310) of the trained rid-
ers stopped riding compared to 47 (9492029) of the untrainedriders They excluded non-active and non-licensed riders before
the study commenced so the differential choice to continue riding
after training did not affect their analyses Leung 1987 reported
that the percentage of participants in the intervention group ob-
taining their licence was 50 (no number provided by authors)
lower than in the control group This may have biased the mea-
surement of outcomes because unlicensed participants will not be
exposed to riding the motorcycle and therefore not be at risk of
being involved in a crash or violation As explained above this
makes it difficult to assess whether the reduction in crashes in the
intervention group is due to a change in driving habits or simply
due to delayed licensure
Mortimer 1984 and Mortimer 1988 describe that about 44 and30 of course graduates respectively were excluded from analysis
because they had not ridden a motorcycle in the previous year
However because these studies were retrospective and the controls
were activeriders at the time of recruitment a decision on whether
the deterrent effect was differential between the two groups could
not be made Though not reported quantitatively Kloeden 1994
also describes the effectof rider training deterringindividuals from
becoming licensed and hence exposure to riding and risk of crash
death or offences involving a motorcycle Osga 1980 reported
that 27 of the trained riders did not ride after the course and
excluded these from the analysis
Use of different data collection strategies may have contributed to
differential loss to follow up in the cohort studies Satten 1980for example interviewed the intervention group over the tele-
phone but mailed questionnaires to the control group Mortimer
1984 and Mortimer 1988 distributed questionnaires in person to
the control group but mailed them to the intervention group
This may have played a role in the observed differential response
rates where more subjects responded in the personally approached
groups
Survival bias
For those studies that assessed offences crashes and injuries but
not death survival bias may have also played a role Motorcyclists
who were fatally or seriously injured in crashes during the follow-
up period after the intervention would have been less likely or
unable to respond to surveys This could bias the effect measures
towards or away from the null None of the studies wrote about
this issue or examined the characteristics of those lost to follow up
in order to address the possibility of survival bias
Intervention integrity
Although the interventions were intended for the entire lsquoexposedrsquo
groups most studies reported difficulty in tracking participants
whoreceivedthe interventionand those whodid not Adams 1985
15Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 1850
estimatedthat 62of thetotalstudysample received the interven-
tion The remaining studies did not report the percentage of thestudy population that received the intervention as was intended
or reported that this percentage was unknown (Raymond 1979
Kloeden 1994 Satten 1980 Jonah 1982 Mortimer 1984 Leung
1987 Cooper 1988 Mortimer 1988 Billheimer 1998 Perrino
2002 Savolainen 2007) Only Raymond 1979 reported the degree
to which their intervention was delivered as intended No studies
reported on possible lsquocontaminationrsquo of participants by other inter-
ventions than the one to which they were allocated Because Osga
1980 Satten 1980 Jonah 1982 Mortimer 1984 Cooper 1988
Mortimer 1988 Waller 1992 and Davis 1997 were retrospective
studies and participants were recruited from a list of graduates it
was known that they had completed and passed the course
Analysis
In some studies some participants who reported at baseline that
they were untrained decided to enroll in a rider training pro-
gramme after the baseline survey The non-randomised trial by
Adams 1985 experienced this move of subjects from the untrained
to the trained group but failed to conduct analyses by lsquointention-
to-treatrsquo This can bias the results in randomised and non-ran-
domised trials if participants who changed groups differ signifi-
cantly with regard to potentially confounding characteristics On
the other hand Perrino 2002 also experienced this move of sub-
jects from theuntrained to the trained groupmore than 100riders(5 of the total sample) originally stated that they were untrained
at baseline but then received training between the baseline and
follow-up surveys Yet since the Perrino 2002 study was a cohort
their analyses then proceeded by including those riders who were
lsquonewlyrsquo trained as exposed Riders who were lsquonewlyrsquo trained since
baseline but with a programme other than the one under study
were excluded from the analysis Nevertheless confounding fac-
tors were not accounted for in the analyses
McDavid 1989 matched exposed with unexposed subjects but
could not find suitable matches for the majority (60) of their
subjects They could therefore only conduct analysis on the result-
ing40 (139346) which is an inappropriate andinefficient anal-
ysis Some studies that reported on the issue of non-active licensedriders(Satten 1980 Jonah 1982 Mortimer 1984 Mortimer 1988
Kloeden 1994) excluded non-active riders from analysis intro-
ducing selection bias as it is likely that riders who continued to
ride differed with regards to potentially confounding character-
istics across intervention and control groups Kloeden 1994 for
example conducted several analyses to assess the effect of various
sub-sets of the control group Original analyses included all con-
trols and led to erroneous results since only a small proportion of
controls hadgone on to get a licence after their learnerrsquos permitand
were actually exposed to the possibility of experiencing a crash
Subsequent analyses compared only intervention and control sub-
jects who had a licence Waller 1992 reported crashes by person-
days of exposure but did not explain how they could be sure that
each study intervention and control participant was an active riderand actually exposed to the risk of experiencing a crash
Potential confounders that are unequally distributed by study arm
at baseline need to be dealt with in the analysis Satten 1980
Adams 1985 and Davis 1997 did not take into account any con-
founding factors either by design or in the analysis McDavid
1989 Waller 1992 and Billheimer 1998 accounted for potential
confounding factors at the design stage by matching the subjects
in the intervention and control groups Other studies (Raymond
1979 Mortimer 1984 Mortimer 1988 Perrino 2002 and also
Waller 1992) adjusted for one or two confounders by stratifica-
tionor standardisation (eg crashesmiles driven age group) Osga
1980 conducteda univariate analysis of crash rates in which no po-
tential confounders were accounted for but also calculated cumu-lative crash rates by mileage travelled Only Jonah 1982 Cooper
1988 and Savolainen 2007 conducted a multivariate analysis
Risk of bias in the case-control studies
A case-control design is suitable for the assessment of risk fac-
tors for outcomes such as motorcycle crashes injuries or deaths
but not appropriate for the evaluation of an intervention such
as motorcycle rider training It can potentially result in spurious
conclusions regarding the efficacy of the intervention because the
intervention has not been applied at random to an intervention
and control group Even with the control of possible confounders
using a case-control design to evaluate an intervention can lead tothese kinds of results since those who choose to undertake rider
training may differ on important prognostic factors that affect the
outcome compared to those who do not choose to participate in
rider training
Selection bias
A potential source of bias in case-control studies as with all other
study types where the intervention is not randomly allocated to
participants is that comparison groups are likely to differ on po-
tentially confounding characteristics This means that ideally all
cases need to be included in the study as well as a random sample
of controls taken from the population that gave rise to the cases If that is not possible cases and controls either need to be carefully
matched to make themcomparable or differences at baseline need
to be adjusted for in the analysis All four case-control studies were
population based Kraus 1975 included all fatalities and injured
cases leading to hospitalisation and appropriately recruited a ran-
dom sample of controls from the same population Hurt 1981 did
not manage to recruit all or a random sample of cases Only half
of the notifications of crashes to the police were received in time
by the researchers the research team for unreported reasons was
only able to respond to half of those again Data collection was
therefore completed for only about a fifth of all cases Neverthe-
less this was still a large sample of 900 cases Haworth 1997 using
16Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 1950
a similar study design was more likely to have included all cases
but this is not entirely clear from the study reported It is unclearfor example whether fatalities were included as they did not use
coronerrsquos reports for the recruitment of cases The MAIDS 2004
study was conducted in five European countries but it was unclear
how cases were recruited whether they were all cases during the
two years that the study lasted or a random sample of all cases
what the response rates were etc Controls were recruited from
randomly selected petrol stations but it was likewise not clear what
the eligibility criteria for inclusion in the study were how they
were recruited (eg randomly) and what the response rate was
Hurt 1981 and Haworth 1997 both matched controls to cases by
recruiting themat the sametime ofdayandsamedayof weekas the
crash occurred Hurt 1981 however recruited controls two years
later andit is likelythat there would have beendifferences betweenthe two comparison groups One important difference was that
during data collection for cases there was very little specialised
motorcycletrainingavailable in the studyregionwhereas two years
later when control data were collected there was some training
available
These four case-control studies also suffered from selection bias
due to potential differences between those that gave consent to
participate in the study and responded to interviews andquestion-
naires and those that did not None of the studies have assessed
such differences although Hurt 1981 did acknowledge that con-
trols giving consent to cooperate may have differed from those
motorcyclists who did not
Detection bias
Measures that can be taken to reduce detection bias in case-con-
trol studies are blinding of interviewers for case or control status
blinding of the participant for the study factor of interest validat-
ing measurement instruments and using these in an identical way
for cases and controls and cross checking self-reported informa-
tion with other more objective sources such as registers None of
the studies reported blinding of the intervention assessors to case
or control status but all used consistent measurement methods for
both case and control groups
All case-control studies used self-report as their main measure toassess whether cases and controls had received rider training and
if so the type of training The major potential source of bias in
case-control studies when self-report is used as a measure is recall
bias One way of reducing recall bias is by double-checking self-
reported data with more objective data Kraus 1975 also used po-
lice accident reports for information on rider characteristics but
such information was not available about rider training Haworth
1997 and MAIDS 2004 did not report on any measures to limit
recall bias Hurt 1981 reported further questioning of cases by
the interviewer when answers did not make sense or when par-
ticipants seemed to contradict themselves as part of the in-depth
unstructured interview technique This qualitative research tech-
nique can result in valuable information but as interviewers were
not blinded to case or control status of the participants it may have led to more thorough examination of cases compared to the
controls and thus differential detection bias
Intervention integrity
The level of detail about the rider training provided in the case-
control studies differed significantly but was minimal for each
Kraus 1975 simply asked whether lsquoanyrsquo motorcycle rider training
had lsquoeverrsquo been carried out Hurt 1981 asked participants whether
they hadbeen self taught informally taught or completed formally
taught rider training courses MAIDS 2004 asked whether partic-
ipants had had pre-license training additional (non-compulsory)
training other training or no training whereas Haworth 1997asked about the number of courses that a participant had com-
pleted whetherthey were basic intermediate or advanced courses
and how long ago the training had been done None of the case-
control studies reported collecting information about further de-
tails about any training programmes (eg type of training cur-
riculum duration of training)
Analysis
Kraus 1975 accounted for confounding factors through the exclu-
sion of females and stratification by age and frequency of motor-
cycle use Hurt 1981 matched cases and controls for time of day
and day of the week that the crash occurred and did not adjust forany confounding factors in the analysis Haworth 1997 matched
groups for time of day and day of the week that the crash occurred
and stratified by age blood alcohol level and licence status in the
analysis None of the studies used multivariate methods MAIDS
2004 did not provide any information about any participant char-
acteristics other than the study exposures of interest and did not
describe any methods to adjust for potential confounding factors
Effects of interventions
We have presented the results in a descriptive manner due to the
substantial heterogeneity in study types interventions and out-comes and the generally poor quality of the studies We have
grouped the interventions by mandatory and non-mandatory pre-
licence rider training and post-licence rider training and further
by study type and by outcome (crashes injurious and fatal crashes
and offences) We discuss results that are reported as statistically
significant (whether a P value was reported or not)
Pre-licence rider training - mandatory
Randomised trials
17Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2050
Crashes
Total group of participants
Anderson 1980 found a reduction in motorcycle crash rates in
the intervention group (MOST skill test with remedial training)
compared to the control group (who received the skill testwithout
a training component) The percent differenceranged from 143
six months after the programme (P lt 005) to 211 (P lt 0001)
after one year and 127 (P lt 001) after two years Hall 1988
found no statistically significant difference in crash rates between
intervention groups at any point in time
Subgroup of participants who obtained their licence only
Anderson 1980 found a reduction in motorcycle crash rates in the
intervention arm but this was statistically significant at one year
only (24 to 218 reduction P lt 005) Hall 1988 found no
statistically significant differences
Offences
Total group of participants
The percentage of offences in Anderson 1980 after six months
and one and two years was slightly higher in the intervention
arm compared to the control group but these differences were
not statistically significant Likewise the study by Hall 1988 did
not find a statistically significant difference in offences between
groups
Subgroup of participants who obtained their licence only
Hall 1988 found no significant difference in offence rates between
the intervention and control groups Anderson 1980 found thatthe percentage of offences increased in the intervention arms com-
pared to the control group This difference was statistically signif-
icant after two years (51 to 115 increase P lt 005)
Cohort studies
Crashes
Neitherof the cohort studies (Kloeden1994 and Billheimer1998)
found a statistically significant difference in crash rates between
the intervention and control groups
Offences
Billheimer 1998 (for the California Motorcyclist Safety Program
course) and Perrino 2002 did not find a statistically significant
difference in offence rates between the intervention and control
groups
Case-control studies
Crashes
MAIDS 2004 found that a slightly higher percentage of controls
(not involved in a crash) had had pre-license rider training than
cases (involved in a crash) This difference was not statistically significant (50 of controls vs 477 of cases OR 091 95 CI
076 to 110)
Pre-licence rider training - non-mandatory
Randomised trial
Injuries and deaths
Swaddiwudhipong 1998 found a significantly lower injury ratein the intervention villages (1051000 population 95 CI 92
to 120) compared to the control villages (1691000 population
95 CI 152 to 185) one year after the intervention This differ-
ence was not maintained two years after the intervention (127
95 CI 112 to 143 intervention 151 95 CI 135 to 167
control) There was no significant difference in death rates be-
tween groups
Cohort studies
Crashes
Cooper 1988 foundno statistically significant differencein crashes
between the intervention and control groups McDavid 1989
found a lower crash rate in the intervention group (14139) com-
pared to the control group (23139) but no test of statistical sig-
nificance was undertaken
Injuries and deaths
McDavid 1989 found that the intervention group had fewer in-
jurious crashes (814 crashes) compared to the control group (16
23 crashes) but no statistical significance test was reported
18Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2150
Offences
Cooper 1988 foundno statistically significantdifference in offencerates between the intervention and control groups McDavid 1989
foundno statistically significantdifference in offence ratesbetween
the intervention and control groups
Non-mandatory rider training
Non-randomised trials
Crashes
Neither Adams 1985 nor Leung 1987 found a statistically signif-icant effect of rider training on crash rates
Offences
Leung 1987 found a statistically significant lower offence rate in
the intervention group compared to the control group for both
males and females and for all age groups except for women 35
years or older (P values not reported)
Cohort studies
Crashes
Osga 1980 in the univariate analysis reported a lower percentage
of crashesin theintervention group (that continued to ride) (28
168595) than in the control group (34 112329) In the cu-
mulative crash rate analysis however Osga1980 found statistically
significant (at the P lt 001 level) higher crash rates in the interven-
tion group than in the control group at allmileage intervals (1000
to 29999 miles) travelled After riding more than 30000 miles
(when over 50 of the trained riders had been involved in a crash
compared to approximately 40 in untrained riders) this differ-
ence was no longer statistically significant Satten 1980 found a
lower percentage of crashes in the intervention group (13 969)
compared to the control group (25 2079) but the differencewas not assessed statistically Davis 1997 found a lower crash rate
in trained riders but it was unclear how this was calculated be-
cause the size of the control group was not reported Jonah 1982
and Billheimer 1998 (for the Experienced Rider Course) found no
statistically significant difference in crashes between the interven-
tion and control groups Although the difference between those
who completed the Experienced Rider Course and no course in
Billheimer 1998 neared significance at six months after the course
(P = 007) large P values were observed for one year after (P =
030) Raymond 1979 found a statistically significant higher crash
rate for the intervention group (1861000 miles ridden) com-
pared to the control group (1181000 miles ridden) in both males
and females (P lt 0001) especially during the first 5000 miles of
riding Waller 1992 found a statistically significant higher crashrate for the intervention group (27 crashes in three years or 070
crashes10000 person days) compared to the control group (5
crashes in three years or 013 crashes10000 person days) with a
crude relative risk of 538 (X 2=1513 plt005) Mortimer 1984
and Mortimer 1988 found no statistically significant difference in
crashes between the intervention and control groups
Savolainen 2007 found that the intervention group (those who
completed the Basic Rider Course) was 44 more likely to be
involved in crashes than the control group and those who com-
pleted the course multiple times were an additional 180 more
likely to be involved in a crash This finding was based on low
numbers and no P value or confidence level was given for either
result
Injuries and deaths
Satten 1980 also found a lower percentage of injurious crashes
(33 39) in the intervention group compared to the control
group(60 1220) although again no statistical significance test
was reported
Offences
Jonah 1982 did find a significantly lower self-reportedoffence rate
in the intervention group (15) compared to the control group(34 P lt 001) (no numbers given) Satten 1980 also found
a lower self-reported offence rate in the intervention group (1
71 versus 19100) but again this was not tested for statistical
significance Mortimer 1984 Mortimer1988 and Billheimer 1998
found no statistically significant differencein offence ratesbetween
the intervention and control groups
Case-control studies
CrashesMAIDS 2004 found that a slightly higher percentage of cases
(involved in a crash) had had additional non-mandatory rider
training than controls (not involved in a crash) This difference
was not statistically significant (17 of cases vs 12 of controls
OR 146 95 CI 064 to 337)
Rider training unspecified
Case-control studies
19Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2250
Injuries and deaths
Hurt 1981 found that the odds of injurious crashes were lower inthose who had received professional training compared to those
who had not Kraus 1975 found that in the younger age group
(less than 20 years) and the older age group (over 35 years) a larger
percentage of injured riders (12 and 21 respectively) reported
prior motorcycle rider training compared with the uninjured con-
trol group (6 and 5 respectively) (P lt 001) (no numbers
given) Haworth 1997 reported several different outcomes one
beingthat having completed at leastone trainingcoursecompared
to no training courses at all had no significant effect on the odds
of crashing (afteradjustment for age blood alcohol level or licence
status (learner permitprobationaryfull)) When comparing three
different levels of training courses (beginner intermediate and ad-
vanced) with no training no significant differences were found
in the odds of crashing However after adjustment for age and
blood alcohol level completing an advanced course as compared
to a beginners course was associated with a significant decrease in
the odds of severe motorcycle crashes resulting in serious injury or
death (OR 02 95 CI 01 to 08)
D I S C U S S I O N
The objective of this review was to quantify the effectiveness of
pre- andpost-licence motorcycle rider training on the reduction of
traffic crash involvement injuries deaths and offences committedby motorcycle riders Most included studies were cohort studies
There were a few randomised and non-randomised trials and case-
control studies
All studies compared either mandatory or non-mandatory formal
training with no formal training except for Haworth 1997rsquos case-
control study that compared levels of training (beginner interme-
diate and advanced) to each other
We found studies that evaluated the effectiveness of manda-
tory pre-licence training in reducing crash and offence rates
(Anderson 1980 Hall 1988 Kloeden 1994 Billheimer 1998
(the basic California Motorcyclist Safety Program component of
the study) Perrino 2002) non-mandatory training that when
passed resulted in a licence (Cooper 1988 McDavid 1989
Swaddiwudhipong 1998) non-mandatory training that could be
taken by motorcycle riders with or without a licence (Raymond
1979 Osga 1980 Satten 1980 Jonah 1982 Mortimer 1984
Adams 1985 Leung 1987 Mortimer 1988 Waller 1992 Davis
1997 Billheimer 1998 (the Experienced Rider Course compo-
nent of the study)) and unspecified rider training on crash injury
and death rates (Kraus 1975 Hurt 1981 Haworth 1997 MAIDS
2004)
The studies provided conflicting evidence regarding motorcycle
rider training which is likely due to the poor design quality and
reporting of the studies Most of the studies included in the review
were likely to be adversely affected by significant selection biasalong with substantial detection bias(measurement and recall bias)
which limits our ability to make conclusive comments
Mandatory pre-licence rider training
There were two RCTs three cohort studies and one case-control
study that examined pre-licence ridertraining The two RCTs eval-
uated the Motorcycle Skills Operating Test (MOST) (Anderson
1980 Hall 1988) Anderson 1980 found a statistically significant
reduction in crashes in the intervention group at six months one
and twoyearsfollow up However when only participantswho ob-
tainedtheirlicence were included in theanalysis thereduction wasstatistically significant only at one year follow up The other RCT
(Hall 1988) found no statistically significant differences in crash
rates at any point of time neither in the total population group
nor in the licensed subgroup Neither of the two cohort stud-
ies assessing crash rates found a statistically significant difference
in crash rates between intervention and control groups ( Kloeden
1994 Billheimer1998) andthe case-control study (MAIDS 2004)
also did not find a statistically significant difference in the odds of
experiencing a crash between trained and untrained riders With
regards to offences neither of the RCTs found a statistically sig-
nificant difference in offence rate between the intervention and
control groups in the analysis of the total group of participants
entering the trial However Anderson 1980 found a statistically significant increase in offence rates in the intervention group but
only after two years of follow up andonly in the analysis of partici-
pants who obtained their licence The two cohort studies assessing
offence rates found no statistically significant difference between
the groups (Billheimer 1998 Perrino 2002)
None of the five studies (RCTs or cohort) found a reduction in
crashes or offences following mandatory pre-licence rider training
except for Anderson 1980 who observed a statistically significant
reduction in crashes in the intervention group at one year in both
analyses It is not clear why both RCTs that evaluated the same
MOST and training intervention observed different results Both
studies enrolled large samples and observed a similar uptake of
licensure following the intervention It is possible that differencesmay have arisen due to different study samples (California ver-
sus New York) or that the intervention delivery differed between
studies
In both trials of the Motorcycle Skills Operating Test including
rider training only a third of motorcycle licence applicants ob-
tained their motorcycle licence It is possible that a licensing pro-
cess including a skills test may reduce crashes through decreased
exposure
Non-mandatory pre-licence training
20Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2350
Three studies assessed the effect of non-mandatory pre-licence
training The Cooper 1988 cohort study found no difference incrash or offence rates between the trained and untrained groups
We cannot however be confident about this result as the study
sufferedfrom selectionbiaswhichcouldhave biasedthe measure of
effect toward the null The RCT conducted by Swaddiwudhipong
1998 examined injury and death rates and found a statistically
significant lower injury rate in the intervention villages after one
year of follow up but no longer at two year follow up There
was no statistically significant difference in death rates between
intervention and control villages Our confidence in the results of
this study are hindered by the inadequacy of the randomisation
process that led to significant differences between the intervention
and control districts on two important variables The intervention
districts were more likely than the control districts to have a validlicence andto wear a helmetSincethese differences were nottaken
into account in later analyses they could explain the differences
observed in injury rates between districts Furthermore with only
six districts randomised the sample size was probably too small
to detect a difference between intervention and control groups
had there been one The McDavid 1989 cohort study found a
lower percentage of injuries in the intervention group although
a statistical test was not conducted McDavid 1989 drew their
intervention and control groups from different populations but
did adjust for confounders although they only analysed 40 of
the initial sample (139 trained and 139 untrained) Because of
these substantial weaknesses we are unable to draw conclusions
from this studyAn overall effect of pre-licence training on the reductionof crashes
injuries deaths and offences was not found although one should
be aware that the research of this issue to date has had significant
flaws However a mandatory skill test (such as the MOST) in
combination with and even without pre-licence training as part
of a licensing process seems to reduce the number of individuals
completing the licensing procedure and therefore indirectly re-
duces crash rates through a reduction in exposure to motorcycle
riding Making licensure a more difficult process may therefore be
seen as advantageous if it delays licensure and therefore reduces
exposure It is possible that the group who chose not to continue
riding a motorcycle would not have been the group to experience a
crash or commit an offence if they had continued to ride anyhowfor example because they are more safety conscious and this was
the reason for their giving up riding in the first place If this is the
case deterring individuals from completing their licence and rid-
ing may not necessarily decrease overall crashes deaths or offence
rates In addition other potentially adverse effects of complicat-
ing a licensing procedure need to be considered such as increases
in unlicensed riding In Thailand non-mandatory rider training
offered as a community intervention may have decreased injury
rates but this needs to be confirmed through further research
Non-mandatory training
The majority of the studies assessed non-mandatory training
(training independent of any licensing procedure) and the results
were varied The two non-randomised trials assessing crash rates found no statistically significant difference in crash rates between
the intervention and control groups (Adams 1985 Leung 1987)
Of the nine cohort studies Satten 1980 found a lower percentage
of crashesin the intervention group although it is unclear whether
this was a statistically significant result Davis 1997 also found a
lower percentage of crashesin theinterventiongroup although it is
unclear with which control group the intervention group wascom-
pared Raymond 1979 Osga 1980 Waller 1992 and Savolainen
2007 found a statistically significant higher crash rate in the in-
tervention group and the remaining cohort studies (Jonah 1982
Mortimer 1984 Mortimer 1988 Billheimer 1998) found no sta-
tistically significant difference in crash rates between intervention
and control groups No conclusions can be drawn from these stud-
ies with confidence The Satten 1980 and Davis 1997 studies did
not adjust for any confounding factors even though important
differences existed between the intervention and control groups
Although Raymond 1979 and Waller 1992 adjusted for a num-
ber of confounders at baseline (Waller 1992) and in the analy-
sis (Raymond 1979 Waller 1992) the intervention and control
groups were drawn from such different populations that it is pos-
sible that unmeasured confounders were responsible for the sta-
tistically significant increased risk of crashes observed in the inter-
vention group The observed difference could also be attributed totheobserved differential loss to followup (Raymond 1979) which
could bias the measure of effect towards or away from the null
The Osga 1980 study found a protective effect of rider training
in the univariate analysis but statistically significant higher crash
rates in the intervention group when accounting for riding expo-
sure Sex (a significantly higher proportion of females in the in-
tervention group) and other potentially confounding factors that
could explain thedifferencebetween the groups were howeverun-
accounted for Finally Savolainen 2007 did control for potential
confounding variables yet it is not clear whether variables were
controlled for simultaneously or one at a time One case-control
study (MAIDS 2004) found no statistically significant reduction
in the odds of experiencing a crash between trained and untrainedriders but the quality of this study was poor mainly because the
analysis did not account for any confounding factors
One cohort study (Satten 1980) assessed the effect of non-manda-
tory rider training on injurious crashes and found a lower percent-
age of injuries in the intervention group although a statistical test
was not conducted As mentioned above the Satten 1980 studyrsquos
validity is reduced due to its lack of adjustment for confounding
factors
The only non-randomised trial that assessed offence rates (Leung
1987) found a statistically significant lower offence rate in the in-
tervention group and of the five cohort studies that assessed of-
fence rates (Satten 1980 Jonah 1982 Mortimer 1984 Mortimer
21Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2450
1988) one study (Jonah 1982) also found a statistically significant
lower offence rate in the intervention group However we have lit-tle confidence in the results from Leung 1987 since potential con-
founding factors were not appropriately addressed Jonah 1982
did conduct a multivariate analysis but was hindered by an initial
low response rate The other four cohort studies did control for
several potential confounders and found no difference in offence
rates following non-mandatory post-licence rider training
The non-randomised trials and cohort studies evaluating non-
mandatory rider training were at high risk of substantial selection
bias due to the non-random allocation of participants as well
as low participation rates and attrition bias and could therefore
not provide conclusive evidence with regard to the effect of the
training It should be noted that despite the biases and other issues
with this group of studies most didnot detectmajor effects relatedto rider training in either direction This may suggest that any
potential impact of rider training will be modest
Studies examining unspecified modes of ridertraining
Three case-control studies assessed whether injured or deceased
motorcyclists were more or less likely to have had rider training
compared to control motorcyclists All had conflicting results
Hurt1981 foundthe odds of havingan injuriouscrash waslower in
those who had received rider training whereas Kraus 1975 found
that the odds of having an injurious crash were higher in youngerand older trained riders Haworth 1997 who compared advanced
versus beginner training found no effect on motorcycle injuries
or deaths although they did find that those who had completed
advanced training had reduced chances of crashing compared to
those who had done a beginnersrsquo training course Similar issues
with uncontrolled confounding and biases of selection as described
above arose with the three case-control studies Hurt 1981 suffers
from inadequate control of confounding via poor matching and
both the Kraus 1975 and the Haworth 1997 findings were only
controlled for limited factors by means of stratified analysis It is
likely that these limitations are the reason for the contradictory
directions of effect observed for the three case-control studies
Methodological weaknesses of the studies
In summary most studies investigating the effectiveness of mo-
torcycle rider training on reducing crashes injuries and death suf-
fered from serious methodological weaknesses
The majority of studies were non-randomised and were there-
fore limited mainly by selection bias Although the approaches of
matching stratification or regression analyses can minimise bias
(Fleiss 2004) they were poorly employed in the reviewed studies
The inadequate adjustment for confounders in particular adds
to our lack of confidence in the studiesrsquo results since it has been
demonstrated that individuals who choose to take motorcycle rider
training differ from those not taking rider training For instance
trained riders often differ from controls in age sex and experi-ence usually ride less and may have a more positive attitude to
safety as indicated by the reported more frequent wearing of seat
belts (Mortimer 1982)
The two large randomised trials reported that a large proportion
of their study group were never exposed to the risk of offences or
crashes because they never attained licensure This made it impos-
sible to assess whether a reduction in crashes in the intervention
group was a result of training or due to delayed or never attained
licensure
In some studies the sample sizes were too small to detect any sta-
tistically significant differences Follow-up time after training dif-
fered between studies and was an impediment to observing any
real effects of training if they existed As riding experience ac-cumulates over time any differences in outcomes between those
trained and untrained would diminish in studies with lengthy fol-
low-up times This dilution of any possible effect of rider training
may have also occurred in studies that mixed both novice and ex-
perienced riders in the study groups since it is hypothesised that
novice riders serve to benefit most from rider training This would
increase the difficulty in finding any effect of rider training par-
ticularly when sample sizes are small
Detection bias may also account for some of the differences ob-
served between trained and untrained riders Detection bias would
have occurred if the self-report hospital driver or police records
used to record the outcomes of crash injury death or offences
differed between the trained and untrained riders Research hasdemonstrated that not all crashes are reported to police for ex-
ample small crashes or off-road crashes (Kraus 1975) even serious
injury crashes have been found to be underreported (Anderson
1980) However this would only introduce detection bias if the
reporting of outcome differs for trained and untrained riders This
does appear possible in that responsible riders may be more likely
to enroll in a rider training course and also may be more likely
to report crashes to police than irresponsible riders meaning that
the number of crashes from riskier riders could be underrepre-
sented on records An analysis to investigate the effectiveness of
rider training may then show that rider training is not effective in
reducing crashes
Rather than using reported outcomes some studies used self-re-ported survey data although this technique suffers from non-
response biases In fact two studies reported differences in de-
mographic characteristics in response rates to questionnaires
(Anderson 1980 Hall 1988) Self-reported data also introduce
the risk of recall bias survival bias and social desirability response
bias Recall bias may have been equally distributed among trained
and untrained riders howeverit is hypothesised that trained riders
may have felt it was lsquoexpectedrsquo to report good driving histories
hence leading the rider training to appear effective
22Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2550
Limitations of this review
Limitations of this review were that we found very few recentstudies (the two most recent ones published in 2002 and 2007)
and that we found very few studies conducted in low-and middle
income countries even though our search strategy was set up to
identify such studies Anotherlimitation wasthat forthe reviewed
studies we could not assess all outcomes of rider training that are
potentially useful Some of the included studies assessed outcomes
such as use of safety clothing including helmets Such outcomes
are valuable since the use of motorcycle helmets has been demon-
strated to be effective in preventing head injuries (Liu 2004) De-
spite this the use of safety clothing was not included as an out-
come in this review because the studies measuring this outcome
were limited by poor data quality due to very low response rates
Another outcome of interest in assessing the effectiveness of rider
training is lsquotype of injuryrsquo As Haworth 2005 state in their review
of motorcycle licensing and training rider training could be con-
sidered effective if injuries become less severe because motorcycle
riders have learned how to avoid more serious injuries through
rider training The studies included in this review did not specify
type of injury and so this outcome could not be assessed Last
changes in knowledge and attitudes following rider training are
other potentially useful outcomes We have excluded such studies
(Osga 1979 Clayton 1990 Goldenbeld 2004) as these outcomes
are proxies rather than direct measures of the effectiveness of mo-
torcycle rider training We opted to report on the outcomes of
crashes injury death and offences alone since there is little ev-
idence about how changes in knowledge and attitudes relate to
these outcomes
A U T H O R S rsquo C O N C L U S I O N S
Implications for practice
Due to the poor quality of studies included in this review we were
unable to draw any conclusions about the effectiveness of rider
training on crash rates injurious or fatal or offence rates The
findings do suggest that mandatory pre-licence training may be
an impediment to completing a motorcycle licensing process thuspossibly indirectly reducing crash injury death and offence rates
through a reduction in exposure to riding a motorcycle On the
basis of the existing evidence it is not clear if and what type of
training reduces the risk of crashes injuries deaths or offences in
motorcyclists and the selection of the best rider training practice
can therefore not be recommended The suggestion that manda-
tory pre-licence training indirectly reduces exposure to riding the
motorcycle because trainees are impeded from completing the li-
censing process does not help those motorcyclists who persist in
obtaining their licence and riding the motorcycle Some type of
rider training is likely necessary to teach motorcyclists to ride a
motorcycle safely Good policy is based on sound evidence and
evaluation but such evidence for motorcycle rider training is seri-
ously lacking Policy makers are therefore encouraged to include arigorous evaluation component to any new or existing rider train-
ing programme Such an evaluation should ideally be randomized
and assess whether the training programme is effective and if so
which component(s) of the programme work best
Implications for research
Given theresults of this review which found littlereliable evidence
about the effectiveness of motorcycle rider training it is necessary
to build up an evidence base of the effectiveness of motorcycle
rider training in the reduction of crash injury death and offence
rates To this end well designed randomised controlled trials need
to be conducted that examine the impact of rider training oncrashes and offences A randomised controlled trial is the preferred
method as it would significantly reduce most biases described in
this review To minimise the impact of participants delaying or
never attaining their licence and therefore exposure as found in
the randomised trials in this review such a trial should include
riders who are already licensed and reliable measures of riding
exposure in a reasonably lengthy follow up period It is expected
that a randomised controlled trial of sufficient sample size would
observe equal balance on possible confounders by study arm how-
ever an assessment of potentially important confounders (such as
helmet wearing) by study group would enable control for these
variables in the analyses if necessary It is imperative that the effec-
tiveness of each component of the rider training (such as theorypractice knowledge skills and modification of behaviour) and the
synergy of such components are evaluated Outcomes should be
objective ideally those that are routinely collected rather than self-
reported outcomes Even though riding the motorcycle poses dif-
ferent challenges and incurs different risks lessons may be learned
from research into the effects of car driver training
Motorcycles are a common means of transport in low- and middle
income countries and usage is increasing As the burden of mo-
torcycle crash victims is mostly borne in such countries research
should be focused on the effectiveness of types of rider training on
the reduction of crashes injuries and fatalities in such countries
and be supported by higher income countries This review found
only one study conducted in a middle income country that sug-gested decreasing injury rates It would be beneficial to conduct
additional research to confirm this suggestion and to enhance our
knowledge of the role of rider training in low- and middle-income
settings
March 2006 saw the launch of the University of North Carolina
Highway Safety Research Centre partnership with the Motorcycle
Safety Foundation (MSF) and plans were revealed to conduct a
study of the potential benefits of motorcycle rider training The
study is funded by the MSF and the National Highway Traffic Sa-
fety Administration (NHTSA) Called rsquoThe Discovery Projectrsquo it
will compare data on crash rates from riders participating in a basic
23Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2650
motorcycle safety class with riders participating in a basic motor-
cycle safety class as well as a variety of continuing education classesThe collection of pilot data was due to start in March 2007 and
the data collection portion of the study is expected to take three
years (Highway Safety Research Center) It is vital that researchers
and policy makers collaborate to develop furtherhigh quality eval-
uations of motorcycle rider training programmes to help inform
development of new and effective training programmes
A C K N O W L E D G E M E N T S
Staff from The University of Sydney library for helping us locate
and retrieve full reports of many papers
R E F E R E N C E S
References to studies included in this review
Adams 1985 published data only
Adams A Collingwood V Job RFS Evaluation of the 1982 Australia
Post motorcycle rider training scheme NSW Australia Traffic
Accident Research Unit Traffic Authority NSW 1985 [ RN 85]
Anderson 1980 published data only
Anderson J Ford JL Peck RC Improved motorcyclist licensing and testing project Washington DC National Highway Traffic Safety
Administration (NHTSA) USA 19801ndash217 [
DOTndashHSndash805ndash598]
Billheimer 1998 published data only
Billheimer JW Evaluation of the California Motorcyclist Safety
Program Transportation Research Record 19981640100ndash9 [
TRB Report Number 980652]
Cooper 1988 published data only
Cooper PJ Rothe JP Motorcyclists who they are and why they do
what they do Transportation Research Record 1988116878ndash85
Davis 1997 published data only
David CF Evaluation of community traffic safety programs and
motorcycle operator training programs Connecticut Department of Transportation 199763
Hall 1988 published data only
Hall CJ Buchanan L Engle R Motorcycle rider education evaluationproject Washington DC USA National Highway Traffic Safety
Administration (NHTSA) US Department of Transportation
1988 [ DTNH 22ndash80ndashCndash05123]
Haworth 1997 published data only
Haworth N Smith R Brumen I Pronk N In Monash University
Accident Research Centre editor(s) Case-control study of motorcycle
crashes Victoria Australia Department of Transport and Regional
Development The Federal Office of Road Safety 19971ndash149 [
ISBN 0 642 25501 6 ]
Hurt 1981 published data only
Hurt H Ouellet J Thom D Motorcycle accident cause factors and
identification of countermeasures USA Department of
Transportation 1981 [ Technical Report DOT HS 806ndash862]
Jonah 1982 published data onlylowast Jonah BA Dawson N Bragg B Are formally trained motorcyclists
safer Accident Analysis amp Prevention 198214(4)247ndash55
Jonah BA Dawson NE Bragg BW Evaluation of the Canada
Safety Councilrsquos Motorcycle Training Program Proceedings of theInternational Motorcycle Safety Conference Washington DC
USA 198059ndash93
Kloeden 1994 published data only
Kloeden CN Moore VM McLean AJ Evaluation of the pre-license
training program for motorcyclists in South Australia Walkerville
South Australia Australia NHMRC Road Accident Research Unit
The University of Adelaide 19941ndash37 [ ISBN 0 7308 0607 3]
Kraus 1975 published data only
Kraus JF Riggins RS Franti CE Some epidemiological features of
motorcycle collision injuries American Journal of Epidemiology 1975102(1)74ndash98
Leung 1987 published data only
Leung KS Reding VA Evaluation of the Wisconsin motorcycle rider course Wisconsin Bureau of Policy Planning and Analysis
Wisconsin Department of Transportation 1987
MAIDS 2004 published and unpublished data
MAIDS In-depth investigations of accidents involving powered
two wheelers ACEM (Acemacemeu) 2004 Vol Final report
20179
McDavid 1989 published data only
McDavid JC Lohrmann BA Lohrmann G Does motorcycle
training reduce accidents Evidence from a longitudinal quasi-
experimental study Journal of Safety Research 198920(2)61ndash72
Mortimer 1984 published data only
Mortimer R Evaluation of the motorcycle rider course Accident
Analysis amp Prevention 198416(1)63ndash71
24Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2750
Mortimer 1988 published data only
Mortimer RG A further evaluation of the motorcycle rider courseJournal of Safety Research 198819(4)187ndash96
Osga 1980 published data only
Osga GA An investigation of the riding experiences of MSF rider
course participants Dissertation South Dakota USA The
University of South Dakota 19801ndash224
Perrino 2002 published data only
Perrino CS Ahmed A Callender A Rozier E Cantwell A Stewart
O The role of Marylandrsquos motorcycle rider course in promoting saver
behaviors and attitudes Baltimore USA National Transportation
Center Morgan State University 20021ndash52
Raymond 1979 published data only
Raymond S Tatum S An evaluation of the effectiveness of the RAC
ACU motorcycle training scheme - final report Road Safety ResearchUnit Dept of Civil Engineering University of Salford London
19791ndash131
Satten 1980 published data only
Satten RS Analysis and evaluation of the motorcycle rider courses
in thirteen Illinois counties International Motorcycle Safety
Conference Washington DC 1980145ndash93
Savolainen 2007 published and unpublished data
Savolainen Peter Mannering Fred Additional evidence on the
effectiveness of motorcycle training and motorcyclistsrsquo risk-taking
behaviour Transportation Research Board 2007 Vol March 20
19
Swaddiwudhipong 1998 published data only
Swaddiwudhipong W Boonmak C Nguntra P Mahasakpan P
Effect of motorcycle rider education on changes in risk behavioursand motorcycle-related injuries in rural Thailand Tropical Medicine
and International Health 19983(10)767ndash70
Waller 1992 published data only
Waller FL An analysis of the impact of the Texas Department of
Public Safety Motorcycle Operator Training Program upon
motocycle related traffic accidents Texas State University Public
Administration Program Applied Research Projects 199277
References to studies excluded from this review
Anderson 1980a published data only
Anderson LC Better biking a motorcycle rider improvement
program International Motorcycle Safety Conference Washington
DC USA Motorcycle Safety Foundation 1980231ndash53Aupetit 2008 published and unpublished data
Aupetit S Riff J Buttelli O Espie S Ergonomics research program
on motorcycle learning Contributions to the development of a
motorcycle simulator for training Transport Research Arena
Ljubljana 20089
Awane 1999 published data only
Awane T Integrating simulators in motorcycle safety education
Journal of International Association of Traffic and Safety Sciences 199923(1)26ndash35
Baldi 2005 published data only
Baldi S Baer JD Cook AL Identifying best practices states in
motorcycle rider education and licensing Journal of Safety Research200536(1)19ndash32
Batchler 1988 published data only
Batchler MW Evaluation of the pre-license motorcycle training course at the New South Wales traffic education centre Federal Office of
Road Safety (FORS) 19881ndash122 [ Report No CR 77 ]
Blanchard 2006 published data only
Blanchard HT Tabloski PA Motorcycle safety educating riders at
the teachable moment Mosby (CV) Company 200632330ndash2
Bonnett 2005 published data only
Bonnett J Mature age skills training for experienced motorcycle
riders Motorcycle ridersrsquo association of the Australian Capital
Territory 2005
Braver 2007 published data only
Braver ER et alPersuasion and licensure a randomized controlled
intervention trial to increase licensure rates among Maryland
motorcycle owners Traffic Injury Prevention 20078(1)39ndash46Clayton 1980 published data only
Clayton AB Sudlow DE An evaluation of the effectiveness of a one day training programme for learner motorcyclists Berkshire UK
Transport and Road Research Laboratory 19801ndash23 [ ISSN
0266ndash7045]lowast Clayton AB Sudlow DE An evaluation of the effectiveness of a one-day training programme for learner motorcyclists Berkshire UK
Road Safety Division Berkshire 19801ndash23 [ISSN 0266ndash7045]
Council 1978 published data only
Council FM Desper LP Dutt AK Driver education for motorcycle operation University of North Carolina Highway Safety Research
Center 19781ndash40
Crowther 2005 published data onlylowast Crowther G Engaging with motorcyclists UK police and theBikeSafe Road Safety Programme Published proceedings of the
Australasian Road Safety Research Policing Education Conference
Wellington New Zealand 2005 issue ISBN 0ndash473ndash10636ndash11ndash11
Daniello 2009 published data only
Daniello A Gabler HC Mehta YA Effectiveness of motorcycle
training and licensing Transportation Research Board 88th Annual
Meeting 200913
Dewanti 2007 published data only
Dewanti M Challenging efforts in promoting young motorcyclist
safety in Indonesia Road Safety on Four Continents 14th
International Conference Swedish National Road and Transport
Research Institute 2007
Du Plooy 1990 published data onlyDu Plooy DW Motorcycle training in South Africa The Human
Element International Motorcycle Safety Conference Orlando
Florida USA Motorcycle Safety Foundation 19904-1 to 4-9
Engel 1989 published data only
Engel U Krogsgaringrd Thomsen L [Faeligrdselsundervisning og
propaganda ndash en faeligrdselssikkerhedsfremmende foranstaltning]
Dansk Vejtidsskrift 198912295ndash7
Goldenbeld 2002 published data only
Goldenbeld C Houwing S de Craen S The development of
driving skills of young moped riders [De ontwikkeling van de
rijvaardigheid van jonge bromfietsers] SWOV Institute for Road
Safety Research (Stichting Wetenschappelijk Onderzoek
Verkeersveiligheid) Leidschendam the Netherlands 200263
25Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2850
Goldenbeld 2004 published data only
Goldenbeld C Twisk D de Craen S Short and long term effects of moped training a field experiment Transportation Research Part F 20047(1)1ndash16
Haworth 2000 published data only
Haworth N Smith R Kowadlo N Motorcycle training in Victoria
evaluation of rider training curriculum in Victoria Monash
University Accident Research Centre Reports 2000 Vol 165 issue
3131
Haworth 2005 published data only
Haworth N Mulvihill C Review of motorcycle licensing and
training Monash University Accident Research Centre 2005 [
Report no240]
Haworth 2006 published data only
Haworth N Mulvihill C Rowdon P Teaching old dogs new tricksTraining and older motorcyclists Journal of the Australasian College of Road Safety 200718(4)20ndash5lowast Haworth NL Mulvihill C Rowden PJ Teaching old dogs new
tricks Training and older motorcyclists Australasian Road Safety
Research Policing and Education Conference Gold Coast
Queensland 2006
Hill 1983 published data only
Hill D Motorcycle rider training testing and licensing Motor
Cycle Council of New South Wales Sydney 1983
Imatake 1980 published data only
Imatake Y On the activities in Japan of motorcycle safe riding
promotion committee (MSRPC) - especially the education of
unskilled riders of class 1 motorbikes International Motorcycle
Safety Conference Washington DC USA Motorcycle Safety
Foundation 1980255ndash68
Ingebritsen 1990 published and unpublished data
Ingebrigtsen S [Risikofaktorer ved ferdsel med moped og
motorsykkel] TOslashI-rapport 66 Oslo Transportoslashkonomisk
institutt 1990
Jain 2009 published and unpublished data
Jain A et alTwo wheeler accidents on Indian roads a study from
Mangalore India Elsevier 200916130ndash3
Jonah 1979 published data only
Jonah BA Dawson NE Validation of the motorcycle operator skill
test Accident Analysis amp Prevention 197911(3)163ndash71
Jonah 1981 published data onlyJonah BA Dawson NE Bragg WE Predicting accident
involvement with the Motorcycle Operator Skill Test Accident Analysis amp Prevention 198113(4)307ndash418
Kadar 1990 published data only
Kadar KJ Countermeasures motorcycle rider education delivery
system The Human Element International Motorcycle Safety
Conference Motorcycle Safety Foundation 19906-27 to 6-41
Kelsey 1986 published data only
Kelsey SL Liddicoat C Ratz M Licensing novice motorcyclists a
comparison of Californiarsquos standard test and the MOST II (Motorcycle
Operator Skill Test) administered at centralized testing offices Sacramento California USA California Department of Motor
Vehicles Research and Development Office 19861ndash74
Koch 1980 published data only
Koch H Advanced course for motorcycle riders InternationalMotorcycle Safety Conference Washington DC USA Motorcycle
Safety Foundation 1980269ndash81
Koch 1990 published data only
Koch H Brendicke R How to influence risk taking behaviour
within motorcycle rider training programs The Human Element
International Motorcycle Safety Conference Orlando Florida
USA Motorcycle Safety Foundation 1990Motorcycle Safety
Foundation 1990 p 9-1 to 9-25A
Labbett 2005 published data only
Labbett S Langham M Training can make the problem worse
Road Safety Congress 20051ndash12
Law 2005 published and unpublished data
Law TH Radin Umar RS Zulkaurnain S Kulanthayan S Impactof the effect of economic crisis and the targeted motorcycle safety
programme on motorcycle-related accidents injuries and fatalities
in Malaysia International Journal of Injury Control and Safety
Promotion 200512(1)9ndash21
Lowes 1990 published data only
Lowes BF In-traffic rider training The Human Element
International Motorcycle Safety Conference Orlando Florida
USA Motorcycle Safety Foundation 19907-46 to 7-57
Mahon 1976 published data only
Mahon J Teaching accident avoidance on the motorcycle Journal of Traffic Safety Education 1976231ndash29
Maxwell 1985 published data only
Maxwell GS Evaluation of a Program for Training MotorcycleRiders Evaluation and Program Planning 19858(4)339ndash47
Mayhew 1996 published data only
Mayhew DR Simpson SH Effectiveness and role of driver educationand training in a graduated licensing system Ottowa Canada The
Traffic Injury Research Foundation 19961ndash89 [ ISBN
0ndash920071ndash12ndash0]
McKnight 1987 published data only
McKnight AJ Milburn BG An evaluation of the modified
motorcycle operator skill test exercises NHTSA Report 1987 [
Report No DOT HS 807 212 ]
McPherson 1976 published data only
McPherson K McKnight AJ The development and evaluation of a
motorcycle skill test manual and knowledge test Washington DCUSA National Highway Safety Administration 1976
Mohan 2004 published data only
Mohan D Evidence-based interventions for road traffic injuries in
South Asia Journal of the College of Physicians and SurgeonsPakistan 200414(12)746ndash7
Morgan 1980 published data only
Morgan JK Advanced motorcycle training course content and
study evaluation International Motorcycle Safety Conference
Washington DC USA Motorcycle Safety Foundation 1980
283ndash312
Motoki 1990 published data only
Motoki M Yamazaki S A study on effective motorcycle rider
education The Human Element International Motorcycle Safety
26Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2950
Conference Orlando Florida USA Motorcycle Safety
Foundation 19909-26 to 9-57A
Nairn 1992 published data onlylowast Nairn RJ Motorcycle safety research literature review 1987 to
1991 Federal Office of Road Safety Australia 1993 Vol Report
CR 117
Newland 1980 published data only
Newland R Darby J Automotive practicespre motorcycle rider
program International Motorcycle Safety Conference Washington
DC USA Motorcycle Safety Foundation 1980127ndash9
Newland 1982 published data only
Newland R Motorcycle rider training and licensing requirements
Driver training steering a course for the future Melbourne
Victoria Transport Regulation Board 1982
Newland 1990 published data only
Newland R Motorcycle training and licensing in Victoria 1980-
1990 The Human Element International Motorcycle Safety
Conference Orlando Florida USA Motorcycle Safety
Foundation 1990
Ormston 2003 published data only
Ormston R Dudleston A Pearson S Stradling Evaluation of
Bikesafe Scotland Scottish Executive Social Research Edinburgh
2003
Osga 1979 published data only
Osga GS Ellingstad VS An evaluation of the effectiveness of the 1978
South Dakota motorcycle rider course program South Dakota USA
University of South Dakota 19791ndash89
Panou 2007 published data only
Panou M Bekiaris E A new concept on the integration of driving
simulators in driving training the train-all approach Road Safety
on Four Continents 14th International Conference Swedish
National Road and Transport Research Institute 2007
Peterson 1990 published data only
Peterson A Kadar K Status report 1980-1989 Minnesota rider
education statistics The Human Element International
Motorcycle Safety Conference Orlando Florida USA 199013-
113 to 13-129
Prem 1984 published data only
Prem H Good MC Motorcycle rider skills assessment ACT
Australia Office of Road Safety Department of Transport 1984
1ndash377 [ Report No CR 34 ]
Radin Umar 2006 published data only
Radin Umar RS Motorcycle safety programmes in Malaysia how
effective are they International Journal of Injury Control and Safety Promotion 200613(2)71ndash9
Rockwell 1990 published data only
Rockwell TH Kiger S Carnot M An evaluation of the Ohio
motorcyclist enrichment program Ohio Department of Highway
Safety 1990
Rothe 1987 published data only
Rothe JP A motorcycle riding course Motorcyclists Image and reality Canada Insurance Corporation of British Columbia 1987
59ndash83
Rowden 2007 published data only
Rowden P Watson B Haworth N What can riders tell us aboutmotorcycle training A view from the other side of the fence
Australasian Road Safety Research Policing and Education
Conference Melbourne Australia Melbourne 2007
RSC Victoria 1993 published data only
Road Safety Committee Inquiry into Motorcycle Safety in Victoria
Parliament of Victoria 1993
Russam 1979 published data only
Russam K Motorcycle training and licensing in Japan UK Transport
and Road Research Laboratory (TRRL) 1979 [ Report 916 ]
Saffron 1981 published data only
Saffron DG Driver and rider education training and licensing a
brief review Traffic Accident Research Unit Traffic Authority of
New South Wales NSW Australia Traffic Accident Research UnitTraffic Authority of New South Wales 1981
Schulz 1990 published data only
Schulz U Motorcycle accidents among learner riders Motorcycle
licensing Evaluation of the effectiveness of a Graduated Licensing Scheme Forshungshefte des Institut fur Zweiradsicherheit 1990
Sexton 2004 published data only (unpublished sought but not used)
Sexton B Baughan C Elliott M Maycock G The accident risk of
motorcyclists 2004TRL Report TRL 607(ISSN 09684107)58
Shephard 1986 published data only
Shepherd R Do Motorcycle Safety Foundation programs work
Driver 198619(9)16ndash8
Simpson 1990 published data onlySimpson HM Mayhew DR The promotion of motorcycle safety
training education and awareness Health Education Research
Theory and Practice 19905(2)257ndash64
Sokolov 1990 published data only
Sokolov O Safe motorcycle riding some aspects of motorcycle
riding in the USSR The Human Element International
Motorcycle Safety Conference Motorcycle Safety Foundation
19904ndash37 - 4-48
Steffens 1988 published data only
Steffens U Gawatz R Willmes G Effectiveness of mofa courses
[Wirksamkeit von Mofakursen] Unfall- und sicherheitsforschung
strasenverkehr 198867168
Sudlow 2003 published data only
Sudlow D Scoping study on motorcycle training Road Safety
Research Report London UK The British Institute of Traffic
Education Research Department for Transport 2003 Vol June
[ Report No 36]
Tasmania 1995 published data only
Mannering FL Grodsky LL Motorcycle rider training scheme
Traffic Education Research Centre Armidale Australia Driver
Training Australia Limited 1995 Vol April1ndash56
Tetard 1985 published data only
Tetard C Motorcycle ridersrsquo training to detect indices in an
intersection situation [Lrsquoentrainement a la prise drsquoinformation dans
lrsquoapprentissage de la conduite motocycliste] Le Travail Humain198548(2)147ndash59
27Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3050
Thackray 1980 published data only
Thackray RM Prescott JC Field test of the motorcycle ridercourse International Motorcycle Safety Conference Washington
DC USA Motorcycle Safety Foundation 1980195ndash7
Thompson 1994 published data only
Thompson M Evaluation of compulsory basic training for
motorcyclists Transport Research Laboratory Crowthorne
Berkshire UK 1994 [ Report 63]
Tobler 1990 published data only
Tobler U Individual driving instruction for motorcycle driving
learners the Swiss experiment of motorcycle driving school The
Human Element International Motorcycle Safety Conference
Orlando Florida USA Motorcycle Safety Foundation 1990
Transport 1989 published data only
Transport Communications Road Crash Statistics Australia Bureauof Transport and Communications Economics 198914ndash20
Tuan 2007 published data only
Tuan VA Shimizu T Towards development and evaluation of the
motorcycle drivers re-education program in Vietnam modeling of
motorcycle driverrsquos undesired behaviors 11th World Conference
on Transport Research 200725
Victor 1980 published data only
Victor DJ Traffic safety education for motorcyclists International
Motorcycle Safety Conference Washington DC USA Motorcycle
Safety Foundation 1980221ndash8
Waterer 1990 published data only
Waterer TH Compulsory training for motorcyclists from liberty
to compulsion The Human Element International Motorcycle
Safety Conference Orlando Florida USA Motorcycle Safety
Foundation 19904-49 to 4-57
Weller 1989 published data only
Weller RB Chandler EW Motorcycle safety and motorcycle
education past research and survey results Journal of Public Policy
amp Marketing 1989893ndash108
Wells 1981 published data only
Wells P Observations of some RACACU motorcycle training
schemes Transport and Road Research Laboratory Crowthorne
Berkshire UK 1981
Winn 1986 published data only
Winn G Two important trends in motorcycle safety regulations
rider education and conspicuity improvement American
Motorcyclist Association 1986147ndash52
Wisher 1988 published data only
Wisher G Reid J Motorcycle accidents in Norfolk a three-year study of motorcycle training and the Part 1 test Norfolk UK Road Safety
Division Highways Department Norfolk County Council 1988
Wood 1987 published data only
Wood HT Bowen R Evaluation of the revised learner permit scheme
July 1983 to December 1985 Victoria Australia Road Traffic
Authority 19871ndash27 [ ISBN 0 7306 0247 8]
Yamazaki 1990 published data only
Yamazaki S Nakanishi M Survey on effect of Honda rider
education The Human Element International Motorcycle Safety
Conference Orlando Florida USA 199012-167 to 12-187
References to ongoing studies
Burgess 2010 unpublished data only
Burgess C Broughton P Fylan F Stradling Interim evaluation of
the UKs National Ride program In L Dorn editor(s) Driver
behaviour and Training Vol IV Aldershot Ashgate Publishing
Limited 2010
Additional references
Anderson 1980
Anderson J Ford JL Peck RC Improved motorcyclist licensing and
testing project Volume I and II National Highway Traffic Safety
Administration US department of Transportation Washington
DC 1980
Bjornstig 1985
Bjornstig U Bylund P Lekander T Brorsson B Motorcycle fatalitiesin Sweden Acta Chirurgica Scandinavica 1985151(7)577ndash81
Clayton 1990
Clayton A Sudlow DE An evaluation of the effectiveness of a one-
day training program for learner motorcyclists The Human
Element International Motorcycle Safety Conference Orlando
Florida USA 199013-105 to 13-111
Cochrane Collaboration 2008
Higgins JP Green S Cochrane Handbook for Systematic Reviews of
Interventions 500 The Cochrane Collaboration 2008
Elliot 2003
Elliott MA Baughan CJT Broughton J Grayson G Knowles J et
alMotorcycle Safety a Scoping Study TRL [ Report No
0968ndash4107]Fleiss 2004
Fleiss JL Levin B Paik MC Statistical methods for rates and proportions 3rd Edition John Wiley amp Sons Inc 2004 [ ISBN
978ndash0ndash471ndash52629ndash2 ]
FORS 1999
FORS Federal Office of Road Safety Road risk for sober licensed
motorcyclists Vol Monograph 27 Federal Office of Road Safety
19991ndash2
Haworth 2005
Haworth N Mulvihill C Review of motorcycle licensing and training
Victoria Australia Monash University Accident Research Centre
20051ndash83 [ Report No 240]
Huwiler-Muumlntener 2002Huwiler-Muumlntener K Juumlni P Junker C Egger M Quality of
Reporting of Randomized Trials as a Measure of Methodologic
Quality JAMA 20022872801ndash4
Johnston 2008
Johnston P Brooks C Savage H Fatal and serious road crashes
involving motorcyclists Road Safety Research and Analysis Report
ATSB Monograph 20 Canberra Department of Infrastructure
Transport Regional Development and Local Government 0205
081ndash26
Juumlni 2001
Juumlni P Altman DG Egger M Systematic reviews in health care
Assessing the quality of controlled clinical trials BMJ 2001323
42ndash6
28Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3150
Kopits 2003
Kopits E Cropper M Traffic fatalities and economic growth TheWorld Bank Development Research Group Infrastructure and
Environment Policy research working paper 3035 2003
Leung 1983
Leung P Chau BTW Motorcycle accidents in Hong Kong pattern
of injury treatment causative factors and socio-economic
implications Injury 198315(1)23ndash9
Liu 2004
Liu B Ivers R Norton R Blows S Lo SK Helmets for preventing
injury in motorcycle riders Cochrane Database of Systematic Reviews 2004 Issue 2 [DOI 10100214651858CD004333pub3]
Mohan 2002
Mohan D Road safety in less-motorized environments future
concerns International Journal of Epidemiology 200231(3)527ndash32Mortimer 1982
Mortimer RG An evaluation of the motorcycle rider course
Proceedings of the 20th conference of the American Association for
Automotive Medicine Arlington Heights IL USA American
Association for Automotive Medicine 198277ndash96
Mullin 1997
Mullin BT Risk factors for motorcycle injury the role of age
gender experience training and alcohol PhD thesis Auckland
New Zealand University of Auckland 1997
Noordzij 2001
Noordzij PC Forke E Brendicke R Chinn BP Integration of needs
of moped and motorcycle riders into safety measures review and
statistical analysis in the framework of the European research
project PROMISING SWOV Institute for Road Safety Research
Leidschendam The Netherlands 2001 [ Report No Dndash2001ndash5]
Osga 1979
Osga G Ellingstad VS An evaluation of the effectiveness of the 1978 South Dakota Motorcycle Riders Course Program South Dakota
University of South Dakota Vermillion Human Factors Lab 1979
Oxford English Dictionary Online
httpoxforddictionariescom Oxford English Dictionary Online
Oxford University Press 2010
Peden 2004
Peden M Scurfield R Sleet D World Report on Road TrafficInjury Prevention Geneva World Health Organization Available
at httpwwwwhointviolence˙injury˙preventionpublications
road˙trafficworld˙reporten 2004Chapter 3
Radin Umar 1995
Radin Umar R Macakay G Hills B Preliminary analysis of
motorcycle accidents Short-term impacts of the running
headlights campaign and regulations in Malaysia J Traffic Med 199523(1)17ndash27
Tsai 1995
Tsai YJ Wang JD et alCase-control study of the effectiveness of
different types of helmets for the prevention of head injuries among
motorcycle riders in Taipei Taiwan American Journal of
Epidemiology 1995142(9)974ndash81
Vis 1995
Vis AA In-depth study of the hazards of motorcycling [De
onveiligheid van motorrijden nader bekeken] SWOV Institute for
Road Safety Research Leidschendam the Netherlands 1995 [
Report No Rndash95ndash69]
Weaver 1990
Weaver EA Motorcycle safety education in the United States
Motorcycle Safety Foundation Orlando Florida USA 1990
Winn 1987
Winn G Two important trends in motorcycle safety regulations
rider education and conspicuity improvement No more
information available Detroit MI US Society of Automotive
Engineers 1987147ndash52
Yuan 2000
Yuan W The effectiveness of the rsquoride-brightrsquo legislation for
motorcycles in Singapore Accident Analysis amp Prevention 200032
(4)559ndash63
Zambon 2006
Zambon F Hasselberg M Factors affecting the severity of injuries
among young motorcyclists a Swedish nationwide cohort study
Traffic Injury Prevention 20067(2)143ndash9lowast Indicates the major publication for the study
29Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3250
C H A R A C T E R I S T I C S O F S T U D I E S
Characteristics of included studies [ordered by study ID]
Adams 1985
Methods Non-randomised trial
Participants Initial sample size
Total of 162 Australia Post telegram delivery riders
Not clear how many in intervention and control groups
Analysis sample size
Intervention group 63Control group 35
Total 98 (61)
Interventions No formal training vs voluntary rider training course of 12 hours over 2 consecutive days (theory 4 hours practice
8 hours)
Intervention integrity
No information given
Outcomes Australia Post recorded crashes
Length of follow up 1 year
Notes Study conducted in NSW Australia
Anderson 1980
Methods Randomised trial
Participants Initial sample size
Total 41188 motorcycle licence applicants
Analysis sample size
Intervention group with skills training 13160
Intervention group without skills training 12634
Control group 15080
Total 40874 (99)
Interventions Standard licensing skill test vs the Motorcycle Operators Skill Test (MOST) with or without a training component
MOST is not to be confused with rider training because it is a licensing test The training component was for
participants who failed the skill test on the first attempt was of 3 hours duration and consisted of instructions about
the skills needed to pass the skill test The training consisted of 30 min theory and 2 frac12 hours practice on the
motorcycle
Intervention integrity
33 in MOST with skills training 31 in MOST without skills training and 43 in control group obtained their
motorcycle licence in the end
Outcomes Police recorded crashes and offences
Length of follow up 3 years
30Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3350
Anderson 1980 (Continued)
Notes Study conducted in California USA
Billheimer 1998
Methods Cohort study
Participants Initial sample size
Intervention group Unclear numbers of CMSP graduates and ERC graduates
Control group Over 16000 untrained motorcyclists
Analysis sample size
1139 matched CMSP-pairs and 1182 ERC-pairs
Interventions No formal training vs the mandatory (for those younger than 18 years) pre-licence 16-hours basic California
Motorcyclist Safety Program (CMSP) and the 8-hour Experienced Rider Course (ERC)
Intervention integrity
No information given
Outcomes Police recorded crashes and offences
Length of follow up 2 years
Notes Study conducted in California USA
Cooper 1988
Methods Cohort study
Participants Initial sample size
Intervention group 418 graduates and 57 who failed
Control group 402
Analysis sample size
Intervention group unclear
Control group unclear
Total 863 (984)
Interventions No formaltrainingvs non-mandatory pre-licence rider trainingThe content of thetrainingwas notfurther explainedIntervention integrity
Participants had all completed the rider training
Outcomes Police recorded crashes and offences
Length of follow up 7 years
Notes Study conducted in British Columbia Canada
31Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3450
Davis 1997
Methods Retrospective cohort study
Participants Initial sample size
Intervention group 9320 graduates
Control group unclear
Analysis sample size
Intervention group 9320 graduates
Control group unclear
Interventions No formal training vs the non-mandatory Connecticut Motorcycle Operator Training Program The content of
training was not further explained
Intervention integrity Participants had all completed the rider training
Outcomes Police recorded crashes severity of crashes
Length of follow up 1 year
Notes Study conducted in Connecticut USA
Hall 1988
Methods Randomised trial
Participants Initial sample sizeIntervention group A 6248
Intervention group B 6616
Intervention group C 6996
Control group 6604
Total 26464 motorcycle licence applicants
Analysis sample size
Intervention group A 6012 (96)
Intervention group B 6428 (97)
Intervention group C 6778 (97)
Control group 6502 (98)
Interventions Standard licensing skill test vs the Motorcycle Operator Skill Test (MOST II) without training (A) MOST II with
3 hours training (B) or MOST II with 20 hours training (C) The 3 hours training comprised instructions aboutthe skills needed to pass the skill test (30 min theory and 2 frac12 hours practice on the motorcycle) The 20 hours
training consisted of the Motorcycle Safety Foundationrsquos Motorcycle Rider Course (MRC) comprising of 8 hours
theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-road practice
Intervention integrity
25 and 26 obtained their motorcycle licence in the lsquoMOST 3-hour trainingrsquo and lsquoMOST 20-hour trainingrsquo
intervention arm respectively In the lsquoMOST testrsquo intervention arm 29 obtained their licence and in the control
group 35
Outcomes Police recorded crashes and offences
Length of follow up 3 years
32Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3550
Hall 1988 (Continued)
Notes Study conducted in New York USA
Haworth 1997
Methods Population based case-control study
Participants Cases Motorcycle riders or pillion riders involved in motorcycle crash (on road) Recruited from hospital and police
records Not clear whether all cases (especially fatalities) were included
Community controls Multiple matched sample of motorcycle rider or pillion who passed the crash site at the same
time of day and day of week that the crash occurred
Initial sample size
Cases 336
Controls 1195
Analysis sample size
Cases 234 (70)
Controls 561 (47)
Interventions Basic handling skills course beginner intermediate or advanced course measured through self-report not validated
Outcomes Injuries with an Injury Severity Score of 5 or greater or admission for at least four hours and fatalities from a
motorcycle crash (from hospital and police records)
Length of follow up two years or more prior to the interview
Notes Study conducted in Victoria Australia
Hurt 1981
Methods Population based case-control study
Participants Cases Motorcycle riders who had a motorcycle crash that was reported to police (traffic accident reports) Not all
cases included
Community controls Multiple matched sample of motorcycle riders who passed the crash site at the same time of
day and day of week as the crash occurred but 2 years later
Initial sample size
Cases 4500 notifications of crashes to police (only 50 received in time research team able to respond to 1126 for
unreported reasons data collection completed for 900)
Controls Number approached unreported 2310 consented and interviewed
Analysis sample size
Cases 808 (808900 data on rider training) - 18 of initial sample
Controls 670 (6702310 data on rider training) - lt 29 of initial sample
Interventions Training experience self taught friends-family school-club formal-professional others measured through self-
report validated
Outcomes Injuries and fatalities from a motorcycle crash (from hospital records and coronerrsquos office)
33Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3650
Hurt 1981 (Continued)
Notes Study conducted in Los Angeles USA
Jonah 1982
Methods Cohort study
Participants Initial sample size
Intervention group
2310 licensed MTP graduates who are active riders
Control group
2029 licensed motorcyclists who are active riders
Non-licensed and non-active riders excluded
Analysis sample size
Intervention group
811 interviews (35) 533 driver records (23)
Control group 1080 interviews (71) 905 driver records (45)
Interventions No training vs the non-mandatory Canada Safety Council Motorcycle Training Program (MTP) comprising of 4
hours theory and 16 hours practise (6 hrs on-road) over 2 weekends
Intervention integrity
Participants had all successfully completed the MTP
Outcomes Self-reported and police reported crashes and offencesLength of follow up 4 years
Notes Study conducted in Ontario Canada
Kloeden 1994
Methods Cohort study
Participants Initial sample size
Intervention group 2533 licence applicants
Control group Unclear
Sample of riders obtaining probationary licenceIntervention group 1625 (64)
Control group 5015 (unclear)
Non-active riders excluded from control group (~66)
Analysis sample size
Intervention group 631 (39)
Control group 318 (6)
Interventions No formal training vs the mandatory pre-licence Ridersafe course consisting of two levelsthe first is a training course
in the basics of motorcycle riding where a pass entitles the rider to obtain a learnerrsquos permit The second is a skill
test where a pass entitles the rider to obtain a probationary motorcycle licence The training course consists of two 4
hour training sessions on separate days involving instruction videos and actual riding on motorcycles The skill test
consists of a 4 hour preparation course of 74 minutes of lecturediscussion 38 minutes of video and 75 minutes of
34Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3750
Kloeden 1994 (Continued)
on-course riding which is the testing phase
Intervention integrity
Unclear what percentage in the Ridersafe group and the control group obtained their probationary licence halfway
and their full motorcycle licence in the end
Outcomes Police reported crashes
Length of follow up 1 year
Notes Study conducted in SA Australia
Kraus 1975
Methods Population based case-control study
Participants Cases All motorcycle riders with a confirmed medically treated motorcycle injury (including fatalities) from all
hospitals in study base and peripheral area hospitals and all deaths due to motorcycle crashes in the study base All
serious injuries (leading to hospitalisation) and fatalities likely to be included
Community controls A random sample of registered motorcycle owners in the same temporary and geographical
study base as the cases arose from
Initial sample size Cases 1273
Controls 738
Analysis sample size Cases 741 (58)
Controls 434 (59)
Interventions Motorcycle rider training (not specified) measured through self-report
Outcomes Injuries leading to hospitalisation fatalities
Length of follow up not specified
Notes Study conducted in Sacramento county USA
Leung 1987
Methods Non-randomised trial
Participants Initial sample sizeIntervention group unknown number of MRC graduates
Control group unknown number of licensed motorcyclists
Analysis sample size
Intervention group 2941
Control group 43094
Interventions No formal training vs non-mandatory Wisconsin motorcycle rider course a 22 hours off-street programme that
teaches the basic rules of safe motorcycling
Intervention integrity
50 in the MRC group and 100 in the control group obtained their motorcycle licence in the end
35Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3850
Leung 1987 (Continued)
Outcomes Police recorded crashes and offences
Length of follow up up to 3 years
Notes Study conducted in Wisconsin USA
MAIDS 2004
Methods Population based case-control study
Participants Cases Motorcycle riders who had a motorcycle crash Unclear whether all cases included
Community controls Motorcycle riders identified at petrol stations Unclear whether matched to cases
Initial sample size
Cases not reported
Controls not reported
Analysis sample size
Cases 921 (Italy 200 Netherlands 200 Spain 121 Germany 250 France 150)
Controls 923 (Italy 200 Netherlands 200 Spain123 Germany 250 France 150)
Interventions Pre-license rider training or additional training (non-compulsory)
Outcomes Crashes
Notes Study in five sampling areas located in five European countries (France Germany Netherlands Spain and Italy)
McDavid 1989
Methods Cohort study
Participants Sample size after first match
Intervention group 346 male training graduates
Control group 346 male motorcycle riders licensed in same year
Sample size after second match
Intervention group 139 (40)
Control group 139 (40)
Analysis sample size
Intervention group 139 (40)
Control group 139 (40)
Interventions No formal training vs the non-mandatory pre-licence British Columbia Safety Councilrsquos 37-hour motorcycle safety
programme
Intervention integrity
Participants had all completed and passed the training
Outcomes Police reported crashes injurious crashes
Length of follow up 1-6 years
36Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3950
McDavid 1989 (Continued)
Notes Study conducted in British Columbia Canada
Mortimer 1984
Methods Cohort study
Participants Initial sample size
Intervention group 608 MRC graduates
Control group 360 motorcycle riders
Response rates
Intervention group 360 (59)
Control group 324 (90)
Non-active riders excluded
Analysis sample size
Intervention group 213 (35)
Control group 303 (84)
Interventions No formal training vs the non-mandatory Motorcycle Safety Foundationrsquos Motorcycle Rider Course comprising of
8 hours theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-road practice
Intervention integrity
Participants had all completed and passed the MRC
Outcomes Self-reported crashes and offencesLength of follow up 1 year
Notes Study conducted in Illinois USA
Mortimer 1988
Methods Cohort study
Participants Initial sample size
Intervention group 2307 MRC graduates
Control group 556 active licensed motorcycle riders
Response ratesIntervention group 1298 (56)
Control group 500 (90)
Non-active riders excluded
Analysis sample size
Intervention group 913 (40)
Control group 500 (90)
Interventions No formal training vs the non-mandatory Motorcycle Safety Foundationrsquos Motorcycle Rider Course comprising of
8 hours theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-road practice
Intervention integrity
Participants had all completed and passed the MRC
37Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4050
Mortimer 1988 (Continued)
Outcomes Self reported crashes and offences
Length of follow up 1 year
Notes Study conducted in Illinois USA
Osga 1980
Methods Cohort study
Participants Initial sample sizeIntervention group 1163 MSF MRC participants invited 818 (74) responded
Control group 329 motorcycle riders responded to recruitment strategies
Non-active riders excluded from analysis
Analysis sample size
Intervention group 595 (73)
Control group 329 (100)
Interventions No formal training vs the non-mandatory Motorcycle Safety Foundationrsquos Motorcycle Rider Course (MRC) com-
prising of 8 hours theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-road
practice
Intervention integrity
Participants had all successfully completed the MRC
Outcomes Self-reported crashes police recorded crash reports
Length of follow up 6 months - 3 12 years
Notes Study conducted in South Dakota USA
Perrino 2002
Methods Cohort study
Participants Initial sample size
Intervention group 2144 (unclear) MRC graduates (motorcycle licence applicants)
Control group 2601 (unclear) regular licensing test graduates (motorcycle licence applicants)Analysis sample size
Intervention group 1070 (unclear)
Control group 867 (unclear)
Interventions No formal training vs the mandatory (for those younger than 18 years) pre-licence Maryland Motorcycle Rider
Course (MRC) consisting of 8 hours theory and 12 hours practice
Intervention integrity
No information given
Outcomes Self-reported and police recorded crashes and offences
Length of follow up 1 year
38Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4150
Perrino 2002 (Continued)
Notes Study conducted in Maryland USA
Raymond 1979
Methods Cohort study
Participants Initial sample size
Intervention group
258 volunteers mainly (but not all) novice riders
Control group
1104 provisional licence applicants
Analysis sample size
Intervention group 190 (74)
Control group 627 (57)
Interventions No formal training vs the non-mandatory RACACU (Royal Automobile ClubAuto-Cycle Union) training scheme
The training program consisted of either 24 one-hour sessions or 13 two-hour sessions Twelve hours of the training
focused on theory about basic operational procedures traffic laws and highway code The practical session focused
on basic operational skills riding on a training area and riding in traffic with an instructor A proficiency test was
administered at the end of the course
Intervention integrity
No information given
Outcomes Self-reported crashes (including minor non-injury non-damage and off-road crashes)
Length of follow up 1 year
Notes Study conducted in the UK
Satten 1980
Methods Cohort study
Participants Initial sample size
Intervention group interviewers kept interviewing until they had 100 riders selected from a population of 891
enrollees in the MSF MRC course (unknown how many were asked and how many refused to participate)Control group unknown number of registered motorcycle riders
Non-active riders excluded from analysis
Analysis sample size
Intervention group 71
Control group 100
Interventions No formal training vs the non-mandatory Motorcycle Safety Foundationrsquos (MSF) Motorcycle Rider Course (MRC)
comprising of 8 hours theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-
road practice
Intervention integrity
Participants had all successfully completed the MRC
39Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4250
Satten 1980 (Continued)
Outcomes Self-reported crashes injuries and offences
Length of follow up 3 years
Notes Study conducted in Illinois USA
Savolainen 2007
Methods Cohort study
Participants Initial sample size
Intervention group 4000 riders that had conducted the MSF MRC course (passed or failed)
Control group 4000 motorcycle owners or permit or endorsement holders from the Indiana Bureau of Motor
Vehicles database
Analysis sample size
Intervention group 739 (18)
Control group 588 (15)
Interventions Intervention integrity
No information given
Outcomes Probability of crash involvement
Length of follow up up to 5 years
Notes Study conducted in Indiana USA
Swaddiwudhipong 1998
Methods Randomised trial
Participants Initial sample size
Intervention group
all 16 villages of 3 sub-districts
(participants were practising motorcyclists licensed and unlicensed)
Control group
18 villages of 3 other sub-districtsAnalysis sample size
Intervention group 3 districts (100)
Control group 3 districts (100)
Interventions No rider education vs non-mandatory community based rider education consisting of theory on motorcycle injury
risk helmet use traffic laws vehicle regulations and traffic signs and a written and skill test leading to a licence
Intervention integrity
No information given
Outcomes Injuries fatalities
Length of follow up 2 years
40Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4350
Swaddiwudhipong 1998 (Continued)
Notes Study conducted in Tak Province rural Thailand
Waller 1992
Methods Cohort study
Participants Initial sample size
Intervention group 349 riders who succesfully completed the Department of Public Safety motorcycle operator
training in 1988
Control group 349 motorcyclists who never completed the DPS motorcycle operator training
Analysis sample size
Intervention group 349
Control group 349
Interventions No DPS motorcycle operator training vs non-mandatory DPS motorcycle operator training The DPS training is
the Motorcycle Safety Foundationrsquos Motorcycle Rider Course comprising eight hours of theory (lectures discussions
films and multiple choice tests) a textbook and 12 hours off-road practice
Intervention integrity
All 349 riders succesfully completed the rider training
Outcomes Police recorded crashes
Length of follow up 3 years
Notes Study conducted in
Characteristics of excluded studies [ordered by study ID]
Study Reason for exclusion
Anderson 1980a Study type review and process evaluation of a rider course
Aupetit 2008 Outcome proposal of a rider training curriculum that combines traditional training and a motorcycle riding
simulator
Awane 1999 Study type review on use of simulators in rider and driver training
Baldi 2005 Intervention not rider training but intervention integrity of rider training availabilty of rider education
implementation of graduated licensing system etc on a state-wide level
Batchler 1988 Study type description of a training course
Blanchard 2006 Study type not a study assessing the association between rider training and outcomes
Bonnett 2005 Outcome qualitative data regarding satisfaction with the training programme
41Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4450
(Continued)
Braver 2007 Outcome licensure rates
Clayton 1980 Outcome knowledge skills attitude
Council 1978 Outcome crash reduction in ANY vehicle not motorcycles exclusively
Crowther 2005 Study type qualitative study
Daniello 2009 Study type review of the effectiveness of motorcycle training and licensing
Dewanti 2007 Intervention emotional characteristics behaviour (ie not motocycle rider training)
Du Plooy 1990 Study type review of types of rider training courses in South Africa
Engel 1989 Study type uncontrolled before-after study
Goldenbeld 2002 Outcome skills
Goldenbeld 2004 Outcome knowledge and skills
Haworth 2000 Intervention balance between attitudinal andvehicle control skills acrosscurriculla of motorcycle rider training
courses
Haworth 2005 Study type review
Haworth 2006 Study type cross-sectional survey
Hill 1983 Study type review and description of training system
Imatake 1980 Study type course descriptionreview
Ingebritsen 1990 Intervention risk factors for motorcycle crashes
Jain 2009 Study type cross-sectional descriptive study of motorcycle crashes over a 5 year period
Jonah 1979 Study type cross-sectional survey Outcome scores on motorcycle operator skill test
Jonah 1981 Intervention evaluation of a motorcycle rider skill test not motorcycle rider training
Kadar 1990 Study type review of components of rider training courses
Kelsey 1986 Intervention evaluation of a motorcycle rider skill test not motorcycle rider training
Koch 1980 Study type reviewcourse description
Koch 1990 Study type description of course content
42Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4550
(Continued)
Labbett 2005 Study type essay about possible explanations for motorcycle deaths
Law 2005 Intervention targeted motorcycle safety programme not specifically rider training
Lowes 1990 Study type review and descriptions of rider training
Mahon 1976 Study type review of motorcycle rider training methods
Maxwell 1985 Outcomesrecommendations forprogramme operatorsand otherson howthe motorcycle trainingprogramme
works and how it can be improved
Mayhew 1996 Study type review
McKnight 1987 Study type evaluation of methodological and administrative aspects of a motorcycle operator skill test
McPherson 1976 Outcome knowledge and skills
Mohan 2004 Study type review of evidence of interventions to reduce road traffic injuries
Morgan 1980 Study type reviewcourse description
Motoki 1990 Study type proposal and development of rider training curriculum
Nairn 1992 Study type literature review
Newland 1980 Study type description of development of motorcycle safety course
Newland 1982 Study type descriptionreview of a rider training course
Newland 1990 Study type uncontrolled BeforeAfter study
Ormston 2003 Outcome changes in attitude and self reported behaviour
Osga 1979 Outcome knowledge attitudes and skills
Panou 2007 Intervention new simulation tool to teach driverrider
Peterson 1990 Study factor demographic details of course participants completing the course vs those not completing the
course
Prem 1984 Study type experimental
Study factor examines association between skills and performance in tests
Radin Umar 2006 Intervention motorcycle safety programme not specifically rider training
Rockwell 1990 Study type methodologically flawed study type
43Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4650
(Continued)
Rothe 1987 Study type qualitative study of rider training in general
Rowden 2007 Study type qualitative study
RSC Victoria 1993 Study type review of motorcycle rider training and education in Victoria
Russam 1979 Study type review of training programmes in Japan
Saffron 1981 Study type literature review
Schulz 1990 Study factor predictors of motorcycle crashes
Sexton 2004 Study type cross-sectional analytic analysis
Shephard 1986 Study factor reviewdescription of programme
Simpson 1990 Study type review of motorcycle rider trainingeducation
Sokolov 1990 Study factor review of motorcycle riding courses in USSR
Steffens 1988 Outcome self-reported behaviour experiences and researcher observed riding behaviour
Sudlow 2003 Study type review of motorcycle training in the UK
Tasmania 1995 Study type cross-sectional descriptiveprocess evaluation
Tetard 1985 Study type experimental
Outcome detection of indices
Thackray 1980 Outcome instructional effectiveness user acceptance administrative feasibility
Thompson 1994 Study type cross-sectional survey
Outcome attitudes and knowledge
Tobler 1990 Study type review of course curricula
Transport 1989 Intervention syllabus for rider training
Tuan 2007 Study type cross-sectional study (survey) which does not assess the effects of an intervention but examines
motocyclistsrsquo behaviour
Victor 1980 Study type review
Waterer 1990 Study type review of UK rider training issues
Weller 1989 Study type review of studies evaluating motorcycle safety and education measures
44Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4750
(Continued)
Wells 1981 Study type description of different methods of training
Winn 1986 Study type review of rsquohistoryrsquo of rider training
Wisher 1988 Study type cross-sectional survey
Wood 1987 Study type Interrupted Time Series but did not meet the criteria for an ITS as published by the Cochrane
Effective Practice and Organisation of Care Group
Yamazaki 1990 Study type cross-sectional survey
Outcome riding skills skills retention and attitude
Characteristics of ongoing studies [ordered by study ID]
Burgess 2010
Trial name or title UKs National Ride program
Methods Cohort study
Participants Motorcycle offenders
Interventions Police led diversion program
Outcomes Self reported attitudinal and behavioural measures
Starting date
Contact information
Notes Expected to be completed in early 2010
45Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4850
D A T A A N D A N A L Y S E S
This review has no analyses
A P P E N D I C E S
Appendix 1 Search strategy
We searched the following electronic databases
bull Cochrane Injuries Group Specialised Register (searched 22 Sept 2008)
bull CENTRAL (The Cochrane Library 2008 Issue 3)
bull TRANSPORT 1988 to 200706 (latest date available)
bull ATRI Australian Transport Index (to Sept 2008)
bull MEDLINE (Ovid SP) 1950 to Sept 2008
bull EMBASE (Ovid SP) 1980 to Sept 2008
bull CINAHL (EBSCO) 1982 to Sept 2008
bull WHOLIS (World Health Organization Library Information System) (to Sept 2008)
bull PsycINFO (Ovid SP) 1806 to Sept 2008
bull LILACS (BIREME) (Latin American and Caribbean Health Sciences) (1982 to Sept 2008)
bull ISI Web of Science Social Sciences Citation Index (SSCI) 1970 to Sept 2008
bull ERIC (searched 25 Sept 2008)
bull ZETOC (searched 25 Sept 2008)
bull SIGLE (searched 25 Sept 2008)
All of the databases were searched using the following text-word search strategy The strategy was adapted using either rsquorsquo or rsquo$rsquo as atruncation symbol
1 motor-cycl or motorcycl or motor-bik or motorbik or motor-scooter or motorscooter or scooter or moped or motocycl
or motocicl or moto
2 train or educat or instruct or program or course or qualif or learn or improv
3 1 and 2
H I S T O R Y
Protocol first published Issue 2 2005
Review first published Issue 10 2010
46Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4950
C O N T R I B U T I O N S O F A U T H O R S
Katina Kardamanidis was the first author of the protocol conducted the literature search extracted full studies assessed the studies
according to inclusion criteria assessed quality of the studies compared findings with the second author discussed disagreements with
the third author wrote the drafts and the final review
Rebecca Ivers was the second author of the protocol assessed the studies according to inclusion criteria assessed quality of the studies
compared findings with the first author edited the final draft and provided overall guidance for the process
Alexandra Martiniuk assessed quality of the studies compared findings with the first author discussed disagreements with the third
author and helped write the first draft and the final review
Mark Stevenson contributed to the protocol reviewed methods of studies and contributed to the final draft
Katrina Thistlethwaite conducted a second literature search extracted full studies and contributed to the final draft
D E C L A R A T I O N S O F I N T E R E S T
None known
S O U R C E S O F S U P P O R T
Internal sources
bull No sources of support supplied
External sources
bull National Health and Medical Research Council of Australia Australia
Salary support for Martiniuk Ivers and Stevenson
bull University of Sydney Sesqui Research Grant Australia
Direct research costs and salary support for Kardamanidis
D I F F E R E N C E S B E T W E E N P R O T O C O L A N D R E V I E W
There are no differences between the protocol and the review
I N D E X T E R M S
Medical Subject Headings (MeSH)
lowastMotorcycles Accident Prevention [lowastmethods] Accidents Traffic [lowastprevention amp control] Licensure Program Evaluation Wounds
and Injuries [mortality lowastprevention amp control]
47Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 5050
MeSH check words
Humans
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 450
Main results
We reviewed 23 studies three randomised trials two non-randomised trials 14 cohort studies and four case-control studies Five
examined mandatory pre-licence training 14 assessed non-mandatory training three of the case-control studies assessed lsquoanyrsquo type of
rider training and one case-control study assessed mandatory pre-licence training and non-mandatory training The types of assessed
rider training varied in duration and content
Most studies suffered from serious methodological weaknesses Most studies were non-randomised and controlled poorly for con-
founders Most studies also suffered from detection bias due to the poor use of outcome measurement tools such as the sole reliance
upon police records or self-reported data Small sample sizes and short follow-up time after training were also common
Authorsrsquo conclusions
Due to the poor quality of studies identified we were unable to draw any conclusions about the effectiveness of rider training on crash
injury or offence rates The findings suggest that mandatory pre-licence training may be an impediment to completing a motorcycle
licensing process possibly indirectly reducing crashes through a reduction in exposure It is not clear if training (or what type) reducesthe risk of crashes injuries or offences in motorcyclists and a best rider training practice can therefore not be recommended As some
type of rider training is likely to be necessary to teach motorcyclists to ride a motorcycle safely rigorous research is needed
P L A I N L A N G U A G E S U M M A R Y
Motorcycle rider training for preventing road traffic crashes
Riders of motorcycles (a two-wheeled vehicle that is powered by a motor and has no pedals - Oxford English Dictionary Online)
especially novice riders have an increased risk of being involved in fatal crashes compared to other road users Motorcycle rider training
could be an important way of reducing the number of crashes and the severity of injuries
The authors of this review examined all research studies that report an evaluation of the effectiveness of motorcycle rider coursesin reducing the number of traffic offences motorcycle rider crashes injuries and deaths This review included 23 research studies
including three randomised trials two non-randomised trials 14 cohort studies and four case-control studies The types of rider training
that were evaluated varied in content and duration The findings suggest that mandatory pre-licence training may present a barrier to
completing a motorcycle licensing process thus possibly indirectly reducing crash injury death and offence rates through a reduction
in exposure to riding a motorcycle However on the basis of the existing evidence it is not clear if (or what type of) training reduces
the risk of crashes injuries deaths or offences in motorcyclists and the selection of the best rider training practice can therefore not be
recommended
It is likely that some type of rider training is necessary to teach motorcyclists basic motorcycle handling techniques and to ride a
motorcycle safely It is therefore important that further research work be conducted to rigorously evaluate motorcycle rider training
courses particularly in low income countries where the main burden of motorcycle injuries and deaths occur
B A C K G R O U N D
Each year anestimated 12million peopleare killedin road crashes
worldwide and an additional 50 million people are injured Of
these deaths 85 occur in low-income and middle-income coun-
tries It has been predicted that between the year 2000 and 2020
without increased efforts the number of road traffic deaths in
high-income countries will decrease by approximately 30 yet
in low- and middle-income countries the number is expected to
increase by over 80 (Kopits 2003 Peden 2004) Road traffic
injuries are likely to become the third leading cause of a global
burden of disease and injury by 2020 if further action is not taken
(Peden 2004)
Riding a motorcycle (a two-wheeled vehicle that is powered by a
motor and has no pedals - Oxford English Dictionary Online) is
associated with a higher risk of fatalities compared with driving a
car (Leung 1983 Bjornstig 1985 Radin Umar 1995 Tsai 1995
FORS 1999 Yuan 2000 Johnston 2008) In Australia motorcy-
2Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 550
clesaccounted for45 of allAustralian passenger vehicle registra-
tions and 09 of vehicle kilometres travelled in 2007 Howevermotorcycle riders accounted for approximately 15 of all road
crash deaths and an even higher proportion of serious injuries
Per distance travelled the Australian rate of motorcyclist deaths
was approximately 30 times the rate for car occupants The cor-
responding rate for a serious injury was approximately 41 times
higher Similar elevated rates were also found in other developed
countries (Johnston 2008) In low and middle-income countries
the motorcycle is one of the most frequently used means of trans-
port and consequently motorcyclists constitute a large proportion
of road crash victims (Mohan 2002)
The lack of experience of many motorcycle riders has been cited
as one potential cause of motorcycle crashes (Winn 1987 Mullin1997) Training of motorcycle riders has therefore been suggested
as an important countermeasure for reducing both the number of
crashes and the severity of injury (Vis 1995 Noordzij 2001)
Pre-licence motorcycle rider trainingcourses have been made com-
pulsory in several countries (Mullin 1997) Pre-licence training
courses are aimed at providing motorcycle riders with the skills
and knowledge needed to improve control of the motorcycle and
to identify and avoid critical situations in traffic (Elliot 2003)
Many pre-licence motorcycle rider training courses are aimed at
novice riders and in many jurisdictions in high-income countries
such as Australia the USA and Canada the completion of such
courses is a mandatory requirement of licensing
Post-licence rider training is not linked with licensing and such
training is generally not compulsory It usually focuses on riding
skills and other topics that have not been covered or dealt with
sufficiently in pre-licence rider training Post-licence rider train-
ing is used to reinforce fundamental skills and knowledge or to
provide some training for motorcyclists who are self-taught and
have never received formal training (Anderson 1980)
Estimates of the number of motorcyclists attending post-licence
rider training are not readily available but existing data suggest
such courses are not well attended although attendance varies by
country In the USA 24 of the total number of motorcyclists
trained in 1988 and 1989 attended an experienced rider course
(Weaver 1990)
The aim of this systematic review is to quantify the effectiveness
of pre-and post-licence rider training for motorcycle riders in re-
ducing traffic offences road traffic crash involvement injury and
death
O B J E C T I V E S
To quantify the effectiveness of pre- and post-licence motorcycle
rider training on the reduction of traffic offences traffic crash
involvement injuries and deaths of motorcycle riders
M E T H O D S
Criteria for considering studies for this review
Types of studies
All relevant intervention studies such as randomised and non-ran-
domised controlled trials and interrupted time series studies and
observational studies including cohort and case-control studies
Types of participants
Motorcyclists and riders of mopeds or motor scooters of all ages
Types of interventions
Motorcycle ridertraining (pre-or post-licence mandatory or non-
mandatory) versus no training
One form of pre-licence rider training versus another form (eg
on-road off-road theory or a combination)
One form of post-licence rider training versus another form (eg
on-road off-road theory or a combination)
Types of outcome measures
Offences crashes injuries and death
Search methods for identification of studies
We did not restrict the searches by date language or publication
status
Electronic searches
We searched the following electronic databases
bull Cochrane Injuries Group Specialised Register (searched 22
Sept 2008)
bull CENTRAL (The Cochrane Library 2008 Issue 3)bull TRANSPORT 1988 to 200706 (latest date available)
bull ATRI Australian Transport Index (to Sept 2008)
bull MEDLINE (Ovid SP) 1950 to Sept 2008
bull EMBASE (Ovid SP) 1980 to Sept 2008
bull CINAHL (EBSCO) 1982 to Sept 2008
bull WHOLIS (World Health Organization Library
Information System) (to Sept 2008)
bull PsycINFO (Ovid SP) 1806 to Sept 2008
bull LILACS (BIREME) (Latin American and Caribbean
Health Sciences) (1982 to Sept 2008)
bull ISI Web of Science Social Sciences Citation Index (SSCI)
1970 to Sept 2008
3Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 650
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 750
Table 1 Risk of selection bias (Continued)
Osga 1980 Differences at baseline reported on sex and age
Confounding factors were not dealt with in univariate analysis Riding exposure was accounted for in the
cumulative survival analysis
Selection intervention group were all training participants on the register and the control group were
respondents to a diverse range of recruitment strategies
Satten 1980 Differences at baseline reported on age sex helmet use riding experience and exposure and previous
training
Confounding factors were not dealt with in design or analysis
Selection intervention group was selected from a list of training graduates and the control group was
partly selected from a state register of motorcyclists
Jonah 1982 Differences at baseline reported on age sex riding experience time licensed previous training and alcohol
use
Confounding factors were dealt with through multivariate analysis
Selection intervention group was selected from a list of training graduates and the control group was
randomly selected from the Ontario motorcycle operator file
Mortimer 1984 Differences reported at baseline on age licence status riding experience years licensed and helmet use
Confounding factors age years licensed and rider experience were dealt with through stratification and
riding exposure through standardisation
Selection intervention group was selected from a list of training enrolees and the control group was
selected from motorcycle dealerships during three successive Saturdays
Cooper 1988 Differences reported at baseline on age gender licensure and highway experience
Confounding factors age gender riding exposure rider characteristics rider attitudes and behaviour were
accounted for in a multivariate analysis
Selection there was no information on how the intervention and control groups were selected and from
where
Mortimer 1988 Differences reported at baseline on years licensed riding experience and helmet use
Confounding factors age years licensed and rider experience were dealt with through stratification and
standardisation of riding exposure A regression analysis was also conducted but poorly reported and not
usable
Selection intervention group was selected from a list of training enrolees and the control group was
selected from motorcycle dealerships
Adams 1985 Differences at baseline reported on previous crash rates
Confounding factors not controlled for in analysis
Selection the intervention group was a non-random sample of those offered training within the Australia
Post motorcycle delivery riders the control group were the remaining Post riders
Leung 1987 Differences at baseline reported on age and sex
Confounding factors age and sex controlled for in analysis by stratification
Selection those exposed to training were MRC graduates those unexposed were taken from the licensing
records
5Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 850
Table 1 Risk of selection bias (Continued)
McDavid 1989 Differences at baseline other than matched characteristics were not reported
Confounding factors age ownership of car licence years licensed previous crashesand offences accounted
for through matching Other confounding effects not controlled
Selection intervention group was selected from a list of training graduates and the control group was
selected from Bureau of Motor Vehicle records
Kloeden 1994 Differences at baseline reported were in geographical area of residence and in riding exposure
Confounding factors were accounted for through exclusion of participants from rural areas
Other confounding factors were not reported
Selection the Ridersafe course was phased in based on postcode of riderrsquos home address analyses were
based on a database of those who obtained motorcycle rider learnerrsquos permits of which about 2000 were
Ridersafe riders and about 5000 were control riders
Billheimer 1998 Differences at baseline reported on age and previous riding experience
Other potential confounding factors assessed were riding exposure primary purpose of riding and helmet
use
Confounding factors accounted for by matching (age sex primary purpose of riding) standardisation
(riding exposure) and stratification (previous riding experience)
Selection the intervention group were those who had taken the rider training programme the control
group were California motorcycle riders matched on the above variables but who had not taken a riding
course
Perrino 2002 Differences at baseline reported were a higher number of females in the trained group and a higher
proportion of experienced riders in the untrained group
Confounding factors were not accounted for in the analysis
Selection the intervention group were those who had done the Motorcycle Riding Course the control
group had not done the course Both groups were identified through a register of licensed motorcyclists
Kraus 1975 Population based case-control study
Cases all motorcycle riders with a confirmed medically treated motorcycle injury (including fatalities)
from all hospitals in study base and peripheral area hospitals and all deaths due to motorcycle crashes in
the study base All serious injuries (leading to hospitalisation) and fatalities likely to be included
Community controls a random sample of registered motorcycle owners in the study base
Potential confounders assessed were age sex weight height riding experience motorcycle use charac-
teristics of motorcycle and previous crashes and offences
Confounding factors were accounted for through exclusion of females and stratification by age and
motorcycle useSelection cases and controls drawn from the same population
Hurt 1981 Population based case-control study
Cases and controls not drawn from the same population
Cases motorcycle riders who had a motorcycle crash Not all reported cases included
Community controls motorcycle riders who passed the crash site at the same time of day and day of week
that the crash occurred but 2 years later
Potential confounders assessed were age sex time type of motorcycle roadway alignment helmet use
weather conditions road surface conditions
Confounding factors were accounted for through matching for time of day and day of week that crash
occurred and stratification
Selection cases were recruited from police reports not all cases were included controls were matched
6Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 950
Table 1 Risk of selection bias (Continued)
community controls from randomly selected sites matched for same time of dayday of week as the crash
occurred
Haworth 1997 Population based case-control study
Cases and controls drawn from the same population
Cases motorcycle riders or pillion riders with an injury from a motorcycle crash from hospital and police
records Not clear whether all cases were included
Community controls motorcycle rider or pillion who passed the crash site at the same time of day and
day of week that the crash occurred
Potential confounders assessed were rider factors age marital status licence status riding experience
and exposure alcohol and drugs use of protective gear including helmet use pillion factors motorcycle
factors trip factorsConfounding factors were accounted for through matching for time of day and day of week that crash
occurred and stratification for age blood alcohol level and licence status
Selectioncases were riders or pillions involved in on-road crash recruited from hospitals and police records
not sure if all cases were included controls were matched community controls who passed the crash site
at the same time of dayday of week as the crash
Savolainen 2007 Differences at baseline reported on sex safety helmet use riding ability riding experience and licence
status
Adjusted for these confounders through statistical modelling although exact process unclear
Selection intervention group from list of trainees control group from Indiana Bureau of Motor Vehicles
database Trainees and the control group were invited to participate in surveys
Davis 1997 Differences at baseline not reported but likely No adjustment for any confounders at baseline or during analysis
Selection intervention group from list of riders involved in crashes who were also on a list of training
graduates control group from the same list of riders involved in crashes who were not also on a list of
training graduates
Waller 1992 Differences at baseline not reported but likely
Adjusted for confounders through exclusion of riders younger than 18 years riders without a motorcycle
operator driver license riders with location of residence outside of the Texas metropolitan areas of Austin
through matching by age sex and location of residence through stratification by age and sex and through
standardisation (number of crashes by person-days at risk)
Selection intervention group from the Motorcycle Operator Training class roster (number of trained
riders that the sample was selected from is unclear) control group through a random number generator
generating driver license numbers until an equal sample size was selected as the intervention group
MAIDS 2004 Population based case-control study
Unclear whether cases and controls not drawn from the same population
Cases motorcycle riders who had a motorcycle crash Unclear whether all cases included
Community controls motorcycle riders recruited at petrol stations Unclear whether and how matched
to cases
No potential confounders were reported
There was no report of whether and how potentially confounding factors were accounted for in the study
7Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 1050
Table 2 Risk of allocation bias
Anderson 1980 Allocation concealment unclear
Hall 1988 Allocation concealment unclear
Swaddiwudhipong 1998 Allocation concealment unclear
Table 3 Risk of detection bias
Anderson 1980 Blinding of outcome assessors not reported
Driver records holding police recorded crashes and offences not validated
Hall 1988 Blinding of outcome assessors not reported
Driver records holding police recorded crashes and offences not validated
Swaddiwudhipong 1998 Blinding of outcome assessors not reported
Police and hospital records Post mortem examination for deaths not validated
Adams 1985 Blinding of outcome assessors not reported
Australia Post crash records not validated but argued that underreporting is unlikely
Leung 1987 Blinding of outcome assessors not reported
Driver records holding police recorded crashes and offences
Raymond 1979 Blinding of outcome assessors not reported
Self-report through personal interviews or questionnaires - both methods validated
Osga 1980 Blinding of outcome assessors not reported
Self-report through structured questionnaires validated by cross-checking with police
recorded crash records
Satten 1980 Blinding of outcome assessors not reported
Self-report through structured interviews or questionnaires - both methods were not
validated
Intervention group were given telephone interviews but the control group mailed ques-
tionnaires
Jonah 1982 Blinding of outcome assessors not reported
Self-report through structured interviews not validated Driver records holding police
recorded crashes and offences
Mortimer 1984 Blinding of outcome assessors not reported
Self-report through questionnaires not validated
Intervention group received mailed questionnaires but the control group received ques-
tionnaires personally at the motorcycle dealership
Cooper 1988 Blinding of outcome assessors not reported
Driver records holding police recorded crashes and offences
8Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 1150
Table 3 Risk of detection bias (Continued)
Mortimer 1988 Blinding of outcome assessors not reported
Self-report through questionnaires not validated
Intervention group received mailed questionnaires but the control group received ques-
tionnaires personally at the motorcycle dealership
McDavid 1989 Blinding of outcome assessors not reported
Driver records holding police recorded crashes and offences
Kloeden 1994 Blinding of outcome assessors not reported
Driver records holding police recorded crashes and offences
Billheimer 1998 Blinding of outcome assessors not reported
Police recorded crashes and offences
Perrino 2002 Blinding of outcome assessors not reported
Self-report through mailed out questionnaires police records not validated
Kraus 1975 Blinding of intervention assessors to case or control status not reported
Intervention measured police records self-report though questionnaires not validated
Length of follow up not specified
Hurt 1981 Blinding of intervention assessors to case or control status not reported
Intervention measured face-to-face in-depth interviews (ie no structured question-
naires) Validity and reliability of data ensured by reported high level of quality controlLength of follow up not specified
Haworth 1997 Blinding of intervention assessors to case or control status not reported
Intervention measured self-report either by personal or by telephone interview for
both cases and controls not validated
Length of follow up in the 2 years prior to the interview or more than 2 years ago
Savolainen 2007 Blinding of outcome assessors not reported
Source of outcome data not reported
Length of follow up up to 5 years
Davis 1997 Blinding of outcome assessors not reported
Source of outcome data police records
Length of follow up 1 year
Waller 1992 Blinding of outcome assessors not reported
Source of outcome data Department of Public Safety driver history file and official
accident reports
Length of follow up 3 years
MAIDS 2004 Blinding of intervention assessors to case or control status not reported
Intervention measured face-to-face in-depth interviews (ie no structured question-
naires) No report on whether and how validity and reliability of data was ensured
Length of follow up 2 years
9Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 1250
Table 4 Risk of attrition bias
Anderson 1980 Percentage of participants in analysis 99 forthe total group Unclear forintervention andcontrol groups
separately
Hall 1988 Percentage of participants in analysis 96-97 forall intervention groups and 98for the control group
Swaddiwudhipong 1998 Percentage of participating sub-districts in analysis 100 for both intervention and control sub-districts
Raymond 1979 Percentage of participants in analysis 74 for the intervention group and less than 57 for the control
group
Osga 1980 Percentage of participants in analysis 73 for the intervention group and 100 for the control group
Satten 1980 Percentage of participants in analysis is unknown Initial sample size is unclear
Jonah 1982 Percentage of participants in analysis 23 for the intervention group and 45 for the control group
Mortimer 1984 Percentage of participants in analysis 35 for the intervention group and 84 for the control group
Cooper 1988 Percentage of participants in analysis 98 for the total group Unclear for intervention and control
groups separately
Mortimer 1988 Percentage of participants in analysis 40 for the intervention group and 90 in the control group
Adams 1985 Percentage of participants in analysis 61 forthe total group Unclear forintervention andcontrol groups
separately
Leung 1987 Percentage of participants in analysis unclear because initial sample size was not reported
McDavid 1989 Percentage of participants in analysis 40 of final matched cohort
Kloeden 1994 Percentage of participants in analysis 39 for the intervention group and 6 for the control group
Billheimer 1998 Percentage of participants in analysis unclear percentage of the initial sample size
Perrino 2002 Percentage of participants in analysis unclear percentage of the initial sample size
Kraus 1975 Percentage of participants in analysis 58 of cases 59 of controls
Hurt 1981 Percentage of participants in analysis 18 of cases lt 29 of controls
Haworth 1997 Percentage of participants in analysis 70 of cases 47 of controls
Savolainen 2007 Percentage of participants in analysis 18 for the intervention group and 15 for the control group
Davis 1997 Percentage of participants in analysis 100 for the intervention group and unclear for the control group
10Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 1350
Table 4 Risk of attrition bias (Continued)
Waller 1992 Percentage of participants in analysis unclear percentage for the intervention group and 100 for the
control group
MAIDS 2004 Percentrage of participants in analysis unclear for both cases and controls
R E S U L T S
Description of studies
See Characteristicsof included studies Characteristics of excluded
studies Characteristics of ongoing studies
Our search resulted in 102 studies that seemed relevant of which
we excluded 79 studies because they did not meet the criteria for
considering studies for this review (see Characteristics of excluded
studies)
Basedon ourcriteriawe includeda total of 23 studies in this review
There were three randomised trials (Anderson 1980 Hall 1988
Swaddiwudhipong 1998) two non-randomised trials (Adams
1985 Leung 1987) 14 cohort studies (Raymond 1979 Osga
1980 Satten 1980 Jonah 1982 Mortimer 1984 Cooper 1988
Mortimer 1988 McDavid 1989 Waller 1992 Kloeden 1994
Davis 1997 Billheimer 1998 Perrino 2002 Savolainen 2007)
and four case-control studies (Kraus 1975 Hurt 1981 Haworth
1997 MAIDS 2004)
Over half of the studies were conducted 20 or more years
ago and the oldest was conducted 31 years ago Only three
studies were from this decade (Perrino 2002 MAIDS 2004
Savolainen 2007) Over 60 of the studies were conducted in
the USA (Kraus 1975 Anderson 1980 Osga 1980 Satten 1980
Hurt 1981 Mortimer 1984 Leung 1987 Hall 1988 Mortimer
1988 Waller 1992 Davis 1997 Billheimer 1998 Perrino 2002
Savolainen2007) three inAustralia (Adams 1985 Haworth 1997
Kloeden 1994) three in Canada (Jonah 1982 Cooper 1988
McDavid 1989) onefromthe United Kingdom (Raymond 1979)one from Europe (with sampling areas located in France Ger-
many Netherlands Spain and Italy) (MAIDS 2004) and one
in Thailand (Swaddiwudhipong 1998) Only seven studies were
published in peer reviewed journals (Kraus 1975 Jonah 1982
Mortimer 1984 Cooper 1988 Mortimer 1988 McDavid 1989
Swaddiwudhipong 1998) The other studies were reports forhigh-
way safety and road traffic organisations
The studies by Anderson 1980 Hall 1988 Kloeden 1994
Billheimer 1998 (the basic California Motorcyclist Safety Program
component of the study) and Perrino 2002 examined manda-
tory pre-licence training Cooper 1988 McDavid 1989 and
Swaddiwudhipong 1998 assessed non-mandatory training that
when passed resulted in a licence Raymond 1979 Osga 1980
Satten 1980 Jonah 1982 Mortimer 1984 Adams 1985 Leung
1987 Mortimer 1988 Waller 1992 Davis 1997 Billheimer 1998
(the Experienced Rider Course component of the study) andSavolainen 2007 assessed non-mandatory training that could be
taken by motorcycle riders with and without a licence The case-
control studies (Kraus 1975 Hurt 1981 Haworth 1997) assessed
lsquoanyrsquo type of rider training and therefore could not be grouped by
pre- or post-licence rider training The MAIDS 2004 case-control
study did make a distinction between participants having received
compulsory pre-licence training or additional (non-compulsory)
training We did not find a study that compared one form of pre-
or post-licence training (eg on-road off-road theory or a com-
bination) versus another form of training All studies compared
eithermandatory or non-mandatory formal training versus no for-
mal training with the exception of Haworth 1997 who compared
different levels of training courses (beginner intermediate and ad-vanced)
We have presented a more detailed description of the individual
studies in Characteristics of included studies We have provided
a description below of the types of motorcycle rider training that
were evaluated in the included studies
Types of motorcycle rider training evaluated in the
studies
The types of training were varied Duration of training ranged
from two or three hours to 24 hours administered over two or
three days Most training consisted of a theoretical and a practical
component usually off-road The content ranged from provid-ing riders with manuals lectures discussions showing films and
administering knowledge and skill tests Little information was
provided about the qualifications of trainers who conducted the
training
Six studies (Osga 1980 Satten 1980 Mortimer 1984 Mortimer
1988 Waller 1992 Savolainen 2007) assessed the effectiveness of
the USA-based Motorcycle Safety Foundationrsquos Motorcycle Rider
Course (MSF MRC) This course was released in 1976 It com-
prises eight hours of theory (lectures discussions films and multi-
plechoice tests) a textbook and 12 hours of off-road practice The
Motorcycle Safety Foundation is a national not-for-profit organ-
11Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 1450
isation sponsored by manufacturers and distributors of motorcy-
cles Many USA states have adopted the MSF Basic Rider Courseand Experienced Rider Course (wwwmsf-usaorg) since its release
in 1976
The Motorcycle Operators Skill Test was also developed
by the USA Motorcycle Safety Foundation in coopera-
tion with others (httpwwwwebbikeworldcommotorcycle-
trainingmotorcycle-test) The MOST is a licensing test consist-
ing of a manual knowledge and skill test (comprising eight exer-
cises of increasing difficulty and skill requirements) and a mini-
skill test conducted with a simulator Some studies compared this
alternative licensing test to the standard licensing test However
as the MOST is not a type of rider training we have included only
the two studies that assessed the MOST in combination with one
or more training components in this review Both Anderson 1980and Hall 1988 assessed the MOST in combination with a train-
ing component of three hoursrsquo duration consisting of instructions
about the skills needed to pass the MOST (30 minutes of the-
ory and two and a half hours on the motorcycle) Hall 1988 also
assessed the MOST in combination with the Motorcycle Safety
Foundationrsquos Motorcycle Rider Course as described above
Other USA studies evaluated the Wisconsin motorcycle rider
course a 22-hour off-street programme that teachesthe basic rules
of safe motorcycling (Leung 1987) the USA Maryland Motorcy-
cle Rider Course (MRC) a mandatory (for those younger than 18
years)pre-licence 20-hour coursecomprising eight hours of theory
and12 hours of practice and created by the Maryland Motor Vehi-
cle Administration in 1985 (Perrino 2002) the California Motor-cyclist Safety Program (CMSP) a mandatory (for those younger
than 18 years) pre-licence 16-hour basic course (Billheimer 1998)
and Experienced Rider Course (ERC) of eight hoursrsquo duration
(Billheimer 1998) The CMSP still offers these courses ( http
wwwca-msporg) but they are now MSF courses as described
above Davis 1997 did not give any information about the assessed
rider training
The Canadian studies evaluated the British Colombia Safety
Councilrsquos 37-hour motorcycle safety programme (no further in-
formation on the content was provided) (McDavid 1989) and
the Canada Safety Council Motorcycle Training Program (MTP)
comprising four hours of theory and 16 hours of practice (six
hours on-road) over twoweekends (Jonah 1982) The MTPcoursewas made available in 1974 and is still being offered in Canada
(httpwwwsafety-councilorgtrainingmtpmtphtm) Cooper
1988 did not give any information about the assessed rider train-
ing However the study was conducted in British Colombia in
Canada in 1988 the same time as the McDavid 1989 study was
conducted
In the United Kingdom Raymond 1979 assessed the non-manda-
tory RACACU (Royal Automobile ClubAuto-Cycle Union)
training scheme The training programme consisted of either 24
one-hour sessions or 13 two-hour sessions Twelve hours of the
training focused on theory about basic operational procedures
traffic laws and highway codes The practical session focused on
basic operational skills riding on a training area and riding in traf-fic with an instructor A proficiency test was administered at the
end of the course
One study from rural Thailand (Swaddiwudhipong 1998) evalu-
ated a community-based rider education programme consisting of
theory on motorcycle injury risk helmet use traffic laws vehicle
regulations and traffic signs and a written and skill test leading to
a licence
TheAustralian studies evaluateda rider trainingcourse of 12 hours
over two consecutive days (four hours of theory and eight hours of
practice)administeredto AustraliaPosttelegram deliverers(Adams
1985) and a mandatory pre-licence Ridersafe course consisting
of two tiers the first tier being a training course on the basics
of motorcycle riding where a pass entitled the rider to obtain alearnerrsquos permit and the second tier being a skill test where a pass
entitled the riderto obtain a probationary motorcycle licence The
training course consisted of two four-hour training sessions on
separate days involving instruction videos and actual riding on
motorcycles The skill test consisted of a four-hour preparation
course of 74 minutes of lecturediscussion 38 minutes of video
and 75 minutes of on-course riding which was the testing phase
(Kloeden 1994) The Rider Safe course is still being provided by
Transport South Australia a state government organisation (http
wwwtransportsagovau )
The duration or content of rider training in the four case-control
studies (Kraus 1975 Hurt 1981 Haworth 1997 MAIDS 2004)
was not described
Risk of bias in included studies
We describe the methodological quality of the included studies
by study type below More detailed information can be found in
Table 1 Table 2 Table 3 and Table 4
When considering the findings from these studies it is important
to note the quality (or lack thereof) of reporting of the studies
Quality of reporting is the extent to which a report provides in-
formation about the design conduct and analysis of a study ( Juumlni2001) Huwiler-Muumlntener 2002 found that even though trials
may be well reported it does not mean that the methodological
quality is of equal standard and well-conducted trials may never-
theless be reported badly The reporting quality of most studies in
this review was poor Most studies were old and did not meet con-
temporary reporting standards This made data extraction com-
plicated It also made it difficult to contact authors for additional
information
Because of significant heterogeneity between studies as well as the
poor quality of most studiespooling of studyresults to calculatean
overall measure of effect was not appropriate Instead we provided
a description of our judgement of risk of bias in the text
12Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 1550
Risk of bias in the randomised trials
There were three randomised trials (Anderson 1980 Hall 1988Swaddiwudhipong 1998) Anderson 1980 and Hall 1988 had an
initial sample size of 41188 and 26464 motorcycle licence ap-
plicants respectively Both studies assessed the effectiveness of the
pre-licence MOST with a trainingcomponent Swaddiwudhipong
1998 evaluated the effectiveness of a community-based rider ed-
ucation programme in six districts
Allocation bias
Anderson 1980 reported using random birth date tables to allocate
participants to either the intervention or control group Hall 1988
reported the use of a computerised algorithm set in a central of-
fice but also used manual random assignment when the computer
went down Swaddiwudhipong 1998 reported lsquosimple random
samplingrsquo for the delivery of the intervention where the sampling
frame included nine districts Three districts were excluded since
they were too hard to access The remaining six districts were ran-
domly allocated to the intervention (three) and the control (three)
groups Systematic sampling wasthen used ordered by village size
for the villages within districts Allocation concealment was not
reported in any of the studies The descriptions of randomisation
in these studies suggest that there may have been limitations to
the process resulting in a lack of equivalence of the intervention
and control groups at baseline and therefore the possibility of
allocation bias Swaddiwudhipong 1998 reported that before theintervention the annual incidence rate of motorcycle-related in-
juries was slightly (but not statistically significant) higher in the
intervention areas than in the control areas Participants in the
intervention districts were also more likely to have a valid licence
and to wear a helmet more often than participants in the control
districts Anderson 1980 measured differences at baseline on the
participant and prior driving record variables and found these to
be negligible Hall 1988 did not report on baseline differences
between the intervention and control groups As the Anderson
1980 and Hall 1988 studies had sample sizes of over 40000 and
26000 respectively and used some form of randomisation it is
likely that any confounding factors at baseline would have been
equally distributed between comparison and intervention groups
Selection bias
After randomisation Anderson 1980 and Hall 1988 excluded
08 and 3 participants respectively because of assignment to
the wrong group administration of the wrong intervention or
other reasons The resulting lack of analysis by rsquointention-to-treatrsquo
can introduce biasby removing the effectivenessof randomisation
Nonetheless the percentage excluded was small and the number
of overall participants large so it is unlikely to have had a major
effect on the final results In the Swaddiwudhipong 1998 study
villages were systematically sampled into intervention or control
groups The intervention was implemented and outcomes of in-
jury and death were measured at a community level Interventionand control districts were randomly selected however there were
only three intervention and three control communities
Detection bias
None of the randomised trials reported blinding of outcome as-
sessors Data collection tools used by Anderson 1980 and Hall
1988 were driver records holding police recorded crashes and of-
fences Swaddiwudhipong 1998 used police and hospital records
at a district level and post-mortem examination for deaths None
reported on validity or reliability for any of the instruments
Attrition bias
Anderson 1980 and Hall 1988 had driver records available for
96 to 100 of participants They also collected self-reported
data but the response rate to the questionnaires was very low (low-
est response rate 42 and 30 respectively) Attrition bias in
Swaddiwudhipong 1998 was not an issue as outcome data were
available for all sub-districts included in the trial
The intervention reported in Anderson 1980 and Hall 1988 tri-
als had an unanticipated effect on the participants In both trials
an average of only 28 of licence applicants in the intervention
group eventually obtained their licence compared to 35 in the
control group This indicates that the licensing process in generalseems to have deterred applicants from obtaining a licence and
although no statistical tests of the difference were conducted this
deterrence appears even greater for those who had rider training
This is not true attrition bias as the studies did not experience
differential loss to follow up from outcome measurement Instead
differential attainment of the intermediary step of licensure pre-
cluded participants from being at risk for the outcomes of interest
they are not driving a motorcycle hence they are not at risk for
motorcycle-related crashes injury death or offences In short it is
not possible to assess whether a reduction in crashes in the inter-
vention group would be due to changes in riding habits as a result
of training or simply due to delayed or never attained licensure
Intervention integrity
Neither Anderson 1980 nor Hall 1988 reported on the consis-
tency in the way the intervention was administered the possible
lsquocontaminationrsquo of participants in the intervention group by other
types of formal or informal rider training or of participants in the
control group by the intervention Swaddiwudhipong 1998 did
not report on the number of villagers in each group that actually
attended the intervention However they did report on a number
of other potential training courses that were being implemented
at the same time
13Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 1650
Analysis
Anderson 1980 and Hall 1988 acknowledging the lsquodeterrent ef-fectrsquo described above conducted two separate analyses The first
analysis included the total study population of licence applicants
over two years from the time of their application (the population
denominator) The second analysis was performed on the licensed
(intervention and control) subgroups only two years from the
time of obtaining their licence (the licensed rider denominator)
As only a third were licensed and exposed to riding the motorcy-
cle it was impossible to determine whether reported differences
in crash rates across groups were due to the intervention or to
other lsquoconfoundingrsquo characteristics of those who were determined
in their efforts to obtain the licence When analysis was restricted
only to those who gained their licence the same difficulties re-
mained The benefits of the randomisation process were thus lost
due to this lack of licensure of riders during the process As there
were only three intervention and three control groups in the study
reported by Swaddiwudhipong 1998 analyses were hindered by
smallnumbers for outcome comparisons betweenthe intervention
and the control districts
Risk of bias in the cohort studies and non-randomised
trials
We discuss the methodological quality of cohort studies (seven
prospective studies (Raymond 1979 McDavid 1989 Waller 1992
Kloeden 1994 Billheimer 1998 Perrino 2002 Savolainen 2007)and seven retrospective studies (Osga 1980 Satten 1980 Jonah
1982 Mortimer 1984 Cooper 1988 Mortimer 1988 Davis
1997)) and non-randomised trials (Adams 1985 Leung 1987) in
the one section as the biases for the two study types are similar
The quality of these studies was in general poor
Selection bias
The confounding factors expected to have been addressed in the
studies were sex age riding experience prior driving records (of-
fences crashes) previous training and the following proxies for
risk taking behaviour helmet use speed and alcoholdrug use All
of these factors have been shown to be associated with an increasedrisk of motorcycle crash (Peden 2004 Zambon 2006) The pos-
sible source of bias in cohort studies and non-randomised trials
is that groups may differ on important characteristics at baseline
participants lsquoself-selectrsquo into either the rider training group by de-
ciding to undertake a rider training course or into the lsquocontrol
grouprsquo by deciding not to undertake a course Participants who
undertake a non-mandatory rider training course may differ from
those who choose not to on characteristics such as age sex rid-
ing experience and riding behaviour as well as other possible con-
founders For example riders who take a training course may be
more safety-conscious or on the contrary may do so because they
have previously experienced a crash or because they perceive that
they have poor riding skills Therefore ideally all participants in
the intervention group need to be included in the study as wellas a random sample of controls taken from the same population
as the intervention group If that is not possible intervention and
control groups either need to be carefully matched to make them
comparable or differences at baseline need to be adjusted for in
the analysis
Most of the studies using a non-randomised trial or cohort design
compared an intervention group to a control group that differed
on a number of important factors The studies in general failed to
adjust for important differences found at baseline and their results
cannot be meaningfully interpreted Ten studies (Raymond 1979
Osga 1980 Satten 1980 Jonah 1982 Mortimer 1984 Mortimer
1988 McDavid 1989 Waller 1992 Davis 1997 Savolainen 2007)
assessed intervention and control groups that were not drawnfrom the same population In most of these cases the intervention
group was selected from a list of training graduates who were usu-
ally young female and less experienced riders whereas controls
were recruited from a general population of motorcycle riders who
were usually older male and experienced riders For three stud-
ies it was unclear how they were selected (Leung 1987 Cooper
1988 Perrino 2002) Only Kloeden 1994 recruited groups from
the same population of motorcycle applicants and Adams 1985
from a group of professional riders Nevertheless due to non-ran-
dom selection of intervention and control groups differences at
baseline are likely even though study groups were from the same
population Three studies adjusted for some confounding fac-
tors found at baseline through matching (McDavid 1989 Waller1992 Billheimer 1998) However matching has limitations in
termsof controlling for multiple confounders especially those that
are unknown to influence the effect of rider training on the out-
comes One study matched the control and intervention groups
by excluding controls (approximately 3500 (66 of initial con-
trol group) that differed from the intervention group) (Kloeden
1994) Five studies (Osga 1980 Satten 1980 Adams 1985 Davis
1997 Perrino 2002) did not adjust for differences between study
groups at baseline either in the study design or analysis Results
from these studies are therefore not reliable The other six stud-
ies (Raymond 1979 Jonah 1982 Mortimer 1984 Cooper 1988
Mortimer 1988 Leung 1987 McDavid 1989 Billheimer 1998)
adjusted for some differences between groups at baseline by eitherbi-variate or multi-variate analyses or stratification and standardi-
sation of riding exposure (outcomemiles driven) or stratification
by age group The limitation of stratification is that it is difficult to
control for many variables simultaneously because a large number
of strata will need to be generated resulting in small numbers that
cannot be meaningfully analysed Savolainen 2007 developed sta-
tistical models to assess the probability of a crash after taking the
rider training course in an intervention and control group while
examining differences by sub-groups (eg age) Participation rates
were also very low in this study with only 5584000 (1395)
in the eligible intervention group and 5884000 (147) for the
14Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 1750
control group completing surveys
Detection bias
None of the studies reported blinding of outcome assessors to the
intervention or control status of the participants Seven studies
used driver records containing police recorded data as their main
measure of outcome (Leung 1987 Cooper 1988 McDavid 1989
Waller 1992 Kloeden 1994 Davis 1997 Billheimer 1998) An-
other four studies (Raymond 1979 Satten 1980 Mortimer 1984
Mortimer 1988) only used self-report Raymond 1979 was the
only study using self-report as the main outcome measurement in
which study instruments were validated Self-report might prevent
the underreporting that has been described for the use of police-
driver records (see rsquoRisk of biasrsquo in the randomised trials above)but has other risks of bias especially if questionnaires or inter-
view methods have not been validated and tested for reliability
Recall bias arises from mistakes in recollecting events both be-
cause of failures of memory and because of looking at things lsquowith
hindsightrsquo and possibly changed views (httpwwwcochraneorg
resourcesglossaryhtm) As many studies were retrospective risk
of recall bias is likely and may have increased with length of follow
up For studies that used self-reported outcomes only it is also
possible that social desirability response bias played a role where
participants in the training group may have felt it was lsquoexpectedrsquo
to report good driving histories Multiple methods to measure the
same outcome for both groups decreases the risk of detection bias
for example because self-reported answers can be cross checkedwith other sources of information such as driver records
In this regard three studies (Osga 1980 Jonah 1982 Perrino
2002) used both self-report and driver records Osga 1980 found
that for both intervention and control groups more crashes were
reported through self-report than on state crash records an aver-
age of 54 of self-reported crashes could be found back on state
crash records One industry-based study (Adams 1985) used oc-
cupational health records which were not validated but which are
unlikely to suffer from underreporting due to work cover claims
Savolainen 2007 used information on rider crashes for a five-year
period but did not specify from where this information was ob-
tained
Attrition bias
The majority of studies (Raymond 1979 Satten 1980 Adams
1985 Leung 1987 Kloeden 1994 Davis 1997 Billheimer 1998
Perrino 2002) inadequately reported on losses of participants to
the study We can therefore not be certain that there were no sys-
tematic differences in the characteristics between the intervention
and control group participants Savolainen 2007 reported the re-
sponse rates to their baseline survey but did not report on the per-
centage of outcome data that was available for their study sample
A similar deterrent effect of rider training as described for the ran-
domised trials (Anderson 1980 Hall 1988) occurred in two stud-
ies Jonah 1982 reported that 65(14992310) of the trained rid-
ers stopped riding compared to 47 (9492029) of the untrainedriders They excluded non-active and non-licensed riders before
the study commenced so the differential choice to continue riding
after training did not affect their analyses Leung 1987 reported
that the percentage of participants in the intervention group ob-
taining their licence was 50 (no number provided by authors)
lower than in the control group This may have biased the mea-
surement of outcomes because unlicensed participants will not be
exposed to riding the motorcycle and therefore not be at risk of
being involved in a crash or violation As explained above this
makes it difficult to assess whether the reduction in crashes in the
intervention group is due to a change in driving habits or simply
due to delayed licensure
Mortimer 1984 and Mortimer 1988 describe that about 44 and30 of course graduates respectively were excluded from analysis
because they had not ridden a motorcycle in the previous year
However because these studies were retrospective and the controls
were activeriders at the time of recruitment a decision on whether
the deterrent effect was differential between the two groups could
not be made Though not reported quantitatively Kloeden 1994
also describes the effectof rider training deterringindividuals from
becoming licensed and hence exposure to riding and risk of crash
death or offences involving a motorcycle Osga 1980 reported
that 27 of the trained riders did not ride after the course and
excluded these from the analysis
Use of different data collection strategies may have contributed to
differential loss to follow up in the cohort studies Satten 1980for example interviewed the intervention group over the tele-
phone but mailed questionnaires to the control group Mortimer
1984 and Mortimer 1988 distributed questionnaires in person to
the control group but mailed them to the intervention group
This may have played a role in the observed differential response
rates where more subjects responded in the personally approached
groups
Survival bias
For those studies that assessed offences crashes and injuries but
not death survival bias may have also played a role Motorcyclists
who were fatally or seriously injured in crashes during the follow-
up period after the intervention would have been less likely or
unable to respond to surveys This could bias the effect measures
towards or away from the null None of the studies wrote about
this issue or examined the characteristics of those lost to follow up
in order to address the possibility of survival bias
Intervention integrity
Although the interventions were intended for the entire lsquoexposedrsquo
groups most studies reported difficulty in tracking participants
whoreceivedthe interventionand those whodid not Adams 1985
15Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 1850
estimatedthat 62of thetotalstudysample received the interven-
tion The remaining studies did not report the percentage of thestudy population that received the intervention as was intended
or reported that this percentage was unknown (Raymond 1979
Kloeden 1994 Satten 1980 Jonah 1982 Mortimer 1984 Leung
1987 Cooper 1988 Mortimer 1988 Billheimer 1998 Perrino
2002 Savolainen 2007) Only Raymond 1979 reported the degree
to which their intervention was delivered as intended No studies
reported on possible lsquocontaminationrsquo of participants by other inter-
ventions than the one to which they were allocated Because Osga
1980 Satten 1980 Jonah 1982 Mortimer 1984 Cooper 1988
Mortimer 1988 Waller 1992 and Davis 1997 were retrospective
studies and participants were recruited from a list of graduates it
was known that they had completed and passed the course
Analysis
In some studies some participants who reported at baseline that
they were untrained decided to enroll in a rider training pro-
gramme after the baseline survey The non-randomised trial by
Adams 1985 experienced this move of subjects from the untrained
to the trained group but failed to conduct analyses by lsquointention-
to-treatrsquo This can bias the results in randomised and non-ran-
domised trials if participants who changed groups differ signifi-
cantly with regard to potentially confounding characteristics On
the other hand Perrino 2002 also experienced this move of sub-
jects from theuntrained to the trained groupmore than 100riders(5 of the total sample) originally stated that they were untrained
at baseline but then received training between the baseline and
follow-up surveys Yet since the Perrino 2002 study was a cohort
their analyses then proceeded by including those riders who were
lsquonewlyrsquo trained as exposed Riders who were lsquonewlyrsquo trained since
baseline but with a programme other than the one under study
were excluded from the analysis Nevertheless confounding fac-
tors were not accounted for in the analyses
McDavid 1989 matched exposed with unexposed subjects but
could not find suitable matches for the majority (60) of their
subjects They could therefore only conduct analysis on the result-
ing40 (139346) which is an inappropriate andinefficient anal-
ysis Some studies that reported on the issue of non-active licensedriders(Satten 1980 Jonah 1982 Mortimer 1984 Mortimer 1988
Kloeden 1994) excluded non-active riders from analysis intro-
ducing selection bias as it is likely that riders who continued to
ride differed with regards to potentially confounding character-
istics across intervention and control groups Kloeden 1994 for
example conducted several analyses to assess the effect of various
sub-sets of the control group Original analyses included all con-
trols and led to erroneous results since only a small proportion of
controls hadgone on to get a licence after their learnerrsquos permitand
were actually exposed to the possibility of experiencing a crash
Subsequent analyses compared only intervention and control sub-
jects who had a licence Waller 1992 reported crashes by person-
days of exposure but did not explain how they could be sure that
each study intervention and control participant was an active riderand actually exposed to the risk of experiencing a crash
Potential confounders that are unequally distributed by study arm
at baseline need to be dealt with in the analysis Satten 1980
Adams 1985 and Davis 1997 did not take into account any con-
founding factors either by design or in the analysis McDavid
1989 Waller 1992 and Billheimer 1998 accounted for potential
confounding factors at the design stage by matching the subjects
in the intervention and control groups Other studies (Raymond
1979 Mortimer 1984 Mortimer 1988 Perrino 2002 and also
Waller 1992) adjusted for one or two confounders by stratifica-
tionor standardisation (eg crashesmiles driven age group) Osga
1980 conducteda univariate analysis of crash rates in which no po-
tential confounders were accounted for but also calculated cumu-lative crash rates by mileage travelled Only Jonah 1982 Cooper
1988 and Savolainen 2007 conducted a multivariate analysis
Risk of bias in the case-control studies
A case-control design is suitable for the assessment of risk fac-
tors for outcomes such as motorcycle crashes injuries or deaths
but not appropriate for the evaluation of an intervention such
as motorcycle rider training It can potentially result in spurious
conclusions regarding the efficacy of the intervention because the
intervention has not been applied at random to an intervention
and control group Even with the control of possible confounders
using a case-control design to evaluate an intervention can lead tothese kinds of results since those who choose to undertake rider
training may differ on important prognostic factors that affect the
outcome compared to those who do not choose to participate in
rider training
Selection bias
A potential source of bias in case-control studies as with all other
study types where the intervention is not randomly allocated to
participants is that comparison groups are likely to differ on po-
tentially confounding characteristics This means that ideally all
cases need to be included in the study as well as a random sample
of controls taken from the population that gave rise to the cases If that is not possible cases and controls either need to be carefully
matched to make themcomparable or differences at baseline need
to be adjusted for in the analysis All four case-control studies were
population based Kraus 1975 included all fatalities and injured
cases leading to hospitalisation and appropriately recruited a ran-
dom sample of controls from the same population Hurt 1981 did
not manage to recruit all or a random sample of cases Only half
of the notifications of crashes to the police were received in time
by the researchers the research team for unreported reasons was
only able to respond to half of those again Data collection was
therefore completed for only about a fifth of all cases Neverthe-
less this was still a large sample of 900 cases Haworth 1997 using
16Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 1950
a similar study design was more likely to have included all cases
but this is not entirely clear from the study reported It is unclearfor example whether fatalities were included as they did not use
coronerrsquos reports for the recruitment of cases The MAIDS 2004
study was conducted in five European countries but it was unclear
how cases were recruited whether they were all cases during the
two years that the study lasted or a random sample of all cases
what the response rates were etc Controls were recruited from
randomly selected petrol stations but it was likewise not clear what
the eligibility criteria for inclusion in the study were how they
were recruited (eg randomly) and what the response rate was
Hurt 1981 and Haworth 1997 both matched controls to cases by
recruiting themat the sametime ofdayandsamedayof weekas the
crash occurred Hurt 1981 however recruited controls two years
later andit is likelythat there would have beendifferences betweenthe two comparison groups One important difference was that
during data collection for cases there was very little specialised
motorcycletrainingavailable in the studyregionwhereas two years
later when control data were collected there was some training
available
These four case-control studies also suffered from selection bias
due to potential differences between those that gave consent to
participate in the study and responded to interviews andquestion-
naires and those that did not None of the studies have assessed
such differences although Hurt 1981 did acknowledge that con-
trols giving consent to cooperate may have differed from those
motorcyclists who did not
Detection bias
Measures that can be taken to reduce detection bias in case-con-
trol studies are blinding of interviewers for case or control status
blinding of the participant for the study factor of interest validat-
ing measurement instruments and using these in an identical way
for cases and controls and cross checking self-reported informa-
tion with other more objective sources such as registers None of
the studies reported blinding of the intervention assessors to case
or control status but all used consistent measurement methods for
both case and control groups
All case-control studies used self-report as their main measure toassess whether cases and controls had received rider training and
if so the type of training The major potential source of bias in
case-control studies when self-report is used as a measure is recall
bias One way of reducing recall bias is by double-checking self-
reported data with more objective data Kraus 1975 also used po-
lice accident reports for information on rider characteristics but
such information was not available about rider training Haworth
1997 and MAIDS 2004 did not report on any measures to limit
recall bias Hurt 1981 reported further questioning of cases by
the interviewer when answers did not make sense or when par-
ticipants seemed to contradict themselves as part of the in-depth
unstructured interview technique This qualitative research tech-
nique can result in valuable information but as interviewers were
not blinded to case or control status of the participants it may have led to more thorough examination of cases compared to the
controls and thus differential detection bias
Intervention integrity
The level of detail about the rider training provided in the case-
control studies differed significantly but was minimal for each
Kraus 1975 simply asked whether lsquoanyrsquo motorcycle rider training
had lsquoeverrsquo been carried out Hurt 1981 asked participants whether
they hadbeen self taught informally taught or completed formally
taught rider training courses MAIDS 2004 asked whether partic-
ipants had had pre-license training additional (non-compulsory)
training other training or no training whereas Haworth 1997asked about the number of courses that a participant had com-
pleted whetherthey were basic intermediate or advanced courses
and how long ago the training had been done None of the case-
control studies reported collecting information about further de-
tails about any training programmes (eg type of training cur-
riculum duration of training)
Analysis
Kraus 1975 accounted for confounding factors through the exclu-
sion of females and stratification by age and frequency of motor-
cycle use Hurt 1981 matched cases and controls for time of day
and day of the week that the crash occurred and did not adjust forany confounding factors in the analysis Haworth 1997 matched
groups for time of day and day of the week that the crash occurred
and stratified by age blood alcohol level and licence status in the
analysis None of the studies used multivariate methods MAIDS
2004 did not provide any information about any participant char-
acteristics other than the study exposures of interest and did not
describe any methods to adjust for potential confounding factors
Effects of interventions
We have presented the results in a descriptive manner due to the
substantial heterogeneity in study types interventions and out-comes and the generally poor quality of the studies We have
grouped the interventions by mandatory and non-mandatory pre-
licence rider training and post-licence rider training and further
by study type and by outcome (crashes injurious and fatal crashes
and offences) We discuss results that are reported as statistically
significant (whether a P value was reported or not)
Pre-licence rider training - mandatory
Randomised trials
17Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2050
Crashes
Total group of participants
Anderson 1980 found a reduction in motorcycle crash rates in
the intervention group (MOST skill test with remedial training)
compared to the control group (who received the skill testwithout
a training component) The percent differenceranged from 143
six months after the programme (P lt 005) to 211 (P lt 0001)
after one year and 127 (P lt 001) after two years Hall 1988
found no statistically significant difference in crash rates between
intervention groups at any point in time
Subgroup of participants who obtained their licence only
Anderson 1980 found a reduction in motorcycle crash rates in the
intervention arm but this was statistically significant at one year
only (24 to 218 reduction P lt 005) Hall 1988 found no
statistically significant differences
Offences
Total group of participants
The percentage of offences in Anderson 1980 after six months
and one and two years was slightly higher in the intervention
arm compared to the control group but these differences were
not statistically significant Likewise the study by Hall 1988 did
not find a statistically significant difference in offences between
groups
Subgroup of participants who obtained their licence only
Hall 1988 found no significant difference in offence rates between
the intervention and control groups Anderson 1980 found thatthe percentage of offences increased in the intervention arms com-
pared to the control group This difference was statistically signif-
icant after two years (51 to 115 increase P lt 005)
Cohort studies
Crashes
Neitherof the cohort studies (Kloeden1994 and Billheimer1998)
found a statistically significant difference in crash rates between
the intervention and control groups
Offences
Billheimer 1998 (for the California Motorcyclist Safety Program
course) and Perrino 2002 did not find a statistically significant
difference in offence rates between the intervention and control
groups
Case-control studies
Crashes
MAIDS 2004 found that a slightly higher percentage of controls
(not involved in a crash) had had pre-license rider training than
cases (involved in a crash) This difference was not statistically significant (50 of controls vs 477 of cases OR 091 95 CI
076 to 110)
Pre-licence rider training - non-mandatory
Randomised trial
Injuries and deaths
Swaddiwudhipong 1998 found a significantly lower injury ratein the intervention villages (1051000 population 95 CI 92
to 120) compared to the control villages (1691000 population
95 CI 152 to 185) one year after the intervention This differ-
ence was not maintained two years after the intervention (127
95 CI 112 to 143 intervention 151 95 CI 135 to 167
control) There was no significant difference in death rates be-
tween groups
Cohort studies
Crashes
Cooper 1988 foundno statistically significant differencein crashes
between the intervention and control groups McDavid 1989
found a lower crash rate in the intervention group (14139) com-
pared to the control group (23139) but no test of statistical sig-
nificance was undertaken
Injuries and deaths
McDavid 1989 found that the intervention group had fewer in-
jurious crashes (814 crashes) compared to the control group (16
23 crashes) but no statistical significance test was reported
18Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2150
Offences
Cooper 1988 foundno statistically significantdifference in offencerates between the intervention and control groups McDavid 1989
foundno statistically significantdifference in offence ratesbetween
the intervention and control groups
Non-mandatory rider training
Non-randomised trials
Crashes
Neither Adams 1985 nor Leung 1987 found a statistically signif-icant effect of rider training on crash rates
Offences
Leung 1987 found a statistically significant lower offence rate in
the intervention group compared to the control group for both
males and females and for all age groups except for women 35
years or older (P values not reported)
Cohort studies
Crashes
Osga 1980 in the univariate analysis reported a lower percentage
of crashesin theintervention group (that continued to ride) (28
168595) than in the control group (34 112329) In the cu-
mulative crash rate analysis however Osga1980 found statistically
significant (at the P lt 001 level) higher crash rates in the interven-
tion group than in the control group at allmileage intervals (1000
to 29999 miles) travelled After riding more than 30000 miles
(when over 50 of the trained riders had been involved in a crash
compared to approximately 40 in untrained riders) this differ-
ence was no longer statistically significant Satten 1980 found a
lower percentage of crashes in the intervention group (13 969)
compared to the control group (25 2079) but the differencewas not assessed statistically Davis 1997 found a lower crash rate
in trained riders but it was unclear how this was calculated be-
cause the size of the control group was not reported Jonah 1982
and Billheimer 1998 (for the Experienced Rider Course) found no
statistically significant difference in crashes between the interven-
tion and control groups Although the difference between those
who completed the Experienced Rider Course and no course in
Billheimer 1998 neared significance at six months after the course
(P = 007) large P values were observed for one year after (P =
030) Raymond 1979 found a statistically significant higher crash
rate for the intervention group (1861000 miles ridden) com-
pared to the control group (1181000 miles ridden) in both males
and females (P lt 0001) especially during the first 5000 miles of
riding Waller 1992 found a statistically significant higher crashrate for the intervention group (27 crashes in three years or 070
crashes10000 person days) compared to the control group (5
crashes in three years or 013 crashes10000 person days) with a
crude relative risk of 538 (X 2=1513 plt005) Mortimer 1984
and Mortimer 1988 found no statistically significant difference in
crashes between the intervention and control groups
Savolainen 2007 found that the intervention group (those who
completed the Basic Rider Course) was 44 more likely to be
involved in crashes than the control group and those who com-
pleted the course multiple times were an additional 180 more
likely to be involved in a crash This finding was based on low
numbers and no P value or confidence level was given for either
result
Injuries and deaths
Satten 1980 also found a lower percentage of injurious crashes
(33 39) in the intervention group compared to the control
group(60 1220) although again no statistical significance test
was reported
Offences
Jonah 1982 did find a significantly lower self-reportedoffence rate
in the intervention group (15) compared to the control group(34 P lt 001) (no numbers given) Satten 1980 also found
a lower self-reported offence rate in the intervention group (1
71 versus 19100) but again this was not tested for statistical
significance Mortimer 1984 Mortimer1988 and Billheimer 1998
found no statistically significant differencein offence ratesbetween
the intervention and control groups
Case-control studies
CrashesMAIDS 2004 found that a slightly higher percentage of cases
(involved in a crash) had had additional non-mandatory rider
training than controls (not involved in a crash) This difference
was not statistically significant (17 of cases vs 12 of controls
OR 146 95 CI 064 to 337)
Rider training unspecified
Case-control studies
19Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2250
Injuries and deaths
Hurt 1981 found that the odds of injurious crashes were lower inthose who had received professional training compared to those
who had not Kraus 1975 found that in the younger age group
(less than 20 years) and the older age group (over 35 years) a larger
percentage of injured riders (12 and 21 respectively) reported
prior motorcycle rider training compared with the uninjured con-
trol group (6 and 5 respectively) (P lt 001) (no numbers
given) Haworth 1997 reported several different outcomes one
beingthat having completed at leastone trainingcoursecompared
to no training courses at all had no significant effect on the odds
of crashing (afteradjustment for age blood alcohol level or licence
status (learner permitprobationaryfull)) When comparing three
different levels of training courses (beginner intermediate and ad-
vanced) with no training no significant differences were found
in the odds of crashing However after adjustment for age and
blood alcohol level completing an advanced course as compared
to a beginners course was associated with a significant decrease in
the odds of severe motorcycle crashes resulting in serious injury or
death (OR 02 95 CI 01 to 08)
D I S C U S S I O N
The objective of this review was to quantify the effectiveness of
pre- andpost-licence motorcycle rider training on the reduction of
traffic crash involvement injuries deaths and offences committedby motorcycle riders Most included studies were cohort studies
There were a few randomised and non-randomised trials and case-
control studies
All studies compared either mandatory or non-mandatory formal
training with no formal training except for Haworth 1997rsquos case-
control study that compared levels of training (beginner interme-
diate and advanced) to each other
We found studies that evaluated the effectiveness of manda-
tory pre-licence training in reducing crash and offence rates
(Anderson 1980 Hall 1988 Kloeden 1994 Billheimer 1998
(the basic California Motorcyclist Safety Program component of
the study) Perrino 2002) non-mandatory training that when
passed resulted in a licence (Cooper 1988 McDavid 1989
Swaddiwudhipong 1998) non-mandatory training that could be
taken by motorcycle riders with or without a licence (Raymond
1979 Osga 1980 Satten 1980 Jonah 1982 Mortimer 1984
Adams 1985 Leung 1987 Mortimer 1988 Waller 1992 Davis
1997 Billheimer 1998 (the Experienced Rider Course compo-
nent of the study)) and unspecified rider training on crash injury
and death rates (Kraus 1975 Hurt 1981 Haworth 1997 MAIDS
2004)
The studies provided conflicting evidence regarding motorcycle
rider training which is likely due to the poor design quality and
reporting of the studies Most of the studies included in the review
were likely to be adversely affected by significant selection biasalong with substantial detection bias(measurement and recall bias)
which limits our ability to make conclusive comments
Mandatory pre-licence rider training
There were two RCTs three cohort studies and one case-control
study that examined pre-licence ridertraining The two RCTs eval-
uated the Motorcycle Skills Operating Test (MOST) (Anderson
1980 Hall 1988) Anderson 1980 found a statistically significant
reduction in crashes in the intervention group at six months one
and twoyearsfollow up However when only participantswho ob-
tainedtheirlicence were included in theanalysis thereduction wasstatistically significant only at one year follow up The other RCT
(Hall 1988) found no statistically significant differences in crash
rates at any point of time neither in the total population group
nor in the licensed subgroup Neither of the two cohort stud-
ies assessing crash rates found a statistically significant difference
in crash rates between intervention and control groups ( Kloeden
1994 Billheimer1998) andthe case-control study (MAIDS 2004)
also did not find a statistically significant difference in the odds of
experiencing a crash between trained and untrained riders With
regards to offences neither of the RCTs found a statistically sig-
nificant difference in offence rate between the intervention and
control groups in the analysis of the total group of participants
entering the trial However Anderson 1980 found a statistically significant increase in offence rates in the intervention group but
only after two years of follow up andonly in the analysis of partici-
pants who obtained their licence The two cohort studies assessing
offence rates found no statistically significant difference between
the groups (Billheimer 1998 Perrino 2002)
None of the five studies (RCTs or cohort) found a reduction in
crashes or offences following mandatory pre-licence rider training
except for Anderson 1980 who observed a statistically significant
reduction in crashes in the intervention group at one year in both
analyses It is not clear why both RCTs that evaluated the same
MOST and training intervention observed different results Both
studies enrolled large samples and observed a similar uptake of
licensure following the intervention It is possible that differencesmay have arisen due to different study samples (California ver-
sus New York) or that the intervention delivery differed between
studies
In both trials of the Motorcycle Skills Operating Test including
rider training only a third of motorcycle licence applicants ob-
tained their motorcycle licence It is possible that a licensing pro-
cess including a skills test may reduce crashes through decreased
exposure
Non-mandatory pre-licence training
20Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2350
Three studies assessed the effect of non-mandatory pre-licence
training The Cooper 1988 cohort study found no difference incrash or offence rates between the trained and untrained groups
We cannot however be confident about this result as the study
sufferedfrom selectionbiaswhichcouldhave biasedthe measure of
effect toward the null The RCT conducted by Swaddiwudhipong
1998 examined injury and death rates and found a statistically
significant lower injury rate in the intervention villages after one
year of follow up but no longer at two year follow up There
was no statistically significant difference in death rates between
intervention and control villages Our confidence in the results of
this study are hindered by the inadequacy of the randomisation
process that led to significant differences between the intervention
and control districts on two important variables The intervention
districts were more likely than the control districts to have a validlicence andto wear a helmetSincethese differences were nottaken
into account in later analyses they could explain the differences
observed in injury rates between districts Furthermore with only
six districts randomised the sample size was probably too small
to detect a difference between intervention and control groups
had there been one The McDavid 1989 cohort study found a
lower percentage of injuries in the intervention group although
a statistical test was not conducted McDavid 1989 drew their
intervention and control groups from different populations but
did adjust for confounders although they only analysed 40 of
the initial sample (139 trained and 139 untrained) Because of
these substantial weaknesses we are unable to draw conclusions
from this studyAn overall effect of pre-licence training on the reductionof crashes
injuries deaths and offences was not found although one should
be aware that the research of this issue to date has had significant
flaws However a mandatory skill test (such as the MOST) in
combination with and even without pre-licence training as part
of a licensing process seems to reduce the number of individuals
completing the licensing procedure and therefore indirectly re-
duces crash rates through a reduction in exposure to motorcycle
riding Making licensure a more difficult process may therefore be
seen as advantageous if it delays licensure and therefore reduces
exposure It is possible that the group who chose not to continue
riding a motorcycle would not have been the group to experience a
crash or commit an offence if they had continued to ride anyhowfor example because they are more safety conscious and this was
the reason for their giving up riding in the first place If this is the
case deterring individuals from completing their licence and rid-
ing may not necessarily decrease overall crashes deaths or offence
rates In addition other potentially adverse effects of complicat-
ing a licensing procedure need to be considered such as increases
in unlicensed riding In Thailand non-mandatory rider training
offered as a community intervention may have decreased injury
rates but this needs to be confirmed through further research
Non-mandatory training
The majority of the studies assessed non-mandatory training
(training independent of any licensing procedure) and the results
were varied The two non-randomised trials assessing crash rates found no statistically significant difference in crash rates between
the intervention and control groups (Adams 1985 Leung 1987)
Of the nine cohort studies Satten 1980 found a lower percentage
of crashesin the intervention group although it is unclear whether
this was a statistically significant result Davis 1997 also found a
lower percentage of crashesin theinterventiongroup although it is
unclear with which control group the intervention group wascom-
pared Raymond 1979 Osga 1980 Waller 1992 and Savolainen
2007 found a statistically significant higher crash rate in the in-
tervention group and the remaining cohort studies (Jonah 1982
Mortimer 1984 Mortimer 1988 Billheimer 1998) found no sta-
tistically significant difference in crash rates between intervention
and control groups No conclusions can be drawn from these stud-
ies with confidence The Satten 1980 and Davis 1997 studies did
not adjust for any confounding factors even though important
differences existed between the intervention and control groups
Although Raymond 1979 and Waller 1992 adjusted for a num-
ber of confounders at baseline (Waller 1992) and in the analy-
sis (Raymond 1979 Waller 1992) the intervention and control
groups were drawn from such different populations that it is pos-
sible that unmeasured confounders were responsible for the sta-
tistically significant increased risk of crashes observed in the inter-
vention group The observed difference could also be attributed totheobserved differential loss to followup (Raymond 1979) which
could bias the measure of effect towards or away from the null
The Osga 1980 study found a protective effect of rider training
in the univariate analysis but statistically significant higher crash
rates in the intervention group when accounting for riding expo-
sure Sex (a significantly higher proportion of females in the in-
tervention group) and other potentially confounding factors that
could explain thedifferencebetween the groups were howeverun-
accounted for Finally Savolainen 2007 did control for potential
confounding variables yet it is not clear whether variables were
controlled for simultaneously or one at a time One case-control
study (MAIDS 2004) found no statistically significant reduction
in the odds of experiencing a crash between trained and untrainedriders but the quality of this study was poor mainly because the
analysis did not account for any confounding factors
One cohort study (Satten 1980) assessed the effect of non-manda-
tory rider training on injurious crashes and found a lower percent-
age of injuries in the intervention group although a statistical test
was not conducted As mentioned above the Satten 1980 studyrsquos
validity is reduced due to its lack of adjustment for confounding
factors
The only non-randomised trial that assessed offence rates (Leung
1987) found a statistically significant lower offence rate in the in-
tervention group and of the five cohort studies that assessed of-
fence rates (Satten 1980 Jonah 1982 Mortimer 1984 Mortimer
21Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2450
1988) one study (Jonah 1982) also found a statistically significant
lower offence rate in the intervention group However we have lit-tle confidence in the results from Leung 1987 since potential con-
founding factors were not appropriately addressed Jonah 1982
did conduct a multivariate analysis but was hindered by an initial
low response rate The other four cohort studies did control for
several potential confounders and found no difference in offence
rates following non-mandatory post-licence rider training
The non-randomised trials and cohort studies evaluating non-
mandatory rider training were at high risk of substantial selection
bias due to the non-random allocation of participants as well
as low participation rates and attrition bias and could therefore
not provide conclusive evidence with regard to the effect of the
training It should be noted that despite the biases and other issues
with this group of studies most didnot detectmajor effects relatedto rider training in either direction This may suggest that any
potential impact of rider training will be modest
Studies examining unspecified modes of ridertraining
Three case-control studies assessed whether injured or deceased
motorcyclists were more or less likely to have had rider training
compared to control motorcyclists All had conflicting results
Hurt1981 foundthe odds of havingan injuriouscrash waslower in
those who had received rider training whereas Kraus 1975 found
that the odds of having an injurious crash were higher in youngerand older trained riders Haworth 1997 who compared advanced
versus beginner training found no effect on motorcycle injuries
or deaths although they did find that those who had completed
advanced training had reduced chances of crashing compared to
those who had done a beginnersrsquo training course Similar issues
with uncontrolled confounding and biases of selection as described
above arose with the three case-control studies Hurt 1981 suffers
from inadequate control of confounding via poor matching and
both the Kraus 1975 and the Haworth 1997 findings were only
controlled for limited factors by means of stratified analysis It is
likely that these limitations are the reason for the contradictory
directions of effect observed for the three case-control studies
Methodological weaknesses of the studies
In summary most studies investigating the effectiveness of mo-
torcycle rider training on reducing crashes injuries and death suf-
fered from serious methodological weaknesses
The majority of studies were non-randomised and were there-
fore limited mainly by selection bias Although the approaches of
matching stratification or regression analyses can minimise bias
(Fleiss 2004) they were poorly employed in the reviewed studies
The inadequate adjustment for confounders in particular adds
to our lack of confidence in the studiesrsquo results since it has been
demonstrated that individuals who choose to take motorcycle rider
training differ from those not taking rider training For instance
trained riders often differ from controls in age sex and experi-ence usually ride less and may have a more positive attitude to
safety as indicated by the reported more frequent wearing of seat
belts (Mortimer 1982)
The two large randomised trials reported that a large proportion
of their study group were never exposed to the risk of offences or
crashes because they never attained licensure This made it impos-
sible to assess whether a reduction in crashes in the intervention
group was a result of training or due to delayed or never attained
licensure
In some studies the sample sizes were too small to detect any sta-
tistically significant differences Follow-up time after training dif-
fered between studies and was an impediment to observing any
real effects of training if they existed As riding experience ac-cumulates over time any differences in outcomes between those
trained and untrained would diminish in studies with lengthy fol-
low-up times This dilution of any possible effect of rider training
may have also occurred in studies that mixed both novice and ex-
perienced riders in the study groups since it is hypothesised that
novice riders serve to benefit most from rider training This would
increase the difficulty in finding any effect of rider training par-
ticularly when sample sizes are small
Detection bias may also account for some of the differences ob-
served between trained and untrained riders Detection bias would
have occurred if the self-report hospital driver or police records
used to record the outcomes of crash injury death or offences
differed between the trained and untrained riders Research hasdemonstrated that not all crashes are reported to police for ex-
ample small crashes or off-road crashes (Kraus 1975) even serious
injury crashes have been found to be underreported (Anderson
1980) However this would only introduce detection bias if the
reporting of outcome differs for trained and untrained riders This
does appear possible in that responsible riders may be more likely
to enroll in a rider training course and also may be more likely
to report crashes to police than irresponsible riders meaning that
the number of crashes from riskier riders could be underrepre-
sented on records An analysis to investigate the effectiveness of
rider training may then show that rider training is not effective in
reducing crashes
Rather than using reported outcomes some studies used self-re-ported survey data although this technique suffers from non-
response biases In fact two studies reported differences in de-
mographic characteristics in response rates to questionnaires
(Anderson 1980 Hall 1988) Self-reported data also introduce
the risk of recall bias survival bias and social desirability response
bias Recall bias may have been equally distributed among trained
and untrained riders howeverit is hypothesised that trained riders
may have felt it was lsquoexpectedrsquo to report good driving histories
hence leading the rider training to appear effective
22Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2550
Limitations of this review
Limitations of this review were that we found very few recentstudies (the two most recent ones published in 2002 and 2007)
and that we found very few studies conducted in low-and middle
income countries even though our search strategy was set up to
identify such studies Anotherlimitation wasthat forthe reviewed
studies we could not assess all outcomes of rider training that are
potentially useful Some of the included studies assessed outcomes
such as use of safety clothing including helmets Such outcomes
are valuable since the use of motorcycle helmets has been demon-
strated to be effective in preventing head injuries (Liu 2004) De-
spite this the use of safety clothing was not included as an out-
come in this review because the studies measuring this outcome
were limited by poor data quality due to very low response rates
Another outcome of interest in assessing the effectiveness of rider
training is lsquotype of injuryrsquo As Haworth 2005 state in their review
of motorcycle licensing and training rider training could be con-
sidered effective if injuries become less severe because motorcycle
riders have learned how to avoid more serious injuries through
rider training The studies included in this review did not specify
type of injury and so this outcome could not be assessed Last
changes in knowledge and attitudes following rider training are
other potentially useful outcomes We have excluded such studies
(Osga 1979 Clayton 1990 Goldenbeld 2004) as these outcomes
are proxies rather than direct measures of the effectiveness of mo-
torcycle rider training We opted to report on the outcomes of
crashes injury death and offences alone since there is little ev-
idence about how changes in knowledge and attitudes relate to
these outcomes
A U T H O R S rsquo C O N C L U S I O N S
Implications for practice
Due to the poor quality of studies included in this review we were
unable to draw any conclusions about the effectiveness of rider
training on crash rates injurious or fatal or offence rates The
findings do suggest that mandatory pre-licence training may be
an impediment to completing a motorcycle licensing process thuspossibly indirectly reducing crash injury death and offence rates
through a reduction in exposure to riding a motorcycle On the
basis of the existing evidence it is not clear if and what type of
training reduces the risk of crashes injuries deaths or offences in
motorcyclists and the selection of the best rider training practice
can therefore not be recommended The suggestion that manda-
tory pre-licence training indirectly reduces exposure to riding the
motorcycle because trainees are impeded from completing the li-
censing process does not help those motorcyclists who persist in
obtaining their licence and riding the motorcycle Some type of
rider training is likely necessary to teach motorcyclists to ride a
motorcycle safely Good policy is based on sound evidence and
evaluation but such evidence for motorcycle rider training is seri-
ously lacking Policy makers are therefore encouraged to include arigorous evaluation component to any new or existing rider train-
ing programme Such an evaluation should ideally be randomized
and assess whether the training programme is effective and if so
which component(s) of the programme work best
Implications for research
Given theresults of this review which found littlereliable evidence
about the effectiveness of motorcycle rider training it is necessary
to build up an evidence base of the effectiveness of motorcycle
rider training in the reduction of crash injury death and offence
rates To this end well designed randomised controlled trials need
to be conducted that examine the impact of rider training oncrashes and offences A randomised controlled trial is the preferred
method as it would significantly reduce most biases described in
this review To minimise the impact of participants delaying or
never attaining their licence and therefore exposure as found in
the randomised trials in this review such a trial should include
riders who are already licensed and reliable measures of riding
exposure in a reasonably lengthy follow up period It is expected
that a randomised controlled trial of sufficient sample size would
observe equal balance on possible confounders by study arm how-
ever an assessment of potentially important confounders (such as
helmet wearing) by study group would enable control for these
variables in the analyses if necessary It is imperative that the effec-
tiveness of each component of the rider training (such as theorypractice knowledge skills and modification of behaviour) and the
synergy of such components are evaluated Outcomes should be
objective ideally those that are routinely collected rather than self-
reported outcomes Even though riding the motorcycle poses dif-
ferent challenges and incurs different risks lessons may be learned
from research into the effects of car driver training
Motorcycles are a common means of transport in low- and middle
income countries and usage is increasing As the burden of mo-
torcycle crash victims is mostly borne in such countries research
should be focused on the effectiveness of types of rider training on
the reduction of crashes injuries and fatalities in such countries
and be supported by higher income countries This review found
only one study conducted in a middle income country that sug-gested decreasing injury rates It would be beneficial to conduct
additional research to confirm this suggestion and to enhance our
knowledge of the role of rider training in low- and middle-income
settings
March 2006 saw the launch of the University of North Carolina
Highway Safety Research Centre partnership with the Motorcycle
Safety Foundation (MSF) and plans were revealed to conduct a
study of the potential benefits of motorcycle rider training The
study is funded by the MSF and the National Highway Traffic Sa-
fety Administration (NHTSA) Called rsquoThe Discovery Projectrsquo it
will compare data on crash rates from riders participating in a basic
23Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2650
motorcycle safety class with riders participating in a basic motor-
cycle safety class as well as a variety of continuing education classesThe collection of pilot data was due to start in March 2007 and
the data collection portion of the study is expected to take three
years (Highway Safety Research Center) It is vital that researchers
and policy makers collaborate to develop furtherhigh quality eval-
uations of motorcycle rider training programmes to help inform
development of new and effective training programmes
A C K N O W L E D G E M E N T S
Staff from The University of Sydney library for helping us locate
and retrieve full reports of many papers
R E F E R E N C E S
References to studies included in this review
Adams 1985 published data only
Adams A Collingwood V Job RFS Evaluation of the 1982 Australia
Post motorcycle rider training scheme NSW Australia Traffic
Accident Research Unit Traffic Authority NSW 1985 [ RN 85]
Anderson 1980 published data only
Anderson J Ford JL Peck RC Improved motorcyclist licensing and testing project Washington DC National Highway Traffic Safety
Administration (NHTSA) USA 19801ndash217 [
DOTndashHSndash805ndash598]
Billheimer 1998 published data only
Billheimer JW Evaluation of the California Motorcyclist Safety
Program Transportation Research Record 19981640100ndash9 [
TRB Report Number 980652]
Cooper 1988 published data only
Cooper PJ Rothe JP Motorcyclists who they are and why they do
what they do Transportation Research Record 1988116878ndash85
Davis 1997 published data only
David CF Evaluation of community traffic safety programs and
motorcycle operator training programs Connecticut Department of Transportation 199763
Hall 1988 published data only
Hall CJ Buchanan L Engle R Motorcycle rider education evaluationproject Washington DC USA National Highway Traffic Safety
Administration (NHTSA) US Department of Transportation
1988 [ DTNH 22ndash80ndashCndash05123]
Haworth 1997 published data only
Haworth N Smith R Brumen I Pronk N In Monash University
Accident Research Centre editor(s) Case-control study of motorcycle
crashes Victoria Australia Department of Transport and Regional
Development The Federal Office of Road Safety 19971ndash149 [
ISBN 0 642 25501 6 ]
Hurt 1981 published data only
Hurt H Ouellet J Thom D Motorcycle accident cause factors and
identification of countermeasures USA Department of
Transportation 1981 [ Technical Report DOT HS 806ndash862]
Jonah 1982 published data onlylowast Jonah BA Dawson N Bragg B Are formally trained motorcyclists
safer Accident Analysis amp Prevention 198214(4)247ndash55
Jonah BA Dawson NE Bragg BW Evaluation of the Canada
Safety Councilrsquos Motorcycle Training Program Proceedings of theInternational Motorcycle Safety Conference Washington DC
USA 198059ndash93
Kloeden 1994 published data only
Kloeden CN Moore VM McLean AJ Evaluation of the pre-license
training program for motorcyclists in South Australia Walkerville
South Australia Australia NHMRC Road Accident Research Unit
The University of Adelaide 19941ndash37 [ ISBN 0 7308 0607 3]
Kraus 1975 published data only
Kraus JF Riggins RS Franti CE Some epidemiological features of
motorcycle collision injuries American Journal of Epidemiology 1975102(1)74ndash98
Leung 1987 published data only
Leung KS Reding VA Evaluation of the Wisconsin motorcycle rider course Wisconsin Bureau of Policy Planning and Analysis
Wisconsin Department of Transportation 1987
MAIDS 2004 published and unpublished data
MAIDS In-depth investigations of accidents involving powered
two wheelers ACEM (Acemacemeu) 2004 Vol Final report
20179
McDavid 1989 published data only
McDavid JC Lohrmann BA Lohrmann G Does motorcycle
training reduce accidents Evidence from a longitudinal quasi-
experimental study Journal of Safety Research 198920(2)61ndash72
Mortimer 1984 published data only
Mortimer R Evaluation of the motorcycle rider course Accident
Analysis amp Prevention 198416(1)63ndash71
24Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2750
Mortimer 1988 published data only
Mortimer RG A further evaluation of the motorcycle rider courseJournal of Safety Research 198819(4)187ndash96
Osga 1980 published data only
Osga GA An investigation of the riding experiences of MSF rider
course participants Dissertation South Dakota USA The
University of South Dakota 19801ndash224
Perrino 2002 published data only
Perrino CS Ahmed A Callender A Rozier E Cantwell A Stewart
O The role of Marylandrsquos motorcycle rider course in promoting saver
behaviors and attitudes Baltimore USA National Transportation
Center Morgan State University 20021ndash52
Raymond 1979 published data only
Raymond S Tatum S An evaluation of the effectiveness of the RAC
ACU motorcycle training scheme - final report Road Safety ResearchUnit Dept of Civil Engineering University of Salford London
19791ndash131
Satten 1980 published data only
Satten RS Analysis and evaluation of the motorcycle rider courses
in thirteen Illinois counties International Motorcycle Safety
Conference Washington DC 1980145ndash93
Savolainen 2007 published and unpublished data
Savolainen Peter Mannering Fred Additional evidence on the
effectiveness of motorcycle training and motorcyclistsrsquo risk-taking
behaviour Transportation Research Board 2007 Vol March 20
19
Swaddiwudhipong 1998 published data only
Swaddiwudhipong W Boonmak C Nguntra P Mahasakpan P
Effect of motorcycle rider education on changes in risk behavioursand motorcycle-related injuries in rural Thailand Tropical Medicine
and International Health 19983(10)767ndash70
Waller 1992 published data only
Waller FL An analysis of the impact of the Texas Department of
Public Safety Motorcycle Operator Training Program upon
motocycle related traffic accidents Texas State University Public
Administration Program Applied Research Projects 199277
References to studies excluded from this review
Anderson 1980a published data only
Anderson LC Better biking a motorcycle rider improvement
program International Motorcycle Safety Conference Washington
DC USA Motorcycle Safety Foundation 1980231ndash53Aupetit 2008 published and unpublished data
Aupetit S Riff J Buttelli O Espie S Ergonomics research program
on motorcycle learning Contributions to the development of a
motorcycle simulator for training Transport Research Arena
Ljubljana 20089
Awane 1999 published data only
Awane T Integrating simulators in motorcycle safety education
Journal of International Association of Traffic and Safety Sciences 199923(1)26ndash35
Baldi 2005 published data only
Baldi S Baer JD Cook AL Identifying best practices states in
motorcycle rider education and licensing Journal of Safety Research200536(1)19ndash32
Batchler 1988 published data only
Batchler MW Evaluation of the pre-license motorcycle training course at the New South Wales traffic education centre Federal Office of
Road Safety (FORS) 19881ndash122 [ Report No CR 77 ]
Blanchard 2006 published data only
Blanchard HT Tabloski PA Motorcycle safety educating riders at
the teachable moment Mosby (CV) Company 200632330ndash2
Bonnett 2005 published data only
Bonnett J Mature age skills training for experienced motorcycle
riders Motorcycle ridersrsquo association of the Australian Capital
Territory 2005
Braver 2007 published data only
Braver ER et alPersuasion and licensure a randomized controlled
intervention trial to increase licensure rates among Maryland
motorcycle owners Traffic Injury Prevention 20078(1)39ndash46Clayton 1980 published data only
Clayton AB Sudlow DE An evaluation of the effectiveness of a one day training programme for learner motorcyclists Berkshire UK
Transport and Road Research Laboratory 19801ndash23 [ ISSN
0266ndash7045]lowast Clayton AB Sudlow DE An evaluation of the effectiveness of a one-day training programme for learner motorcyclists Berkshire UK
Road Safety Division Berkshire 19801ndash23 [ISSN 0266ndash7045]
Council 1978 published data only
Council FM Desper LP Dutt AK Driver education for motorcycle operation University of North Carolina Highway Safety Research
Center 19781ndash40
Crowther 2005 published data onlylowast Crowther G Engaging with motorcyclists UK police and theBikeSafe Road Safety Programme Published proceedings of the
Australasian Road Safety Research Policing Education Conference
Wellington New Zealand 2005 issue ISBN 0ndash473ndash10636ndash11ndash11
Daniello 2009 published data only
Daniello A Gabler HC Mehta YA Effectiveness of motorcycle
training and licensing Transportation Research Board 88th Annual
Meeting 200913
Dewanti 2007 published data only
Dewanti M Challenging efforts in promoting young motorcyclist
safety in Indonesia Road Safety on Four Continents 14th
International Conference Swedish National Road and Transport
Research Institute 2007
Du Plooy 1990 published data onlyDu Plooy DW Motorcycle training in South Africa The Human
Element International Motorcycle Safety Conference Orlando
Florida USA Motorcycle Safety Foundation 19904-1 to 4-9
Engel 1989 published data only
Engel U Krogsgaringrd Thomsen L [Faeligrdselsundervisning og
propaganda ndash en faeligrdselssikkerhedsfremmende foranstaltning]
Dansk Vejtidsskrift 198912295ndash7
Goldenbeld 2002 published data only
Goldenbeld C Houwing S de Craen S The development of
driving skills of young moped riders [De ontwikkeling van de
rijvaardigheid van jonge bromfietsers] SWOV Institute for Road
Safety Research (Stichting Wetenschappelijk Onderzoek
Verkeersveiligheid) Leidschendam the Netherlands 200263
25Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2850
Goldenbeld 2004 published data only
Goldenbeld C Twisk D de Craen S Short and long term effects of moped training a field experiment Transportation Research Part F 20047(1)1ndash16
Haworth 2000 published data only
Haworth N Smith R Kowadlo N Motorcycle training in Victoria
evaluation of rider training curriculum in Victoria Monash
University Accident Research Centre Reports 2000 Vol 165 issue
3131
Haworth 2005 published data only
Haworth N Mulvihill C Review of motorcycle licensing and
training Monash University Accident Research Centre 2005 [
Report no240]
Haworth 2006 published data only
Haworth N Mulvihill C Rowdon P Teaching old dogs new tricksTraining and older motorcyclists Journal of the Australasian College of Road Safety 200718(4)20ndash5lowast Haworth NL Mulvihill C Rowden PJ Teaching old dogs new
tricks Training and older motorcyclists Australasian Road Safety
Research Policing and Education Conference Gold Coast
Queensland 2006
Hill 1983 published data only
Hill D Motorcycle rider training testing and licensing Motor
Cycle Council of New South Wales Sydney 1983
Imatake 1980 published data only
Imatake Y On the activities in Japan of motorcycle safe riding
promotion committee (MSRPC) - especially the education of
unskilled riders of class 1 motorbikes International Motorcycle
Safety Conference Washington DC USA Motorcycle Safety
Foundation 1980255ndash68
Ingebritsen 1990 published and unpublished data
Ingebrigtsen S [Risikofaktorer ved ferdsel med moped og
motorsykkel] TOslashI-rapport 66 Oslo Transportoslashkonomisk
institutt 1990
Jain 2009 published and unpublished data
Jain A et alTwo wheeler accidents on Indian roads a study from
Mangalore India Elsevier 200916130ndash3
Jonah 1979 published data only
Jonah BA Dawson NE Validation of the motorcycle operator skill
test Accident Analysis amp Prevention 197911(3)163ndash71
Jonah 1981 published data onlyJonah BA Dawson NE Bragg WE Predicting accident
involvement with the Motorcycle Operator Skill Test Accident Analysis amp Prevention 198113(4)307ndash418
Kadar 1990 published data only
Kadar KJ Countermeasures motorcycle rider education delivery
system The Human Element International Motorcycle Safety
Conference Motorcycle Safety Foundation 19906-27 to 6-41
Kelsey 1986 published data only
Kelsey SL Liddicoat C Ratz M Licensing novice motorcyclists a
comparison of Californiarsquos standard test and the MOST II (Motorcycle
Operator Skill Test) administered at centralized testing offices Sacramento California USA California Department of Motor
Vehicles Research and Development Office 19861ndash74
Koch 1980 published data only
Koch H Advanced course for motorcycle riders InternationalMotorcycle Safety Conference Washington DC USA Motorcycle
Safety Foundation 1980269ndash81
Koch 1990 published data only
Koch H Brendicke R How to influence risk taking behaviour
within motorcycle rider training programs The Human Element
International Motorcycle Safety Conference Orlando Florida
USA Motorcycle Safety Foundation 1990Motorcycle Safety
Foundation 1990 p 9-1 to 9-25A
Labbett 2005 published data only
Labbett S Langham M Training can make the problem worse
Road Safety Congress 20051ndash12
Law 2005 published and unpublished data
Law TH Radin Umar RS Zulkaurnain S Kulanthayan S Impactof the effect of economic crisis and the targeted motorcycle safety
programme on motorcycle-related accidents injuries and fatalities
in Malaysia International Journal of Injury Control and Safety
Promotion 200512(1)9ndash21
Lowes 1990 published data only
Lowes BF In-traffic rider training The Human Element
International Motorcycle Safety Conference Orlando Florida
USA Motorcycle Safety Foundation 19907-46 to 7-57
Mahon 1976 published data only
Mahon J Teaching accident avoidance on the motorcycle Journal of Traffic Safety Education 1976231ndash29
Maxwell 1985 published data only
Maxwell GS Evaluation of a Program for Training MotorcycleRiders Evaluation and Program Planning 19858(4)339ndash47
Mayhew 1996 published data only
Mayhew DR Simpson SH Effectiveness and role of driver educationand training in a graduated licensing system Ottowa Canada The
Traffic Injury Research Foundation 19961ndash89 [ ISBN
0ndash920071ndash12ndash0]
McKnight 1987 published data only
McKnight AJ Milburn BG An evaluation of the modified
motorcycle operator skill test exercises NHTSA Report 1987 [
Report No DOT HS 807 212 ]
McPherson 1976 published data only
McPherson K McKnight AJ The development and evaluation of a
motorcycle skill test manual and knowledge test Washington DCUSA National Highway Safety Administration 1976
Mohan 2004 published data only
Mohan D Evidence-based interventions for road traffic injuries in
South Asia Journal of the College of Physicians and SurgeonsPakistan 200414(12)746ndash7
Morgan 1980 published data only
Morgan JK Advanced motorcycle training course content and
study evaluation International Motorcycle Safety Conference
Washington DC USA Motorcycle Safety Foundation 1980
283ndash312
Motoki 1990 published data only
Motoki M Yamazaki S A study on effective motorcycle rider
education The Human Element International Motorcycle Safety
26Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2950
Conference Orlando Florida USA Motorcycle Safety
Foundation 19909-26 to 9-57A
Nairn 1992 published data onlylowast Nairn RJ Motorcycle safety research literature review 1987 to
1991 Federal Office of Road Safety Australia 1993 Vol Report
CR 117
Newland 1980 published data only
Newland R Darby J Automotive practicespre motorcycle rider
program International Motorcycle Safety Conference Washington
DC USA Motorcycle Safety Foundation 1980127ndash9
Newland 1982 published data only
Newland R Motorcycle rider training and licensing requirements
Driver training steering a course for the future Melbourne
Victoria Transport Regulation Board 1982
Newland 1990 published data only
Newland R Motorcycle training and licensing in Victoria 1980-
1990 The Human Element International Motorcycle Safety
Conference Orlando Florida USA Motorcycle Safety
Foundation 1990
Ormston 2003 published data only
Ormston R Dudleston A Pearson S Stradling Evaluation of
Bikesafe Scotland Scottish Executive Social Research Edinburgh
2003
Osga 1979 published data only
Osga GS Ellingstad VS An evaluation of the effectiveness of the 1978
South Dakota motorcycle rider course program South Dakota USA
University of South Dakota 19791ndash89
Panou 2007 published data only
Panou M Bekiaris E A new concept on the integration of driving
simulators in driving training the train-all approach Road Safety
on Four Continents 14th International Conference Swedish
National Road and Transport Research Institute 2007
Peterson 1990 published data only
Peterson A Kadar K Status report 1980-1989 Minnesota rider
education statistics The Human Element International
Motorcycle Safety Conference Orlando Florida USA 199013-
113 to 13-129
Prem 1984 published data only
Prem H Good MC Motorcycle rider skills assessment ACT
Australia Office of Road Safety Department of Transport 1984
1ndash377 [ Report No CR 34 ]
Radin Umar 2006 published data only
Radin Umar RS Motorcycle safety programmes in Malaysia how
effective are they International Journal of Injury Control and Safety Promotion 200613(2)71ndash9
Rockwell 1990 published data only
Rockwell TH Kiger S Carnot M An evaluation of the Ohio
motorcyclist enrichment program Ohio Department of Highway
Safety 1990
Rothe 1987 published data only
Rothe JP A motorcycle riding course Motorcyclists Image and reality Canada Insurance Corporation of British Columbia 1987
59ndash83
Rowden 2007 published data only
Rowden P Watson B Haworth N What can riders tell us aboutmotorcycle training A view from the other side of the fence
Australasian Road Safety Research Policing and Education
Conference Melbourne Australia Melbourne 2007
RSC Victoria 1993 published data only
Road Safety Committee Inquiry into Motorcycle Safety in Victoria
Parliament of Victoria 1993
Russam 1979 published data only
Russam K Motorcycle training and licensing in Japan UK Transport
and Road Research Laboratory (TRRL) 1979 [ Report 916 ]
Saffron 1981 published data only
Saffron DG Driver and rider education training and licensing a
brief review Traffic Accident Research Unit Traffic Authority of
New South Wales NSW Australia Traffic Accident Research UnitTraffic Authority of New South Wales 1981
Schulz 1990 published data only
Schulz U Motorcycle accidents among learner riders Motorcycle
licensing Evaluation of the effectiveness of a Graduated Licensing Scheme Forshungshefte des Institut fur Zweiradsicherheit 1990
Sexton 2004 published data only (unpublished sought but not used)
Sexton B Baughan C Elliott M Maycock G The accident risk of
motorcyclists 2004TRL Report TRL 607(ISSN 09684107)58
Shephard 1986 published data only
Shepherd R Do Motorcycle Safety Foundation programs work
Driver 198619(9)16ndash8
Simpson 1990 published data onlySimpson HM Mayhew DR The promotion of motorcycle safety
training education and awareness Health Education Research
Theory and Practice 19905(2)257ndash64
Sokolov 1990 published data only
Sokolov O Safe motorcycle riding some aspects of motorcycle
riding in the USSR The Human Element International
Motorcycle Safety Conference Motorcycle Safety Foundation
19904ndash37 - 4-48
Steffens 1988 published data only
Steffens U Gawatz R Willmes G Effectiveness of mofa courses
[Wirksamkeit von Mofakursen] Unfall- und sicherheitsforschung
strasenverkehr 198867168
Sudlow 2003 published data only
Sudlow D Scoping study on motorcycle training Road Safety
Research Report London UK The British Institute of Traffic
Education Research Department for Transport 2003 Vol June
[ Report No 36]
Tasmania 1995 published data only
Mannering FL Grodsky LL Motorcycle rider training scheme
Traffic Education Research Centre Armidale Australia Driver
Training Australia Limited 1995 Vol April1ndash56
Tetard 1985 published data only
Tetard C Motorcycle ridersrsquo training to detect indices in an
intersection situation [Lrsquoentrainement a la prise drsquoinformation dans
lrsquoapprentissage de la conduite motocycliste] Le Travail Humain198548(2)147ndash59
27Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3050
Thackray 1980 published data only
Thackray RM Prescott JC Field test of the motorcycle ridercourse International Motorcycle Safety Conference Washington
DC USA Motorcycle Safety Foundation 1980195ndash7
Thompson 1994 published data only
Thompson M Evaluation of compulsory basic training for
motorcyclists Transport Research Laboratory Crowthorne
Berkshire UK 1994 [ Report 63]
Tobler 1990 published data only
Tobler U Individual driving instruction for motorcycle driving
learners the Swiss experiment of motorcycle driving school The
Human Element International Motorcycle Safety Conference
Orlando Florida USA Motorcycle Safety Foundation 1990
Transport 1989 published data only
Transport Communications Road Crash Statistics Australia Bureauof Transport and Communications Economics 198914ndash20
Tuan 2007 published data only
Tuan VA Shimizu T Towards development and evaluation of the
motorcycle drivers re-education program in Vietnam modeling of
motorcycle driverrsquos undesired behaviors 11th World Conference
on Transport Research 200725
Victor 1980 published data only
Victor DJ Traffic safety education for motorcyclists International
Motorcycle Safety Conference Washington DC USA Motorcycle
Safety Foundation 1980221ndash8
Waterer 1990 published data only
Waterer TH Compulsory training for motorcyclists from liberty
to compulsion The Human Element International Motorcycle
Safety Conference Orlando Florida USA Motorcycle Safety
Foundation 19904-49 to 4-57
Weller 1989 published data only
Weller RB Chandler EW Motorcycle safety and motorcycle
education past research and survey results Journal of Public Policy
amp Marketing 1989893ndash108
Wells 1981 published data only
Wells P Observations of some RACACU motorcycle training
schemes Transport and Road Research Laboratory Crowthorne
Berkshire UK 1981
Winn 1986 published data only
Winn G Two important trends in motorcycle safety regulations
rider education and conspicuity improvement American
Motorcyclist Association 1986147ndash52
Wisher 1988 published data only
Wisher G Reid J Motorcycle accidents in Norfolk a three-year study of motorcycle training and the Part 1 test Norfolk UK Road Safety
Division Highways Department Norfolk County Council 1988
Wood 1987 published data only
Wood HT Bowen R Evaluation of the revised learner permit scheme
July 1983 to December 1985 Victoria Australia Road Traffic
Authority 19871ndash27 [ ISBN 0 7306 0247 8]
Yamazaki 1990 published data only
Yamazaki S Nakanishi M Survey on effect of Honda rider
education The Human Element International Motorcycle Safety
Conference Orlando Florida USA 199012-167 to 12-187
References to ongoing studies
Burgess 2010 unpublished data only
Burgess C Broughton P Fylan F Stradling Interim evaluation of
the UKs National Ride program In L Dorn editor(s) Driver
behaviour and Training Vol IV Aldershot Ashgate Publishing
Limited 2010
Additional references
Anderson 1980
Anderson J Ford JL Peck RC Improved motorcyclist licensing and
testing project Volume I and II National Highway Traffic Safety
Administration US department of Transportation Washington
DC 1980
Bjornstig 1985
Bjornstig U Bylund P Lekander T Brorsson B Motorcycle fatalitiesin Sweden Acta Chirurgica Scandinavica 1985151(7)577ndash81
Clayton 1990
Clayton A Sudlow DE An evaluation of the effectiveness of a one-
day training program for learner motorcyclists The Human
Element International Motorcycle Safety Conference Orlando
Florida USA 199013-105 to 13-111
Cochrane Collaboration 2008
Higgins JP Green S Cochrane Handbook for Systematic Reviews of
Interventions 500 The Cochrane Collaboration 2008
Elliot 2003
Elliott MA Baughan CJT Broughton J Grayson G Knowles J et
alMotorcycle Safety a Scoping Study TRL [ Report No
0968ndash4107]Fleiss 2004
Fleiss JL Levin B Paik MC Statistical methods for rates and proportions 3rd Edition John Wiley amp Sons Inc 2004 [ ISBN
978ndash0ndash471ndash52629ndash2 ]
FORS 1999
FORS Federal Office of Road Safety Road risk for sober licensed
motorcyclists Vol Monograph 27 Federal Office of Road Safety
19991ndash2
Haworth 2005
Haworth N Mulvihill C Review of motorcycle licensing and training
Victoria Australia Monash University Accident Research Centre
20051ndash83 [ Report No 240]
Huwiler-Muumlntener 2002Huwiler-Muumlntener K Juumlni P Junker C Egger M Quality of
Reporting of Randomized Trials as a Measure of Methodologic
Quality JAMA 20022872801ndash4
Johnston 2008
Johnston P Brooks C Savage H Fatal and serious road crashes
involving motorcyclists Road Safety Research and Analysis Report
ATSB Monograph 20 Canberra Department of Infrastructure
Transport Regional Development and Local Government 0205
081ndash26
Juumlni 2001
Juumlni P Altman DG Egger M Systematic reviews in health care
Assessing the quality of controlled clinical trials BMJ 2001323
42ndash6
28Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3150
Kopits 2003
Kopits E Cropper M Traffic fatalities and economic growth TheWorld Bank Development Research Group Infrastructure and
Environment Policy research working paper 3035 2003
Leung 1983
Leung P Chau BTW Motorcycle accidents in Hong Kong pattern
of injury treatment causative factors and socio-economic
implications Injury 198315(1)23ndash9
Liu 2004
Liu B Ivers R Norton R Blows S Lo SK Helmets for preventing
injury in motorcycle riders Cochrane Database of Systematic Reviews 2004 Issue 2 [DOI 10100214651858CD004333pub3]
Mohan 2002
Mohan D Road safety in less-motorized environments future
concerns International Journal of Epidemiology 200231(3)527ndash32Mortimer 1982
Mortimer RG An evaluation of the motorcycle rider course
Proceedings of the 20th conference of the American Association for
Automotive Medicine Arlington Heights IL USA American
Association for Automotive Medicine 198277ndash96
Mullin 1997
Mullin BT Risk factors for motorcycle injury the role of age
gender experience training and alcohol PhD thesis Auckland
New Zealand University of Auckland 1997
Noordzij 2001
Noordzij PC Forke E Brendicke R Chinn BP Integration of needs
of moped and motorcycle riders into safety measures review and
statistical analysis in the framework of the European research
project PROMISING SWOV Institute for Road Safety Research
Leidschendam The Netherlands 2001 [ Report No Dndash2001ndash5]
Osga 1979
Osga G Ellingstad VS An evaluation of the effectiveness of the 1978 South Dakota Motorcycle Riders Course Program South Dakota
University of South Dakota Vermillion Human Factors Lab 1979
Oxford English Dictionary Online
httpoxforddictionariescom Oxford English Dictionary Online
Oxford University Press 2010
Peden 2004
Peden M Scurfield R Sleet D World Report on Road TrafficInjury Prevention Geneva World Health Organization Available
at httpwwwwhointviolence˙injury˙preventionpublications
road˙trafficworld˙reporten 2004Chapter 3
Radin Umar 1995
Radin Umar R Macakay G Hills B Preliminary analysis of
motorcycle accidents Short-term impacts of the running
headlights campaign and regulations in Malaysia J Traffic Med 199523(1)17ndash27
Tsai 1995
Tsai YJ Wang JD et alCase-control study of the effectiveness of
different types of helmets for the prevention of head injuries among
motorcycle riders in Taipei Taiwan American Journal of
Epidemiology 1995142(9)974ndash81
Vis 1995
Vis AA In-depth study of the hazards of motorcycling [De
onveiligheid van motorrijden nader bekeken] SWOV Institute for
Road Safety Research Leidschendam the Netherlands 1995 [
Report No Rndash95ndash69]
Weaver 1990
Weaver EA Motorcycle safety education in the United States
Motorcycle Safety Foundation Orlando Florida USA 1990
Winn 1987
Winn G Two important trends in motorcycle safety regulations
rider education and conspicuity improvement No more
information available Detroit MI US Society of Automotive
Engineers 1987147ndash52
Yuan 2000
Yuan W The effectiveness of the rsquoride-brightrsquo legislation for
motorcycles in Singapore Accident Analysis amp Prevention 200032
(4)559ndash63
Zambon 2006
Zambon F Hasselberg M Factors affecting the severity of injuries
among young motorcyclists a Swedish nationwide cohort study
Traffic Injury Prevention 20067(2)143ndash9lowast Indicates the major publication for the study
29Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3250
C H A R A C T E R I S T I C S O F S T U D I E S
Characteristics of included studies [ordered by study ID]
Adams 1985
Methods Non-randomised trial
Participants Initial sample size
Total of 162 Australia Post telegram delivery riders
Not clear how many in intervention and control groups
Analysis sample size
Intervention group 63Control group 35
Total 98 (61)
Interventions No formal training vs voluntary rider training course of 12 hours over 2 consecutive days (theory 4 hours practice
8 hours)
Intervention integrity
No information given
Outcomes Australia Post recorded crashes
Length of follow up 1 year
Notes Study conducted in NSW Australia
Anderson 1980
Methods Randomised trial
Participants Initial sample size
Total 41188 motorcycle licence applicants
Analysis sample size
Intervention group with skills training 13160
Intervention group without skills training 12634
Control group 15080
Total 40874 (99)
Interventions Standard licensing skill test vs the Motorcycle Operators Skill Test (MOST) with or without a training component
MOST is not to be confused with rider training because it is a licensing test The training component was for
participants who failed the skill test on the first attempt was of 3 hours duration and consisted of instructions about
the skills needed to pass the skill test The training consisted of 30 min theory and 2 frac12 hours practice on the
motorcycle
Intervention integrity
33 in MOST with skills training 31 in MOST without skills training and 43 in control group obtained their
motorcycle licence in the end
Outcomes Police recorded crashes and offences
Length of follow up 3 years
30Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3350
Anderson 1980 (Continued)
Notes Study conducted in California USA
Billheimer 1998
Methods Cohort study
Participants Initial sample size
Intervention group Unclear numbers of CMSP graduates and ERC graduates
Control group Over 16000 untrained motorcyclists
Analysis sample size
1139 matched CMSP-pairs and 1182 ERC-pairs
Interventions No formal training vs the mandatory (for those younger than 18 years) pre-licence 16-hours basic California
Motorcyclist Safety Program (CMSP) and the 8-hour Experienced Rider Course (ERC)
Intervention integrity
No information given
Outcomes Police recorded crashes and offences
Length of follow up 2 years
Notes Study conducted in California USA
Cooper 1988
Methods Cohort study
Participants Initial sample size
Intervention group 418 graduates and 57 who failed
Control group 402
Analysis sample size
Intervention group unclear
Control group unclear
Total 863 (984)
Interventions No formaltrainingvs non-mandatory pre-licence rider trainingThe content of thetrainingwas notfurther explainedIntervention integrity
Participants had all completed the rider training
Outcomes Police recorded crashes and offences
Length of follow up 7 years
Notes Study conducted in British Columbia Canada
31Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3450
Davis 1997
Methods Retrospective cohort study
Participants Initial sample size
Intervention group 9320 graduates
Control group unclear
Analysis sample size
Intervention group 9320 graduates
Control group unclear
Interventions No formal training vs the non-mandatory Connecticut Motorcycle Operator Training Program The content of
training was not further explained
Intervention integrity Participants had all completed the rider training
Outcomes Police recorded crashes severity of crashes
Length of follow up 1 year
Notes Study conducted in Connecticut USA
Hall 1988
Methods Randomised trial
Participants Initial sample sizeIntervention group A 6248
Intervention group B 6616
Intervention group C 6996
Control group 6604
Total 26464 motorcycle licence applicants
Analysis sample size
Intervention group A 6012 (96)
Intervention group B 6428 (97)
Intervention group C 6778 (97)
Control group 6502 (98)
Interventions Standard licensing skill test vs the Motorcycle Operator Skill Test (MOST II) without training (A) MOST II with
3 hours training (B) or MOST II with 20 hours training (C) The 3 hours training comprised instructions aboutthe skills needed to pass the skill test (30 min theory and 2 frac12 hours practice on the motorcycle) The 20 hours
training consisted of the Motorcycle Safety Foundationrsquos Motorcycle Rider Course (MRC) comprising of 8 hours
theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-road practice
Intervention integrity
25 and 26 obtained their motorcycle licence in the lsquoMOST 3-hour trainingrsquo and lsquoMOST 20-hour trainingrsquo
intervention arm respectively In the lsquoMOST testrsquo intervention arm 29 obtained their licence and in the control
group 35
Outcomes Police recorded crashes and offences
Length of follow up 3 years
32Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3550
Hall 1988 (Continued)
Notes Study conducted in New York USA
Haworth 1997
Methods Population based case-control study
Participants Cases Motorcycle riders or pillion riders involved in motorcycle crash (on road) Recruited from hospital and police
records Not clear whether all cases (especially fatalities) were included
Community controls Multiple matched sample of motorcycle rider or pillion who passed the crash site at the same
time of day and day of week that the crash occurred
Initial sample size
Cases 336
Controls 1195
Analysis sample size
Cases 234 (70)
Controls 561 (47)
Interventions Basic handling skills course beginner intermediate or advanced course measured through self-report not validated
Outcomes Injuries with an Injury Severity Score of 5 or greater or admission for at least four hours and fatalities from a
motorcycle crash (from hospital and police records)
Length of follow up two years or more prior to the interview
Notes Study conducted in Victoria Australia
Hurt 1981
Methods Population based case-control study
Participants Cases Motorcycle riders who had a motorcycle crash that was reported to police (traffic accident reports) Not all
cases included
Community controls Multiple matched sample of motorcycle riders who passed the crash site at the same time of
day and day of week as the crash occurred but 2 years later
Initial sample size
Cases 4500 notifications of crashes to police (only 50 received in time research team able to respond to 1126 for
unreported reasons data collection completed for 900)
Controls Number approached unreported 2310 consented and interviewed
Analysis sample size
Cases 808 (808900 data on rider training) - 18 of initial sample
Controls 670 (6702310 data on rider training) - lt 29 of initial sample
Interventions Training experience self taught friends-family school-club formal-professional others measured through self-
report validated
Outcomes Injuries and fatalities from a motorcycle crash (from hospital records and coronerrsquos office)
33Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3650
Hurt 1981 (Continued)
Notes Study conducted in Los Angeles USA
Jonah 1982
Methods Cohort study
Participants Initial sample size
Intervention group
2310 licensed MTP graduates who are active riders
Control group
2029 licensed motorcyclists who are active riders
Non-licensed and non-active riders excluded
Analysis sample size
Intervention group
811 interviews (35) 533 driver records (23)
Control group 1080 interviews (71) 905 driver records (45)
Interventions No training vs the non-mandatory Canada Safety Council Motorcycle Training Program (MTP) comprising of 4
hours theory and 16 hours practise (6 hrs on-road) over 2 weekends
Intervention integrity
Participants had all successfully completed the MTP
Outcomes Self-reported and police reported crashes and offencesLength of follow up 4 years
Notes Study conducted in Ontario Canada
Kloeden 1994
Methods Cohort study
Participants Initial sample size
Intervention group 2533 licence applicants
Control group Unclear
Sample of riders obtaining probationary licenceIntervention group 1625 (64)
Control group 5015 (unclear)
Non-active riders excluded from control group (~66)
Analysis sample size
Intervention group 631 (39)
Control group 318 (6)
Interventions No formal training vs the mandatory pre-licence Ridersafe course consisting of two levelsthe first is a training course
in the basics of motorcycle riding where a pass entitles the rider to obtain a learnerrsquos permit The second is a skill
test where a pass entitles the rider to obtain a probationary motorcycle licence The training course consists of two 4
hour training sessions on separate days involving instruction videos and actual riding on motorcycles The skill test
consists of a 4 hour preparation course of 74 minutes of lecturediscussion 38 minutes of video and 75 minutes of
34Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3750
Kloeden 1994 (Continued)
on-course riding which is the testing phase
Intervention integrity
Unclear what percentage in the Ridersafe group and the control group obtained their probationary licence halfway
and their full motorcycle licence in the end
Outcomes Police reported crashes
Length of follow up 1 year
Notes Study conducted in SA Australia
Kraus 1975
Methods Population based case-control study
Participants Cases All motorcycle riders with a confirmed medically treated motorcycle injury (including fatalities) from all
hospitals in study base and peripheral area hospitals and all deaths due to motorcycle crashes in the study base All
serious injuries (leading to hospitalisation) and fatalities likely to be included
Community controls A random sample of registered motorcycle owners in the same temporary and geographical
study base as the cases arose from
Initial sample size Cases 1273
Controls 738
Analysis sample size Cases 741 (58)
Controls 434 (59)
Interventions Motorcycle rider training (not specified) measured through self-report
Outcomes Injuries leading to hospitalisation fatalities
Length of follow up not specified
Notes Study conducted in Sacramento county USA
Leung 1987
Methods Non-randomised trial
Participants Initial sample sizeIntervention group unknown number of MRC graduates
Control group unknown number of licensed motorcyclists
Analysis sample size
Intervention group 2941
Control group 43094
Interventions No formal training vs non-mandatory Wisconsin motorcycle rider course a 22 hours off-street programme that
teaches the basic rules of safe motorcycling
Intervention integrity
50 in the MRC group and 100 in the control group obtained their motorcycle licence in the end
35Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3850
Leung 1987 (Continued)
Outcomes Police recorded crashes and offences
Length of follow up up to 3 years
Notes Study conducted in Wisconsin USA
MAIDS 2004
Methods Population based case-control study
Participants Cases Motorcycle riders who had a motorcycle crash Unclear whether all cases included
Community controls Motorcycle riders identified at petrol stations Unclear whether matched to cases
Initial sample size
Cases not reported
Controls not reported
Analysis sample size
Cases 921 (Italy 200 Netherlands 200 Spain 121 Germany 250 France 150)
Controls 923 (Italy 200 Netherlands 200 Spain123 Germany 250 France 150)
Interventions Pre-license rider training or additional training (non-compulsory)
Outcomes Crashes
Notes Study in five sampling areas located in five European countries (France Germany Netherlands Spain and Italy)
McDavid 1989
Methods Cohort study
Participants Sample size after first match
Intervention group 346 male training graduates
Control group 346 male motorcycle riders licensed in same year
Sample size after second match
Intervention group 139 (40)
Control group 139 (40)
Analysis sample size
Intervention group 139 (40)
Control group 139 (40)
Interventions No formal training vs the non-mandatory pre-licence British Columbia Safety Councilrsquos 37-hour motorcycle safety
programme
Intervention integrity
Participants had all completed and passed the training
Outcomes Police reported crashes injurious crashes
Length of follow up 1-6 years
36Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3950
McDavid 1989 (Continued)
Notes Study conducted in British Columbia Canada
Mortimer 1984
Methods Cohort study
Participants Initial sample size
Intervention group 608 MRC graduates
Control group 360 motorcycle riders
Response rates
Intervention group 360 (59)
Control group 324 (90)
Non-active riders excluded
Analysis sample size
Intervention group 213 (35)
Control group 303 (84)
Interventions No formal training vs the non-mandatory Motorcycle Safety Foundationrsquos Motorcycle Rider Course comprising of
8 hours theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-road practice
Intervention integrity
Participants had all completed and passed the MRC
Outcomes Self-reported crashes and offencesLength of follow up 1 year
Notes Study conducted in Illinois USA
Mortimer 1988
Methods Cohort study
Participants Initial sample size
Intervention group 2307 MRC graduates
Control group 556 active licensed motorcycle riders
Response ratesIntervention group 1298 (56)
Control group 500 (90)
Non-active riders excluded
Analysis sample size
Intervention group 913 (40)
Control group 500 (90)
Interventions No formal training vs the non-mandatory Motorcycle Safety Foundationrsquos Motorcycle Rider Course comprising of
8 hours theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-road practice
Intervention integrity
Participants had all completed and passed the MRC
37Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4050
Mortimer 1988 (Continued)
Outcomes Self reported crashes and offences
Length of follow up 1 year
Notes Study conducted in Illinois USA
Osga 1980
Methods Cohort study
Participants Initial sample sizeIntervention group 1163 MSF MRC participants invited 818 (74) responded
Control group 329 motorcycle riders responded to recruitment strategies
Non-active riders excluded from analysis
Analysis sample size
Intervention group 595 (73)
Control group 329 (100)
Interventions No formal training vs the non-mandatory Motorcycle Safety Foundationrsquos Motorcycle Rider Course (MRC) com-
prising of 8 hours theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-road
practice
Intervention integrity
Participants had all successfully completed the MRC
Outcomes Self-reported crashes police recorded crash reports
Length of follow up 6 months - 3 12 years
Notes Study conducted in South Dakota USA
Perrino 2002
Methods Cohort study
Participants Initial sample size
Intervention group 2144 (unclear) MRC graduates (motorcycle licence applicants)
Control group 2601 (unclear) regular licensing test graduates (motorcycle licence applicants)Analysis sample size
Intervention group 1070 (unclear)
Control group 867 (unclear)
Interventions No formal training vs the mandatory (for those younger than 18 years) pre-licence Maryland Motorcycle Rider
Course (MRC) consisting of 8 hours theory and 12 hours practice
Intervention integrity
No information given
Outcomes Self-reported and police recorded crashes and offences
Length of follow up 1 year
38Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4150
Perrino 2002 (Continued)
Notes Study conducted in Maryland USA
Raymond 1979
Methods Cohort study
Participants Initial sample size
Intervention group
258 volunteers mainly (but not all) novice riders
Control group
1104 provisional licence applicants
Analysis sample size
Intervention group 190 (74)
Control group 627 (57)
Interventions No formal training vs the non-mandatory RACACU (Royal Automobile ClubAuto-Cycle Union) training scheme
The training program consisted of either 24 one-hour sessions or 13 two-hour sessions Twelve hours of the training
focused on theory about basic operational procedures traffic laws and highway code The practical session focused
on basic operational skills riding on a training area and riding in traffic with an instructor A proficiency test was
administered at the end of the course
Intervention integrity
No information given
Outcomes Self-reported crashes (including minor non-injury non-damage and off-road crashes)
Length of follow up 1 year
Notes Study conducted in the UK
Satten 1980
Methods Cohort study
Participants Initial sample size
Intervention group interviewers kept interviewing until they had 100 riders selected from a population of 891
enrollees in the MSF MRC course (unknown how many were asked and how many refused to participate)Control group unknown number of registered motorcycle riders
Non-active riders excluded from analysis
Analysis sample size
Intervention group 71
Control group 100
Interventions No formal training vs the non-mandatory Motorcycle Safety Foundationrsquos (MSF) Motorcycle Rider Course (MRC)
comprising of 8 hours theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-
road practice
Intervention integrity
Participants had all successfully completed the MRC
39Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4250
Satten 1980 (Continued)
Outcomes Self-reported crashes injuries and offences
Length of follow up 3 years
Notes Study conducted in Illinois USA
Savolainen 2007
Methods Cohort study
Participants Initial sample size
Intervention group 4000 riders that had conducted the MSF MRC course (passed or failed)
Control group 4000 motorcycle owners or permit or endorsement holders from the Indiana Bureau of Motor
Vehicles database
Analysis sample size
Intervention group 739 (18)
Control group 588 (15)
Interventions Intervention integrity
No information given
Outcomes Probability of crash involvement
Length of follow up up to 5 years
Notes Study conducted in Indiana USA
Swaddiwudhipong 1998
Methods Randomised trial
Participants Initial sample size
Intervention group
all 16 villages of 3 sub-districts
(participants were practising motorcyclists licensed and unlicensed)
Control group
18 villages of 3 other sub-districtsAnalysis sample size
Intervention group 3 districts (100)
Control group 3 districts (100)
Interventions No rider education vs non-mandatory community based rider education consisting of theory on motorcycle injury
risk helmet use traffic laws vehicle regulations and traffic signs and a written and skill test leading to a licence
Intervention integrity
No information given
Outcomes Injuries fatalities
Length of follow up 2 years
40Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4350
Swaddiwudhipong 1998 (Continued)
Notes Study conducted in Tak Province rural Thailand
Waller 1992
Methods Cohort study
Participants Initial sample size
Intervention group 349 riders who succesfully completed the Department of Public Safety motorcycle operator
training in 1988
Control group 349 motorcyclists who never completed the DPS motorcycle operator training
Analysis sample size
Intervention group 349
Control group 349
Interventions No DPS motorcycle operator training vs non-mandatory DPS motorcycle operator training The DPS training is
the Motorcycle Safety Foundationrsquos Motorcycle Rider Course comprising eight hours of theory (lectures discussions
films and multiple choice tests) a textbook and 12 hours off-road practice
Intervention integrity
All 349 riders succesfully completed the rider training
Outcomes Police recorded crashes
Length of follow up 3 years
Notes Study conducted in
Characteristics of excluded studies [ordered by study ID]
Study Reason for exclusion
Anderson 1980a Study type review and process evaluation of a rider course
Aupetit 2008 Outcome proposal of a rider training curriculum that combines traditional training and a motorcycle riding
simulator
Awane 1999 Study type review on use of simulators in rider and driver training
Baldi 2005 Intervention not rider training but intervention integrity of rider training availabilty of rider education
implementation of graduated licensing system etc on a state-wide level
Batchler 1988 Study type description of a training course
Blanchard 2006 Study type not a study assessing the association between rider training and outcomes
Bonnett 2005 Outcome qualitative data regarding satisfaction with the training programme
41Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4450
(Continued)
Braver 2007 Outcome licensure rates
Clayton 1980 Outcome knowledge skills attitude
Council 1978 Outcome crash reduction in ANY vehicle not motorcycles exclusively
Crowther 2005 Study type qualitative study
Daniello 2009 Study type review of the effectiveness of motorcycle training and licensing
Dewanti 2007 Intervention emotional characteristics behaviour (ie not motocycle rider training)
Du Plooy 1990 Study type review of types of rider training courses in South Africa
Engel 1989 Study type uncontrolled before-after study
Goldenbeld 2002 Outcome skills
Goldenbeld 2004 Outcome knowledge and skills
Haworth 2000 Intervention balance between attitudinal andvehicle control skills acrosscurriculla of motorcycle rider training
courses
Haworth 2005 Study type review
Haworth 2006 Study type cross-sectional survey
Hill 1983 Study type review and description of training system
Imatake 1980 Study type course descriptionreview
Ingebritsen 1990 Intervention risk factors for motorcycle crashes
Jain 2009 Study type cross-sectional descriptive study of motorcycle crashes over a 5 year period
Jonah 1979 Study type cross-sectional survey Outcome scores on motorcycle operator skill test
Jonah 1981 Intervention evaluation of a motorcycle rider skill test not motorcycle rider training
Kadar 1990 Study type review of components of rider training courses
Kelsey 1986 Intervention evaluation of a motorcycle rider skill test not motorcycle rider training
Koch 1980 Study type reviewcourse description
Koch 1990 Study type description of course content
42Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4550
(Continued)
Labbett 2005 Study type essay about possible explanations for motorcycle deaths
Law 2005 Intervention targeted motorcycle safety programme not specifically rider training
Lowes 1990 Study type review and descriptions of rider training
Mahon 1976 Study type review of motorcycle rider training methods
Maxwell 1985 Outcomesrecommendations forprogramme operatorsand otherson howthe motorcycle trainingprogramme
works and how it can be improved
Mayhew 1996 Study type review
McKnight 1987 Study type evaluation of methodological and administrative aspects of a motorcycle operator skill test
McPherson 1976 Outcome knowledge and skills
Mohan 2004 Study type review of evidence of interventions to reduce road traffic injuries
Morgan 1980 Study type reviewcourse description
Motoki 1990 Study type proposal and development of rider training curriculum
Nairn 1992 Study type literature review
Newland 1980 Study type description of development of motorcycle safety course
Newland 1982 Study type descriptionreview of a rider training course
Newland 1990 Study type uncontrolled BeforeAfter study
Ormston 2003 Outcome changes in attitude and self reported behaviour
Osga 1979 Outcome knowledge attitudes and skills
Panou 2007 Intervention new simulation tool to teach driverrider
Peterson 1990 Study factor demographic details of course participants completing the course vs those not completing the
course
Prem 1984 Study type experimental
Study factor examines association between skills and performance in tests
Radin Umar 2006 Intervention motorcycle safety programme not specifically rider training
Rockwell 1990 Study type methodologically flawed study type
43Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4650
(Continued)
Rothe 1987 Study type qualitative study of rider training in general
Rowden 2007 Study type qualitative study
RSC Victoria 1993 Study type review of motorcycle rider training and education in Victoria
Russam 1979 Study type review of training programmes in Japan
Saffron 1981 Study type literature review
Schulz 1990 Study factor predictors of motorcycle crashes
Sexton 2004 Study type cross-sectional analytic analysis
Shephard 1986 Study factor reviewdescription of programme
Simpson 1990 Study type review of motorcycle rider trainingeducation
Sokolov 1990 Study factor review of motorcycle riding courses in USSR
Steffens 1988 Outcome self-reported behaviour experiences and researcher observed riding behaviour
Sudlow 2003 Study type review of motorcycle training in the UK
Tasmania 1995 Study type cross-sectional descriptiveprocess evaluation
Tetard 1985 Study type experimental
Outcome detection of indices
Thackray 1980 Outcome instructional effectiveness user acceptance administrative feasibility
Thompson 1994 Study type cross-sectional survey
Outcome attitudes and knowledge
Tobler 1990 Study type review of course curricula
Transport 1989 Intervention syllabus for rider training
Tuan 2007 Study type cross-sectional study (survey) which does not assess the effects of an intervention but examines
motocyclistsrsquo behaviour
Victor 1980 Study type review
Waterer 1990 Study type review of UK rider training issues
Weller 1989 Study type review of studies evaluating motorcycle safety and education measures
44Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4750
(Continued)
Wells 1981 Study type description of different methods of training
Winn 1986 Study type review of rsquohistoryrsquo of rider training
Wisher 1988 Study type cross-sectional survey
Wood 1987 Study type Interrupted Time Series but did not meet the criteria for an ITS as published by the Cochrane
Effective Practice and Organisation of Care Group
Yamazaki 1990 Study type cross-sectional survey
Outcome riding skills skills retention and attitude
Characteristics of ongoing studies [ordered by study ID]
Burgess 2010
Trial name or title UKs National Ride program
Methods Cohort study
Participants Motorcycle offenders
Interventions Police led diversion program
Outcomes Self reported attitudinal and behavioural measures
Starting date
Contact information
Notes Expected to be completed in early 2010
45Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4850
D A T A A N D A N A L Y S E S
This review has no analyses
A P P E N D I C E S
Appendix 1 Search strategy
We searched the following electronic databases
bull Cochrane Injuries Group Specialised Register (searched 22 Sept 2008)
bull CENTRAL (The Cochrane Library 2008 Issue 3)
bull TRANSPORT 1988 to 200706 (latest date available)
bull ATRI Australian Transport Index (to Sept 2008)
bull MEDLINE (Ovid SP) 1950 to Sept 2008
bull EMBASE (Ovid SP) 1980 to Sept 2008
bull CINAHL (EBSCO) 1982 to Sept 2008
bull WHOLIS (World Health Organization Library Information System) (to Sept 2008)
bull PsycINFO (Ovid SP) 1806 to Sept 2008
bull LILACS (BIREME) (Latin American and Caribbean Health Sciences) (1982 to Sept 2008)
bull ISI Web of Science Social Sciences Citation Index (SSCI) 1970 to Sept 2008
bull ERIC (searched 25 Sept 2008)
bull ZETOC (searched 25 Sept 2008)
bull SIGLE (searched 25 Sept 2008)
All of the databases were searched using the following text-word search strategy The strategy was adapted using either rsquorsquo or rsquo$rsquo as atruncation symbol
1 motor-cycl or motorcycl or motor-bik or motorbik or motor-scooter or motorscooter or scooter or moped or motocycl
or motocicl or moto
2 train or educat or instruct or program or course or qualif or learn or improv
3 1 and 2
H I S T O R Y
Protocol first published Issue 2 2005
Review first published Issue 10 2010
46Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4950
C O N T R I B U T I O N S O F A U T H O R S
Katina Kardamanidis was the first author of the protocol conducted the literature search extracted full studies assessed the studies
according to inclusion criteria assessed quality of the studies compared findings with the second author discussed disagreements with
the third author wrote the drafts and the final review
Rebecca Ivers was the second author of the protocol assessed the studies according to inclusion criteria assessed quality of the studies
compared findings with the first author edited the final draft and provided overall guidance for the process
Alexandra Martiniuk assessed quality of the studies compared findings with the first author discussed disagreements with the third
author and helped write the first draft and the final review
Mark Stevenson contributed to the protocol reviewed methods of studies and contributed to the final draft
Katrina Thistlethwaite conducted a second literature search extracted full studies and contributed to the final draft
D E C L A R A T I O N S O F I N T E R E S T
None known
S O U R C E S O F S U P P O R T
Internal sources
bull No sources of support supplied
External sources
bull National Health and Medical Research Council of Australia Australia
Salary support for Martiniuk Ivers and Stevenson
bull University of Sydney Sesqui Research Grant Australia
Direct research costs and salary support for Kardamanidis
D I F F E R E N C E S B E T W E E N P R O T O C O L A N D R E V I E W
There are no differences between the protocol and the review
I N D E X T E R M S
Medical Subject Headings (MeSH)
lowastMotorcycles Accident Prevention [lowastmethods] Accidents Traffic [lowastprevention amp control] Licensure Program Evaluation Wounds
and Injuries [mortality lowastprevention amp control]
47Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 5050
MeSH check words
Humans
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 550
clesaccounted for45 of allAustralian passenger vehicle registra-
tions and 09 of vehicle kilometres travelled in 2007 Howevermotorcycle riders accounted for approximately 15 of all road
crash deaths and an even higher proportion of serious injuries
Per distance travelled the Australian rate of motorcyclist deaths
was approximately 30 times the rate for car occupants The cor-
responding rate for a serious injury was approximately 41 times
higher Similar elevated rates were also found in other developed
countries (Johnston 2008) In low and middle-income countries
the motorcycle is one of the most frequently used means of trans-
port and consequently motorcyclists constitute a large proportion
of road crash victims (Mohan 2002)
The lack of experience of many motorcycle riders has been cited
as one potential cause of motorcycle crashes (Winn 1987 Mullin1997) Training of motorcycle riders has therefore been suggested
as an important countermeasure for reducing both the number of
crashes and the severity of injury (Vis 1995 Noordzij 2001)
Pre-licence motorcycle rider trainingcourses have been made com-
pulsory in several countries (Mullin 1997) Pre-licence training
courses are aimed at providing motorcycle riders with the skills
and knowledge needed to improve control of the motorcycle and
to identify and avoid critical situations in traffic (Elliot 2003)
Many pre-licence motorcycle rider training courses are aimed at
novice riders and in many jurisdictions in high-income countries
such as Australia the USA and Canada the completion of such
courses is a mandatory requirement of licensing
Post-licence rider training is not linked with licensing and such
training is generally not compulsory It usually focuses on riding
skills and other topics that have not been covered or dealt with
sufficiently in pre-licence rider training Post-licence rider train-
ing is used to reinforce fundamental skills and knowledge or to
provide some training for motorcyclists who are self-taught and
have never received formal training (Anderson 1980)
Estimates of the number of motorcyclists attending post-licence
rider training are not readily available but existing data suggest
such courses are not well attended although attendance varies by
country In the USA 24 of the total number of motorcyclists
trained in 1988 and 1989 attended an experienced rider course
(Weaver 1990)
The aim of this systematic review is to quantify the effectiveness
of pre-and post-licence rider training for motorcycle riders in re-
ducing traffic offences road traffic crash involvement injury and
death
O B J E C T I V E S
To quantify the effectiveness of pre- and post-licence motorcycle
rider training on the reduction of traffic offences traffic crash
involvement injuries and deaths of motorcycle riders
M E T H O D S
Criteria for considering studies for this review
Types of studies
All relevant intervention studies such as randomised and non-ran-
domised controlled trials and interrupted time series studies and
observational studies including cohort and case-control studies
Types of participants
Motorcyclists and riders of mopeds or motor scooters of all ages
Types of interventions
Motorcycle ridertraining (pre-or post-licence mandatory or non-
mandatory) versus no training
One form of pre-licence rider training versus another form (eg
on-road off-road theory or a combination)
One form of post-licence rider training versus another form (eg
on-road off-road theory or a combination)
Types of outcome measures
Offences crashes injuries and death
Search methods for identification of studies
We did not restrict the searches by date language or publication
status
Electronic searches
We searched the following electronic databases
bull Cochrane Injuries Group Specialised Register (searched 22
Sept 2008)
bull CENTRAL (The Cochrane Library 2008 Issue 3)bull TRANSPORT 1988 to 200706 (latest date available)
bull ATRI Australian Transport Index (to Sept 2008)
bull MEDLINE (Ovid SP) 1950 to Sept 2008
bull EMBASE (Ovid SP) 1980 to Sept 2008
bull CINAHL (EBSCO) 1982 to Sept 2008
bull WHOLIS (World Health Organization Library
Information System) (to Sept 2008)
bull PsycINFO (Ovid SP) 1806 to Sept 2008
bull LILACS (BIREME) (Latin American and Caribbean
Health Sciences) (1982 to Sept 2008)
bull ISI Web of Science Social Sciences Citation Index (SSCI)
1970 to Sept 2008
3Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 650
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 750
Table 1 Risk of selection bias (Continued)
Osga 1980 Differences at baseline reported on sex and age
Confounding factors were not dealt with in univariate analysis Riding exposure was accounted for in the
cumulative survival analysis
Selection intervention group were all training participants on the register and the control group were
respondents to a diverse range of recruitment strategies
Satten 1980 Differences at baseline reported on age sex helmet use riding experience and exposure and previous
training
Confounding factors were not dealt with in design or analysis
Selection intervention group was selected from a list of training graduates and the control group was
partly selected from a state register of motorcyclists
Jonah 1982 Differences at baseline reported on age sex riding experience time licensed previous training and alcohol
use
Confounding factors were dealt with through multivariate analysis
Selection intervention group was selected from a list of training graduates and the control group was
randomly selected from the Ontario motorcycle operator file
Mortimer 1984 Differences reported at baseline on age licence status riding experience years licensed and helmet use
Confounding factors age years licensed and rider experience were dealt with through stratification and
riding exposure through standardisation
Selection intervention group was selected from a list of training enrolees and the control group was
selected from motorcycle dealerships during three successive Saturdays
Cooper 1988 Differences reported at baseline on age gender licensure and highway experience
Confounding factors age gender riding exposure rider characteristics rider attitudes and behaviour were
accounted for in a multivariate analysis
Selection there was no information on how the intervention and control groups were selected and from
where
Mortimer 1988 Differences reported at baseline on years licensed riding experience and helmet use
Confounding factors age years licensed and rider experience were dealt with through stratification and
standardisation of riding exposure A regression analysis was also conducted but poorly reported and not
usable
Selection intervention group was selected from a list of training enrolees and the control group was
selected from motorcycle dealerships
Adams 1985 Differences at baseline reported on previous crash rates
Confounding factors not controlled for in analysis
Selection the intervention group was a non-random sample of those offered training within the Australia
Post motorcycle delivery riders the control group were the remaining Post riders
Leung 1987 Differences at baseline reported on age and sex
Confounding factors age and sex controlled for in analysis by stratification
Selection those exposed to training were MRC graduates those unexposed were taken from the licensing
records
5Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 850
Table 1 Risk of selection bias (Continued)
McDavid 1989 Differences at baseline other than matched characteristics were not reported
Confounding factors age ownership of car licence years licensed previous crashesand offences accounted
for through matching Other confounding effects not controlled
Selection intervention group was selected from a list of training graduates and the control group was
selected from Bureau of Motor Vehicle records
Kloeden 1994 Differences at baseline reported were in geographical area of residence and in riding exposure
Confounding factors were accounted for through exclusion of participants from rural areas
Other confounding factors were not reported
Selection the Ridersafe course was phased in based on postcode of riderrsquos home address analyses were
based on a database of those who obtained motorcycle rider learnerrsquos permits of which about 2000 were
Ridersafe riders and about 5000 were control riders
Billheimer 1998 Differences at baseline reported on age and previous riding experience
Other potential confounding factors assessed were riding exposure primary purpose of riding and helmet
use
Confounding factors accounted for by matching (age sex primary purpose of riding) standardisation
(riding exposure) and stratification (previous riding experience)
Selection the intervention group were those who had taken the rider training programme the control
group were California motorcycle riders matched on the above variables but who had not taken a riding
course
Perrino 2002 Differences at baseline reported were a higher number of females in the trained group and a higher
proportion of experienced riders in the untrained group
Confounding factors were not accounted for in the analysis
Selection the intervention group were those who had done the Motorcycle Riding Course the control
group had not done the course Both groups were identified through a register of licensed motorcyclists
Kraus 1975 Population based case-control study
Cases all motorcycle riders with a confirmed medically treated motorcycle injury (including fatalities)
from all hospitals in study base and peripheral area hospitals and all deaths due to motorcycle crashes in
the study base All serious injuries (leading to hospitalisation) and fatalities likely to be included
Community controls a random sample of registered motorcycle owners in the study base
Potential confounders assessed were age sex weight height riding experience motorcycle use charac-
teristics of motorcycle and previous crashes and offences
Confounding factors were accounted for through exclusion of females and stratification by age and
motorcycle useSelection cases and controls drawn from the same population
Hurt 1981 Population based case-control study
Cases and controls not drawn from the same population
Cases motorcycle riders who had a motorcycle crash Not all reported cases included
Community controls motorcycle riders who passed the crash site at the same time of day and day of week
that the crash occurred but 2 years later
Potential confounders assessed were age sex time type of motorcycle roadway alignment helmet use
weather conditions road surface conditions
Confounding factors were accounted for through matching for time of day and day of week that crash
occurred and stratification
Selection cases were recruited from police reports not all cases were included controls were matched
6Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 950
Table 1 Risk of selection bias (Continued)
community controls from randomly selected sites matched for same time of dayday of week as the crash
occurred
Haworth 1997 Population based case-control study
Cases and controls drawn from the same population
Cases motorcycle riders or pillion riders with an injury from a motorcycle crash from hospital and police
records Not clear whether all cases were included
Community controls motorcycle rider or pillion who passed the crash site at the same time of day and
day of week that the crash occurred
Potential confounders assessed were rider factors age marital status licence status riding experience
and exposure alcohol and drugs use of protective gear including helmet use pillion factors motorcycle
factors trip factorsConfounding factors were accounted for through matching for time of day and day of week that crash
occurred and stratification for age blood alcohol level and licence status
Selectioncases were riders or pillions involved in on-road crash recruited from hospitals and police records
not sure if all cases were included controls were matched community controls who passed the crash site
at the same time of dayday of week as the crash
Savolainen 2007 Differences at baseline reported on sex safety helmet use riding ability riding experience and licence
status
Adjusted for these confounders through statistical modelling although exact process unclear
Selection intervention group from list of trainees control group from Indiana Bureau of Motor Vehicles
database Trainees and the control group were invited to participate in surveys
Davis 1997 Differences at baseline not reported but likely No adjustment for any confounders at baseline or during analysis
Selection intervention group from list of riders involved in crashes who were also on a list of training
graduates control group from the same list of riders involved in crashes who were not also on a list of
training graduates
Waller 1992 Differences at baseline not reported but likely
Adjusted for confounders through exclusion of riders younger than 18 years riders without a motorcycle
operator driver license riders with location of residence outside of the Texas metropolitan areas of Austin
through matching by age sex and location of residence through stratification by age and sex and through
standardisation (number of crashes by person-days at risk)
Selection intervention group from the Motorcycle Operator Training class roster (number of trained
riders that the sample was selected from is unclear) control group through a random number generator
generating driver license numbers until an equal sample size was selected as the intervention group
MAIDS 2004 Population based case-control study
Unclear whether cases and controls not drawn from the same population
Cases motorcycle riders who had a motorcycle crash Unclear whether all cases included
Community controls motorcycle riders recruited at petrol stations Unclear whether and how matched
to cases
No potential confounders were reported
There was no report of whether and how potentially confounding factors were accounted for in the study
7Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 1050
Table 2 Risk of allocation bias
Anderson 1980 Allocation concealment unclear
Hall 1988 Allocation concealment unclear
Swaddiwudhipong 1998 Allocation concealment unclear
Table 3 Risk of detection bias
Anderson 1980 Blinding of outcome assessors not reported
Driver records holding police recorded crashes and offences not validated
Hall 1988 Blinding of outcome assessors not reported
Driver records holding police recorded crashes and offences not validated
Swaddiwudhipong 1998 Blinding of outcome assessors not reported
Police and hospital records Post mortem examination for deaths not validated
Adams 1985 Blinding of outcome assessors not reported
Australia Post crash records not validated but argued that underreporting is unlikely
Leung 1987 Blinding of outcome assessors not reported
Driver records holding police recorded crashes and offences
Raymond 1979 Blinding of outcome assessors not reported
Self-report through personal interviews or questionnaires - both methods validated
Osga 1980 Blinding of outcome assessors not reported
Self-report through structured questionnaires validated by cross-checking with police
recorded crash records
Satten 1980 Blinding of outcome assessors not reported
Self-report through structured interviews or questionnaires - both methods were not
validated
Intervention group were given telephone interviews but the control group mailed ques-
tionnaires
Jonah 1982 Blinding of outcome assessors not reported
Self-report through structured interviews not validated Driver records holding police
recorded crashes and offences
Mortimer 1984 Blinding of outcome assessors not reported
Self-report through questionnaires not validated
Intervention group received mailed questionnaires but the control group received ques-
tionnaires personally at the motorcycle dealership
Cooper 1988 Blinding of outcome assessors not reported
Driver records holding police recorded crashes and offences
8Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 1150
Table 3 Risk of detection bias (Continued)
Mortimer 1988 Blinding of outcome assessors not reported
Self-report through questionnaires not validated
Intervention group received mailed questionnaires but the control group received ques-
tionnaires personally at the motorcycle dealership
McDavid 1989 Blinding of outcome assessors not reported
Driver records holding police recorded crashes and offences
Kloeden 1994 Blinding of outcome assessors not reported
Driver records holding police recorded crashes and offences
Billheimer 1998 Blinding of outcome assessors not reported
Police recorded crashes and offences
Perrino 2002 Blinding of outcome assessors not reported
Self-report through mailed out questionnaires police records not validated
Kraus 1975 Blinding of intervention assessors to case or control status not reported
Intervention measured police records self-report though questionnaires not validated
Length of follow up not specified
Hurt 1981 Blinding of intervention assessors to case or control status not reported
Intervention measured face-to-face in-depth interviews (ie no structured question-
naires) Validity and reliability of data ensured by reported high level of quality controlLength of follow up not specified
Haworth 1997 Blinding of intervention assessors to case or control status not reported
Intervention measured self-report either by personal or by telephone interview for
both cases and controls not validated
Length of follow up in the 2 years prior to the interview or more than 2 years ago
Savolainen 2007 Blinding of outcome assessors not reported
Source of outcome data not reported
Length of follow up up to 5 years
Davis 1997 Blinding of outcome assessors not reported
Source of outcome data police records
Length of follow up 1 year
Waller 1992 Blinding of outcome assessors not reported
Source of outcome data Department of Public Safety driver history file and official
accident reports
Length of follow up 3 years
MAIDS 2004 Blinding of intervention assessors to case or control status not reported
Intervention measured face-to-face in-depth interviews (ie no structured question-
naires) No report on whether and how validity and reliability of data was ensured
Length of follow up 2 years
9Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 1250
Table 4 Risk of attrition bias
Anderson 1980 Percentage of participants in analysis 99 forthe total group Unclear forintervention andcontrol groups
separately
Hall 1988 Percentage of participants in analysis 96-97 forall intervention groups and 98for the control group
Swaddiwudhipong 1998 Percentage of participating sub-districts in analysis 100 for both intervention and control sub-districts
Raymond 1979 Percentage of participants in analysis 74 for the intervention group and less than 57 for the control
group
Osga 1980 Percentage of participants in analysis 73 for the intervention group and 100 for the control group
Satten 1980 Percentage of participants in analysis is unknown Initial sample size is unclear
Jonah 1982 Percentage of participants in analysis 23 for the intervention group and 45 for the control group
Mortimer 1984 Percentage of participants in analysis 35 for the intervention group and 84 for the control group
Cooper 1988 Percentage of participants in analysis 98 for the total group Unclear for intervention and control
groups separately
Mortimer 1988 Percentage of participants in analysis 40 for the intervention group and 90 in the control group
Adams 1985 Percentage of participants in analysis 61 forthe total group Unclear forintervention andcontrol groups
separately
Leung 1987 Percentage of participants in analysis unclear because initial sample size was not reported
McDavid 1989 Percentage of participants in analysis 40 of final matched cohort
Kloeden 1994 Percentage of participants in analysis 39 for the intervention group and 6 for the control group
Billheimer 1998 Percentage of participants in analysis unclear percentage of the initial sample size
Perrino 2002 Percentage of participants in analysis unclear percentage of the initial sample size
Kraus 1975 Percentage of participants in analysis 58 of cases 59 of controls
Hurt 1981 Percentage of participants in analysis 18 of cases lt 29 of controls
Haworth 1997 Percentage of participants in analysis 70 of cases 47 of controls
Savolainen 2007 Percentage of participants in analysis 18 for the intervention group and 15 for the control group
Davis 1997 Percentage of participants in analysis 100 for the intervention group and unclear for the control group
10Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 1350
Table 4 Risk of attrition bias (Continued)
Waller 1992 Percentage of participants in analysis unclear percentage for the intervention group and 100 for the
control group
MAIDS 2004 Percentrage of participants in analysis unclear for both cases and controls
R E S U L T S
Description of studies
See Characteristicsof included studies Characteristics of excluded
studies Characteristics of ongoing studies
Our search resulted in 102 studies that seemed relevant of which
we excluded 79 studies because they did not meet the criteria for
considering studies for this review (see Characteristics of excluded
studies)
Basedon ourcriteriawe includeda total of 23 studies in this review
There were three randomised trials (Anderson 1980 Hall 1988
Swaddiwudhipong 1998) two non-randomised trials (Adams
1985 Leung 1987) 14 cohort studies (Raymond 1979 Osga
1980 Satten 1980 Jonah 1982 Mortimer 1984 Cooper 1988
Mortimer 1988 McDavid 1989 Waller 1992 Kloeden 1994
Davis 1997 Billheimer 1998 Perrino 2002 Savolainen 2007)
and four case-control studies (Kraus 1975 Hurt 1981 Haworth
1997 MAIDS 2004)
Over half of the studies were conducted 20 or more years
ago and the oldest was conducted 31 years ago Only three
studies were from this decade (Perrino 2002 MAIDS 2004
Savolainen 2007) Over 60 of the studies were conducted in
the USA (Kraus 1975 Anderson 1980 Osga 1980 Satten 1980
Hurt 1981 Mortimer 1984 Leung 1987 Hall 1988 Mortimer
1988 Waller 1992 Davis 1997 Billheimer 1998 Perrino 2002
Savolainen2007) three inAustralia (Adams 1985 Haworth 1997
Kloeden 1994) three in Canada (Jonah 1982 Cooper 1988
McDavid 1989) onefromthe United Kingdom (Raymond 1979)one from Europe (with sampling areas located in France Ger-
many Netherlands Spain and Italy) (MAIDS 2004) and one
in Thailand (Swaddiwudhipong 1998) Only seven studies were
published in peer reviewed journals (Kraus 1975 Jonah 1982
Mortimer 1984 Cooper 1988 Mortimer 1988 McDavid 1989
Swaddiwudhipong 1998) The other studies were reports forhigh-
way safety and road traffic organisations
The studies by Anderson 1980 Hall 1988 Kloeden 1994
Billheimer 1998 (the basic California Motorcyclist Safety Program
component of the study) and Perrino 2002 examined manda-
tory pre-licence training Cooper 1988 McDavid 1989 and
Swaddiwudhipong 1998 assessed non-mandatory training that
when passed resulted in a licence Raymond 1979 Osga 1980
Satten 1980 Jonah 1982 Mortimer 1984 Adams 1985 Leung
1987 Mortimer 1988 Waller 1992 Davis 1997 Billheimer 1998
(the Experienced Rider Course component of the study) andSavolainen 2007 assessed non-mandatory training that could be
taken by motorcycle riders with and without a licence The case-
control studies (Kraus 1975 Hurt 1981 Haworth 1997) assessed
lsquoanyrsquo type of rider training and therefore could not be grouped by
pre- or post-licence rider training The MAIDS 2004 case-control
study did make a distinction between participants having received
compulsory pre-licence training or additional (non-compulsory)
training We did not find a study that compared one form of pre-
or post-licence training (eg on-road off-road theory or a com-
bination) versus another form of training All studies compared
eithermandatory or non-mandatory formal training versus no for-
mal training with the exception of Haworth 1997 who compared
different levels of training courses (beginner intermediate and ad-vanced)
We have presented a more detailed description of the individual
studies in Characteristics of included studies We have provided
a description below of the types of motorcycle rider training that
were evaluated in the included studies
Types of motorcycle rider training evaluated in the
studies
The types of training were varied Duration of training ranged
from two or three hours to 24 hours administered over two or
three days Most training consisted of a theoretical and a practical
component usually off-road The content ranged from provid-ing riders with manuals lectures discussions showing films and
administering knowledge and skill tests Little information was
provided about the qualifications of trainers who conducted the
training
Six studies (Osga 1980 Satten 1980 Mortimer 1984 Mortimer
1988 Waller 1992 Savolainen 2007) assessed the effectiveness of
the USA-based Motorcycle Safety Foundationrsquos Motorcycle Rider
Course (MSF MRC) This course was released in 1976 It com-
prises eight hours of theory (lectures discussions films and multi-
plechoice tests) a textbook and 12 hours of off-road practice The
Motorcycle Safety Foundation is a national not-for-profit organ-
11Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 1450
isation sponsored by manufacturers and distributors of motorcy-
cles Many USA states have adopted the MSF Basic Rider Courseand Experienced Rider Course (wwwmsf-usaorg) since its release
in 1976
The Motorcycle Operators Skill Test was also developed
by the USA Motorcycle Safety Foundation in coopera-
tion with others (httpwwwwebbikeworldcommotorcycle-
trainingmotorcycle-test) The MOST is a licensing test consist-
ing of a manual knowledge and skill test (comprising eight exer-
cises of increasing difficulty and skill requirements) and a mini-
skill test conducted with a simulator Some studies compared this
alternative licensing test to the standard licensing test However
as the MOST is not a type of rider training we have included only
the two studies that assessed the MOST in combination with one
or more training components in this review Both Anderson 1980and Hall 1988 assessed the MOST in combination with a train-
ing component of three hoursrsquo duration consisting of instructions
about the skills needed to pass the MOST (30 minutes of the-
ory and two and a half hours on the motorcycle) Hall 1988 also
assessed the MOST in combination with the Motorcycle Safety
Foundationrsquos Motorcycle Rider Course as described above
Other USA studies evaluated the Wisconsin motorcycle rider
course a 22-hour off-street programme that teachesthe basic rules
of safe motorcycling (Leung 1987) the USA Maryland Motorcy-
cle Rider Course (MRC) a mandatory (for those younger than 18
years)pre-licence 20-hour coursecomprising eight hours of theory
and12 hours of practice and created by the Maryland Motor Vehi-
cle Administration in 1985 (Perrino 2002) the California Motor-cyclist Safety Program (CMSP) a mandatory (for those younger
than 18 years) pre-licence 16-hour basic course (Billheimer 1998)
and Experienced Rider Course (ERC) of eight hoursrsquo duration
(Billheimer 1998) The CMSP still offers these courses ( http
wwwca-msporg) but they are now MSF courses as described
above Davis 1997 did not give any information about the assessed
rider training
The Canadian studies evaluated the British Colombia Safety
Councilrsquos 37-hour motorcycle safety programme (no further in-
formation on the content was provided) (McDavid 1989) and
the Canada Safety Council Motorcycle Training Program (MTP)
comprising four hours of theory and 16 hours of practice (six
hours on-road) over twoweekends (Jonah 1982) The MTPcoursewas made available in 1974 and is still being offered in Canada
(httpwwwsafety-councilorgtrainingmtpmtphtm) Cooper
1988 did not give any information about the assessed rider train-
ing However the study was conducted in British Colombia in
Canada in 1988 the same time as the McDavid 1989 study was
conducted
In the United Kingdom Raymond 1979 assessed the non-manda-
tory RACACU (Royal Automobile ClubAuto-Cycle Union)
training scheme The training programme consisted of either 24
one-hour sessions or 13 two-hour sessions Twelve hours of the
training focused on theory about basic operational procedures
traffic laws and highway codes The practical session focused on
basic operational skills riding on a training area and riding in traf-fic with an instructor A proficiency test was administered at the
end of the course
One study from rural Thailand (Swaddiwudhipong 1998) evalu-
ated a community-based rider education programme consisting of
theory on motorcycle injury risk helmet use traffic laws vehicle
regulations and traffic signs and a written and skill test leading to
a licence
TheAustralian studies evaluateda rider trainingcourse of 12 hours
over two consecutive days (four hours of theory and eight hours of
practice)administeredto AustraliaPosttelegram deliverers(Adams
1985) and a mandatory pre-licence Ridersafe course consisting
of two tiers the first tier being a training course on the basics
of motorcycle riding where a pass entitled the rider to obtain alearnerrsquos permit and the second tier being a skill test where a pass
entitled the riderto obtain a probationary motorcycle licence The
training course consisted of two four-hour training sessions on
separate days involving instruction videos and actual riding on
motorcycles The skill test consisted of a four-hour preparation
course of 74 minutes of lecturediscussion 38 minutes of video
and 75 minutes of on-course riding which was the testing phase
(Kloeden 1994) The Rider Safe course is still being provided by
Transport South Australia a state government organisation (http
wwwtransportsagovau )
The duration or content of rider training in the four case-control
studies (Kraus 1975 Hurt 1981 Haworth 1997 MAIDS 2004)
was not described
Risk of bias in included studies
We describe the methodological quality of the included studies
by study type below More detailed information can be found in
Table 1 Table 2 Table 3 and Table 4
When considering the findings from these studies it is important
to note the quality (or lack thereof) of reporting of the studies
Quality of reporting is the extent to which a report provides in-
formation about the design conduct and analysis of a study ( Juumlni2001) Huwiler-Muumlntener 2002 found that even though trials
may be well reported it does not mean that the methodological
quality is of equal standard and well-conducted trials may never-
theless be reported badly The reporting quality of most studies in
this review was poor Most studies were old and did not meet con-
temporary reporting standards This made data extraction com-
plicated It also made it difficult to contact authors for additional
information
Because of significant heterogeneity between studies as well as the
poor quality of most studiespooling of studyresults to calculatean
overall measure of effect was not appropriate Instead we provided
a description of our judgement of risk of bias in the text
12Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 1550
Risk of bias in the randomised trials
There were three randomised trials (Anderson 1980 Hall 1988Swaddiwudhipong 1998) Anderson 1980 and Hall 1988 had an
initial sample size of 41188 and 26464 motorcycle licence ap-
plicants respectively Both studies assessed the effectiveness of the
pre-licence MOST with a trainingcomponent Swaddiwudhipong
1998 evaluated the effectiveness of a community-based rider ed-
ucation programme in six districts
Allocation bias
Anderson 1980 reported using random birth date tables to allocate
participants to either the intervention or control group Hall 1988
reported the use of a computerised algorithm set in a central of-
fice but also used manual random assignment when the computer
went down Swaddiwudhipong 1998 reported lsquosimple random
samplingrsquo for the delivery of the intervention where the sampling
frame included nine districts Three districts were excluded since
they were too hard to access The remaining six districts were ran-
domly allocated to the intervention (three) and the control (three)
groups Systematic sampling wasthen used ordered by village size
for the villages within districts Allocation concealment was not
reported in any of the studies The descriptions of randomisation
in these studies suggest that there may have been limitations to
the process resulting in a lack of equivalence of the intervention
and control groups at baseline and therefore the possibility of
allocation bias Swaddiwudhipong 1998 reported that before theintervention the annual incidence rate of motorcycle-related in-
juries was slightly (but not statistically significant) higher in the
intervention areas than in the control areas Participants in the
intervention districts were also more likely to have a valid licence
and to wear a helmet more often than participants in the control
districts Anderson 1980 measured differences at baseline on the
participant and prior driving record variables and found these to
be negligible Hall 1988 did not report on baseline differences
between the intervention and control groups As the Anderson
1980 and Hall 1988 studies had sample sizes of over 40000 and
26000 respectively and used some form of randomisation it is
likely that any confounding factors at baseline would have been
equally distributed between comparison and intervention groups
Selection bias
After randomisation Anderson 1980 and Hall 1988 excluded
08 and 3 participants respectively because of assignment to
the wrong group administration of the wrong intervention or
other reasons The resulting lack of analysis by rsquointention-to-treatrsquo
can introduce biasby removing the effectivenessof randomisation
Nonetheless the percentage excluded was small and the number
of overall participants large so it is unlikely to have had a major
effect on the final results In the Swaddiwudhipong 1998 study
villages were systematically sampled into intervention or control
groups The intervention was implemented and outcomes of in-
jury and death were measured at a community level Interventionand control districts were randomly selected however there were
only three intervention and three control communities
Detection bias
None of the randomised trials reported blinding of outcome as-
sessors Data collection tools used by Anderson 1980 and Hall
1988 were driver records holding police recorded crashes and of-
fences Swaddiwudhipong 1998 used police and hospital records
at a district level and post-mortem examination for deaths None
reported on validity or reliability for any of the instruments
Attrition bias
Anderson 1980 and Hall 1988 had driver records available for
96 to 100 of participants They also collected self-reported
data but the response rate to the questionnaires was very low (low-
est response rate 42 and 30 respectively) Attrition bias in
Swaddiwudhipong 1998 was not an issue as outcome data were
available for all sub-districts included in the trial
The intervention reported in Anderson 1980 and Hall 1988 tri-
als had an unanticipated effect on the participants In both trials
an average of only 28 of licence applicants in the intervention
group eventually obtained their licence compared to 35 in the
control group This indicates that the licensing process in generalseems to have deterred applicants from obtaining a licence and
although no statistical tests of the difference were conducted this
deterrence appears even greater for those who had rider training
This is not true attrition bias as the studies did not experience
differential loss to follow up from outcome measurement Instead
differential attainment of the intermediary step of licensure pre-
cluded participants from being at risk for the outcomes of interest
they are not driving a motorcycle hence they are not at risk for
motorcycle-related crashes injury death or offences In short it is
not possible to assess whether a reduction in crashes in the inter-
vention group would be due to changes in riding habits as a result
of training or simply due to delayed or never attained licensure
Intervention integrity
Neither Anderson 1980 nor Hall 1988 reported on the consis-
tency in the way the intervention was administered the possible
lsquocontaminationrsquo of participants in the intervention group by other
types of formal or informal rider training or of participants in the
control group by the intervention Swaddiwudhipong 1998 did
not report on the number of villagers in each group that actually
attended the intervention However they did report on a number
of other potential training courses that were being implemented
at the same time
13Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 1650
Analysis
Anderson 1980 and Hall 1988 acknowledging the lsquodeterrent ef-fectrsquo described above conducted two separate analyses The first
analysis included the total study population of licence applicants
over two years from the time of their application (the population
denominator) The second analysis was performed on the licensed
(intervention and control) subgroups only two years from the
time of obtaining their licence (the licensed rider denominator)
As only a third were licensed and exposed to riding the motorcy-
cle it was impossible to determine whether reported differences
in crash rates across groups were due to the intervention or to
other lsquoconfoundingrsquo characteristics of those who were determined
in their efforts to obtain the licence When analysis was restricted
only to those who gained their licence the same difficulties re-
mained The benefits of the randomisation process were thus lost
due to this lack of licensure of riders during the process As there
were only three intervention and three control groups in the study
reported by Swaddiwudhipong 1998 analyses were hindered by
smallnumbers for outcome comparisons betweenthe intervention
and the control districts
Risk of bias in the cohort studies and non-randomised
trials
We discuss the methodological quality of cohort studies (seven
prospective studies (Raymond 1979 McDavid 1989 Waller 1992
Kloeden 1994 Billheimer 1998 Perrino 2002 Savolainen 2007)and seven retrospective studies (Osga 1980 Satten 1980 Jonah
1982 Mortimer 1984 Cooper 1988 Mortimer 1988 Davis
1997)) and non-randomised trials (Adams 1985 Leung 1987) in
the one section as the biases for the two study types are similar
The quality of these studies was in general poor
Selection bias
The confounding factors expected to have been addressed in the
studies were sex age riding experience prior driving records (of-
fences crashes) previous training and the following proxies for
risk taking behaviour helmet use speed and alcoholdrug use All
of these factors have been shown to be associated with an increasedrisk of motorcycle crash (Peden 2004 Zambon 2006) The pos-
sible source of bias in cohort studies and non-randomised trials
is that groups may differ on important characteristics at baseline
participants lsquoself-selectrsquo into either the rider training group by de-
ciding to undertake a rider training course or into the lsquocontrol
grouprsquo by deciding not to undertake a course Participants who
undertake a non-mandatory rider training course may differ from
those who choose not to on characteristics such as age sex rid-
ing experience and riding behaviour as well as other possible con-
founders For example riders who take a training course may be
more safety-conscious or on the contrary may do so because they
have previously experienced a crash or because they perceive that
they have poor riding skills Therefore ideally all participants in
the intervention group need to be included in the study as wellas a random sample of controls taken from the same population
as the intervention group If that is not possible intervention and
control groups either need to be carefully matched to make them
comparable or differences at baseline need to be adjusted for in
the analysis
Most of the studies using a non-randomised trial or cohort design
compared an intervention group to a control group that differed
on a number of important factors The studies in general failed to
adjust for important differences found at baseline and their results
cannot be meaningfully interpreted Ten studies (Raymond 1979
Osga 1980 Satten 1980 Jonah 1982 Mortimer 1984 Mortimer
1988 McDavid 1989 Waller 1992 Davis 1997 Savolainen 2007)
assessed intervention and control groups that were not drawnfrom the same population In most of these cases the intervention
group was selected from a list of training graduates who were usu-
ally young female and less experienced riders whereas controls
were recruited from a general population of motorcycle riders who
were usually older male and experienced riders For three stud-
ies it was unclear how they were selected (Leung 1987 Cooper
1988 Perrino 2002) Only Kloeden 1994 recruited groups from
the same population of motorcycle applicants and Adams 1985
from a group of professional riders Nevertheless due to non-ran-
dom selection of intervention and control groups differences at
baseline are likely even though study groups were from the same
population Three studies adjusted for some confounding fac-
tors found at baseline through matching (McDavid 1989 Waller1992 Billheimer 1998) However matching has limitations in
termsof controlling for multiple confounders especially those that
are unknown to influence the effect of rider training on the out-
comes One study matched the control and intervention groups
by excluding controls (approximately 3500 (66 of initial con-
trol group) that differed from the intervention group) (Kloeden
1994) Five studies (Osga 1980 Satten 1980 Adams 1985 Davis
1997 Perrino 2002) did not adjust for differences between study
groups at baseline either in the study design or analysis Results
from these studies are therefore not reliable The other six stud-
ies (Raymond 1979 Jonah 1982 Mortimer 1984 Cooper 1988
Mortimer 1988 Leung 1987 McDavid 1989 Billheimer 1998)
adjusted for some differences between groups at baseline by eitherbi-variate or multi-variate analyses or stratification and standardi-
sation of riding exposure (outcomemiles driven) or stratification
by age group The limitation of stratification is that it is difficult to
control for many variables simultaneously because a large number
of strata will need to be generated resulting in small numbers that
cannot be meaningfully analysed Savolainen 2007 developed sta-
tistical models to assess the probability of a crash after taking the
rider training course in an intervention and control group while
examining differences by sub-groups (eg age) Participation rates
were also very low in this study with only 5584000 (1395)
in the eligible intervention group and 5884000 (147) for the
14Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 1750
control group completing surveys
Detection bias
None of the studies reported blinding of outcome assessors to the
intervention or control status of the participants Seven studies
used driver records containing police recorded data as their main
measure of outcome (Leung 1987 Cooper 1988 McDavid 1989
Waller 1992 Kloeden 1994 Davis 1997 Billheimer 1998) An-
other four studies (Raymond 1979 Satten 1980 Mortimer 1984
Mortimer 1988) only used self-report Raymond 1979 was the
only study using self-report as the main outcome measurement in
which study instruments were validated Self-report might prevent
the underreporting that has been described for the use of police-
driver records (see rsquoRisk of biasrsquo in the randomised trials above)but has other risks of bias especially if questionnaires or inter-
view methods have not been validated and tested for reliability
Recall bias arises from mistakes in recollecting events both be-
cause of failures of memory and because of looking at things lsquowith
hindsightrsquo and possibly changed views (httpwwwcochraneorg
resourcesglossaryhtm) As many studies were retrospective risk
of recall bias is likely and may have increased with length of follow
up For studies that used self-reported outcomes only it is also
possible that social desirability response bias played a role where
participants in the training group may have felt it was lsquoexpectedrsquo
to report good driving histories Multiple methods to measure the
same outcome for both groups decreases the risk of detection bias
for example because self-reported answers can be cross checkedwith other sources of information such as driver records
In this regard three studies (Osga 1980 Jonah 1982 Perrino
2002) used both self-report and driver records Osga 1980 found
that for both intervention and control groups more crashes were
reported through self-report than on state crash records an aver-
age of 54 of self-reported crashes could be found back on state
crash records One industry-based study (Adams 1985) used oc-
cupational health records which were not validated but which are
unlikely to suffer from underreporting due to work cover claims
Savolainen 2007 used information on rider crashes for a five-year
period but did not specify from where this information was ob-
tained
Attrition bias
The majority of studies (Raymond 1979 Satten 1980 Adams
1985 Leung 1987 Kloeden 1994 Davis 1997 Billheimer 1998
Perrino 2002) inadequately reported on losses of participants to
the study We can therefore not be certain that there were no sys-
tematic differences in the characteristics between the intervention
and control group participants Savolainen 2007 reported the re-
sponse rates to their baseline survey but did not report on the per-
centage of outcome data that was available for their study sample
A similar deterrent effect of rider training as described for the ran-
domised trials (Anderson 1980 Hall 1988) occurred in two stud-
ies Jonah 1982 reported that 65(14992310) of the trained rid-
ers stopped riding compared to 47 (9492029) of the untrainedriders They excluded non-active and non-licensed riders before
the study commenced so the differential choice to continue riding
after training did not affect their analyses Leung 1987 reported
that the percentage of participants in the intervention group ob-
taining their licence was 50 (no number provided by authors)
lower than in the control group This may have biased the mea-
surement of outcomes because unlicensed participants will not be
exposed to riding the motorcycle and therefore not be at risk of
being involved in a crash or violation As explained above this
makes it difficult to assess whether the reduction in crashes in the
intervention group is due to a change in driving habits or simply
due to delayed licensure
Mortimer 1984 and Mortimer 1988 describe that about 44 and30 of course graduates respectively were excluded from analysis
because they had not ridden a motorcycle in the previous year
However because these studies were retrospective and the controls
were activeriders at the time of recruitment a decision on whether
the deterrent effect was differential between the two groups could
not be made Though not reported quantitatively Kloeden 1994
also describes the effectof rider training deterringindividuals from
becoming licensed and hence exposure to riding and risk of crash
death or offences involving a motorcycle Osga 1980 reported
that 27 of the trained riders did not ride after the course and
excluded these from the analysis
Use of different data collection strategies may have contributed to
differential loss to follow up in the cohort studies Satten 1980for example interviewed the intervention group over the tele-
phone but mailed questionnaires to the control group Mortimer
1984 and Mortimer 1988 distributed questionnaires in person to
the control group but mailed them to the intervention group
This may have played a role in the observed differential response
rates where more subjects responded in the personally approached
groups
Survival bias
For those studies that assessed offences crashes and injuries but
not death survival bias may have also played a role Motorcyclists
who were fatally or seriously injured in crashes during the follow-
up period after the intervention would have been less likely or
unable to respond to surveys This could bias the effect measures
towards or away from the null None of the studies wrote about
this issue or examined the characteristics of those lost to follow up
in order to address the possibility of survival bias
Intervention integrity
Although the interventions were intended for the entire lsquoexposedrsquo
groups most studies reported difficulty in tracking participants
whoreceivedthe interventionand those whodid not Adams 1985
15Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 1850
estimatedthat 62of thetotalstudysample received the interven-
tion The remaining studies did not report the percentage of thestudy population that received the intervention as was intended
or reported that this percentage was unknown (Raymond 1979
Kloeden 1994 Satten 1980 Jonah 1982 Mortimer 1984 Leung
1987 Cooper 1988 Mortimer 1988 Billheimer 1998 Perrino
2002 Savolainen 2007) Only Raymond 1979 reported the degree
to which their intervention was delivered as intended No studies
reported on possible lsquocontaminationrsquo of participants by other inter-
ventions than the one to which they were allocated Because Osga
1980 Satten 1980 Jonah 1982 Mortimer 1984 Cooper 1988
Mortimer 1988 Waller 1992 and Davis 1997 were retrospective
studies and participants were recruited from a list of graduates it
was known that they had completed and passed the course
Analysis
In some studies some participants who reported at baseline that
they were untrained decided to enroll in a rider training pro-
gramme after the baseline survey The non-randomised trial by
Adams 1985 experienced this move of subjects from the untrained
to the trained group but failed to conduct analyses by lsquointention-
to-treatrsquo This can bias the results in randomised and non-ran-
domised trials if participants who changed groups differ signifi-
cantly with regard to potentially confounding characteristics On
the other hand Perrino 2002 also experienced this move of sub-
jects from theuntrained to the trained groupmore than 100riders(5 of the total sample) originally stated that they were untrained
at baseline but then received training between the baseline and
follow-up surveys Yet since the Perrino 2002 study was a cohort
their analyses then proceeded by including those riders who were
lsquonewlyrsquo trained as exposed Riders who were lsquonewlyrsquo trained since
baseline but with a programme other than the one under study
were excluded from the analysis Nevertheless confounding fac-
tors were not accounted for in the analyses
McDavid 1989 matched exposed with unexposed subjects but
could not find suitable matches for the majority (60) of their
subjects They could therefore only conduct analysis on the result-
ing40 (139346) which is an inappropriate andinefficient anal-
ysis Some studies that reported on the issue of non-active licensedriders(Satten 1980 Jonah 1982 Mortimer 1984 Mortimer 1988
Kloeden 1994) excluded non-active riders from analysis intro-
ducing selection bias as it is likely that riders who continued to
ride differed with regards to potentially confounding character-
istics across intervention and control groups Kloeden 1994 for
example conducted several analyses to assess the effect of various
sub-sets of the control group Original analyses included all con-
trols and led to erroneous results since only a small proportion of
controls hadgone on to get a licence after their learnerrsquos permitand
were actually exposed to the possibility of experiencing a crash
Subsequent analyses compared only intervention and control sub-
jects who had a licence Waller 1992 reported crashes by person-
days of exposure but did not explain how they could be sure that
each study intervention and control participant was an active riderand actually exposed to the risk of experiencing a crash
Potential confounders that are unequally distributed by study arm
at baseline need to be dealt with in the analysis Satten 1980
Adams 1985 and Davis 1997 did not take into account any con-
founding factors either by design or in the analysis McDavid
1989 Waller 1992 and Billheimer 1998 accounted for potential
confounding factors at the design stage by matching the subjects
in the intervention and control groups Other studies (Raymond
1979 Mortimer 1984 Mortimer 1988 Perrino 2002 and also
Waller 1992) adjusted for one or two confounders by stratifica-
tionor standardisation (eg crashesmiles driven age group) Osga
1980 conducteda univariate analysis of crash rates in which no po-
tential confounders were accounted for but also calculated cumu-lative crash rates by mileage travelled Only Jonah 1982 Cooper
1988 and Savolainen 2007 conducted a multivariate analysis
Risk of bias in the case-control studies
A case-control design is suitable for the assessment of risk fac-
tors for outcomes such as motorcycle crashes injuries or deaths
but not appropriate for the evaluation of an intervention such
as motorcycle rider training It can potentially result in spurious
conclusions regarding the efficacy of the intervention because the
intervention has not been applied at random to an intervention
and control group Even with the control of possible confounders
using a case-control design to evaluate an intervention can lead tothese kinds of results since those who choose to undertake rider
training may differ on important prognostic factors that affect the
outcome compared to those who do not choose to participate in
rider training
Selection bias
A potential source of bias in case-control studies as with all other
study types where the intervention is not randomly allocated to
participants is that comparison groups are likely to differ on po-
tentially confounding characteristics This means that ideally all
cases need to be included in the study as well as a random sample
of controls taken from the population that gave rise to the cases If that is not possible cases and controls either need to be carefully
matched to make themcomparable or differences at baseline need
to be adjusted for in the analysis All four case-control studies were
population based Kraus 1975 included all fatalities and injured
cases leading to hospitalisation and appropriately recruited a ran-
dom sample of controls from the same population Hurt 1981 did
not manage to recruit all or a random sample of cases Only half
of the notifications of crashes to the police were received in time
by the researchers the research team for unreported reasons was
only able to respond to half of those again Data collection was
therefore completed for only about a fifth of all cases Neverthe-
less this was still a large sample of 900 cases Haworth 1997 using
16Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 1950
a similar study design was more likely to have included all cases
but this is not entirely clear from the study reported It is unclearfor example whether fatalities were included as they did not use
coronerrsquos reports for the recruitment of cases The MAIDS 2004
study was conducted in five European countries but it was unclear
how cases were recruited whether they were all cases during the
two years that the study lasted or a random sample of all cases
what the response rates were etc Controls were recruited from
randomly selected petrol stations but it was likewise not clear what
the eligibility criteria for inclusion in the study were how they
were recruited (eg randomly) and what the response rate was
Hurt 1981 and Haworth 1997 both matched controls to cases by
recruiting themat the sametime ofdayandsamedayof weekas the
crash occurred Hurt 1981 however recruited controls two years
later andit is likelythat there would have beendifferences betweenthe two comparison groups One important difference was that
during data collection for cases there was very little specialised
motorcycletrainingavailable in the studyregionwhereas two years
later when control data were collected there was some training
available
These four case-control studies also suffered from selection bias
due to potential differences between those that gave consent to
participate in the study and responded to interviews andquestion-
naires and those that did not None of the studies have assessed
such differences although Hurt 1981 did acknowledge that con-
trols giving consent to cooperate may have differed from those
motorcyclists who did not
Detection bias
Measures that can be taken to reduce detection bias in case-con-
trol studies are blinding of interviewers for case or control status
blinding of the participant for the study factor of interest validat-
ing measurement instruments and using these in an identical way
for cases and controls and cross checking self-reported informa-
tion with other more objective sources such as registers None of
the studies reported blinding of the intervention assessors to case
or control status but all used consistent measurement methods for
both case and control groups
All case-control studies used self-report as their main measure toassess whether cases and controls had received rider training and
if so the type of training The major potential source of bias in
case-control studies when self-report is used as a measure is recall
bias One way of reducing recall bias is by double-checking self-
reported data with more objective data Kraus 1975 also used po-
lice accident reports for information on rider characteristics but
such information was not available about rider training Haworth
1997 and MAIDS 2004 did not report on any measures to limit
recall bias Hurt 1981 reported further questioning of cases by
the interviewer when answers did not make sense or when par-
ticipants seemed to contradict themselves as part of the in-depth
unstructured interview technique This qualitative research tech-
nique can result in valuable information but as interviewers were
not blinded to case or control status of the participants it may have led to more thorough examination of cases compared to the
controls and thus differential detection bias
Intervention integrity
The level of detail about the rider training provided in the case-
control studies differed significantly but was minimal for each
Kraus 1975 simply asked whether lsquoanyrsquo motorcycle rider training
had lsquoeverrsquo been carried out Hurt 1981 asked participants whether
they hadbeen self taught informally taught or completed formally
taught rider training courses MAIDS 2004 asked whether partic-
ipants had had pre-license training additional (non-compulsory)
training other training or no training whereas Haworth 1997asked about the number of courses that a participant had com-
pleted whetherthey were basic intermediate or advanced courses
and how long ago the training had been done None of the case-
control studies reported collecting information about further de-
tails about any training programmes (eg type of training cur-
riculum duration of training)
Analysis
Kraus 1975 accounted for confounding factors through the exclu-
sion of females and stratification by age and frequency of motor-
cycle use Hurt 1981 matched cases and controls for time of day
and day of the week that the crash occurred and did not adjust forany confounding factors in the analysis Haworth 1997 matched
groups for time of day and day of the week that the crash occurred
and stratified by age blood alcohol level and licence status in the
analysis None of the studies used multivariate methods MAIDS
2004 did not provide any information about any participant char-
acteristics other than the study exposures of interest and did not
describe any methods to adjust for potential confounding factors
Effects of interventions
We have presented the results in a descriptive manner due to the
substantial heterogeneity in study types interventions and out-comes and the generally poor quality of the studies We have
grouped the interventions by mandatory and non-mandatory pre-
licence rider training and post-licence rider training and further
by study type and by outcome (crashes injurious and fatal crashes
and offences) We discuss results that are reported as statistically
significant (whether a P value was reported or not)
Pre-licence rider training - mandatory
Randomised trials
17Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2050
Crashes
Total group of participants
Anderson 1980 found a reduction in motorcycle crash rates in
the intervention group (MOST skill test with remedial training)
compared to the control group (who received the skill testwithout
a training component) The percent differenceranged from 143
six months after the programme (P lt 005) to 211 (P lt 0001)
after one year and 127 (P lt 001) after two years Hall 1988
found no statistically significant difference in crash rates between
intervention groups at any point in time
Subgroup of participants who obtained their licence only
Anderson 1980 found a reduction in motorcycle crash rates in the
intervention arm but this was statistically significant at one year
only (24 to 218 reduction P lt 005) Hall 1988 found no
statistically significant differences
Offences
Total group of participants
The percentage of offences in Anderson 1980 after six months
and one and two years was slightly higher in the intervention
arm compared to the control group but these differences were
not statistically significant Likewise the study by Hall 1988 did
not find a statistically significant difference in offences between
groups
Subgroup of participants who obtained their licence only
Hall 1988 found no significant difference in offence rates between
the intervention and control groups Anderson 1980 found thatthe percentage of offences increased in the intervention arms com-
pared to the control group This difference was statistically signif-
icant after two years (51 to 115 increase P lt 005)
Cohort studies
Crashes
Neitherof the cohort studies (Kloeden1994 and Billheimer1998)
found a statistically significant difference in crash rates between
the intervention and control groups
Offences
Billheimer 1998 (for the California Motorcyclist Safety Program
course) and Perrino 2002 did not find a statistically significant
difference in offence rates between the intervention and control
groups
Case-control studies
Crashes
MAIDS 2004 found that a slightly higher percentage of controls
(not involved in a crash) had had pre-license rider training than
cases (involved in a crash) This difference was not statistically significant (50 of controls vs 477 of cases OR 091 95 CI
076 to 110)
Pre-licence rider training - non-mandatory
Randomised trial
Injuries and deaths
Swaddiwudhipong 1998 found a significantly lower injury ratein the intervention villages (1051000 population 95 CI 92
to 120) compared to the control villages (1691000 population
95 CI 152 to 185) one year after the intervention This differ-
ence was not maintained two years after the intervention (127
95 CI 112 to 143 intervention 151 95 CI 135 to 167
control) There was no significant difference in death rates be-
tween groups
Cohort studies
Crashes
Cooper 1988 foundno statistically significant differencein crashes
between the intervention and control groups McDavid 1989
found a lower crash rate in the intervention group (14139) com-
pared to the control group (23139) but no test of statistical sig-
nificance was undertaken
Injuries and deaths
McDavid 1989 found that the intervention group had fewer in-
jurious crashes (814 crashes) compared to the control group (16
23 crashes) but no statistical significance test was reported
18Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2150
Offences
Cooper 1988 foundno statistically significantdifference in offencerates between the intervention and control groups McDavid 1989
foundno statistically significantdifference in offence ratesbetween
the intervention and control groups
Non-mandatory rider training
Non-randomised trials
Crashes
Neither Adams 1985 nor Leung 1987 found a statistically signif-icant effect of rider training on crash rates
Offences
Leung 1987 found a statistically significant lower offence rate in
the intervention group compared to the control group for both
males and females and for all age groups except for women 35
years or older (P values not reported)
Cohort studies
Crashes
Osga 1980 in the univariate analysis reported a lower percentage
of crashesin theintervention group (that continued to ride) (28
168595) than in the control group (34 112329) In the cu-
mulative crash rate analysis however Osga1980 found statistically
significant (at the P lt 001 level) higher crash rates in the interven-
tion group than in the control group at allmileage intervals (1000
to 29999 miles) travelled After riding more than 30000 miles
(when over 50 of the trained riders had been involved in a crash
compared to approximately 40 in untrained riders) this differ-
ence was no longer statistically significant Satten 1980 found a
lower percentage of crashes in the intervention group (13 969)
compared to the control group (25 2079) but the differencewas not assessed statistically Davis 1997 found a lower crash rate
in trained riders but it was unclear how this was calculated be-
cause the size of the control group was not reported Jonah 1982
and Billheimer 1998 (for the Experienced Rider Course) found no
statistically significant difference in crashes between the interven-
tion and control groups Although the difference between those
who completed the Experienced Rider Course and no course in
Billheimer 1998 neared significance at six months after the course
(P = 007) large P values were observed for one year after (P =
030) Raymond 1979 found a statistically significant higher crash
rate for the intervention group (1861000 miles ridden) com-
pared to the control group (1181000 miles ridden) in both males
and females (P lt 0001) especially during the first 5000 miles of
riding Waller 1992 found a statistically significant higher crashrate for the intervention group (27 crashes in three years or 070
crashes10000 person days) compared to the control group (5
crashes in three years or 013 crashes10000 person days) with a
crude relative risk of 538 (X 2=1513 plt005) Mortimer 1984
and Mortimer 1988 found no statistically significant difference in
crashes between the intervention and control groups
Savolainen 2007 found that the intervention group (those who
completed the Basic Rider Course) was 44 more likely to be
involved in crashes than the control group and those who com-
pleted the course multiple times were an additional 180 more
likely to be involved in a crash This finding was based on low
numbers and no P value or confidence level was given for either
result
Injuries and deaths
Satten 1980 also found a lower percentage of injurious crashes
(33 39) in the intervention group compared to the control
group(60 1220) although again no statistical significance test
was reported
Offences
Jonah 1982 did find a significantly lower self-reportedoffence rate
in the intervention group (15) compared to the control group(34 P lt 001) (no numbers given) Satten 1980 also found
a lower self-reported offence rate in the intervention group (1
71 versus 19100) but again this was not tested for statistical
significance Mortimer 1984 Mortimer1988 and Billheimer 1998
found no statistically significant differencein offence ratesbetween
the intervention and control groups
Case-control studies
CrashesMAIDS 2004 found that a slightly higher percentage of cases
(involved in a crash) had had additional non-mandatory rider
training than controls (not involved in a crash) This difference
was not statistically significant (17 of cases vs 12 of controls
OR 146 95 CI 064 to 337)
Rider training unspecified
Case-control studies
19Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2250
Injuries and deaths
Hurt 1981 found that the odds of injurious crashes were lower inthose who had received professional training compared to those
who had not Kraus 1975 found that in the younger age group
(less than 20 years) and the older age group (over 35 years) a larger
percentage of injured riders (12 and 21 respectively) reported
prior motorcycle rider training compared with the uninjured con-
trol group (6 and 5 respectively) (P lt 001) (no numbers
given) Haworth 1997 reported several different outcomes one
beingthat having completed at leastone trainingcoursecompared
to no training courses at all had no significant effect on the odds
of crashing (afteradjustment for age blood alcohol level or licence
status (learner permitprobationaryfull)) When comparing three
different levels of training courses (beginner intermediate and ad-
vanced) with no training no significant differences were found
in the odds of crashing However after adjustment for age and
blood alcohol level completing an advanced course as compared
to a beginners course was associated with a significant decrease in
the odds of severe motorcycle crashes resulting in serious injury or
death (OR 02 95 CI 01 to 08)
D I S C U S S I O N
The objective of this review was to quantify the effectiveness of
pre- andpost-licence motorcycle rider training on the reduction of
traffic crash involvement injuries deaths and offences committedby motorcycle riders Most included studies were cohort studies
There were a few randomised and non-randomised trials and case-
control studies
All studies compared either mandatory or non-mandatory formal
training with no formal training except for Haworth 1997rsquos case-
control study that compared levels of training (beginner interme-
diate and advanced) to each other
We found studies that evaluated the effectiveness of manda-
tory pre-licence training in reducing crash and offence rates
(Anderson 1980 Hall 1988 Kloeden 1994 Billheimer 1998
(the basic California Motorcyclist Safety Program component of
the study) Perrino 2002) non-mandatory training that when
passed resulted in a licence (Cooper 1988 McDavid 1989
Swaddiwudhipong 1998) non-mandatory training that could be
taken by motorcycle riders with or without a licence (Raymond
1979 Osga 1980 Satten 1980 Jonah 1982 Mortimer 1984
Adams 1985 Leung 1987 Mortimer 1988 Waller 1992 Davis
1997 Billheimer 1998 (the Experienced Rider Course compo-
nent of the study)) and unspecified rider training on crash injury
and death rates (Kraus 1975 Hurt 1981 Haworth 1997 MAIDS
2004)
The studies provided conflicting evidence regarding motorcycle
rider training which is likely due to the poor design quality and
reporting of the studies Most of the studies included in the review
were likely to be adversely affected by significant selection biasalong with substantial detection bias(measurement and recall bias)
which limits our ability to make conclusive comments
Mandatory pre-licence rider training
There were two RCTs three cohort studies and one case-control
study that examined pre-licence ridertraining The two RCTs eval-
uated the Motorcycle Skills Operating Test (MOST) (Anderson
1980 Hall 1988) Anderson 1980 found a statistically significant
reduction in crashes in the intervention group at six months one
and twoyearsfollow up However when only participantswho ob-
tainedtheirlicence were included in theanalysis thereduction wasstatistically significant only at one year follow up The other RCT
(Hall 1988) found no statistically significant differences in crash
rates at any point of time neither in the total population group
nor in the licensed subgroup Neither of the two cohort stud-
ies assessing crash rates found a statistically significant difference
in crash rates between intervention and control groups ( Kloeden
1994 Billheimer1998) andthe case-control study (MAIDS 2004)
also did not find a statistically significant difference in the odds of
experiencing a crash between trained and untrained riders With
regards to offences neither of the RCTs found a statistically sig-
nificant difference in offence rate between the intervention and
control groups in the analysis of the total group of participants
entering the trial However Anderson 1980 found a statistically significant increase in offence rates in the intervention group but
only after two years of follow up andonly in the analysis of partici-
pants who obtained their licence The two cohort studies assessing
offence rates found no statistically significant difference between
the groups (Billheimer 1998 Perrino 2002)
None of the five studies (RCTs or cohort) found a reduction in
crashes or offences following mandatory pre-licence rider training
except for Anderson 1980 who observed a statistically significant
reduction in crashes in the intervention group at one year in both
analyses It is not clear why both RCTs that evaluated the same
MOST and training intervention observed different results Both
studies enrolled large samples and observed a similar uptake of
licensure following the intervention It is possible that differencesmay have arisen due to different study samples (California ver-
sus New York) or that the intervention delivery differed between
studies
In both trials of the Motorcycle Skills Operating Test including
rider training only a third of motorcycle licence applicants ob-
tained their motorcycle licence It is possible that a licensing pro-
cess including a skills test may reduce crashes through decreased
exposure
Non-mandatory pre-licence training
20Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2350
Three studies assessed the effect of non-mandatory pre-licence
training The Cooper 1988 cohort study found no difference incrash or offence rates between the trained and untrained groups
We cannot however be confident about this result as the study
sufferedfrom selectionbiaswhichcouldhave biasedthe measure of
effect toward the null The RCT conducted by Swaddiwudhipong
1998 examined injury and death rates and found a statistically
significant lower injury rate in the intervention villages after one
year of follow up but no longer at two year follow up There
was no statistically significant difference in death rates between
intervention and control villages Our confidence in the results of
this study are hindered by the inadequacy of the randomisation
process that led to significant differences between the intervention
and control districts on two important variables The intervention
districts were more likely than the control districts to have a validlicence andto wear a helmetSincethese differences were nottaken
into account in later analyses they could explain the differences
observed in injury rates between districts Furthermore with only
six districts randomised the sample size was probably too small
to detect a difference between intervention and control groups
had there been one The McDavid 1989 cohort study found a
lower percentage of injuries in the intervention group although
a statistical test was not conducted McDavid 1989 drew their
intervention and control groups from different populations but
did adjust for confounders although they only analysed 40 of
the initial sample (139 trained and 139 untrained) Because of
these substantial weaknesses we are unable to draw conclusions
from this studyAn overall effect of pre-licence training on the reductionof crashes
injuries deaths and offences was not found although one should
be aware that the research of this issue to date has had significant
flaws However a mandatory skill test (such as the MOST) in
combination with and even without pre-licence training as part
of a licensing process seems to reduce the number of individuals
completing the licensing procedure and therefore indirectly re-
duces crash rates through a reduction in exposure to motorcycle
riding Making licensure a more difficult process may therefore be
seen as advantageous if it delays licensure and therefore reduces
exposure It is possible that the group who chose not to continue
riding a motorcycle would not have been the group to experience a
crash or commit an offence if they had continued to ride anyhowfor example because they are more safety conscious and this was
the reason for their giving up riding in the first place If this is the
case deterring individuals from completing their licence and rid-
ing may not necessarily decrease overall crashes deaths or offence
rates In addition other potentially adverse effects of complicat-
ing a licensing procedure need to be considered such as increases
in unlicensed riding In Thailand non-mandatory rider training
offered as a community intervention may have decreased injury
rates but this needs to be confirmed through further research
Non-mandatory training
The majority of the studies assessed non-mandatory training
(training independent of any licensing procedure) and the results
were varied The two non-randomised trials assessing crash rates found no statistically significant difference in crash rates between
the intervention and control groups (Adams 1985 Leung 1987)
Of the nine cohort studies Satten 1980 found a lower percentage
of crashesin the intervention group although it is unclear whether
this was a statistically significant result Davis 1997 also found a
lower percentage of crashesin theinterventiongroup although it is
unclear with which control group the intervention group wascom-
pared Raymond 1979 Osga 1980 Waller 1992 and Savolainen
2007 found a statistically significant higher crash rate in the in-
tervention group and the remaining cohort studies (Jonah 1982
Mortimer 1984 Mortimer 1988 Billheimer 1998) found no sta-
tistically significant difference in crash rates between intervention
and control groups No conclusions can be drawn from these stud-
ies with confidence The Satten 1980 and Davis 1997 studies did
not adjust for any confounding factors even though important
differences existed between the intervention and control groups
Although Raymond 1979 and Waller 1992 adjusted for a num-
ber of confounders at baseline (Waller 1992) and in the analy-
sis (Raymond 1979 Waller 1992) the intervention and control
groups were drawn from such different populations that it is pos-
sible that unmeasured confounders were responsible for the sta-
tistically significant increased risk of crashes observed in the inter-
vention group The observed difference could also be attributed totheobserved differential loss to followup (Raymond 1979) which
could bias the measure of effect towards or away from the null
The Osga 1980 study found a protective effect of rider training
in the univariate analysis but statistically significant higher crash
rates in the intervention group when accounting for riding expo-
sure Sex (a significantly higher proportion of females in the in-
tervention group) and other potentially confounding factors that
could explain thedifferencebetween the groups were howeverun-
accounted for Finally Savolainen 2007 did control for potential
confounding variables yet it is not clear whether variables were
controlled for simultaneously or one at a time One case-control
study (MAIDS 2004) found no statistically significant reduction
in the odds of experiencing a crash between trained and untrainedriders but the quality of this study was poor mainly because the
analysis did not account for any confounding factors
One cohort study (Satten 1980) assessed the effect of non-manda-
tory rider training on injurious crashes and found a lower percent-
age of injuries in the intervention group although a statistical test
was not conducted As mentioned above the Satten 1980 studyrsquos
validity is reduced due to its lack of adjustment for confounding
factors
The only non-randomised trial that assessed offence rates (Leung
1987) found a statistically significant lower offence rate in the in-
tervention group and of the five cohort studies that assessed of-
fence rates (Satten 1980 Jonah 1982 Mortimer 1984 Mortimer
21Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2450
1988) one study (Jonah 1982) also found a statistically significant
lower offence rate in the intervention group However we have lit-tle confidence in the results from Leung 1987 since potential con-
founding factors were not appropriately addressed Jonah 1982
did conduct a multivariate analysis but was hindered by an initial
low response rate The other four cohort studies did control for
several potential confounders and found no difference in offence
rates following non-mandatory post-licence rider training
The non-randomised trials and cohort studies evaluating non-
mandatory rider training were at high risk of substantial selection
bias due to the non-random allocation of participants as well
as low participation rates and attrition bias and could therefore
not provide conclusive evidence with regard to the effect of the
training It should be noted that despite the biases and other issues
with this group of studies most didnot detectmajor effects relatedto rider training in either direction This may suggest that any
potential impact of rider training will be modest
Studies examining unspecified modes of ridertraining
Three case-control studies assessed whether injured or deceased
motorcyclists were more or less likely to have had rider training
compared to control motorcyclists All had conflicting results
Hurt1981 foundthe odds of havingan injuriouscrash waslower in
those who had received rider training whereas Kraus 1975 found
that the odds of having an injurious crash were higher in youngerand older trained riders Haworth 1997 who compared advanced
versus beginner training found no effect on motorcycle injuries
or deaths although they did find that those who had completed
advanced training had reduced chances of crashing compared to
those who had done a beginnersrsquo training course Similar issues
with uncontrolled confounding and biases of selection as described
above arose with the three case-control studies Hurt 1981 suffers
from inadequate control of confounding via poor matching and
both the Kraus 1975 and the Haworth 1997 findings were only
controlled for limited factors by means of stratified analysis It is
likely that these limitations are the reason for the contradictory
directions of effect observed for the three case-control studies
Methodological weaknesses of the studies
In summary most studies investigating the effectiveness of mo-
torcycle rider training on reducing crashes injuries and death suf-
fered from serious methodological weaknesses
The majority of studies were non-randomised and were there-
fore limited mainly by selection bias Although the approaches of
matching stratification or regression analyses can minimise bias
(Fleiss 2004) they were poorly employed in the reviewed studies
The inadequate adjustment for confounders in particular adds
to our lack of confidence in the studiesrsquo results since it has been
demonstrated that individuals who choose to take motorcycle rider
training differ from those not taking rider training For instance
trained riders often differ from controls in age sex and experi-ence usually ride less and may have a more positive attitude to
safety as indicated by the reported more frequent wearing of seat
belts (Mortimer 1982)
The two large randomised trials reported that a large proportion
of their study group were never exposed to the risk of offences or
crashes because they never attained licensure This made it impos-
sible to assess whether a reduction in crashes in the intervention
group was a result of training or due to delayed or never attained
licensure
In some studies the sample sizes were too small to detect any sta-
tistically significant differences Follow-up time after training dif-
fered between studies and was an impediment to observing any
real effects of training if they existed As riding experience ac-cumulates over time any differences in outcomes between those
trained and untrained would diminish in studies with lengthy fol-
low-up times This dilution of any possible effect of rider training
may have also occurred in studies that mixed both novice and ex-
perienced riders in the study groups since it is hypothesised that
novice riders serve to benefit most from rider training This would
increase the difficulty in finding any effect of rider training par-
ticularly when sample sizes are small
Detection bias may also account for some of the differences ob-
served between trained and untrained riders Detection bias would
have occurred if the self-report hospital driver or police records
used to record the outcomes of crash injury death or offences
differed between the trained and untrained riders Research hasdemonstrated that not all crashes are reported to police for ex-
ample small crashes or off-road crashes (Kraus 1975) even serious
injury crashes have been found to be underreported (Anderson
1980) However this would only introduce detection bias if the
reporting of outcome differs for trained and untrained riders This
does appear possible in that responsible riders may be more likely
to enroll in a rider training course and also may be more likely
to report crashes to police than irresponsible riders meaning that
the number of crashes from riskier riders could be underrepre-
sented on records An analysis to investigate the effectiveness of
rider training may then show that rider training is not effective in
reducing crashes
Rather than using reported outcomes some studies used self-re-ported survey data although this technique suffers from non-
response biases In fact two studies reported differences in de-
mographic characteristics in response rates to questionnaires
(Anderson 1980 Hall 1988) Self-reported data also introduce
the risk of recall bias survival bias and social desirability response
bias Recall bias may have been equally distributed among trained
and untrained riders howeverit is hypothesised that trained riders
may have felt it was lsquoexpectedrsquo to report good driving histories
hence leading the rider training to appear effective
22Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2550
Limitations of this review
Limitations of this review were that we found very few recentstudies (the two most recent ones published in 2002 and 2007)
and that we found very few studies conducted in low-and middle
income countries even though our search strategy was set up to
identify such studies Anotherlimitation wasthat forthe reviewed
studies we could not assess all outcomes of rider training that are
potentially useful Some of the included studies assessed outcomes
such as use of safety clothing including helmets Such outcomes
are valuable since the use of motorcycle helmets has been demon-
strated to be effective in preventing head injuries (Liu 2004) De-
spite this the use of safety clothing was not included as an out-
come in this review because the studies measuring this outcome
were limited by poor data quality due to very low response rates
Another outcome of interest in assessing the effectiveness of rider
training is lsquotype of injuryrsquo As Haworth 2005 state in their review
of motorcycle licensing and training rider training could be con-
sidered effective if injuries become less severe because motorcycle
riders have learned how to avoid more serious injuries through
rider training The studies included in this review did not specify
type of injury and so this outcome could not be assessed Last
changes in knowledge and attitudes following rider training are
other potentially useful outcomes We have excluded such studies
(Osga 1979 Clayton 1990 Goldenbeld 2004) as these outcomes
are proxies rather than direct measures of the effectiveness of mo-
torcycle rider training We opted to report on the outcomes of
crashes injury death and offences alone since there is little ev-
idence about how changes in knowledge and attitudes relate to
these outcomes
A U T H O R S rsquo C O N C L U S I O N S
Implications for practice
Due to the poor quality of studies included in this review we were
unable to draw any conclusions about the effectiveness of rider
training on crash rates injurious or fatal or offence rates The
findings do suggest that mandatory pre-licence training may be
an impediment to completing a motorcycle licensing process thuspossibly indirectly reducing crash injury death and offence rates
through a reduction in exposure to riding a motorcycle On the
basis of the existing evidence it is not clear if and what type of
training reduces the risk of crashes injuries deaths or offences in
motorcyclists and the selection of the best rider training practice
can therefore not be recommended The suggestion that manda-
tory pre-licence training indirectly reduces exposure to riding the
motorcycle because trainees are impeded from completing the li-
censing process does not help those motorcyclists who persist in
obtaining their licence and riding the motorcycle Some type of
rider training is likely necessary to teach motorcyclists to ride a
motorcycle safely Good policy is based on sound evidence and
evaluation but such evidence for motorcycle rider training is seri-
ously lacking Policy makers are therefore encouraged to include arigorous evaluation component to any new or existing rider train-
ing programme Such an evaluation should ideally be randomized
and assess whether the training programme is effective and if so
which component(s) of the programme work best
Implications for research
Given theresults of this review which found littlereliable evidence
about the effectiveness of motorcycle rider training it is necessary
to build up an evidence base of the effectiveness of motorcycle
rider training in the reduction of crash injury death and offence
rates To this end well designed randomised controlled trials need
to be conducted that examine the impact of rider training oncrashes and offences A randomised controlled trial is the preferred
method as it would significantly reduce most biases described in
this review To minimise the impact of participants delaying or
never attaining their licence and therefore exposure as found in
the randomised trials in this review such a trial should include
riders who are already licensed and reliable measures of riding
exposure in a reasonably lengthy follow up period It is expected
that a randomised controlled trial of sufficient sample size would
observe equal balance on possible confounders by study arm how-
ever an assessment of potentially important confounders (such as
helmet wearing) by study group would enable control for these
variables in the analyses if necessary It is imperative that the effec-
tiveness of each component of the rider training (such as theorypractice knowledge skills and modification of behaviour) and the
synergy of such components are evaluated Outcomes should be
objective ideally those that are routinely collected rather than self-
reported outcomes Even though riding the motorcycle poses dif-
ferent challenges and incurs different risks lessons may be learned
from research into the effects of car driver training
Motorcycles are a common means of transport in low- and middle
income countries and usage is increasing As the burden of mo-
torcycle crash victims is mostly borne in such countries research
should be focused on the effectiveness of types of rider training on
the reduction of crashes injuries and fatalities in such countries
and be supported by higher income countries This review found
only one study conducted in a middle income country that sug-gested decreasing injury rates It would be beneficial to conduct
additional research to confirm this suggestion and to enhance our
knowledge of the role of rider training in low- and middle-income
settings
March 2006 saw the launch of the University of North Carolina
Highway Safety Research Centre partnership with the Motorcycle
Safety Foundation (MSF) and plans were revealed to conduct a
study of the potential benefits of motorcycle rider training The
study is funded by the MSF and the National Highway Traffic Sa-
fety Administration (NHTSA) Called rsquoThe Discovery Projectrsquo it
will compare data on crash rates from riders participating in a basic
23Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2650
motorcycle safety class with riders participating in a basic motor-
cycle safety class as well as a variety of continuing education classesThe collection of pilot data was due to start in March 2007 and
the data collection portion of the study is expected to take three
years (Highway Safety Research Center) It is vital that researchers
and policy makers collaborate to develop furtherhigh quality eval-
uations of motorcycle rider training programmes to help inform
development of new and effective training programmes
A C K N O W L E D G E M E N T S
Staff from The University of Sydney library for helping us locate
and retrieve full reports of many papers
R E F E R E N C E S
References to studies included in this review
Adams 1985 published data only
Adams A Collingwood V Job RFS Evaluation of the 1982 Australia
Post motorcycle rider training scheme NSW Australia Traffic
Accident Research Unit Traffic Authority NSW 1985 [ RN 85]
Anderson 1980 published data only
Anderson J Ford JL Peck RC Improved motorcyclist licensing and testing project Washington DC National Highway Traffic Safety
Administration (NHTSA) USA 19801ndash217 [
DOTndashHSndash805ndash598]
Billheimer 1998 published data only
Billheimer JW Evaluation of the California Motorcyclist Safety
Program Transportation Research Record 19981640100ndash9 [
TRB Report Number 980652]
Cooper 1988 published data only
Cooper PJ Rothe JP Motorcyclists who they are and why they do
what they do Transportation Research Record 1988116878ndash85
Davis 1997 published data only
David CF Evaluation of community traffic safety programs and
motorcycle operator training programs Connecticut Department of Transportation 199763
Hall 1988 published data only
Hall CJ Buchanan L Engle R Motorcycle rider education evaluationproject Washington DC USA National Highway Traffic Safety
Administration (NHTSA) US Department of Transportation
1988 [ DTNH 22ndash80ndashCndash05123]
Haworth 1997 published data only
Haworth N Smith R Brumen I Pronk N In Monash University
Accident Research Centre editor(s) Case-control study of motorcycle
crashes Victoria Australia Department of Transport and Regional
Development The Federal Office of Road Safety 19971ndash149 [
ISBN 0 642 25501 6 ]
Hurt 1981 published data only
Hurt H Ouellet J Thom D Motorcycle accident cause factors and
identification of countermeasures USA Department of
Transportation 1981 [ Technical Report DOT HS 806ndash862]
Jonah 1982 published data onlylowast Jonah BA Dawson N Bragg B Are formally trained motorcyclists
safer Accident Analysis amp Prevention 198214(4)247ndash55
Jonah BA Dawson NE Bragg BW Evaluation of the Canada
Safety Councilrsquos Motorcycle Training Program Proceedings of theInternational Motorcycle Safety Conference Washington DC
USA 198059ndash93
Kloeden 1994 published data only
Kloeden CN Moore VM McLean AJ Evaluation of the pre-license
training program for motorcyclists in South Australia Walkerville
South Australia Australia NHMRC Road Accident Research Unit
The University of Adelaide 19941ndash37 [ ISBN 0 7308 0607 3]
Kraus 1975 published data only
Kraus JF Riggins RS Franti CE Some epidemiological features of
motorcycle collision injuries American Journal of Epidemiology 1975102(1)74ndash98
Leung 1987 published data only
Leung KS Reding VA Evaluation of the Wisconsin motorcycle rider course Wisconsin Bureau of Policy Planning and Analysis
Wisconsin Department of Transportation 1987
MAIDS 2004 published and unpublished data
MAIDS In-depth investigations of accidents involving powered
two wheelers ACEM (Acemacemeu) 2004 Vol Final report
20179
McDavid 1989 published data only
McDavid JC Lohrmann BA Lohrmann G Does motorcycle
training reduce accidents Evidence from a longitudinal quasi-
experimental study Journal of Safety Research 198920(2)61ndash72
Mortimer 1984 published data only
Mortimer R Evaluation of the motorcycle rider course Accident
Analysis amp Prevention 198416(1)63ndash71
24Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2750
Mortimer 1988 published data only
Mortimer RG A further evaluation of the motorcycle rider courseJournal of Safety Research 198819(4)187ndash96
Osga 1980 published data only
Osga GA An investigation of the riding experiences of MSF rider
course participants Dissertation South Dakota USA The
University of South Dakota 19801ndash224
Perrino 2002 published data only
Perrino CS Ahmed A Callender A Rozier E Cantwell A Stewart
O The role of Marylandrsquos motorcycle rider course in promoting saver
behaviors and attitudes Baltimore USA National Transportation
Center Morgan State University 20021ndash52
Raymond 1979 published data only
Raymond S Tatum S An evaluation of the effectiveness of the RAC
ACU motorcycle training scheme - final report Road Safety ResearchUnit Dept of Civil Engineering University of Salford London
19791ndash131
Satten 1980 published data only
Satten RS Analysis and evaluation of the motorcycle rider courses
in thirteen Illinois counties International Motorcycle Safety
Conference Washington DC 1980145ndash93
Savolainen 2007 published and unpublished data
Savolainen Peter Mannering Fred Additional evidence on the
effectiveness of motorcycle training and motorcyclistsrsquo risk-taking
behaviour Transportation Research Board 2007 Vol March 20
19
Swaddiwudhipong 1998 published data only
Swaddiwudhipong W Boonmak C Nguntra P Mahasakpan P
Effect of motorcycle rider education on changes in risk behavioursand motorcycle-related injuries in rural Thailand Tropical Medicine
and International Health 19983(10)767ndash70
Waller 1992 published data only
Waller FL An analysis of the impact of the Texas Department of
Public Safety Motorcycle Operator Training Program upon
motocycle related traffic accidents Texas State University Public
Administration Program Applied Research Projects 199277
References to studies excluded from this review
Anderson 1980a published data only
Anderson LC Better biking a motorcycle rider improvement
program International Motorcycle Safety Conference Washington
DC USA Motorcycle Safety Foundation 1980231ndash53Aupetit 2008 published and unpublished data
Aupetit S Riff J Buttelli O Espie S Ergonomics research program
on motorcycle learning Contributions to the development of a
motorcycle simulator for training Transport Research Arena
Ljubljana 20089
Awane 1999 published data only
Awane T Integrating simulators in motorcycle safety education
Journal of International Association of Traffic and Safety Sciences 199923(1)26ndash35
Baldi 2005 published data only
Baldi S Baer JD Cook AL Identifying best practices states in
motorcycle rider education and licensing Journal of Safety Research200536(1)19ndash32
Batchler 1988 published data only
Batchler MW Evaluation of the pre-license motorcycle training course at the New South Wales traffic education centre Federal Office of
Road Safety (FORS) 19881ndash122 [ Report No CR 77 ]
Blanchard 2006 published data only
Blanchard HT Tabloski PA Motorcycle safety educating riders at
the teachable moment Mosby (CV) Company 200632330ndash2
Bonnett 2005 published data only
Bonnett J Mature age skills training for experienced motorcycle
riders Motorcycle ridersrsquo association of the Australian Capital
Territory 2005
Braver 2007 published data only
Braver ER et alPersuasion and licensure a randomized controlled
intervention trial to increase licensure rates among Maryland
motorcycle owners Traffic Injury Prevention 20078(1)39ndash46Clayton 1980 published data only
Clayton AB Sudlow DE An evaluation of the effectiveness of a one day training programme for learner motorcyclists Berkshire UK
Transport and Road Research Laboratory 19801ndash23 [ ISSN
0266ndash7045]lowast Clayton AB Sudlow DE An evaluation of the effectiveness of a one-day training programme for learner motorcyclists Berkshire UK
Road Safety Division Berkshire 19801ndash23 [ISSN 0266ndash7045]
Council 1978 published data only
Council FM Desper LP Dutt AK Driver education for motorcycle operation University of North Carolina Highway Safety Research
Center 19781ndash40
Crowther 2005 published data onlylowast Crowther G Engaging with motorcyclists UK police and theBikeSafe Road Safety Programme Published proceedings of the
Australasian Road Safety Research Policing Education Conference
Wellington New Zealand 2005 issue ISBN 0ndash473ndash10636ndash11ndash11
Daniello 2009 published data only
Daniello A Gabler HC Mehta YA Effectiveness of motorcycle
training and licensing Transportation Research Board 88th Annual
Meeting 200913
Dewanti 2007 published data only
Dewanti M Challenging efforts in promoting young motorcyclist
safety in Indonesia Road Safety on Four Continents 14th
International Conference Swedish National Road and Transport
Research Institute 2007
Du Plooy 1990 published data onlyDu Plooy DW Motorcycle training in South Africa The Human
Element International Motorcycle Safety Conference Orlando
Florida USA Motorcycle Safety Foundation 19904-1 to 4-9
Engel 1989 published data only
Engel U Krogsgaringrd Thomsen L [Faeligrdselsundervisning og
propaganda ndash en faeligrdselssikkerhedsfremmende foranstaltning]
Dansk Vejtidsskrift 198912295ndash7
Goldenbeld 2002 published data only
Goldenbeld C Houwing S de Craen S The development of
driving skills of young moped riders [De ontwikkeling van de
rijvaardigheid van jonge bromfietsers] SWOV Institute for Road
Safety Research (Stichting Wetenschappelijk Onderzoek
Verkeersveiligheid) Leidschendam the Netherlands 200263
25Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2850
Goldenbeld 2004 published data only
Goldenbeld C Twisk D de Craen S Short and long term effects of moped training a field experiment Transportation Research Part F 20047(1)1ndash16
Haworth 2000 published data only
Haworth N Smith R Kowadlo N Motorcycle training in Victoria
evaluation of rider training curriculum in Victoria Monash
University Accident Research Centre Reports 2000 Vol 165 issue
3131
Haworth 2005 published data only
Haworth N Mulvihill C Review of motorcycle licensing and
training Monash University Accident Research Centre 2005 [
Report no240]
Haworth 2006 published data only
Haworth N Mulvihill C Rowdon P Teaching old dogs new tricksTraining and older motorcyclists Journal of the Australasian College of Road Safety 200718(4)20ndash5lowast Haworth NL Mulvihill C Rowden PJ Teaching old dogs new
tricks Training and older motorcyclists Australasian Road Safety
Research Policing and Education Conference Gold Coast
Queensland 2006
Hill 1983 published data only
Hill D Motorcycle rider training testing and licensing Motor
Cycle Council of New South Wales Sydney 1983
Imatake 1980 published data only
Imatake Y On the activities in Japan of motorcycle safe riding
promotion committee (MSRPC) - especially the education of
unskilled riders of class 1 motorbikes International Motorcycle
Safety Conference Washington DC USA Motorcycle Safety
Foundation 1980255ndash68
Ingebritsen 1990 published and unpublished data
Ingebrigtsen S [Risikofaktorer ved ferdsel med moped og
motorsykkel] TOslashI-rapport 66 Oslo Transportoslashkonomisk
institutt 1990
Jain 2009 published and unpublished data
Jain A et alTwo wheeler accidents on Indian roads a study from
Mangalore India Elsevier 200916130ndash3
Jonah 1979 published data only
Jonah BA Dawson NE Validation of the motorcycle operator skill
test Accident Analysis amp Prevention 197911(3)163ndash71
Jonah 1981 published data onlyJonah BA Dawson NE Bragg WE Predicting accident
involvement with the Motorcycle Operator Skill Test Accident Analysis amp Prevention 198113(4)307ndash418
Kadar 1990 published data only
Kadar KJ Countermeasures motorcycle rider education delivery
system The Human Element International Motorcycle Safety
Conference Motorcycle Safety Foundation 19906-27 to 6-41
Kelsey 1986 published data only
Kelsey SL Liddicoat C Ratz M Licensing novice motorcyclists a
comparison of Californiarsquos standard test and the MOST II (Motorcycle
Operator Skill Test) administered at centralized testing offices Sacramento California USA California Department of Motor
Vehicles Research and Development Office 19861ndash74
Koch 1980 published data only
Koch H Advanced course for motorcycle riders InternationalMotorcycle Safety Conference Washington DC USA Motorcycle
Safety Foundation 1980269ndash81
Koch 1990 published data only
Koch H Brendicke R How to influence risk taking behaviour
within motorcycle rider training programs The Human Element
International Motorcycle Safety Conference Orlando Florida
USA Motorcycle Safety Foundation 1990Motorcycle Safety
Foundation 1990 p 9-1 to 9-25A
Labbett 2005 published data only
Labbett S Langham M Training can make the problem worse
Road Safety Congress 20051ndash12
Law 2005 published and unpublished data
Law TH Radin Umar RS Zulkaurnain S Kulanthayan S Impactof the effect of economic crisis and the targeted motorcycle safety
programme on motorcycle-related accidents injuries and fatalities
in Malaysia International Journal of Injury Control and Safety
Promotion 200512(1)9ndash21
Lowes 1990 published data only
Lowes BF In-traffic rider training The Human Element
International Motorcycle Safety Conference Orlando Florida
USA Motorcycle Safety Foundation 19907-46 to 7-57
Mahon 1976 published data only
Mahon J Teaching accident avoidance on the motorcycle Journal of Traffic Safety Education 1976231ndash29
Maxwell 1985 published data only
Maxwell GS Evaluation of a Program for Training MotorcycleRiders Evaluation and Program Planning 19858(4)339ndash47
Mayhew 1996 published data only
Mayhew DR Simpson SH Effectiveness and role of driver educationand training in a graduated licensing system Ottowa Canada The
Traffic Injury Research Foundation 19961ndash89 [ ISBN
0ndash920071ndash12ndash0]
McKnight 1987 published data only
McKnight AJ Milburn BG An evaluation of the modified
motorcycle operator skill test exercises NHTSA Report 1987 [
Report No DOT HS 807 212 ]
McPherson 1976 published data only
McPherson K McKnight AJ The development and evaluation of a
motorcycle skill test manual and knowledge test Washington DCUSA National Highway Safety Administration 1976
Mohan 2004 published data only
Mohan D Evidence-based interventions for road traffic injuries in
South Asia Journal of the College of Physicians and SurgeonsPakistan 200414(12)746ndash7
Morgan 1980 published data only
Morgan JK Advanced motorcycle training course content and
study evaluation International Motorcycle Safety Conference
Washington DC USA Motorcycle Safety Foundation 1980
283ndash312
Motoki 1990 published data only
Motoki M Yamazaki S A study on effective motorcycle rider
education The Human Element International Motorcycle Safety
26Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2950
Conference Orlando Florida USA Motorcycle Safety
Foundation 19909-26 to 9-57A
Nairn 1992 published data onlylowast Nairn RJ Motorcycle safety research literature review 1987 to
1991 Federal Office of Road Safety Australia 1993 Vol Report
CR 117
Newland 1980 published data only
Newland R Darby J Automotive practicespre motorcycle rider
program International Motorcycle Safety Conference Washington
DC USA Motorcycle Safety Foundation 1980127ndash9
Newland 1982 published data only
Newland R Motorcycle rider training and licensing requirements
Driver training steering a course for the future Melbourne
Victoria Transport Regulation Board 1982
Newland 1990 published data only
Newland R Motorcycle training and licensing in Victoria 1980-
1990 The Human Element International Motorcycle Safety
Conference Orlando Florida USA Motorcycle Safety
Foundation 1990
Ormston 2003 published data only
Ormston R Dudleston A Pearson S Stradling Evaluation of
Bikesafe Scotland Scottish Executive Social Research Edinburgh
2003
Osga 1979 published data only
Osga GS Ellingstad VS An evaluation of the effectiveness of the 1978
South Dakota motorcycle rider course program South Dakota USA
University of South Dakota 19791ndash89
Panou 2007 published data only
Panou M Bekiaris E A new concept on the integration of driving
simulators in driving training the train-all approach Road Safety
on Four Continents 14th International Conference Swedish
National Road and Transport Research Institute 2007
Peterson 1990 published data only
Peterson A Kadar K Status report 1980-1989 Minnesota rider
education statistics The Human Element International
Motorcycle Safety Conference Orlando Florida USA 199013-
113 to 13-129
Prem 1984 published data only
Prem H Good MC Motorcycle rider skills assessment ACT
Australia Office of Road Safety Department of Transport 1984
1ndash377 [ Report No CR 34 ]
Radin Umar 2006 published data only
Radin Umar RS Motorcycle safety programmes in Malaysia how
effective are they International Journal of Injury Control and Safety Promotion 200613(2)71ndash9
Rockwell 1990 published data only
Rockwell TH Kiger S Carnot M An evaluation of the Ohio
motorcyclist enrichment program Ohio Department of Highway
Safety 1990
Rothe 1987 published data only
Rothe JP A motorcycle riding course Motorcyclists Image and reality Canada Insurance Corporation of British Columbia 1987
59ndash83
Rowden 2007 published data only
Rowden P Watson B Haworth N What can riders tell us aboutmotorcycle training A view from the other side of the fence
Australasian Road Safety Research Policing and Education
Conference Melbourne Australia Melbourne 2007
RSC Victoria 1993 published data only
Road Safety Committee Inquiry into Motorcycle Safety in Victoria
Parliament of Victoria 1993
Russam 1979 published data only
Russam K Motorcycle training and licensing in Japan UK Transport
and Road Research Laboratory (TRRL) 1979 [ Report 916 ]
Saffron 1981 published data only
Saffron DG Driver and rider education training and licensing a
brief review Traffic Accident Research Unit Traffic Authority of
New South Wales NSW Australia Traffic Accident Research UnitTraffic Authority of New South Wales 1981
Schulz 1990 published data only
Schulz U Motorcycle accidents among learner riders Motorcycle
licensing Evaluation of the effectiveness of a Graduated Licensing Scheme Forshungshefte des Institut fur Zweiradsicherheit 1990
Sexton 2004 published data only (unpublished sought but not used)
Sexton B Baughan C Elliott M Maycock G The accident risk of
motorcyclists 2004TRL Report TRL 607(ISSN 09684107)58
Shephard 1986 published data only
Shepherd R Do Motorcycle Safety Foundation programs work
Driver 198619(9)16ndash8
Simpson 1990 published data onlySimpson HM Mayhew DR The promotion of motorcycle safety
training education and awareness Health Education Research
Theory and Practice 19905(2)257ndash64
Sokolov 1990 published data only
Sokolov O Safe motorcycle riding some aspects of motorcycle
riding in the USSR The Human Element International
Motorcycle Safety Conference Motorcycle Safety Foundation
19904ndash37 - 4-48
Steffens 1988 published data only
Steffens U Gawatz R Willmes G Effectiveness of mofa courses
[Wirksamkeit von Mofakursen] Unfall- und sicherheitsforschung
strasenverkehr 198867168
Sudlow 2003 published data only
Sudlow D Scoping study on motorcycle training Road Safety
Research Report London UK The British Institute of Traffic
Education Research Department for Transport 2003 Vol June
[ Report No 36]
Tasmania 1995 published data only
Mannering FL Grodsky LL Motorcycle rider training scheme
Traffic Education Research Centre Armidale Australia Driver
Training Australia Limited 1995 Vol April1ndash56
Tetard 1985 published data only
Tetard C Motorcycle ridersrsquo training to detect indices in an
intersection situation [Lrsquoentrainement a la prise drsquoinformation dans
lrsquoapprentissage de la conduite motocycliste] Le Travail Humain198548(2)147ndash59
27Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3050
Thackray 1980 published data only
Thackray RM Prescott JC Field test of the motorcycle ridercourse International Motorcycle Safety Conference Washington
DC USA Motorcycle Safety Foundation 1980195ndash7
Thompson 1994 published data only
Thompson M Evaluation of compulsory basic training for
motorcyclists Transport Research Laboratory Crowthorne
Berkshire UK 1994 [ Report 63]
Tobler 1990 published data only
Tobler U Individual driving instruction for motorcycle driving
learners the Swiss experiment of motorcycle driving school The
Human Element International Motorcycle Safety Conference
Orlando Florida USA Motorcycle Safety Foundation 1990
Transport 1989 published data only
Transport Communications Road Crash Statistics Australia Bureauof Transport and Communications Economics 198914ndash20
Tuan 2007 published data only
Tuan VA Shimizu T Towards development and evaluation of the
motorcycle drivers re-education program in Vietnam modeling of
motorcycle driverrsquos undesired behaviors 11th World Conference
on Transport Research 200725
Victor 1980 published data only
Victor DJ Traffic safety education for motorcyclists International
Motorcycle Safety Conference Washington DC USA Motorcycle
Safety Foundation 1980221ndash8
Waterer 1990 published data only
Waterer TH Compulsory training for motorcyclists from liberty
to compulsion The Human Element International Motorcycle
Safety Conference Orlando Florida USA Motorcycle Safety
Foundation 19904-49 to 4-57
Weller 1989 published data only
Weller RB Chandler EW Motorcycle safety and motorcycle
education past research and survey results Journal of Public Policy
amp Marketing 1989893ndash108
Wells 1981 published data only
Wells P Observations of some RACACU motorcycle training
schemes Transport and Road Research Laboratory Crowthorne
Berkshire UK 1981
Winn 1986 published data only
Winn G Two important trends in motorcycle safety regulations
rider education and conspicuity improvement American
Motorcyclist Association 1986147ndash52
Wisher 1988 published data only
Wisher G Reid J Motorcycle accidents in Norfolk a three-year study of motorcycle training and the Part 1 test Norfolk UK Road Safety
Division Highways Department Norfolk County Council 1988
Wood 1987 published data only
Wood HT Bowen R Evaluation of the revised learner permit scheme
July 1983 to December 1985 Victoria Australia Road Traffic
Authority 19871ndash27 [ ISBN 0 7306 0247 8]
Yamazaki 1990 published data only
Yamazaki S Nakanishi M Survey on effect of Honda rider
education The Human Element International Motorcycle Safety
Conference Orlando Florida USA 199012-167 to 12-187
References to ongoing studies
Burgess 2010 unpublished data only
Burgess C Broughton P Fylan F Stradling Interim evaluation of
the UKs National Ride program In L Dorn editor(s) Driver
behaviour and Training Vol IV Aldershot Ashgate Publishing
Limited 2010
Additional references
Anderson 1980
Anderson J Ford JL Peck RC Improved motorcyclist licensing and
testing project Volume I and II National Highway Traffic Safety
Administration US department of Transportation Washington
DC 1980
Bjornstig 1985
Bjornstig U Bylund P Lekander T Brorsson B Motorcycle fatalitiesin Sweden Acta Chirurgica Scandinavica 1985151(7)577ndash81
Clayton 1990
Clayton A Sudlow DE An evaluation of the effectiveness of a one-
day training program for learner motorcyclists The Human
Element International Motorcycle Safety Conference Orlando
Florida USA 199013-105 to 13-111
Cochrane Collaboration 2008
Higgins JP Green S Cochrane Handbook for Systematic Reviews of
Interventions 500 The Cochrane Collaboration 2008
Elliot 2003
Elliott MA Baughan CJT Broughton J Grayson G Knowles J et
alMotorcycle Safety a Scoping Study TRL [ Report No
0968ndash4107]Fleiss 2004
Fleiss JL Levin B Paik MC Statistical methods for rates and proportions 3rd Edition John Wiley amp Sons Inc 2004 [ ISBN
978ndash0ndash471ndash52629ndash2 ]
FORS 1999
FORS Federal Office of Road Safety Road risk for sober licensed
motorcyclists Vol Monograph 27 Federal Office of Road Safety
19991ndash2
Haworth 2005
Haworth N Mulvihill C Review of motorcycle licensing and training
Victoria Australia Monash University Accident Research Centre
20051ndash83 [ Report No 240]
Huwiler-Muumlntener 2002Huwiler-Muumlntener K Juumlni P Junker C Egger M Quality of
Reporting of Randomized Trials as a Measure of Methodologic
Quality JAMA 20022872801ndash4
Johnston 2008
Johnston P Brooks C Savage H Fatal and serious road crashes
involving motorcyclists Road Safety Research and Analysis Report
ATSB Monograph 20 Canberra Department of Infrastructure
Transport Regional Development and Local Government 0205
081ndash26
Juumlni 2001
Juumlni P Altman DG Egger M Systematic reviews in health care
Assessing the quality of controlled clinical trials BMJ 2001323
42ndash6
28Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3150
Kopits 2003
Kopits E Cropper M Traffic fatalities and economic growth TheWorld Bank Development Research Group Infrastructure and
Environment Policy research working paper 3035 2003
Leung 1983
Leung P Chau BTW Motorcycle accidents in Hong Kong pattern
of injury treatment causative factors and socio-economic
implications Injury 198315(1)23ndash9
Liu 2004
Liu B Ivers R Norton R Blows S Lo SK Helmets for preventing
injury in motorcycle riders Cochrane Database of Systematic Reviews 2004 Issue 2 [DOI 10100214651858CD004333pub3]
Mohan 2002
Mohan D Road safety in less-motorized environments future
concerns International Journal of Epidemiology 200231(3)527ndash32Mortimer 1982
Mortimer RG An evaluation of the motorcycle rider course
Proceedings of the 20th conference of the American Association for
Automotive Medicine Arlington Heights IL USA American
Association for Automotive Medicine 198277ndash96
Mullin 1997
Mullin BT Risk factors for motorcycle injury the role of age
gender experience training and alcohol PhD thesis Auckland
New Zealand University of Auckland 1997
Noordzij 2001
Noordzij PC Forke E Brendicke R Chinn BP Integration of needs
of moped and motorcycle riders into safety measures review and
statistical analysis in the framework of the European research
project PROMISING SWOV Institute for Road Safety Research
Leidschendam The Netherlands 2001 [ Report No Dndash2001ndash5]
Osga 1979
Osga G Ellingstad VS An evaluation of the effectiveness of the 1978 South Dakota Motorcycle Riders Course Program South Dakota
University of South Dakota Vermillion Human Factors Lab 1979
Oxford English Dictionary Online
httpoxforddictionariescom Oxford English Dictionary Online
Oxford University Press 2010
Peden 2004
Peden M Scurfield R Sleet D World Report on Road TrafficInjury Prevention Geneva World Health Organization Available
at httpwwwwhointviolence˙injury˙preventionpublications
road˙trafficworld˙reporten 2004Chapter 3
Radin Umar 1995
Radin Umar R Macakay G Hills B Preliminary analysis of
motorcycle accidents Short-term impacts of the running
headlights campaign and regulations in Malaysia J Traffic Med 199523(1)17ndash27
Tsai 1995
Tsai YJ Wang JD et alCase-control study of the effectiveness of
different types of helmets for the prevention of head injuries among
motorcycle riders in Taipei Taiwan American Journal of
Epidemiology 1995142(9)974ndash81
Vis 1995
Vis AA In-depth study of the hazards of motorcycling [De
onveiligheid van motorrijden nader bekeken] SWOV Institute for
Road Safety Research Leidschendam the Netherlands 1995 [
Report No Rndash95ndash69]
Weaver 1990
Weaver EA Motorcycle safety education in the United States
Motorcycle Safety Foundation Orlando Florida USA 1990
Winn 1987
Winn G Two important trends in motorcycle safety regulations
rider education and conspicuity improvement No more
information available Detroit MI US Society of Automotive
Engineers 1987147ndash52
Yuan 2000
Yuan W The effectiveness of the rsquoride-brightrsquo legislation for
motorcycles in Singapore Accident Analysis amp Prevention 200032
(4)559ndash63
Zambon 2006
Zambon F Hasselberg M Factors affecting the severity of injuries
among young motorcyclists a Swedish nationwide cohort study
Traffic Injury Prevention 20067(2)143ndash9lowast Indicates the major publication for the study
29Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3250
C H A R A C T E R I S T I C S O F S T U D I E S
Characteristics of included studies [ordered by study ID]
Adams 1985
Methods Non-randomised trial
Participants Initial sample size
Total of 162 Australia Post telegram delivery riders
Not clear how many in intervention and control groups
Analysis sample size
Intervention group 63Control group 35
Total 98 (61)
Interventions No formal training vs voluntary rider training course of 12 hours over 2 consecutive days (theory 4 hours practice
8 hours)
Intervention integrity
No information given
Outcomes Australia Post recorded crashes
Length of follow up 1 year
Notes Study conducted in NSW Australia
Anderson 1980
Methods Randomised trial
Participants Initial sample size
Total 41188 motorcycle licence applicants
Analysis sample size
Intervention group with skills training 13160
Intervention group without skills training 12634
Control group 15080
Total 40874 (99)
Interventions Standard licensing skill test vs the Motorcycle Operators Skill Test (MOST) with or without a training component
MOST is not to be confused with rider training because it is a licensing test The training component was for
participants who failed the skill test on the first attempt was of 3 hours duration and consisted of instructions about
the skills needed to pass the skill test The training consisted of 30 min theory and 2 frac12 hours practice on the
motorcycle
Intervention integrity
33 in MOST with skills training 31 in MOST without skills training and 43 in control group obtained their
motorcycle licence in the end
Outcomes Police recorded crashes and offences
Length of follow up 3 years
30Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3350
Anderson 1980 (Continued)
Notes Study conducted in California USA
Billheimer 1998
Methods Cohort study
Participants Initial sample size
Intervention group Unclear numbers of CMSP graduates and ERC graduates
Control group Over 16000 untrained motorcyclists
Analysis sample size
1139 matched CMSP-pairs and 1182 ERC-pairs
Interventions No formal training vs the mandatory (for those younger than 18 years) pre-licence 16-hours basic California
Motorcyclist Safety Program (CMSP) and the 8-hour Experienced Rider Course (ERC)
Intervention integrity
No information given
Outcomes Police recorded crashes and offences
Length of follow up 2 years
Notes Study conducted in California USA
Cooper 1988
Methods Cohort study
Participants Initial sample size
Intervention group 418 graduates and 57 who failed
Control group 402
Analysis sample size
Intervention group unclear
Control group unclear
Total 863 (984)
Interventions No formaltrainingvs non-mandatory pre-licence rider trainingThe content of thetrainingwas notfurther explainedIntervention integrity
Participants had all completed the rider training
Outcomes Police recorded crashes and offences
Length of follow up 7 years
Notes Study conducted in British Columbia Canada
31Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3450
Davis 1997
Methods Retrospective cohort study
Participants Initial sample size
Intervention group 9320 graduates
Control group unclear
Analysis sample size
Intervention group 9320 graduates
Control group unclear
Interventions No formal training vs the non-mandatory Connecticut Motorcycle Operator Training Program The content of
training was not further explained
Intervention integrity Participants had all completed the rider training
Outcomes Police recorded crashes severity of crashes
Length of follow up 1 year
Notes Study conducted in Connecticut USA
Hall 1988
Methods Randomised trial
Participants Initial sample sizeIntervention group A 6248
Intervention group B 6616
Intervention group C 6996
Control group 6604
Total 26464 motorcycle licence applicants
Analysis sample size
Intervention group A 6012 (96)
Intervention group B 6428 (97)
Intervention group C 6778 (97)
Control group 6502 (98)
Interventions Standard licensing skill test vs the Motorcycle Operator Skill Test (MOST II) without training (A) MOST II with
3 hours training (B) or MOST II with 20 hours training (C) The 3 hours training comprised instructions aboutthe skills needed to pass the skill test (30 min theory and 2 frac12 hours practice on the motorcycle) The 20 hours
training consisted of the Motorcycle Safety Foundationrsquos Motorcycle Rider Course (MRC) comprising of 8 hours
theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-road practice
Intervention integrity
25 and 26 obtained their motorcycle licence in the lsquoMOST 3-hour trainingrsquo and lsquoMOST 20-hour trainingrsquo
intervention arm respectively In the lsquoMOST testrsquo intervention arm 29 obtained their licence and in the control
group 35
Outcomes Police recorded crashes and offences
Length of follow up 3 years
32Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3550
Hall 1988 (Continued)
Notes Study conducted in New York USA
Haworth 1997
Methods Population based case-control study
Participants Cases Motorcycle riders or pillion riders involved in motorcycle crash (on road) Recruited from hospital and police
records Not clear whether all cases (especially fatalities) were included
Community controls Multiple matched sample of motorcycle rider or pillion who passed the crash site at the same
time of day and day of week that the crash occurred
Initial sample size
Cases 336
Controls 1195
Analysis sample size
Cases 234 (70)
Controls 561 (47)
Interventions Basic handling skills course beginner intermediate or advanced course measured through self-report not validated
Outcomes Injuries with an Injury Severity Score of 5 or greater or admission for at least four hours and fatalities from a
motorcycle crash (from hospital and police records)
Length of follow up two years or more prior to the interview
Notes Study conducted in Victoria Australia
Hurt 1981
Methods Population based case-control study
Participants Cases Motorcycle riders who had a motorcycle crash that was reported to police (traffic accident reports) Not all
cases included
Community controls Multiple matched sample of motorcycle riders who passed the crash site at the same time of
day and day of week as the crash occurred but 2 years later
Initial sample size
Cases 4500 notifications of crashes to police (only 50 received in time research team able to respond to 1126 for
unreported reasons data collection completed for 900)
Controls Number approached unreported 2310 consented and interviewed
Analysis sample size
Cases 808 (808900 data on rider training) - 18 of initial sample
Controls 670 (6702310 data on rider training) - lt 29 of initial sample
Interventions Training experience self taught friends-family school-club formal-professional others measured through self-
report validated
Outcomes Injuries and fatalities from a motorcycle crash (from hospital records and coronerrsquos office)
33Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3650
Hurt 1981 (Continued)
Notes Study conducted in Los Angeles USA
Jonah 1982
Methods Cohort study
Participants Initial sample size
Intervention group
2310 licensed MTP graduates who are active riders
Control group
2029 licensed motorcyclists who are active riders
Non-licensed and non-active riders excluded
Analysis sample size
Intervention group
811 interviews (35) 533 driver records (23)
Control group 1080 interviews (71) 905 driver records (45)
Interventions No training vs the non-mandatory Canada Safety Council Motorcycle Training Program (MTP) comprising of 4
hours theory and 16 hours practise (6 hrs on-road) over 2 weekends
Intervention integrity
Participants had all successfully completed the MTP
Outcomes Self-reported and police reported crashes and offencesLength of follow up 4 years
Notes Study conducted in Ontario Canada
Kloeden 1994
Methods Cohort study
Participants Initial sample size
Intervention group 2533 licence applicants
Control group Unclear
Sample of riders obtaining probationary licenceIntervention group 1625 (64)
Control group 5015 (unclear)
Non-active riders excluded from control group (~66)
Analysis sample size
Intervention group 631 (39)
Control group 318 (6)
Interventions No formal training vs the mandatory pre-licence Ridersafe course consisting of two levelsthe first is a training course
in the basics of motorcycle riding where a pass entitles the rider to obtain a learnerrsquos permit The second is a skill
test where a pass entitles the rider to obtain a probationary motorcycle licence The training course consists of two 4
hour training sessions on separate days involving instruction videos and actual riding on motorcycles The skill test
consists of a 4 hour preparation course of 74 minutes of lecturediscussion 38 minutes of video and 75 minutes of
34Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3750
Kloeden 1994 (Continued)
on-course riding which is the testing phase
Intervention integrity
Unclear what percentage in the Ridersafe group and the control group obtained their probationary licence halfway
and their full motorcycle licence in the end
Outcomes Police reported crashes
Length of follow up 1 year
Notes Study conducted in SA Australia
Kraus 1975
Methods Population based case-control study
Participants Cases All motorcycle riders with a confirmed medically treated motorcycle injury (including fatalities) from all
hospitals in study base and peripheral area hospitals and all deaths due to motorcycle crashes in the study base All
serious injuries (leading to hospitalisation) and fatalities likely to be included
Community controls A random sample of registered motorcycle owners in the same temporary and geographical
study base as the cases arose from
Initial sample size Cases 1273
Controls 738
Analysis sample size Cases 741 (58)
Controls 434 (59)
Interventions Motorcycle rider training (not specified) measured through self-report
Outcomes Injuries leading to hospitalisation fatalities
Length of follow up not specified
Notes Study conducted in Sacramento county USA
Leung 1987
Methods Non-randomised trial
Participants Initial sample sizeIntervention group unknown number of MRC graduates
Control group unknown number of licensed motorcyclists
Analysis sample size
Intervention group 2941
Control group 43094
Interventions No formal training vs non-mandatory Wisconsin motorcycle rider course a 22 hours off-street programme that
teaches the basic rules of safe motorcycling
Intervention integrity
50 in the MRC group and 100 in the control group obtained their motorcycle licence in the end
35Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3850
Leung 1987 (Continued)
Outcomes Police recorded crashes and offences
Length of follow up up to 3 years
Notes Study conducted in Wisconsin USA
MAIDS 2004
Methods Population based case-control study
Participants Cases Motorcycle riders who had a motorcycle crash Unclear whether all cases included
Community controls Motorcycle riders identified at petrol stations Unclear whether matched to cases
Initial sample size
Cases not reported
Controls not reported
Analysis sample size
Cases 921 (Italy 200 Netherlands 200 Spain 121 Germany 250 France 150)
Controls 923 (Italy 200 Netherlands 200 Spain123 Germany 250 France 150)
Interventions Pre-license rider training or additional training (non-compulsory)
Outcomes Crashes
Notes Study in five sampling areas located in five European countries (France Germany Netherlands Spain and Italy)
McDavid 1989
Methods Cohort study
Participants Sample size after first match
Intervention group 346 male training graduates
Control group 346 male motorcycle riders licensed in same year
Sample size after second match
Intervention group 139 (40)
Control group 139 (40)
Analysis sample size
Intervention group 139 (40)
Control group 139 (40)
Interventions No formal training vs the non-mandatory pre-licence British Columbia Safety Councilrsquos 37-hour motorcycle safety
programme
Intervention integrity
Participants had all completed and passed the training
Outcomes Police reported crashes injurious crashes
Length of follow up 1-6 years
36Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3950
McDavid 1989 (Continued)
Notes Study conducted in British Columbia Canada
Mortimer 1984
Methods Cohort study
Participants Initial sample size
Intervention group 608 MRC graduates
Control group 360 motorcycle riders
Response rates
Intervention group 360 (59)
Control group 324 (90)
Non-active riders excluded
Analysis sample size
Intervention group 213 (35)
Control group 303 (84)
Interventions No formal training vs the non-mandatory Motorcycle Safety Foundationrsquos Motorcycle Rider Course comprising of
8 hours theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-road practice
Intervention integrity
Participants had all completed and passed the MRC
Outcomes Self-reported crashes and offencesLength of follow up 1 year
Notes Study conducted in Illinois USA
Mortimer 1988
Methods Cohort study
Participants Initial sample size
Intervention group 2307 MRC graduates
Control group 556 active licensed motorcycle riders
Response ratesIntervention group 1298 (56)
Control group 500 (90)
Non-active riders excluded
Analysis sample size
Intervention group 913 (40)
Control group 500 (90)
Interventions No formal training vs the non-mandatory Motorcycle Safety Foundationrsquos Motorcycle Rider Course comprising of
8 hours theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-road practice
Intervention integrity
Participants had all completed and passed the MRC
37Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4050
Mortimer 1988 (Continued)
Outcomes Self reported crashes and offences
Length of follow up 1 year
Notes Study conducted in Illinois USA
Osga 1980
Methods Cohort study
Participants Initial sample sizeIntervention group 1163 MSF MRC participants invited 818 (74) responded
Control group 329 motorcycle riders responded to recruitment strategies
Non-active riders excluded from analysis
Analysis sample size
Intervention group 595 (73)
Control group 329 (100)
Interventions No formal training vs the non-mandatory Motorcycle Safety Foundationrsquos Motorcycle Rider Course (MRC) com-
prising of 8 hours theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-road
practice
Intervention integrity
Participants had all successfully completed the MRC
Outcomes Self-reported crashes police recorded crash reports
Length of follow up 6 months - 3 12 years
Notes Study conducted in South Dakota USA
Perrino 2002
Methods Cohort study
Participants Initial sample size
Intervention group 2144 (unclear) MRC graduates (motorcycle licence applicants)
Control group 2601 (unclear) regular licensing test graduates (motorcycle licence applicants)Analysis sample size
Intervention group 1070 (unclear)
Control group 867 (unclear)
Interventions No formal training vs the mandatory (for those younger than 18 years) pre-licence Maryland Motorcycle Rider
Course (MRC) consisting of 8 hours theory and 12 hours practice
Intervention integrity
No information given
Outcomes Self-reported and police recorded crashes and offences
Length of follow up 1 year
38Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4150
Perrino 2002 (Continued)
Notes Study conducted in Maryland USA
Raymond 1979
Methods Cohort study
Participants Initial sample size
Intervention group
258 volunteers mainly (but not all) novice riders
Control group
1104 provisional licence applicants
Analysis sample size
Intervention group 190 (74)
Control group 627 (57)
Interventions No formal training vs the non-mandatory RACACU (Royal Automobile ClubAuto-Cycle Union) training scheme
The training program consisted of either 24 one-hour sessions or 13 two-hour sessions Twelve hours of the training
focused on theory about basic operational procedures traffic laws and highway code The practical session focused
on basic operational skills riding on a training area and riding in traffic with an instructor A proficiency test was
administered at the end of the course
Intervention integrity
No information given
Outcomes Self-reported crashes (including minor non-injury non-damage and off-road crashes)
Length of follow up 1 year
Notes Study conducted in the UK
Satten 1980
Methods Cohort study
Participants Initial sample size
Intervention group interviewers kept interviewing until they had 100 riders selected from a population of 891
enrollees in the MSF MRC course (unknown how many were asked and how many refused to participate)Control group unknown number of registered motorcycle riders
Non-active riders excluded from analysis
Analysis sample size
Intervention group 71
Control group 100
Interventions No formal training vs the non-mandatory Motorcycle Safety Foundationrsquos (MSF) Motorcycle Rider Course (MRC)
comprising of 8 hours theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-
road practice
Intervention integrity
Participants had all successfully completed the MRC
39Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4250
Satten 1980 (Continued)
Outcomes Self-reported crashes injuries and offences
Length of follow up 3 years
Notes Study conducted in Illinois USA
Savolainen 2007
Methods Cohort study
Participants Initial sample size
Intervention group 4000 riders that had conducted the MSF MRC course (passed or failed)
Control group 4000 motorcycle owners or permit or endorsement holders from the Indiana Bureau of Motor
Vehicles database
Analysis sample size
Intervention group 739 (18)
Control group 588 (15)
Interventions Intervention integrity
No information given
Outcomes Probability of crash involvement
Length of follow up up to 5 years
Notes Study conducted in Indiana USA
Swaddiwudhipong 1998
Methods Randomised trial
Participants Initial sample size
Intervention group
all 16 villages of 3 sub-districts
(participants were practising motorcyclists licensed and unlicensed)
Control group
18 villages of 3 other sub-districtsAnalysis sample size
Intervention group 3 districts (100)
Control group 3 districts (100)
Interventions No rider education vs non-mandatory community based rider education consisting of theory on motorcycle injury
risk helmet use traffic laws vehicle regulations and traffic signs and a written and skill test leading to a licence
Intervention integrity
No information given
Outcomes Injuries fatalities
Length of follow up 2 years
40Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4350
Swaddiwudhipong 1998 (Continued)
Notes Study conducted in Tak Province rural Thailand
Waller 1992
Methods Cohort study
Participants Initial sample size
Intervention group 349 riders who succesfully completed the Department of Public Safety motorcycle operator
training in 1988
Control group 349 motorcyclists who never completed the DPS motorcycle operator training
Analysis sample size
Intervention group 349
Control group 349
Interventions No DPS motorcycle operator training vs non-mandatory DPS motorcycle operator training The DPS training is
the Motorcycle Safety Foundationrsquos Motorcycle Rider Course comprising eight hours of theory (lectures discussions
films and multiple choice tests) a textbook and 12 hours off-road practice
Intervention integrity
All 349 riders succesfully completed the rider training
Outcomes Police recorded crashes
Length of follow up 3 years
Notes Study conducted in
Characteristics of excluded studies [ordered by study ID]
Study Reason for exclusion
Anderson 1980a Study type review and process evaluation of a rider course
Aupetit 2008 Outcome proposal of a rider training curriculum that combines traditional training and a motorcycle riding
simulator
Awane 1999 Study type review on use of simulators in rider and driver training
Baldi 2005 Intervention not rider training but intervention integrity of rider training availabilty of rider education
implementation of graduated licensing system etc on a state-wide level
Batchler 1988 Study type description of a training course
Blanchard 2006 Study type not a study assessing the association between rider training and outcomes
Bonnett 2005 Outcome qualitative data regarding satisfaction with the training programme
41Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4450
(Continued)
Braver 2007 Outcome licensure rates
Clayton 1980 Outcome knowledge skills attitude
Council 1978 Outcome crash reduction in ANY vehicle not motorcycles exclusively
Crowther 2005 Study type qualitative study
Daniello 2009 Study type review of the effectiveness of motorcycle training and licensing
Dewanti 2007 Intervention emotional characteristics behaviour (ie not motocycle rider training)
Du Plooy 1990 Study type review of types of rider training courses in South Africa
Engel 1989 Study type uncontrolled before-after study
Goldenbeld 2002 Outcome skills
Goldenbeld 2004 Outcome knowledge and skills
Haworth 2000 Intervention balance between attitudinal andvehicle control skills acrosscurriculla of motorcycle rider training
courses
Haworth 2005 Study type review
Haworth 2006 Study type cross-sectional survey
Hill 1983 Study type review and description of training system
Imatake 1980 Study type course descriptionreview
Ingebritsen 1990 Intervention risk factors for motorcycle crashes
Jain 2009 Study type cross-sectional descriptive study of motorcycle crashes over a 5 year period
Jonah 1979 Study type cross-sectional survey Outcome scores on motorcycle operator skill test
Jonah 1981 Intervention evaluation of a motorcycle rider skill test not motorcycle rider training
Kadar 1990 Study type review of components of rider training courses
Kelsey 1986 Intervention evaluation of a motorcycle rider skill test not motorcycle rider training
Koch 1980 Study type reviewcourse description
Koch 1990 Study type description of course content
42Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4550
(Continued)
Labbett 2005 Study type essay about possible explanations for motorcycle deaths
Law 2005 Intervention targeted motorcycle safety programme not specifically rider training
Lowes 1990 Study type review and descriptions of rider training
Mahon 1976 Study type review of motorcycle rider training methods
Maxwell 1985 Outcomesrecommendations forprogramme operatorsand otherson howthe motorcycle trainingprogramme
works and how it can be improved
Mayhew 1996 Study type review
McKnight 1987 Study type evaluation of methodological and administrative aspects of a motorcycle operator skill test
McPherson 1976 Outcome knowledge and skills
Mohan 2004 Study type review of evidence of interventions to reduce road traffic injuries
Morgan 1980 Study type reviewcourse description
Motoki 1990 Study type proposal and development of rider training curriculum
Nairn 1992 Study type literature review
Newland 1980 Study type description of development of motorcycle safety course
Newland 1982 Study type descriptionreview of a rider training course
Newland 1990 Study type uncontrolled BeforeAfter study
Ormston 2003 Outcome changes in attitude and self reported behaviour
Osga 1979 Outcome knowledge attitudes and skills
Panou 2007 Intervention new simulation tool to teach driverrider
Peterson 1990 Study factor demographic details of course participants completing the course vs those not completing the
course
Prem 1984 Study type experimental
Study factor examines association between skills and performance in tests
Radin Umar 2006 Intervention motorcycle safety programme not specifically rider training
Rockwell 1990 Study type methodologically flawed study type
43Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4650
(Continued)
Rothe 1987 Study type qualitative study of rider training in general
Rowden 2007 Study type qualitative study
RSC Victoria 1993 Study type review of motorcycle rider training and education in Victoria
Russam 1979 Study type review of training programmes in Japan
Saffron 1981 Study type literature review
Schulz 1990 Study factor predictors of motorcycle crashes
Sexton 2004 Study type cross-sectional analytic analysis
Shephard 1986 Study factor reviewdescription of programme
Simpson 1990 Study type review of motorcycle rider trainingeducation
Sokolov 1990 Study factor review of motorcycle riding courses in USSR
Steffens 1988 Outcome self-reported behaviour experiences and researcher observed riding behaviour
Sudlow 2003 Study type review of motorcycle training in the UK
Tasmania 1995 Study type cross-sectional descriptiveprocess evaluation
Tetard 1985 Study type experimental
Outcome detection of indices
Thackray 1980 Outcome instructional effectiveness user acceptance administrative feasibility
Thompson 1994 Study type cross-sectional survey
Outcome attitudes and knowledge
Tobler 1990 Study type review of course curricula
Transport 1989 Intervention syllabus for rider training
Tuan 2007 Study type cross-sectional study (survey) which does not assess the effects of an intervention but examines
motocyclistsrsquo behaviour
Victor 1980 Study type review
Waterer 1990 Study type review of UK rider training issues
Weller 1989 Study type review of studies evaluating motorcycle safety and education measures
44Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4750
(Continued)
Wells 1981 Study type description of different methods of training
Winn 1986 Study type review of rsquohistoryrsquo of rider training
Wisher 1988 Study type cross-sectional survey
Wood 1987 Study type Interrupted Time Series but did not meet the criteria for an ITS as published by the Cochrane
Effective Practice and Organisation of Care Group
Yamazaki 1990 Study type cross-sectional survey
Outcome riding skills skills retention and attitude
Characteristics of ongoing studies [ordered by study ID]
Burgess 2010
Trial name or title UKs National Ride program
Methods Cohort study
Participants Motorcycle offenders
Interventions Police led diversion program
Outcomes Self reported attitudinal and behavioural measures
Starting date
Contact information
Notes Expected to be completed in early 2010
45Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4850
D A T A A N D A N A L Y S E S
This review has no analyses
A P P E N D I C E S
Appendix 1 Search strategy
We searched the following electronic databases
bull Cochrane Injuries Group Specialised Register (searched 22 Sept 2008)
bull CENTRAL (The Cochrane Library 2008 Issue 3)
bull TRANSPORT 1988 to 200706 (latest date available)
bull ATRI Australian Transport Index (to Sept 2008)
bull MEDLINE (Ovid SP) 1950 to Sept 2008
bull EMBASE (Ovid SP) 1980 to Sept 2008
bull CINAHL (EBSCO) 1982 to Sept 2008
bull WHOLIS (World Health Organization Library Information System) (to Sept 2008)
bull PsycINFO (Ovid SP) 1806 to Sept 2008
bull LILACS (BIREME) (Latin American and Caribbean Health Sciences) (1982 to Sept 2008)
bull ISI Web of Science Social Sciences Citation Index (SSCI) 1970 to Sept 2008
bull ERIC (searched 25 Sept 2008)
bull ZETOC (searched 25 Sept 2008)
bull SIGLE (searched 25 Sept 2008)
All of the databases were searched using the following text-word search strategy The strategy was adapted using either rsquorsquo or rsquo$rsquo as atruncation symbol
1 motor-cycl or motorcycl or motor-bik or motorbik or motor-scooter or motorscooter or scooter or moped or motocycl
or motocicl or moto
2 train or educat or instruct or program or course or qualif or learn or improv
3 1 and 2
H I S T O R Y
Protocol first published Issue 2 2005
Review first published Issue 10 2010
46Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4950
C O N T R I B U T I O N S O F A U T H O R S
Katina Kardamanidis was the first author of the protocol conducted the literature search extracted full studies assessed the studies
according to inclusion criteria assessed quality of the studies compared findings with the second author discussed disagreements with
the third author wrote the drafts and the final review
Rebecca Ivers was the second author of the protocol assessed the studies according to inclusion criteria assessed quality of the studies
compared findings with the first author edited the final draft and provided overall guidance for the process
Alexandra Martiniuk assessed quality of the studies compared findings with the first author discussed disagreements with the third
author and helped write the first draft and the final review
Mark Stevenson contributed to the protocol reviewed methods of studies and contributed to the final draft
Katrina Thistlethwaite conducted a second literature search extracted full studies and contributed to the final draft
D E C L A R A T I O N S O F I N T E R E S T
None known
S O U R C E S O F S U P P O R T
Internal sources
bull No sources of support supplied
External sources
bull National Health and Medical Research Council of Australia Australia
Salary support for Martiniuk Ivers and Stevenson
bull University of Sydney Sesqui Research Grant Australia
Direct research costs and salary support for Kardamanidis
D I F F E R E N C E S B E T W E E N P R O T O C O L A N D R E V I E W
There are no differences between the protocol and the review
I N D E X T E R M S
Medical Subject Headings (MeSH)
lowastMotorcycles Accident Prevention [lowastmethods] Accidents Traffic [lowastprevention amp control] Licensure Program Evaluation Wounds
and Injuries [mortality lowastprevention amp control]
47Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 5050
MeSH check words
Humans
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 650
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 750
Table 1 Risk of selection bias (Continued)
Osga 1980 Differences at baseline reported on sex and age
Confounding factors were not dealt with in univariate analysis Riding exposure was accounted for in the
cumulative survival analysis
Selection intervention group were all training participants on the register and the control group were
respondents to a diverse range of recruitment strategies
Satten 1980 Differences at baseline reported on age sex helmet use riding experience and exposure and previous
training
Confounding factors were not dealt with in design or analysis
Selection intervention group was selected from a list of training graduates and the control group was
partly selected from a state register of motorcyclists
Jonah 1982 Differences at baseline reported on age sex riding experience time licensed previous training and alcohol
use
Confounding factors were dealt with through multivariate analysis
Selection intervention group was selected from a list of training graduates and the control group was
randomly selected from the Ontario motorcycle operator file
Mortimer 1984 Differences reported at baseline on age licence status riding experience years licensed and helmet use
Confounding factors age years licensed and rider experience were dealt with through stratification and
riding exposure through standardisation
Selection intervention group was selected from a list of training enrolees and the control group was
selected from motorcycle dealerships during three successive Saturdays
Cooper 1988 Differences reported at baseline on age gender licensure and highway experience
Confounding factors age gender riding exposure rider characteristics rider attitudes and behaviour were
accounted for in a multivariate analysis
Selection there was no information on how the intervention and control groups were selected and from
where
Mortimer 1988 Differences reported at baseline on years licensed riding experience and helmet use
Confounding factors age years licensed and rider experience were dealt with through stratification and
standardisation of riding exposure A regression analysis was also conducted but poorly reported and not
usable
Selection intervention group was selected from a list of training enrolees and the control group was
selected from motorcycle dealerships
Adams 1985 Differences at baseline reported on previous crash rates
Confounding factors not controlled for in analysis
Selection the intervention group was a non-random sample of those offered training within the Australia
Post motorcycle delivery riders the control group were the remaining Post riders
Leung 1987 Differences at baseline reported on age and sex
Confounding factors age and sex controlled for in analysis by stratification
Selection those exposed to training were MRC graduates those unexposed were taken from the licensing
records
5Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 850
Table 1 Risk of selection bias (Continued)
McDavid 1989 Differences at baseline other than matched characteristics were not reported
Confounding factors age ownership of car licence years licensed previous crashesand offences accounted
for through matching Other confounding effects not controlled
Selection intervention group was selected from a list of training graduates and the control group was
selected from Bureau of Motor Vehicle records
Kloeden 1994 Differences at baseline reported were in geographical area of residence and in riding exposure
Confounding factors were accounted for through exclusion of participants from rural areas
Other confounding factors were not reported
Selection the Ridersafe course was phased in based on postcode of riderrsquos home address analyses were
based on a database of those who obtained motorcycle rider learnerrsquos permits of which about 2000 were
Ridersafe riders and about 5000 were control riders
Billheimer 1998 Differences at baseline reported on age and previous riding experience
Other potential confounding factors assessed were riding exposure primary purpose of riding and helmet
use
Confounding factors accounted for by matching (age sex primary purpose of riding) standardisation
(riding exposure) and stratification (previous riding experience)
Selection the intervention group were those who had taken the rider training programme the control
group were California motorcycle riders matched on the above variables but who had not taken a riding
course
Perrino 2002 Differences at baseline reported were a higher number of females in the trained group and a higher
proportion of experienced riders in the untrained group
Confounding factors were not accounted for in the analysis
Selection the intervention group were those who had done the Motorcycle Riding Course the control
group had not done the course Both groups were identified through a register of licensed motorcyclists
Kraus 1975 Population based case-control study
Cases all motorcycle riders with a confirmed medically treated motorcycle injury (including fatalities)
from all hospitals in study base and peripheral area hospitals and all deaths due to motorcycle crashes in
the study base All serious injuries (leading to hospitalisation) and fatalities likely to be included
Community controls a random sample of registered motorcycle owners in the study base
Potential confounders assessed were age sex weight height riding experience motorcycle use charac-
teristics of motorcycle and previous crashes and offences
Confounding factors were accounted for through exclusion of females and stratification by age and
motorcycle useSelection cases and controls drawn from the same population
Hurt 1981 Population based case-control study
Cases and controls not drawn from the same population
Cases motorcycle riders who had a motorcycle crash Not all reported cases included
Community controls motorcycle riders who passed the crash site at the same time of day and day of week
that the crash occurred but 2 years later
Potential confounders assessed were age sex time type of motorcycle roadway alignment helmet use
weather conditions road surface conditions
Confounding factors were accounted for through matching for time of day and day of week that crash
occurred and stratification
Selection cases were recruited from police reports not all cases were included controls were matched
6Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 950
Table 1 Risk of selection bias (Continued)
community controls from randomly selected sites matched for same time of dayday of week as the crash
occurred
Haworth 1997 Population based case-control study
Cases and controls drawn from the same population
Cases motorcycle riders or pillion riders with an injury from a motorcycle crash from hospital and police
records Not clear whether all cases were included
Community controls motorcycle rider or pillion who passed the crash site at the same time of day and
day of week that the crash occurred
Potential confounders assessed were rider factors age marital status licence status riding experience
and exposure alcohol and drugs use of protective gear including helmet use pillion factors motorcycle
factors trip factorsConfounding factors were accounted for through matching for time of day and day of week that crash
occurred and stratification for age blood alcohol level and licence status
Selectioncases were riders or pillions involved in on-road crash recruited from hospitals and police records
not sure if all cases were included controls were matched community controls who passed the crash site
at the same time of dayday of week as the crash
Savolainen 2007 Differences at baseline reported on sex safety helmet use riding ability riding experience and licence
status
Adjusted for these confounders through statistical modelling although exact process unclear
Selection intervention group from list of trainees control group from Indiana Bureau of Motor Vehicles
database Trainees and the control group were invited to participate in surveys
Davis 1997 Differences at baseline not reported but likely No adjustment for any confounders at baseline or during analysis
Selection intervention group from list of riders involved in crashes who were also on a list of training
graduates control group from the same list of riders involved in crashes who were not also on a list of
training graduates
Waller 1992 Differences at baseline not reported but likely
Adjusted for confounders through exclusion of riders younger than 18 years riders without a motorcycle
operator driver license riders with location of residence outside of the Texas metropolitan areas of Austin
through matching by age sex and location of residence through stratification by age and sex and through
standardisation (number of crashes by person-days at risk)
Selection intervention group from the Motorcycle Operator Training class roster (number of trained
riders that the sample was selected from is unclear) control group through a random number generator
generating driver license numbers until an equal sample size was selected as the intervention group
MAIDS 2004 Population based case-control study
Unclear whether cases and controls not drawn from the same population
Cases motorcycle riders who had a motorcycle crash Unclear whether all cases included
Community controls motorcycle riders recruited at petrol stations Unclear whether and how matched
to cases
No potential confounders were reported
There was no report of whether and how potentially confounding factors were accounted for in the study
7Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 1050
Table 2 Risk of allocation bias
Anderson 1980 Allocation concealment unclear
Hall 1988 Allocation concealment unclear
Swaddiwudhipong 1998 Allocation concealment unclear
Table 3 Risk of detection bias
Anderson 1980 Blinding of outcome assessors not reported
Driver records holding police recorded crashes and offences not validated
Hall 1988 Blinding of outcome assessors not reported
Driver records holding police recorded crashes and offences not validated
Swaddiwudhipong 1998 Blinding of outcome assessors not reported
Police and hospital records Post mortem examination for deaths not validated
Adams 1985 Blinding of outcome assessors not reported
Australia Post crash records not validated but argued that underreporting is unlikely
Leung 1987 Blinding of outcome assessors not reported
Driver records holding police recorded crashes and offences
Raymond 1979 Blinding of outcome assessors not reported
Self-report through personal interviews or questionnaires - both methods validated
Osga 1980 Blinding of outcome assessors not reported
Self-report through structured questionnaires validated by cross-checking with police
recorded crash records
Satten 1980 Blinding of outcome assessors not reported
Self-report through structured interviews or questionnaires - both methods were not
validated
Intervention group were given telephone interviews but the control group mailed ques-
tionnaires
Jonah 1982 Blinding of outcome assessors not reported
Self-report through structured interviews not validated Driver records holding police
recorded crashes and offences
Mortimer 1984 Blinding of outcome assessors not reported
Self-report through questionnaires not validated
Intervention group received mailed questionnaires but the control group received ques-
tionnaires personally at the motorcycle dealership
Cooper 1988 Blinding of outcome assessors not reported
Driver records holding police recorded crashes and offences
8Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 1150
Table 3 Risk of detection bias (Continued)
Mortimer 1988 Blinding of outcome assessors not reported
Self-report through questionnaires not validated
Intervention group received mailed questionnaires but the control group received ques-
tionnaires personally at the motorcycle dealership
McDavid 1989 Blinding of outcome assessors not reported
Driver records holding police recorded crashes and offences
Kloeden 1994 Blinding of outcome assessors not reported
Driver records holding police recorded crashes and offences
Billheimer 1998 Blinding of outcome assessors not reported
Police recorded crashes and offences
Perrino 2002 Blinding of outcome assessors not reported
Self-report through mailed out questionnaires police records not validated
Kraus 1975 Blinding of intervention assessors to case or control status not reported
Intervention measured police records self-report though questionnaires not validated
Length of follow up not specified
Hurt 1981 Blinding of intervention assessors to case or control status not reported
Intervention measured face-to-face in-depth interviews (ie no structured question-
naires) Validity and reliability of data ensured by reported high level of quality controlLength of follow up not specified
Haworth 1997 Blinding of intervention assessors to case or control status not reported
Intervention measured self-report either by personal or by telephone interview for
both cases and controls not validated
Length of follow up in the 2 years prior to the interview or more than 2 years ago
Savolainen 2007 Blinding of outcome assessors not reported
Source of outcome data not reported
Length of follow up up to 5 years
Davis 1997 Blinding of outcome assessors not reported
Source of outcome data police records
Length of follow up 1 year
Waller 1992 Blinding of outcome assessors not reported
Source of outcome data Department of Public Safety driver history file and official
accident reports
Length of follow up 3 years
MAIDS 2004 Blinding of intervention assessors to case or control status not reported
Intervention measured face-to-face in-depth interviews (ie no structured question-
naires) No report on whether and how validity and reliability of data was ensured
Length of follow up 2 years
9Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 1250
Table 4 Risk of attrition bias
Anderson 1980 Percentage of participants in analysis 99 forthe total group Unclear forintervention andcontrol groups
separately
Hall 1988 Percentage of participants in analysis 96-97 forall intervention groups and 98for the control group
Swaddiwudhipong 1998 Percentage of participating sub-districts in analysis 100 for both intervention and control sub-districts
Raymond 1979 Percentage of participants in analysis 74 for the intervention group and less than 57 for the control
group
Osga 1980 Percentage of participants in analysis 73 for the intervention group and 100 for the control group
Satten 1980 Percentage of participants in analysis is unknown Initial sample size is unclear
Jonah 1982 Percentage of participants in analysis 23 for the intervention group and 45 for the control group
Mortimer 1984 Percentage of participants in analysis 35 for the intervention group and 84 for the control group
Cooper 1988 Percentage of participants in analysis 98 for the total group Unclear for intervention and control
groups separately
Mortimer 1988 Percentage of participants in analysis 40 for the intervention group and 90 in the control group
Adams 1985 Percentage of participants in analysis 61 forthe total group Unclear forintervention andcontrol groups
separately
Leung 1987 Percentage of participants in analysis unclear because initial sample size was not reported
McDavid 1989 Percentage of participants in analysis 40 of final matched cohort
Kloeden 1994 Percentage of participants in analysis 39 for the intervention group and 6 for the control group
Billheimer 1998 Percentage of participants in analysis unclear percentage of the initial sample size
Perrino 2002 Percentage of participants in analysis unclear percentage of the initial sample size
Kraus 1975 Percentage of participants in analysis 58 of cases 59 of controls
Hurt 1981 Percentage of participants in analysis 18 of cases lt 29 of controls
Haworth 1997 Percentage of participants in analysis 70 of cases 47 of controls
Savolainen 2007 Percentage of participants in analysis 18 for the intervention group and 15 for the control group
Davis 1997 Percentage of participants in analysis 100 for the intervention group and unclear for the control group
10Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 1350
Table 4 Risk of attrition bias (Continued)
Waller 1992 Percentage of participants in analysis unclear percentage for the intervention group and 100 for the
control group
MAIDS 2004 Percentrage of participants in analysis unclear for both cases and controls
R E S U L T S
Description of studies
See Characteristicsof included studies Characteristics of excluded
studies Characteristics of ongoing studies
Our search resulted in 102 studies that seemed relevant of which
we excluded 79 studies because they did not meet the criteria for
considering studies for this review (see Characteristics of excluded
studies)
Basedon ourcriteriawe includeda total of 23 studies in this review
There were three randomised trials (Anderson 1980 Hall 1988
Swaddiwudhipong 1998) two non-randomised trials (Adams
1985 Leung 1987) 14 cohort studies (Raymond 1979 Osga
1980 Satten 1980 Jonah 1982 Mortimer 1984 Cooper 1988
Mortimer 1988 McDavid 1989 Waller 1992 Kloeden 1994
Davis 1997 Billheimer 1998 Perrino 2002 Savolainen 2007)
and four case-control studies (Kraus 1975 Hurt 1981 Haworth
1997 MAIDS 2004)
Over half of the studies were conducted 20 or more years
ago and the oldest was conducted 31 years ago Only three
studies were from this decade (Perrino 2002 MAIDS 2004
Savolainen 2007) Over 60 of the studies were conducted in
the USA (Kraus 1975 Anderson 1980 Osga 1980 Satten 1980
Hurt 1981 Mortimer 1984 Leung 1987 Hall 1988 Mortimer
1988 Waller 1992 Davis 1997 Billheimer 1998 Perrino 2002
Savolainen2007) three inAustralia (Adams 1985 Haworth 1997
Kloeden 1994) three in Canada (Jonah 1982 Cooper 1988
McDavid 1989) onefromthe United Kingdom (Raymond 1979)one from Europe (with sampling areas located in France Ger-
many Netherlands Spain and Italy) (MAIDS 2004) and one
in Thailand (Swaddiwudhipong 1998) Only seven studies were
published in peer reviewed journals (Kraus 1975 Jonah 1982
Mortimer 1984 Cooper 1988 Mortimer 1988 McDavid 1989
Swaddiwudhipong 1998) The other studies were reports forhigh-
way safety and road traffic organisations
The studies by Anderson 1980 Hall 1988 Kloeden 1994
Billheimer 1998 (the basic California Motorcyclist Safety Program
component of the study) and Perrino 2002 examined manda-
tory pre-licence training Cooper 1988 McDavid 1989 and
Swaddiwudhipong 1998 assessed non-mandatory training that
when passed resulted in a licence Raymond 1979 Osga 1980
Satten 1980 Jonah 1982 Mortimer 1984 Adams 1985 Leung
1987 Mortimer 1988 Waller 1992 Davis 1997 Billheimer 1998
(the Experienced Rider Course component of the study) andSavolainen 2007 assessed non-mandatory training that could be
taken by motorcycle riders with and without a licence The case-
control studies (Kraus 1975 Hurt 1981 Haworth 1997) assessed
lsquoanyrsquo type of rider training and therefore could not be grouped by
pre- or post-licence rider training The MAIDS 2004 case-control
study did make a distinction between participants having received
compulsory pre-licence training or additional (non-compulsory)
training We did not find a study that compared one form of pre-
or post-licence training (eg on-road off-road theory or a com-
bination) versus another form of training All studies compared
eithermandatory or non-mandatory formal training versus no for-
mal training with the exception of Haworth 1997 who compared
different levels of training courses (beginner intermediate and ad-vanced)
We have presented a more detailed description of the individual
studies in Characteristics of included studies We have provided
a description below of the types of motorcycle rider training that
were evaluated in the included studies
Types of motorcycle rider training evaluated in the
studies
The types of training were varied Duration of training ranged
from two or three hours to 24 hours administered over two or
three days Most training consisted of a theoretical and a practical
component usually off-road The content ranged from provid-ing riders with manuals lectures discussions showing films and
administering knowledge and skill tests Little information was
provided about the qualifications of trainers who conducted the
training
Six studies (Osga 1980 Satten 1980 Mortimer 1984 Mortimer
1988 Waller 1992 Savolainen 2007) assessed the effectiveness of
the USA-based Motorcycle Safety Foundationrsquos Motorcycle Rider
Course (MSF MRC) This course was released in 1976 It com-
prises eight hours of theory (lectures discussions films and multi-
plechoice tests) a textbook and 12 hours of off-road practice The
Motorcycle Safety Foundation is a national not-for-profit organ-
11Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 1450
isation sponsored by manufacturers and distributors of motorcy-
cles Many USA states have adopted the MSF Basic Rider Courseand Experienced Rider Course (wwwmsf-usaorg) since its release
in 1976
The Motorcycle Operators Skill Test was also developed
by the USA Motorcycle Safety Foundation in coopera-
tion with others (httpwwwwebbikeworldcommotorcycle-
trainingmotorcycle-test) The MOST is a licensing test consist-
ing of a manual knowledge and skill test (comprising eight exer-
cises of increasing difficulty and skill requirements) and a mini-
skill test conducted with a simulator Some studies compared this
alternative licensing test to the standard licensing test However
as the MOST is not a type of rider training we have included only
the two studies that assessed the MOST in combination with one
or more training components in this review Both Anderson 1980and Hall 1988 assessed the MOST in combination with a train-
ing component of three hoursrsquo duration consisting of instructions
about the skills needed to pass the MOST (30 minutes of the-
ory and two and a half hours on the motorcycle) Hall 1988 also
assessed the MOST in combination with the Motorcycle Safety
Foundationrsquos Motorcycle Rider Course as described above
Other USA studies evaluated the Wisconsin motorcycle rider
course a 22-hour off-street programme that teachesthe basic rules
of safe motorcycling (Leung 1987) the USA Maryland Motorcy-
cle Rider Course (MRC) a mandatory (for those younger than 18
years)pre-licence 20-hour coursecomprising eight hours of theory
and12 hours of practice and created by the Maryland Motor Vehi-
cle Administration in 1985 (Perrino 2002) the California Motor-cyclist Safety Program (CMSP) a mandatory (for those younger
than 18 years) pre-licence 16-hour basic course (Billheimer 1998)
and Experienced Rider Course (ERC) of eight hoursrsquo duration
(Billheimer 1998) The CMSP still offers these courses ( http
wwwca-msporg) but they are now MSF courses as described
above Davis 1997 did not give any information about the assessed
rider training
The Canadian studies evaluated the British Colombia Safety
Councilrsquos 37-hour motorcycle safety programme (no further in-
formation on the content was provided) (McDavid 1989) and
the Canada Safety Council Motorcycle Training Program (MTP)
comprising four hours of theory and 16 hours of practice (six
hours on-road) over twoweekends (Jonah 1982) The MTPcoursewas made available in 1974 and is still being offered in Canada
(httpwwwsafety-councilorgtrainingmtpmtphtm) Cooper
1988 did not give any information about the assessed rider train-
ing However the study was conducted in British Colombia in
Canada in 1988 the same time as the McDavid 1989 study was
conducted
In the United Kingdom Raymond 1979 assessed the non-manda-
tory RACACU (Royal Automobile ClubAuto-Cycle Union)
training scheme The training programme consisted of either 24
one-hour sessions or 13 two-hour sessions Twelve hours of the
training focused on theory about basic operational procedures
traffic laws and highway codes The practical session focused on
basic operational skills riding on a training area and riding in traf-fic with an instructor A proficiency test was administered at the
end of the course
One study from rural Thailand (Swaddiwudhipong 1998) evalu-
ated a community-based rider education programme consisting of
theory on motorcycle injury risk helmet use traffic laws vehicle
regulations and traffic signs and a written and skill test leading to
a licence
TheAustralian studies evaluateda rider trainingcourse of 12 hours
over two consecutive days (four hours of theory and eight hours of
practice)administeredto AustraliaPosttelegram deliverers(Adams
1985) and a mandatory pre-licence Ridersafe course consisting
of two tiers the first tier being a training course on the basics
of motorcycle riding where a pass entitled the rider to obtain alearnerrsquos permit and the second tier being a skill test where a pass
entitled the riderto obtain a probationary motorcycle licence The
training course consisted of two four-hour training sessions on
separate days involving instruction videos and actual riding on
motorcycles The skill test consisted of a four-hour preparation
course of 74 minutes of lecturediscussion 38 minutes of video
and 75 minutes of on-course riding which was the testing phase
(Kloeden 1994) The Rider Safe course is still being provided by
Transport South Australia a state government organisation (http
wwwtransportsagovau )
The duration or content of rider training in the four case-control
studies (Kraus 1975 Hurt 1981 Haworth 1997 MAIDS 2004)
was not described
Risk of bias in included studies
We describe the methodological quality of the included studies
by study type below More detailed information can be found in
Table 1 Table 2 Table 3 and Table 4
When considering the findings from these studies it is important
to note the quality (or lack thereof) of reporting of the studies
Quality of reporting is the extent to which a report provides in-
formation about the design conduct and analysis of a study ( Juumlni2001) Huwiler-Muumlntener 2002 found that even though trials
may be well reported it does not mean that the methodological
quality is of equal standard and well-conducted trials may never-
theless be reported badly The reporting quality of most studies in
this review was poor Most studies were old and did not meet con-
temporary reporting standards This made data extraction com-
plicated It also made it difficult to contact authors for additional
information
Because of significant heterogeneity between studies as well as the
poor quality of most studiespooling of studyresults to calculatean
overall measure of effect was not appropriate Instead we provided
a description of our judgement of risk of bias in the text
12Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 1550
Risk of bias in the randomised trials
There were three randomised trials (Anderson 1980 Hall 1988Swaddiwudhipong 1998) Anderson 1980 and Hall 1988 had an
initial sample size of 41188 and 26464 motorcycle licence ap-
plicants respectively Both studies assessed the effectiveness of the
pre-licence MOST with a trainingcomponent Swaddiwudhipong
1998 evaluated the effectiveness of a community-based rider ed-
ucation programme in six districts
Allocation bias
Anderson 1980 reported using random birth date tables to allocate
participants to either the intervention or control group Hall 1988
reported the use of a computerised algorithm set in a central of-
fice but also used manual random assignment when the computer
went down Swaddiwudhipong 1998 reported lsquosimple random
samplingrsquo for the delivery of the intervention where the sampling
frame included nine districts Three districts were excluded since
they were too hard to access The remaining six districts were ran-
domly allocated to the intervention (three) and the control (three)
groups Systematic sampling wasthen used ordered by village size
for the villages within districts Allocation concealment was not
reported in any of the studies The descriptions of randomisation
in these studies suggest that there may have been limitations to
the process resulting in a lack of equivalence of the intervention
and control groups at baseline and therefore the possibility of
allocation bias Swaddiwudhipong 1998 reported that before theintervention the annual incidence rate of motorcycle-related in-
juries was slightly (but not statistically significant) higher in the
intervention areas than in the control areas Participants in the
intervention districts were also more likely to have a valid licence
and to wear a helmet more often than participants in the control
districts Anderson 1980 measured differences at baseline on the
participant and prior driving record variables and found these to
be negligible Hall 1988 did not report on baseline differences
between the intervention and control groups As the Anderson
1980 and Hall 1988 studies had sample sizes of over 40000 and
26000 respectively and used some form of randomisation it is
likely that any confounding factors at baseline would have been
equally distributed between comparison and intervention groups
Selection bias
After randomisation Anderson 1980 and Hall 1988 excluded
08 and 3 participants respectively because of assignment to
the wrong group administration of the wrong intervention or
other reasons The resulting lack of analysis by rsquointention-to-treatrsquo
can introduce biasby removing the effectivenessof randomisation
Nonetheless the percentage excluded was small and the number
of overall participants large so it is unlikely to have had a major
effect on the final results In the Swaddiwudhipong 1998 study
villages were systematically sampled into intervention or control
groups The intervention was implemented and outcomes of in-
jury and death were measured at a community level Interventionand control districts were randomly selected however there were
only three intervention and three control communities
Detection bias
None of the randomised trials reported blinding of outcome as-
sessors Data collection tools used by Anderson 1980 and Hall
1988 were driver records holding police recorded crashes and of-
fences Swaddiwudhipong 1998 used police and hospital records
at a district level and post-mortem examination for deaths None
reported on validity or reliability for any of the instruments
Attrition bias
Anderson 1980 and Hall 1988 had driver records available for
96 to 100 of participants They also collected self-reported
data but the response rate to the questionnaires was very low (low-
est response rate 42 and 30 respectively) Attrition bias in
Swaddiwudhipong 1998 was not an issue as outcome data were
available for all sub-districts included in the trial
The intervention reported in Anderson 1980 and Hall 1988 tri-
als had an unanticipated effect on the participants In both trials
an average of only 28 of licence applicants in the intervention
group eventually obtained their licence compared to 35 in the
control group This indicates that the licensing process in generalseems to have deterred applicants from obtaining a licence and
although no statistical tests of the difference were conducted this
deterrence appears even greater for those who had rider training
This is not true attrition bias as the studies did not experience
differential loss to follow up from outcome measurement Instead
differential attainment of the intermediary step of licensure pre-
cluded participants from being at risk for the outcomes of interest
they are not driving a motorcycle hence they are not at risk for
motorcycle-related crashes injury death or offences In short it is
not possible to assess whether a reduction in crashes in the inter-
vention group would be due to changes in riding habits as a result
of training or simply due to delayed or never attained licensure
Intervention integrity
Neither Anderson 1980 nor Hall 1988 reported on the consis-
tency in the way the intervention was administered the possible
lsquocontaminationrsquo of participants in the intervention group by other
types of formal or informal rider training or of participants in the
control group by the intervention Swaddiwudhipong 1998 did
not report on the number of villagers in each group that actually
attended the intervention However they did report on a number
of other potential training courses that were being implemented
at the same time
13Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 1650
Analysis
Anderson 1980 and Hall 1988 acknowledging the lsquodeterrent ef-fectrsquo described above conducted two separate analyses The first
analysis included the total study population of licence applicants
over two years from the time of their application (the population
denominator) The second analysis was performed on the licensed
(intervention and control) subgroups only two years from the
time of obtaining their licence (the licensed rider denominator)
As only a third were licensed and exposed to riding the motorcy-
cle it was impossible to determine whether reported differences
in crash rates across groups were due to the intervention or to
other lsquoconfoundingrsquo characteristics of those who were determined
in their efforts to obtain the licence When analysis was restricted
only to those who gained their licence the same difficulties re-
mained The benefits of the randomisation process were thus lost
due to this lack of licensure of riders during the process As there
were only three intervention and three control groups in the study
reported by Swaddiwudhipong 1998 analyses were hindered by
smallnumbers for outcome comparisons betweenthe intervention
and the control districts
Risk of bias in the cohort studies and non-randomised
trials
We discuss the methodological quality of cohort studies (seven
prospective studies (Raymond 1979 McDavid 1989 Waller 1992
Kloeden 1994 Billheimer 1998 Perrino 2002 Savolainen 2007)and seven retrospective studies (Osga 1980 Satten 1980 Jonah
1982 Mortimer 1984 Cooper 1988 Mortimer 1988 Davis
1997)) and non-randomised trials (Adams 1985 Leung 1987) in
the one section as the biases for the two study types are similar
The quality of these studies was in general poor
Selection bias
The confounding factors expected to have been addressed in the
studies were sex age riding experience prior driving records (of-
fences crashes) previous training and the following proxies for
risk taking behaviour helmet use speed and alcoholdrug use All
of these factors have been shown to be associated with an increasedrisk of motorcycle crash (Peden 2004 Zambon 2006) The pos-
sible source of bias in cohort studies and non-randomised trials
is that groups may differ on important characteristics at baseline
participants lsquoself-selectrsquo into either the rider training group by de-
ciding to undertake a rider training course or into the lsquocontrol
grouprsquo by deciding not to undertake a course Participants who
undertake a non-mandatory rider training course may differ from
those who choose not to on characteristics such as age sex rid-
ing experience and riding behaviour as well as other possible con-
founders For example riders who take a training course may be
more safety-conscious or on the contrary may do so because they
have previously experienced a crash or because they perceive that
they have poor riding skills Therefore ideally all participants in
the intervention group need to be included in the study as wellas a random sample of controls taken from the same population
as the intervention group If that is not possible intervention and
control groups either need to be carefully matched to make them
comparable or differences at baseline need to be adjusted for in
the analysis
Most of the studies using a non-randomised trial or cohort design
compared an intervention group to a control group that differed
on a number of important factors The studies in general failed to
adjust for important differences found at baseline and their results
cannot be meaningfully interpreted Ten studies (Raymond 1979
Osga 1980 Satten 1980 Jonah 1982 Mortimer 1984 Mortimer
1988 McDavid 1989 Waller 1992 Davis 1997 Savolainen 2007)
assessed intervention and control groups that were not drawnfrom the same population In most of these cases the intervention
group was selected from a list of training graduates who were usu-
ally young female and less experienced riders whereas controls
were recruited from a general population of motorcycle riders who
were usually older male and experienced riders For three stud-
ies it was unclear how they were selected (Leung 1987 Cooper
1988 Perrino 2002) Only Kloeden 1994 recruited groups from
the same population of motorcycle applicants and Adams 1985
from a group of professional riders Nevertheless due to non-ran-
dom selection of intervention and control groups differences at
baseline are likely even though study groups were from the same
population Three studies adjusted for some confounding fac-
tors found at baseline through matching (McDavid 1989 Waller1992 Billheimer 1998) However matching has limitations in
termsof controlling for multiple confounders especially those that
are unknown to influence the effect of rider training on the out-
comes One study matched the control and intervention groups
by excluding controls (approximately 3500 (66 of initial con-
trol group) that differed from the intervention group) (Kloeden
1994) Five studies (Osga 1980 Satten 1980 Adams 1985 Davis
1997 Perrino 2002) did not adjust for differences between study
groups at baseline either in the study design or analysis Results
from these studies are therefore not reliable The other six stud-
ies (Raymond 1979 Jonah 1982 Mortimer 1984 Cooper 1988
Mortimer 1988 Leung 1987 McDavid 1989 Billheimer 1998)
adjusted for some differences between groups at baseline by eitherbi-variate or multi-variate analyses or stratification and standardi-
sation of riding exposure (outcomemiles driven) or stratification
by age group The limitation of stratification is that it is difficult to
control for many variables simultaneously because a large number
of strata will need to be generated resulting in small numbers that
cannot be meaningfully analysed Savolainen 2007 developed sta-
tistical models to assess the probability of a crash after taking the
rider training course in an intervention and control group while
examining differences by sub-groups (eg age) Participation rates
were also very low in this study with only 5584000 (1395)
in the eligible intervention group and 5884000 (147) for the
14Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 1750
control group completing surveys
Detection bias
None of the studies reported blinding of outcome assessors to the
intervention or control status of the participants Seven studies
used driver records containing police recorded data as their main
measure of outcome (Leung 1987 Cooper 1988 McDavid 1989
Waller 1992 Kloeden 1994 Davis 1997 Billheimer 1998) An-
other four studies (Raymond 1979 Satten 1980 Mortimer 1984
Mortimer 1988) only used self-report Raymond 1979 was the
only study using self-report as the main outcome measurement in
which study instruments were validated Self-report might prevent
the underreporting that has been described for the use of police-
driver records (see rsquoRisk of biasrsquo in the randomised trials above)but has other risks of bias especially if questionnaires or inter-
view methods have not been validated and tested for reliability
Recall bias arises from mistakes in recollecting events both be-
cause of failures of memory and because of looking at things lsquowith
hindsightrsquo and possibly changed views (httpwwwcochraneorg
resourcesglossaryhtm) As many studies were retrospective risk
of recall bias is likely and may have increased with length of follow
up For studies that used self-reported outcomes only it is also
possible that social desirability response bias played a role where
participants in the training group may have felt it was lsquoexpectedrsquo
to report good driving histories Multiple methods to measure the
same outcome for both groups decreases the risk of detection bias
for example because self-reported answers can be cross checkedwith other sources of information such as driver records
In this regard three studies (Osga 1980 Jonah 1982 Perrino
2002) used both self-report and driver records Osga 1980 found
that for both intervention and control groups more crashes were
reported through self-report than on state crash records an aver-
age of 54 of self-reported crashes could be found back on state
crash records One industry-based study (Adams 1985) used oc-
cupational health records which were not validated but which are
unlikely to suffer from underreporting due to work cover claims
Savolainen 2007 used information on rider crashes for a five-year
period but did not specify from where this information was ob-
tained
Attrition bias
The majority of studies (Raymond 1979 Satten 1980 Adams
1985 Leung 1987 Kloeden 1994 Davis 1997 Billheimer 1998
Perrino 2002) inadequately reported on losses of participants to
the study We can therefore not be certain that there were no sys-
tematic differences in the characteristics between the intervention
and control group participants Savolainen 2007 reported the re-
sponse rates to their baseline survey but did not report on the per-
centage of outcome data that was available for their study sample
A similar deterrent effect of rider training as described for the ran-
domised trials (Anderson 1980 Hall 1988) occurred in two stud-
ies Jonah 1982 reported that 65(14992310) of the trained rid-
ers stopped riding compared to 47 (9492029) of the untrainedriders They excluded non-active and non-licensed riders before
the study commenced so the differential choice to continue riding
after training did not affect their analyses Leung 1987 reported
that the percentage of participants in the intervention group ob-
taining their licence was 50 (no number provided by authors)
lower than in the control group This may have biased the mea-
surement of outcomes because unlicensed participants will not be
exposed to riding the motorcycle and therefore not be at risk of
being involved in a crash or violation As explained above this
makes it difficult to assess whether the reduction in crashes in the
intervention group is due to a change in driving habits or simply
due to delayed licensure
Mortimer 1984 and Mortimer 1988 describe that about 44 and30 of course graduates respectively were excluded from analysis
because they had not ridden a motorcycle in the previous year
However because these studies were retrospective and the controls
were activeriders at the time of recruitment a decision on whether
the deterrent effect was differential between the two groups could
not be made Though not reported quantitatively Kloeden 1994
also describes the effectof rider training deterringindividuals from
becoming licensed and hence exposure to riding and risk of crash
death or offences involving a motorcycle Osga 1980 reported
that 27 of the trained riders did not ride after the course and
excluded these from the analysis
Use of different data collection strategies may have contributed to
differential loss to follow up in the cohort studies Satten 1980for example interviewed the intervention group over the tele-
phone but mailed questionnaires to the control group Mortimer
1984 and Mortimer 1988 distributed questionnaires in person to
the control group but mailed them to the intervention group
This may have played a role in the observed differential response
rates where more subjects responded in the personally approached
groups
Survival bias
For those studies that assessed offences crashes and injuries but
not death survival bias may have also played a role Motorcyclists
who were fatally or seriously injured in crashes during the follow-
up period after the intervention would have been less likely or
unable to respond to surveys This could bias the effect measures
towards or away from the null None of the studies wrote about
this issue or examined the characteristics of those lost to follow up
in order to address the possibility of survival bias
Intervention integrity
Although the interventions were intended for the entire lsquoexposedrsquo
groups most studies reported difficulty in tracking participants
whoreceivedthe interventionand those whodid not Adams 1985
15Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 1850
estimatedthat 62of thetotalstudysample received the interven-
tion The remaining studies did not report the percentage of thestudy population that received the intervention as was intended
or reported that this percentage was unknown (Raymond 1979
Kloeden 1994 Satten 1980 Jonah 1982 Mortimer 1984 Leung
1987 Cooper 1988 Mortimer 1988 Billheimer 1998 Perrino
2002 Savolainen 2007) Only Raymond 1979 reported the degree
to which their intervention was delivered as intended No studies
reported on possible lsquocontaminationrsquo of participants by other inter-
ventions than the one to which they were allocated Because Osga
1980 Satten 1980 Jonah 1982 Mortimer 1984 Cooper 1988
Mortimer 1988 Waller 1992 and Davis 1997 were retrospective
studies and participants were recruited from a list of graduates it
was known that they had completed and passed the course
Analysis
In some studies some participants who reported at baseline that
they were untrained decided to enroll in a rider training pro-
gramme after the baseline survey The non-randomised trial by
Adams 1985 experienced this move of subjects from the untrained
to the trained group but failed to conduct analyses by lsquointention-
to-treatrsquo This can bias the results in randomised and non-ran-
domised trials if participants who changed groups differ signifi-
cantly with regard to potentially confounding characteristics On
the other hand Perrino 2002 also experienced this move of sub-
jects from theuntrained to the trained groupmore than 100riders(5 of the total sample) originally stated that they were untrained
at baseline but then received training between the baseline and
follow-up surveys Yet since the Perrino 2002 study was a cohort
their analyses then proceeded by including those riders who were
lsquonewlyrsquo trained as exposed Riders who were lsquonewlyrsquo trained since
baseline but with a programme other than the one under study
were excluded from the analysis Nevertheless confounding fac-
tors were not accounted for in the analyses
McDavid 1989 matched exposed with unexposed subjects but
could not find suitable matches for the majority (60) of their
subjects They could therefore only conduct analysis on the result-
ing40 (139346) which is an inappropriate andinefficient anal-
ysis Some studies that reported on the issue of non-active licensedriders(Satten 1980 Jonah 1982 Mortimer 1984 Mortimer 1988
Kloeden 1994) excluded non-active riders from analysis intro-
ducing selection bias as it is likely that riders who continued to
ride differed with regards to potentially confounding character-
istics across intervention and control groups Kloeden 1994 for
example conducted several analyses to assess the effect of various
sub-sets of the control group Original analyses included all con-
trols and led to erroneous results since only a small proportion of
controls hadgone on to get a licence after their learnerrsquos permitand
were actually exposed to the possibility of experiencing a crash
Subsequent analyses compared only intervention and control sub-
jects who had a licence Waller 1992 reported crashes by person-
days of exposure but did not explain how they could be sure that
each study intervention and control participant was an active riderand actually exposed to the risk of experiencing a crash
Potential confounders that are unequally distributed by study arm
at baseline need to be dealt with in the analysis Satten 1980
Adams 1985 and Davis 1997 did not take into account any con-
founding factors either by design or in the analysis McDavid
1989 Waller 1992 and Billheimer 1998 accounted for potential
confounding factors at the design stage by matching the subjects
in the intervention and control groups Other studies (Raymond
1979 Mortimer 1984 Mortimer 1988 Perrino 2002 and also
Waller 1992) adjusted for one or two confounders by stratifica-
tionor standardisation (eg crashesmiles driven age group) Osga
1980 conducteda univariate analysis of crash rates in which no po-
tential confounders were accounted for but also calculated cumu-lative crash rates by mileage travelled Only Jonah 1982 Cooper
1988 and Savolainen 2007 conducted a multivariate analysis
Risk of bias in the case-control studies
A case-control design is suitable for the assessment of risk fac-
tors for outcomes such as motorcycle crashes injuries or deaths
but not appropriate for the evaluation of an intervention such
as motorcycle rider training It can potentially result in spurious
conclusions regarding the efficacy of the intervention because the
intervention has not been applied at random to an intervention
and control group Even with the control of possible confounders
using a case-control design to evaluate an intervention can lead tothese kinds of results since those who choose to undertake rider
training may differ on important prognostic factors that affect the
outcome compared to those who do not choose to participate in
rider training
Selection bias
A potential source of bias in case-control studies as with all other
study types where the intervention is not randomly allocated to
participants is that comparison groups are likely to differ on po-
tentially confounding characteristics This means that ideally all
cases need to be included in the study as well as a random sample
of controls taken from the population that gave rise to the cases If that is not possible cases and controls either need to be carefully
matched to make themcomparable or differences at baseline need
to be adjusted for in the analysis All four case-control studies were
population based Kraus 1975 included all fatalities and injured
cases leading to hospitalisation and appropriately recruited a ran-
dom sample of controls from the same population Hurt 1981 did
not manage to recruit all or a random sample of cases Only half
of the notifications of crashes to the police were received in time
by the researchers the research team for unreported reasons was
only able to respond to half of those again Data collection was
therefore completed for only about a fifth of all cases Neverthe-
less this was still a large sample of 900 cases Haworth 1997 using
16Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 1950
a similar study design was more likely to have included all cases
but this is not entirely clear from the study reported It is unclearfor example whether fatalities were included as they did not use
coronerrsquos reports for the recruitment of cases The MAIDS 2004
study was conducted in five European countries but it was unclear
how cases were recruited whether they were all cases during the
two years that the study lasted or a random sample of all cases
what the response rates were etc Controls were recruited from
randomly selected petrol stations but it was likewise not clear what
the eligibility criteria for inclusion in the study were how they
were recruited (eg randomly) and what the response rate was
Hurt 1981 and Haworth 1997 both matched controls to cases by
recruiting themat the sametime ofdayandsamedayof weekas the
crash occurred Hurt 1981 however recruited controls two years
later andit is likelythat there would have beendifferences betweenthe two comparison groups One important difference was that
during data collection for cases there was very little specialised
motorcycletrainingavailable in the studyregionwhereas two years
later when control data were collected there was some training
available
These four case-control studies also suffered from selection bias
due to potential differences between those that gave consent to
participate in the study and responded to interviews andquestion-
naires and those that did not None of the studies have assessed
such differences although Hurt 1981 did acknowledge that con-
trols giving consent to cooperate may have differed from those
motorcyclists who did not
Detection bias
Measures that can be taken to reduce detection bias in case-con-
trol studies are blinding of interviewers for case or control status
blinding of the participant for the study factor of interest validat-
ing measurement instruments and using these in an identical way
for cases and controls and cross checking self-reported informa-
tion with other more objective sources such as registers None of
the studies reported blinding of the intervention assessors to case
or control status but all used consistent measurement methods for
both case and control groups
All case-control studies used self-report as their main measure toassess whether cases and controls had received rider training and
if so the type of training The major potential source of bias in
case-control studies when self-report is used as a measure is recall
bias One way of reducing recall bias is by double-checking self-
reported data with more objective data Kraus 1975 also used po-
lice accident reports for information on rider characteristics but
such information was not available about rider training Haworth
1997 and MAIDS 2004 did not report on any measures to limit
recall bias Hurt 1981 reported further questioning of cases by
the interviewer when answers did not make sense or when par-
ticipants seemed to contradict themselves as part of the in-depth
unstructured interview technique This qualitative research tech-
nique can result in valuable information but as interviewers were
not blinded to case or control status of the participants it may have led to more thorough examination of cases compared to the
controls and thus differential detection bias
Intervention integrity
The level of detail about the rider training provided in the case-
control studies differed significantly but was minimal for each
Kraus 1975 simply asked whether lsquoanyrsquo motorcycle rider training
had lsquoeverrsquo been carried out Hurt 1981 asked participants whether
they hadbeen self taught informally taught or completed formally
taught rider training courses MAIDS 2004 asked whether partic-
ipants had had pre-license training additional (non-compulsory)
training other training or no training whereas Haworth 1997asked about the number of courses that a participant had com-
pleted whetherthey were basic intermediate or advanced courses
and how long ago the training had been done None of the case-
control studies reported collecting information about further de-
tails about any training programmes (eg type of training cur-
riculum duration of training)
Analysis
Kraus 1975 accounted for confounding factors through the exclu-
sion of females and stratification by age and frequency of motor-
cycle use Hurt 1981 matched cases and controls for time of day
and day of the week that the crash occurred and did not adjust forany confounding factors in the analysis Haworth 1997 matched
groups for time of day and day of the week that the crash occurred
and stratified by age blood alcohol level and licence status in the
analysis None of the studies used multivariate methods MAIDS
2004 did not provide any information about any participant char-
acteristics other than the study exposures of interest and did not
describe any methods to adjust for potential confounding factors
Effects of interventions
We have presented the results in a descriptive manner due to the
substantial heterogeneity in study types interventions and out-comes and the generally poor quality of the studies We have
grouped the interventions by mandatory and non-mandatory pre-
licence rider training and post-licence rider training and further
by study type and by outcome (crashes injurious and fatal crashes
and offences) We discuss results that are reported as statistically
significant (whether a P value was reported or not)
Pre-licence rider training - mandatory
Randomised trials
17Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2050
Crashes
Total group of participants
Anderson 1980 found a reduction in motorcycle crash rates in
the intervention group (MOST skill test with remedial training)
compared to the control group (who received the skill testwithout
a training component) The percent differenceranged from 143
six months after the programme (P lt 005) to 211 (P lt 0001)
after one year and 127 (P lt 001) after two years Hall 1988
found no statistically significant difference in crash rates between
intervention groups at any point in time
Subgroup of participants who obtained their licence only
Anderson 1980 found a reduction in motorcycle crash rates in the
intervention arm but this was statistically significant at one year
only (24 to 218 reduction P lt 005) Hall 1988 found no
statistically significant differences
Offences
Total group of participants
The percentage of offences in Anderson 1980 after six months
and one and two years was slightly higher in the intervention
arm compared to the control group but these differences were
not statistically significant Likewise the study by Hall 1988 did
not find a statistically significant difference in offences between
groups
Subgroup of participants who obtained their licence only
Hall 1988 found no significant difference in offence rates between
the intervention and control groups Anderson 1980 found thatthe percentage of offences increased in the intervention arms com-
pared to the control group This difference was statistically signif-
icant after two years (51 to 115 increase P lt 005)
Cohort studies
Crashes
Neitherof the cohort studies (Kloeden1994 and Billheimer1998)
found a statistically significant difference in crash rates between
the intervention and control groups
Offences
Billheimer 1998 (for the California Motorcyclist Safety Program
course) and Perrino 2002 did not find a statistically significant
difference in offence rates between the intervention and control
groups
Case-control studies
Crashes
MAIDS 2004 found that a slightly higher percentage of controls
(not involved in a crash) had had pre-license rider training than
cases (involved in a crash) This difference was not statistically significant (50 of controls vs 477 of cases OR 091 95 CI
076 to 110)
Pre-licence rider training - non-mandatory
Randomised trial
Injuries and deaths
Swaddiwudhipong 1998 found a significantly lower injury ratein the intervention villages (1051000 population 95 CI 92
to 120) compared to the control villages (1691000 population
95 CI 152 to 185) one year after the intervention This differ-
ence was not maintained two years after the intervention (127
95 CI 112 to 143 intervention 151 95 CI 135 to 167
control) There was no significant difference in death rates be-
tween groups
Cohort studies
Crashes
Cooper 1988 foundno statistically significant differencein crashes
between the intervention and control groups McDavid 1989
found a lower crash rate in the intervention group (14139) com-
pared to the control group (23139) but no test of statistical sig-
nificance was undertaken
Injuries and deaths
McDavid 1989 found that the intervention group had fewer in-
jurious crashes (814 crashes) compared to the control group (16
23 crashes) but no statistical significance test was reported
18Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2150
Offences
Cooper 1988 foundno statistically significantdifference in offencerates between the intervention and control groups McDavid 1989
foundno statistically significantdifference in offence ratesbetween
the intervention and control groups
Non-mandatory rider training
Non-randomised trials
Crashes
Neither Adams 1985 nor Leung 1987 found a statistically signif-icant effect of rider training on crash rates
Offences
Leung 1987 found a statistically significant lower offence rate in
the intervention group compared to the control group for both
males and females and for all age groups except for women 35
years or older (P values not reported)
Cohort studies
Crashes
Osga 1980 in the univariate analysis reported a lower percentage
of crashesin theintervention group (that continued to ride) (28
168595) than in the control group (34 112329) In the cu-
mulative crash rate analysis however Osga1980 found statistically
significant (at the P lt 001 level) higher crash rates in the interven-
tion group than in the control group at allmileage intervals (1000
to 29999 miles) travelled After riding more than 30000 miles
(when over 50 of the trained riders had been involved in a crash
compared to approximately 40 in untrained riders) this differ-
ence was no longer statistically significant Satten 1980 found a
lower percentage of crashes in the intervention group (13 969)
compared to the control group (25 2079) but the differencewas not assessed statistically Davis 1997 found a lower crash rate
in trained riders but it was unclear how this was calculated be-
cause the size of the control group was not reported Jonah 1982
and Billheimer 1998 (for the Experienced Rider Course) found no
statistically significant difference in crashes between the interven-
tion and control groups Although the difference between those
who completed the Experienced Rider Course and no course in
Billheimer 1998 neared significance at six months after the course
(P = 007) large P values were observed for one year after (P =
030) Raymond 1979 found a statistically significant higher crash
rate for the intervention group (1861000 miles ridden) com-
pared to the control group (1181000 miles ridden) in both males
and females (P lt 0001) especially during the first 5000 miles of
riding Waller 1992 found a statistically significant higher crashrate for the intervention group (27 crashes in three years or 070
crashes10000 person days) compared to the control group (5
crashes in three years or 013 crashes10000 person days) with a
crude relative risk of 538 (X 2=1513 plt005) Mortimer 1984
and Mortimer 1988 found no statistically significant difference in
crashes between the intervention and control groups
Savolainen 2007 found that the intervention group (those who
completed the Basic Rider Course) was 44 more likely to be
involved in crashes than the control group and those who com-
pleted the course multiple times were an additional 180 more
likely to be involved in a crash This finding was based on low
numbers and no P value or confidence level was given for either
result
Injuries and deaths
Satten 1980 also found a lower percentage of injurious crashes
(33 39) in the intervention group compared to the control
group(60 1220) although again no statistical significance test
was reported
Offences
Jonah 1982 did find a significantly lower self-reportedoffence rate
in the intervention group (15) compared to the control group(34 P lt 001) (no numbers given) Satten 1980 also found
a lower self-reported offence rate in the intervention group (1
71 versus 19100) but again this was not tested for statistical
significance Mortimer 1984 Mortimer1988 and Billheimer 1998
found no statistically significant differencein offence ratesbetween
the intervention and control groups
Case-control studies
CrashesMAIDS 2004 found that a slightly higher percentage of cases
(involved in a crash) had had additional non-mandatory rider
training than controls (not involved in a crash) This difference
was not statistically significant (17 of cases vs 12 of controls
OR 146 95 CI 064 to 337)
Rider training unspecified
Case-control studies
19Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2250
Injuries and deaths
Hurt 1981 found that the odds of injurious crashes were lower inthose who had received professional training compared to those
who had not Kraus 1975 found that in the younger age group
(less than 20 years) and the older age group (over 35 years) a larger
percentage of injured riders (12 and 21 respectively) reported
prior motorcycle rider training compared with the uninjured con-
trol group (6 and 5 respectively) (P lt 001) (no numbers
given) Haworth 1997 reported several different outcomes one
beingthat having completed at leastone trainingcoursecompared
to no training courses at all had no significant effect on the odds
of crashing (afteradjustment for age blood alcohol level or licence
status (learner permitprobationaryfull)) When comparing three
different levels of training courses (beginner intermediate and ad-
vanced) with no training no significant differences were found
in the odds of crashing However after adjustment for age and
blood alcohol level completing an advanced course as compared
to a beginners course was associated with a significant decrease in
the odds of severe motorcycle crashes resulting in serious injury or
death (OR 02 95 CI 01 to 08)
D I S C U S S I O N
The objective of this review was to quantify the effectiveness of
pre- andpost-licence motorcycle rider training on the reduction of
traffic crash involvement injuries deaths and offences committedby motorcycle riders Most included studies were cohort studies
There were a few randomised and non-randomised trials and case-
control studies
All studies compared either mandatory or non-mandatory formal
training with no formal training except for Haworth 1997rsquos case-
control study that compared levels of training (beginner interme-
diate and advanced) to each other
We found studies that evaluated the effectiveness of manda-
tory pre-licence training in reducing crash and offence rates
(Anderson 1980 Hall 1988 Kloeden 1994 Billheimer 1998
(the basic California Motorcyclist Safety Program component of
the study) Perrino 2002) non-mandatory training that when
passed resulted in a licence (Cooper 1988 McDavid 1989
Swaddiwudhipong 1998) non-mandatory training that could be
taken by motorcycle riders with or without a licence (Raymond
1979 Osga 1980 Satten 1980 Jonah 1982 Mortimer 1984
Adams 1985 Leung 1987 Mortimer 1988 Waller 1992 Davis
1997 Billheimer 1998 (the Experienced Rider Course compo-
nent of the study)) and unspecified rider training on crash injury
and death rates (Kraus 1975 Hurt 1981 Haworth 1997 MAIDS
2004)
The studies provided conflicting evidence regarding motorcycle
rider training which is likely due to the poor design quality and
reporting of the studies Most of the studies included in the review
were likely to be adversely affected by significant selection biasalong with substantial detection bias(measurement and recall bias)
which limits our ability to make conclusive comments
Mandatory pre-licence rider training
There were two RCTs three cohort studies and one case-control
study that examined pre-licence ridertraining The two RCTs eval-
uated the Motorcycle Skills Operating Test (MOST) (Anderson
1980 Hall 1988) Anderson 1980 found a statistically significant
reduction in crashes in the intervention group at six months one
and twoyearsfollow up However when only participantswho ob-
tainedtheirlicence were included in theanalysis thereduction wasstatistically significant only at one year follow up The other RCT
(Hall 1988) found no statistically significant differences in crash
rates at any point of time neither in the total population group
nor in the licensed subgroup Neither of the two cohort stud-
ies assessing crash rates found a statistically significant difference
in crash rates between intervention and control groups ( Kloeden
1994 Billheimer1998) andthe case-control study (MAIDS 2004)
also did not find a statistically significant difference in the odds of
experiencing a crash between trained and untrained riders With
regards to offences neither of the RCTs found a statistically sig-
nificant difference in offence rate between the intervention and
control groups in the analysis of the total group of participants
entering the trial However Anderson 1980 found a statistically significant increase in offence rates in the intervention group but
only after two years of follow up andonly in the analysis of partici-
pants who obtained their licence The two cohort studies assessing
offence rates found no statistically significant difference between
the groups (Billheimer 1998 Perrino 2002)
None of the five studies (RCTs or cohort) found a reduction in
crashes or offences following mandatory pre-licence rider training
except for Anderson 1980 who observed a statistically significant
reduction in crashes in the intervention group at one year in both
analyses It is not clear why both RCTs that evaluated the same
MOST and training intervention observed different results Both
studies enrolled large samples and observed a similar uptake of
licensure following the intervention It is possible that differencesmay have arisen due to different study samples (California ver-
sus New York) or that the intervention delivery differed between
studies
In both trials of the Motorcycle Skills Operating Test including
rider training only a third of motorcycle licence applicants ob-
tained their motorcycle licence It is possible that a licensing pro-
cess including a skills test may reduce crashes through decreased
exposure
Non-mandatory pre-licence training
20Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2350
Three studies assessed the effect of non-mandatory pre-licence
training The Cooper 1988 cohort study found no difference incrash or offence rates between the trained and untrained groups
We cannot however be confident about this result as the study
sufferedfrom selectionbiaswhichcouldhave biasedthe measure of
effect toward the null The RCT conducted by Swaddiwudhipong
1998 examined injury and death rates and found a statistically
significant lower injury rate in the intervention villages after one
year of follow up but no longer at two year follow up There
was no statistically significant difference in death rates between
intervention and control villages Our confidence in the results of
this study are hindered by the inadequacy of the randomisation
process that led to significant differences between the intervention
and control districts on two important variables The intervention
districts were more likely than the control districts to have a validlicence andto wear a helmetSincethese differences were nottaken
into account in later analyses they could explain the differences
observed in injury rates between districts Furthermore with only
six districts randomised the sample size was probably too small
to detect a difference between intervention and control groups
had there been one The McDavid 1989 cohort study found a
lower percentage of injuries in the intervention group although
a statistical test was not conducted McDavid 1989 drew their
intervention and control groups from different populations but
did adjust for confounders although they only analysed 40 of
the initial sample (139 trained and 139 untrained) Because of
these substantial weaknesses we are unable to draw conclusions
from this studyAn overall effect of pre-licence training on the reductionof crashes
injuries deaths and offences was not found although one should
be aware that the research of this issue to date has had significant
flaws However a mandatory skill test (such as the MOST) in
combination with and even without pre-licence training as part
of a licensing process seems to reduce the number of individuals
completing the licensing procedure and therefore indirectly re-
duces crash rates through a reduction in exposure to motorcycle
riding Making licensure a more difficult process may therefore be
seen as advantageous if it delays licensure and therefore reduces
exposure It is possible that the group who chose not to continue
riding a motorcycle would not have been the group to experience a
crash or commit an offence if they had continued to ride anyhowfor example because they are more safety conscious and this was
the reason for their giving up riding in the first place If this is the
case deterring individuals from completing their licence and rid-
ing may not necessarily decrease overall crashes deaths or offence
rates In addition other potentially adverse effects of complicat-
ing a licensing procedure need to be considered such as increases
in unlicensed riding In Thailand non-mandatory rider training
offered as a community intervention may have decreased injury
rates but this needs to be confirmed through further research
Non-mandatory training
The majority of the studies assessed non-mandatory training
(training independent of any licensing procedure) and the results
were varied The two non-randomised trials assessing crash rates found no statistically significant difference in crash rates between
the intervention and control groups (Adams 1985 Leung 1987)
Of the nine cohort studies Satten 1980 found a lower percentage
of crashesin the intervention group although it is unclear whether
this was a statistically significant result Davis 1997 also found a
lower percentage of crashesin theinterventiongroup although it is
unclear with which control group the intervention group wascom-
pared Raymond 1979 Osga 1980 Waller 1992 and Savolainen
2007 found a statistically significant higher crash rate in the in-
tervention group and the remaining cohort studies (Jonah 1982
Mortimer 1984 Mortimer 1988 Billheimer 1998) found no sta-
tistically significant difference in crash rates between intervention
and control groups No conclusions can be drawn from these stud-
ies with confidence The Satten 1980 and Davis 1997 studies did
not adjust for any confounding factors even though important
differences existed between the intervention and control groups
Although Raymond 1979 and Waller 1992 adjusted for a num-
ber of confounders at baseline (Waller 1992) and in the analy-
sis (Raymond 1979 Waller 1992) the intervention and control
groups were drawn from such different populations that it is pos-
sible that unmeasured confounders were responsible for the sta-
tistically significant increased risk of crashes observed in the inter-
vention group The observed difference could also be attributed totheobserved differential loss to followup (Raymond 1979) which
could bias the measure of effect towards or away from the null
The Osga 1980 study found a protective effect of rider training
in the univariate analysis but statistically significant higher crash
rates in the intervention group when accounting for riding expo-
sure Sex (a significantly higher proportion of females in the in-
tervention group) and other potentially confounding factors that
could explain thedifferencebetween the groups were howeverun-
accounted for Finally Savolainen 2007 did control for potential
confounding variables yet it is not clear whether variables were
controlled for simultaneously or one at a time One case-control
study (MAIDS 2004) found no statistically significant reduction
in the odds of experiencing a crash between trained and untrainedriders but the quality of this study was poor mainly because the
analysis did not account for any confounding factors
One cohort study (Satten 1980) assessed the effect of non-manda-
tory rider training on injurious crashes and found a lower percent-
age of injuries in the intervention group although a statistical test
was not conducted As mentioned above the Satten 1980 studyrsquos
validity is reduced due to its lack of adjustment for confounding
factors
The only non-randomised trial that assessed offence rates (Leung
1987) found a statistically significant lower offence rate in the in-
tervention group and of the five cohort studies that assessed of-
fence rates (Satten 1980 Jonah 1982 Mortimer 1984 Mortimer
21Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2450
1988) one study (Jonah 1982) also found a statistically significant
lower offence rate in the intervention group However we have lit-tle confidence in the results from Leung 1987 since potential con-
founding factors were not appropriately addressed Jonah 1982
did conduct a multivariate analysis but was hindered by an initial
low response rate The other four cohort studies did control for
several potential confounders and found no difference in offence
rates following non-mandatory post-licence rider training
The non-randomised trials and cohort studies evaluating non-
mandatory rider training were at high risk of substantial selection
bias due to the non-random allocation of participants as well
as low participation rates and attrition bias and could therefore
not provide conclusive evidence with regard to the effect of the
training It should be noted that despite the biases and other issues
with this group of studies most didnot detectmajor effects relatedto rider training in either direction This may suggest that any
potential impact of rider training will be modest
Studies examining unspecified modes of ridertraining
Three case-control studies assessed whether injured or deceased
motorcyclists were more or less likely to have had rider training
compared to control motorcyclists All had conflicting results
Hurt1981 foundthe odds of havingan injuriouscrash waslower in
those who had received rider training whereas Kraus 1975 found
that the odds of having an injurious crash were higher in youngerand older trained riders Haworth 1997 who compared advanced
versus beginner training found no effect on motorcycle injuries
or deaths although they did find that those who had completed
advanced training had reduced chances of crashing compared to
those who had done a beginnersrsquo training course Similar issues
with uncontrolled confounding and biases of selection as described
above arose with the three case-control studies Hurt 1981 suffers
from inadequate control of confounding via poor matching and
both the Kraus 1975 and the Haworth 1997 findings were only
controlled for limited factors by means of stratified analysis It is
likely that these limitations are the reason for the contradictory
directions of effect observed for the three case-control studies
Methodological weaknesses of the studies
In summary most studies investigating the effectiveness of mo-
torcycle rider training on reducing crashes injuries and death suf-
fered from serious methodological weaknesses
The majority of studies were non-randomised and were there-
fore limited mainly by selection bias Although the approaches of
matching stratification or regression analyses can minimise bias
(Fleiss 2004) they were poorly employed in the reviewed studies
The inadequate adjustment for confounders in particular adds
to our lack of confidence in the studiesrsquo results since it has been
demonstrated that individuals who choose to take motorcycle rider
training differ from those not taking rider training For instance
trained riders often differ from controls in age sex and experi-ence usually ride less and may have a more positive attitude to
safety as indicated by the reported more frequent wearing of seat
belts (Mortimer 1982)
The two large randomised trials reported that a large proportion
of their study group were never exposed to the risk of offences or
crashes because they never attained licensure This made it impos-
sible to assess whether a reduction in crashes in the intervention
group was a result of training or due to delayed or never attained
licensure
In some studies the sample sizes were too small to detect any sta-
tistically significant differences Follow-up time after training dif-
fered between studies and was an impediment to observing any
real effects of training if they existed As riding experience ac-cumulates over time any differences in outcomes between those
trained and untrained would diminish in studies with lengthy fol-
low-up times This dilution of any possible effect of rider training
may have also occurred in studies that mixed both novice and ex-
perienced riders in the study groups since it is hypothesised that
novice riders serve to benefit most from rider training This would
increase the difficulty in finding any effect of rider training par-
ticularly when sample sizes are small
Detection bias may also account for some of the differences ob-
served between trained and untrained riders Detection bias would
have occurred if the self-report hospital driver or police records
used to record the outcomes of crash injury death or offences
differed between the trained and untrained riders Research hasdemonstrated that not all crashes are reported to police for ex-
ample small crashes or off-road crashes (Kraus 1975) even serious
injury crashes have been found to be underreported (Anderson
1980) However this would only introduce detection bias if the
reporting of outcome differs for trained and untrained riders This
does appear possible in that responsible riders may be more likely
to enroll in a rider training course and also may be more likely
to report crashes to police than irresponsible riders meaning that
the number of crashes from riskier riders could be underrepre-
sented on records An analysis to investigate the effectiveness of
rider training may then show that rider training is not effective in
reducing crashes
Rather than using reported outcomes some studies used self-re-ported survey data although this technique suffers from non-
response biases In fact two studies reported differences in de-
mographic characteristics in response rates to questionnaires
(Anderson 1980 Hall 1988) Self-reported data also introduce
the risk of recall bias survival bias and social desirability response
bias Recall bias may have been equally distributed among trained
and untrained riders howeverit is hypothesised that trained riders
may have felt it was lsquoexpectedrsquo to report good driving histories
hence leading the rider training to appear effective
22Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2550
Limitations of this review
Limitations of this review were that we found very few recentstudies (the two most recent ones published in 2002 and 2007)
and that we found very few studies conducted in low-and middle
income countries even though our search strategy was set up to
identify such studies Anotherlimitation wasthat forthe reviewed
studies we could not assess all outcomes of rider training that are
potentially useful Some of the included studies assessed outcomes
such as use of safety clothing including helmets Such outcomes
are valuable since the use of motorcycle helmets has been demon-
strated to be effective in preventing head injuries (Liu 2004) De-
spite this the use of safety clothing was not included as an out-
come in this review because the studies measuring this outcome
were limited by poor data quality due to very low response rates
Another outcome of interest in assessing the effectiveness of rider
training is lsquotype of injuryrsquo As Haworth 2005 state in their review
of motorcycle licensing and training rider training could be con-
sidered effective if injuries become less severe because motorcycle
riders have learned how to avoid more serious injuries through
rider training The studies included in this review did not specify
type of injury and so this outcome could not be assessed Last
changes in knowledge and attitudes following rider training are
other potentially useful outcomes We have excluded such studies
(Osga 1979 Clayton 1990 Goldenbeld 2004) as these outcomes
are proxies rather than direct measures of the effectiveness of mo-
torcycle rider training We opted to report on the outcomes of
crashes injury death and offences alone since there is little ev-
idence about how changes in knowledge and attitudes relate to
these outcomes
A U T H O R S rsquo C O N C L U S I O N S
Implications for practice
Due to the poor quality of studies included in this review we were
unable to draw any conclusions about the effectiveness of rider
training on crash rates injurious or fatal or offence rates The
findings do suggest that mandatory pre-licence training may be
an impediment to completing a motorcycle licensing process thuspossibly indirectly reducing crash injury death and offence rates
through a reduction in exposure to riding a motorcycle On the
basis of the existing evidence it is not clear if and what type of
training reduces the risk of crashes injuries deaths or offences in
motorcyclists and the selection of the best rider training practice
can therefore not be recommended The suggestion that manda-
tory pre-licence training indirectly reduces exposure to riding the
motorcycle because trainees are impeded from completing the li-
censing process does not help those motorcyclists who persist in
obtaining their licence and riding the motorcycle Some type of
rider training is likely necessary to teach motorcyclists to ride a
motorcycle safely Good policy is based on sound evidence and
evaluation but such evidence for motorcycle rider training is seri-
ously lacking Policy makers are therefore encouraged to include arigorous evaluation component to any new or existing rider train-
ing programme Such an evaluation should ideally be randomized
and assess whether the training programme is effective and if so
which component(s) of the programme work best
Implications for research
Given theresults of this review which found littlereliable evidence
about the effectiveness of motorcycle rider training it is necessary
to build up an evidence base of the effectiveness of motorcycle
rider training in the reduction of crash injury death and offence
rates To this end well designed randomised controlled trials need
to be conducted that examine the impact of rider training oncrashes and offences A randomised controlled trial is the preferred
method as it would significantly reduce most biases described in
this review To minimise the impact of participants delaying or
never attaining their licence and therefore exposure as found in
the randomised trials in this review such a trial should include
riders who are already licensed and reliable measures of riding
exposure in a reasonably lengthy follow up period It is expected
that a randomised controlled trial of sufficient sample size would
observe equal balance on possible confounders by study arm how-
ever an assessment of potentially important confounders (such as
helmet wearing) by study group would enable control for these
variables in the analyses if necessary It is imperative that the effec-
tiveness of each component of the rider training (such as theorypractice knowledge skills and modification of behaviour) and the
synergy of such components are evaluated Outcomes should be
objective ideally those that are routinely collected rather than self-
reported outcomes Even though riding the motorcycle poses dif-
ferent challenges and incurs different risks lessons may be learned
from research into the effects of car driver training
Motorcycles are a common means of transport in low- and middle
income countries and usage is increasing As the burden of mo-
torcycle crash victims is mostly borne in such countries research
should be focused on the effectiveness of types of rider training on
the reduction of crashes injuries and fatalities in such countries
and be supported by higher income countries This review found
only one study conducted in a middle income country that sug-gested decreasing injury rates It would be beneficial to conduct
additional research to confirm this suggestion and to enhance our
knowledge of the role of rider training in low- and middle-income
settings
March 2006 saw the launch of the University of North Carolina
Highway Safety Research Centre partnership with the Motorcycle
Safety Foundation (MSF) and plans were revealed to conduct a
study of the potential benefits of motorcycle rider training The
study is funded by the MSF and the National Highway Traffic Sa-
fety Administration (NHTSA) Called rsquoThe Discovery Projectrsquo it
will compare data on crash rates from riders participating in a basic
23Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2650
motorcycle safety class with riders participating in a basic motor-
cycle safety class as well as a variety of continuing education classesThe collection of pilot data was due to start in March 2007 and
the data collection portion of the study is expected to take three
years (Highway Safety Research Center) It is vital that researchers
and policy makers collaborate to develop furtherhigh quality eval-
uations of motorcycle rider training programmes to help inform
development of new and effective training programmes
A C K N O W L E D G E M E N T S
Staff from The University of Sydney library for helping us locate
and retrieve full reports of many papers
R E F E R E N C E S
References to studies included in this review
Adams 1985 published data only
Adams A Collingwood V Job RFS Evaluation of the 1982 Australia
Post motorcycle rider training scheme NSW Australia Traffic
Accident Research Unit Traffic Authority NSW 1985 [ RN 85]
Anderson 1980 published data only
Anderson J Ford JL Peck RC Improved motorcyclist licensing and testing project Washington DC National Highway Traffic Safety
Administration (NHTSA) USA 19801ndash217 [
DOTndashHSndash805ndash598]
Billheimer 1998 published data only
Billheimer JW Evaluation of the California Motorcyclist Safety
Program Transportation Research Record 19981640100ndash9 [
TRB Report Number 980652]
Cooper 1988 published data only
Cooper PJ Rothe JP Motorcyclists who they are and why they do
what they do Transportation Research Record 1988116878ndash85
Davis 1997 published data only
David CF Evaluation of community traffic safety programs and
motorcycle operator training programs Connecticut Department of Transportation 199763
Hall 1988 published data only
Hall CJ Buchanan L Engle R Motorcycle rider education evaluationproject Washington DC USA National Highway Traffic Safety
Administration (NHTSA) US Department of Transportation
1988 [ DTNH 22ndash80ndashCndash05123]
Haworth 1997 published data only
Haworth N Smith R Brumen I Pronk N In Monash University
Accident Research Centre editor(s) Case-control study of motorcycle
crashes Victoria Australia Department of Transport and Regional
Development The Federal Office of Road Safety 19971ndash149 [
ISBN 0 642 25501 6 ]
Hurt 1981 published data only
Hurt H Ouellet J Thom D Motorcycle accident cause factors and
identification of countermeasures USA Department of
Transportation 1981 [ Technical Report DOT HS 806ndash862]
Jonah 1982 published data onlylowast Jonah BA Dawson N Bragg B Are formally trained motorcyclists
safer Accident Analysis amp Prevention 198214(4)247ndash55
Jonah BA Dawson NE Bragg BW Evaluation of the Canada
Safety Councilrsquos Motorcycle Training Program Proceedings of theInternational Motorcycle Safety Conference Washington DC
USA 198059ndash93
Kloeden 1994 published data only
Kloeden CN Moore VM McLean AJ Evaluation of the pre-license
training program for motorcyclists in South Australia Walkerville
South Australia Australia NHMRC Road Accident Research Unit
The University of Adelaide 19941ndash37 [ ISBN 0 7308 0607 3]
Kraus 1975 published data only
Kraus JF Riggins RS Franti CE Some epidemiological features of
motorcycle collision injuries American Journal of Epidemiology 1975102(1)74ndash98
Leung 1987 published data only
Leung KS Reding VA Evaluation of the Wisconsin motorcycle rider course Wisconsin Bureau of Policy Planning and Analysis
Wisconsin Department of Transportation 1987
MAIDS 2004 published and unpublished data
MAIDS In-depth investigations of accidents involving powered
two wheelers ACEM (Acemacemeu) 2004 Vol Final report
20179
McDavid 1989 published data only
McDavid JC Lohrmann BA Lohrmann G Does motorcycle
training reduce accidents Evidence from a longitudinal quasi-
experimental study Journal of Safety Research 198920(2)61ndash72
Mortimer 1984 published data only
Mortimer R Evaluation of the motorcycle rider course Accident
Analysis amp Prevention 198416(1)63ndash71
24Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2750
Mortimer 1988 published data only
Mortimer RG A further evaluation of the motorcycle rider courseJournal of Safety Research 198819(4)187ndash96
Osga 1980 published data only
Osga GA An investigation of the riding experiences of MSF rider
course participants Dissertation South Dakota USA The
University of South Dakota 19801ndash224
Perrino 2002 published data only
Perrino CS Ahmed A Callender A Rozier E Cantwell A Stewart
O The role of Marylandrsquos motorcycle rider course in promoting saver
behaviors and attitudes Baltimore USA National Transportation
Center Morgan State University 20021ndash52
Raymond 1979 published data only
Raymond S Tatum S An evaluation of the effectiveness of the RAC
ACU motorcycle training scheme - final report Road Safety ResearchUnit Dept of Civil Engineering University of Salford London
19791ndash131
Satten 1980 published data only
Satten RS Analysis and evaluation of the motorcycle rider courses
in thirteen Illinois counties International Motorcycle Safety
Conference Washington DC 1980145ndash93
Savolainen 2007 published and unpublished data
Savolainen Peter Mannering Fred Additional evidence on the
effectiveness of motorcycle training and motorcyclistsrsquo risk-taking
behaviour Transportation Research Board 2007 Vol March 20
19
Swaddiwudhipong 1998 published data only
Swaddiwudhipong W Boonmak C Nguntra P Mahasakpan P
Effect of motorcycle rider education on changes in risk behavioursand motorcycle-related injuries in rural Thailand Tropical Medicine
and International Health 19983(10)767ndash70
Waller 1992 published data only
Waller FL An analysis of the impact of the Texas Department of
Public Safety Motorcycle Operator Training Program upon
motocycle related traffic accidents Texas State University Public
Administration Program Applied Research Projects 199277
References to studies excluded from this review
Anderson 1980a published data only
Anderson LC Better biking a motorcycle rider improvement
program International Motorcycle Safety Conference Washington
DC USA Motorcycle Safety Foundation 1980231ndash53Aupetit 2008 published and unpublished data
Aupetit S Riff J Buttelli O Espie S Ergonomics research program
on motorcycle learning Contributions to the development of a
motorcycle simulator for training Transport Research Arena
Ljubljana 20089
Awane 1999 published data only
Awane T Integrating simulators in motorcycle safety education
Journal of International Association of Traffic and Safety Sciences 199923(1)26ndash35
Baldi 2005 published data only
Baldi S Baer JD Cook AL Identifying best practices states in
motorcycle rider education and licensing Journal of Safety Research200536(1)19ndash32
Batchler 1988 published data only
Batchler MW Evaluation of the pre-license motorcycle training course at the New South Wales traffic education centre Federal Office of
Road Safety (FORS) 19881ndash122 [ Report No CR 77 ]
Blanchard 2006 published data only
Blanchard HT Tabloski PA Motorcycle safety educating riders at
the teachable moment Mosby (CV) Company 200632330ndash2
Bonnett 2005 published data only
Bonnett J Mature age skills training for experienced motorcycle
riders Motorcycle ridersrsquo association of the Australian Capital
Territory 2005
Braver 2007 published data only
Braver ER et alPersuasion and licensure a randomized controlled
intervention trial to increase licensure rates among Maryland
motorcycle owners Traffic Injury Prevention 20078(1)39ndash46Clayton 1980 published data only
Clayton AB Sudlow DE An evaluation of the effectiveness of a one day training programme for learner motorcyclists Berkshire UK
Transport and Road Research Laboratory 19801ndash23 [ ISSN
0266ndash7045]lowast Clayton AB Sudlow DE An evaluation of the effectiveness of a one-day training programme for learner motorcyclists Berkshire UK
Road Safety Division Berkshire 19801ndash23 [ISSN 0266ndash7045]
Council 1978 published data only
Council FM Desper LP Dutt AK Driver education for motorcycle operation University of North Carolina Highway Safety Research
Center 19781ndash40
Crowther 2005 published data onlylowast Crowther G Engaging with motorcyclists UK police and theBikeSafe Road Safety Programme Published proceedings of the
Australasian Road Safety Research Policing Education Conference
Wellington New Zealand 2005 issue ISBN 0ndash473ndash10636ndash11ndash11
Daniello 2009 published data only
Daniello A Gabler HC Mehta YA Effectiveness of motorcycle
training and licensing Transportation Research Board 88th Annual
Meeting 200913
Dewanti 2007 published data only
Dewanti M Challenging efforts in promoting young motorcyclist
safety in Indonesia Road Safety on Four Continents 14th
International Conference Swedish National Road and Transport
Research Institute 2007
Du Plooy 1990 published data onlyDu Plooy DW Motorcycle training in South Africa The Human
Element International Motorcycle Safety Conference Orlando
Florida USA Motorcycle Safety Foundation 19904-1 to 4-9
Engel 1989 published data only
Engel U Krogsgaringrd Thomsen L [Faeligrdselsundervisning og
propaganda ndash en faeligrdselssikkerhedsfremmende foranstaltning]
Dansk Vejtidsskrift 198912295ndash7
Goldenbeld 2002 published data only
Goldenbeld C Houwing S de Craen S The development of
driving skills of young moped riders [De ontwikkeling van de
rijvaardigheid van jonge bromfietsers] SWOV Institute for Road
Safety Research (Stichting Wetenschappelijk Onderzoek
Verkeersveiligheid) Leidschendam the Netherlands 200263
25Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2850
Goldenbeld 2004 published data only
Goldenbeld C Twisk D de Craen S Short and long term effects of moped training a field experiment Transportation Research Part F 20047(1)1ndash16
Haworth 2000 published data only
Haworth N Smith R Kowadlo N Motorcycle training in Victoria
evaluation of rider training curriculum in Victoria Monash
University Accident Research Centre Reports 2000 Vol 165 issue
3131
Haworth 2005 published data only
Haworth N Mulvihill C Review of motorcycle licensing and
training Monash University Accident Research Centre 2005 [
Report no240]
Haworth 2006 published data only
Haworth N Mulvihill C Rowdon P Teaching old dogs new tricksTraining and older motorcyclists Journal of the Australasian College of Road Safety 200718(4)20ndash5lowast Haworth NL Mulvihill C Rowden PJ Teaching old dogs new
tricks Training and older motorcyclists Australasian Road Safety
Research Policing and Education Conference Gold Coast
Queensland 2006
Hill 1983 published data only
Hill D Motorcycle rider training testing and licensing Motor
Cycle Council of New South Wales Sydney 1983
Imatake 1980 published data only
Imatake Y On the activities in Japan of motorcycle safe riding
promotion committee (MSRPC) - especially the education of
unskilled riders of class 1 motorbikes International Motorcycle
Safety Conference Washington DC USA Motorcycle Safety
Foundation 1980255ndash68
Ingebritsen 1990 published and unpublished data
Ingebrigtsen S [Risikofaktorer ved ferdsel med moped og
motorsykkel] TOslashI-rapport 66 Oslo Transportoslashkonomisk
institutt 1990
Jain 2009 published and unpublished data
Jain A et alTwo wheeler accidents on Indian roads a study from
Mangalore India Elsevier 200916130ndash3
Jonah 1979 published data only
Jonah BA Dawson NE Validation of the motorcycle operator skill
test Accident Analysis amp Prevention 197911(3)163ndash71
Jonah 1981 published data onlyJonah BA Dawson NE Bragg WE Predicting accident
involvement with the Motorcycle Operator Skill Test Accident Analysis amp Prevention 198113(4)307ndash418
Kadar 1990 published data only
Kadar KJ Countermeasures motorcycle rider education delivery
system The Human Element International Motorcycle Safety
Conference Motorcycle Safety Foundation 19906-27 to 6-41
Kelsey 1986 published data only
Kelsey SL Liddicoat C Ratz M Licensing novice motorcyclists a
comparison of Californiarsquos standard test and the MOST II (Motorcycle
Operator Skill Test) administered at centralized testing offices Sacramento California USA California Department of Motor
Vehicles Research and Development Office 19861ndash74
Koch 1980 published data only
Koch H Advanced course for motorcycle riders InternationalMotorcycle Safety Conference Washington DC USA Motorcycle
Safety Foundation 1980269ndash81
Koch 1990 published data only
Koch H Brendicke R How to influence risk taking behaviour
within motorcycle rider training programs The Human Element
International Motorcycle Safety Conference Orlando Florida
USA Motorcycle Safety Foundation 1990Motorcycle Safety
Foundation 1990 p 9-1 to 9-25A
Labbett 2005 published data only
Labbett S Langham M Training can make the problem worse
Road Safety Congress 20051ndash12
Law 2005 published and unpublished data
Law TH Radin Umar RS Zulkaurnain S Kulanthayan S Impactof the effect of economic crisis and the targeted motorcycle safety
programme on motorcycle-related accidents injuries and fatalities
in Malaysia International Journal of Injury Control and Safety
Promotion 200512(1)9ndash21
Lowes 1990 published data only
Lowes BF In-traffic rider training The Human Element
International Motorcycle Safety Conference Orlando Florida
USA Motorcycle Safety Foundation 19907-46 to 7-57
Mahon 1976 published data only
Mahon J Teaching accident avoidance on the motorcycle Journal of Traffic Safety Education 1976231ndash29
Maxwell 1985 published data only
Maxwell GS Evaluation of a Program for Training MotorcycleRiders Evaluation and Program Planning 19858(4)339ndash47
Mayhew 1996 published data only
Mayhew DR Simpson SH Effectiveness and role of driver educationand training in a graduated licensing system Ottowa Canada The
Traffic Injury Research Foundation 19961ndash89 [ ISBN
0ndash920071ndash12ndash0]
McKnight 1987 published data only
McKnight AJ Milburn BG An evaluation of the modified
motorcycle operator skill test exercises NHTSA Report 1987 [
Report No DOT HS 807 212 ]
McPherson 1976 published data only
McPherson K McKnight AJ The development and evaluation of a
motorcycle skill test manual and knowledge test Washington DCUSA National Highway Safety Administration 1976
Mohan 2004 published data only
Mohan D Evidence-based interventions for road traffic injuries in
South Asia Journal of the College of Physicians and SurgeonsPakistan 200414(12)746ndash7
Morgan 1980 published data only
Morgan JK Advanced motorcycle training course content and
study evaluation International Motorcycle Safety Conference
Washington DC USA Motorcycle Safety Foundation 1980
283ndash312
Motoki 1990 published data only
Motoki M Yamazaki S A study on effective motorcycle rider
education The Human Element International Motorcycle Safety
26Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2950
Conference Orlando Florida USA Motorcycle Safety
Foundation 19909-26 to 9-57A
Nairn 1992 published data onlylowast Nairn RJ Motorcycle safety research literature review 1987 to
1991 Federal Office of Road Safety Australia 1993 Vol Report
CR 117
Newland 1980 published data only
Newland R Darby J Automotive practicespre motorcycle rider
program International Motorcycle Safety Conference Washington
DC USA Motorcycle Safety Foundation 1980127ndash9
Newland 1982 published data only
Newland R Motorcycle rider training and licensing requirements
Driver training steering a course for the future Melbourne
Victoria Transport Regulation Board 1982
Newland 1990 published data only
Newland R Motorcycle training and licensing in Victoria 1980-
1990 The Human Element International Motorcycle Safety
Conference Orlando Florida USA Motorcycle Safety
Foundation 1990
Ormston 2003 published data only
Ormston R Dudleston A Pearson S Stradling Evaluation of
Bikesafe Scotland Scottish Executive Social Research Edinburgh
2003
Osga 1979 published data only
Osga GS Ellingstad VS An evaluation of the effectiveness of the 1978
South Dakota motorcycle rider course program South Dakota USA
University of South Dakota 19791ndash89
Panou 2007 published data only
Panou M Bekiaris E A new concept on the integration of driving
simulators in driving training the train-all approach Road Safety
on Four Continents 14th International Conference Swedish
National Road and Transport Research Institute 2007
Peterson 1990 published data only
Peterson A Kadar K Status report 1980-1989 Minnesota rider
education statistics The Human Element International
Motorcycle Safety Conference Orlando Florida USA 199013-
113 to 13-129
Prem 1984 published data only
Prem H Good MC Motorcycle rider skills assessment ACT
Australia Office of Road Safety Department of Transport 1984
1ndash377 [ Report No CR 34 ]
Radin Umar 2006 published data only
Radin Umar RS Motorcycle safety programmes in Malaysia how
effective are they International Journal of Injury Control and Safety Promotion 200613(2)71ndash9
Rockwell 1990 published data only
Rockwell TH Kiger S Carnot M An evaluation of the Ohio
motorcyclist enrichment program Ohio Department of Highway
Safety 1990
Rothe 1987 published data only
Rothe JP A motorcycle riding course Motorcyclists Image and reality Canada Insurance Corporation of British Columbia 1987
59ndash83
Rowden 2007 published data only
Rowden P Watson B Haworth N What can riders tell us aboutmotorcycle training A view from the other side of the fence
Australasian Road Safety Research Policing and Education
Conference Melbourne Australia Melbourne 2007
RSC Victoria 1993 published data only
Road Safety Committee Inquiry into Motorcycle Safety in Victoria
Parliament of Victoria 1993
Russam 1979 published data only
Russam K Motorcycle training and licensing in Japan UK Transport
and Road Research Laboratory (TRRL) 1979 [ Report 916 ]
Saffron 1981 published data only
Saffron DG Driver and rider education training and licensing a
brief review Traffic Accident Research Unit Traffic Authority of
New South Wales NSW Australia Traffic Accident Research UnitTraffic Authority of New South Wales 1981
Schulz 1990 published data only
Schulz U Motorcycle accidents among learner riders Motorcycle
licensing Evaluation of the effectiveness of a Graduated Licensing Scheme Forshungshefte des Institut fur Zweiradsicherheit 1990
Sexton 2004 published data only (unpublished sought but not used)
Sexton B Baughan C Elliott M Maycock G The accident risk of
motorcyclists 2004TRL Report TRL 607(ISSN 09684107)58
Shephard 1986 published data only
Shepherd R Do Motorcycle Safety Foundation programs work
Driver 198619(9)16ndash8
Simpson 1990 published data onlySimpson HM Mayhew DR The promotion of motorcycle safety
training education and awareness Health Education Research
Theory and Practice 19905(2)257ndash64
Sokolov 1990 published data only
Sokolov O Safe motorcycle riding some aspects of motorcycle
riding in the USSR The Human Element International
Motorcycle Safety Conference Motorcycle Safety Foundation
19904ndash37 - 4-48
Steffens 1988 published data only
Steffens U Gawatz R Willmes G Effectiveness of mofa courses
[Wirksamkeit von Mofakursen] Unfall- und sicherheitsforschung
strasenverkehr 198867168
Sudlow 2003 published data only
Sudlow D Scoping study on motorcycle training Road Safety
Research Report London UK The British Institute of Traffic
Education Research Department for Transport 2003 Vol June
[ Report No 36]
Tasmania 1995 published data only
Mannering FL Grodsky LL Motorcycle rider training scheme
Traffic Education Research Centre Armidale Australia Driver
Training Australia Limited 1995 Vol April1ndash56
Tetard 1985 published data only
Tetard C Motorcycle ridersrsquo training to detect indices in an
intersection situation [Lrsquoentrainement a la prise drsquoinformation dans
lrsquoapprentissage de la conduite motocycliste] Le Travail Humain198548(2)147ndash59
27Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3050
Thackray 1980 published data only
Thackray RM Prescott JC Field test of the motorcycle ridercourse International Motorcycle Safety Conference Washington
DC USA Motorcycle Safety Foundation 1980195ndash7
Thompson 1994 published data only
Thompson M Evaluation of compulsory basic training for
motorcyclists Transport Research Laboratory Crowthorne
Berkshire UK 1994 [ Report 63]
Tobler 1990 published data only
Tobler U Individual driving instruction for motorcycle driving
learners the Swiss experiment of motorcycle driving school The
Human Element International Motorcycle Safety Conference
Orlando Florida USA Motorcycle Safety Foundation 1990
Transport 1989 published data only
Transport Communications Road Crash Statistics Australia Bureauof Transport and Communications Economics 198914ndash20
Tuan 2007 published data only
Tuan VA Shimizu T Towards development and evaluation of the
motorcycle drivers re-education program in Vietnam modeling of
motorcycle driverrsquos undesired behaviors 11th World Conference
on Transport Research 200725
Victor 1980 published data only
Victor DJ Traffic safety education for motorcyclists International
Motorcycle Safety Conference Washington DC USA Motorcycle
Safety Foundation 1980221ndash8
Waterer 1990 published data only
Waterer TH Compulsory training for motorcyclists from liberty
to compulsion The Human Element International Motorcycle
Safety Conference Orlando Florida USA Motorcycle Safety
Foundation 19904-49 to 4-57
Weller 1989 published data only
Weller RB Chandler EW Motorcycle safety and motorcycle
education past research and survey results Journal of Public Policy
amp Marketing 1989893ndash108
Wells 1981 published data only
Wells P Observations of some RACACU motorcycle training
schemes Transport and Road Research Laboratory Crowthorne
Berkshire UK 1981
Winn 1986 published data only
Winn G Two important trends in motorcycle safety regulations
rider education and conspicuity improvement American
Motorcyclist Association 1986147ndash52
Wisher 1988 published data only
Wisher G Reid J Motorcycle accidents in Norfolk a three-year study of motorcycle training and the Part 1 test Norfolk UK Road Safety
Division Highways Department Norfolk County Council 1988
Wood 1987 published data only
Wood HT Bowen R Evaluation of the revised learner permit scheme
July 1983 to December 1985 Victoria Australia Road Traffic
Authority 19871ndash27 [ ISBN 0 7306 0247 8]
Yamazaki 1990 published data only
Yamazaki S Nakanishi M Survey on effect of Honda rider
education The Human Element International Motorcycle Safety
Conference Orlando Florida USA 199012-167 to 12-187
References to ongoing studies
Burgess 2010 unpublished data only
Burgess C Broughton P Fylan F Stradling Interim evaluation of
the UKs National Ride program In L Dorn editor(s) Driver
behaviour and Training Vol IV Aldershot Ashgate Publishing
Limited 2010
Additional references
Anderson 1980
Anderson J Ford JL Peck RC Improved motorcyclist licensing and
testing project Volume I and II National Highway Traffic Safety
Administration US department of Transportation Washington
DC 1980
Bjornstig 1985
Bjornstig U Bylund P Lekander T Brorsson B Motorcycle fatalitiesin Sweden Acta Chirurgica Scandinavica 1985151(7)577ndash81
Clayton 1990
Clayton A Sudlow DE An evaluation of the effectiveness of a one-
day training program for learner motorcyclists The Human
Element International Motorcycle Safety Conference Orlando
Florida USA 199013-105 to 13-111
Cochrane Collaboration 2008
Higgins JP Green S Cochrane Handbook for Systematic Reviews of
Interventions 500 The Cochrane Collaboration 2008
Elliot 2003
Elliott MA Baughan CJT Broughton J Grayson G Knowles J et
alMotorcycle Safety a Scoping Study TRL [ Report No
0968ndash4107]Fleiss 2004
Fleiss JL Levin B Paik MC Statistical methods for rates and proportions 3rd Edition John Wiley amp Sons Inc 2004 [ ISBN
978ndash0ndash471ndash52629ndash2 ]
FORS 1999
FORS Federal Office of Road Safety Road risk for sober licensed
motorcyclists Vol Monograph 27 Federal Office of Road Safety
19991ndash2
Haworth 2005
Haworth N Mulvihill C Review of motorcycle licensing and training
Victoria Australia Monash University Accident Research Centre
20051ndash83 [ Report No 240]
Huwiler-Muumlntener 2002Huwiler-Muumlntener K Juumlni P Junker C Egger M Quality of
Reporting of Randomized Trials as a Measure of Methodologic
Quality JAMA 20022872801ndash4
Johnston 2008
Johnston P Brooks C Savage H Fatal and serious road crashes
involving motorcyclists Road Safety Research and Analysis Report
ATSB Monograph 20 Canberra Department of Infrastructure
Transport Regional Development and Local Government 0205
081ndash26
Juumlni 2001
Juumlni P Altman DG Egger M Systematic reviews in health care
Assessing the quality of controlled clinical trials BMJ 2001323
42ndash6
28Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3150
Kopits 2003
Kopits E Cropper M Traffic fatalities and economic growth TheWorld Bank Development Research Group Infrastructure and
Environment Policy research working paper 3035 2003
Leung 1983
Leung P Chau BTW Motorcycle accidents in Hong Kong pattern
of injury treatment causative factors and socio-economic
implications Injury 198315(1)23ndash9
Liu 2004
Liu B Ivers R Norton R Blows S Lo SK Helmets for preventing
injury in motorcycle riders Cochrane Database of Systematic Reviews 2004 Issue 2 [DOI 10100214651858CD004333pub3]
Mohan 2002
Mohan D Road safety in less-motorized environments future
concerns International Journal of Epidemiology 200231(3)527ndash32Mortimer 1982
Mortimer RG An evaluation of the motorcycle rider course
Proceedings of the 20th conference of the American Association for
Automotive Medicine Arlington Heights IL USA American
Association for Automotive Medicine 198277ndash96
Mullin 1997
Mullin BT Risk factors for motorcycle injury the role of age
gender experience training and alcohol PhD thesis Auckland
New Zealand University of Auckland 1997
Noordzij 2001
Noordzij PC Forke E Brendicke R Chinn BP Integration of needs
of moped and motorcycle riders into safety measures review and
statistical analysis in the framework of the European research
project PROMISING SWOV Institute for Road Safety Research
Leidschendam The Netherlands 2001 [ Report No Dndash2001ndash5]
Osga 1979
Osga G Ellingstad VS An evaluation of the effectiveness of the 1978 South Dakota Motorcycle Riders Course Program South Dakota
University of South Dakota Vermillion Human Factors Lab 1979
Oxford English Dictionary Online
httpoxforddictionariescom Oxford English Dictionary Online
Oxford University Press 2010
Peden 2004
Peden M Scurfield R Sleet D World Report on Road TrafficInjury Prevention Geneva World Health Organization Available
at httpwwwwhointviolence˙injury˙preventionpublications
road˙trafficworld˙reporten 2004Chapter 3
Radin Umar 1995
Radin Umar R Macakay G Hills B Preliminary analysis of
motorcycle accidents Short-term impacts of the running
headlights campaign and regulations in Malaysia J Traffic Med 199523(1)17ndash27
Tsai 1995
Tsai YJ Wang JD et alCase-control study of the effectiveness of
different types of helmets for the prevention of head injuries among
motorcycle riders in Taipei Taiwan American Journal of
Epidemiology 1995142(9)974ndash81
Vis 1995
Vis AA In-depth study of the hazards of motorcycling [De
onveiligheid van motorrijden nader bekeken] SWOV Institute for
Road Safety Research Leidschendam the Netherlands 1995 [
Report No Rndash95ndash69]
Weaver 1990
Weaver EA Motorcycle safety education in the United States
Motorcycle Safety Foundation Orlando Florida USA 1990
Winn 1987
Winn G Two important trends in motorcycle safety regulations
rider education and conspicuity improvement No more
information available Detroit MI US Society of Automotive
Engineers 1987147ndash52
Yuan 2000
Yuan W The effectiveness of the rsquoride-brightrsquo legislation for
motorcycles in Singapore Accident Analysis amp Prevention 200032
(4)559ndash63
Zambon 2006
Zambon F Hasselberg M Factors affecting the severity of injuries
among young motorcyclists a Swedish nationwide cohort study
Traffic Injury Prevention 20067(2)143ndash9lowast Indicates the major publication for the study
29Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3250
C H A R A C T E R I S T I C S O F S T U D I E S
Characteristics of included studies [ordered by study ID]
Adams 1985
Methods Non-randomised trial
Participants Initial sample size
Total of 162 Australia Post telegram delivery riders
Not clear how many in intervention and control groups
Analysis sample size
Intervention group 63Control group 35
Total 98 (61)
Interventions No formal training vs voluntary rider training course of 12 hours over 2 consecutive days (theory 4 hours practice
8 hours)
Intervention integrity
No information given
Outcomes Australia Post recorded crashes
Length of follow up 1 year
Notes Study conducted in NSW Australia
Anderson 1980
Methods Randomised trial
Participants Initial sample size
Total 41188 motorcycle licence applicants
Analysis sample size
Intervention group with skills training 13160
Intervention group without skills training 12634
Control group 15080
Total 40874 (99)
Interventions Standard licensing skill test vs the Motorcycle Operators Skill Test (MOST) with or without a training component
MOST is not to be confused with rider training because it is a licensing test The training component was for
participants who failed the skill test on the first attempt was of 3 hours duration and consisted of instructions about
the skills needed to pass the skill test The training consisted of 30 min theory and 2 frac12 hours practice on the
motorcycle
Intervention integrity
33 in MOST with skills training 31 in MOST without skills training and 43 in control group obtained their
motorcycle licence in the end
Outcomes Police recorded crashes and offences
Length of follow up 3 years
30Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3350
Anderson 1980 (Continued)
Notes Study conducted in California USA
Billheimer 1998
Methods Cohort study
Participants Initial sample size
Intervention group Unclear numbers of CMSP graduates and ERC graduates
Control group Over 16000 untrained motorcyclists
Analysis sample size
1139 matched CMSP-pairs and 1182 ERC-pairs
Interventions No formal training vs the mandatory (for those younger than 18 years) pre-licence 16-hours basic California
Motorcyclist Safety Program (CMSP) and the 8-hour Experienced Rider Course (ERC)
Intervention integrity
No information given
Outcomes Police recorded crashes and offences
Length of follow up 2 years
Notes Study conducted in California USA
Cooper 1988
Methods Cohort study
Participants Initial sample size
Intervention group 418 graduates and 57 who failed
Control group 402
Analysis sample size
Intervention group unclear
Control group unclear
Total 863 (984)
Interventions No formaltrainingvs non-mandatory pre-licence rider trainingThe content of thetrainingwas notfurther explainedIntervention integrity
Participants had all completed the rider training
Outcomes Police recorded crashes and offences
Length of follow up 7 years
Notes Study conducted in British Columbia Canada
31Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3450
Davis 1997
Methods Retrospective cohort study
Participants Initial sample size
Intervention group 9320 graduates
Control group unclear
Analysis sample size
Intervention group 9320 graduates
Control group unclear
Interventions No formal training vs the non-mandatory Connecticut Motorcycle Operator Training Program The content of
training was not further explained
Intervention integrity Participants had all completed the rider training
Outcomes Police recorded crashes severity of crashes
Length of follow up 1 year
Notes Study conducted in Connecticut USA
Hall 1988
Methods Randomised trial
Participants Initial sample sizeIntervention group A 6248
Intervention group B 6616
Intervention group C 6996
Control group 6604
Total 26464 motorcycle licence applicants
Analysis sample size
Intervention group A 6012 (96)
Intervention group B 6428 (97)
Intervention group C 6778 (97)
Control group 6502 (98)
Interventions Standard licensing skill test vs the Motorcycle Operator Skill Test (MOST II) without training (A) MOST II with
3 hours training (B) or MOST II with 20 hours training (C) The 3 hours training comprised instructions aboutthe skills needed to pass the skill test (30 min theory and 2 frac12 hours practice on the motorcycle) The 20 hours
training consisted of the Motorcycle Safety Foundationrsquos Motorcycle Rider Course (MRC) comprising of 8 hours
theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-road practice
Intervention integrity
25 and 26 obtained their motorcycle licence in the lsquoMOST 3-hour trainingrsquo and lsquoMOST 20-hour trainingrsquo
intervention arm respectively In the lsquoMOST testrsquo intervention arm 29 obtained their licence and in the control
group 35
Outcomes Police recorded crashes and offences
Length of follow up 3 years
32Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3550
Hall 1988 (Continued)
Notes Study conducted in New York USA
Haworth 1997
Methods Population based case-control study
Participants Cases Motorcycle riders or pillion riders involved in motorcycle crash (on road) Recruited from hospital and police
records Not clear whether all cases (especially fatalities) were included
Community controls Multiple matched sample of motorcycle rider or pillion who passed the crash site at the same
time of day and day of week that the crash occurred
Initial sample size
Cases 336
Controls 1195
Analysis sample size
Cases 234 (70)
Controls 561 (47)
Interventions Basic handling skills course beginner intermediate or advanced course measured through self-report not validated
Outcomes Injuries with an Injury Severity Score of 5 or greater or admission for at least four hours and fatalities from a
motorcycle crash (from hospital and police records)
Length of follow up two years or more prior to the interview
Notes Study conducted in Victoria Australia
Hurt 1981
Methods Population based case-control study
Participants Cases Motorcycle riders who had a motorcycle crash that was reported to police (traffic accident reports) Not all
cases included
Community controls Multiple matched sample of motorcycle riders who passed the crash site at the same time of
day and day of week as the crash occurred but 2 years later
Initial sample size
Cases 4500 notifications of crashes to police (only 50 received in time research team able to respond to 1126 for
unreported reasons data collection completed for 900)
Controls Number approached unreported 2310 consented and interviewed
Analysis sample size
Cases 808 (808900 data on rider training) - 18 of initial sample
Controls 670 (6702310 data on rider training) - lt 29 of initial sample
Interventions Training experience self taught friends-family school-club formal-professional others measured through self-
report validated
Outcomes Injuries and fatalities from a motorcycle crash (from hospital records and coronerrsquos office)
33Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3650
Hurt 1981 (Continued)
Notes Study conducted in Los Angeles USA
Jonah 1982
Methods Cohort study
Participants Initial sample size
Intervention group
2310 licensed MTP graduates who are active riders
Control group
2029 licensed motorcyclists who are active riders
Non-licensed and non-active riders excluded
Analysis sample size
Intervention group
811 interviews (35) 533 driver records (23)
Control group 1080 interviews (71) 905 driver records (45)
Interventions No training vs the non-mandatory Canada Safety Council Motorcycle Training Program (MTP) comprising of 4
hours theory and 16 hours practise (6 hrs on-road) over 2 weekends
Intervention integrity
Participants had all successfully completed the MTP
Outcomes Self-reported and police reported crashes and offencesLength of follow up 4 years
Notes Study conducted in Ontario Canada
Kloeden 1994
Methods Cohort study
Participants Initial sample size
Intervention group 2533 licence applicants
Control group Unclear
Sample of riders obtaining probationary licenceIntervention group 1625 (64)
Control group 5015 (unclear)
Non-active riders excluded from control group (~66)
Analysis sample size
Intervention group 631 (39)
Control group 318 (6)
Interventions No formal training vs the mandatory pre-licence Ridersafe course consisting of two levelsthe first is a training course
in the basics of motorcycle riding where a pass entitles the rider to obtain a learnerrsquos permit The second is a skill
test where a pass entitles the rider to obtain a probationary motorcycle licence The training course consists of two 4
hour training sessions on separate days involving instruction videos and actual riding on motorcycles The skill test
consists of a 4 hour preparation course of 74 minutes of lecturediscussion 38 minutes of video and 75 minutes of
34Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3750
Kloeden 1994 (Continued)
on-course riding which is the testing phase
Intervention integrity
Unclear what percentage in the Ridersafe group and the control group obtained their probationary licence halfway
and their full motorcycle licence in the end
Outcomes Police reported crashes
Length of follow up 1 year
Notes Study conducted in SA Australia
Kraus 1975
Methods Population based case-control study
Participants Cases All motorcycle riders with a confirmed medically treated motorcycle injury (including fatalities) from all
hospitals in study base and peripheral area hospitals and all deaths due to motorcycle crashes in the study base All
serious injuries (leading to hospitalisation) and fatalities likely to be included
Community controls A random sample of registered motorcycle owners in the same temporary and geographical
study base as the cases arose from
Initial sample size Cases 1273
Controls 738
Analysis sample size Cases 741 (58)
Controls 434 (59)
Interventions Motorcycle rider training (not specified) measured through self-report
Outcomes Injuries leading to hospitalisation fatalities
Length of follow up not specified
Notes Study conducted in Sacramento county USA
Leung 1987
Methods Non-randomised trial
Participants Initial sample sizeIntervention group unknown number of MRC graduates
Control group unknown number of licensed motorcyclists
Analysis sample size
Intervention group 2941
Control group 43094
Interventions No formal training vs non-mandatory Wisconsin motorcycle rider course a 22 hours off-street programme that
teaches the basic rules of safe motorcycling
Intervention integrity
50 in the MRC group and 100 in the control group obtained their motorcycle licence in the end
35Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3850
Leung 1987 (Continued)
Outcomes Police recorded crashes and offences
Length of follow up up to 3 years
Notes Study conducted in Wisconsin USA
MAIDS 2004
Methods Population based case-control study
Participants Cases Motorcycle riders who had a motorcycle crash Unclear whether all cases included
Community controls Motorcycle riders identified at petrol stations Unclear whether matched to cases
Initial sample size
Cases not reported
Controls not reported
Analysis sample size
Cases 921 (Italy 200 Netherlands 200 Spain 121 Germany 250 France 150)
Controls 923 (Italy 200 Netherlands 200 Spain123 Germany 250 France 150)
Interventions Pre-license rider training or additional training (non-compulsory)
Outcomes Crashes
Notes Study in five sampling areas located in five European countries (France Germany Netherlands Spain and Italy)
McDavid 1989
Methods Cohort study
Participants Sample size after first match
Intervention group 346 male training graduates
Control group 346 male motorcycle riders licensed in same year
Sample size after second match
Intervention group 139 (40)
Control group 139 (40)
Analysis sample size
Intervention group 139 (40)
Control group 139 (40)
Interventions No formal training vs the non-mandatory pre-licence British Columbia Safety Councilrsquos 37-hour motorcycle safety
programme
Intervention integrity
Participants had all completed and passed the training
Outcomes Police reported crashes injurious crashes
Length of follow up 1-6 years
36Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3950
McDavid 1989 (Continued)
Notes Study conducted in British Columbia Canada
Mortimer 1984
Methods Cohort study
Participants Initial sample size
Intervention group 608 MRC graduates
Control group 360 motorcycle riders
Response rates
Intervention group 360 (59)
Control group 324 (90)
Non-active riders excluded
Analysis sample size
Intervention group 213 (35)
Control group 303 (84)
Interventions No formal training vs the non-mandatory Motorcycle Safety Foundationrsquos Motorcycle Rider Course comprising of
8 hours theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-road practice
Intervention integrity
Participants had all completed and passed the MRC
Outcomes Self-reported crashes and offencesLength of follow up 1 year
Notes Study conducted in Illinois USA
Mortimer 1988
Methods Cohort study
Participants Initial sample size
Intervention group 2307 MRC graduates
Control group 556 active licensed motorcycle riders
Response ratesIntervention group 1298 (56)
Control group 500 (90)
Non-active riders excluded
Analysis sample size
Intervention group 913 (40)
Control group 500 (90)
Interventions No formal training vs the non-mandatory Motorcycle Safety Foundationrsquos Motorcycle Rider Course comprising of
8 hours theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-road practice
Intervention integrity
Participants had all completed and passed the MRC
37Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4050
Mortimer 1988 (Continued)
Outcomes Self reported crashes and offences
Length of follow up 1 year
Notes Study conducted in Illinois USA
Osga 1980
Methods Cohort study
Participants Initial sample sizeIntervention group 1163 MSF MRC participants invited 818 (74) responded
Control group 329 motorcycle riders responded to recruitment strategies
Non-active riders excluded from analysis
Analysis sample size
Intervention group 595 (73)
Control group 329 (100)
Interventions No formal training vs the non-mandatory Motorcycle Safety Foundationrsquos Motorcycle Rider Course (MRC) com-
prising of 8 hours theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-road
practice
Intervention integrity
Participants had all successfully completed the MRC
Outcomes Self-reported crashes police recorded crash reports
Length of follow up 6 months - 3 12 years
Notes Study conducted in South Dakota USA
Perrino 2002
Methods Cohort study
Participants Initial sample size
Intervention group 2144 (unclear) MRC graduates (motorcycle licence applicants)
Control group 2601 (unclear) regular licensing test graduates (motorcycle licence applicants)Analysis sample size
Intervention group 1070 (unclear)
Control group 867 (unclear)
Interventions No formal training vs the mandatory (for those younger than 18 years) pre-licence Maryland Motorcycle Rider
Course (MRC) consisting of 8 hours theory and 12 hours practice
Intervention integrity
No information given
Outcomes Self-reported and police recorded crashes and offences
Length of follow up 1 year
38Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4150
Perrino 2002 (Continued)
Notes Study conducted in Maryland USA
Raymond 1979
Methods Cohort study
Participants Initial sample size
Intervention group
258 volunteers mainly (but not all) novice riders
Control group
1104 provisional licence applicants
Analysis sample size
Intervention group 190 (74)
Control group 627 (57)
Interventions No formal training vs the non-mandatory RACACU (Royal Automobile ClubAuto-Cycle Union) training scheme
The training program consisted of either 24 one-hour sessions or 13 two-hour sessions Twelve hours of the training
focused on theory about basic operational procedures traffic laws and highway code The practical session focused
on basic operational skills riding on a training area and riding in traffic with an instructor A proficiency test was
administered at the end of the course
Intervention integrity
No information given
Outcomes Self-reported crashes (including minor non-injury non-damage and off-road crashes)
Length of follow up 1 year
Notes Study conducted in the UK
Satten 1980
Methods Cohort study
Participants Initial sample size
Intervention group interviewers kept interviewing until they had 100 riders selected from a population of 891
enrollees in the MSF MRC course (unknown how many were asked and how many refused to participate)Control group unknown number of registered motorcycle riders
Non-active riders excluded from analysis
Analysis sample size
Intervention group 71
Control group 100
Interventions No formal training vs the non-mandatory Motorcycle Safety Foundationrsquos (MSF) Motorcycle Rider Course (MRC)
comprising of 8 hours theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-
road practice
Intervention integrity
Participants had all successfully completed the MRC
39Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4250
Satten 1980 (Continued)
Outcomes Self-reported crashes injuries and offences
Length of follow up 3 years
Notes Study conducted in Illinois USA
Savolainen 2007
Methods Cohort study
Participants Initial sample size
Intervention group 4000 riders that had conducted the MSF MRC course (passed or failed)
Control group 4000 motorcycle owners or permit or endorsement holders from the Indiana Bureau of Motor
Vehicles database
Analysis sample size
Intervention group 739 (18)
Control group 588 (15)
Interventions Intervention integrity
No information given
Outcomes Probability of crash involvement
Length of follow up up to 5 years
Notes Study conducted in Indiana USA
Swaddiwudhipong 1998
Methods Randomised trial
Participants Initial sample size
Intervention group
all 16 villages of 3 sub-districts
(participants were practising motorcyclists licensed and unlicensed)
Control group
18 villages of 3 other sub-districtsAnalysis sample size
Intervention group 3 districts (100)
Control group 3 districts (100)
Interventions No rider education vs non-mandatory community based rider education consisting of theory on motorcycle injury
risk helmet use traffic laws vehicle regulations and traffic signs and a written and skill test leading to a licence
Intervention integrity
No information given
Outcomes Injuries fatalities
Length of follow up 2 years
40Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4350
Swaddiwudhipong 1998 (Continued)
Notes Study conducted in Tak Province rural Thailand
Waller 1992
Methods Cohort study
Participants Initial sample size
Intervention group 349 riders who succesfully completed the Department of Public Safety motorcycle operator
training in 1988
Control group 349 motorcyclists who never completed the DPS motorcycle operator training
Analysis sample size
Intervention group 349
Control group 349
Interventions No DPS motorcycle operator training vs non-mandatory DPS motorcycle operator training The DPS training is
the Motorcycle Safety Foundationrsquos Motorcycle Rider Course comprising eight hours of theory (lectures discussions
films and multiple choice tests) a textbook and 12 hours off-road practice
Intervention integrity
All 349 riders succesfully completed the rider training
Outcomes Police recorded crashes
Length of follow up 3 years
Notes Study conducted in
Characteristics of excluded studies [ordered by study ID]
Study Reason for exclusion
Anderson 1980a Study type review and process evaluation of a rider course
Aupetit 2008 Outcome proposal of a rider training curriculum that combines traditional training and a motorcycle riding
simulator
Awane 1999 Study type review on use of simulators in rider and driver training
Baldi 2005 Intervention not rider training but intervention integrity of rider training availabilty of rider education
implementation of graduated licensing system etc on a state-wide level
Batchler 1988 Study type description of a training course
Blanchard 2006 Study type not a study assessing the association between rider training and outcomes
Bonnett 2005 Outcome qualitative data regarding satisfaction with the training programme
41Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4450
(Continued)
Braver 2007 Outcome licensure rates
Clayton 1980 Outcome knowledge skills attitude
Council 1978 Outcome crash reduction in ANY vehicle not motorcycles exclusively
Crowther 2005 Study type qualitative study
Daniello 2009 Study type review of the effectiveness of motorcycle training and licensing
Dewanti 2007 Intervention emotional characteristics behaviour (ie not motocycle rider training)
Du Plooy 1990 Study type review of types of rider training courses in South Africa
Engel 1989 Study type uncontrolled before-after study
Goldenbeld 2002 Outcome skills
Goldenbeld 2004 Outcome knowledge and skills
Haworth 2000 Intervention balance between attitudinal andvehicle control skills acrosscurriculla of motorcycle rider training
courses
Haworth 2005 Study type review
Haworth 2006 Study type cross-sectional survey
Hill 1983 Study type review and description of training system
Imatake 1980 Study type course descriptionreview
Ingebritsen 1990 Intervention risk factors for motorcycle crashes
Jain 2009 Study type cross-sectional descriptive study of motorcycle crashes over a 5 year period
Jonah 1979 Study type cross-sectional survey Outcome scores on motorcycle operator skill test
Jonah 1981 Intervention evaluation of a motorcycle rider skill test not motorcycle rider training
Kadar 1990 Study type review of components of rider training courses
Kelsey 1986 Intervention evaluation of a motorcycle rider skill test not motorcycle rider training
Koch 1980 Study type reviewcourse description
Koch 1990 Study type description of course content
42Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4550
(Continued)
Labbett 2005 Study type essay about possible explanations for motorcycle deaths
Law 2005 Intervention targeted motorcycle safety programme not specifically rider training
Lowes 1990 Study type review and descriptions of rider training
Mahon 1976 Study type review of motorcycle rider training methods
Maxwell 1985 Outcomesrecommendations forprogramme operatorsand otherson howthe motorcycle trainingprogramme
works and how it can be improved
Mayhew 1996 Study type review
McKnight 1987 Study type evaluation of methodological and administrative aspects of a motorcycle operator skill test
McPherson 1976 Outcome knowledge and skills
Mohan 2004 Study type review of evidence of interventions to reduce road traffic injuries
Morgan 1980 Study type reviewcourse description
Motoki 1990 Study type proposal and development of rider training curriculum
Nairn 1992 Study type literature review
Newland 1980 Study type description of development of motorcycle safety course
Newland 1982 Study type descriptionreview of a rider training course
Newland 1990 Study type uncontrolled BeforeAfter study
Ormston 2003 Outcome changes in attitude and self reported behaviour
Osga 1979 Outcome knowledge attitudes and skills
Panou 2007 Intervention new simulation tool to teach driverrider
Peterson 1990 Study factor demographic details of course participants completing the course vs those not completing the
course
Prem 1984 Study type experimental
Study factor examines association between skills and performance in tests
Radin Umar 2006 Intervention motorcycle safety programme not specifically rider training
Rockwell 1990 Study type methodologically flawed study type
43Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4650
(Continued)
Rothe 1987 Study type qualitative study of rider training in general
Rowden 2007 Study type qualitative study
RSC Victoria 1993 Study type review of motorcycle rider training and education in Victoria
Russam 1979 Study type review of training programmes in Japan
Saffron 1981 Study type literature review
Schulz 1990 Study factor predictors of motorcycle crashes
Sexton 2004 Study type cross-sectional analytic analysis
Shephard 1986 Study factor reviewdescription of programme
Simpson 1990 Study type review of motorcycle rider trainingeducation
Sokolov 1990 Study factor review of motorcycle riding courses in USSR
Steffens 1988 Outcome self-reported behaviour experiences and researcher observed riding behaviour
Sudlow 2003 Study type review of motorcycle training in the UK
Tasmania 1995 Study type cross-sectional descriptiveprocess evaluation
Tetard 1985 Study type experimental
Outcome detection of indices
Thackray 1980 Outcome instructional effectiveness user acceptance administrative feasibility
Thompson 1994 Study type cross-sectional survey
Outcome attitudes and knowledge
Tobler 1990 Study type review of course curricula
Transport 1989 Intervention syllabus for rider training
Tuan 2007 Study type cross-sectional study (survey) which does not assess the effects of an intervention but examines
motocyclistsrsquo behaviour
Victor 1980 Study type review
Waterer 1990 Study type review of UK rider training issues
Weller 1989 Study type review of studies evaluating motorcycle safety and education measures
44Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4750
(Continued)
Wells 1981 Study type description of different methods of training
Winn 1986 Study type review of rsquohistoryrsquo of rider training
Wisher 1988 Study type cross-sectional survey
Wood 1987 Study type Interrupted Time Series but did not meet the criteria for an ITS as published by the Cochrane
Effective Practice and Organisation of Care Group
Yamazaki 1990 Study type cross-sectional survey
Outcome riding skills skills retention and attitude
Characteristics of ongoing studies [ordered by study ID]
Burgess 2010
Trial name or title UKs National Ride program
Methods Cohort study
Participants Motorcycle offenders
Interventions Police led diversion program
Outcomes Self reported attitudinal and behavioural measures
Starting date
Contact information
Notes Expected to be completed in early 2010
45Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4850
D A T A A N D A N A L Y S E S
This review has no analyses
A P P E N D I C E S
Appendix 1 Search strategy
We searched the following electronic databases
bull Cochrane Injuries Group Specialised Register (searched 22 Sept 2008)
bull CENTRAL (The Cochrane Library 2008 Issue 3)
bull TRANSPORT 1988 to 200706 (latest date available)
bull ATRI Australian Transport Index (to Sept 2008)
bull MEDLINE (Ovid SP) 1950 to Sept 2008
bull EMBASE (Ovid SP) 1980 to Sept 2008
bull CINAHL (EBSCO) 1982 to Sept 2008
bull WHOLIS (World Health Organization Library Information System) (to Sept 2008)
bull PsycINFO (Ovid SP) 1806 to Sept 2008
bull LILACS (BIREME) (Latin American and Caribbean Health Sciences) (1982 to Sept 2008)
bull ISI Web of Science Social Sciences Citation Index (SSCI) 1970 to Sept 2008
bull ERIC (searched 25 Sept 2008)
bull ZETOC (searched 25 Sept 2008)
bull SIGLE (searched 25 Sept 2008)
All of the databases were searched using the following text-word search strategy The strategy was adapted using either rsquorsquo or rsquo$rsquo as atruncation symbol
1 motor-cycl or motorcycl or motor-bik or motorbik or motor-scooter or motorscooter or scooter or moped or motocycl
or motocicl or moto
2 train or educat or instruct or program or course or qualif or learn or improv
3 1 and 2
H I S T O R Y
Protocol first published Issue 2 2005
Review first published Issue 10 2010
46Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4950
C O N T R I B U T I O N S O F A U T H O R S
Katina Kardamanidis was the first author of the protocol conducted the literature search extracted full studies assessed the studies
according to inclusion criteria assessed quality of the studies compared findings with the second author discussed disagreements with
the third author wrote the drafts and the final review
Rebecca Ivers was the second author of the protocol assessed the studies according to inclusion criteria assessed quality of the studies
compared findings with the first author edited the final draft and provided overall guidance for the process
Alexandra Martiniuk assessed quality of the studies compared findings with the first author discussed disagreements with the third
author and helped write the first draft and the final review
Mark Stevenson contributed to the protocol reviewed methods of studies and contributed to the final draft
Katrina Thistlethwaite conducted a second literature search extracted full studies and contributed to the final draft
D E C L A R A T I O N S O F I N T E R E S T
None known
S O U R C E S O F S U P P O R T
Internal sources
bull No sources of support supplied
External sources
bull National Health and Medical Research Council of Australia Australia
Salary support for Martiniuk Ivers and Stevenson
bull University of Sydney Sesqui Research Grant Australia
Direct research costs and salary support for Kardamanidis
D I F F E R E N C E S B E T W E E N P R O T O C O L A N D R E V I E W
There are no differences between the protocol and the review
I N D E X T E R M S
Medical Subject Headings (MeSH)
lowastMotorcycles Accident Prevention [lowastmethods] Accidents Traffic [lowastprevention amp control] Licensure Program Evaluation Wounds
and Injuries [mortality lowastprevention amp control]
47Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 5050
MeSH check words
Humans
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 750
Table 1 Risk of selection bias (Continued)
Osga 1980 Differences at baseline reported on sex and age
Confounding factors were not dealt with in univariate analysis Riding exposure was accounted for in the
cumulative survival analysis
Selection intervention group were all training participants on the register and the control group were
respondents to a diverse range of recruitment strategies
Satten 1980 Differences at baseline reported on age sex helmet use riding experience and exposure and previous
training
Confounding factors were not dealt with in design or analysis
Selection intervention group was selected from a list of training graduates and the control group was
partly selected from a state register of motorcyclists
Jonah 1982 Differences at baseline reported on age sex riding experience time licensed previous training and alcohol
use
Confounding factors were dealt with through multivariate analysis
Selection intervention group was selected from a list of training graduates and the control group was
randomly selected from the Ontario motorcycle operator file
Mortimer 1984 Differences reported at baseline on age licence status riding experience years licensed and helmet use
Confounding factors age years licensed and rider experience were dealt with through stratification and
riding exposure through standardisation
Selection intervention group was selected from a list of training enrolees and the control group was
selected from motorcycle dealerships during three successive Saturdays
Cooper 1988 Differences reported at baseline on age gender licensure and highway experience
Confounding factors age gender riding exposure rider characteristics rider attitudes and behaviour were
accounted for in a multivariate analysis
Selection there was no information on how the intervention and control groups were selected and from
where
Mortimer 1988 Differences reported at baseline on years licensed riding experience and helmet use
Confounding factors age years licensed and rider experience were dealt with through stratification and
standardisation of riding exposure A regression analysis was also conducted but poorly reported and not
usable
Selection intervention group was selected from a list of training enrolees and the control group was
selected from motorcycle dealerships
Adams 1985 Differences at baseline reported on previous crash rates
Confounding factors not controlled for in analysis
Selection the intervention group was a non-random sample of those offered training within the Australia
Post motorcycle delivery riders the control group were the remaining Post riders
Leung 1987 Differences at baseline reported on age and sex
Confounding factors age and sex controlled for in analysis by stratification
Selection those exposed to training were MRC graduates those unexposed were taken from the licensing
records
5Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 850
Table 1 Risk of selection bias (Continued)
McDavid 1989 Differences at baseline other than matched characteristics were not reported
Confounding factors age ownership of car licence years licensed previous crashesand offences accounted
for through matching Other confounding effects not controlled
Selection intervention group was selected from a list of training graduates and the control group was
selected from Bureau of Motor Vehicle records
Kloeden 1994 Differences at baseline reported were in geographical area of residence and in riding exposure
Confounding factors were accounted for through exclusion of participants from rural areas
Other confounding factors were not reported
Selection the Ridersafe course was phased in based on postcode of riderrsquos home address analyses were
based on a database of those who obtained motorcycle rider learnerrsquos permits of which about 2000 were
Ridersafe riders and about 5000 were control riders
Billheimer 1998 Differences at baseline reported on age and previous riding experience
Other potential confounding factors assessed were riding exposure primary purpose of riding and helmet
use
Confounding factors accounted for by matching (age sex primary purpose of riding) standardisation
(riding exposure) and stratification (previous riding experience)
Selection the intervention group were those who had taken the rider training programme the control
group were California motorcycle riders matched on the above variables but who had not taken a riding
course
Perrino 2002 Differences at baseline reported were a higher number of females in the trained group and a higher
proportion of experienced riders in the untrained group
Confounding factors were not accounted for in the analysis
Selection the intervention group were those who had done the Motorcycle Riding Course the control
group had not done the course Both groups were identified through a register of licensed motorcyclists
Kraus 1975 Population based case-control study
Cases all motorcycle riders with a confirmed medically treated motorcycle injury (including fatalities)
from all hospitals in study base and peripheral area hospitals and all deaths due to motorcycle crashes in
the study base All serious injuries (leading to hospitalisation) and fatalities likely to be included
Community controls a random sample of registered motorcycle owners in the study base
Potential confounders assessed were age sex weight height riding experience motorcycle use charac-
teristics of motorcycle and previous crashes and offences
Confounding factors were accounted for through exclusion of females and stratification by age and
motorcycle useSelection cases and controls drawn from the same population
Hurt 1981 Population based case-control study
Cases and controls not drawn from the same population
Cases motorcycle riders who had a motorcycle crash Not all reported cases included
Community controls motorcycle riders who passed the crash site at the same time of day and day of week
that the crash occurred but 2 years later
Potential confounders assessed were age sex time type of motorcycle roadway alignment helmet use
weather conditions road surface conditions
Confounding factors were accounted for through matching for time of day and day of week that crash
occurred and stratification
Selection cases were recruited from police reports not all cases were included controls were matched
6Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 950
Table 1 Risk of selection bias (Continued)
community controls from randomly selected sites matched for same time of dayday of week as the crash
occurred
Haworth 1997 Population based case-control study
Cases and controls drawn from the same population
Cases motorcycle riders or pillion riders with an injury from a motorcycle crash from hospital and police
records Not clear whether all cases were included
Community controls motorcycle rider or pillion who passed the crash site at the same time of day and
day of week that the crash occurred
Potential confounders assessed were rider factors age marital status licence status riding experience
and exposure alcohol and drugs use of protective gear including helmet use pillion factors motorcycle
factors trip factorsConfounding factors were accounted for through matching for time of day and day of week that crash
occurred and stratification for age blood alcohol level and licence status
Selectioncases were riders or pillions involved in on-road crash recruited from hospitals and police records
not sure if all cases were included controls were matched community controls who passed the crash site
at the same time of dayday of week as the crash
Savolainen 2007 Differences at baseline reported on sex safety helmet use riding ability riding experience and licence
status
Adjusted for these confounders through statistical modelling although exact process unclear
Selection intervention group from list of trainees control group from Indiana Bureau of Motor Vehicles
database Trainees and the control group were invited to participate in surveys
Davis 1997 Differences at baseline not reported but likely No adjustment for any confounders at baseline or during analysis
Selection intervention group from list of riders involved in crashes who were also on a list of training
graduates control group from the same list of riders involved in crashes who were not also on a list of
training graduates
Waller 1992 Differences at baseline not reported but likely
Adjusted for confounders through exclusion of riders younger than 18 years riders without a motorcycle
operator driver license riders with location of residence outside of the Texas metropolitan areas of Austin
through matching by age sex and location of residence through stratification by age and sex and through
standardisation (number of crashes by person-days at risk)
Selection intervention group from the Motorcycle Operator Training class roster (number of trained
riders that the sample was selected from is unclear) control group through a random number generator
generating driver license numbers until an equal sample size was selected as the intervention group
MAIDS 2004 Population based case-control study
Unclear whether cases and controls not drawn from the same population
Cases motorcycle riders who had a motorcycle crash Unclear whether all cases included
Community controls motorcycle riders recruited at petrol stations Unclear whether and how matched
to cases
No potential confounders were reported
There was no report of whether and how potentially confounding factors were accounted for in the study
7Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 1050
Table 2 Risk of allocation bias
Anderson 1980 Allocation concealment unclear
Hall 1988 Allocation concealment unclear
Swaddiwudhipong 1998 Allocation concealment unclear
Table 3 Risk of detection bias
Anderson 1980 Blinding of outcome assessors not reported
Driver records holding police recorded crashes and offences not validated
Hall 1988 Blinding of outcome assessors not reported
Driver records holding police recorded crashes and offences not validated
Swaddiwudhipong 1998 Blinding of outcome assessors not reported
Police and hospital records Post mortem examination for deaths not validated
Adams 1985 Blinding of outcome assessors not reported
Australia Post crash records not validated but argued that underreporting is unlikely
Leung 1987 Blinding of outcome assessors not reported
Driver records holding police recorded crashes and offences
Raymond 1979 Blinding of outcome assessors not reported
Self-report through personal interviews or questionnaires - both methods validated
Osga 1980 Blinding of outcome assessors not reported
Self-report through structured questionnaires validated by cross-checking with police
recorded crash records
Satten 1980 Blinding of outcome assessors not reported
Self-report through structured interviews or questionnaires - both methods were not
validated
Intervention group were given telephone interviews but the control group mailed ques-
tionnaires
Jonah 1982 Blinding of outcome assessors not reported
Self-report through structured interviews not validated Driver records holding police
recorded crashes and offences
Mortimer 1984 Blinding of outcome assessors not reported
Self-report through questionnaires not validated
Intervention group received mailed questionnaires but the control group received ques-
tionnaires personally at the motorcycle dealership
Cooper 1988 Blinding of outcome assessors not reported
Driver records holding police recorded crashes and offences
8Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 1150
Table 3 Risk of detection bias (Continued)
Mortimer 1988 Blinding of outcome assessors not reported
Self-report through questionnaires not validated
Intervention group received mailed questionnaires but the control group received ques-
tionnaires personally at the motorcycle dealership
McDavid 1989 Blinding of outcome assessors not reported
Driver records holding police recorded crashes and offences
Kloeden 1994 Blinding of outcome assessors not reported
Driver records holding police recorded crashes and offences
Billheimer 1998 Blinding of outcome assessors not reported
Police recorded crashes and offences
Perrino 2002 Blinding of outcome assessors not reported
Self-report through mailed out questionnaires police records not validated
Kraus 1975 Blinding of intervention assessors to case or control status not reported
Intervention measured police records self-report though questionnaires not validated
Length of follow up not specified
Hurt 1981 Blinding of intervention assessors to case or control status not reported
Intervention measured face-to-face in-depth interviews (ie no structured question-
naires) Validity and reliability of data ensured by reported high level of quality controlLength of follow up not specified
Haworth 1997 Blinding of intervention assessors to case or control status not reported
Intervention measured self-report either by personal or by telephone interview for
both cases and controls not validated
Length of follow up in the 2 years prior to the interview or more than 2 years ago
Savolainen 2007 Blinding of outcome assessors not reported
Source of outcome data not reported
Length of follow up up to 5 years
Davis 1997 Blinding of outcome assessors not reported
Source of outcome data police records
Length of follow up 1 year
Waller 1992 Blinding of outcome assessors not reported
Source of outcome data Department of Public Safety driver history file and official
accident reports
Length of follow up 3 years
MAIDS 2004 Blinding of intervention assessors to case or control status not reported
Intervention measured face-to-face in-depth interviews (ie no structured question-
naires) No report on whether and how validity and reliability of data was ensured
Length of follow up 2 years
9Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 1250
Table 4 Risk of attrition bias
Anderson 1980 Percentage of participants in analysis 99 forthe total group Unclear forintervention andcontrol groups
separately
Hall 1988 Percentage of participants in analysis 96-97 forall intervention groups and 98for the control group
Swaddiwudhipong 1998 Percentage of participating sub-districts in analysis 100 for both intervention and control sub-districts
Raymond 1979 Percentage of participants in analysis 74 for the intervention group and less than 57 for the control
group
Osga 1980 Percentage of participants in analysis 73 for the intervention group and 100 for the control group
Satten 1980 Percentage of participants in analysis is unknown Initial sample size is unclear
Jonah 1982 Percentage of participants in analysis 23 for the intervention group and 45 for the control group
Mortimer 1984 Percentage of participants in analysis 35 for the intervention group and 84 for the control group
Cooper 1988 Percentage of participants in analysis 98 for the total group Unclear for intervention and control
groups separately
Mortimer 1988 Percentage of participants in analysis 40 for the intervention group and 90 in the control group
Adams 1985 Percentage of participants in analysis 61 forthe total group Unclear forintervention andcontrol groups
separately
Leung 1987 Percentage of participants in analysis unclear because initial sample size was not reported
McDavid 1989 Percentage of participants in analysis 40 of final matched cohort
Kloeden 1994 Percentage of participants in analysis 39 for the intervention group and 6 for the control group
Billheimer 1998 Percentage of participants in analysis unclear percentage of the initial sample size
Perrino 2002 Percentage of participants in analysis unclear percentage of the initial sample size
Kraus 1975 Percentage of participants in analysis 58 of cases 59 of controls
Hurt 1981 Percentage of participants in analysis 18 of cases lt 29 of controls
Haworth 1997 Percentage of participants in analysis 70 of cases 47 of controls
Savolainen 2007 Percentage of participants in analysis 18 for the intervention group and 15 for the control group
Davis 1997 Percentage of participants in analysis 100 for the intervention group and unclear for the control group
10Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 1350
Table 4 Risk of attrition bias (Continued)
Waller 1992 Percentage of participants in analysis unclear percentage for the intervention group and 100 for the
control group
MAIDS 2004 Percentrage of participants in analysis unclear for both cases and controls
R E S U L T S
Description of studies
See Characteristicsof included studies Characteristics of excluded
studies Characteristics of ongoing studies
Our search resulted in 102 studies that seemed relevant of which
we excluded 79 studies because they did not meet the criteria for
considering studies for this review (see Characteristics of excluded
studies)
Basedon ourcriteriawe includeda total of 23 studies in this review
There were three randomised trials (Anderson 1980 Hall 1988
Swaddiwudhipong 1998) two non-randomised trials (Adams
1985 Leung 1987) 14 cohort studies (Raymond 1979 Osga
1980 Satten 1980 Jonah 1982 Mortimer 1984 Cooper 1988
Mortimer 1988 McDavid 1989 Waller 1992 Kloeden 1994
Davis 1997 Billheimer 1998 Perrino 2002 Savolainen 2007)
and four case-control studies (Kraus 1975 Hurt 1981 Haworth
1997 MAIDS 2004)
Over half of the studies were conducted 20 or more years
ago and the oldest was conducted 31 years ago Only three
studies were from this decade (Perrino 2002 MAIDS 2004
Savolainen 2007) Over 60 of the studies were conducted in
the USA (Kraus 1975 Anderson 1980 Osga 1980 Satten 1980
Hurt 1981 Mortimer 1984 Leung 1987 Hall 1988 Mortimer
1988 Waller 1992 Davis 1997 Billheimer 1998 Perrino 2002
Savolainen2007) three inAustralia (Adams 1985 Haworth 1997
Kloeden 1994) three in Canada (Jonah 1982 Cooper 1988
McDavid 1989) onefromthe United Kingdom (Raymond 1979)one from Europe (with sampling areas located in France Ger-
many Netherlands Spain and Italy) (MAIDS 2004) and one
in Thailand (Swaddiwudhipong 1998) Only seven studies were
published in peer reviewed journals (Kraus 1975 Jonah 1982
Mortimer 1984 Cooper 1988 Mortimer 1988 McDavid 1989
Swaddiwudhipong 1998) The other studies were reports forhigh-
way safety and road traffic organisations
The studies by Anderson 1980 Hall 1988 Kloeden 1994
Billheimer 1998 (the basic California Motorcyclist Safety Program
component of the study) and Perrino 2002 examined manda-
tory pre-licence training Cooper 1988 McDavid 1989 and
Swaddiwudhipong 1998 assessed non-mandatory training that
when passed resulted in a licence Raymond 1979 Osga 1980
Satten 1980 Jonah 1982 Mortimer 1984 Adams 1985 Leung
1987 Mortimer 1988 Waller 1992 Davis 1997 Billheimer 1998
(the Experienced Rider Course component of the study) andSavolainen 2007 assessed non-mandatory training that could be
taken by motorcycle riders with and without a licence The case-
control studies (Kraus 1975 Hurt 1981 Haworth 1997) assessed
lsquoanyrsquo type of rider training and therefore could not be grouped by
pre- or post-licence rider training The MAIDS 2004 case-control
study did make a distinction between participants having received
compulsory pre-licence training or additional (non-compulsory)
training We did not find a study that compared one form of pre-
or post-licence training (eg on-road off-road theory or a com-
bination) versus another form of training All studies compared
eithermandatory or non-mandatory formal training versus no for-
mal training with the exception of Haworth 1997 who compared
different levels of training courses (beginner intermediate and ad-vanced)
We have presented a more detailed description of the individual
studies in Characteristics of included studies We have provided
a description below of the types of motorcycle rider training that
were evaluated in the included studies
Types of motorcycle rider training evaluated in the
studies
The types of training were varied Duration of training ranged
from two or three hours to 24 hours administered over two or
three days Most training consisted of a theoretical and a practical
component usually off-road The content ranged from provid-ing riders with manuals lectures discussions showing films and
administering knowledge and skill tests Little information was
provided about the qualifications of trainers who conducted the
training
Six studies (Osga 1980 Satten 1980 Mortimer 1984 Mortimer
1988 Waller 1992 Savolainen 2007) assessed the effectiveness of
the USA-based Motorcycle Safety Foundationrsquos Motorcycle Rider
Course (MSF MRC) This course was released in 1976 It com-
prises eight hours of theory (lectures discussions films and multi-
plechoice tests) a textbook and 12 hours of off-road practice The
Motorcycle Safety Foundation is a national not-for-profit organ-
11Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 1450
isation sponsored by manufacturers and distributors of motorcy-
cles Many USA states have adopted the MSF Basic Rider Courseand Experienced Rider Course (wwwmsf-usaorg) since its release
in 1976
The Motorcycle Operators Skill Test was also developed
by the USA Motorcycle Safety Foundation in coopera-
tion with others (httpwwwwebbikeworldcommotorcycle-
trainingmotorcycle-test) The MOST is a licensing test consist-
ing of a manual knowledge and skill test (comprising eight exer-
cises of increasing difficulty and skill requirements) and a mini-
skill test conducted with a simulator Some studies compared this
alternative licensing test to the standard licensing test However
as the MOST is not a type of rider training we have included only
the two studies that assessed the MOST in combination with one
or more training components in this review Both Anderson 1980and Hall 1988 assessed the MOST in combination with a train-
ing component of three hoursrsquo duration consisting of instructions
about the skills needed to pass the MOST (30 minutes of the-
ory and two and a half hours on the motorcycle) Hall 1988 also
assessed the MOST in combination with the Motorcycle Safety
Foundationrsquos Motorcycle Rider Course as described above
Other USA studies evaluated the Wisconsin motorcycle rider
course a 22-hour off-street programme that teachesthe basic rules
of safe motorcycling (Leung 1987) the USA Maryland Motorcy-
cle Rider Course (MRC) a mandatory (for those younger than 18
years)pre-licence 20-hour coursecomprising eight hours of theory
and12 hours of practice and created by the Maryland Motor Vehi-
cle Administration in 1985 (Perrino 2002) the California Motor-cyclist Safety Program (CMSP) a mandatory (for those younger
than 18 years) pre-licence 16-hour basic course (Billheimer 1998)
and Experienced Rider Course (ERC) of eight hoursrsquo duration
(Billheimer 1998) The CMSP still offers these courses ( http
wwwca-msporg) but they are now MSF courses as described
above Davis 1997 did not give any information about the assessed
rider training
The Canadian studies evaluated the British Colombia Safety
Councilrsquos 37-hour motorcycle safety programme (no further in-
formation on the content was provided) (McDavid 1989) and
the Canada Safety Council Motorcycle Training Program (MTP)
comprising four hours of theory and 16 hours of practice (six
hours on-road) over twoweekends (Jonah 1982) The MTPcoursewas made available in 1974 and is still being offered in Canada
(httpwwwsafety-councilorgtrainingmtpmtphtm) Cooper
1988 did not give any information about the assessed rider train-
ing However the study was conducted in British Colombia in
Canada in 1988 the same time as the McDavid 1989 study was
conducted
In the United Kingdom Raymond 1979 assessed the non-manda-
tory RACACU (Royal Automobile ClubAuto-Cycle Union)
training scheme The training programme consisted of either 24
one-hour sessions or 13 two-hour sessions Twelve hours of the
training focused on theory about basic operational procedures
traffic laws and highway codes The practical session focused on
basic operational skills riding on a training area and riding in traf-fic with an instructor A proficiency test was administered at the
end of the course
One study from rural Thailand (Swaddiwudhipong 1998) evalu-
ated a community-based rider education programme consisting of
theory on motorcycle injury risk helmet use traffic laws vehicle
regulations and traffic signs and a written and skill test leading to
a licence
TheAustralian studies evaluateda rider trainingcourse of 12 hours
over two consecutive days (four hours of theory and eight hours of
practice)administeredto AustraliaPosttelegram deliverers(Adams
1985) and a mandatory pre-licence Ridersafe course consisting
of two tiers the first tier being a training course on the basics
of motorcycle riding where a pass entitled the rider to obtain alearnerrsquos permit and the second tier being a skill test where a pass
entitled the riderto obtain a probationary motorcycle licence The
training course consisted of two four-hour training sessions on
separate days involving instruction videos and actual riding on
motorcycles The skill test consisted of a four-hour preparation
course of 74 minutes of lecturediscussion 38 minutes of video
and 75 minutes of on-course riding which was the testing phase
(Kloeden 1994) The Rider Safe course is still being provided by
Transport South Australia a state government organisation (http
wwwtransportsagovau )
The duration or content of rider training in the four case-control
studies (Kraus 1975 Hurt 1981 Haworth 1997 MAIDS 2004)
was not described
Risk of bias in included studies
We describe the methodological quality of the included studies
by study type below More detailed information can be found in
Table 1 Table 2 Table 3 and Table 4
When considering the findings from these studies it is important
to note the quality (or lack thereof) of reporting of the studies
Quality of reporting is the extent to which a report provides in-
formation about the design conduct and analysis of a study ( Juumlni2001) Huwiler-Muumlntener 2002 found that even though trials
may be well reported it does not mean that the methodological
quality is of equal standard and well-conducted trials may never-
theless be reported badly The reporting quality of most studies in
this review was poor Most studies were old and did not meet con-
temporary reporting standards This made data extraction com-
plicated It also made it difficult to contact authors for additional
information
Because of significant heterogeneity between studies as well as the
poor quality of most studiespooling of studyresults to calculatean
overall measure of effect was not appropriate Instead we provided
a description of our judgement of risk of bias in the text
12Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 1550
Risk of bias in the randomised trials
There were three randomised trials (Anderson 1980 Hall 1988Swaddiwudhipong 1998) Anderson 1980 and Hall 1988 had an
initial sample size of 41188 and 26464 motorcycle licence ap-
plicants respectively Both studies assessed the effectiveness of the
pre-licence MOST with a trainingcomponent Swaddiwudhipong
1998 evaluated the effectiveness of a community-based rider ed-
ucation programme in six districts
Allocation bias
Anderson 1980 reported using random birth date tables to allocate
participants to either the intervention or control group Hall 1988
reported the use of a computerised algorithm set in a central of-
fice but also used manual random assignment when the computer
went down Swaddiwudhipong 1998 reported lsquosimple random
samplingrsquo for the delivery of the intervention where the sampling
frame included nine districts Three districts were excluded since
they were too hard to access The remaining six districts were ran-
domly allocated to the intervention (three) and the control (three)
groups Systematic sampling wasthen used ordered by village size
for the villages within districts Allocation concealment was not
reported in any of the studies The descriptions of randomisation
in these studies suggest that there may have been limitations to
the process resulting in a lack of equivalence of the intervention
and control groups at baseline and therefore the possibility of
allocation bias Swaddiwudhipong 1998 reported that before theintervention the annual incidence rate of motorcycle-related in-
juries was slightly (but not statistically significant) higher in the
intervention areas than in the control areas Participants in the
intervention districts were also more likely to have a valid licence
and to wear a helmet more often than participants in the control
districts Anderson 1980 measured differences at baseline on the
participant and prior driving record variables and found these to
be negligible Hall 1988 did not report on baseline differences
between the intervention and control groups As the Anderson
1980 and Hall 1988 studies had sample sizes of over 40000 and
26000 respectively and used some form of randomisation it is
likely that any confounding factors at baseline would have been
equally distributed between comparison and intervention groups
Selection bias
After randomisation Anderson 1980 and Hall 1988 excluded
08 and 3 participants respectively because of assignment to
the wrong group administration of the wrong intervention or
other reasons The resulting lack of analysis by rsquointention-to-treatrsquo
can introduce biasby removing the effectivenessof randomisation
Nonetheless the percentage excluded was small and the number
of overall participants large so it is unlikely to have had a major
effect on the final results In the Swaddiwudhipong 1998 study
villages were systematically sampled into intervention or control
groups The intervention was implemented and outcomes of in-
jury and death were measured at a community level Interventionand control districts were randomly selected however there were
only three intervention and three control communities
Detection bias
None of the randomised trials reported blinding of outcome as-
sessors Data collection tools used by Anderson 1980 and Hall
1988 were driver records holding police recorded crashes and of-
fences Swaddiwudhipong 1998 used police and hospital records
at a district level and post-mortem examination for deaths None
reported on validity or reliability for any of the instruments
Attrition bias
Anderson 1980 and Hall 1988 had driver records available for
96 to 100 of participants They also collected self-reported
data but the response rate to the questionnaires was very low (low-
est response rate 42 and 30 respectively) Attrition bias in
Swaddiwudhipong 1998 was not an issue as outcome data were
available for all sub-districts included in the trial
The intervention reported in Anderson 1980 and Hall 1988 tri-
als had an unanticipated effect on the participants In both trials
an average of only 28 of licence applicants in the intervention
group eventually obtained their licence compared to 35 in the
control group This indicates that the licensing process in generalseems to have deterred applicants from obtaining a licence and
although no statistical tests of the difference were conducted this
deterrence appears even greater for those who had rider training
This is not true attrition bias as the studies did not experience
differential loss to follow up from outcome measurement Instead
differential attainment of the intermediary step of licensure pre-
cluded participants from being at risk for the outcomes of interest
they are not driving a motorcycle hence they are not at risk for
motorcycle-related crashes injury death or offences In short it is
not possible to assess whether a reduction in crashes in the inter-
vention group would be due to changes in riding habits as a result
of training or simply due to delayed or never attained licensure
Intervention integrity
Neither Anderson 1980 nor Hall 1988 reported on the consis-
tency in the way the intervention was administered the possible
lsquocontaminationrsquo of participants in the intervention group by other
types of formal or informal rider training or of participants in the
control group by the intervention Swaddiwudhipong 1998 did
not report on the number of villagers in each group that actually
attended the intervention However they did report on a number
of other potential training courses that were being implemented
at the same time
13Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 1650
Analysis
Anderson 1980 and Hall 1988 acknowledging the lsquodeterrent ef-fectrsquo described above conducted two separate analyses The first
analysis included the total study population of licence applicants
over two years from the time of their application (the population
denominator) The second analysis was performed on the licensed
(intervention and control) subgroups only two years from the
time of obtaining their licence (the licensed rider denominator)
As only a third were licensed and exposed to riding the motorcy-
cle it was impossible to determine whether reported differences
in crash rates across groups were due to the intervention or to
other lsquoconfoundingrsquo characteristics of those who were determined
in their efforts to obtain the licence When analysis was restricted
only to those who gained their licence the same difficulties re-
mained The benefits of the randomisation process were thus lost
due to this lack of licensure of riders during the process As there
were only three intervention and three control groups in the study
reported by Swaddiwudhipong 1998 analyses were hindered by
smallnumbers for outcome comparisons betweenthe intervention
and the control districts
Risk of bias in the cohort studies and non-randomised
trials
We discuss the methodological quality of cohort studies (seven
prospective studies (Raymond 1979 McDavid 1989 Waller 1992
Kloeden 1994 Billheimer 1998 Perrino 2002 Savolainen 2007)and seven retrospective studies (Osga 1980 Satten 1980 Jonah
1982 Mortimer 1984 Cooper 1988 Mortimer 1988 Davis
1997)) and non-randomised trials (Adams 1985 Leung 1987) in
the one section as the biases for the two study types are similar
The quality of these studies was in general poor
Selection bias
The confounding factors expected to have been addressed in the
studies were sex age riding experience prior driving records (of-
fences crashes) previous training and the following proxies for
risk taking behaviour helmet use speed and alcoholdrug use All
of these factors have been shown to be associated with an increasedrisk of motorcycle crash (Peden 2004 Zambon 2006) The pos-
sible source of bias in cohort studies and non-randomised trials
is that groups may differ on important characteristics at baseline
participants lsquoself-selectrsquo into either the rider training group by de-
ciding to undertake a rider training course or into the lsquocontrol
grouprsquo by deciding not to undertake a course Participants who
undertake a non-mandatory rider training course may differ from
those who choose not to on characteristics such as age sex rid-
ing experience and riding behaviour as well as other possible con-
founders For example riders who take a training course may be
more safety-conscious or on the contrary may do so because they
have previously experienced a crash or because they perceive that
they have poor riding skills Therefore ideally all participants in
the intervention group need to be included in the study as wellas a random sample of controls taken from the same population
as the intervention group If that is not possible intervention and
control groups either need to be carefully matched to make them
comparable or differences at baseline need to be adjusted for in
the analysis
Most of the studies using a non-randomised trial or cohort design
compared an intervention group to a control group that differed
on a number of important factors The studies in general failed to
adjust for important differences found at baseline and their results
cannot be meaningfully interpreted Ten studies (Raymond 1979
Osga 1980 Satten 1980 Jonah 1982 Mortimer 1984 Mortimer
1988 McDavid 1989 Waller 1992 Davis 1997 Savolainen 2007)
assessed intervention and control groups that were not drawnfrom the same population In most of these cases the intervention
group was selected from a list of training graduates who were usu-
ally young female and less experienced riders whereas controls
were recruited from a general population of motorcycle riders who
were usually older male and experienced riders For three stud-
ies it was unclear how they were selected (Leung 1987 Cooper
1988 Perrino 2002) Only Kloeden 1994 recruited groups from
the same population of motorcycle applicants and Adams 1985
from a group of professional riders Nevertheless due to non-ran-
dom selection of intervention and control groups differences at
baseline are likely even though study groups were from the same
population Three studies adjusted for some confounding fac-
tors found at baseline through matching (McDavid 1989 Waller1992 Billheimer 1998) However matching has limitations in
termsof controlling for multiple confounders especially those that
are unknown to influence the effect of rider training on the out-
comes One study matched the control and intervention groups
by excluding controls (approximately 3500 (66 of initial con-
trol group) that differed from the intervention group) (Kloeden
1994) Five studies (Osga 1980 Satten 1980 Adams 1985 Davis
1997 Perrino 2002) did not adjust for differences between study
groups at baseline either in the study design or analysis Results
from these studies are therefore not reliable The other six stud-
ies (Raymond 1979 Jonah 1982 Mortimer 1984 Cooper 1988
Mortimer 1988 Leung 1987 McDavid 1989 Billheimer 1998)
adjusted for some differences between groups at baseline by eitherbi-variate or multi-variate analyses or stratification and standardi-
sation of riding exposure (outcomemiles driven) or stratification
by age group The limitation of stratification is that it is difficult to
control for many variables simultaneously because a large number
of strata will need to be generated resulting in small numbers that
cannot be meaningfully analysed Savolainen 2007 developed sta-
tistical models to assess the probability of a crash after taking the
rider training course in an intervention and control group while
examining differences by sub-groups (eg age) Participation rates
were also very low in this study with only 5584000 (1395)
in the eligible intervention group and 5884000 (147) for the
14Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 1750
control group completing surveys
Detection bias
None of the studies reported blinding of outcome assessors to the
intervention or control status of the participants Seven studies
used driver records containing police recorded data as their main
measure of outcome (Leung 1987 Cooper 1988 McDavid 1989
Waller 1992 Kloeden 1994 Davis 1997 Billheimer 1998) An-
other four studies (Raymond 1979 Satten 1980 Mortimer 1984
Mortimer 1988) only used self-report Raymond 1979 was the
only study using self-report as the main outcome measurement in
which study instruments were validated Self-report might prevent
the underreporting that has been described for the use of police-
driver records (see rsquoRisk of biasrsquo in the randomised trials above)but has other risks of bias especially if questionnaires or inter-
view methods have not been validated and tested for reliability
Recall bias arises from mistakes in recollecting events both be-
cause of failures of memory and because of looking at things lsquowith
hindsightrsquo and possibly changed views (httpwwwcochraneorg
resourcesglossaryhtm) As many studies were retrospective risk
of recall bias is likely and may have increased with length of follow
up For studies that used self-reported outcomes only it is also
possible that social desirability response bias played a role where
participants in the training group may have felt it was lsquoexpectedrsquo
to report good driving histories Multiple methods to measure the
same outcome for both groups decreases the risk of detection bias
for example because self-reported answers can be cross checkedwith other sources of information such as driver records
In this regard three studies (Osga 1980 Jonah 1982 Perrino
2002) used both self-report and driver records Osga 1980 found
that for both intervention and control groups more crashes were
reported through self-report than on state crash records an aver-
age of 54 of self-reported crashes could be found back on state
crash records One industry-based study (Adams 1985) used oc-
cupational health records which were not validated but which are
unlikely to suffer from underreporting due to work cover claims
Savolainen 2007 used information on rider crashes for a five-year
period but did not specify from where this information was ob-
tained
Attrition bias
The majority of studies (Raymond 1979 Satten 1980 Adams
1985 Leung 1987 Kloeden 1994 Davis 1997 Billheimer 1998
Perrino 2002) inadequately reported on losses of participants to
the study We can therefore not be certain that there were no sys-
tematic differences in the characteristics between the intervention
and control group participants Savolainen 2007 reported the re-
sponse rates to their baseline survey but did not report on the per-
centage of outcome data that was available for their study sample
A similar deterrent effect of rider training as described for the ran-
domised trials (Anderson 1980 Hall 1988) occurred in two stud-
ies Jonah 1982 reported that 65(14992310) of the trained rid-
ers stopped riding compared to 47 (9492029) of the untrainedriders They excluded non-active and non-licensed riders before
the study commenced so the differential choice to continue riding
after training did not affect their analyses Leung 1987 reported
that the percentage of participants in the intervention group ob-
taining their licence was 50 (no number provided by authors)
lower than in the control group This may have biased the mea-
surement of outcomes because unlicensed participants will not be
exposed to riding the motorcycle and therefore not be at risk of
being involved in a crash or violation As explained above this
makes it difficult to assess whether the reduction in crashes in the
intervention group is due to a change in driving habits or simply
due to delayed licensure
Mortimer 1984 and Mortimer 1988 describe that about 44 and30 of course graduates respectively were excluded from analysis
because they had not ridden a motorcycle in the previous year
However because these studies were retrospective and the controls
were activeriders at the time of recruitment a decision on whether
the deterrent effect was differential between the two groups could
not be made Though not reported quantitatively Kloeden 1994
also describes the effectof rider training deterringindividuals from
becoming licensed and hence exposure to riding and risk of crash
death or offences involving a motorcycle Osga 1980 reported
that 27 of the trained riders did not ride after the course and
excluded these from the analysis
Use of different data collection strategies may have contributed to
differential loss to follow up in the cohort studies Satten 1980for example interviewed the intervention group over the tele-
phone but mailed questionnaires to the control group Mortimer
1984 and Mortimer 1988 distributed questionnaires in person to
the control group but mailed them to the intervention group
This may have played a role in the observed differential response
rates where more subjects responded in the personally approached
groups
Survival bias
For those studies that assessed offences crashes and injuries but
not death survival bias may have also played a role Motorcyclists
who were fatally or seriously injured in crashes during the follow-
up period after the intervention would have been less likely or
unable to respond to surveys This could bias the effect measures
towards or away from the null None of the studies wrote about
this issue or examined the characteristics of those lost to follow up
in order to address the possibility of survival bias
Intervention integrity
Although the interventions were intended for the entire lsquoexposedrsquo
groups most studies reported difficulty in tracking participants
whoreceivedthe interventionand those whodid not Adams 1985
15Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 1850
estimatedthat 62of thetotalstudysample received the interven-
tion The remaining studies did not report the percentage of thestudy population that received the intervention as was intended
or reported that this percentage was unknown (Raymond 1979
Kloeden 1994 Satten 1980 Jonah 1982 Mortimer 1984 Leung
1987 Cooper 1988 Mortimer 1988 Billheimer 1998 Perrino
2002 Savolainen 2007) Only Raymond 1979 reported the degree
to which their intervention was delivered as intended No studies
reported on possible lsquocontaminationrsquo of participants by other inter-
ventions than the one to which they were allocated Because Osga
1980 Satten 1980 Jonah 1982 Mortimer 1984 Cooper 1988
Mortimer 1988 Waller 1992 and Davis 1997 were retrospective
studies and participants were recruited from a list of graduates it
was known that they had completed and passed the course
Analysis
In some studies some participants who reported at baseline that
they were untrained decided to enroll in a rider training pro-
gramme after the baseline survey The non-randomised trial by
Adams 1985 experienced this move of subjects from the untrained
to the trained group but failed to conduct analyses by lsquointention-
to-treatrsquo This can bias the results in randomised and non-ran-
domised trials if participants who changed groups differ signifi-
cantly with regard to potentially confounding characteristics On
the other hand Perrino 2002 also experienced this move of sub-
jects from theuntrained to the trained groupmore than 100riders(5 of the total sample) originally stated that they were untrained
at baseline but then received training between the baseline and
follow-up surveys Yet since the Perrino 2002 study was a cohort
their analyses then proceeded by including those riders who were
lsquonewlyrsquo trained as exposed Riders who were lsquonewlyrsquo trained since
baseline but with a programme other than the one under study
were excluded from the analysis Nevertheless confounding fac-
tors were not accounted for in the analyses
McDavid 1989 matched exposed with unexposed subjects but
could not find suitable matches for the majority (60) of their
subjects They could therefore only conduct analysis on the result-
ing40 (139346) which is an inappropriate andinefficient anal-
ysis Some studies that reported on the issue of non-active licensedriders(Satten 1980 Jonah 1982 Mortimer 1984 Mortimer 1988
Kloeden 1994) excluded non-active riders from analysis intro-
ducing selection bias as it is likely that riders who continued to
ride differed with regards to potentially confounding character-
istics across intervention and control groups Kloeden 1994 for
example conducted several analyses to assess the effect of various
sub-sets of the control group Original analyses included all con-
trols and led to erroneous results since only a small proportion of
controls hadgone on to get a licence after their learnerrsquos permitand
were actually exposed to the possibility of experiencing a crash
Subsequent analyses compared only intervention and control sub-
jects who had a licence Waller 1992 reported crashes by person-
days of exposure but did not explain how they could be sure that
each study intervention and control participant was an active riderand actually exposed to the risk of experiencing a crash
Potential confounders that are unequally distributed by study arm
at baseline need to be dealt with in the analysis Satten 1980
Adams 1985 and Davis 1997 did not take into account any con-
founding factors either by design or in the analysis McDavid
1989 Waller 1992 and Billheimer 1998 accounted for potential
confounding factors at the design stage by matching the subjects
in the intervention and control groups Other studies (Raymond
1979 Mortimer 1984 Mortimer 1988 Perrino 2002 and also
Waller 1992) adjusted for one or two confounders by stratifica-
tionor standardisation (eg crashesmiles driven age group) Osga
1980 conducteda univariate analysis of crash rates in which no po-
tential confounders were accounted for but also calculated cumu-lative crash rates by mileage travelled Only Jonah 1982 Cooper
1988 and Savolainen 2007 conducted a multivariate analysis
Risk of bias in the case-control studies
A case-control design is suitable for the assessment of risk fac-
tors for outcomes such as motorcycle crashes injuries or deaths
but not appropriate for the evaluation of an intervention such
as motorcycle rider training It can potentially result in spurious
conclusions regarding the efficacy of the intervention because the
intervention has not been applied at random to an intervention
and control group Even with the control of possible confounders
using a case-control design to evaluate an intervention can lead tothese kinds of results since those who choose to undertake rider
training may differ on important prognostic factors that affect the
outcome compared to those who do not choose to participate in
rider training
Selection bias
A potential source of bias in case-control studies as with all other
study types where the intervention is not randomly allocated to
participants is that comparison groups are likely to differ on po-
tentially confounding characteristics This means that ideally all
cases need to be included in the study as well as a random sample
of controls taken from the population that gave rise to the cases If that is not possible cases and controls either need to be carefully
matched to make themcomparable or differences at baseline need
to be adjusted for in the analysis All four case-control studies were
population based Kraus 1975 included all fatalities and injured
cases leading to hospitalisation and appropriately recruited a ran-
dom sample of controls from the same population Hurt 1981 did
not manage to recruit all or a random sample of cases Only half
of the notifications of crashes to the police were received in time
by the researchers the research team for unreported reasons was
only able to respond to half of those again Data collection was
therefore completed for only about a fifth of all cases Neverthe-
less this was still a large sample of 900 cases Haworth 1997 using
16Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 1950
a similar study design was more likely to have included all cases
but this is not entirely clear from the study reported It is unclearfor example whether fatalities were included as they did not use
coronerrsquos reports for the recruitment of cases The MAIDS 2004
study was conducted in five European countries but it was unclear
how cases were recruited whether they were all cases during the
two years that the study lasted or a random sample of all cases
what the response rates were etc Controls were recruited from
randomly selected petrol stations but it was likewise not clear what
the eligibility criteria for inclusion in the study were how they
were recruited (eg randomly) and what the response rate was
Hurt 1981 and Haworth 1997 both matched controls to cases by
recruiting themat the sametime ofdayandsamedayof weekas the
crash occurred Hurt 1981 however recruited controls two years
later andit is likelythat there would have beendifferences betweenthe two comparison groups One important difference was that
during data collection for cases there was very little specialised
motorcycletrainingavailable in the studyregionwhereas two years
later when control data were collected there was some training
available
These four case-control studies also suffered from selection bias
due to potential differences between those that gave consent to
participate in the study and responded to interviews andquestion-
naires and those that did not None of the studies have assessed
such differences although Hurt 1981 did acknowledge that con-
trols giving consent to cooperate may have differed from those
motorcyclists who did not
Detection bias
Measures that can be taken to reduce detection bias in case-con-
trol studies are blinding of interviewers for case or control status
blinding of the participant for the study factor of interest validat-
ing measurement instruments and using these in an identical way
for cases and controls and cross checking self-reported informa-
tion with other more objective sources such as registers None of
the studies reported blinding of the intervention assessors to case
or control status but all used consistent measurement methods for
both case and control groups
All case-control studies used self-report as their main measure toassess whether cases and controls had received rider training and
if so the type of training The major potential source of bias in
case-control studies when self-report is used as a measure is recall
bias One way of reducing recall bias is by double-checking self-
reported data with more objective data Kraus 1975 also used po-
lice accident reports for information on rider characteristics but
such information was not available about rider training Haworth
1997 and MAIDS 2004 did not report on any measures to limit
recall bias Hurt 1981 reported further questioning of cases by
the interviewer when answers did not make sense or when par-
ticipants seemed to contradict themselves as part of the in-depth
unstructured interview technique This qualitative research tech-
nique can result in valuable information but as interviewers were
not blinded to case or control status of the participants it may have led to more thorough examination of cases compared to the
controls and thus differential detection bias
Intervention integrity
The level of detail about the rider training provided in the case-
control studies differed significantly but was minimal for each
Kraus 1975 simply asked whether lsquoanyrsquo motorcycle rider training
had lsquoeverrsquo been carried out Hurt 1981 asked participants whether
they hadbeen self taught informally taught or completed formally
taught rider training courses MAIDS 2004 asked whether partic-
ipants had had pre-license training additional (non-compulsory)
training other training or no training whereas Haworth 1997asked about the number of courses that a participant had com-
pleted whetherthey were basic intermediate or advanced courses
and how long ago the training had been done None of the case-
control studies reported collecting information about further de-
tails about any training programmes (eg type of training cur-
riculum duration of training)
Analysis
Kraus 1975 accounted for confounding factors through the exclu-
sion of females and stratification by age and frequency of motor-
cycle use Hurt 1981 matched cases and controls for time of day
and day of the week that the crash occurred and did not adjust forany confounding factors in the analysis Haworth 1997 matched
groups for time of day and day of the week that the crash occurred
and stratified by age blood alcohol level and licence status in the
analysis None of the studies used multivariate methods MAIDS
2004 did not provide any information about any participant char-
acteristics other than the study exposures of interest and did not
describe any methods to adjust for potential confounding factors
Effects of interventions
We have presented the results in a descriptive manner due to the
substantial heterogeneity in study types interventions and out-comes and the generally poor quality of the studies We have
grouped the interventions by mandatory and non-mandatory pre-
licence rider training and post-licence rider training and further
by study type and by outcome (crashes injurious and fatal crashes
and offences) We discuss results that are reported as statistically
significant (whether a P value was reported or not)
Pre-licence rider training - mandatory
Randomised trials
17Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2050
Crashes
Total group of participants
Anderson 1980 found a reduction in motorcycle crash rates in
the intervention group (MOST skill test with remedial training)
compared to the control group (who received the skill testwithout
a training component) The percent differenceranged from 143
six months after the programme (P lt 005) to 211 (P lt 0001)
after one year and 127 (P lt 001) after two years Hall 1988
found no statistically significant difference in crash rates between
intervention groups at any point in time
Subgroup of participants who obtained their licence only
Anderson 1980 found a reduction in motorcycle crash rates in the
intervention arm but this was statistically significant at one year
only (24 to 218 reduction P lt 005) Hall 1988 found no
statistically significant differences
Offences
Total group of participants
The percentage of offences in Anderson 1980 after six months
and one and two years was slightly higher in the intervention
arm compared to the control group but these differences were
not statistically significant Likewise the study by Hall 1988 did
not find a statistically significant difference in offences between
groups
Subgroup of participants who obtained their licence only
Hall 1988 found no significant difference in offence rates between
the intervention and control groups Anderson 1980 found thatthe percentage of offences increased in the intervention arms com-
pared to the control group This difference was statistically signif-
icant after two years (51 to 115 increase P lt 005)
Cohort studies
Crashes
Neitherof the cohort studies (Kloeden1994 and Billheimer1998)
found a statistically significant difference in crash rates between
the intervention and control groups
Offences
Billheimer 1998 (for the California Motorcyclist Safety Program
course) and Perrino 2002 did not find a statistically significant
difference in offence rates between the intervention and control
groups
Case-control studies
Crashes
MAIDS 2004 found that a slightly higher percentage of controls
(not involved in a crash) had had pre-license rider training than
cases (involved in a crash) This difference was not statistically significant (50 of controls vs 477 of cases OR 091 95 CI
076 to 110)
Pre-licence rider training - non-mandatory
Randomised trial
Injuries and deaths
Swaddiwudhipong 1998 found a significantly lower injury ratein the intervention villages (1051000 population 95 CI 92
to 120) compared to the control villages (1691000 population
95 CI 152 to 185) one year after the intervention This differ-
ence was not maintained two years after the intervention (127
95 CI 112 to 143 intervention 151 95 CI 135 to 167
control) There was no significant difference in death rates be-
tween groups
Cohort studies
Crashes
Cooper 1988 foundno statistically significant differencein crashes
between the intervention and control groups McDavid 1989
found a lower crash rate in the intervention group (14139) com-
pared to the control group (23139) but no test of statistical sig-
nificance was undertaken
Injuries and deaths
McDavid 1989 found that the intervention group had fewer in-
jurious crashes (814 crashes) compared to the control group (16
23 crashes) but no statistical significance test was reported
18Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2150
Offences
Cooper 1988 foundno statistically significantdifference in offencerates between the intervention and control groups McDavid 1989
foundno statistically significantdifference in offence ratesbetween
the intervention and control groups
Non-mandatory rider training
Non-randomised trials
Crashes
Neither Adams 1985 nor Leung 1987 found a statistically signif-icant effect of rider training on crash rates
Offences
Leung 1987 found a statistically significant lower offence rate in
the intervention group compared to the control group for both
males and females and for all age groups except for women 35
years or older (P values not reported)
Cohort studies
Crashes
Osga 1980 in the univariate analysis reported a lower percentage
of crashesin theintervention group (that continued to ride) (28
168595) than in the control group (34 112329) In the cu-
mulative crash rate analysis however Osga1980 found statistically
significant (at the P lt 001 level) higher crash rates in the interven-
tion group than in the control group at allmileage intervals (1000
to 29999 miles) travelled After riding more than 30000 miles
(when over 50 of the trained riders had been involved in a crash
compared to approximately 40 in untrained riders) this differ-
ence was no longer statistically significant Satten 1980 found a
lower percentage of crashes in the intervention group (13 969)
compared to the control group (25 2079) but the differencewas not assessed statistically Davis 1997 found a lower crash rate
in trained riders but it was unclear how this was calculated be-
cause the size of the control group was not reported Jonah 1982
and Billheimer 1998 (for the Experienced Rider Course) found no
statistically significant difference in crashes between the interven-
tion and control groups Although the difference between those
who completed the Experienced Rider Course and no course in
Billheimer 1998 neared significance at six months after the course
(P = 007) large P values were observed for one year after (P =
030) Raymond 1979 found a statistically significant higher crash
rate for the intervention group (1861000 miles ridden) com-
pared to the control group (1181000 miles ridden) in both males
and females (P lt 0001) especially during the first 5000 miles of
riding Waller 1992 found a statistically significant higher crashrate for the intervention group (27 crashes in three years or 070
crashes10000 person days) compared to the control group (5
crashes in three years or 013 crashes10000 person days) with a
crude relative risk of 538 (X 2=1513 plt005) Mortimer 1984
and Mortimer 1988 found no statistically significant difference in
crashes between the intervention and control groups
Savolainen 2007 found that the intervention group (those who
completed the Basic Rider Course) was 44 more likely to be
involved in crashes than the control group and those who com-
pleted the course multiple times were an additional 180 more
likely to be involved in a crash This finding was based on low
numbers and no P value or confidence level was given for either
result
Injuries and deaths
Satten 1980 also found a lower percentage of injurious crashes
(33 39) in the intervention group compared to the control
group(60 1220) although again no statistical significance test
was reported
Offences
Jonah 1982 did find a significantly lower self-reportedoffence rate
in the intervention group (15) compared to the control group(34 P lt 001) (no numbers given) Satten 1980 also found
a lower self-reported offence rate in the intervention group (1
71 versus 19100) but again this was not tested for statistical
significance Mortimer 1984 Mortimer1988 and Billheimer 1998
found no statistically significant differencein offence ratesbetween
the intervention and control groups
Case-control studies
CrashesMAIDS 2004 found that a slightly higher percentage of cases
(involved in a crash) had had additional non-mandatory rider
training than controls (not involved in a crash) This difference
was not statistically significant (17 of cases vs 12 of controls
OR 146 95 CI 064 to 337)
Rider training unspecified
Case-control studies
19Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2250
Injuries and deaths
Hurt 1981 found that the odds of injurious crashes were lower inthose who had received professional training compared to those
who had not Kraus 1975 found that in the younger age group
(less than 20 years) and the older age group (over 35 years) a larger
percentage of injured riders (12 and 21 respectively) reported
prior motorcycle rider training compared with the uninjured con-
trol group (6 and 5 respectively) (P lt 001) (no numbers
given) Haworth 1997 reported several different outcomes one
beingthat having completed at leastone trainingcoursecompared
to no training courses at all had no significant effect on the odds
of crashing (afteradjustment for age blood alcohol level or licence
status (learner permitprobationaryfull)) When comparing three
different levels of training courses (beginner intermediate and ad-
vanced) with no training no significant differences were found
in the odds of crashing However after adjustment for age and
blood alcohol level completing an advanced course as compared
to a beginners course was associated with a significant decrease in
the odds of severe motorcycle crashes resulting in serious injury or
death (OR 02 95 CI 01 to 08)
D I S C U S S I O N
The objective of this review was to quantify the effectiveness of
pre- andpost-licence motorcycle rider training on the reduction of
traffic crash involvement injuries deaths and offences committedby motorcycle riders Most included studies were cohort studies
There were a few randomised and non-randomised trials and case-
control studies
All studies compared either mandatory or non-mandatory formal
training with no formal training except for Haworth 1997rsquos case-
control study that compared levels of training (beginner interme-
diate and advanced) to each other
We found studies that evaluated the effectiveness of manda-
tory pre-licence training in reducing crash and offence rates
(Anderson 1980 Hall 1988 Kloeden 1994 Billheimer 1998
(the basic California Motorcyclist Safety Program component of
the study) Perrino 2002) non-mandatory training that when
passed resulted in a licence (Cooper 1988 McDavid 1989
Swaddiwudhipong 1998) non-mandatory training that could be
taken by motorcycle riders with or without a licence (Raymond
1979 Osga 1980 Satten 1980 Jonah 1982 Mortimer 1984
Adams 1985 Leung 1987 Mortimer 1988 Waller 1992 Davis
1997 Billheimer 1998 (the Experienced Rider Course compo-
nent of the study)) and unspecified rider training on crash injury
and death rates (Kraus 1975 Hurt 1981 Haworth 1997 MAIDS
2004)
The studies provided conflicting evidence regarding motorcycle
rider training which is likely due to the poor design quality and
reporting of the studies Most of the studies included in the review
were likely to be adversely affected by significant selection biasalong with substantial detection bias(measurement and recall bias)
which limits our ability to make conclusive comments
Mandatory pre-licence rider training
There were two RCTs three cohort studies and one case-control
study that examined pre-licence ridertraining The two RCTs eval-
uated the Motorcycle Skills Operating Test (MOST) (Anderson
1980 Hall 1988) Anderson 1980 found a statistically significant
reduction in crashes in the intervention group at six months one
and twoyearsfollow up However when only participantswho ob-
tainedtheirlicence were included in theanalysis thereduction wasstatistically significant only at one year follow up The other RCT
(Hall 1988) found no statistically significant differences in crash
rates at any point of time neither in the total population group
nor in the licensed subgroup Neither of the two cohort stud-
ies assessing crash rates found a statistically significant difference
in crash rates between intervention and control groups ( Kloeden
1994 Billheimer1998) andthe case-control study (MAIDS 2004)
also did not find a statistically significant difference in the odds of
experiencing a crash between trained and untrained riders With
regards to offences neither of the RCTs found a statistically sig-
nificant difference in offence rate between the intervention and
control groups in the analysis of the total group of participants
entering the trial However Anderson 1980 found a statistically significant increase in offence rates in the intervention group but
only after two years of follow up andonly in the analysis of partici-
pants who obtained their licence The two cohort studies assessing
offence rates found no statistically significant difference between
the groups (Billheimer 1998 Perrino 2002)
None of the five studies (RCTs or cohort) found a reduction in
crashes or offences following mandatory pre-licence rider training
except for Anderson 1980 who observed a statistically significant
reduction in crashes in the intervention group at one year in both
analyses It is not clear why both RCTs that evaluated the same
MOST and training intervention observed different results Both
studies enrolled large samples and observed a similar uptake of
licensure following the intervention It is possible that differencesmay have arisen due to different study samples (California ver-
sus New York) or that the intervention delivery differed between
studies
In both trials of the Motorcycle Skills Operating Test including
rider training only a third of motorcycle licence applicants ob-
tained their motorcycle licence It is possible that a licensing pro-
cess including a skills test may reduce crashes through decreased
exposure
Non-mandatory pre-licence training
20Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2350
Three studies assessed the effect of non-mandatory pre-licence
training The Cooper 1988 cohort study found no difference incrash or offence rates between the trained and untrained groups
We cannot however be confident about this result as the study
sufferedfrom selectionbiaswhichcouldhave biasedthe measure of
effect toward the null The RCT conducted by Swaddiwudhipong
1998 examined injury and death rates and found a statistically
significant lower injury rate in the intervention villages after one
year of follow up but no longer at two year follow up There
was no statistically significant difference in death rates between
intervention and control villages Our confidence in the results of
this study are hindered by the inadequacy of the randomisation
process that led to significant differences between the intervention
and control districts on two important variables The intervention
districts were more likely than the control districts to have a validlicence andto wear a helmetSincethese differences were nottaken
into account in later analyses they could explain the differences
observed in injury rates between districts Furthermore with only
six districts randomised the sample size was probably too small
to detect a difference between intervention and control groups
had there been one The McDavid 1989 cohort study found a
lower percentage of injuries in the intervention group although
a statistical test was not conducted McDavid 1989 drew their
intervention and control groups from different populations but
did adjust for confounders although they only analysed 40 of
the initial sample (139 trained and 139 untrained) Because of
these substantial weaknesses we are unable to draw conclusions
from this studyAn overall effect of pre-licence training on the reductionof crashes
injuries deaths and offences was not found although one should
be aware that the research of this issue to date has had significant
flaws However a mandatory skill test (such as the MOST) in
combination with and even without pre-licence training as part
of a licensing process seems to reduce the number of individuals
completing the licensing procedure and therefore indirectly re-
duces crash rates through a reduction in exposure to motorcycle
riding Making licensure a more difficult process may therefore be
seen as advantageous if it delays licensure and therefore reduces
exposure It is possible that the group who chose not to continue
riding a motorcycle would not have been the group to experience a
crash or commit an offence if they had continued to ride anyhowfor example because they are more safety conscious and this was
the reason for their giving up riding in the first place If this is the
case deterring individuals from completing their licence and rid-
ing may not necessarily decrease overall crashes deaths or offence
rates In addition other potentially adverse effects of complicat-
ing a licensing procedure need to be considered such as increases
in unlicensed riding In Thailand non-mandatory rider training
offered as a community intervention may have decreased injury
rates but this needs to be confirmed through further research
Non-mandatory training
The majority of the studies assessed non-mandatory training
(training independent of any licensing procedure) and the results
were varied The two non-randomised trials assessing crash rates found no statistically significant difference in crash rates between
the intervention and control groups (Adams 1985 Leung 1987)
Of the nine cohort studies Satten 1980 found a lower percentage
of crashesin the intervention group although it is unclear whether
this was a statistically significant result Davis 1997 also found a
lower percentage of crashesin theinterventiongroup although it is
unclear with which control group the intervention group wascom-
pared Raymond 1979 Osga 1980 Waller 1992 and Savolainen
2007 found a statistically significant higher crash rate in the in-
tervention group and the remaining cohort studies (Jonah 1982
Mortimer 1984 Mortimer 1988 Billheimer 1998) found no sta-
tistically significant difference in crash rates between intervention
and control groups No conclusions can be drawn from these stud-
ies with confidence The Satten 1980 and Davis 1997 studies did
not adjust for any confounding factors even though important
differences existed between the intervention and control groups
Although Raymond 1979 and Waller 1992 adjusted for a num-
ber of confounders at baseline (Waller 1992) and in the analy-
sis (Raymond 1979 Waller 1992) the intervention and control
groups were drawn from such different populations that it is pos-
sible that unmeasured confounders were responsible for the sta-
tistically significant increased risk of crashes observed in the inter-
vention group The observed difference could also be attributed totheobserved differential loss to followup (Raymond 1979) which
could bias the measure of effect towards or away from the null
The Osga 1980 study found a protective effect of rider training
in the univariate analysis but statistically significant higher crash
rates in the intervention group when accounting for riding expo-
sure Sex (a significantly higher proportion of females in the in-
tervention group) and other potentially confounding factors that
could explain thedifferencebetween the groups were howeverun-
accounted for Finally Savolainen 2007 did control for potential
confounding variables yet it is not clear whether variables were
controlled for simultaneously or one at a time One case-control
study (MAIDS 2004) found no statistically significant reduction
in the odds of experiencing a crash between trained and untrainedriders but the quality of this study was poor mainly because the
analysis did not account for any confounding factors
One cohort study (Satten 1980) assessed the effect of non-manda-
tory rider training on injurious crashes and found a lower percent-
age of injuries in the intervention group although a statistical test
was not conducted As mentioned above the Satten 1980 studyrsquos
validity is reduced due to its lack of adjustment for confounding
factors
The only non-randomised trial that assessed offence rates (Leung
1987) found a statistically significant lower offence rate in the in-
tervention group and of the five cohort studies that assessed of-
fence rates (Satten 1980 Jonah 1982 Mortimer 1984 Mortimer
21Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2450
1988) one study (Jonah 1982) also found a statistically significant
lower offence rate in the intervention group However we have lit-tle confidence in the results from Leung 1987 since potential con-
founding factors were not appropriately addressed Jonah 1982
did conduct a multivariate analysis but was hindered by an initial
low response rate The other four cohort studies did control for
several potential confounders and found no difference in offence
rates following non-mandatory post-licence rider training
The non-randomised trials and cohort studies evaluating non-
mandatory rider training were at high risk of substantial selection
bias due to the non-random allocation of participants as well
as low participation rates and attrition bias and could therefore
not provide conclusive evidence with regard to the effect of the
training It should be noted that despite the biases and other issues
with this group of studies most didnot detectmajor effects relatedto rider training in either direction This may suggest that any
potential impact of rider training will be modest
Studies examining unspecified modes of ridertraining
Three case-control studies assessed whether injured or deceased
motorcyclists were more or less likely to have had rider training
compared to control motorcyclists All had conflicting results
Hurt1981 foundthe odds of havingan injuriouscrash waslower in
those who had received rider training whereas Kraus 1975 found
that the odds of having an injurious crash were higher in youngerand older trained riders Haworth 1997 who compared advanced
versus beginner training found no effect on motorcycle injuries
or deaths although they did find that those who had completed
advanced training had reduced chances of crashing compared to
those who had done a beginnersrsquo training course Similar issues
with uncontrolled confounding and biases of selection as described
above arose with the three case-control studies Hurt 1981 suffers
from inadequate control of confounding via poor matching and
both the Kraus 1975 and the Haworth 1997 findings were only
controlled for limited factors by means of stratified analysis It is
likely that these limitations are the reason for the contradictory
directions of effect observed for the three case-control studies
Methodological weaknesses of the studies
In summary most studies investigating the effectiveness of mo-
torcycle rider training on reducing crashes injuries and death suf-
fered from serious methodological weaknesses
The majority of studies were non-randomised and were there-
fore limited mainly by selection bias Although the approaches of
matching stratification or regression analyses can minimise bias
(Fleiss 2004) they were poorly employed in the reviewed studies
The inadequate adjustment for confounders in particular adds
to our lack of confidence in the studiesrsquo results since it has been
demonstrated that individuals who choose to take motorcycle rider
training differ from those not taking rider training For instance
trained riders often differ from controls in age sex and experi-ence usually ride less and may have a more positive attitude to
safety as indicated by the reported more frequent wearing of seat
belts (Mortimer 1982)
The two large randomised trials reported that a large proportion
of their study group were never exposed to the risk of offences or
crashes because they never attained licensure This made it impos-
sible to assess whether a reduction in crashes in the intervention
group was a result of training or due to delayed or never attained
licensure
In some studies the sample sizes were too small to detect any sta-
tistically significant differences Follow-up time after training dif-
fered between studies and was an impediment to observing any
real effects of training if they existed As riding experience ac-cumulates over time any differences in outcomes between those
trained and untrained would diminish in studies with lengthy fol-
low-up times This dilution of any possible effect of rider training
may have also occurred in studies that mixed both novice and ex-
perienced riders in the study groups since it is hypothesised that
novice riders serve to benefit most from rider training This would
increase the difficulty in finding any effect of rider training par-
ticularly when sample sizes are small
Detection bias may also account for some of the differences ob-
served between trained and untrained riders Detection bias would
have occurred if the self-report hospital driver or police records
used to record the outcomes of crash injury death or offences
differed between the trained and untrained riders Research hasdemonstrated that not all crashes are reported to police for ex-
ample small crashes or off-road crashes (Kraus 1975) even serious
injury crashes have been found to be underreported (Anderson
1980) However this would only introduce detection bias if the
reporting of outcome differs for trained and untrained riders This
does appear possible in that responsible riders may be more likely
to enroll in a rider training course and also may be more likely
to report crashes to police than irresponsible riders meaning that
the number of crashes from riskier riders could be underrepre-
sented on records An analysis to investigate the effectiveness of
rider training may then show that rider training is not effective in
reducing crashes
Rather than using reported outcomes some studies used self-re-ported survey data although this technique suffers from non-
response biases In fact two studies reported differences in de-
mographic characteristics in response rates to questionnaires
(Anderson 1980 Hall 1988) Self-reported data also introduce
the risk of recall bias survival bias and social desirability response
bias Recall bias may have been equally distributed among trained
and untrained riders howeverit is hypothesised that trained riders
may have felt it was lsquoexpectedrsquo to report good driving histories
hence leading the rider training to appear effective
22Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2550
Limitations of this review
Limitations of this review were that we found very few recentstudies (the two most recent ones published in 2002 and 2007)
and that we found very few studies conducted in low-and middle
income countries even though our search strategy was set up to
identify such studies Anotherlimitation wasthat forthe reviewed
studies we could not assess all outcomes of rider training that are
potentially useful Some of the included studies assessed outcomes
such as use of safety clothing including helmets Such outcomes
are valuable since the use of motorcycle helmets has been demon-
strated to be effective in preventing head injuries (Liu 2004) De-
spite this the use of safety clothing was not included as an out-
come in this review because the studies measuring this outcome
were limited by poor data quality due to very low response rates
Another outcome of interest in assessing the effectiveness of rider
training is lsquotype of injuryrsquo As Haworth 2005 state in their review
of motorcycle licensing and training rider training could be con-
sidered effective if injuries become less severe because motorcycle
riders have learned how to avoid more serious injuries through
rider training The studies included in this review did not specify
type of injury and so this outcome could not be assessed Last
changes in knowledge and attitudes following rider training are
other potentially useful outcomes We have excluded such studies
(Osga 1979 Clayton 1990 Goldenbeld 2004) as these outcomes
are proxies rather than direct measures of the effectiveness of mo-
torcycle rider training We opted to report on the outcomes of
crashes injury death and offences alone since there is little ev-
idence about how changes in knowledge and attitudes relate to
these outcomes
A U T H O R S rsquo C O N C L U S I O N S
Implications for practice
Due to the poor quality of studies included in this review we were
unable to draw any conclusions about the effectiveness of rider
training on crash rates injurious or fatal or offence rates The
findings do suggest that mandatory pre-licence training may be
an impediment to completing a motorcycle licensing process thuspossibly indirectly reducing crash injury death and offence rates
through a reduction in exposure to riding a motorcycle On the
basis of the existing evidence it is not clear if and what type of
training reduces the risk of crashes injuries deaths or offences in
motorcyclists and the selection of the best rider training practice
can therefore not be recommended The suggestion that manda-
tory pre-licence training indirectly reduces exposure to riding the
motorcycle because trainees are impeded from completing the li-
censing process does not help those motorcyclists who persist in
obtaining their licence and riding the motorcycle Some type of
rider training is likely necessary to teach motorcyclists to ride a
motorcycle safely Good policy is based on sound evidence and
evaluation but such evidence for motorcycle rider training is seri-
ously lacking Policy makers are therefore encouraged to include arigorous evaluation component to any new or existing rider train-
ing programme Such an evaluation should ideally be randomized
and assess whether the training programme is effective and if so
which component(s) of the programme work best
Implications for research
Given theresults of this review which found littlereliable evidence
about the effectiveness of motorcycle rider training it is necessary
to build up an evidence base of the effectiveness of motorcycle
rider training in the reduction of crash injury death and offence
rates To this end well designed randomised controlled trials need
to be conducted that examine the impact of rider training oncrashes and offences A randomised controlled trial is the preferred
method as it would significantly reduce most biases described in
this review To minimise the impact of participants delaying or
never attaining their licence and therefore exposure as found in
the randomised trials in this review such a trial should include
riders who are already licensed and reliable measures of riding
exposure in a reasonably lengthy follow up period It is expected
that a randomised controlled trial of sufficient sample size would
observe equal balance on possible confounders by study arm how-
ever an assessment of potentially important confounders (such as
helmet wearing) by study group would enable control for these
variables in the analyses if necessary It is imperative that the effec-
tiveness of each component of the rider training (such as theorypractice knowledge skills and modification of behaviour) and the
synergy of such components are evaluated Outcomes should be
objective ideally those that are routinely collected rather than self-
reported outcomes Even though riding the motorcycle poses dif-
ferent challenges and incurs different risks lessons may be learned
from research into the effects of car driver training
Motorcycles are a common means of transport in low- and middle
income countries and usage is increasing As the burden of mo-
torcycle crash victims is mostly borne in such countries research
should be focused on the effectiveness of types of rider training on
the reduction of crashes injuries and fatalities in such countries
and be supported by higher income countries This review found
only one study conducted in a middle income country that sug-gested decreasing injury rates It would be beneficial to conduct
additional research to confirm this suggestion and to enhance our
knowledge of the role of rider training in low- and middle-income
settings
March 2006 saw the launch of the University of North Carolina
Highway Safety Research Centre partnership with the Motorcycle
Safety Foundation (MSF) and plans were revealed to conduct a
study of the potential benefits of motorcycle rider training The
study is funded by the MSF and the National Highway Traffic Sa-
fety Administration (NHTSA) Called rsquoThe Discovery Projectrsquo it
will compare data on crash rates from riders participating in a basic
23Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2650
motorcycle safety class with riders participating in a basic motor-
cycle safety class as well as a variety of continuing education classesThe collection of pilot data was due to start in March 2007 and
the data collection portion of the study is expected to take three
years (Highway Safety Research Center) It is vital that researchers
and policy makers collaborate to develop furtherhigh quality eval-
uations of motorcycle rider training programmes to help inform
development of new and effective training programmes
A C K N O W L E D G E M E N T S
Staff from The University of Sydney library for helping us locate
and retrieve full reports of many papers
R E F E R E N C E S
References to studies included in this review
Adams 1985 published data only
Adams A Collingwood V Job RFS Evaluation of the 1982 Australia
Post motorcycle rider training scheme NSW Australia Traffic
Accident Research Unit Traffic Authority NSW 1985 [ RN 85]
Anderson 1980 published data only
Anderson J Ford JL Peck RC Improved motorcyclist licensing and testing project Washington DC National Highway Traffic Safety
Administration (NHTSA) USA 19801ndash217 [
DOTndashHSndash805ndash598]
Billheimer 1998 published data only
Billheimer JW Evaluation of the California Motorcyclist Safety
Program Transportation Research Record 19981640100ndash9 [
TRB Report Number 980652]
Cooper 1988 published data only
Cooper PJ Rothe JP Motorcyclists who they are and why they do
what they do Transportation Research Record 1988116878ndash85
Davis 1997 published data only
David CF Evaluation of community traffic safety programs and
motorcycle operator training programs Connecticut Department of Transportation 199763
Hall 1988 published data only
Hall CJ Buchanan L Engle R Motorcycle rider education evaluationproject Washington DC USA National Highway Traffic Safety
Administration (NHTSA) US Department of Transportation
1988 [ DTNH 22ndash80ndashCndash05123]
Haworth 1997 published data only
Haworth N Smith R Brumen I Pronk N In Monash University
Accident Research Centre editor(s) Case-control study of motorcycle
crashes Victoria Australia Department of Transport and Regional
Development The Federal Office of Road Safety 19971ndash149 [
ISBN 0 642 25501 6 ]
Hurt 1981 published data only
Hurt H Ouellet J Thom D Motorcycle accident cause factors and
identification of countermeasures USA Department of
Transportation 1981 [ Technical Report DOT HS 806ndash862]
Jonah 1982 published data onlylowast Jonah BA Dawson N Bragg B Are formally trained motorcyclists
safer Accident Analysis amp Prevention 198214(4)247ndash55
Jonah BA Dawson NE Bragg BW Evaluation of the Canada
Safety Councilrsquos Motorcycle Training Program Proceedings of theInternational Motorcycle Safety Conference Washington DC
USA 198059ndash93
Kloeden 1994 published data only
Kloeden CN Moore VM McLean AJ Evaluation of the pre-license
training program for motorcyclists in South Australia Walkerville
South Australia Australia NHMRC Road Accident Research Unit
The University of Adelaide 19941ndash37 [ ISBN 0 7308 0607 3]
Kraus 1975 published data only
Kraus JF Riggins RS Franti CE Some epidemiological features of
motorcycle collision injuries American Journal of Epidemiology 1975102(1)74ndash98
Leung 1987 published data only
Leung KS Reding VA Evaluation of the Wisconsin motorcycle rider course Wisconsin Bureau of Policy Planning and Analysis
Wisconsin Department of Transportation 1987
MAIDS 2004 published and unpublished data
MAIDS In-depth investigations of accidents involving powered
two wheelers ACEM (Acemacemeu) 2004 Vol Final report
20179
McDavid 1989 published data only
McDavid JC Lohrmann BA Lohrmann G Does motorcycle
training reduce accidents Evidence from a longitudinal quasi-
experimental study Journal of Safety Research 198920(2)61ndash72
Mortimer 1984 published data only
Mortimer R Evaluation of the motorcycle rider course Accident
Analysis amp Prevention 198416(1)63ndash71
24Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2750
Mortimer 1988 published data only
Mortimer RG A further evaluation of the motorcycle rider courseJournal of Safety Research 198819(4)187ndash96
Osga 1980 published data only
Osga GA An investigation of the riding experiences of MSF rider
course participants Dissertation South Dakota USA The
University of South Dakota 19801ndash224
Perrino 2002 published data only
Perrino CS Ahmed A Callender A Rozier E Cantwell A Stewart
O The role of Marylandrsquos motorcycle rider course in promoting saver
behaviors and attitudes Baltimore USA National Transportation
Center Morgan State University 20021ndash52
Raymond 1979 published data only
Raymond S Tatum S An evaluation of the effectiveness of the RAC
ACU motorcycle training scheme - final report Road Safety ResearchUnit Dept of Civil Engineering University of Salford London
19791ndash131
Satten 1980 published data only
Satten RS Analysis and evaluation of the motorcycle rider courses
in thirteen Illinois counties International Motorcycle Safety
Conference Washington DC 1980145ndash93
Savolainen 2007 published and unpublished data
Savolainen Peter Mannering Fred Additional evidence on the
effectiveness of motorcycle training and motorcyclistsrsquo risk-taking
behaviour Transportation Research Board 2007 Vol March 20
19
Swaddiwudhipong 1998 published data only
Swaddiwudhipong W Boonmak C Nguntra P Mahasakpan P
Effect of motorcycle rider education on changes in risk behavioursand motorcycle-related injuries in rural Thailand Tropical Medicine
and International Health 19983(10)767ndash70
Waller 1992 published data only
Waller FL An analysis of the impact of the Texas Department of
Public Safety Motorcycle Operator Training Program upon
motocycle related traffic accidents Texas State University Public
Administration Program Applied Research Projects 199277
References to studies excluded from this review
Anderson 1980a published data only
Anderson LC Better biking a motorcycle rider improvement
program International Motorcycle Safety Conference Washington
DC USA Motorcycle Safety Foundation 1980231ndash53Aupetit 2008 published and unpublished data
Aupetit S Riff J Buttelli O Espie S Ergonomics research program
on motorcycle learning Contributions to the development of a
motorcycle simulator for training Transport Research Arena
Ljubljana 20089
Awane 1999 published data only
Awane T Integrating simulators in motorcycle safety education
Journal of International Association of Traffic and Safety Sciences 199923(1)26ndash35
Baldi 2005 published data only
Baldi S Baer JD Cook AL Identifying best practices states in
motorcycle rider education and licensing Journal of Safety Research200536(1)19ndash32
Batchler 1988 published data only
Batchler MW Evaluation of the pre-license motorcycle training course at the New South Wales traffic education centre Federal Office of
Road Safety (FORS) 19881ndash122 [ Report No CR 77 ]
Blanchard 2006 published data only
Blanchard HT Tabloski PA Motorcycle safety educating riders at
the teachable moment Mosby (CV) Company 200632330ndash2
Bonnett 2005 published data only
Bonnett J Mature age skills training for experienced motorcycle
riders Motorcycle ridersrsquo association of the Australian Capital
Territory 2005
Braver 2007 published data only
Braver ER et alPersuasion and licensure a randomized controlled
intervention trial to increase licensure rates among Maryland
motorcycle owners Traffic Injury Prevention 20078(1)39ndash46Clayton 1980 published data only
Clayton AB Sudlow DE An evaluation of the effectiveness of a one day training programme for learner motorcyclists Berkshire UK
Transport and Road Research Laboratory 19801ndash23 [ ISSN
0266ndash7045]lowast Clayton AB Sudlow DE An evaluation of the effectiveness of a one-day training programme for learner motorcyclists Berkshire UK
Road Safety Division Berkshire 19801ndash23 [ISSN 0266ndash7045]
Council 1978 published data only
Council FM Desper LP Dutt AK Driver education for motorcycle operation University of North Carolina Highway Safety Research
Center 19781ndash40
Crowther 2005 published data onlylowast Crowther G Engaging with motorcyclists UK police and theBikeSafe Road Safety Programme Published proceedings of the
Australasian Road Safety Research Policing Education Conference
Wellington New Zealand 2005 issue ISBN 0ndash473ndash10636ndash11ndash11
Daniello 2009 published data only
Daniello A Gabler HC Mehta YA Effectiveness of motorcycle
training and licensing Transportation Research Board 88th Annual
Meeting 200913
Dewanti 2007 published data only
Dewanti M Challenging efforts in promoting young motorcyclist
safety in Indonesia Road Safety on Four Continents 14th
International Conference Swedish National Road and Transport
Research Institute 2007
Du Plooy 1990 published data onlyDu Plooy DW Motorcycle training in South Africa The Human
Element International Motorcycle Safety Conference Orlando
Florida USA Motorcycle Safety Foundation 19904-1 to 4-9
Engel 1989 published data only
Engel U Krogsgaringrd Thomsen L [Faeligrdselsundervisning og
propaganda ndash en faeligrdselssikkerhedsfremmende foranstaltning]
Dansk Vejtidsskrift 198912295ndash7
Goldenbeld 2002 published data only
Goldenbeld C Houwing S de Craen S The development of
driving skills of young moped riders [De ontwikkeling van de
rijvaardigheid van jonge bromfietsers] SWOV Institute for Road
Safety Research (Stichting Wetenschappelijk Onderzoek
Verkeersveiligheid) Leidschendam the Netherlands 200263
25Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2850
Goldenbeld 2004 published data only
Goldenbeld C Twisk D de Craen S Short and long term effects of moped training a field experiment Transportation Research Part F 20047(1)1ndash16
Haworth 2000 published data only
Haworth N Smith R Kowadlo N Motorcycle training in Victoria
evaluation of rider training curriculum in Victoria Monash
University Accident Research Centre Reports 2000 Vol 165 issue
3131
Haworth 2005 published data only
Haworth N Mulvihill C Review of motorcycle licensing and
training Monash University Accident Research Centre 2005 [
Report no240]
Haworth 2006 published data only
Haworth N Mulvihill C Rowdon P Teaching old dogs new tricksTraining and older motorcyclists Journal of the Australasian College of Road Safety 200718(4)20ndash5lowast Haworth NL Mulvihill C Rowden PJ Teaching old dogs new
tricks Training and older motorcyclists Australasian Road Safety
Research Policing and Education Conference Gold Coast
Queensland 2006
Hill 1983 published data only
Hill D Motorcycle rider training testing and licensing Motor
Cycle Council of New South Wales Sydney 1983
Imatake 1980 published data only
Imatake Y On the activities in Japan of motorcycle safe riding
promotion committee (MSRPC) - especially the education of
unskilled riders of class 1 motorbikes International Motorcycle
Safety Conference Washington DC USA Motorcycle Safety
Foundation 1980255ndash68
Ingebritsen 1990 published and unpublished data
Ingebrigtsen S [Risikofaktorer ved ferdsel med moped og
motorsykkel] TOslashI-rapport 66 Oslo Transportoslashkonomisk
institutt 1990
Jain 2009 published and unpublished data
Jain A et alTwo wheeler accidents on Indian roads a study from
Mangalore India Elsevier 200916130ndash3
Jonah 1979 published data only
Jonah BA Dawson NE Validation of the motorcycle operator skill
test Accident Analysis amp Prevention 197911(3)163ndash71
Jonah 1981 published data onlyJonah BA Dawson NE Bragg WE Predicting accident
involvement with the Motorcycle Operator Skill Test Accident Analysis amp Prevention 198113(4)307ndash418
Kadar 1990 published data only
Kadar KJ Countermeasures motorcycle rider education delivery
system The Human Element International Motorcycle Safety
Conference Motorcycle Safety Foundation 19906-27 to 6-41
Kelsey 1986 published data only
Kelsey SL Liddicoat C Ratz M Licensing novice motorcyclists a
comparison of Californiarsquos standard test and the MOST II (Motorcycle
Operator Skill Test) administered at centralized testing offices Sacramento California USA California Department of Motor
Vehicles Research and Development Office 19861ndash74
Koch 1980 published data only
Koch H Advanced course for motorcycle riders InternationalMotorcycle Safety Conference Washington DC USA Motorcycle
Safety Foundation 1980269ndash81
Koch 1990 published data only
Koch H Brendicke R How to influence risk taking behaviour
within motorcycle rider training programs The Human Element
International Motorcycle Safety Conference Orlando Florida
USA Motorcycle Safety Foundation 1990Motorcycle Safety
Foundation 1990 p 9-1 to 9-25A
Labbett 2005 published data only
Labbett S Langham M Training can make the problem worse
Road Safety Congress 20051ndash12
Law 2005 published and unpublished data
Law TH Radin Umar RS Zulkaurnain S Kulanthayan S Impactof the effect of economic crisis and the targeted motorcycle safety
programme on motorcycle-related accidents injuries and fatalities
in Malaysia International Journal of Injury Control and Safety
Promotion 200512(1)9ndash21
Lowes 1990 published data only
Lowes BF In-traffic rider training The Human Element
International Motorcycle Safety Conference Orlando Florida
USA Motorcycle Safety Foundation 19907-46 to 7-57
Mahon 1976 published data only
Mahon J Teaching accident avoidance on the motorcycle Journal of Traffic Safety Education 1976231ndash29
Maxwell 1985 published data only
Maxwell GS Evaluation of a Program for Training MotorcycleRiders Evaluation and Program Planning 19858(4)339ndash47
Mayhew 1996 published data only
Mayhew DR Simpson SH Effectiveness and role of driver educationand training in a graduated licensing system Ottowa Canada The
Traffic Injury Research Foundation 19961ndash89 [ ISBN
0ndash920071ndash12ndash0]
McKnight 1987 published data only
McKnight AJ Milburn BG An evaluation of the modified
motorcycle operator skill test exercises NHTSA Report 1987 [
Report No DOT HS 807 212 ]
McPherson 1976 published data only
McPherson K McKnight AJ The development and evaluation of a
motorcycle skill test manual and knowledge test Washington DCUSA National Highway Safety Administration 1976
Mohan 2004 published data only
Mohan D Evidence-based interventions for road traffic injuries in
South Asia Journal of the College of Physicians and SurgeonsPakistan 200414(12)746ndash7
Morgan 1980 published data only
Morgan JK Advanced motorcycle training course content and
study evaluation International Motorcycle Safety Conference
Washington DC USA Motorcycle Safety Foundation 1980
283ndash312
Motoki 1990 published data only
Motoki M Yamazaki S A study on effective motorcycle rider
education The Human Element International Motorcycle Safety
26Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2950
Conference Orlando Florida USA Motorcycle Safety
Foundation 19909-26 to 9-57A
Nairn 1992 published data onlylowast Nairn RJ Motorcycle safety research literature review 1987 to
1991 Federal Office of Road Safety Australia 1993 Vol Report
CR 117
Newland 1980 published data only
Newland R Darby J Automotive practicespre motorcycle rider
program International Motorcycle Safety Conference Washington
DC USA Motorcycle Safety Foundation 1980127ndash9
Newland 1982 published data only
Newland R Motorcycle rider training and licensing requirements
Driver training steering a course for the future Melbourne
Victoria Transport Regulation Board 1982
Newland 1990 published data only
Newland R Motorcycle training and licensing in Victoria 1980-
1990 The Human Element International Motorcycle Safety
Conference Orlando Florida USA Motorcycle Safety
Foundation 1990
Ormston 2003 published data only
Ormston R Dudleston A Pearson S Stradling Evaluation of
Bikesafe Scotland Scottish Executive Social Research Edinburgh
2003
Osga 1979 published data only
Osga GS Ellingstad VS An evaluation of the effectiveness of the 1978
South Dakota motorcycle rider course program South Dakota USA
University of South Dakota 19791ndash89
Panou 2007 published data only
Panou M Bekiaris E A new concept on the integration of driving
simulators in driving training the train-all approach Road Safety
on Four Continents 14th International Conference Swedish
National Road and Transport Research Institute 2007
Peterson 1990 published data only
Peterson A Kadar K Status report 1980-1989 Minnesota rider
education statistics The Human Element International
Motorcycle Safety Conference Orlando Florida USA 199013-
113 to 13-129
Prem 1984 published data only
Prem H Good MC Motorcycle rider skills assessment ACT
Australia Office of Road Safety Department of Transport 1984
1ndash377 [ Report No CR 34 ]
Radin Umar 2006 published data only
Radin Umar RS Motorcycle safety programmes in Malaysia how
effective are they International Journal of Injury Control and Safety Promotion 200613(2)71ndash9
Rockwell 1990 published data only
Rockwell TH Kiger S Carnot M An evaluation of the Ohio
motorcyclist enrichment program Ohio Department of Highway
Safety 1990
Rothe 1987 published data only
Rothe JP A motorcycle riding course Motorcyclists Image and reality Canada Insurance Corporation of British Columbia 1987
59ndash83
Rowden 2007 published data only
Rowden P Watson B Haworth N What can riders tell us aboutmotorcycle training A view from the other side of the fence
Australasian Road Safety Research Policing and Education
Conference Melbourne Australia Melbourne 2007
RSC Victoria 1993 published data only
Road Safety Committee Inquiry into Motorcycle Safety in Victoria
Parliament of Victoria 1993
Russam 1979 published data only
Russam K Motorcycle training and licensing in Japan UK Transport
and Road Research Laboratory (TRRL) 1979 [ Report 916 ]
Saffron 1981 published data only
Saffron DG Driver and rider education training and licensing a
brief review Traffic Accident Research Unit Traffic Authority of
New South Wales NSW Australia Traffic Accident Research UnitTraffic Authority of New South Wales 1981
Schulz 1990 published data only
Schulz U Motorcycle accidents among learner riders Motorcycle
licensing Evaluation of the effectiveness of a Graduated Licensing Scheme Forshungshefte des Institut fur Zweiradsicherheit 1990
Sexton 2004 published data only (unpublished sought but not used)
Sexton B Baughan C Elliott M Maycock G The accident risk of
motorcyclists 2004TRL Report TRL 607(ISSN 09684107)58
Shephard 1986 published data only
Shepherd R Do Motorcycle Safety Foundation programs work
Driver 198619(9)16ndash8
Simpson 1990 published data onlySimpson HM Mayhew DR The promotion of motorcycle safety
training education and awareness Health Education Research
Theory and Practice 19905(2)257ndash64
Sokolov 1990 published data only
Sokolov O Safe motorcycle riding some aspects of motorcycle
riding in the USSR The Human Element International
Motorcycle Safety Conference Motorcycle Safety Foundation
19904ndash37 - 4-48
Steffens 1988 published data only
Steffens U Gawatz R Willmes G Effectiveness of mofa courses
[Wirksamkeit von Mofakursen] Unfall- und sicherheitsforschung
strasenverkehr 198867168
Sudlow 2003 published data only
Sudlow D Scoping study on motorcycle training Road Safety
Research Report London UK The British Institute of Traffic
Education Research Department for Transport 2003 Vol June
[ Report No 36]
Tasmania 1995 published data only
Mannering FL Grodsky LL Motorcycle rider training scheme
Traffic Education Research Centre Armidale Australia Driver
Training Australia Limited 1995 Vol April1ndash56
Tetard 1985 published data only
Tetard C Motorcycle ridersrsquo training to detect indices in an
intersection situation [Lrsquoentrainement a la prise drsquoinformation dans
lrsquoapprentissage de la conduite motocycliste] Le Travail Humain198548(2)147ndash59
27Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3050
Thackray 1980 published data only
Thackray RM Prescott JC Field test of the motorcycle ridercourse International Motorcycle Safety Conference Washington
DC USA Motorcycle Safety Foundation 1980195ndash7
Thompson 1994 published data only
Thompson M Evaluation of compulsory basic training for
motorcyclists Transport Research Laboratory Crowthorne
Berkshire UK 1994 [ Report 63]
Tobler 1990 published data only
Tobler U Individual driving instruction for motorcycle driving
learners the Swiss experiment of motorcycle driving school The
Human Element International Motorcycle Safety Conference
Orlando Florida USA Motorcycle Safety Foundation 1990
Transport 1989 published data only
Transport Communications Road Crash Statistics Australia Bureauof Transport and Communications Economics 198914ndash20
Tuan 2007 published data only
Tuan VA Shimizu T Towards development and evaluation of the
motorcycle drivers re-education program in Vietnam modeling of
motorcycle driverrsquos undesired behaviors 11th World Conference
on Transport Research 200725
Victor 1980 published data only
Victor DJ Traffic safety education for motorcyclists International
Motorcycle Safety Conference Washington DC USA Motorcycle
Safety Foundation 1980221ndash8
Waterer 1990 published data only
Waterer TH Compulsory training for motorcyclists from liberty
to compulsion The Human Element International Motorcycle
Safety Conference Orlando Florida USA Motorcycle Safety
Foundation 19904-49 to 4-57
Weller 1989 published data only
Weller RB Chandler EW Motorcycle safety and motorcycle
education past research and survey results Journal of Public Policy
amp Marketing 1989893ndash108
Wells 1981 published data only
Wells P Observations of some RACACU motorcycle training
schemes Transport and Road Research Laboratory Crowthorne
Berkshire UK 1981
Winn 1986 published data only
Winn G Two important trends in motorcycle safety regulations
rider education and conspicuity improvement American
Motorcyclist Association 1986147ndash52
Wisher 1988 published data only
Wisher G Reid J Motorcycle accidents in Norfolk a three-year study of motorcycle training and the Part 1 test Norfolk UK Road Safety
Division Highways Department Norfolk County Council 1988
Wood 1987 published data only
Wood HT Bowen R Evaluation of the revised learner permit scheme
July 1983 to December 1985 Victoria Australia Road Traffic
Authority 19871ndash27 [ ISBN 0 7306 0247 8]
Yamazaki 1990 published data only
Yamazaki S Nakanishi M Survey on effect of Honda rider
education The Human Element International Motorcycle Safety
Conference Orlando Florida USA 199012-167 to 12-187
References to ongoing studies
Burgess 2010 unpublished data only
Burgess C Broughton P Fylan F Stradling Interim evaluation of
the UKs National Ride program In L Dorn editor(s) Driver
behaviour and Training Vol IV Aldershot Ashgate Publishing
Limited 2010
Additional references
Anderson 1980
Anderson J Ford JL Peck RC Improved motorcyclist licensing and
testing project Volume I and II National Highway Traffic Safety
Administration US department of Transportation Washington
DC 1980
Bjornstig 1985
Bjornstig U Bylund P Lekander T Brorsson B Motorcycle fatalitiesin Sweden Acta Chirurgica Scandinavica 1985151(7)577ndash81
Clayton 1990
Clayton A Sudlow DE An evaluation of the effectiveness of a one-
day training program for learner motorcyclists The Human
Element International Motorcycle Safety Conference Orlando
Florida USA 199013-105 to 13-111
Cochrane Collaboration 2008
Higgins JP Green S Cochrane Handbook for Systematic Reviews of
Interventions 500 The Cochrane Collaboration 2008
Elliot 2003
Elliott MA Baughan CJT Broughton J Grayson G Knowles J et
alMotorcycle Safety a Scoping Study TRL [ Report No
0968ndash4107]Fleiss 2004
Fleiss JL Levin B Paik MC Statistical methods for rates and proportions 3rd Edition John Wiley amp Sons Inc 2004 [ ISBN
978ndash0ndash471ndash52629ndash2 ]
FORS 1999
FORS Federal Office of Road Safety Road risk for sober licensed
motorcyclists Vol Monograph 27 Federal Office of Road Safety
19991ndash2
Haworth 2005
Haworth N Mulvihill C Review of motorcycle licensing and training
Victoria Australia Monash University Accident Research Centre
20051ndash83 [ Report No 240]
Huwiler-Muumlntener 2002Huwiler-Muumlntener K Juumlni P Junker C Egger M Quality of
Reporting of Randomized Trials as a Measure of Methodologic
Quality JAMA 20022872801ndash4
Johnston 2008
Johnston P Brooks C Savage H Fatal and serious road crashes
involving motorcyclists Road Safety Research and Analysis Report
ATSB Monograph 20 Canberra Department of Infrastructure
Transport Regional Development and Local Government 0205
081ndash26
Juumlni 2001
Juumlni P Altman DG Egger M Systematic reviews in health care
Assessing the quality of controlled clinical trials BMJ 2001323
42ndash6
28Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3150
Kopits 2003
Kopits E Cropper M Traffic fatalities and economic growth TheWorld Bank Development Research Group Infrastructure and
Environment Policy research working paper 3035 2003
Leung 1983
Leung P Chau BTW Motorcycle accidents in Hong Kong pattern
of injury treatment causative factors and socio-economic
implications Injury 198315(1)23ndash9
Liu 2004
Liu B Ivers R Norton R Blows S Lo SK Helmets for preventing
injury in motorcycle riders Cochrane Database of Systematic Reviews 2004 Issue 2 [DOI 10100214651858CD004333pub3]
Mohan 2002
Mohan D Road safety in less-motorized environments future
concerns International Journal of Epidemiology 200231(3)527ndash32Mortimer 1982
Mortimer RG An evaluation of the motorcycle rider course
Proceedings of the 20th conference of the American Association for
Automotive Medicine Arlington Heights IL USA American
Association for Automotive Medicine 198277ndash96
Mullin 1997
Mullin BT Risk factors for motorcycle injury the role of age
gender experience training and alcohol PhD thesis Auckland
New Zealand University of Auckland 1997
Noordzij 2001
Noordzij PC Forke E Brendicke R Chinn BP Integration of needs
of moped and motorcycle riders into safety measures review and
statistical analysis in the framework of the European research
project PROMISING SWOV Institute for Road Safety Research
Leidschendam The Netherlands 2001 [ Report No Dndash2001ndash5]
Osga 1979
Osga G Ellingstad VS An evaluation of the effectiveness of the 1978 South Dakota Motorcycle Riders Course Program South Dakota
University of South Dakota Vermillion Human Factors Lab 1979
Oxford English Dictionary Online
httpoxforddictionariescom Oxford English Dictionary Online
Oxford University Press 2010
Peden 2004
Peden M Scurfield R Sleet D World Report on Road TrafficInjury Prevention Geneva World Health Organization Available
at httpwwwwhointviolence˙injury˙preventionpublications
road˙trafficworld˙reporten 2004Chapter 3
Radin Umar 1995
Radin Umar R Macakay G Hills B Preliminary analysis of
motorcycle accidents Short-term impacts of the running
headlights campaign and regulations in Malaysia J Traffic Med 199523(1)17ndash27
Tsai 1995
Tsai YJ Wang JD et alCase-control study of the effectiveness of
different types of helmets for the prevention of head injuries among
motorcycle riders in Taipei Taiwan American Journal of
Epidemiology 1995142(9)974ndash81
Vis 1995
Vis AA In-depth study of the hazards of motorcycling [De
onveiligheid van motorrijden nader bekeken] SWOV Institute for
Road Safety Research Leidschendam the Netherlands 1995 [
Report No Rndash95ndash69]
Weaver 1990
Weaver EA Motorcycle safety education in the United States
Motorcycle Safety Foundation Orlando Florida USA 1990
Winn 1987
Winn G Two important trends in motorcycle safety regulations
rider education and conspicuity improvement No more
information available Detroit MI US Society of Automotive
Engineers 1987147ndash52
Yuan 2000
Yuan W The effectiveness of the rsquoride-brightrsquo legislation for
motorcycles in Singapore Accident Analysis amp Prevention 200032
(4)559ndash63
Zambon 2006
Zambon F Hasselberg M Factors affecting the severity of injuries
among young motorcyclists a Swedish nationwide cohort study
Traffic Injury Prevention 20067(2)143ndash9lowast Indicates the major publication for the study
29Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3250
C H A R A C T E R I S T I C S O F S T U D I E S
Characteristics of included studies [ordered by study ID]
Adams 1985
Methods Non-randomised trial
Participants Initial sample size
Total of 162 Australia Post telegram delivery riders
Not clear how many in intervention and control groups
Analysis sample size
Intervention group 63Control group 35
Total 98 (61)
Interventions No formal training vs voluntary rider training course of 12 hours over 2 consecutive days (theory 4 hours practice
8 hours)
Intervention integrity
No information given
Outcomes Australia Post recorded crashes
Length of follow up 1 year
Notes Study conducted in NSW Australia
Anderson 1980
Methods Randomised trial
Participants Initial sample size
Total 41188 motorcycle licence applicants
Analysis sample size
Intervention group with skills training 13160
Intervention group without skills training 12634
Control group 15080
Total 40874 (99)
Interventions Standard licensing skill test vs the Motorcycle Operators Skill Test (MOST) with or without a training component
MOST is not to be confused with rider training because it is a licensing test The training component was for
participants who failed the skill test on the first attempt was of 3 hours duration and consisted of instructions about
the skills needed to pass the skill test The training consisted of 30 min theory and 2 frac12 hours practice on the
motorcycle
Intervention integrity
33 in MOST with skills training 31 in MOST without skills training and 43 in control group obtained their
motorcycle licence in the end
Outcomes Police recorded crashes and offences
Length of follow up 3 years
30Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3350
Anderson 1980 (Continued)
Notes Study conducted in California USA
Billheimer 1998
Methods Cohort study
Participants Initial sample size
Intervention group Unclear numbers of CMSP graduates and ERC graduates
Control group Over 16000 untrained motorcyclists
Analysis sample size
1139 matched CMSP-pairs and 1182 ERC-pairs
Interventions No formal training vs the mandatory (for those younger than 18 years) pre-licence 16-hours basic California
Motorcyclist Safety Program (CMSP) and the 8-hour Experienced Rider Course (ERC)
Intervention integrity
No information given
Outcomes Police recorded crashes and offences
Length of follow up 2 years
Notes Study conducted in California USA
Cooper 1988
Methods Cohort study
Participants Initial sample size
Intervention group 418 graduates and 57 who failed
Control group 402
Analysis sample size
Intervention group unclear
Control group unclear
Total 863 (984)
Interventions No formaltrainingvs non-mandatory pre-licence rider trainingThe content of thetrainingwas notfurther explainedIntervention integrity
Participants had all completed the rider training
Outcomes Police recorded crashes and offences
Length of follow up 7 years
Notes Study conducted in British Columbia Canada
31Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3450
Davis 1997
Methods Retrospective cohort study
Participants Initial sample size
Intervention group 9320 graduates
Control group unclear
Analysis sample size
Intervention group 9320 graduates
Control group unclear
Interventions No formal training vs the non-mandatory Connecticut Motorcycle Operator Training Program The content of
training was not further explained
Intervention integrity Participants had all completed the rider training
Outcomes Police recorded crashes severity of crashes
Length of follow up 1 year
Notes Study conducted in Connecticut USA
Hall 1988
Methods Randomised trial
Participants Initial sample sizeIntervention group A 6248
Intervention group B 6616
Intervention group C 6996
Control group 6604
Total 26464 motorcycle licence applicants
Analysis sample size
Intervention group A 6012 (96)
Intervention group B 6428 (97)
Intervention group C 6778 (97)
Control group 6502 (98)
Interventions Standard licensing skill test vs the Motorcycle Operator Skill Test (MOST II) without training (A) MOST II with
3 hours training (B) or MOST II with 20 hours training (C) The 3 hours training comprised instructions aboutthe skills needed to pass the skill test (30 min theory and 2 frac12 hours practice on the motorcycle) The 20 hours
training consisted of the Motorcycle Safety Foundationrsquos Motorcycle Rider Course (MRC) comprising of 8 hours
theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-road practice
Intervention integrity
25 and 26 obtained their motorcycle licence in the lsquoMOST 3-hour trainingrsquo and lsquoMOST 20-hour trainingrsquo
intervention arm respectively In the lsquoMOST testrsquo intervention arm 29 obtained their licence and in the control
group 35
Outcomes Police recorded crashes and offences
Length of follow up 3 years
32Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3550
Hall 1988 (Continued)
Notes Study conducted in New York USA
Haworth 1997
Methods Population based case-control study
Participants Cases Motorcycle riders or pillion riders involved in motorcycle crash (on road) Recruited from hospital and police
records Not clear whether all cases (especially fatalities) were included
Community controls Multiple matched sample of motorcycle rider or pillion who passed the crash site at the same
time of day and day of week that the crash occurred
Initial sample size
Cases 336
Controls 1195
Analysis sample size
Cases 234 (70)
Controls 561 (47)
Interventions Basic handling skills course beginner intermediate or advanced course measured through self-report not validated
Outcomes Injuries with an Injury Severity Score of 5 or greater or admission for at least four hours and fatalities from a
motorcycle crash (from hospital and police records)
Length of follow up two years or more prior to the interview
Notes Study conducted in Victoria Australia
Hurt 1981
Methods Population based case-control study
Participants Cases Motorcycle riders who had a motorcycle crash that was reported to police (traffic accident reports) Not all
cases included
Community controls Multiple matched sample of motorcycle riders who passed the crash site at the same time of
day and day of week as the crash occurred but 2 years later
Initial sample size
Cases 4500 notifications of crashes to police (only 50 received in time research team able to respond to 1126 for
unreported reasons data collection completed for 900)
Controls Number approached unreported 2310 consented and interviewed
Analysis sample size
Cases 808 (808900 data on rider training) - 18 of initial sample
Controls 670 (6702310 data on rider training) - lt 29 of initial sample
Interventions Training experience self taught friends-family school-club formal-professional others measured through self-
report validated
Outcomes Injuries and fatalities from a motorcycle crash (from hospital records and coronerrsquos office)
33Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3650
Hurt 1981 (Continued)
Notes Study conducted in Los Angeles USA
Jonah 1982
Methods Cohort study
Participants Initial sample size
Intervention group
2310 licensed MTP graduates who are active riders
Control group
2029 licensed motorcyclists who are active riders
Non-licensed and non-active riders excluded
Analysis sample size
Intervention group
811 interviews (35) 533 driver records (23)
Control group 1080 interviews (71) 905 driver records (45)
Interventions No training vs the non-mandatory Canada Safety Council Motorcycle Training Program (MTP) comprising of 4
hours theory and 16 hours practise (6 hrs on-road) over 2 weekends
Intervention integrity
Participants had all successfully completed the MTP
Outcomes Self-reported and police reported crashes and offencesLength of follow up 4 years
Notes Study conducted in Ontario Canada
Kloeden 1994
Methods Cohort study
Participants Initial sample size
Intervention group 2533 licence applicants
Control group Unclear
Sample of riders obtaining probationary licenceIntervention group 1625 (64)
Control group 5015 (unclear)
Non-active riders excluded from control group (~66)
Analysis sample size
Intervention group 631 (39)
Control group 318 (6)
Interventions No formal training vs the mandatory pre-licence Ridersafe course consisting of two levelsthe first is a training course
in the basics of motorcycle riding where a pass entitles the rider to obtain a learnerrsquos permit The second is a skill
test where a pass entitles the rider to obtain a probationary motorcycle licence The training course consists of two 4
hour training sessions on separate days involving instruction videos and actual riding on motorcycles The skill test
consists of a 4 hour preparation course of 74 minutes of lecturediscussion 38 minutes of video and 75 minutes of
34Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3750
Kloeden 1994 (Continued)
on-course riding which is the testing phase
Intervention integrity
Unclear what percentage in the Ridersafe group and the control group obtained their probationary licence halfway
and their full motorcycle licence in the end
Outcomes Police reported crashes
Length of follow up 1 year
Notes Study conducted in SA Australia
Kraus 1975
Methods Population based case-control study
Participants Cases All motorcycle riders with a confirmed medically treated motorcycle injury (including fatalities) from all
hospitals in study base and peripheral area hospitals and all deaths due to motorcycle crashes in the study base All
serious injuries (leading to hospitalisation) and fatalities likely to be included
Community controls A random sample of registered motorcycle owners in the same temporary and geographical
study base as the cases arose from
Initial sample size Cases 1273
Controls 738
Analysis sample size Cases 741 (58)
Controls 434 (59)
Interventions Motorcycle rider training (not specified) measured through self-report
Outcomes Injuries leading to hospitalisation fatalities
Length of follow up not specified
Notes Study conducted in Sacramento county USA
Leung 1987
Methods Non-randomised trial
Participants Initial sample sizeIntervention group unknown number of MRC graduates
Control group unknown number of licensed motorcyclists
Analysis sample size
Intervention group 2941
Control group 43094
Interventions No formal training vs non-mandatory Wisconsin motorcycle rider course a 22 hours off-street programme that
teaches the basic rules of safe motorcycling
Intervention integrity
50 in the MRC group and 100 in the control group obtained their motorcycle licence in the end
35Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3850
Leung 1987 (Continued)
Outcomes Police recorded crashes and offences
Length of follow up up to 3 years
Notes Study conducted in Wisconsin USA
MAIDS 2004
Methods Population based case-control study
Participants Cases Motorcycle riders who had a motorcycle crash Unclear whether all cases included
Community controls Motorcycle riders identified at petrol stations Unclear whether matched to cases
Initial sample size
Cases not reported
Controls not reported
Analysis sample size
Cases 921 (Italy 200 Netherlands 200 Spain 121 Germany 250 France 150)
Controls 923 (Italy 200 Netherlands 200 Spain123 Germany 250 France 150)
Interventions Pre-license rider training or additional training (non-compulsory)
Outcomes Crashes
Notes Study in five sampling areas located in five European countries (France Germany Netherlands Spain and Italy)
McDavid 1989
Methods Cohort study
Participants Sample size after first match
Intervention group 346 male training graduates
Control group 346 male motorcycle riders licensed in same year
Sample size after second match
Intervention group 139 (40)
Control group 139 (40)
Analysis sample size
Intervention group 139 (40)
Control group 139 (40)
Interventions No formal training vs the non-mandatory pre-licence British Columbia Safety Councilrsquos 37-hour motorcycle safety
programme
Intervention integrity
Participants had all completed and passed the training
Outcomes Police reported crashes injurious crashes
Length of follow up 1-6 years
36Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3950
McDavid 1989 (Continued)
Notes Study conducted in British Columbia Canada
Mortimer 1984
Methods Cohort study
Participants Initial sample size
Intervention group 608 MRC graduates
Control group 360 motorcycle riders
Response rates
Intervention group 360 (59)
Control group 324 (90)
Non-active riders excluded
Analysis sample size
Intervention group 213 (35)
Control group 303 (84)
Interventions No formal training vs the non-mandatory Motorcycle Safety Foundationrsquos Motorcycle Rider Course comprising of
8 hours theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-road practice
Intervention integrity
Participants had all completed and passed the MRC
Outcomes Self-reported crashes and offencesLength of follow up 1 year
Notes Study conducted in Illinois USA
Mortimer 1988
Methods Cohort study
Participants Initial sample size
Intervention group 2307 MRC graduates
Control group 556 active licensed motorcycle riders
Response ratesIntervention group 1298 (56)
Control group 500 (90)
Non-active riders excluded
Analysis sample size
Intervention group 913 (40)
Control group 500 (90)
Interventions No formal training vs the non-mandatory Motorcycle Safety Foundationrsquos Motorcycle Rider Course comprising of
8 hours theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-road practice
Intervention integrity
Participants had all completed and passed the MRC
37Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4050
Mortimer 1988 (Continued)
Outcomes Self reported crashes and offences
Length of follow up 1 year
Notes Study conducted in Illinois USA
Osga 1980
Methods Cohort study
Participants Initial sample sizeIntervention group 1163 MSF MRC participants invited 818 (74) responded
Control group 329 motorcycle riders responded to recruitment strategies
Non-active riders excluded from analysis
Analysis sample size
Intervention group 595 (73)
Control group 329 (100)
Interventions No formal training vs the non-mandatory Motorcycle Safety Foundationrsquos Motorcycle Rider Course (MRC) com-
prising of 8 hours theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-road
practice
Intervention integrity
Participants had all successfully completed the MRC
Outcomes Self-reported crashes police recorded crash reports
Length of follow up 6 months - 3 12 years
Notes Study conducted in South Dakota USA
Perrino 2002
Methods Cohort study
Participants Initial sample size
Intervention group 2144 (unclear) MRC graduates (motorcycle licence applicants)
Control group 2601 (unclear) regular licensing test graduates (motorcycle licence applicants)Analysis sample size
Intervention group 1070 (unclear)
Control group 867 (unclear)
Interventions No formal training vs the mandatory (for those younger than 18 years) pre-licence Maryland Motorcycle Rider
Course (MRC) consisting of 8 hours theory and 12 hours practice
Intervention integrity
No information given
Outcomes Self-reported and police recorded crashes and offences
Length of follow up 1 year
38Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4150
Perrino 2002 (Continued)
Notes Study conducted in Maryland USA
Raymond 1979
Methods Cohort study
Participants Initial sample size
Intervention group
258 volunteers mainly (but not all) novice riders
Control group
1104 provisional licence applicants
Analysis sample size
Intervention group 190 (74)
Control group 627 (57)
Interventions No formal training vs the non-mandatory RACACU (Royal Automobile ClubAuto-Cycle Union) training scheme
The training program consisted of either 24 one-hour sessions or 13 two-hour sessions Twelve hours of the training
focused on theory about basic operational procedures traffic laws and highway code The practical session focused
on basic operational skills riding on a training area and riding in traffic with an instructor A proficiency test was
administered at the end of the course
Intervention integrity
No information given
Outcomes Self-reported crashes (including minor non-injury non-damage and off-road crashes)
Length of follow up 1 year
Notes Study conducted in the UK
Satten 1980
Methods Cohort study
Participants Initial sample size
Intervention group interviewers kept interviewing until they had 100 riders selected from a population of 891
enrollees in the MSF MRC course (unknown how many were asked and how many refused to participate)Control group unknown number of registered motorcycle riders
Non-active riders excluded from analysis
Analysis sample size
Intervention group 71
Control group 100
Interventions No formal training vs the non-mandatory Motorcycle Safety Foundationrsquos (MSF) Motorcycle Rider Course (MRC)
comprising of 8 hours theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-
road practice
Intervention integrity
Participants had all successfully completed the MRC
39Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4250
Satten 1980 (Continued)
Outcomes Self-reported crashes injuries and offences
Length of follow up 3 years
Notes Study conducted in Illinois USA
Savolainen 2007
Methods Cohort study
Participants Initial sample size
Intervention group 4000 riders that had conducted the MSF MRC course (passed or failed)
Control group 4000 motorcycle owners or permit or endorsement holders from the Indiana Bureau of Motor
Vehicles database
Analysis sample size
Intervention group 739 (18)
Control group 588 (15)
Interventions Intervention integrity
No information given
Outcomes Probability of crash involvement
Length of follow up up to 5 years
Notes Study conducted in Indiana USA
Swaddiwudhipong 1998
Methods Randomised trial
Participants Initial sample size
Intervention group
all 16 villages of 3 sub-districts
(participants were practising motorcyclists licensed and unlicensed)
Control group
18 villages of 3 other sub-districtsAnalysis sample size
Intervention group 3 districts (100)
Control group 3 districts (100)
Interventions No rider education vs non-mandatory community based rider education consisting of theory on motorcycle injury
risk helmet use traffic laws vehicle regulations and traffic signs and a written and skill test leading to a licence
Intervention integrity
No information given
Outcomes Injuries fatalities
Length of follow up 2 years
40Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4350
Swaddiwudhipong 1998 (Continued)
Notes Study conducted in Tak Province rural Thailand
Waller 1992
Methods Cohort study
Participants Initial sample size
Intervention group 349 riders who succesfully completed the Department of Public Safety motorcycle operator
training in 1988
Control group 349 motorcyclists who never completed the DPS motorcycle operator training
Analysis sample size
Intervention group 349
Control group 349
Interventions No DPS motorcycle operator training vs non-mandatory DPS motorcycle operator training The DPS training is
the Motorcycle Safety Foundationrsquos Motorcycle Rider Course comprising eight hours of theory (lectures discussions
films and multiple choice tests) a textbook and 12 hours off-road practice
Intervention integrity
All 349 riders succesfully completed the rider training
Outcomes Police recorded crashes
Length of follow up 3 years
Notes Study conducted in
Characteristics of excluded studies [ordered by study ID]
Study Reason for exclusion
Anderson 1980a Study type review and process evaluation of a rider course
Aupetit 2008 Outcome proposal of a rider training curriculum that combines traditional training and a motorcycle riding
simulator
Awane 1999 Study type review on use of simulators in rider and driver training
Baldi 2005 Intervention not rider training but intervention integrity of rider training availabilty of rider education
implementation of graduated licensing system etc on a state-wide level
Batchler 1988 Study type description of a training course
Blanchard 2006 Study type not a study assessing the association between rider training and outcomes
Bonnett 2005 Outcome qualitative data regarding satisfaction with the training programme
41Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4450
(Continued)
Braver 2007 Outcome licensure rates
Clayton 1980 Outcome knowledge skills attitude
Council 1978 Outcome crash reduction in ANY vehicle not motorcycles exclusively
Crowther 2005 Study type qualitative study
Daniello 2009 Study type review of the effectiveness of motorcycle training and licensing
Dewanti 2007 Intervention emotional characteristics behaviour (ie not motocycle rider training)
Du Plooy 1990 Study type review of types of rider training courses in South Africa
Engel 1989 Study type uncontrolled before-after study
Goldenbeld 2002 Outcome skills
Goldenbeld 2004 Outcome knowledge and skills
Haworth 2000 Intervention balance between attitudinal andvehicle control skills acrosscurriculla of motorcycle rider training
courses
Haworth 2005 Study type review
Haworth 2006 Study type cross-sectional survey
Hill 1983 Study type review and description of training system
Imatake 1980 Study type course descriptionreview
Ingebritsen 1990 Intervention risk factors for motorcycle crashes
Jain 2009 Study type cross-sectional descriptive study of motorcycle crashes over a 5 year period
Jonah 1979 Study type cross-sectional survey Outcome scores on motorcycle operator skill test
Jonah 1981 Intervention evaluation of a motorcycle rider skill test not motorcycle rider training
Kadar 1990 Study type review of components of rider training courses
Kelsey 1986 Intervention evaluation of a motorcycle rider skill test not motorcycle rider training
Koch 1980 Study type reviewcourse description
Koch 1990 Study type description of course content
42Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4550
(Continued)
Labbett 2005 Study type essay about possible explanations for motorcycle deaths
Law 2005 Intervention targeted motorcycle safety programme not specifically rider training
Lowes 1990 Study type review and descriptions of rider training
Mahon 1976 Study type review of motorcycle rider training methods
Maxwell 1985 Outcomesrecommendations forprogramme operatorsand otherson howthe motorcycle trainingprogramme
works and how it can be improved
Mayhew 1996 Study type review
McKnight 1987 Study type evaluation of methodological and administrative aspects of a motorcycle operator skill test
McPherson 1976 Outcome knowledge and skills
Mohan 2004 Study type review of evidence of interventions to reduce road traffic injuries
Morgan 1980 Study type reviewcourse description
Motoki 1990 Study type proposal and development of rider training curriculum
Nairn 1992 Study type literature review
Newland 1980 Study type description of development of motorcycle safety course
Newland 1982 Study type descriptionreview of a rider training course
Newland 1990 Study type uncontrolled BeforeAfter study
Ormston 2003 Outcome changes in attitude and self reported behaviour
Osga 1979 Outcome knowledge attitudes and skills
Panou 2007 Intervention new simulation tool to teach driverrider
Peterson 1990 Study factor demographic details of course participants completing the course vs those not completing the
course
Prem 1984 Study type experimental
Study factor examines association between skills and performance in tests
Radin Umar 2006 Intervention motorcycle safety programme not specifically rider training
Rockwell 1990 Study type methodologically flawed study type
43Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4650
(Continued)
Rothe 1987 Study type qualitative study of rider training in general
Rowden 2007 Study type qualitative study
RSC Victoria 1993 Study type review of motorcycle rider training and education in Victoria
Russam 1979 Study type review of training programmes in Japan
Saffron 1981 Study type literature review
Schulz 1990 Study factor predictors of motorcycle crashes
Sexton 2004 Study type cross-sectional analytic analysis
Shephard 1986 Study factor reviewdescription of programme
Simpson 1990 Study type review of motorcycle rider trainingeducation
Sokolov 1990 Study factor review of motorcycle riding courses in USSR
Steffens 1988 Outcome self-reported behaviour experiences and researcher observed riding behaviour
Sudlow 2003 Study type review of motorcycle training in the UK
Tasmania 1995 Study type cross-sectional descriptiveprocess evaluation
Tetard 1985 Study type experimental
Outcome detection of indices
Thackray 1980 Outcome instructional effectiveness user acceptance administrative feasibility
Thompson 1994 Study type cross-sectional survey
Outcome attitudes and knowledge
Tobler 1990 Study type review of course curricula
Transport 1989 Intervention syllabus for rider training
Tuan 2007 Study type cross-sectional study (survey) which does not assess the effects of an intervention but examines
motocyclistsrsquo behaviour
Victor 1980 Study type review
Waterer 1990 Study type review of UK rider training issues
Weller 1989 Study type review of studies evaluating motorcycle safety and education measures
44Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4750
(Continued)
Wells 1981 Study type description of different methods of training
Winn 1986 Study type review of rsquohistoryrsquo of rider training
Wisher 1988 Study type cross-sectional survey
Wood 1987 Study type Interrupted Time Series but did not meet the criteria for an ITS as published by the Cochrane
Effective Practice and Organisation of Care Group
Yamazaki 1990 Study type cross-sectional survey
Outcome riding skills skills retention and attitude
Characteristics of ongoing studies [ordered by study ID]
Burgess 2010
Trial name or title UKs National Ride program
Methods Cohort study
Participants Motorcycle offenders
Interventions Police led diversion program
Outcomes Self reported attitudinal and behavioural measures
Starting date
Contact information
Notes Expected to be completed in early 2010
45Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4850
D A T A A N D A N A L Y S E S
This review has no analyses
A P P E N D I C E S
Appendix 1 Search strategy
We searched the following electronic databases
bull Cochrane Injuries Group Specialised Register (searched 22 Sept 2008)
bull CENTRAL (The Cochrane Library 2008 Issue 3)
bull TRANSPORT 1988 to 200706 (latest date available)
bull ATRI Australian Transport Index (to Sept 2008)
bull MEDLINE (Ovid SP) 1950 to Sept 2008
bull EMBASE (Ovid SP) 1980 to Sept 2008
bull CINAHL (EBSCO) 1982 to Sept 2008
bull WHOLIS (World Health Organization Library Information System) (to Sept 2008)
bull PsycINFO (Ovid SP) 1806 to Sept 2008
bull LILACS (BIREME) (Latin American and Caribbean Health Sciences) (1982 to Sept 2008)
bull ISI Web of Science Social Sciences Citation Index (SSCI) 1970 to Sept 2008
bull ERIC (searched 25 Sept 2008)
bull ZETOC (searched 25 Sept 2008)
bull SIGLE (searched 25 Sept 2008)
All of the databases were searched using the following text-word search strategy The strategy was adapted using either rsquorsquo or rsquo$rsquo as atruncation symbol
1 motor-cycl or motorcycl or motor-bik or motorbik or motor-scooter or motorscooter or scooter or moped or motocycl
or motocicl or moto
2 train or educat or instruct or program or course or qualif or learn or improv
3 1 and 2
H I S T O R Y
Protocol first published Issue 2 2005
Review first published Issue 10 2010
46Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4950
C O N T R I B U T I O N S O F A U T H O R S
Katina Kardamanidis was the first author of the protocol conducted the literature search extracted full studies assessed the studies
according to inclusion criteria assessed quality of the studies compared findings with the second author discussed disagreements with
the third author wrote the drafts and the final review
Rebecca Ivers was the second author of the protocol assessed the studies according to inclusion criteria assessed quality of the studies
compared findings with the first author edited the final draft and provided overall guidance for the process
Alexandra Martiniuk assessed quality of the studies compared findings with the first author discussed disagreements with the third
author and helped write the first draft and the final review
Mark Stevenson contributed to the protocol reviewed methods of studies and contributed to the final draft
Katrina Thistlethwaite conducted a second literature search extracted full studies and contributed to the final draft
D E C L A R A T I O N S O F I N T E R E S T
None known
S O U R C E S O F S U P P O R T
Internal sources
bull No sources of support supplied
External sources
bull National Health and Medical Research Council of Australia Australia
Salary support for Martiniuk Ivers and Stevenson
bull University of Sydney Sesqui Research Grant Australia
Direct research costs and salary support for Kardamanidis
D I F F E R E N C E S B E T W E E N P R O T O C O L A N D R E V I E W
There are no differences between the protocol and the review
I N D E X T E R M S
Medical Subject Headings (MeSH)
lowastMotorcycles Accident Prevention [lowastmethods] Accidents Traffic [lowastprevention amp control] Licensure Program Evaluation Wounds
and Injuries [mortality lowastprevention amp control]
47Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 5050
MeSH check words
Humans
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 850
Table 1 Risk of selection bias (Continued)
McDavid 1989 Differences at baseline other than matched characteristics were not reported
Confounding factors age ownership of car licence years licensed previous crashesand offences accounted
for through matching Other confounding effects not controlled
Selection intervention group was selected from a list of training graduates and the control group was
selected from Bureau of Motor Vehicle records
Kloeden 1994 Differences at baseline reported were in geographical area of residence and in riding exposure
Confounding factors were accounted for through exclusion of participants from rural areas
Other confounding factors were not reported
Selection the Ridersafe course was phased in based on postcode of riderrsquos home address analyses were
based on a database of those who obtained motorcycle rider learnerrsquos permits of which about 2000 were
Ridersafe riders and about 5000 were control riders
Billheimer 1998 Differences at baseline reported on age and previous riding experience
Other potential confounding factors assessed were riding exposure primary purpose of riding and helmet
use
Confounding factors accounted for by matching (age sex primary purpose of riding) standardisation
(riding exposure) and stratification (previous riding experience)
Selection the intervention group were those who had taken the rider training programme the control
group were California motorcycle riders matched on the above variables but who had not taken a riding
course
Perrino 2002 Differences at baseline reported were a higher number of females in the trained group and a higher
proportion of experienced riders in the untrained group
Confounding factors were not accounted for in the analysis
Selection the intervention group were those who had done the Motorcycle Riding Course the control
group had not done the course Both groups were identified through a register of licensed motorcyclists
Kraus 1975 Population based case-control study
Cases all motorcycle riders with a confirmed medically treated motorcycle injury (including fatalities)
from all hospitals in study base and peripheral area hospitals and all deaths due to motorcycle crashes in
the study base All serious injuries (leading to hospitalisation) and fatalities likely to be included
Community controls a random sample of registered motorcycle owners in the study base
Potential confounders assessed were age sex weight height riding experience motorcycle use charac-
teristics of motorcycle and previous crashes and offences
Confounding factors were accounted for through exclusion of females and stratification by age and
motorcycle useSelection cases and controls drawn from the same population
Hurt 1981 Population based case-control study
Cases and controls not drawn from the same population
Cases motorcycle riders who had a motorcycle crash Not all reported cases included
Community controls motorcycle riders who passed the crash site at the same time of day and day of week
that the crash occurred but 2 years later
Potential confounders assessed were age sex time type of motorcycle roadway alignment helmet use
weather conditions road surface conditions
Confounding factors were accounted for through matching for time of day and day of week that crash
occurred and stratification
Selection cases were recruited from police reports not all cases were included controls were matched
6Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 950
Table 1 Risk of selection bias (Continued)
community controls from randomly selected sites matched for same time of dayday of week as the crash
occurred
Haworth 1997 Population based case-control study
Cases and controls drawn from the same population
Cases motorcycle riders or pillion riders with an injury from a motorcycle crash from hospital and police
records Not clear whether all cases were included
Community controls motorcycle rider or pillion who passed the crash site at the same time of day and
day of week that the crash occurred
Potential confounders assessed were rider factors age marital status licence status riding experience
and exposure alcohol and drugs use of protective gear including helmet use pillion factors motorcycle
factors trip factorsConfounding factors were accounted for through matching for time of day and day of week that crash
occurred and stratification for age blood alcohol level and licence status
Selectioncases were riders or pillions involved in on-road crash recruited from hospitals and police records
not sure if all cases were included controls were matched community controls who passed the crash site
at the same time of dayday of week as the crash
Savolainen 2007 Differences at baseline reported on sex safety helmet use riding ability riding experience and licence
status
Adjusted for these confounders through statistical modelling although exact process unclear
Selection intervention group from list of trainees control group from Indiana Bureau of Motor Vehicles
database Trainees and the control group were invited to participate in surveys
Davis 1997 Differences at baseline not reported but likely No adjustment for any confounders at baseline or during analysis
Selection intervention group from list of riders involved in crashes who were also on a list of training
graduates control group from the same list of riders involved in crashes who were not also on a list of
training graduates
Waller 1992 Differences at baseline not reported but likely
Adjusted for confounders through exclusion of riders younger than 18 years riders without a motorcycle
operator driver license riders with location of residence outside of the Texas metropolitan areas of Austin
through matching by age sex and location of residence through stratification by age and sex and through
standardisation (number of crashes by person-days at risk)
Selection intervention group from the Motorcycle Operator Training class roster (number of trained
riders that the sample was selected from is unclear) control group through a random number generator
generating driver license numbers until an equal sample size was selected as the intervention group
MAIDS 2004 Population based case-control study
Unclear whether cases and controls not drawn from the same population
Cases motorcycle riders who had a motorcycle crash Unclear whether all cases included
Community controls motorcycle riders recruited at petrol stations Unclear whether and how matched
to cases
No potential confounders were reported
There was no report of whether and how potentially confounding factors were accounted for in the study
7Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 1050
Table 2 Risk of allocation bias
Anderson 1980 Allocation concealment unclear
Hall 1988 Allocation concealment unclear
Swaddiwudhipong 1998 Allocation concealment unclear
Table 3 Risk of detection bias
Anderson 1980 Blinding of outcome assessors not reported
Driver records holding police recorded crashes and offences not validated
Hall 1988 Blinding of outcome assessors not reported
Driver records holding police recorded crashes and offences not validated
Swaddiwudhipong 1998 Blinding of outcome assessors not reported
Police and hospital records Post mortem examination for deaths not validated
Adams 1985 Blinding of outcome assessors not reported
Australia Post crash records not validated but argued that underreporting is unlikely
Leung 1987 Blinding of outcome assessors not reported
Driver records holding police recorded crashes and offences
Raymond 1979 Blinding of outcome assessors not reported
Self-report through personal interviews or questionnaires - both methods validated
Osga 1980 Blinding of outcome assessors not reported
Self-report through structured questionnaires validated by cross-checking with police
recorded crash records
Satten 1980 Blinding of outcome assessors not reported
Self-report through structured interviews or questionnaires - both methods were not
validated
Intervention group were given telephone interviews but the control group mailed ques-
tionnaires
Jonah 1982 Blinding of outcome assessors not reported
Self-report through structured interviews not validated Driver records holding police
recorded crashes and offences
Mortimer 1984 Blinding of outcome assessors not reported
Self-report through questionnaires not validated
Intervention group received mailed questionnaires but the control group received ques-
tionnaires personally at the motorcycle dealership
Cooper 1988 Blinding of outcome assessors not reported
Driver records holding police recorded crashes and offences
8Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 1150
Table 3 Risk of detection bias (Continued)
Mortimer 1988 Blinding of outcome assessors not reported
Self-report through questionnaires not validated
Intervention group received mailed questionnaires but the control group received ques-
tionnaires personally at the motorcycle dealership
McDavid 1989 Blinding of outcome assessors not reported
Driver records holding police recorded crashes and offences
Kloeden 1994 Blinding of outcome assessors not reported
Driver records holding police recorded crashes and offences
Billheimer 1998 Blinding of outcome assessors not reported
Police recorded crashes and offences
Perrino 2002 Blinding of outcome assessors not reported
Self-report through mailed out questionnaires police records not validated
Kraus 1975 Blinding of intervention assessors to case or control status not reported
Intervention measured police records self-report though questionnaires not validated
Length of follow up not specified
Hurt 1981 Blinding of intervention assessors to case or control status not reported
Intervention measured face-to-face in-depth interviews (ie no structured question-
naires) Validity and reliability of data ensured by reported high level of quality controlLength of follow up not specified
Haworth 1997 Blinding of intervention assessors to case or control status not reported
Intervention measured self-report either by personal or by telephone interview for
both cases and controls not validated
Length of follow up in the 2 years prior to the interview or more than 2 years ago
Savolainen 2007 Blinding of outcome assessors not reported
Source of outcome data not reported
Length of follow up up to 5 years
Davis 1997 Blinding of outcome assessors not reported
Source of outcome data police records
Length of follow up 1 year
Waller 1992 Blinding of outcome assessors not reported
Source of outcome data Department of Public Safety driver history file and official
accident reports
Length of follow up 3 years
MAIDS 2004 Blinding of intervention assessors to case or control status not reported
Intervention measured face-to-face in-depth interviews (ie no structured question-
naires) No report on whether and how validity and reliability of data was ensured
Length of follow up 2 years
9Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 1250
Table 4 Risk of attrition bias
Anderson 1980 Percentage of participants in analysis 99 forthe total group Unclear forintervention andcontrol groups
separately
Hall 1988 Percentage of participants in analysis 96-97 forall intervention groups and 98for the control group
Swaddiwudhipong 1998 Percentage of participating sub-districts in analysis 100 for both intervention and control sub-districts
Raymond 1979 Percentage of participants in analysis 74 for the intervention group and less than 57 for the control
group
Osga 1980 Percentage of participants in analysis 73 for the intervention group and 100 for the control group
Satten 1980 Percentage of participants in analysis is unknown Initial sample size is unclear
Jonah 1982 Percentage of participants in analysis 23 for the intervention group and 45 for the control group
Mortimer 1984 Percentage of participants in analysis 35 for the intervention group and 84 for the control group
Cooper 1988 Percentage of participants in analysis 98 for the total group Unclear for intervention and control
groups separately
Mortimer 1988 Percentage of participants in analysis 40 for the intervention group and 90 in the control group
Adams 1985 Percentage of participants in analysis 61 forthe total group Unclear forintervention andcontrol groups
separately
Leung 1987 Percentage of participants in analysis unclear because initial sample size was not reported
McDavid 1989 Percentage of participants in analysis 40 of final matched cohort
Kloeden 1994 Percentage of participants in analysis 39 for the intervention group and 6 for the control group
Billheimer 1998 Percentage of participants in analysis unclear percentage of the initial sample size
Perrino 2002 Percentage of participants in analysis unclear percentage of the initial sample size
Kraus 1975 Percentage of participants in analysis 58 of cases 59 of controls
Hurt 1981 Percentage of participants in analysis 18 of cases lt 29 of controls
Haworth 1997 Percentage of participants in analysis 70 of cases 47 of controls
Savolainen 2007 Percentage of participants in analysis 18 for the intervention group and 15 for the control group
Davis 1997 Percentage of participants in analysis 100 for the intervention group and unclear for the control group
10Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 1350
Table 4 Risk of attrition bias (Continued)
Waller 1992 Percentage of participants in analysis unclear percentage for the intervention group and 100 for the
control group
MAIDS 2004 Percentrage of participants in analysis unclear for both cases and controls
R E S U L T S
Description of studies
See Characteristicsof included studies Characteristics of excluded
studies Characteristics of ongoing studies
Our search resulted in 102 studies that seemed relevant of which
we excluded 79 studies because they did not meet the criteria for
considering studies for this review (see Characteristics of excluded
studies)
Basedon ourcriteriawe includeda total of 23 studies in this review
There were three randomised trials (Anderson 1980 Hall 1988
Swaddiwudhipong 1998) two non-randomised trials (Adams
1985 Leung 1987) 14 cohort studies (Raymond 1979 Osga
1980 Satten 1980 Jonah 1982 Mortimer 1984 Cooper 1988
Mortimer 1988 McDavid 1989 Waller 1992 Kloeden 1994
Davis 1997 Billheimer 1998 Perrino 2002 Savolainen 2007)
and four case-control studies (Kraus 1975 Hurt 1981 Haworth
1997 MAIDS 2004)
Over half of the studies were conducted 20 or more years
ago and the oldest was conducted 31 years ago Only three
studies were from this decade (Perrino 2002 MAIDS 2004
Savolainen 2007) Over 60 of the studies were conducted in
the USA (Kraus 1975 Anderson 1980 Osga 1980 Satten 1980
Hurt 1981 Mortimer 1984 Leung 1987 Hall 1988 Mortimer
1988 Waller 1992 Davis 1997 Billheimer 1998 Perrino 2002
Savolainen2007) three inAustralia (Adams 1985 Haworth 1997
Kloeden 1994) three in Canada (Jonah 1982 Cooper 1988
McDavid 1989) onefromthe United Kingdom (Raymond 1979)one from Europe (with sampling areas located in France Ger-
many Netherlands Spain and Italy) (MAIDS 2004) and one
in Thailand (Swaddiwudhipong 1998) Only seven studies were
published in peer reviewed journals (Kraus 1975 Jonah 1982
Mortimer 1984 Cooper 1988 Mortimer 1988 McDavid 1989
Swaddiwudhipong 1998) The other studies were reports forhigh-
way safety and road traffic organisations
The studies by Anderson 1980 Hall 1988 Kloeden 1994
Billheimer 1998 (the basic California Motorcyclist Safety Program
component of the study) and Perrino 2002 examined manda-
tory pre-licence training Cooper 1988 McDavid 1989 and
Swaddiwudhipong 1998 assessed non-mandatory training that
when passed resulted in a licence Raymond 1979 Osga 1980
Satten 1980 Jonah 1982 Mortimer 1984 Adams 1985 Leung
1987 Mortimer 1988 Waller 1992 Davis 1997 Billheimer 1998
(the Experienced Rider Course component of the study) andSavolainen 2007 assessed non-mandatory training that could be
taken by motorcycle riders with and without a licence The case-
control studies (Kraus 1975 Hurt 1981 Haworth 1997) assessed
lsquoanyrsquo type of rider training and therefore could not be grouped by
pre- or post-licence rider training The MAIDS 2004 case-control
study did make a distinction between participants having received
compulsory pre-licence training or additional (non-compulsory)
training We did not find a study that compared one form of pre-
or post-licence training (eg on-road off-road theory or a com-
bination) versus another form of training All studies compared
eithermandatory or non-mandatory formal training versus no for-
mal training with the exception of Haworth 1997 who compared
different levels of training courses (beginner intermediate and ad-vanced)
We have presented a more detailed description of the individual
studies in Characteristics of included studies We have provided
a description below of the types of motorcycle rider training that
were evaluated in the included studies
Types of motorcycle rider training evaluated in the
studies
The types of training were varied Duration of training ranged
from two or three hours to 24 hours administered over two or
three days Most training consisted of a theoretical and a practical
component usually off-road The content ranged from provid-ing riders with manuals lectures discussions showing films and
administering knowledge and skill tests Little information was
provided about the qualifications of trainers who conducted the
training
Six studies (Osga 1980 Satten 1980 Mortimer 1984 Mortimer
1988 Waller 1992 Savolainen 2007) assessed the effectiveness of
the USA-based Motorcycle Safety Foundationrsquos Motorcycle Rider
Course (MSF MRC) This course was released in 1976 It com-
prises eight hours of theory (lectures discussions films and multi-
plechoice tests) a textbook and 12 hours of off-road practice The
Motorcycle Safety Foundation is a national not-for-profit organ-
11Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 1450
isation sponsored by manufacturers and distributors of motorcy-
cles Many USA states have adopted the MSF Basic Rider Courseand Experienced Rider Course (wwwmsf-usaorg) since its release
in 1976
The Motorcycle Operators Skill Test was also developed
by the USA Motorcycle Safety Foundation in coopera-
tion with others (httpwwwwebbikeworldcommotorcycle-
trainingmotorcycle-test) The MOST is a licensing test consist-
ing of a manual knowledge and skill test (comprising eight exer-
cises of increasing difficulty and skill requirements) and a mini-
skill test conducted with a simulator Some studies compared this
alternative licensing test to the standard licensing test However
as the MOST is not a type of rider training we have included only
the two studies that assessed the MOST in combination with one
or more training components in this review Both Anderson 1980and Hall 1988 assessed the MOST in combination with a train-
ing component of three hoursrsquo duration consisting of instructions
about the skills needed to pass the MOST (30 minutes of the-
ory and two and a half hours on the motorcycle) Hall 1988 also
assessed the MOST in combination with the Motorcycle Safety
Foundationrsquos Motorcycle Rider Course as described above
Other USA studies evaluated the Wisconsin motorcycle rider
course a 22-hour off-street programme that teachesthe basic rules
of safe motorcycling (Leung 1987) the USA Maryland Motorcy-
cle Rider Course (MRC) a mandatory (for those younger than 18
years)pre-licence 20-hour coursecomprising eight hours of theory
and12 hours of practice and created by the Maryland Motor Vehi-
cle Administration in 1985 (Perrino 2002) the California Motor-cyclist Safety Program (CMSP) a mandatory (for those younger
than 18 years) pre-licence 16-hour basic course (Billheimer 1998)
and Experienced Rider Course (ERC) of eight hoursrsquo duration
(Billheimer 1998) The CMSP still offers these courses ( http
wwwca-msporg) but they are now MSF courses as described
above Davis 1997 did not give any information about the assessed
rider training
The Canadian studies evaluated the British Colombia Safety
Councilrsquos 37-hour motorcycle safety programme (no further in-
formation on the content was provided) (McDavid 1989) and
the Canada Safety Council Motorcycle Training Program (MTP)
comprising four hours of theory and 16 hours of practice (six
hours on-road) over twoweekends (Jonah 1982) The MTPcoursewas made available in 1974 and is still being offered in Canada
(httpwwwsafety-councilorgtrainingmtpmtphtm) Cooper
1988 did not give any information about the assessed rider train-
ing However the study was conducted in British Colombia in
Canada in 1988 the same time as the McDavid 1989 study was
conducted
In the United Kingdom Raymond 1979 assessed the non-manda-
tory RACACU (Royal Automobile ClubAuto-Cycle Union)
training scheme The training programme consisted of either 24
one-hour sessions or 13 two-hour sessions Twelve hours of the
training focused on theory about basic operational procedures
traffic laws and highway codes The practical session focused on
basic operational skills riding on a training area and riding in traf-fic with an instructor A proficiency test was administered at the
end of the course
One study from rural Thailand (Swaddiwudhipong 1998) evalu-
ated a community-based rider education programme consisting of
theory on motorcycle injury risk helmet use traffic laws vehicle
regulations and traffic signs and a written and skill test leading to
a licence
TheAustralian studies evaluateda rider trainingcourse of 12 hours
over two consecutive days (four hours of theory and eight hours of
practice)administeredto AustraliaPosttelegram deliverers(Adams
1985) and a mandatory pre-licence Ridersafe course consisting
of two tiers the first tier being a training course on the basics
of motorcycle riding where a pass entitled the rider to obtain alearnerrsquos permit and the second tier being a skill test where a pass
entitled the riderto obtain a probationary motorcycle licence The
training course consisted of two four-hour training sessions on
separate days involving instruction videos and actual riding on
motorcycles The skill test consisted of a four-hour preparation
course of 74 minutes of lecturediscussion 38 minutes of video
and 75 minutes of on-course riding which was the testing phase
(Kloeden 1994) The Rider Safe course is still being provided by
Transport South Australia a state government organisation (http
wwwtransportsagovau )
The duration or content of rider training in the four case-control
studies (Kraus 1975 Hurt 1981 Haworth 1997 MAIDS 2004)
was not described
Risk of bias in included studies
We describe the methodological quality of the included studies
by study type below More detailed information can be found in
Table 1 Table 2 Table 3 and Table 4
When considering the findings from these studies it is important
to note the quality (or lack thereof) of reporting of the studies
Quality of reporting is the extent to which a report provides in-
formation about the design conduct and analysis of a study ( Juumlni2001) Huwiler-Muumlntener 2002 found that even though trials
may be well reported it does not mean that the methodological
quality is of equal standard and well-conducted trials may never-
theless be reported badly The reporting quality of most studies in
this review was poor Most studies were old and did not meet con-
temporary reporting standards This made data extraction com-
plicated It also made it difficult to contact authors for additional
information
Because of significant heterogeneity between studies as well as the
poor quality of most studiespooling of studyresults to calculatean
overall measure of effect was not appropriate Instead we provided
a description of our judgement of risk of bias in the text
12Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 1550
Risk of bias in the randomised trials
There were three randomised trials (Anderson 1980 Hall 1988Swaddiwudhipong 1998) Anderson 1980 and Hall 1988 had an
initial sample size of 41188 and 26464 motorcycle licence ap-
plicants respectively Both studies assessed the effectiveness of the
pre-licence MOST with a trainingcomponent Swaddiwudhipong
1998 evaluated the effectiveness of a community-based rider ed-
ucation programme in six districts
Allocation bias
Anderson 1980 reported using random birth date tables to allocate
participants to either the intervention or control group Hall 1988
reported the use of a computerised algorithm set in a central of-
fice but also used manual random assignment when the computer
went down Swaddiwudhipong 1998 reported lsquosimple random
samplingrsquo for the delivery of the intervention where the sampling
frame included nine districts Three districts were excluded since
they were too hard to access The remaining six districts were ran-
domly allocated to the intervention (three) and the control (three)
groups Systematic sampling wasthen used ordered by village size
for the villages within districts Allocation concealment was not
reported in any of the studies The descriptions of randomisation
in these studies suggest that there may have been limitations to
the process resulting in a lack of equivalence of the intervention
and control groups at baseline and therefore the possibility of
allocation bias Swaddiwudhipong 1998 reported that before theintervention the annual incidence rate of motorcycle-related in-
juries was slightly (but not statistically significant) higher in the
intervention areas than in the control areas Participants in the
intervention districts were also more likely to have a valid licence
and to wear a helmet more often than participants in the control
districts Anderson 1980 measured differences at baseline on the
participant and prior driving record variables and found these to
be negligible Hall 1988 did not report on baseline differences
between the intervention and control groups As the Anderson
1980 and Hall 1988 studies had sample sizes of over 40000 and
26000 respectively and used some form of randomisation it is
likely that any confounding factors at baseline would have been
equally distributed between comparison and intervention groups
Selection bias
After randomisation Anderson 1980 and Hall 1988 excluded
08 and 3 participants respectively because of assignment to
the wrong group administration of the wrong intervention or
other reasons The resulting lack of analysis by rsquointention-to-treatrsquo
can introduce biasby removing the effectivenessof randomisation
Nonetheless the percentage excluded was small and the number
of overall participants large so it is unlikely to have had a major
effect on the final results In the Swaddiwudhipong 1998 study
villages were systematically sampled into intervention or control
groups The intervention was implemented and outcomes of in-
jury and death were measured at a community level Interventionand control districts were randomly selected however there were
only three intervention and three control communities
Detection bias
None of the randomised trials reported blinding of outcome as-
sessors Data collection tools used by Anderson 1980 and Hall
1988 were driver records holding police recorded crashes and of-
fences Swaddiwudhipong 1998 used police and hospital records
at a district level and post-mortem examination for deaths None
reported on validity or reliability for any of the instruments
Attrition bias
Anderson 1980 and Hall 1988 had driver records available for
96 to 100 of participants They also collected self-reported
data but the response rate to the questionnaires was very low (low-
est response rate 42 and 30 respectively) Attrition bias in
Swaddiwudhipong 1998 was not an issue as outcome data were
available for all sub-districts included in the trial
The intervention reported in Anderson 1980 and Hall 1988 tri-
als had an unanticipated effect on the participants In both trials
an average of only 28 of licence applicants in the intervention
group eventually obtained their licence compared to 35 in the
control group This indicates that the licensing process in generalseems to have deterred applicants from obtaining a licence and
although no statistical tests of the difference were conducted this
deterrence appears even greater for those who had rider training
This is not true attrition bias as the studies did not experience
differential loss to follow up from outcome measurement Instead
differential attainment of the intermediary step of licensure pre-
cluded participants from being at risk for the outcomes of interest
they are not driving a motorcycle hence they are not at risk for
motorcycle-related crashes injury death or offences In short it is
not possible to assess whether a reduction in crashes in the inter-
vention group would be due to changes in riding habits as a result
of training or simply due to delayed or never attained licensure
Intervention integrity
Neither Anderson 1980 nor Hall 1988 reported on the consis-
tency in the way the intervention was administered the possible
lsquocontaminationrsquo of participants in the intervention group by other
types of formal or informal rider training or of participants in the
control group by the intervention Swaddiwudhipong 1998 did
not report on the number of villagers in each group that actually
attended the intervention However they did report on a number
of other potential training courses that were being implemented
at the same time
13Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 1650
Analysis
Anderson 1980 and Hall 1988 acknowledging the lsquodeterrent ef-fectrsquo described above conducted two separate analyses The first
analysis included the total study population of licence applicants
over two years from the time of their application (the population
denominator) The second analysis was performed on the licensed
(intervention and control) subgroups only two years from the
time of obtaining their licence (the licensed rider denominator)
As only a third were licensed and exposed to riding the motorcy-
cle it was impossible to determine whether reported differences
in crash rates across groups were due to the intervention or to
other lsquoconfoundingrsquo characteristics of those who were determined
in their efforts to obtain the licence When analysis was restricted
only to those who gained their licence the same difficulties re-
mained The benefits of the randomisation process were thus lost
due to this lack of licensure of riders during the process As there
were only three intervention and three control groups in the study
reported by Swaddiwudhipong 1998 analyses were hindered by
smallnumbers for outcome comparisons betweenthe intervention
and the control districts
Risk of bias in the cohort studies and non-randomised
trials
We discuss the methodological quality of cohort studies (seven
prospective studies (Raymond 1979 McDavid 1989 Waller 1992
Kloeden 1994 Billheimer 1998 Perrino 2002 Savolainen 2007)and seven retrospective studies (Osga 1980 Satten 1980 Jonah
1982 Mortimer 1984 Cooper 1988 Mortimer 1988 Davis
1997)) and non-randomised trials (Adams 1985 Leung 1987) in
the one section as the biases for the two study types are similar
The quality of these studies was in general poor
Selection bias
The confounding factors expected to have been addressed in the
studies were sex age riding experience prior driving records (of-
fences crashes) previous training and the following proxies for
risk taking behaviour helmet use speed and alcoholdrug use All
of these factors have been shown to be associated with an increasedrisk of motorcycle crash (Peden 2004 Zambon 2006) The pos-
sible source of bias in cohort studies and non-randomised trials
is that groups may differ on important characteristics at baseline
participants lsquoself-selectrsquo into either the rider training group by de-
ciding to undertake a rider training course or into the lsquocontrol
grouprsquo by deciding not to undertake a course Participants who
undertake a non-mandatory rider training course may differ from
those who choose not to on characteristics such as age sex rid-
ing experience and riding behaviour as well as other possible con-
founders For example riders who take a training course may be
more safety-conscious or on the contrary may do so because they
have previously experienced a crash or because they perceive that
they have poor riding skills Therefore ideally all participants in
the intervention group need to be included in the study as wellas a random sample of controls taken from the same population
as the intervention group If that is not possible intervention and
control groups either need to be carefully matched to make them
comparable or differences at baseline need to be adjusted for in
the analysis
Most of the studies using a non-randomised trial or cohort design
compared an intervention group to a control group that differed
on a number of important factors The studies in general failed to
adjust for important differences found at baseline and their results
cannot be meaningfully interpreted Ten studies (Raymond 1979
Osga 1980 Satten 1980 Jonah 1982 Mortimer 1984 Mortimer
1988 McDavid 1989 Waller 1992 Davis 1997 Savolainen 2007)
assessed intervention and control groups that were not drawnfrom the same population In most of these cases the intervention
group was selected from a list of training graduates who were usu-
ally young female and less experienced riders whereas controls
were recruited from a general population of motorcycle riders who
were usually older male and experienced riders For three stud-
ies it was unclear how they were selected (Leung 1987 Cooper
1988 Perrino 2002) Only Kloeden 1994 recruited groups from
the same population of motorcycle applicants and Adams 1985
from a group of professional riders Nevertheless due to non-ran-
dom selection of intervention and control groups differences at
baseline are likely even though study groups were from the same
population Three studies adjusted for some confounding fac-
tors found at baseline through matching (McDavid 1989 Waller1992 Billheimer 1998) However matching has limitations in
termsof controlling for multiple confounders especially those that
are unknown to influence the effect of rider training on the out-
comes One study matched the control and intervention groups
by excluding controls (approximately 3500 (66 of initial con-
trol group) that differed from the intervention group) (Kloeden
1994) Five studies (Osga 1980 Satten 1980 Adams 1985 Davis
1997 Perrino 2002) did not adjust for differences between study
groups at baseline either in the study design or analysis Results
from these studies are therefore not reliable The other six stud-
ies (Raymond 1979 Jonah 1982 Mortimer 1984 Cooper 1988
Mortimer 1988 Leung 1987 McDavid 1989 Billheimer 1998)
adjusted for some differences between groups at baseline by eitherbi-variate or multi-variate analyses or stratification and standardi-
sation of riding exposure (outcomemiles driven) or stratification
by age group The limitation of stratification is that it is difficult to
control for many variables simultaneously because a large number
of strata will need to be generated resulting in small numbers that
cannot be meaningfully analysed Savolainen 2007 developed sta-
tistical models to assess the probability of a crash after taking the
rider training course in an intervention and control group while
examining differences by sub-groups (eg age) Participation rates
were also very low in this study with only 5584000 (1395)
in the eligible intervention group and 5884000 (147) for the
14Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 1750
control group completing surveys
Detection bias
None of the studies reported blinding of outcome assessors to the
intervention or control status of the participants Seven studies
used driver records containing police recorded data as their main
measure of outcome (Leung 1987 Cooper 1988 McDavid 1989
Waller 1992 Kloeden 1994 Davis 1997 Billheimer 1998) An-
other four studies (Raymond 1979 Satten 1980 Mortimer 1984
Mortimer 1988) only used self-report Raymond 1979 was the
only study using self-report as the main outcome measurement in
which study instruments were validated Self-report might prevent
the underreporting that has been described for the use of police-
driver records (see rsquoRisk of biasrsquo in the randomised trials above)but has other risks of bias especially if questionnaires or inter-
view methods have not been validated and tested for reliability
Recall bias arises from mistakes in recollecting events both be-
cause of failures of memory and because of looking at things lsquowith
hindsightrsquo and possibly changed views (httpwwwcochraneorg
resourcesglossaryhtm) As many studies were retrospective risk
of recall bias is likely and may have increased with length of follow
up For studies that used self-reported outcomes only it is also
possible that social desirability response bias played a role where
participants in the training group may have felt it was lsquoexpectedrsquo
to report good driving histories Multiple methods to measure the
same outcome for both groups decreases the risk of detection bias
for example because self-reported answers can be cross checkedwith other sources of information such as driver records
In this regard three studies (Osga 1980 Jonah 1982 Perrino
2002) used both self-report and driver records Osga 1980 found
that for both intervention and control groups more crashes were
reported through self-report than on state crash records an aver-
age of 54 of self-reported crashes could be found back on state
crash records One industry-based study (Adams 1985) used oc-
cupational health records which were not validated but which are
unlikely to suffer from underreporting due to work cover claims
Savolainen 2007 used information on rider crashes for a five-year
period but did not specify from where this information was ob-
tained
Attrition bias
The majority of studies (Raymond 1979 Satten 1980 Adams
1985 Leung 1987 Kloeden 1994 Davis 1997 Billheimer 1998
Perrino 2002) inadequately reported on losses of participants to
the study We can therefore not be certain that there were no sys-
tematic differences in the characteristics between the intervention
and control group participants Savolainen 2007 reported the re-
sponse rates to their baseline survey but did not report on the per-
centage of outcome data that was available for their study sample
A similar deterrent effect of rider training as described for the ran-
domised trials (Anderson 1980 Hall 1988) occurred in two stud-
ies Jonah 1982 reported that 65(14992310) of the trained rid-
ers stopped riding compared to 47 (9492029) of the untrainedriders They excluded non-active and non-licensed riders before
the study commenced so the differential choice to continue riding
after training did not affect their analyses Leung 1987 reported
that the percentage of participants in the intervention group ob-
taining their licence was 50 (no number provided by authors)
lower than in the control group This may have biased the mea-
surement of outcomes because unlicensed participants will not be
exposed to riding the motorcycle and therefore not be at risk of
being involved in a crash or violation As explained above this
makes it difficult to assess whether the reduction in crashes in the
intervention group is due to a change in driving habits or simply
due to delayed licensure
Mortimer 1984 and Mortimer 1988 describe that about 44 and30 of course graduates respectively were excluded from analysis
because they had not ridden a motorcycle in the previous year
However because these studies were retrospective and the controls
were activeriders at the time of recruitment a decision on whether
the deterrent effect was differential between the two groups could
not be made Though not reported quantitatively Kloeden 1994
also describes the effectof rider training deterringindividuals from
becoming licensed and hence exposure to riding and risk of crash
death or offences involving a motorcycle Osga 1980 reported
that 27 of the trained riders did not ride after the course and
excluded these from the analysis
Use of different data collection strategies may have contributed to
differential loss to follow up in the cohort studies Satten 1980for example interviewed the intervention group over the tele-
phone but mailed questionnaires to the control group Mortimer
1984 and Mortimer 1988 distributed questionnaires in person to
the control group but mailed them to the intervention group
This may have played a role in the observed differential response
rates where more subjects responded in the personally approached
groups
Survival bias
For those studies that assessed offences crashes and injuries but
not death survival bias may have also played a role Motorcyclists
who were fatally or seriously injured in crashes during the follow-
up period after the intervention would have been less likely or
unable to respond to surveys This could bias the effect measures
towards or away from the null None of the studies wrote about
this issue or examined the characteristics of those lost to follow up
in order to address the possibility of survival bias
Intervention integrity
Although the interventions were intended for the entire lsquoexposedrsquo
groups most studies reported difficulty in tracking participants
whoreceivedthe interventionand those whodid not Adams 1985
15Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 1850
estimatedthat 62of thetotalstudysample received the interven-
tion The remaining studies did not report the percentage of thestudy population that received the intervention as was intended
or reported that this percentage was unknown (Raymond 1979
Kloeden 1994 Satten 1980 Jonah 1982 Mortimer 1984 Leung
1987 Cooper 1988 Mortimer 1988 Billheimer 1998 Perrino
2002 Savolainen 2007) Only Raymond 1979 reported the degree
to which their intervention was delivered as intended No studies
reported on possible lsquocontaminationrsquo of participants by other inter-
ventions than the one to which they were allocated Because Osga
1980 Satten 1980 Jonah 1982 Mortimer 1984 Cooper 1988
Mortimer 1988 Waller 1992 and Davis 1997 were retrospective
studies and participants were recruited from a list of graduates it
was known that they had completed and passed the course
Analysis
In some studies some participants who reported at baseline that
they were untrained decided to enroll in a rider training pro-
gramme after the baseline survey The non-randomised trial by
Adams 1985 experienced this move of subjects from the untrained
to the trained group but failed to conduct analyses by lsquointention-
to-treatrsquo This can bias the results in randomised and non-ran-
domised trials if participants who changed groups differ signifi-
cantly with regard to potentially confounding characteristics On
the other hand Perrino 2002 also experienced this move of sub-
jects from theuntrained to the trained groupmore than 100riders(5 of the total sample) originally stated that they were untrained
at baseline but then received training between the baseline and
follow-up surveys Yet since the Perrino 2002 study was a cohort
their analyses then proceeded by including those riders who were
lsquonewlyrsquo trained as exposed Riders who were lsquonewlyrsquo trained since
baseline but with a programme other than the one under study
were excluded from the analysis Nevertheless confounding fac-
tors were not accounted for in the analyses
McDavid 1989 matched exposed with unexposed subjects but
could not find suitable matches for the majority (60) of their
subjects They could therefore only conduct analysis on the result-
ing40 (139346) which is an inappropriate andinefficient anal-
ysis Some studies that reported on the issue of non-active licensedriders(Satten 1980 Jonah 1982 Mortimer 1984 Mortimer 1988
Kloeden 1994) excluded non-active riders from analysis intro-
ducing selection bias as it is likely that riders who continued to
ride differed with regards to potentially confounding character-
istics across intervention and control groups Kloeden 1994 for
example conducted several analyses to assess the effect of various
sub-sets of the control group Original analyses included all con-
trols and led to erroneous results since only a small proportion of
controls hadgone on to get a licence after their learnerrsquos permitand
were actually exposed to the possibility of experiencing a crash
Subsequent analyses compared only intervention and control sub-
jects who had a licence Waller 1992 reported crashes by person-
days of exposure but did not explain how they could be sure that
each study intervention and control participant was an active riderand actually exposed to the risk of experiencing a crash
Potential confounders that are unequally distributed by study arm
at baseline need to be dealt with in the analysis Satten 1980
Adams 1985 and Davis 1997 did not take into account any con-
founding factors either by design or in the analysis McDavid
1989 Waller 1992 and Billheimer 1998 accounted for potential
confounding factors at the design stage by matching the subjects
in the intervention and control groups Other studies (Raymond
1979 Mortimer 1984 Mortimer 1988 Perrino 2002 and also
Waller 1992) adjusted for one or two confounders by stratifica-
tionor standardisation (eg crashesmiles driven age group) Osga
1980 conducteda univariate analysis of crash rates in which no po-
tential confounders were accounted for but also calculated cumu-lative crash rates by mileage travelled Only Jonah 1982 Cooper
1988 and Savolainen 2007 conducted a multivariate analysis
Risk of bias in the case-control studies
A case-control design is suitable for the assessment of risk fac-
tors for outcomes such as motorcycle crashes injuries or deaths
but not appropriate for the evaluation of an intervention such
as motorcycle rider training It can potentially result in spurious
conclusions regarding the efficacy of the intervention because the
intervention has not been applied at random to an intervention
and control group Even with the control of possible confounders
using a case-control design to evaluate an intervention can lead tothese kinds of results since those who choose to undertake rider
training may differ on important prognostic factors that affect the
outcome compared to those who do not choose to participate in
rider training
Selection bias
A potential source of bias in case-control studies as with all other
study types where the intervention is not randomly allocated to
participants is that comparison groups are likely to differ on po-
tentially confounding characteristics This means that ideally all
cases need to be included in the study as well as a random sample
of controls taken from the population that gave rise to the cases If that is not possible cases and controls either need to be carefully
matched to make themcomparable or differences at baseline need
to be adjusted for in the analysis All four case-control studies were
population based Kraus 1975 included all fatalities and injured
cases leading to hospitalisation and appropriately recruited a ran-
dom sample of controls from the same population Hurt 1981 did
not manage to recruit all or a random sample of cases Only half
of the notifications of crashes to the police were received in time
by the researchers the research team for unreported reasons was
only able to respond to half of those again Data collection was
therefore completed for only about a fifth of all cases Neverthe-
less this was still a large sample of 900 cases Haworth 1997 using
16Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 1950
a similar study design was more likely to have included all cases
but this is not entirely clear from the study reported It is unclearfor example whether fatalities were included as they did not use
coronerrsquos reports for the recruitment of cases The MAIDS 2004
study was conducted in five European countries but it was unclear
how cases were recruited whether they were all cases during the
two years that the study lasted or a random sample of all cases
what the response rates were etc Controls were recruited from
randomly selected petrol stations but it was likewise not clear what
the eligibility criteria for inclusion in the study were how they
were recruited (eg randomly) and what the response rate was
Hurt 1981 and Haworth 1997 both matched controls to cases by
recruiting themat the sametime ofdayandsamedayof weekas the
crash occurred Hurt 1981 however recruited controls two years
later andit is likelythat there would have beendifferences betweenthe two comparison groups One important difference was that
during data collection for cases there was very little specialised
motorcycletrainingavailable in the studyregionwhereas two years
later when control data were collected there was some training
available
These four case-control studies also suffered from selection bias
due to potential differences between those that gave consent to
participate in the study and responded to interviews andquestion-
naires and those that did not None of the studies have assessed
such differences although Hurt 1981 did acknowledge that con-
trols giving consent to cooperate may have differed from those
motorcyclists who did not
Detection bias
Measures that can be taken to reduce detection bias in case-con-
trol studies are blinding of interviewers for case or control status
blinding of the participant for the study factor of interest validat-
ing measurement instruments and using these in an identical way
for cases and controls and cross checking self-reported informa-
tion with other more objective sources such as registers None of
the studies reported blinding of the intervention assessors to case
or control status but all used consistent measurement methods for
both case and control groups
All case-control studies used self-report as their main measure toassess whether cases and controls had received rider training and
if so the type of training The major potential source of bias in
case-control studies when self-report is used as a measure is recall
bias One way of reducing recall bias is by double-checking self-
reported data with more objective data Kraus 1975 also used po-
lice accident reports for information on rider characteristics but
such information was not available about rider training Haworth
1997 and MAIDS 2004 did not report on any measures to limit
recall bias Hurt 1981 reported further questioning of cases by
the interviewer when answers did not make sense or when par-
ticipants seemed to contradict themselves as part of the in-depth
unstructured interview technique This qualitative research tech-
nique can result in valuable information but as interviewers were
not blinded to case or control status of the participants it may have led to more thorough examination of cases compared to the
controls and thus differential detection bias
Intervention integrity
The level of detail about the rider training provided in the case-
control studies differed significantly but was minimal for each
Kraus 1975 simply asked whether lsquoanyrsquo motorcycle rider training
had lsquoeverrsquo been carried out Hurt 1981 asked participants whether
they hadbeen self taught informally taught or completed formally
taught rider training courses MAIDS 2004 asked whether partic-
ipants had had pre-license training additional (non-compulsory)
training other training or no training whereas Haworth 1997asked about the number of courses that a participant had com-
pleted whetherthey were basic intermediate or advanced courses
and how long ago the training had been done None of the case-
control studies reported collecting information about further de-
tails about any training programmes (eg type of training cur-
riculum duration of training)
Analysis
Kraus 1975 accounted for confounding factors through the exclu-
sion of females and stratification by age and frequency of motor-
cycle use Hurt 1981 matched cases and controls for time of day
and day of the week that the crash occurred and did not adjust forany confounding factors in the analysis Haworth 1997 matched
groups for time of day and day of the week that the crash occurred
and stratified by age blood alcohol level and licence status in the
analysis None of the studies used multivariate methods MAIDS
2004 did not provide any information about any participant char-
acteristics other than the study exposures of interest and did not
describe any methods to adjust for potential confounding factors
Effects of interventions
We have presented the results in a descriptive manner due to the
substantial heterogeneity in study types interventions and out-comes and the generally poor quality of the studies We have
grouped the interventions by mandatory and non-mandatory pre-
licence rider training and post-licence rider training and further
by study type and by outcome (crashes injurious and fatal crashes
and offences) We discuss results that are reported as statistically
significant (whether a P value was reported or not)
Pre-licence rider training - mandatory
Randomised trials
17Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2050
Crashes
Total group of participants
Anderson 1980 found a reduction in motorcycle crash rates in
the intervention group (MOST skill test with remedial training)
compared to the control group (who received the skill testwithout
a training component) The percent differenceranged from 143
six months after the programme (P lt 005) to 211 (P lt 0001)
after one year and 127 (P lt 001) after two years Hall 1988
found no statistically significant difference in crash rates between
intervention groups at any point in time
Subgroup of participants who obtained their licence only
Anderson 1980 found a reduction in motorcycle crash rates in the
intervention arm but this was statistically significant at one year
only (24 to 218 reduction P lt 005) Hall 1988 found no
statistically significant differences
Offences
Total group of participants
The percentage of offences in Anderson 1980 after six months
and one and two years was slightly higher in the intervention
arm compared to the control group but these differences were
not statistically significant Likewise the study by Hall 1988 did
not find a statistically significant difference in offences between
groups
Subgroup of participants who obtained their licence only
Hall 1988 found no significant difference in offence rates between
the intervention and control groups Anderson 1980 found thatthe percentage of offences increased in the intervention arms com-
pared to the control group This difference was statistically signif-
icant after two years (51 to 115 increase P lt 005)
Cohort studies
Crashes
Neitherof the cohort studies (Kloeden1994 and Billheimer1998)
found a statistically significant difference in crash rates between
the intervention and control groups
Offences
Billheimer 1998 (for the California Motorcyclist Safety Program
course) and Perrino 2002 did not find a statistically significant
difference in offence rates between the intervention and control
groups
Case-control studies
Crashes
MAIDS 2004 found that a slightly higher percentage of controls
(not involved in a crash) had had pre-license rider training than
cases (involved in a crash) This difference was not statistically significant (50 of controls vs 477 of cases OR 091 95 CI
076 to 110)
Pre-licence rider training - non-mandatory
Randomised trial
Injuries and deaths
Swaddiwudhipong 1998 found a significantly lower injury ratein the intervention villages (1051000 population 95 CI 92
to 120) compared to the control villages (1691000 population
95 CI 152 to 185) one year after the intervention This differ-
ence was not maintained two years after the intervention (127
95 CI 112 to 143 intervention 151 95 CI 135 to 167
control) There was no significant difference in death rates be-
tween groups
Cohort studies
Crashes
Cooper 1988 foundno statistically significant differencein crashes
between the intervention and control groups McDavid 1989
found a lower crash rate in the intervention group (14139) com-
pared to the control group (23139) but no test of statistical sig-
nificance was undertaken
Injuries and deaths
McDavid 1989 found that the intervention group had fewer in-
jurious crashes (814 crashes) compared to the control group (16
23 crashes) but no statistical significance test was reported
18Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2150
Offences
Cooper 1988 foundno statistically significantdifference in offencerates between the intervention and control groups McDavid 1989
foundno statistically significantdifference in offence ratesbetween
the intervention and control groups
Non-mandatory rider training
Non-randomised trials
Crashes
Neither Adams 1985 nor Leung 1987 found a statistically signif-icant effect of rider training on crash rates
Offences
Leung 1987 found a statistically significant lower offence rate in
the intervention group compared to the control group for both
males and females and for all age groups except for women 35
years or older (P values not reported)
Cohort studies
Crashes
Osga 1980 in the univariate analysis reported a lower percentage
of crashesin theintervention group (that continued to ride) (28
168595) than in the control group (34 112329) In the cu-
mulative crash rate analysis however Osga1980 found statistically
significant (at the P lt 001 level) higher crash rates in the interven-
tion group than in the control group at allmileage intervals (1000
to 29999 miles) travelled After riding more than 30000 miles
(when over 50 of the trained riders had been involved in a crash
compared to approximately 40 in untrained riders) this differ-
ence was no longer statistically significant Satten 1980 found a
lower percentage of crashes in the intervention group (13 969)
compared to the control group (25 2079) but the differencewas not assessed statistically Davis 1997 found a lower crash rate
in trained riders but it was unclear how this was calculated be-
cause the size of the control group was not reported Jonah 1982
and Billheimer 1998 (for the Experienced Rider Course) found no
statistically significant difference in crashes between the interven-
tion and control groups Although the difference between those
who completed the Experienced Rider Course and no course in
Billheimer 1998 neared significance at six months after the course
(P = 007) large P values were observed for one year after (P =
030) Raymond 1979 found a statistically significant higher crash
rate for the intervention group (1861000 miles ridden) com-
pared to the control group (1181000 miles ridden) in both males
and females (P lt 0001) especially during the first 5000 miles of
riding Waller 1992 found a statistically significant higher crashrate for the intervention group (27 crashes in three years or 070
crashes10000 person days) compared to the control group (5
crashes in three years or 013 crashes10000 person days) with a
crude relative risk of 538 (X 2=1513 plt005) Mortimer 1984
and Mortimer 1988 found no statistically significant difference in
crashes between the intervention and control groups
Savolainen 2007 found that the intervention group (those who
completed the Basic Rider Course) was 44 more likely to be
involved in crashes than the control group and those who com-
pleted the course multiple times were an additional 180 more
likely to be involved in a crash This finding was based on low
numbers and no P value or confidence level was given for either
result
Injuries and deaths
Satten 1980 also found a lower percentage of injurious crashes
(33 39) in the intervention group compared to the control
group(60 1220) although again no statistical significance test
was reported
Offences
Jonah 1982 did find a significantly lower self-reportedoffence rate
in the intervention group (15) compared to the control group(34 P lt 001) (no numbers given) Satten 1980 also found
a lower self-reported offence rate in the intervention group (1
71 versus 19100) but again this was not tested for statistical
significance Mortimer 1984 Mortimer1988 and Billheimer 1998
found no statistically significant differencein offence ratesbetween
the intervention and control groups
Case-control studies
CrashesMAIDS 2004 found that a slightly higher percentage of cases
(involved in a crash) had had additional non-mandatory rider
training than controls (not involved in a crash) This difference
was not statistically significant (17 of cases vs 12 of controls
OR 146 95 CI 064 to 337)
Rider training unspecified
Case-control studies
19Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2250
Injuries and deaths
Hurt 1981 found that the odds of injurious crashes were lower inthose who had received professional training compared to those
who had not Kraus 1975 found that in the younger age group
(less than 20 years) and the older age group (over 35 years) a larger
percentage of injured riders (12 and 21 respectively) reported
prior motorcycle rider training compared with the uninjured con-
trol group (6 and 5 respectively) (P lt 001) (no numbers
given) Haworth 1997 reported several different outcomes one
beingthat having completed at leastone trainingcoursecompared
to no training courses at all had no significant effect on the odds
of crashing (afteradjustment for age blood alcohol level or licence
status (learner permitprobationaryfull)) When comparing three
different levels of training courses (beginner intermediate and ad-
vanced) with no training no significant differences were found
in the odds of crashing However after adjustment for age and
blood alcohol level completing an advanced course as compared
to a beginners course was associated with a significant decrease in
the odds of severe motorcycle crashes resulting in serious injury or
death (OR 02 95 CI 01 to 08)
D I S C U S S I O N
The objective of this review was to quantify the effectiveness of
pre- andpost-licence motorcycle rider training on the reduction of
traffic crash involvement injuries deaths and offences committedby motorcycle riders Most included studies were cohort studies
There were a few randomised and non-randomised trials and case-
control studies
All studies compared either mandatory or non-mandatory formal
training with no formal training except for Haworth 1997rsquos case-
control study that compared levels of training (beginner interme-
diate and advanced) to each other
We found studies that evaluated the effectiveness of manda-
tory pre-licence training in reducing crash and offence rates
(Anderson 1980 Hall 1988 Kloeden 1994 Billheimer 1998
(the basic California Motorcyclist Safety Program component of
the study) Perrino 2002) non-mandatory training that when
passed resulted in a licence (Cooper 1988 McDavid 1989
Swaddiwudhipong 1998) non-mandatory training that could be
taken by motorcycle riders with or without a licence (Raymond
1979 Osga 1980 Satten 1980 Jonah 1982 Mortimer 1984
Adams 1985 Leung 1987 Mortimer 1988 Waller 1992 Davis
1997 Billheimer 1998 (the Experienced Rider Course compo-
nent of the study)) and unspecified rider training on crash injury
and death rates (Kraus 1975 Hurt 1981 Haworth 1997 MAIDS
2004)
The studies provided conflicting evidence regarding motorcycle
rider training which is likely due to the poor design quality and
reporting of the studies Most of the studies included in the review
were likely to be adversely affected by significant selection biasalong with substantial detection bias(measurement and recall bias)
which limits our ability to make conclusive comments
Mandatory pre-licence rider training
There were two RCTs three cohort studies and one case-control
study that examined pre-licence ridertraining The two RCTs eval-
uated the Motorcycle Skills Operating Test (MOST) (Anderson
1980 Hall 1988) Anderson 1980 found a statistically significant
reduction in crashes in the intervention group at six months one
and twoyearsfollow up However when only participantswho ob-
tainedtheirlicence were included in theanalysis thereduction wasstatistically significant only at one year follow up The other RCT
(Hall 1988) found no statistically significant differences in crash
rates at any point of time neither in the total population group
nor in the licensed subgroup Neither of the two cohort stud-
ies assessing crash rates found a statistically significant difference
in crash rates between intervention and control groups ( Kloeden
1994 Billheimer1998) andthe case-control study (MAIDS 2004)
also did not find a statistically significant difference in the odds of
experiencing a crash between trained and untrained riders With
regards to offences neither of the RCTs found a statistically sig-
nificant difference in offence rate between the intervention and
control groups in the analysis of the total group of participants
entering the trial However Anderson 1980 found a statistically significant increase in offence rates in the intervention group but
only after two years of follow up andonly in the analysis of partici-
pants who obtained their licence The two cohort studies assessing
offence rates found no statistically significant difference between
the groups (Billheimer 1998 Perrino 2002)
None of the five studies (RCTs or cohort) found a reduction in
crashes or offences following mandatory pre-licence rider training
except for Anderson 1980 who observed a statistically significant
reduction in crashes in the intervention group at one year in both
analyses It is not clear why both RCTs that evaluated the same
MOST and training intervention observed different results Both
studies enrolled large samples and observed a similar uptake of
licensure following the intervention It is possible that differencesmay have arisen due to different study samples (California ver-
sus New York) or that the intervention delivery differed between
studies
In both trials of the Motorcycle Skills Operating Test including
rider training only a third of motorcycle licence applicants ob-
tained their motorcycle licence It is possible that a licensing pro-
cess including a skills test may reduce crashes through decreased
exposure
Non-mandatory pre-licence training
20Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2350
Three studies assessed the effect of non-mandatory pre-licence
training The Cooper 1988 cohort study found no difference incrash or offence rates between the trained and untrained groups
We cannot however be confident about this result as the study
sufferedfrom selectionbiaswhichcouldhave biasedthe measure of
effect toward the null The RCT conducted by Swaddiwudhipong
1998 examined injury and death rates and found a statistically
significant lower injury rate in the intervention villages after one
year of follow up but no longer at two year follow up There
was no statistically significant difference in death rates between
intervention and control villages Our confidence in the results of
this study are hindered by the inadequacy of the randomisation
process that led to significant differences between the intervention
and control districts on two important variables The intervention
districts were more likely than the control districts to have a validlicence andto wear a helmetSincethese differences were nottaken
into account in later analyses they could explain the differences
observed in injury rates between districts Furthermore with only
six districts randomised the sample size was probably too small
to detect a difference between intervention and control groups
had there been one The McDavid 1989 cohort study found a
lower percentage of injuries in the intervention group although
a statistical test was not conducted McDavid 1989 drew their
intervention and control groups from different populations but
did adjust for confounders although they only analysed 40 of
the initial sample (139 trained and 139 untrained) Because of
these substantial weaknesses we are unable to draw conclusions
from this studyAn overall effect of pre-licence training on the reductionof crashes
injuries deaths and offences was not found although one should
be aware that the research of this issue to date has had significant
flaws However a mandatory skill test (such as the MOST) in
combination with and even without pre-licence training as part
of a licensing process seems to reduce the number of individuals
completing the licensing procedure and therefore indirectly re-
duces crash rates through a reduction in exposure to motorcycle
riding Making licensure a more difficult process may therefore be
seen as advantageous if it delays licensure and therefore reduces
exposure It is possible that the group who chose not to continue
riding a motorcycle would not have been the group to experience a
crash or commit an offence if they had continued to ride anyhowfor example because they are more safety conscious and this was
the reason for their giving up riding in the first place If this is the
case deterring individuals from completing their licence and rid-
ing may not necessarily decrease overall crashes deaths or offence
rates In addition other potentially adverse effects of complicat-
ing a licensing procedure need to be considered such as increases
in unlicensed riding In Thailand non-mandatory rider training
offered as a community intervention may have decreased injury
rates but this needs to be confirmed through further research
Non-mandatory training
The majority of the studies assessed non-mandatory training
(training independent of any licensing procedure) and the results
were varied The two non-randomised trials assessing crash rates found no statistically significant difference in crash rates between
the intervention and control groups (Adams 1985 Leung 1987)
Of the nine cohort studies Satten 1980 found a lower percentage
of crashesin the intervention group although it is unclear whether
this was a statistically significant result Davis 1997 also found a
lower percentage of crashesin theinterventiongroup although it is
unclear with which control group the intervention group wascom-
pared Raymond 1979 Osga 1980 Waller 1992 and Savolainen
2007 found a statistically significant higher crash rate in the in-
tervention group and the remaining cohort studies (Jonah 1982
Mortimer 1984 Mortimer 1988 Billheimer 1998) found no sta-
tistically significant difference in crash rates between intervention
and control groups No conclusions can be drawn from these stud-
ies with confidence The Satten 1980 and Davis 1997 studies did
not adjust for any confounding factors even though important
differences existed between the intervention and control groups
Although Raymond 1979 and Waller 1992 adjusted for a num-
ber of confounders at baseline (Waller 1992) and in the analy-
sis (Raymond 1979 Waller 1992) the intervention and control
groups were drawn from such different populations that it is pos-
sible that unmeasured confounders were responsible for the sta-
tistically significant increased risk of crashes observed in the inter-
vention group The observed difference could also be attributed totheobserved differential loss to followup (Raymond 1979) which
could bias the measure of effect towards or away from the null
The Osga 1980 study found a protective effect of rider training
in the univariate analysis but statistically significant higher crash
rates in the intervention group when accounting for riding expo-
sure Sex (a significantly higher proportion of females in the in-
tervention group) and other potentially confounding factors that
could explain thedifferencebetween the groups were howeverun-
accounted for Finally Savolainen 2007 did control for potential
confounding variables yet it is not clear whether variables were
controlled for simultaneously or one at a time One case-control
study (MAIDS 2004) found no statistically significant reduction
in the odds of experiencing a crash between trained and untrainedriders but the quality of this study was poor mainly because the
analysis did not account for any confounding factors
One cohort study (Satten 1980) assessed the effect of non-manda-
tory rider training on injurious crashes and found a lower percent-
age of injuries in the intervention group although a statistical test
was not conducted As mentioned above the Satten 1980 studyrsquos
validity is reduced due to its lack of adjustment for confounding
factors
The only non-randomised trial that assessed offence rates (Leung
1987) found a statistically significant lower offence rate in the in-
tervention group and of the five cohort studies that assessed of-
fence rates (Satten 1980 Jonah 1982 Mortimer 1984 Mortimer
21Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2450
1988) one study (Jonah 1982) also found a statistically significant
lower offence rate in the intervention group However we have lit-tle confidence in the results from Leung 1987 since potential con-
founding factors were not appropriately addressed Jonah 1982
did conduct a multivariate analysis but was hindered by an initial
low response rate The other four cohort studies did control for
several potential confounders and found no difference in offence
rates following non-mandatory post-licence rider training
The non-randomised trials and cohort studies evaluating non-
mandatory rider training were at high risk of substantial selection
bias due to the non-random allocation of participants as well
as low participation rates and attrition bias and could therefore
not provide conclusive evidence with regard to the effect of the
training It should be noted that despite the biases and other issues
with this group of studies most didnot detectmajor effects relatedto rider training in either direction This may suggest that any
potential impact of rider training will be modest
Studies examining unspecified modes of ridertraining
Three case-control studies assessed whether injured or deceased
motorcyclists were more or less likely to have had rider training
compared to control motorcyclists All had conflicting results
Hurt1981 foundthe odds of havingan injuriouscrash waslower in
those who had received rider training whereas Kraus 1975 found
that the odds of having an injurious crash were higher in youngerand older trained riders Haworth 1997 who compared advanced
versus beginner training found no effect on motorcycle injuries
or deaths although they did find that those who had completed
advanced training had reduced chances of crashing compared to
those who had done a beginnersrsquo training course Similar issues
with uncontrolled confounding and biases of selection as described
above arose with the three case-control studies Hurt 1981 suffers
from inadequate control of confounding via poor matching and
both the Kraus 1975 and the Haworth 1997 findings were only
controlled for limited factors by means of stratified analysis It is
likely that these limitations are the reason for the contradictory
directions of effect observed for the three case-control studies
Methodological weaknesses of the studies
In summary most studies investigating the effectiveness of mo-
torcycle rider training on reducing crashes injuries and death suf-
fered from serious methodological weaknesses
The majority of studies were non-randomised and were there-
fore limited mainly by selection bias Although the approaches of
matching stratification or regression analyses can minimise bias
(Fleiss 2004) they were poorly employed in the reviewed studies
The inadequate adjustment for confounders in particular adds
to our lack of confidence in the studiesrsquo results since it has been
demonstrated that individuals who choose to take motorcycle rider
training differ from those not taking rider training For instance
trained riders often differ from controls in age sex and experi-ence usually ride less and may have a more positive attitude to
safety as indicated by the reported more frequent wearing of seat
belts (Mortimer 1982)
The two large randomised trials reported that a large proportion
of their study group were never exposed to the risk of offences or
crashes because they never attained licensure This made it impos-
sible to assess whether a reduction in crashes in the intervention
group was a result of training or due to delayed or never attained
licensure
In some studies the sample sizes were too small to detect any sta-
tistically significant differences Follow-up time after training dif-
fered between studies and was an impediment to observing any
real effects of training if they existed As riding experience ac-cumulates over time any differences in outcomes between those
trained and untrained would diminish in studies with lengthy fol-
low-up times This dilution of any possible effect of rider training
may have also occurred in studies that mixed both novice and ex-
perienced riders in the study groups since it is hypothesised that
novice riders serve to benefit most from rider training This would
increase the difficulty in finding any effect of rider training par-
ticularly when sample sizes are small
Detection bias may also account for some of the differences ob-
served between trained and untrained riders Detection bias would
have occurred if the self-report hospital driver or police records
used to record the outcomes of crash injury death or offences
differed between the trained and untrained riders Research hasdemonstrated that not all crashes are reported to police for ex-
ample small crashes or off-road crashes (Kraus 1975) even serious
injury crashes have been found to be underreported (Anderson
1980) However this would only introduce detection bias if the
reporting of outcome differs for trained and untrained riders This
does appear possible in that responsible riders may be more likely
to enroll in a rider training course and also may be more likely
to report crashes to police than irresponsible riders meaning that
the number of crashes from riskier riders could be underrepre-
sented on records An analysis to investigate the effectiveness of
rider training may then show that rider training is not effective in
reducing crashes
Rather than using reported outcomes some studies used self-re-ported survey data although this technique suffers from non-
response biases In fact two studies reported differences in de-
mographic characteristics in response rates to questionnaires
(Anderson 1980 Hall 1988) Self-reported data also introduce
the risk of recall bias survival bias and social desirability response
bias Recall bias may have been equally distributed among trained
and untrained riders howeverit is hypothesised that trained riders
may have felt it was lsquoexpectedrsquo to report good driving histories
hence leading the rider training to appear effective
22Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2550
Limitations of this review
Limitations of this review were that we found very few recentstudies (the two most recent ones published in 2002 and 2007)
and that we found very few studies conducted in low-and middle
income countries even though our search strategy was set up to
identify such studies Anotherlimitation wasthat forthe reviewed
studies we could not assess all outcomes of rider training that are
potentially useful Some of the included studies assessed outcomes
such as use of safety clothing including helmets Such outcomes
are valuable since the use of motorcycle helmets has been demon-
strated to be effective in preventing head injuries (Liu 2004) De-
spite this the use of safety clothing was not included as an out-
come in this review because the studies measuring this outcome
were limited by poor data quality due to very low response rates
Another outcome of interest in assessing the effectiveness of rider
training is lsquotype of injuryrsquo As Haworth 2005 state in their review
of motorcycle licensing and training rider training could be con-
sidered effective if injuries become less severe because motorcycle
riders have learned how to avoid more serious injuries through
rider training The studies included in this review did not specify
type of injury and so this outcome could not be assessed Last
changes in knowledge and attitudes following rider training are
other potentially useful outcomes We have excluded such studies
(Osga 1979 Clayton 1990 Goldenbeld 2004) as these outcomes
are proxies rather than direct measures of the effectiveness of mo-
torcycle rider training We opted to report on the outcomes of
crashes injury death and offences alone since there is little ev-
idence about how changes in knowledge and attitudes relate to
these outcomes
A U T H O R S rsquo C O N C L U S I O N S
Implications for practice
Due to the poor quality of studies included in this review we were
unable to draw any conclusions about the effectiveness of rider
training on crash rates injurious or fatal or offence rates The
findings do suggest that mandatory pre-licence training may be
an impediment to completing a motorcycle licensing process thuspossibly indirectly reducing crash injury death and offence rates
through a reduction in exposure to riding a motorcycle On the
basis of the existing evidence it is not clear if and what type of
training reduces the risk of crashes injuries deaths or offences in
motorcyclists and the selection of the best rider training practice
can therefore not be recommended The suggestion that manda-
tory pre-licence training indirectly reduces exposure to riding the
motorcycle because trainees are impeded from completing the li-
censing process does not help those motorcyclists who persist in
obtaining their licence and riding the motorcycle Some type of
rider training is likely necessary to teach motorcyclists to ride a
motorcycle safely Good policy is based on sound evidence and
evaluation but such evidence for motorcycle rider training is seri-
ously lacking Policy makers are therefore encouraged to include arigorous evaluation component to any new or existing rider train-
ing programme Such an evaluation should ideally be randomized
and assess whether the training programme is effective and if so
which component(s) of the programme work best
Implications for research
Given theresults of this review which found littlereliable evidence
about the effectiveness of motorcycle rider training it is necessary
to build up an evidence base of the effectiveness of motorcycle
rider training in the reduction of crash injury death and offence
rates To this end well designed randomised controlled trials need
to be conducted that examine the impact of rider training oncrashes and offences A randomised controlled trial is the preferred
method as it would significantly reduce most biases described in
this review To minimise the impact of participants delaying or
never attaining their licence and therefore exposure as found in
the randomised trials in this review such a trial should include
riders who are already licensed and reliable measures of riding
exposure in a reasonably lengthy follow up period It is expected
that a randomised controlled trial of sufficient sample size would
observe equal balance on possible confounders by study arm how-
ever an assessment of potentially important confounders (such as
helmet wearing) by study group would enable control for these
variables in the analyses if necessary It is imperative that the effec-
tiveness of each component of the rider training (such as theorypractice knowledge skills and modification of behaviour) and the
synergy of such components are evaluated Outcomes should be
objective ideally those that are routinely collected rather than self-
reported outcomes Even though riding the motorcycle poses dif-
ferent challenges and incurs different risks lessons may be learned
from research into the effects of car driver training
Motorcycles are a common means of transport in low- and middle
income countries and usage is increasing As the burden of mo-
torcycle crash victims is mostly borne in such countries research
should be focused on the effectiveness of types of rider training on
the reduction of crashes injuries and fatalities in such countries
and be supported by higher income countries This review found
only one study conducted in a middle income country that sug-gested decreasing injury rates It would be beneficial to conduct
additional research to confirm this suggestion and to enhance our
knowledge of the role of rider training in low- and middle-income
settings
March 2006 saw the launch of the University of North Carolina
Highway Safety Research Centre partnership with the Motorcycle
Safety Foundation (MSF) and plans were revealed to conduct a
study of the potential benefits of motorcycle rider training The
study is funded by the MSF and the National Highway Traffic Sa-
fety Administration (NHTSA) Called rsquoThe Discovery Projectrsquo it
will compare data on crash rates from riders participating in a basic
23Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2650
motorcycle safety class with riders participating in a basic motor-
cycle safety class as well as a variety of continuing education classesThe collection of pilot data was due to start in March 2007 and
the data collection portion of the study is expected to take three
years (Highway Safety Research Center) It is vital that researchers
and policy makers collaborate to develop furtherhigh quality eval-
uations of motorcycle rider training programmes to help inform
development of new and effective training programmes
A C K N O W L E D G E M E N T S
Staff from The University of Sydney library for helping us locate
and retrieve full reports of many papers
R E F E R E N C E S
References to studies included in this review
Adams 1985 published data only
Adams A Collingwood V Job RFS Evaluation of the 1982 Australia
Post motorcycle rider training scheme NSW Australia Traffic
Accident Research Unit Traffic Authority NSW 1985 [ RN 85]
Anderson 1980 published data only
Anderson J Ford JL Peck RC Improved motorcyclist licensing and testing project Washington DC National Highway Traffic Safety
Administration (NHTSA) USA 19801ndash217 [
DOTndashHSndash805ndash598]
Billheimer 1998 published data only
Billheimer JW Evaluation of the California Motorcyclist Safety
Program Transportation Research Record 19981640100ndash9 [
TRB Report Number 980652]
Cooper 1988 published data only
Cooper PJ Rothe JP Motorcyclists who they are and why they do
what they do Transportation Research Record 1988116878ndash85
Davis 1997 published data only
David CF Evaluation of community traffic safety programs and
motorcycle operator training programs Connecticut Department of Transportation 199763
Hall 1988 published data only
Hall CJ Buchanan L Engle R Motorcycle rider education evaluationproject Washington DC USA National Highway Traffic Safety
Administration (NHTSA) US Department of Transportation
1988 [ DTNH 22ndash80ndashCndash05123]
Haworth 1997 published data only
Haworth N Smith R Brumen I Pronk N In Monash University
Accident Research Centre editor(s) Case-control study of motorcycle
crashes Victoria Australia Department of Transport and Regional
Development The Federal Office of Road Safety 19971ndash149 [
ISBN 0 642 25501 6 ]
Hurt 1981 published data only
Hurt H Ouellet J Thom D Motorcycle accident cause factors and
identification of countermeasures USA Department of
Transportation 1981 [ Technical Report DOT HS 806ndash862]
Jonah 1982 published data onlylowast Jonah BA Dawson N Bragg B Are formally trained motorcyclists
safer Accident Analysis amp Prevention 198214(4)247ndash55
Jonah BA Dawson NE Bragg BW Evaluation of the Canada
Safety Councilrsquos Motorcycle Training Program Proceedings of theInternational Motorcycle Safety Conference Washington DC
USA 198059ndash93
Kloeden 1994 published data only
Kloeden CN Moore VM McLean AJ Evaluation of the pre-license
training program for motorcyclists in South Australia Walkerville
South Australia Australia NHMRC Road Accident Research Unit
The University of Adelaide 19941ndash37 [ ISBN 0 7308 0607 3]
Kraus 1975 published data only
Kraus JF Riggins RS Franti CE Some epidemiological features of
motorcycle collision injuries American Journal of Epidemiology 1975102(1)74ndash98
Leung 1987 published data only
Leung KS Reding VA Evaluation of the Wisconsin motorcycle rider course Wisconsin Bureau of Policy Planning and Analysis
Wisconsin Department of Transportation 1987
MAIDS 2004 published and unpublished data
MAIDS In-depth investigations of accidents involving powered
two wheelers ACEM (Acemacemeu) 2004 Vol Final report
20179
McDavid 1989 published data only
McDavid JC Lohrmann BA Lohrmann G Does motorcycle
training reduce accidents Evidence from a longitudinal quasi-
experimental study Journal of Safety Research 198920(2)61ndash72
Mortimer 1984 published data only
Mortimer R Evaluation of the motorcycle rider course Accident
Analysis amp Prevention 198416(1)63ndash71
24Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2750
Mortimer 1988 published data only
Mortimer RG A further evaluation of the motorcycle rider courseJournal of Safety Research 198819(4)187ndash96
Osga 1980 published data only
Osga GA An investigation of the riding experiences of MSF rider
course participants Dissertation South Dakota USA The
University of South Dakota 19801ndash224
Perrino 2002 published data only
Perrino CS Ahmed A Callender A Rozier E Cantwell A Stewart
O The role of Marylandrsquos motorcycle rider course in promoting saver
behaviors and attitudes Baltimore USA National Transportation
Center Morgan State University 20021ndash52
Raymond 1979 published data only
Raymond S Tatum S An evaluation of the effectiveness of the RAC
ACU motorcycle training scheme - final report Road Safety ResearchUnit Dept of Civil Engineering University of Salford London
19791ndash131
Satten 1980 published data only
Satten RS Analysis and evaluation of the motorcycle rider courses
in thirteen Illinois counties International Motorcycle Safety
Conference Washington DC 1980145ndash93
Savolainen 2007 published and unpublished data
Savolainen Peter Mannering Fred Additional evidence on the
effectiveness of motorcycle training and motorcyclistsrsquo risk-taking
behaviour Transportation Research Board 2007 Vol March 20
19
Swaddiwudhipong 1998 published data only
Swaddiwudhipong W Boonmak C Nguntra P Mahasakpan P
Effect of motorcycle rider education on changes in risk behavioursand motorcycle-related injuries in rural Thailand Tropical Medicine
and International Health 19983(10)767ndash70
Waller 1992 published data only
Waller FL An analysis of the impact of the Texas Department of
Public Safety Motorcycle Operator Training Program upon
motocycle related traffic accidents Texas State University Public
Administration Program Applied Research Projects 199277
References to studies excluded from this review
Anderson 1980a published data only
Anderson LC Better biking a motorcycle rider improvement
program International Motorcycle Safety Conference Washington
DC USA Motorcycle Safety Foundation 1980231ndash53Aupetit 2008 published and unpublished data
Aupetit S Riff J Buttelli O Espie S Ergonomics research program
on motorcycle learning Contributions to the development of a
motorcycle simulator for training Transport Research Arena
Ljubljana 20089
Awane 1999 published data only
Awane T Integrating simulators in motorcycle safety education
Journal of International Association of Traffic and Safety Sciences 199923(1)26ndash35
Baldi 2005 published data only
Baldi S Baer JD Cook AL Identifying best practices states in
motorcycle rider education and licensing Journal of Safety Research200536(1)19ndash32
Batchler 1988 published data only
Batchler MW Evaluation of the pre-license motorcycle training course at the New South Wales traffic education centre Federal Office of
Road Safety (FORS) 19881ndash122 [ Report No CR 77 ]
Blanchard 2006 published data only
Blanchard HT Tabloski PA Motorcycle safety educating riders at
the teachable moment Mosby (CV) Company 200632330ndash2
Bonnett 2005 published data only
Bonnett J Mature age skills training for experienced motorcycle
riders Motorcycle ridersrsquo association of the Australian Capital
Territory 2005
Braver 2007 published data only
Braver ER et alPersuasion and licensure a randomized controlled
intervention trial to increase licensure rates among Maryland
motorcycle owners Traffic Injury Prevention 20078(1)39ndash46Clayton 1980 published data only
Clayton AB Sudlow DE An evaluation of the effectiveness of a one day training programme for learner motorcyclists Berkshire UK
Transport and Road Research Laboratory 19801ndash23 [ ISSN
0266ndash7045]lowast Clayton AB Sudlow DE An evaluation of the effectiveness of a one-day training programme for learner motorcyclists Berkshire UK
Road Safety Division Berkshire 19801ndash23 [ISSN 0266ndash7045]
Council 1978 published data only
Council FM Desper LP Dutt AK Driver education for motorcycle operation University of North Carolina Highway Safety Research
Center 19781ndash40
Crowther 2005 published data onlylowast Crowther G Engaging with motorcyclists UK police and theBikeSafe Road Safety Programme Published proceedings of the
Australasian Road Safety Research Policing Education Conference
Wellington New Zealand 2005 issue ISBN 0ndash473ndash10636ndash11ndash11
Daniello 2009 published data only
Daniello A Gabler HC Mehta YA Effectiveness of motorcycle
training and licensing Transportation Research Board 88th Annual
Meeting 200913
Dewanti 2007 published data only
Dewanti M Challenging efforts in promoting young motorcyclist
safety in Indonesia Road Safety on Four Continents 14th
International Conference Swedish National Road and Transport
Research Institute 2007
Du Plooy 1990 published data onlyDu Plooy DW Motorcycle training in South Africa The Human
Element International Motorcycle Safety Conference Orlando
Florida USA Motorcycle Safety Foundation 19904-1 to 4-9
Engel 1989 published data only
Engel U Krogsgaringrd Thomsen L [Faeligrdselsundervisning og
propaganda ndash en faeligrdselssikkerhedsfremmende foranstaltning]
Dansk Vejtidsskrift 198912295ndash7
Goldenbeld 2002 published data only
Goldenbeld C Houwing S de Craen S The development of
driving skills of young moped riders [De ontwikkeling van de
rijvaardigheid van jonge bromfietsers] SWOV Institute for Road
Safety Research (Stichting Wetenschappelijk Onderzoek
Verkeersveiligheid) Leidschendam the Netherlands 200263
25Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2850
Goldenbeld 2004 published data only
Goldenbeld C Twisk D de Craen S Short and long term effects of moped training a field experiment Transportation Research Part F 20047(1)1ndash16
Haworth 2000 published data only
Haworth N Smith R Kowadlo N Motorcycle training in Victoria
evaluation of rider training curriculum in Victoria Monash
University Accident Research Centre Reports 2000 Vol 165 issue
3131
Haworth 2005 published data only
Haworth N Mulvihill C Review of motorcycle licensing and
training Monash University Accident Research Centre 2005 [
Report no240]
Haworth 2006 published data only
Haworth N Mulvihill C Rowdon P Teaching old dogs new tricksTraining and older motorcyclists Journal of the Australasian College of Road Safety 200718(4)20ndash5lowast Haworth NL Mulvihill C Rowden PJ Teaching old dogs new
tricks Training and older motorcyclists Australasian Road Safety
Research Policing and Education Conference Gold Coast
Queensland 2006
Hill 1983 published data only
Hill D Motorcycle rider training testing and licensing Motor
Cycle Council of New South Wales Sydney 1983
Imatake 1980 published data only
Imatake Y On the activities in Japan of motorcycle safe riding
promotion committee (MSRPC) - especially the education of
unskilled riders of class 1 motorbikes International Motorcycle
Safety Conference Washington DC USA Motorcycle Safety
Foundation 1980255ndash68
Ingebritsen 1990 published and unpublished data
Ingebrigtsen S [Risikofaktorer ved ferdsel med moped og
motorsykkel] TOslashI-rapport 66 Oslo Transportoslashkonomisk
institutt 1990
Jain 2009 published and unpublished data
Jain A et alTwo wheeler accidents on Indian roads a study from
Mangalore India Elsevier 200916130ndash3
Jonah 1979 published data only
Jonah BA Dawson NE Validation of the motorcycle operator skill
test Accident Analysis amp Prevention 197911(3)163ndash71
Jonah 1981 published data onlyJonah BA Dawson NE Bragg WE Predicting accident
involvement with the Motorcycle Operator Skill Test Accident Analysis amp Prevention 198113(4)307ndash418
Kadar 1990 published data only
Kadar KJ Countermeasures motorcycle rider education delivery
system The Human Element International Motorcycle Safety
Conference Motorcycle Safety Foundation 19906-27 to 6-41
Kelsey 1986 published data only
Kelsey SL Liddicoat C Ratz M Licensing novice motorcyclists a
comparison of Californiarsquos standard test and the MOST II (Motorcycle
Operator Skill Test) administered at centralized testing offices Sacramento California USA California Department of Motor
Vehicles Research and Development Office 19861ndash74
Koch 1980 published data only
Koch H Advanced course for motorcycle riders InternationalMotorcycle Safety Conference Washington DC USA Motorcycle
Safety Foundation 1980269ndash81
Koch 1990 published data only
Koch H Brendicke R How to influence risk taking behaviour
within motorcycle rider training programs The Human Element
International Motorcycle Safety Conference Orlando Florida
USA Motorcycle Safety Foundation 1990Motorcycle Safety
Foundation 1990 p 9-1 to 9-25A
Labbett 2005 published data only
Labbett S Langham M Training can make the problem worse
Road Safety Congress 20051ndash12
Law 2005 published and unpublished data
Law TH Radin Umar RS Zulkaurnain S Kulanthayan S Impactof the effect of economic crisis and the targeted motorcycle safety
programme on motorcycle-related accidents injuries and fatalities
in Malaysia International Journal of Injury Control and Safety
Promotion 200512(1)9ndash21
Lowes 1990 published data only
Lowes BF In-traffic rider training The Human Element
International Motorcycle Safety Conference Orlando Florida
USA Motorcycle Safety Foundation 19907-46 to 7-57
Mahon 1976 published data only
Mahon J Teaching accident avoidance on the motorcycle Journal of Traffic Safety Education 1976231ndash29
Maxwell 1985 published data only
Maxwell GS Evaluation of a Program for Training MotorcycleRiders Evaluation and Program Planning 19858(4)339ndash47
Mayhew 1996 published data only
Mayhew DR Simpson SH Effectiveness and role of driver educationand training in a graduated licensing system Ottowa Canada The
Traffic Injury Research Foundation 19961ndash89 [ ISBN
0ndash920071ndash12ndash0]
McKnight 1987 published data only
McKnight AJ Milburn BG An evaluation of the modified
motorcycle operator skill test exercises NHTSA Report 1987 [
Report No DOT HS 807 212 ]
McPherson 1976 published data only
McPherson K McKnight AJ The development and evaluation of a
motorcycle skill test manual and knowledge test Washington DCUSA National Highway Safety Administration 1976
Mohan 2004 published data only
Mohan D Evidence-based interventions for road traffic injuries in
South Asia Journal of the College of Physicians and SurgeonsPakistan 200414(12)746ndash7
Morgan 1980 published data only
Morgan JK Advanced motorcycle training course content and
study evaluation International Motorcycle Safety Conference
Washington DC USA Motorcycle Safety Foundation 1980
283ndash312
Motoki 1990 published data only
Motoki M Yamazaki S A study on effective motorcycle rider
education The Human Element International Motorcycle Safety
26Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2950
Conference Orlando Florida USA Motorcycle Safety
Foundation 19909-26 to 9-57A
Nairn 1992 published data onlylowast Nairn RJ Motorcycle safety research literature review 1987 to
1991 Federal Office of Road Safety Australia 1993 Vol Report
CR 117
Newland 1980 published data only
Newland R Darby J Automotive practicespre motorcycle rider
program International Motorcycle Safety Conference Washington
DC USA Motorcycle Safety Foundation 1980127ndash9
Newland 1982 published data only
Newland R Motorcycle rider training and licensing requirements
Driver training steering a course for the future Melbourne
Victoria Transport Regulation Board 1982
Newland 1990 published data only
Newland R Motorcycle training and licensing in Victoria 1980-
1990 The Human Element International Motorcycle Safety
Conference Orlando Florida USA Motorcycle Safety
Foundation 1990
Ormston 2003 published data only
Ormston R Dudleston A Pearson S Stradling Evaluation of
Bikesafe Scotland Scottish Executive Social Research Edinburgh
2003
Osga 1979 published data only
Osga GS Ellingstad VS An evaluation of the effectiveness of the 1978
South Dakota motorcycle rider course program South Dakota USA
University of South Dakota 19791ndash89
Panou 2007 published data only
Panou M Bekiaris E A new concept on the integration of driving
simulators in driving training the train-all approach Road Safety
on Four Continents 14th International Conference Swedish
National Road and Transport Research Institute 2007
Peterson 1990 published data only
Peterson A Kadar K Status report 1980-1989 Minnesota rider
education statistics The Human Element International
Motorcycle Safety Conference Orlando Florida USA 199013-
113 to 13-129
Prem 1984 published data only
Prem H Good MC Motorcycle rider skills assessment ACT
Australia Office of Road Safety Department of Transport 1984
1ndash377 [ Report No CR 34 ]
Radin Umar 2006 published data only
Radin Umar RS Motorcycle safety programmes in Malaysia how
effective are they International Journal of Injury Control and Safety Promotion 200613(2)71ndash9
Rockwell 1990 published data only
Rockwell TH Kiger S Carnot M An evaluation of the Ohio
motorcyclist enrichment program Ohio Department of Highway
Safety 1990
Rothe 1987 published data only
Rothe JP A motorcycle riding course Motorcyclists Image and reality Canada Insurance Corporation of British Columbia 1987
59ndash83
Rowden 2007 published data only
Rowden P Watson B Haworth N What can riders tell us aboutmotorcycle training A view from the other side of the fence
Australasian Road Safety Research Policing and Education
Conference Melbourne Australia Melbourne 2007
RSC Victoria 1993 published data only
Road Safety Committee Inquiry into Motorcycle Safety in Victoria
Parliament of Victoria 1993
Russam 1979 published data only
Russam K Motorcycle training and licensing in Japan UK Transport
and Road Research Laboratory (TRRL) 1979 [ Report 916 ]
Saffron 1981 published data only
Saffron DG Driver and rider education training and licensing a
brief review Traffic Accident Research Unit Traffic Authority of
New South Wales NSW Australia Traffic Accident Research UnitTraffic Authority of New South Wales 1981
Schulz 1990 published data only
Schulz U Motorcycle accidents among learner riders Motorcycle
licensing Evaluation of the effectiveness of a Graduated Licensing Scheme Forshungshefte des Institut fur Zweiradsicherheit 1990
Sexton 2004 published data only (unpublished sought but not used)
Sexton B Baughan C Elliott M Maycock G The accident risk of
motorcyclists 2004TRL Report TRL 607(ISSN 09684107)58
Shephard 1986 published data only
Shepherd R Do Motorcycle Safety Foundation programs work
Driver 198619(9)16ndash8
Simpson 1990 published data onlySimpson HM Mayhew DR The promotion of motorcycle safety
training education and awareness Health Education Research
Theory and Practice 19905(2)257ndash64
Sokolov 1990 published data only
Sokolov O Safe motorcycle riding some aspects of motorcycle
riding in the USSR The Human Element International
Motorcycle Safety Conference Motorcycle Safety Foundation
19904ndash37 - 4-48
Steffens 1988 published data only
Steffens U Gawatz R Willmes G Effectiveness of mofa courses
[Wirksamkeit von Mofakursen] Unfall- und sicherheitsforschung
strasenverkehr 198867168
Sudlow 2003 published data only
Sudlow D Scoping study on motorcycle training Road Safety
Research Report London UK The British Institute of Traffic
Education Research Department for Transport 2003 Vol June
[ Report No 36]
Tasmania 1995 published data only
Mannering FL Grodsky LL Motorcycle rider training scheme
Traffic Education Research Centre Armidale Australia Driver
Training Australia Limited 1995 Vol April1ndash56
Tetard 1985 published data only
Tetard C Motorcycle ridersrsquo training to detect indices in an
intersection situation [Lrsquoentrainement a la prise drsquoinformation dans
lrsquoapprentissage de la conduite motocycliste] Le Travail Humain198548(2)147ndash59
27Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3050
Thackray 1980 published data only
Thackray RM Prescott JC Field test of the motorcycle ridercourse International Motorcycle Safety Conference Washington
DC USA Motorcycle Safety Foundation 1980195ndash7
Thompson 1994 published data only
Thompson M Evaluation of compulsory basic training for
motorcyclists Transport Research Laboratory Crowthorne
Berkshire UK 1994 [ Report 63]
Tobler 1990 published data only
Tobler U Individual driving instruction for motorcycle driving
learners the Swiss experiment of motorcycle driving school The
Human Element International Motorcycle Safety Conference
Orlando Florida USA Motorcycle Safety Foundation 1990
Transport 1989 published data only
Transport Communications Road Crash Statistics Australia Bureauof Transport and Communications Economics 198914ndash20
Tuan 2007 published data only
Tuan VA Shimizu T Towards development and evaluation of the
motorcycle drivers re-education program in Vietnam modeling of
motorcycle driverrsquos undesired behaviors 11th World Conference
on Transport Research 200725
Victor 1980 published data only
Victor DJ Traffic safety education for motorcyclists International
Motorcycle Safety Conference Washington DC USA Motorcycle
Safety Foundation 1980221ndash8
Waterer 1990 published data only
Waterer TH Compulsory training for motorcyclists from liberty
to compulsion The Human Element International Motorcycle
Safety Conference Orlando Florida USA Motorcycle Safety
Foundation 19904-49 to 4-57
Weller 1989 published data only
Weller RB Chandler EW Motorcycle safety and motorcycle
education past research and survey results Journal of Public Policy
amp Marketing 1989893ndash108
Wells 1981 published data only
Wells P Observations of some RACACU motorcycle training
schemes Transport and Road Research Laboratory Crowthorne
Berkshire UK 1981
Winn 1986 published data only
Winn G Two important trends in motorcycle safety regulations
rider education and conspicuity improvement American
Motorcyclist Association 1986147ndash52
Wisher 1988 published data only
Wisher G Reid J Motorcycle accidents in Norfolk a three-year study of motorcycle training and the Part 1 test Norfolk UK Road Safety
Division Highways Department Norfolk County Council 1988
Wood 1987 published data only
Wood HT Bowen R Evaluation of the revised learner permit scheme
July 1983 to December 1985 Victoria Australia Road Traffic
Authority 19871ndash27 [ ISBN 0 7306 0247 8]
Yamazaki 1990 published data only
Yamazaki S Nakanishi M Survey on effect of Honda rider
education The Human Element International Motorcycle Safety
Conference Orlando Florida USA 199012-167 to 12-187
References to ongoing studies
Burgess 2010 unpublished data only
Burgess C Broughton P Fylan F Stradling Interim evaluation of
the UKs National Ride program In L Dorn editor(s) Driver
behaviour and Training Vol IV Aldershot Ashgate Publishing
Limited 2010
Additional references
Anderson 1980
Anderson J Ford JL Peck RC Improved motorcyclist licensing and
testing project Volume I and II National Highway Traffic Safety
Administration US department of Transportation Washington
DC 1980
Bjornstig 1985
Bjornstig U Bylund P Lekander T Brorsson B Motorcycle fatalitiesin Sweden Acta Chirurgica Scandinavica 1985151(7)577ndash81
Clayton 1990
Clayton A Sudlow DE An evaluation of the effectiveness of a one-
day training program for learner motorcyclists The Human
Element International Motorcycle Safety Conference Orlando
Florida USA 199013-105 to 13-111
Cochrane Collaboration 2008
Higgins JP Green S Cochrane Handbook for Systematic Reviews of
Interventions 500 The Cochrane Collaboration 2008
Elliot 2003
Elliott MA Baughan CJT Broughton J Grayson G Knowles J et
alMotorcycle Safety a Scoping Study TRL [ Report No
0968ndash4107]Fleiss 2004
Fleiss JL Levin B Paik MC Statistical methods for rates and proportions 3rd Edition John Wiley amp Sons Inc 2004 [ ISBN
978ndash0ndash471ndash52629ndash2 ]
FORS 1999
FORS Federal Office of Road Safety Road risk for sober licensed
motorcyclists Vol Monograph 27 Federal Office of Road Safety
19991ndash2
Haworth 2005
Haworth N Mulvihill C Review of motorcycle licensing and training
Victoria Australia Monash University Accident Research Centre
20051ndash83 [ Report No 240]
Huwiler-Muumlntener 2002Huwiler-Muumlntener K Juumlni P Junker C Egger M Quality of
Reporting of Randomized Trials as a Measure of Methodologic
Quality JAMA 20022872801ndash4
Johnston 2008
Johnston P Brooks C Savage H Fatal and serious road crashes
involving motorcyclists Road Safety Research and Analysis Report
ATSB Monograph 20 Canberra Department of Infrastructure
Transport Regional Development and Local Government 0205
081ndash26
Juumlni 2001
Juumlni P Altman DG Egger M Systematic reviews in health care
Assessing the quality of controlled clinical trials BMJ 2001323
42ndash6
28Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3150
Kopits 2003
Kopits E Cropper M Traffic fatalities and economic growth TheWorld Bank Development Research Group Infrastructure and
Environment Policy research working paper 3035 2003
Leung 1983
Leung P Chau BTW Motorcycle accidents in Hong Kong pattern
of injury treatment causative factors and socio-economic
implications Injury 198315(1)23ndash9
Liu 2004
Liu B Ivers R Norton R Blows S Lo SK Helmets for preventing
injury in motorcycle riders Cochrane Database of Systematic Reviews 2004 Issue 2 [DOI 10100214651858CD004333pub3]
Mohan 2002
Mohan D Road safety in less-motorized environments future
concerns International Journal of Epidemiology 200231(3)527ndash32Mortimer 1982
Mortimer RG An evaluation of the motorcycle rider course
Proceedings of the 20th conference of the American Association for
Automotive Medicine Arlington Heights IL USA American
Association for Automotive Medicine 198277ndash96
Mullin 1997
Mullin BT Risk factors for motorcycle injury the role of age
gender experience training and alcohol PhD thesis Auckland
New Zealand University of Auckland 1997
Noordzij 2001
Noordzij PC Forke E Brendicke R Chinn BP Integration of needs
of moped and motorcycle riders into safety measures review and
statistical analysis in the framework of the European research
project PROMISING SWOV Institute for Road Safety Research
Leidschendam The Netherlands 2001 [ Report No Dndash2001ndash5]
Osga 1979
Osga G Ellingstad VS An evaluation of the effectiveness of the 1978 South Dakota Motorcycle Riders Course Program South Dakota
University of South Dakota Vermillion Human Factors Lab 1979
Oxford English Dictionary Online
httpoxforddictionariescom Oxford English Dictionary Online
Oxford University Press 2010
Peden 2004
Peden M Scurfield R Sleet D World Report on Road TrafficInjury Prevention Geneva World Health Organization Available
at httpwwwwhointviolence˙injury˙preventionpublications
road˙trafficworld˙reporten 2004Chapter 3
Radin Umar 1995
Radin Umar R Macakay G Hills B Preliminary analysis of
motorcycle accidents Short-term impacts of the running
headlights campaign and regulations in Malaysia J Traffic Med 199523(1)17ndash27
Tsai 1995
Tsai YJ Wang JD et alCase-control study of the effectiveness of
different types of helmets for the prevention of head injuries among
motorcycle riders in Taipei Taiwan American Journal of
Epidemiology 1995142(9)974ndash81
Vis 1995
Vis AA In-depth study of the hazards of motorcycling [De
onveiligheid van motorrijden nader bekeken] SWOV Institute for
Road Safety Research Leidschendam the Netherlands 1995 [
Report No Rndash95ndash69]
Weaver 1990
Weaver EA Motorcycle safety education in the United States
Motorcycle Safety Foundation Orlando Florida USA 1990
Winn 1987
Winn G Two important trends in motorcycle safety regulations
rider education and conspicuity improvement No more
information available Detroit MI US Society of Automotive
Engineers 1987147ndash52
Yuan 2000
Yuan W The effectiveness of the rsquoride-brightrsquo legislation for
motorcycles in Singapore Accident Analysis amp Prevention 200032
(4)559ndash63
Zambon 2006
Zambon F Hasselberg M Factors affecting the severity of injuries
among young motorcyclists a Swedish nationwide cohort study
Traffic Injury Prevention 20067(2)143ndash9lowast Indicates the major publication for the study
29Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3250
C H A R A C T E R I S T I C S O F S T U D I E S
Characteristics of included studies [ordered by study ID]
Adams 1985
Methods Non-randomised trial
Participants Initial sample size
Total of 162 Australia Post telegram delivery riders
Not clear how many in intervention and control groups
Analysis sample size
Intervention group 63Control group 35
Total 98 (61)
Interventions No formal training vs voluntary rider training course of 12 hours over 2 consecutive days (theory 4 hours practice
8 hours)
Intervention integrity
No information given
Outcomes Australia Post recorded crashes
Length of follow up 1 year
Notes Study conducted in NSW Australia
Anderson 1980
Methods Randomised trial
Participants Initial sample size
Total 41188 motorcycle licence applicants
Analysis sample size
Intervention group with skills training 13160
Intervention group without skills training 12634
Control group 15080
Total 40874 (99)
Interventions Standard licensing skill test vs the Motorcycle Operators Skill Test (MOST) with or without a training component
MOST is not to be confused with rider training because it is a licensing test The training component was for
participants who failed the skill test on the first attempt was of 3 hours duration and consisted of instructions about
the skills needed to pass the skill test The training consisted of 30 min theory and 2 frac12 hours practice on the
motorcycle
Intervention integrity
33 in MOST with skills training 31 in MOST without skills training and 43 in control group obtained their
motorcycle licence in the end
Outcomes Police recorded crashes and offences
Length of follow up 3 years
30Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3350
Anderson 1980 (Continued)
Notes Study conducted in California USA
Billheimer 1998
Methods Cohort study
Participants Initial sample size
Intervention group Unclear numbers of CMSP graduates and ERC graduates
Control group Over 16000 untrained motorcyclists
Analysis sample size
1139 matched CMSP-pairs and 1182 ERC-pairs
Interventions No formal training vs the mandatory (for those younger than 18 years) pre-licence 16-hours basic California
Motorcyclist Safety Program (CMSP) and the 8-hour Experienced Rider Course (ERC)
Intervention integrity
No information given
Outcomes Police recorded crashes and offences
Length of follow up 2 years
Notes Study conducted in California USA
Cooper 1988
Methods Cohort study
Participants Initial sample size
Intervention group 418 graduates and 57 who failed
Control group 402
Analysis sample size
Intervention group unclear
Control group unclear
Total 863 (984)
Interventions No formaltrainingvs non-mandatory pre-licence rider trainingThe content of thetrainingwas notfurther explainedIntervention integrity
Participants had all completed the rider training
Outcomes Police recorded crashes and offences
Length of follow up 7 years
Notes Study conducted in British Columbia Canada
31Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3450
Davis 1997
Methods Retrospective cohort study
Participants Initial sample size
Intervention group 9320 graduates
Control group unclear
Analysis sample size
Intervention group 9320 graduates
Control group unclear
Interventions No formal training vs the non-mandatory Connecticut Motorcycle Operator Training Program The content of
training was not further explained
Intervention integrity Participants had all completed the rider training
Outcomes Police recorded crashes severity of crashes
Length of follow up 1 year
Notes Study conducted in Connecticut USA
Hall 1988
Methods Randomised trial
Participants Initial sample sizeIntervention group A 6248
Intervention group B 6616
Intervention group C 6996
Control group 6604
Total 26464 motorcycle licence applicants
Analysis sample size
Intervention group A 6012 (96)
Intervention group B 6428 (97)
Intervention group C 6778 (97)
Control group 6502 (98)
Interventions Standard licensing skill test vs the Motorcycle Operator Skill Test (MOST II) without training (A) MOST II with
3 hours training (B) or MOST II with 20 hours training (C) The 3 hours training comprised instructions aboutthe skills needed to pass the skill test (30 min theory and 2 frac12 hours practice on the motorcycle) The 20 hours
training consisted of the Motorcycle Safety Foundationrsquos Motorcycle Rider Course (MRC) comprising of 8 hours
theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-road practice
Intervention integrity
25 and 26 obtained their motorcycle licence in the lsquoMOST 3-hour trainingrsquo and lsquoMOST 20-hour trainingrsquo
intervention arm respectively In the lsquoMOST testrsquo intervention arm 29 obtained their licence and in the control
group 35
Outcomes Police recorded crashes and offences
Length of follow up 3 years
32Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3550
Hall 1988 (Continued)
Notes Study conducted in New York USA
Haworth 1997
Methods Population based case-control study
Participants Cases Motorcycle riders or pillion riders involved in motorcycle crash (on road) Recruited from hospital and police
records Not clear whether all cases (especially fatalities) were included
Community controls Multiple matched sample of motorcycle rider or pillion who passed the crash site at the same
time of day and day of week that the crash occurred
Initial sample size
Cases 336
Controls 1195
Analysis sample size
Cases 234 (70)
Controls 561 (47)
Interventions Basic handling skills course beginner intermediate or advanced course measured through self-report not validated
Outcomes Injuries with an Injury Severity Score of 5 or greater or admission for at least four hours and fatalities from a
motorcycle crash (from hospital and police records)
Length of follow up two years or more prior to the interview
Notes Study conducted in Victoria Australia
Hurt 1981
Methods Population based case-control study
Participants Cases Motorcycle riders who had a motorcycle crash that was reported to police (traffic accident reports) Not all
cases included
Community controls Multiple matched sample of motorcycle riders who passed the crash site at the same time of
day and day of week as the crash occurred but 2 years later
Initial sample size
Cases 4500 notifications of crashes to police (only 50 received in time research team able to respond to 1126 for
unreported reasons data collection completed for 900)
Controls Number approached unreported 2310 consented and interviewed
Analysis sample size
Cases 808 (808900 data on rider training) - 18 of initial sample
Controls 670 (6702310 data on rider training) - lt 29 of initial sample
Interventions Training experience self taught friends-family school-club formal-professional others measured through self-
report validated
Outcomes Injuries and fatalities from a motorcycle crash (from hospital records and coronerrsquos office)
33Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3650
Hurt 1981 (Continued)
Notes Study conducted in Los Angeles USA
Jonah 1982
Methods Cohort study
Participants Initial sample size
Intervention group
2310 licensed MTP graduates who are active riders
Control group
2029 licensed motorcyclists who are active riders
Non-licensed and non-active riders excluded
Analysis sample size
Intervention group
811 interviews (35) 533 driver records (23)
Control group 1080 interviews (71) 905 driver records (45)
Interventions No training vs the non-mandatory Canada Safety Council Motorcycle Training Program (MTP) comprising of 4
hours theory and 16 hours practise (6 hrs on-road) over 2 weekends
Intervention integrity
Participants had all successfully completed the MTP
Outcomes Self-reported and police reported crashes and offencesLength of follow up 4 years
Notes Study conducted in Ontario Canada
Kloeden 1994
Methods Cohort study
Participants Initial sample size
Intervention group 2533 licence applicants
Control group Unclear
Sample of riders obtaining probationary licenceIntervention group 1625 (64)
Control group 5015 (unclear)
Non-active riders excluded from control group (~66)
Analysis sample size
Intervention group 631 (39)
Control group 318 (6)
Interventions No formal training vs the mandatory pre-licence Ridersafe course consisting of two levelsthe first is a training course
in the basics of motorcycle riding where a pass entitles the rider to obtain a learnerrsquos permit The second is a skill
test where a pass entitles the rider to obtain a probationary motorcycle licence The training course consists of two 4
hour training sessions on separate days involving instruction videos and actual riding on motorcycles The skill test
consists of a 4 hour preparation course of 74 minutes of lecturediscussion 38 minutes of video and 75 minutes of
34Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3750
Kloeden 1994 (Continued)
on-course riding which is the testing phase
Intervention integrity
Unclear what percentage in the Ridersafe group and the control group obtained their probationary licence halfway
and their full motorcycle licence in the end
Outcomes Police reported crashes
Length of follow up 1 year
Notes Study conducted in SA Australia
Kraus 1975
Methods Population based case-control study
Participants Cases All motorcycle riders with a confirmed medically treated motorcycle injury (including fatalities) from all
hospitals in study base and peripheral area hospitals and all deaths due to motorcycle crashes in the study base All
serious injuries (leading to hospitalisation) and fatalities likely to be included
Community controls A random sample of registered motorcycle owners in the same temporary and geographical
study base as the cases arose from
Initial sample size Cases 1273
Controls 738
Analysis sample size Cases 741 (58)
Controls 434 (59)
Interventions Motorcycle rider training (not specified) measured through self-report
Outcomes Injuries leading to hospitalisation fatalities
Length of follow up not specified
Notes Study conducted in Sacramento county USA
Leung 1987
Methods Non-randomised trial
Participants Initial sample sizeIntervention group unknown number of MRC graduates
Control group unknown number of licensed motorcyclists
Analysis sample size
Intervention group 2941
Control group 43094
Interventions No formal training vs non-mandatory Wisconsin motorcycle rider course a 22 hours off-street programme that
teaches the basic rules of safe motorcycling
Intervention integrity
50 in the MRC group and 100 in the control group obtained their motorcycle licence in the end
35Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3850
Leung 1987 (Continued)
Outcomes Police recorded crashes and offences
Length of follow up up to 3 years
Notes Study conducted in Wisconsin USA
MAIDS 2004
Methods Population based case-control study
Participants Cases Motorcycle riders who had a motorcycle crash Unclear whether all cases included
Community controls Motorcycle riders identified at petrol stations Unclear whether matched to cases
Initial sample size
Cases not reported
Controls not reported
Analysis sample size
Cases 921 (Italy 200 Netherlands 200 Spain 121 Germany 250 France 150)
Controls 923 (Italy 200 Netherlands 200 Spain123 Germany 250 France 150)
Interventions Pre-license rider training or additional training (non-compulsory)
Outcomes Crashes
Notes Study in five sampling areas located in five European countries (France Germany Netherlands Spain and Italy)
McDavid 1989
Methods Cohort study
Participants Sample size after first match
Intervention group 346 male training graduates
Control group 346 male motorcycle riders licensed in same year
Sample size after second match
Intervention group 139 (40)
Control group 139 (40)
Analysis sample size
Intervention group 139 (40)
Control group 139 (40)
Interventions No formal training vs the non-mandatory pre-licence British Columbia Safety Councilrsquos 37-hour motorcycle safety
programme
Intervention integrity
Participants had all completed and passed the training
Outcomes Police reported crashes injurious crashes
Length of follow up 1-6 years
36Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3950
McDavid 1989 (Continued)
Notes Study conducted in British Columbia Canada
Mortimer 1984
Methods Cohort study
Participants Initial sample size
Intervention group 608 MRC graduates
Control group 360 motorcycle riders
Response rates
Intervention group 360 (59)
Control group 324 (90)
Non-active riders excluded
Analysis sample size
Intervention group 213 (35)
Control group 303 (84)
Interventions No formal training vs the non-mandatory Motorcycle Safety Foundationrsquos Motorcycle Rider Course comprising of
8 hours theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-road practice
Intervention integrity
Participants had all completed and passed the MRC
Outcomes Self-reported crashes and offencesLength of follow up 1 year
Notes Study conducted in Illinois USA
Mortimer 1988
Methods Cohort study
Participants Initial sample size
Intervention group 2307 MRC graduates
Control group 556 active licensed motorcycle riders
Response ratesIntervention group 1298 (56)
Control group 500 (90)
Non-active riders excluded
Analysis sample size
Intervention group 913 (40)
Control group 500 (90)
Interventions No formal training vs the non-mandatory Motorcycle Safety Foundationrsquos Motorcycle Rider Course comprising of
8 hours theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-road practice
Intervention integrity
Participants had all completed and passed the MRC
37Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4050
Mortimer 1988 (Continued)
Outcomes Self reported crashes and offences
Length of follow up 1 year
Notes Study conducted in Illinois USA
Osga 1980
Methods Cohort study
Participants Initial sample sizeIntervention group 1163 MSF MRC participants invited 818 (74) responded
Control group 329 motorcycle riders responded to recruitment strategies
Non-active riders excluded from analysis
Analysis sample size
Intervention group 595 (73)
Control group 329 (100)
Interventions No formal training vs the non-mandatory Motorcycle Safety Foundationrsquos Motorcycle Rider Course (MRC) com-
prising of 8 hours theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-road
practice
Intervention integrity
Participants had all successfully completed the MRC
Outcomes Self-reported crashes police recorded crash reports
Length of follow up 6 months - 3 12 years
Notes Study conducted in South Dakota USA
Perrino 2002
Methods Cohort study
Participants Initial sample size
Intervention group 2144 (unclear) MRC graduates (motorcycle licence applicants)
Control group 2601 (unclear) regular licensing test graduates (motorcycle licence applicants)Analysis sample size
Intervention group 1070 (unclear)
Control group 867 (unclear)
Interventions No formal training vs the mandatory (for those younger than 18 years) pre-licence Maryland Motorcycle Rider
Course (MRC) consisting of 8 hours theory and 12 hours practice
Intervention integrity
No information given
Outcomes Self-reported and police recorded crashes and offences
Length of follow up 1 year
38Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4150
Perrino 2002 (Continued)
Notes Study conducted in Maryland USA
Raymond 1979
Methods Cohort study
Participants Initial sample size
Intervention group
258 volunteers mainly (but not all) novice riders
Control group
1104 provisional licence applicants
Analysis sample size
Intervention group 190 (74)
Control group 627 (57)
Interventions No formal training vs the non-mandatory RACACU (Royal Automobile ClubAuto-Cycle Union) training scheme
The training program consisted of either 24 one-hour sessions or 13 two-hour sessions Twelve hours of the training
focused on theory about basic operational procedures traffic laws and highway code The practical session focused
on basic operational skills riding on a training area and riding in traffic with an instructor A proficiency test was
administered at the end of the course
Intervention integrity
No information given
Outcomes Self-reported crashes (including minor non-injury non-damage and off-road crashes)
Length of follow up 1 year
Notes Study conducted in the UK
Satten 1980
Methods Cohort study
Participants Initial sample size
Intervention group interviewers kept interviewing until they had 100 riders selected from a population of 891
enrollees in the MSF MRC course (unknown how many were asked and how many refused to participate)Control group unknown number of registered motorcycle riders
Non-active riders excluded from analysis
Analysis sample size
Intervention group 71
Control group 100
Interventions No formal training vs the non-mandatory Motorcycle Safety Foundationrsquos (MSF) Motorcycle Rider Course (MRC)
comprising of 8 hours theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-
road practice
Intervention integrity
Participants had all successfully completed the MRC
39Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4250
Satten 1980 (Continued)
Outcomes Self-reported crashes injuries and offences
Length of follow up 3 years
Notes Study conducted in Illinois USA
Savolainen 2007
Methods Cohort study
Participants Initial sample size
Intervention group 4000 riders that had conducted the MSF MRC course (passed or failed)
Control group 4000 motorcycle owners or permit or endorsement holders from the Indiana Bureau of Motor
Vehicles database
Analysis sample size
Intervention group 739 (18)
Control group 588 (15)
Interventions Intervention integrity
No information given
Outcomes Probability of crash involvement
Length of follow up up to 5 years
Notes Study conducted in Indiana USA
Swaddiwudhipong 1998
Methods Randomised trial
Participants Initial sample size
Intervention group
all 16 villages of 3 sub-districts
(participants were practising motorcyclists licensed and unlicensed)
Control group
18 villages of 3 other sub-districtsAnalysis sample size
Intervention group 3 districts (100)
Control group 3 districts (100)
Interventions No rider education vs non-mandatory community based rider education consisting of theory on motorcycle injury
risk helmet use traffic laws vehicle regulations and traffic signs and a written and skill test leading to a licence
Intervention integrity
No information given
Outcomes Injuries fatalities
Length of follow up 2 years
40Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4350
Swaddiwudhipong 1998 (Continued)
Notes Study conducted in Tak Province rural Thailand
Waller 1992
Methods Cohort study
Participants Initial sample size
Intervention group 349 riders who succesfully completed the Department of Public Safety motorcycle operator
training in 1988
Control group 349 motorcyclists who never completed the DPS motorcycle operator training
Analysis sample size
Intervention group 349
Control group 349
Interventions No DPS motorcycle operator training vs non-mandatory DPS motorcycle operator training The DPS training is
the Motorcycle Safety Foundationrsquos Motorcycle Rider Course comprising eight hours of theory (lectures discussions
films and multiple choice tests) a textbook and 12 hours off-road practice
Intervention integrity
All 349 riders succesfully completed the rider training
Outcomes Police recorded crashes
Length of follow up 3 years
Notes Study conducted in
Characteristics of excluded studies [ordered by study ID]
Study Reason for exclusion
Anderson 1980a Study type review and process evaluation of a rider course
Aupetit 2008 Outcome proposal of a rider training curriculum that combines traditional training and a motorcycle riding
simulator
Awane 1999 Study type review on use of simulators in rider and driver training
Baldi 2005 Intervention not rider training but intervention integrity of rider training availabilty of rider education
implementation of graduated licensing system etc on a state-wide level
Batchler 1988 Study type description of a training course
Blanchard 2006 Study type not a study assessing the association between rider training and outcomes
Bonnett 2005 Outcome qualitative data regarding satisfaction with the training programme
41Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4450
(Continued)
Braver 2007 Outcome licensure rates
Clayton 1980 Outcome knowledge skills attitude
Council 1978 Outcome crash reduction in ANY vehicle not motorcycles exclusively
Crowther 2005 Study type qualitative study
Daniello 2009 Study type review of the effectiveness of motorcycle training and licensing
Dewanti 2007 Intervention emotional characteristics behaviour (ie not motocycle rider training)
Du Plooy 1990 Study type review of types of rider training courses in South Africa
Engel 1989 Study type uncontrolled before-after study
Goldenbeld 2002 Outcome skills
Goldenbeld 2004 Outcome knowledge and skills
Haworth 2000 Intervention balance between attitudinal andvehicle control skills acrosscurriculla of motorcycle rider training
courses
Haworth 2005 Study type review
Haworth 2006 Study type cross-sectional survey
Hill 1983 Study type review and description of training system
Imatake 1980 Study type course descriptionreview
Ingebritsen 1990 Intervention risk factors for motorcycle crashes
Jain 2009 Study type cross-sectional descriptive study of motorcycle crashes over a 5 year period
Jonah 1979 Study type cross-sectional survey Outcome scores on motorcycle operator skill test
Jonah 1981 Intervention evaluation of a motorcycle rider skill test not motorcycle rider training
Kadar 1990 Study type review of components of rider training courses
Kelsey 1986 Intervention evaluation of a motorcycle rider skill test not motorcycle rider training
Koch 1980 Study type reviewcourse description
Koch 1990 Study type description of course content
42Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4550
(Continued)
Labbett 2005 Study type essay about possible explanations for motorcycle deaths
Law 2005 Intervention targeted motorcycle safety programme not specifically rider training
Lowes 1990 Study type review and descriptions of rider training
Mahon 1976 Study type review of motorcycle rider training methods
Maxwell 1985 Outcomesrecommendations forprogramme operatorsand otherson howthe motorcycle trainingprogramme
works and how it can be improved
Mayhew 1996 Study type review
McKnight 1987 Study type evaluation of methodological and administrative aspects of a motorcycle operator skill test
McPherson 1976 Outcome knowledge and skills
Mohan 2004 Study type review of evidence of interventions to reduce road traffic injuries
Morgan 1980 Study type reviewcourse description
Motoki 1990 Study type proposal and development of rider training curriculum
Nairn 1992 Study type literature review
Newland 1980 Study type description of development of motorcycle safety course
Newland 1982 Study type descriptionreview of a rider training course
Newland 1990 Study type uncontrolled BeforeAfter study
Ormston 2003 Outcome changes in attitude and self reported behaviour
Osga 1979 Outcome knowledge attitudes and skills
Panou 2007 Intervention new simulation tool to teach driverrider
Peterson 1990 Study factor demographic details of course participants completing the course vs those not completing the
course
Prem 1984 Study type experimental
Study factor examines association between skills and performance in tests
Radin Umar 2006 Intervention motorcycle safety programme not specifically rider training
Rockwell 1990 Study type methodologically flawed study type
43Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4650
(Continued)
Rothe 1987 Study type qualitative study of rider training in general
Rowden 2007 Study type qualitative study
RSC Victoria 1993 Study type review of motorcycle rider training and education in Victoria
Russam 1979 Study type review of training programmes in Japan
Saffron 1981 Study type literature review
Schulz 1990 Study factor predictors of motorcycle crashes
Sexton 2004 Study type cross-sectional analytic analysis
Shephard 1986 Study factor reviewdescription of programme
Simpson 1990 Study type review of motorcycle rider trainingeducation
Sokolov 1990 Study factor review of motorcycle riding courses in USSR
Steffens 1988 Outcome self-reported behaviour experiences and researcher observed riding behaviour
Sudlow 2003 Study type review of motorcycle training in the UK
Tasmania 1995 Study type cross-sectional descriptiveprocess evaluation
Tetard 1985 Study type experimental
Outcome detection of indices
Thackray 1980 Outcome instructional effectiveness user acceptance administrative feasibility
Thompson 1994 Study type cross-sectional survey
Outcome attitudes and knowledge
Tobler 1990 Study type review of course curricula
Transport 1989 Intervention syllabus for rider training
Tuan 2007 Study type cross-sectional study (survey) which does not assess the effects of an intervention but examines
motocyclistsrsquo behaviour
Victor 1980 Study type review
Waterer 1990 Study type review of UK rider training issues
Weller 1989 Study type review of studies evaluating motorcycle safety and education measures
44Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4750
(Continued)
Wells 1981 Study type description of different methods of training
Winn 1986 Study type review of rsquohistoryrsquo of rider training
Wisher 1988 Study type cross-sectional survey
Wood 1987 Study type Interrupted Time Series but did not meet the criteria for an ITS as published by the Cochrane
Effective Practice and Organisation of Care Group
Yamazaki 1990 Study type cross-sectional survey
Outcome riding skills skills retention and attitude
Characteristics of ongoing studies [ordered by study ID]
Burgess 2010
Trial name or title UKs National Ride program
Methods Cohort study
Participants Motorcycle offenders
Interventions Police led diversion program
Outcomes Self reported attitudinal and behavioural measures
Starting date
Contact information
Notes Expected to be completed in early 2010
45Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4850
D A T A A N D A N A L Y S E S
This review has no analyses
A P P E N D I C E S
Appendix 1 Search strategy
We searched the following electronic databases
bull Cochrane Injuries Group Specialised Register (searched 22 Sept 2008)
bull CENTRAL (The Cochrane Library 2008 Issue 3)
bull TRANSPORT 1988 to 200706 (latest date available)
bull ATRI Australian Transport Index (to Sept 2008)
bull MEDLINE (Ovid SP) 1950 to Sept 2008
bull EMBASE (Ovid SP) 1980 to Sept 2008
bull CINAHL (EBSCO) 1982 to Sept 2008
bull WHOLIS (World Health Organization Library Information System) (to Sept 2008)
bull PsycINFO (Ovid SP) 1806 to Sept 2008
bull LILACS (BIREME) (Latin American and Caribbean Health Sciences) (1982 to Sept 2008)
bull ISI Web of Science Social Sciences Citation Index (SSCI) 1970 to Sept 2008
bull ERIC (searched 25 Sept 2008)
bull ZETOC (searched 25 Sept 2008)
bull SIGLE (searched 25 Sept 2008)
All of the databases were searched using the following text-word search strategy The strategy was adapted using either rsquorsquo or rsquo$rsquo as atruncation symbol
1 motor-cycl or motorcycl or motor-bik or motorbik or motor-scooter or motorscooter or scooter or moped or motocycl
or motocicl or moto
2 train or educat or instruct or program or course or qualif or learn or improv
3 1 and 2
H I S T O R Y
Protocol first published Issue 2 2005
Review first published Issue 10 2010
46Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4950
C O N T R I B U T I O N S O F A U T H O R S
Katina Kardamanidis was the first author of the protocol conducted the literature search extracted full studies assessed the studies
according to inclusion criteria assessed quality of the studies compared findings with the second author discussed disagreements with
the third author wrote the drafts and the final review
Rebecca Ivers was the second author of the protocol assessed the studies according to inclusion criteria assessed quality of the studies
compared findings with the first author edited the final draft and provided overall guidance for the process
Alexandra Martiniuk assessed quality of the studies compared findings with the first author discussed disagreements with the third
author and helped write the first draft and the final review
Mark Stevenson contributed to the protocol reviewed methods of studies and contributed to the final draft
Katrina Thistlethwaite conducted a second literature search extracted full studies and contributed to the final draft
D E C L A R A T I O N S O F I N T E R E S T
None known
S O U R C E S O F S U P P O R T
Internal sources
bull No sources of support supplied
External sources
bull National Health and Medical Research Council of Australia Australia
Salary support for Martiniuk Ivers and Stevenson
bull University of Sydney Sesqui Research Grant Australia
Direct research costs and salary support for Kardamanidis
D I F F E R E N C E S B E T W E E N P R O T O C O L A N D R E V I E W
There are no differences between the protocol and the review
I N D E X T E R M S
Medical Subject Headings (MeSH)
lowastMotorcycles Accident Prevention [lowastmethods] Accidents Traffic [lowastprevention amp control] Licensure Program Evaluation Wounds
and Injuries [mortality lowastprevention amp control]
47Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 5050
MeSH check words
Humans
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 950
Table 1 Risk of selection bias (Continued)
community controls from randomly selected sites matched for same time of dayday of week as the crash
occurred
Haworth 1997 Population based case-control study
Cases and controls drawn from the same population
Cases motorcycle riders or pillion riders with an injury from a motorcycle crash from hospital and police
records Not clear whether all cases were included
Community controls motorcycle rider or pillion who passed the crash site at the same time of day and
day of week that the crash occurred
Potential confounders assessed were rider factors age marital status licence status riding experience
and exposure alcohol and drugs use of protective gear including helmet use pillion factors motorcycle
factors trip factorsConfounding factors were accounted for through matching for time of day and day of week that crash
occurred and stratification for age blood alcohol level and licence status
Selectioncases were riders or pillions involved in on-road crash recruited from hospitals and police records
not sure if all cases were included controls were matched community controls who passed the crash site
at the same time of dayday of week as the crash
Savolainen 2007 Differences at baseline reported on sex safety helmet use riding ability riding experience and licence
status
Adjusted for these confounders through statistical modelling although exact process unclear
Selection intervention group from list of trainees control group from Indiana Bureau of Motor Vehicles
database Trainees and the control group were invited to participate in surveys
Davis 1997 Differences at baseline not reported but likely No adjustment for any confounders at baseline or during analysis
Selection intervention group from list of riders involved in crashes who were also on a list of training
graduates control group from the same list of riders involved in crashes who were not also on a list of
training graduates
Waller 1992 Differences at baseline not reported but likely
Adjusted for confounders through exclusion of riders younger than 18 years riders without a motorcycle
operator driver license riders with location of residence outside of the Texas metropolitan areas of Austin
through matching by age sex and location of residence through stratification by age and sex and through
standardisation (number of crashes by person-days at risk)
Selection intervention group from the Motorcycle Operator Training class roster (number of trained
riders that the sample was selected from is unclear) control group through a random number generator
generating driver license numbers until an equal sample size was selected as the intervention group
MAIDS 2004 Population based case-control study
Unclear whether cases and controls not drawn from the same population
Cases motorcycle riders who had a motorcycle crash Unclear whether all cases included
Community controls motorcycle riders recruited at petrol stations Unclear whether and how matched
to cases
No potential confounders were reported
There was no report of whether and how potentially confounding factors were accounted for in the study
7Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 1050
Table 2 Risk of allocation bias
Anderson 1980 Allocation concealment unclear
Hall 1988 Allocation concealment unclear
Swaddiwudhipong 1998 Allocation concealment unclear
Table 3 Risk of detection bias
Anderson 1980 Blinding of outcome assessors not reported
Driver records holding police recorded crashes and offences not validated
Hall 1988 Blinding of outcome assessors not reported
Driver records holding police recorded crashes and offences not validated
Swaddiwudhipong 1998 Blinding of outcome assessors not reported
Police and hospital records Post mortem examination for deaths not validated
Adams 1985 Blinding of outcome assessors not reported
Australia Post crash records not validated but argued that underreporting is unlikely
Leung 1987 Blinding of outcome assessors not reported
Driver records holding police recorded crashes and offences
Raymond 1979 Blinding of outcome assessors not reported
Self-report through personal interviews or questionnaires - both methods validated
Osga 1980 Blinding of outcome assessors not reported
Self-report through structured questionnaires validated by cross-checking with police
recorded crash records
Satten 1980 Blinding of outcome assessors not reported
Self-report through structured interviews or questionnaires - both methods were not
validated
Intervention group were given telephone interviews but the control group mailed ques-
tionnaires
Jonah 1982 Blinding of outcome assessors not reported
Self-report through structured interviews not validated Driver records holding police
recorded crashes and offences
Mortimer 1984 Blinding of outcome assessors not reported
Self-report through questionnaires not validated
Intervention group received mailed questionnaires but the control group received ques-
tionnaires personally at the motorcycle dealership
Cooper 1988 Blinding of outcome assessors not reported
Driver records holding police recorded crashes and offences
8Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 1150
Table 3 Risk of detection bias (Continued)
Mortimer 1988 Blinding of outcome assessors not reported
Self-report through questionnaires not validated
Intervention group received mailed questionnaires but the control group received ques-
tionnaires personally at the motorcycle dealership
McDavid 1989 Blinding of outcome assessors not reported
Driver records holding police recorded crashes and offences
Kloeden 1994 Blinding of outcome assessors not reported
Driver records holding police recorded crashes and offences
Billheimer 1998 Blinding of outcome assessors not reported
Police recorded crashes and offences
Perrino 2002 Blinding of outcome assessors not reported
Self-report through mailed out questionnaires police records not validated
Kraus 1975 Blinding of intervention assessors to case or control status not reported
Intervention measured police records self-report though questionnaires not validated
Length of follow up not specified
Hurt 1981 Blinding of intervention assessors to case or control status not reported
Intervention measured face-to-face in-depth interviews (ie no structured question-
naires) Validity and reliability of data ensured by reported high level of quality controlLength of follow up not specified
Haworth 1997 Blinding of intervention assessors to case or control status not reported
Intervention measured self-report either by personal or by telephone interview for
both cases and controls not validated
Length of follow up in the 2 years prior to the interview or more than 2 years ago
Savolainen 2007 Blinding of outcome assessors not reported
Source of outcome data not reported
Length of follow up up to 5 years
Davis 1997 Blinding of outcome assessors not reported
Source of outcome data police records
Length of follow up 1 year
Waller 1992 Blinding of outcome assessors not reported
Source of outcome data Department of Public Safety driver history file and official
accident reports
Length of follow up 3 years
MAIDS 2004 Blinding of intervention assessors to case or control status not reported
Intervention measured face-to-face in-depth interviews (ie no structured question-
naires) No report on whether and how validity and reliability of data was ensured
Length of follow up 2 years
9Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 1250
Table 4 Risk of attrition bias
Anderson 1980 Percentage of participants in analysis 99 forthe total group Unclear forintervention andcontrol groups
separately
Hall 1988 Percentage of participants in analysis 96-97 forall intervention groups and 98for the control group
Swaddiwudhipong 1998 Percentage of participating sub-districts in analysis 100 for both intervention and control sub-districts
Raymond 1979 Percentage of participants in analysis 74 for the intervention group and less than 57 for the control
group
Osga 1980 Percentage of participants in analysis 73 for the intervention group and 100 for the control group
Satten 1980 Percentage of participants in analysis is unknown Initial sample size is unclear
Jonah 1982 Percentage of participants in analysis 23 for the intervention group and 45 for the control group
Mortimer 1984 Percentage of participants in analysis 35 for the intervention group and 84 for the control group
Cooper 1988 Percentage of participants in analysis 98 for the total group Unclear for intervention and control
groups separately
Mortimer 1988 Percentage of participants in analysis 40 for the intervention group and 90 in the control group
Adams 1985 Percentage of participants in analysis 61 forthe total group Unclear forintervention andcontrol groups
separately
Leung 1987 Percentage of participants in analysis unclear because initial sample size was not reported
McDavid 1989 Percentage of participants in analysis 40 of final matched cohort
Kloeden 1994 Percentage of participants in analysis 39 for the intervention group and 6 for the control group
Billheimer 1998 Percentage of participants in analysis unclear percentage of the initial sample size
Perrino 2002 Percentage of participants in analysis unclear percentage of the initial sample size
Kraus 1975 Percentage of participants in analysis 58 of cases 59 of controls
Hurt 1981 Percentage of participants in analysis 18 of cases lt 29 of controls
Haworth 1997 Percentage of participants in analysis 70 of cases 47 of controls
Savolainen 2007 Percentage of participants in analysis 18 for the intervention group and 15 for the control group
Davis 1997 Percentage of participants in analysis 100 for the intervention group and unclear for the control group
10Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 1350
Table 4 Risk of attrition bias (Continued)
Waller 1992 Percentage of participants in analysis unclear percentage for the intervention group and 100 for the
control group
MAIDS 2004 Percentrage of participants in analysis unclear for both cases and controls
R E S U L T S
Description of studies
See Characteristicsof included studies Characteristics of excluded
studies Characteristics of ongoing studies
Our search resulted in 102 studies that seemed relevant of which
we excluded 79 studies because they did not meet the criteria for
considering studies for this review (see Characteristics of excluded
studies)
Basedon ourcriteriawe includeda total of 23 studies in this review
There were three randomised trials (Anderson 1980 Hall 1988
Swaddiwudhipong 1998) two non-randomised trials (Adams
1985 Leung 1987) 14 cohort studies (Raymond 1979 Osga
1980 Satten 1980 Jonah 1982 Mortimer 1984 Cooper 1988
Mortimer 1988 McDavid 1989 Waller 1992 Kloeden 1994
Davis 1997 Billheimer 1998 Perrino 2002 Savolainen 2007)
and four case-control studies (Kraus 1975 Hurt 1981 Haworth
1997 MAIDS 2004)
Over half of the studies were conducted 20 or more years
ago and the oldest was conducted 31 years ago Only three
studies were from this decade (Perrino 2002 MAIDS 2004
Savolainen 2007) Over 60 of the studies were conducted in
the USA (Kraus 1975 Anderson 1980 Osga 1980 Satten 1980
Hurt 1981 Mortimer 1984 Leung 1987 Hall 1988 Mortimer
1988 Waller 1992 Davis 1997 Billheimer 1998 Perrino 2002
Savolainen2007) three inAustralia (Adams 1985 Haworth 1997
Kloeden 1994) three in Canada (Jonah 1982 Cooper 1988
McDavid 1989) onefromthe United Kingdom (Raymond 1979)one from Europe (with sampling areas located in France Ger-
many Netherlands Spain and Italy) (MAIDS 2004) and one
in Thailand (Swaddiwudhipong 1998) Only seven studies were
published in peer reviewed journals (Kraus 1975 Jonah 1982
Mortimer 1984 Cooper 1988 Mortimer 1988 McDavid 1989
Swaddiwudhipong 1998) The other studies were reports forhigh-
way safety and road traffic organisations
The studies by Anderson 1980 Hall 1988 Kloeden 1994
Billheimer 1998 (the basic California Motorcyclist Safety Program
component of the study) and Perrino 2002 examined manda-
tory pre-licence training Cooper 1988 McDavid 1989 and
Swaddiwudhipong 1998 assessed non-mandatory training that
when passed resulted in a licence Raymond 1979 Osga 1980
Satten 1980 Jonah 1982 Mortimer 1984 Adams 1985 Leung
1987 Mortimer 1988 Waller 1992 Davis 1997 Billheimer 1998
(the Experienced Rider Course component of the study) andSavolainen 2007 assessed non-mandatory training that could be
taken by motorcycle riders with and without a licence The case-
control studies (Kraus 1975 Hurt 1981 Haworth 1997) assessed
lsquoanyrsquo type of rider training and therefore could not be grouped by
pre- or post-licence rider training The MAIDS 2004 case-control
study did make a distinction between participants having received
compulsory pre-licence training or additional (non-compulsory)
training We did not find a study that compared one form of pre-
or post-licence training (eg on-road off-road theory or a com-
bination) versus another form of training All studies compared
eithermandatory or non-mandatory formal training versus no for-
mal training with the exception of Haworth 1997 who compared
different levels of training courses (beginner intermediate and ad-vanced)
We have presented a more detailed description of the individual
studies in Characteristics of included studies We have provided
a description below of the types of motorcycle rider training that
were evaluated in the included studies
Types of motorcycle rider training evaluated in the
studies
The types of training were varied Duration of training ranged
from two or three hours to 24 hours administered over two or
three days Most training consisted of a theoretical and a practical
component usually off-road The content ranged from provid-ing riders with manuals lectures discussions showing films and
administering knowledge and skill tests Little information was
provided about the qualifications of trainers who conducted the
training
Six studies (Osga 1980 Satten 1980 Mortimer 1984 Mortimer
1988 Waller 1992 Savolainen 2007) assessed the effectiveness of
the USA-based Motorcycle Safety Foundationrsquos Motorcycle Rider
Course (MSF MRC) This course was released in 1976 It com-
prises eight hours of theory (lectures discussions films and multi-
plechoice tests) a textbook and 12 hours of off-road practice The
Motorcycle Safety Foundation is a national not-for-profit organ-
11Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 1450
isation sponsored by manufacturers and distributors of motorcy-
cles Many USA states have adopted the MSF Basic Rider Courseand Experienced Rider Course (wwwmsf-usaorg) since its release
in 1976
The Motorcycle Operators Skill Test was also developed
by the USA Motorcycle Safety Foundation in coopera-
tion with others (httpwwwwebbikeworldcommotorcycle-
trainingmotorcycle-test) The MOST is a licensing test consist-
ing of a manual knowledge and skill test (comprising eight exer-
cises of increasing difficulty and skill requirements) and a mini-
skill test conducted with a simulator Some studies compared this
alternative licensing test to the standard licensing test However
as the MOST is not a type of rider training we have included only
the two studies that assessed the MOST in combination with one
or more training components in this review Both Anderson 1980and Hall 1988 assessed the MOST in combination with a train-
ing component of three hoursrsquo duration consisting of instructions
about the skills needed to pass the MOST (30 minutes of the-
ory and two and a half hours on the motorcycle) Hall 1988 also
assessed the MOST in combination with the Motorcycle Safety
Foundationrsquos Motorcycle Rider Course as described above
Other USA studies evaluated the Wisconsin motorcycle rider
course a 22-hour off-street programme that teachesthe basic rules
of safe motorcycling (Leung 1987) the USA Maryland Motorcy-
cle Rider Course (MRC) a mandatory (for those younger than 18
years)pre-licence 20-hour coursecomprising eight hours of theory
and12 hours of practice and created by the Maryland Motor Vehi-
cle Administration in 1985 (Perrino 2002) the California Motor-cyclist Safety Program (CMSP) a mandatory (for those younger
than 18 years) pre-licence 16-hour basic course (Billheimer 1998)
and Experienced Rider Course (ERC) of eight hoursrsquo duration
(Billheimer 1998) The CMSP still offers these courses ( http
wwwca-msporg) but they are now MSF courses as described
above Davis 1997 did not give any information about the assessed
rider training
The Canadian studies evaluated the British Colombia Safety
Councilrsquos 37-hour motorcycle safety programme (no further in-
formation on the content was provided) (McDavid 1989) and
the Canada Safety Council Motorcycle Training Program (MTP)
comprising four hours of theory and 16 hours of practice (six
hours on-road) over twoweekends (Jonah 1982) The MTPcoursewas made available in 1974 and is still being offered in Canada
(httpwwwsafety-councilorgtrainingmtpmtphtm) Cooper
1988 did not give any information about the assessed rider train-
ing However the study was conducted in British Colombia in
Canada in 1988 the same time as the McDavid 1989 study was
conducted
In the United Kingdom Raymond 1979 assessed the non-manda-
tory RACACU (Royal Automobile ClubAuto-Cycle Union)
training scheme The training programme consisted of either 24
one-hour sessions or 13 two-hour sessions Twelve hours of the
training focused on theory about basic operational procedures
traffic laws and highway codes The practical session focused on
basic operational skills riding on a training area and riding in traf-fic with an instructor A proficiency test was administered at the
end of the course
One study from rural Thailand (Swaddiwudhipong 1998) evalu-
ated a community-based rider education programme consisting of
theory on motorcycle injury risk helmet use traffic laws vehicle
regulations and traffic signs and a written and skill test leading to
a licence
TheAustralian studies evaluateda rider trainingcourse of 12 hours
over two consecutive days (four hours of theory and eight hours of
practice)administeredto AustraliaPosttelegram deliverers(Adams
1985) and a mandatory pre-licence Ridersafe course consisting
of two tiers the first tier being a training course on the basics
of motorcycle riding where a pass entitled the rider to obtain alearnerrsquos permit and the second tier being a skill test where a pass
entitled the riderto obtain a probationary motorcycle licence The
training course consisted of two four-hour training sessions on
separate days involving instruction videos and actual riding on
motorcycles The skill test consisted of a four-hour preparation
course of 74 minutes of lecturediscussion 38 minutes of video
and 75 minutes of on-course riding which was the testing phase
(Kloeden 1994) The Rider Safe course is still being provided by
Transport South Australia a state government organisation (http
wwwtransportsagovau )
The duration or content of rider training in the four case-control
studies (Kraus 1975 Hurt 1981 Haworth 1997 MAIDS 2004)
was not described
Risk of bias in included studies
We describe the methodological quality of the included studies
by study type below More detailed information can be found in
Table 1 Table 2 Table 3 and Table 4
When considering the findings from these studies it is important
to note the quality (or lack thereof) of reporting of the studies
Quality of reporting is the extent to which a report provides in-
formation about the design conduct and analysis of a study ( Juumlni2001) Huwiler-Muumlntener 2002 found that even though trials
may be well reported it does not mean that the methodological
quality is of equal standard and well-conducted trials may never-
theless be reported badly The reporting quality of most studies in
this review was poor Most studies were old and did not meet con-
temporary reporting standards This made data extraction com-
plicated It also made it difficult to contact authors for additional
information
Because of significant heterogeneity between studies as well as the
poor quality of most studiespooling of studyresults to calculatean
overall measure of effect was not appropriate Instead we provided
a description of our judgement of risk of bias in the text
12Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 1550
Risk of bias in the randomised trials
There were three randomised trials (Anderson 1980 Hall 1988Swaddiwudhipong 1998) Anderson 1980 and Hall 1988 had an
initial sample size of 41188 and 26464 motorcycle licence ap-
plicants respectively Both studies assessed the effectiveness of the
pre-licence MOST with a trainingcomponent Swaddiwudhipong
1998 evaluated the effectiveness of a community-based rider ed-
ucation programme in six districts
Allocation bias
Anderson 1980 reported using random birth date tables to allocate
participants to either the intervention or control group Hall 1988
reported the use of a computerised algorithm set in a central of-
fice but also used manual random assignment when the computer
went down Swaddiwudhipong 1998 reported lsquosimple random
samplingrsquo for the delivery of the intervention where the sampling
frame included nine districts Three districts were excluded since
they were too hard to access The remaining six districts were ran-
domly allocated to the intervention (three) and the control (three)
groups Systematic sampling wasthen used ordered by village size
for the villages within districts Allocation concealment was not
reported in any of the studies The descriptions of randomisation
in these studies suggest that there may have been limitations to
the process resulting in a lack of equivalence of the intervention
and control groups at baseline and therefore the possibility of
allocation bias Swaddiwudhipong 1998 reported that before theintervention the annual incidence rate of motorcycle-related in-
juries was slightly (but not statistically significant) higher in the
intervention areas than in the control areas Participants in the
intervention districts were also more likely to have a valid licence
and to wear a helmet more often than participants in the control
districts Anderson 1980 measured differences at baseline on the
participant and prior driving record variables and found these to
be negligible Hall 1988 did not report on baseline differences
between the intervention and control groups As the Anderson
1980 and Hall 1988 studies had sample sizes of over 40000 and
26000 respectively and used some form of randomisation it is
likely that any confounding factors at baseline would have been
equally distributed between comparison and intervention groups
Selection bias
After randomisation Anderson 1980 and Hall 1988 excluded
08 and 3 participants respectively because of assignment to
the wrong group administration of the wrong intervention or
other reasons The resulting lack of analysis by rsquointention-to-treatrsquo
can introduce biasby removing the effectivenessof randomisation
Nonetheless the percentage excluded was small and the number
of overall participants large so it is unlikely to have had a major
effect on the final results In the Swaddiwudhipong 1998 study
villages were systematically sampled into intervention or control
groups The intervention was implemented and outcomes of in-
jury and death were measured at a community level Interventionand control districts were randomly selected however there were
only three intervention and three control communities
Detection bias
None of the randomised trials reported blinding of outcome as-
sessors Data collection tools used by Anderson 1980 and Hall
1988 were driver records holding police recorded crashes and of-
fences Swaddiwudhipong 1998 used police and hospital records
at a district level and post-mortem examination for deaths None
reported on validity or reliability for any of the instruments
Attrition bias
Anderson 1980 and Hall 1988 had driver records available for
96 to 100 of participants They also collected self-reported
data but the response rate to the questionnaires was very low (low-
est response rate 42 and 30 respectively) Attrition bias in
Swaddiwudhipong 1998 was not an issue as outcome data were
available for all sub-districts included in the trial
The intervention reported in Anderson 1980 and Hall 1988 tri-
als had an unanticipated effect on the participants In both trials
an average of only 28 of licence applicants in the intervention
group eventually obtained their licence compared to 35 in the
control group This indicates that the licensing process in generalseems to have deterred applicants from obtaining a licence and
although no statistical tests of the difference were conducted this
deterrence appears even greater for those who had rider training
This is not true attrition bias as the studies did not experience
differential loss to follow up from outcome measurement Instead
differential attainment of the intermediary step of licensure pre-
cluded participants from being at risk for the outcomes of interest
they are not driving a motorcycle hence they are not at risk for
motorcycle-related crashes injury death or offences In short it is
not possible to assess whether a reduction in crashes in the inter-
vention group would be due to changes in riding habits as a result
of training or simply due to delayed or never attained licensure
Intervention integrity
Neither Anderson 1980 nor Hall 1988 reported on the consis-
tency in the way the intervention was administered the possible
lsquocontaminationrsquo of participants in the intervention group by other
types of formal or informal rider training or of participants in the
control group by the intervention Swaddiwudhipong 1998 did
not report on the number of villagers in each group that actually
attended the intervention However they did report on a number
of other potential training courses that were being implemented
at the same time
13Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 1650
Analysis
Anderson 1980 and Hall 1988 acknowledging the lsquodeterrent ef-fectrsquo described above conducted two separate analyses The first
analysis included the total study population of licence applicants
over two years from the time of their application (the population
denominator) The second analysis was performed on the licensed
(intervention and control) subgroups only two years from the
time of obtaining their licence (the licensed rider denominator)
As only a third were licensed and exposed to riding the motorcy-
cle it was impossible to determine whether reported differences
in crash rates across groups were due to the intervention or to
other lsquoconfoundingrsquo characteristics of those who were determined
in their efforts to obtain the licence When analysis was restricted
only to those who gained their licence the same difficulties re-
mained The benefits of the randomisation process were thus lost
due to this lack of licensure of riders during the process As there
were only three intervention and three control groups in the study
reported by Swaddiwudhipong 1998 analyses were hindered by
smallnumbers for outcome comparisons betweenthe intervention
and the control districts
Risk of bias in the cohort studies and non-randomised
trials
We discuss the methodological quality of cohort studies (seven
prospective studies (Raymond 1979 McDavid 1989 Waller 1992
Kloeden 1994 Billheimer 1998 Perrino 2002 Savolainen 2007)and seven retrospective studies (Osga 1980 Satten 1980 Jonah
1982 Mortimer 1984 Cooper 1988 Mortimer 1988 Davis
1997)) and non-randomised trials (Adams 1985 Leung 1987) in
the one section as the biases for the two study types are similar
The quality of these studies was in general poor
Selection bias
The confounding factors expected to have been addressed in the
studies were sex age riding experience prior driving records (of-
fences crashes) previous training and the following proxies for
risk taking behaviour helmet use speed and alcoholdrug use All
of these factors have been shown to be associated with an increasedrisk of motorcycle crash (Peden 2004 Zambon 2006) The pos-
sible source of bias in cohort studies and non-randomised trials
is that groups may differ on important characteristics at baseline
participants lsquoself-selectrsquo into either the rider training group by de-
ciding to undertake a rider training course or into the lsquocontrol
grouprsquo by deciding not to undertake a course Participants who
undertake a non-mandatory rider training course may differ from
those who choose not to on characteristics such as age sex rid-
ing experience and riding behaviour as well as other possible con-
founders For example riders who take a training course may be
more safety-conscious or on the contrary may do so because they
have previously experienced a crash or because they perceive that
they have poor riding skills Therefore ideally all participants in
the intervention group need to be included in the study as wellas a random sample of controls taken from the same population
as the intervention group If that is not possible intervention and
control groups either need to be carefully matched to make them
comparable or differences at baseline need to be adjusted for in
the analysis
Most of the studies using a non-randomised trial or cohort design
compared an intervention group to a control group that differed
on a number of important factors The studies in general failed to
adjust for important differences found at baseline and their results
cannot be meaningfully interpreted Ten studies (Raymond 1979
Osga 1980 Satten 1980 Jonah 1982 Mortimer 1984 Mortimer
1988 McDavid 1989 Waller 1992 Davis 1997 Savolainen 2007)
assessed intervention and control groups that were not drawnfrom the same population In most of these cases the intervention
group was selected from a list of training graduates who were usu-
ally young female and less experienced riders whereas controls
were recruited from a general population of motorcycle riders who
were usually older male and experienced riders For three stud-
ies it was unclear how they were selected (Leung 1987 Cooper
1988 Perrino 2002) Only Kloeden 1994 recruited groups from
the same population of motorcycle applicants and Adams 1985
from a group of professional riders Nevertheless due to non-ran-
dom selection of intervention and control groups differences at
baseline are likely even though study groups were from the same
population Three studies adjusted for some confounding fac-
tors found at baseline through matching (McDavid 1989 Waller1992 Billheimer 1998) However matching has limitations in
termsof controlling for multiple confounders especially those that
are unknown to influence the effect of rider training on the out-
comes One study matched the control and intervention groups
by excluding controls (approximately 3500 (66 of initial con-
trol group) that differed from the intervention group) (Kloeden
1994) Five studies (Osga 1980 Satten 1980 Adams 1985 Davis
1997 Perrino 2002) did not adjust for differences between study
groups at baseline either in the study design or analysis Results
from these studies are therefore not reliable The other six stud-
ies (Raymond 1979 Jonah 1982 Mortimer 1984 Cooper 1988
Mortimer 1988 Leung 1987 McDavid 1989 Billheimer 1998)
adjusted for some differences between groups at baseline by eitherbi-variate or multi-variate analyses or stratification and standardi-
sation of riding exposure (outcomemiles driven) or stratification
by age group The limitation of stratification is that it is difficult to
control for many variables simultaneously because a large number
of strata will need to be generated resulting in small numbers that
cannot be meaningfully analysed Savolainen 2007 developed sta-
tistical models to assess the probability of a crash after taking the
rider training course in an intervention and control group while
examining differences by sub-groups (eg age) Participation rates
were also very low in this study with only 5584000 (1395)
in the eligible intervention group and 5884000 (147) for the
14Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 1750
control group completing surveys
Detection bias
None of the studies reported blinding of outcome assessors to the
intervention or control status of the participants Seven studies
used driver records containing police recorded data as their main
measure of outcome (Leung 1987 Cooper 1988 McDavid 1989
Waller 1992 Kloeden 1994 Davis 1997 Billheimer 1998) An-
other four studies (Raymond 1979 Satten 1980 Mortimer 1984
Mortimer 1988) only used self-report Raymond 1979 was the
only study using self-report as the main outcome measurement in
which study instruments were validated Self-report might prevent
the underreporting that has been described for the use of police-
driver records (see rsquoRisk of biasrsquo in the randomised trials above)but has other risks of bias especially if questionnaires or inter-
view methods have not been validated and tested for reliability
Recall bias arises from mistakes in recollecting events both be-
cause of failures of memory and because of looking at things lsquowith
hindsightrsquo and possibly changed views (httpwwwcochraneorg
resourcesglossaryhtm) As many studies were retrospective risk
of recall bias is likely and may have increased with length of follow
up For studies that used self-reported outcomes only it is also
possible that social desirability response bias played a role where
participants in the training group may have felt it was lsquoexpectedrsquo
to report good driving histories Multiple methods to measure the
same outcome for both groups decreases the risk of detection bias
for example because self-reported answers can be cross checkedwith other sources of information such as driver records
In this regard three studies (Osga 1980 Jonah 1982 Perrino
2002) used both self-report and driver records Osga 1980 found
that for both intervention and control groups more crashes were
reported through self-report than on state crash records an aver-
age of 54 of self-reported crashes could be found back on state
crash records One industry-based study (Adams 1985) used oc-
cupational health records which were not validated but which are
unlikely to suffer from underreporting due to work cover claims
Savolainen 2007 used information on rider crashes for a five-year
period but did not specify from where this information was ob-
tained
Attrition bias
The majority of studies (Raymond 1979 Satten 1980 Adams
1985 Leung 1987 Kloeden 1994 Davis 1997 Billheimer 1998
Perrino 2002) inadequately reported on losses of participants to
the study We can therefore not be certain that there were no sys-
tematic differences in the characteristics between the intervention
and control group participants Savolainen 2007 reported the re-
sponse rates to their baseline survey but did not report on the per-
centage of outcome data that was available for their study sample
A similar deterrent effect of rider training as described for the ran-
domised trials (Anderson 1980 Hall 1988) occurred in two stud-
ies Jonah 1982 reported that 65(14992310) of the trained rid-
ers stopped riding compared to 47 (9492029) of the untrainedriders They excluded non-active and non-licensed riders before
the study commenced so the differential choice to continue riding
after training did not affect their analyses Leung 1987 reported
that the percentage of participants in the intervention group ob-
taining their licence was 50 (no number provided by authors)
lower than in the control group This may have biased the mea-
surement of outcomes because unlicensed participants will not be
exposed to riding the motorcycle and therefore not be at risk of
being involved in a crash or violation As explained above this
makes it difficult to assess whether the reduction in crashes in the
intervention group is due to a change in driving habits or simply
due to delayed licensure
Mortimer 1984 and Mortimer 1988 describe that about 44 and30 of course graduates respectively were excluded from analysis
because they had not ridden a motorcycle in the previous year
However because these studies were retrospective and the controls
were activeriders at the time of recruitment a decision on whether
the deterrent effect was differential between the two groups could
not be made Though not reported quantitatively Kloeden 1994
also describes the effectof rider training deterringindividuals from
becoming licensed and hence exposure to riding and risk of crash
death or offences involving a motorcycle Osga 1980 reported
that 27 of the trained riders did not ride after the course and
excluded these from the analysis
Use of different data collection strategies may have contributed to
differential loss to follow up in the cohort studies Satten 1980for example interviewed the intervention group over the tele-
phone but mailed questionnaires to the control group Mortimer
1984 and Mortimer 1988 distributed questionnaires in person to
the control group but mailed them to the intervention group
This may have played a role in the observed differential response
rates where more subjects responded in the personally approached
groups
Survival bias
For those studies that assessed offences crashes and injuries but
not death survival bias may have also played a role Motorcyclists
who were fatally or seriously injured in crashes during the follow-
up period after the intervention would have been less likely or
unable to respond to surveys This could bias the effect measures
towards or away from the null None of the studies wrote about
this issue or examined the characteristics of those lost to follow up
in order to address the possibility of survival bias
Intervention integrity
Although the interventions were intended for the entire lsquoexposedrsquo
groups most studies reported difficulty in tracking participants
whoreceivedthe interventionand those whodid not Adams 1985
15Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 1850
estimatedthat 62of thetotalstudysample received the interven-
tion The remaining studies did not report the percentage of thestudy population that received the intervention as was intended
or reported that this percentage was unknown (Raymond 1979
Kloeden 1994 Satten 1980 Jonah 1982 Mortimer 1984 Leung
1987 Cooper 1988 Mortimer 1988 Billheimer 1998 Perrino
2002 Savolainen 2007) Only Raymond 1979 reported the degree
to which their intervention was delivered as intended No studies
reported on possible lsquocontaminationrsquo of participants by other inter-
ventions than the one to which they were allocated Because Osga
1980 Satten 1980 Jonah 1982 Mortimer 1984 Cooper 1988
Mortimer 1988 Waller 1992 and Davis 1997 were retrospective
studies and participants were recruited from a list of graduates it
was known that they had completed and passed the course
Analysis
In some studies some participants who reported at baseline that
they were untrained decided to enroll in a rider training pro-
gramme after the baseline survey The non-randomised trial by
Adams 1985 experienced this move of subjects from the untrained
to the trained group but failed to conduct analyses by lsquointention-
to-treatrsquo This can bias the results in randomised and non-ran-
domised trials if participants who changed groups differ signifi-
cantly with regard to potentially confounding characteristics On
the other hand Perrino 2002 also experienced this move of sub-
jects from theuntrained to the trained groupmore than 100riders(5 of the total sample) originally stated that they were untrained
at baseline but then received training between the baseline and
follow-up surveys Yet since the Perrino 2002 study was a cohort
their analyses then proceeded by including those riders who were
lsquonewlyrsquo trained as exposed Riders who were lsquonewlyrsquo trained since
baseline but with a programme other than the one under study
were excluded from the analysis Nevertheless confounding fac-
tors were not accounted for in the analyses
McDavid 1989 matched exposed with unexposed subjects but
could not find suitable matches for the majority (60) of their
subjects They could therefore only conduct analysis on the result-
ing40 (139346) which is an inappropriate andinefficient anal-
ysis Some studies that reported on the issue of non-active licensedriders(Satten 1980 Jonah 1982 Mortimer 1984 Mortimer 1988
Kloeden 1994) excluded non-active riders from analysis intro-
ducing selection bias as it is likely that riders who continued to
ride differed with regards to potentially confounding character-
istics across intervention and control groups Kloeden 1994 for
example conducted several analyses to assess the effect of various
sub-sets of the control group Original analyses included all con-
trols and led to erroneous results since only a small proportion of
controls hadgone on to get a licence after their learnerrsquos permitand
were actually exposed to the possibility of experiencing a crash
Subsequent analyses compared only intervention and control sub-
jects who had a licence Waller 1992 reported crashes by person-
days of exposure but did not explain how they could be sure that
each study intervention and control participant was an active riderand actually exposed to the risk of experiencing a crash
Potential confounders that are unequally distributed by study arm
at baseline need to be dealt with in the analysis Satten 1980
Adams 1985 and Davis 1997 did not take into account any con-
founding factors either by design or in the analysis McDavid
1989 Waller 1992 and Billheimer 1998 accounted for potential
confounding factors at the design stage by matching the subjects
in the intervention and control groups Other studies (Raymond
1979 Mortimer 1984 Mortimer 1988 Perrino 2002 and also
Waller 1992) adjusted for one or two confounders by stratifica-
tionor standardisation (eg crashesmiles driven age group) Osga
1980 conducteda univariate analysis of crash rates in which no po-
tential confounders were accounted for but also calculated cumu-lative crash rates by mileage travelled Only Jonah 1982 Cooper
1988 and Savolainen 2007 conducted a multivariate analysis
Risk of bias in the case-control studies
A case-control design is suitable for the assessment of risk fac-
tors for outcomes such as motorcycle crashes injuries or deaths
but not appropriate for the evaluation of an intervention such
as motorcycle rider training It can potentially result in spurious
conclusions regarding the efficacy of the intervention because the
intervention has not been applied at random to an intervention
and control group Even with the control of possible confounders
using a case-control design to evaluate an intervention can lead tothese kinds of results since those who choose to undertake rider
training may differ on important prognostic factors that affect the
outcome compared to those who do not choose to participate in
rider training
Selection bias
A potential source of bias in case-control studies as with all other
study types where the intervention is not randomly allocated to
participants is that comparison groups are likely to differ on po-
tentially confounding characteristics This means that ideally all
cases need to be included in the study as well as a random sample
of controls taken from the population that gave rise to the cases If that is not possible cases and controls either need to be carefully
matched to make themcomparable or differences at baseline need
to be adjusted for in the analysis All four case-control studies were
population based Kraus 1975 included all fatalities and injured
cases leading to hospitalisation and appropriately recruited a ran-
dom sample of controls from the same population Hurt 1981 did
not manage to recruit all or a random sample of cases Only half
of the notifications of crashes to the police were received in time
by the researchers the research team for unreported reasons was
only able to respond to half of those again Data collection was
therefore completed for only about a fifth of all cases Neverthe-
less this was still a large sample of 900 cases Haworth 1997 using
16Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 1950
a similar study design was more likely to have included all cases
but this is not entirely clear from the study reported It is unclearfor example whether fatalities were included as they did not use
coronerrsquos reports for the recruitment of cases The MAIDS 2004
study was conducted in five European countries but it was unclear
how cases were recruited whether they were all cases during the
two years that the study lasted or a random sample of all cases
what the response rates were etc Controls were recruited from
randomly selected petrol stations but it was likewise not clear what
the eligibility criteria for inclusion in the study were how they
were recruited (eg randomly) and what the response rate was
Hurt 1981 and Haworth 1997 both matched controls to cases by
recruiting themat the sametime ofdayandsamedayof weekas the
crash occurred Hurt 1981 however recruited controls two years
later andit is likelythat there would have beendifferences betweenthe two comparison groups One important difference was that
during data collection for cases there was very little specialised
motorcycletrainingavailable in the studyregionwhereas two years
later when control data were collected there was some training
available
These four case-control studies also suffered from selection bias
due to potential differences between those that gave consent to
participate in the study and responded to interviews andquestion-
naires and those that did not None of the studies have assessed
such differences although Hurt 1981 did acknowledge that con-
trols giving consent to cooperate may have differed from those
motorcyclists who did not
Detection bias
Measures that can be taken to reduce detection bias in case-con-
trol studies are blinding of interviewers for case or control status
blinding of the participant for the study factor of interest validat-
ing measurement instruments and using these in an identical way
for cases and controls and cross checking self-reported informa-
tion with other more objective sources such as registers None of
the studies reported blinding of the intervention assessors to case
or control status but all used consistent measurement methods for
both case and control groups
All case-control studies used self-report as their main measure toassess whether cases and controls had received rider training and
if so the type of training The major potential source of bias in
case-control studies when self-report is used as a measure is recall
bias One way of reducing recall bias is by double-checking self-
reported data with more objective data Kraus 1975 also used po-
lice accident reports for information on rider characteristics but
such information was not available about rider training Haworth
1997 and MAIDS 2004 did not report on any measures to limit
recall bias Hurt 1981 reported further questioning of cases by
the interviewer when answers did not make sense or when par-
ticipants seemed to contradict themselves as part of the in-depth
unstructured interview technique This qualitative research tech-
nique can result in valuable information but as interviewers were
not blinded to case or control status of the participants it may have led to more thorough examination of cases compared to the
controls and thus differential detection bias
Intervention integrity
The level of detail about the rider training provided in the case-
control studies differed significantly but was minimal for each
Kraus 1975 simply asked whether lsquoanyrsquo motorcycle rider training
had lsquoeverrsquo been carried out Hurt 1981 asked participants whether
they hadbeen self taught informally taught or completed formally
taught rider training courses MAIDS 2004 asked whether partic-
ipants had had pre-license training additional (non-compulsory)
training other training or no training whereas Haworth 1997asked about the number of courses that a participant had com-
pleted whetherthey were basic intermediate or advanced courses
and how long ago the training had been done None of the case-
control studies reported collecting information about further de-
tails about any training programmes (eg type of training cur-
riculum duration of training)
Analysis
Kraus 1975 accounted for confounding factors through the exclu-
sion of females and stratification by age and frequency of motor-
cycle use Hurt 1981 matched cases and controls for time of day
and day of the week that the crash occurred and did not adjust forany confounding factors in the analysis Haworth 1997 matched
groups for time of day and day of the week that the crash occurred
and stratified by age blood alcohol level and licence status in the
analysis None of the studies used multivariate methods MAIDS
2004 did not provide any information about any participant char-
acteristics other than the study exposures of interest and did not
describe any methods to adjust for potential confounding factors
Effects of interventions
We have presented the results in a descriptive manner due to the
substantial heterogeneity in study types interventions and out-comes and the generally poor quality of the studies We have
grouped the interventions by mandatory and non-mandatory pre-
licence rider training and post-licence rider training and further
by study type and by outcome (crashes injurious and fatal crashes
and offences) We discuss results that are reported as statistically
significant (whether a P value was reported or not)
Pre-licence rider training - mandatory
Randomised trials
17Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2050
Crashes
Total group of participants
Anderson 1980 found a reduction in motorcycle crash rates in
the intervention group (MOST skill test with remedial training)
compared to the control group (who received the skill testwithout
a training component) The percent differenceranged from 143
six months after the programme (P lt 005) to 211 (P lt 0001)
after one year and 127 (P lt 001) after two years Hall 1988
found no statistically significant difference in crash rates between
intervention groups at any point in time
Subgroup of participants who obtained their licence only
Anderson 1980 found a reduction in motorcycle crash rates in the
intervention arm but this was statistically significant at one year
only (24 to 218 reduction P lt 005) Hall 1988 found no
statistically significant differences
Offences
Total group of participants
The percentage of offences in Anderson 1980 after six months
and one and two years was slightly higher in the intervention
arm compared to the control group but these differences were
not statistically significant Likewise the study by Hall 1988 did
not find a statistically significant difference in offences between
groups
Subgroup of participants who obtained their licence only
Hall 1988 found no significant difference in offence rates between
the intervention and control groups Anderson 1980 found thatthe percentage of offences increased in the intervention arms com-
pared to the control group This difference was statistically signif-
icant after two years (51 to 115 increase P lt 005)
Cohort studies
Crashes
Neitherof the cohort studies (Kloeden1994 and Billheimer1998)
found a statistically significant difference in crash rates between
the intervention and control groups
Offences
Billheimer 1998 (for the California Motorcyclist Safety Program
course) and Perrino 2002 did not find a statistically significant
difference in offence rates between the intervention and control
groups
Case-control studies
Crashes
MAIDS 2004 found that a slightly higher percentage of controls
(not involved in a crash) had had pre-license rider training than
cases (involved in a crash) This difference was not statistically significant (50 of controls vs 477 of cases OR 091 95 CI
076 to 110)
Pre-licence rider training - non-mandatory
Randomised trial
Injuries and deaths
Swaddiwudhipong 1998 found a significantly lower injury ratein the intervention villages (1051000 population 95 CI 92
to 120) compared to the control villages (1691000 population
95 CI 152 to 185) one year after the intervention This differ-
ence was not maintained two years after the intervention (127
95 CI 112 to 143 intervention 151 95 CI 135 to 167
control) There was no significant difference in death rates be-
tween groups
Cohort studies
Crashes
Cooper 1988 foundno statistically significant differencein crashes
between the intervention and control groups McDavid 1989
found a lower crash rate in the intervention group (14139) com-
pared to the control group (23139) but no test of statistical sig-
nificance was undertaken
Injuries and deaths
McDavid 1989 found that the intervention group had fewer in-
jurious crashes (814 crashes) compared to the control group (16
23 crashes) but no statistical significance test was reported
18Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2150
Offences
Cooper 1988 foundno statistically significantdifference in offencerates between the intervention and control groups McDavid 1989
foundno statistically significantdifference in offence ratesbetween
the intervention and control groups
Non-mandatory rider training
Non-randomised trials
Crashes
Neither Adams 1985 nor Leung 1987 found a statistically signif-icant effect of rider training on crash rates
Offences
Leung 1987 found a statistically significant lower offence rate in
the intervention group compared to the control group for both
males and females and for all age groups except for women 35
years or older (P values not reported)
Cohort studies
Crashes
Osga 1980 in the univariate analysis reported a lower percentage
of crashesin theintervention group (that continued to ride) (28
168595) than in the control group (34 112329) In the cu-
mulative crash rate analysis however Osga1980 found statistically
significant (at the P lt 001 level) higher crash rates in the interven-
tion group than in the control group at allmileage intervals (1000
to 29999 miles) travelled After riding more than 30000 miles
(when over 50 of the trained riders had been involved in a crash
compared to approximately 40 in untrained riders) this differ-
ence was no longer statistically significant Satten 1980 found a
lower percentage of crashes in the intervention group (13 969)
compared to the control group (25 2079) but the differencewas not assessed statistically Davis 1997 found a lower crash rate
in trained riders but it was unclear how this was calculated be-
cause the size of the control group was not reported Jonah 1982
and Billheimer 1998 (for the Experienced Rider Course) found no
statistically significant difference in crashes between the interven-
tion and control groups Although the difference between those
who completed the Experienced Rider Course and no course in
Billheimer 1998 neared significance at six months after the course
(P = 007) large P values were observed for one year after (P =
030) Raymond 1979 found a statistically significant higher crash
rate for the intervention group (1861000 miles ridden) com-
pared to the control group (1181000 miles ridden) in both males
and females (P lt 0001) especially during the first 5000 miles of
riding Waller 1992 found a statistically significant higher crashrate for the intervention group (27 crashes in three years or 070
crashes10000 person days) compared to the control group (5
crashes in three years or 013 crashes10000 person days) with a
crude relative risk of 538 (X 2=1513 plt005) Mortimer 1984
and Mortimer 1988 found no statistically significant difference in
crashes between the intervention and control groups
Savolainen 2007 found that the intervention group (those who
completed the Basic Rider Course) was 44 more likely to be
involved in crashes than the control group and those who com-
pleted the course multiple times were an additional 180 more
likely to be involved in a crash This finding was based on low
numbers and no P value or confidence level was given for either
result
Injuries and deaths
Satten 1980 also found a lower percentage of injurious crashes
(33 39) in the intervention group compared to the control
group(60 1220) although again no statistical significance test
was reported
Offences
Jonah 1982 did find a significantly lower self-reportedoffence rate
in the intervention group (15) compared to the control group(34 P lt 001) (no numbers given) Satten 1980 also found
a lower self-reported offence rate in the intervention group (1
71 versus 19100) but again this was not tested for statistical
significance Mortimer 1984 Mortimer1988 and Billheimer 1998
found no statistically significant differencein offence ratesbetween
the intervention and control groups
Case-control studies
CrashesMAIDS 2004 found that a slightly higher percentage of cases
(involved in a crash) had had additional non-mandatory rider
training than controls (not involved in a crash) This difference
was not statistically significant (17 of cases vs 12 of controls
OR 146 95 CI 064 to 337)
Rider training unspecified
Case-control studies
19Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2250
Injuries and deaths
Hurt 1981 found that the odds of injurious crashes were lower inthose who had received professional training compared to those
who had not Kraus 1975 found that in the younger age group
(less than 20 years) and the older age group (over 35 years) a larger
percentage of injured riders (12 and 21 respectively) reported
prior motorcycle rider training compared with the uninjured con-
trol group (6 and 5 respectively) (P lt 001) (no numbers
given) Haworth 1997 reported several different outcomes one
beingthat having completed at leastone trainingcoursecompared
to no training courses at all had no significant effect on the odds
of crashing (afteradjustment for age blood alcohol level or licence
status (learner permitprobationaryfull)) When comparing three
different levels of training courses (beginner intermediate and ad-
vanced) with no training no significant differences were found
in the odds of crashing However after adjustment for age and
blood alcohol level completing an advanced course as compared
to a beginners course was associated with a significant decrease in
the odds of severe motorcycle crashes resulting in serious injury or
death (OR 02 95 CI 01 to 08)
D I S C U S S I O N
The objective of this review was to quantify the effectiveness of
pre- andpost-licence motorcycle rider training on the reduction of
traffic crash involvement injuries deaths and offences committedby motorcycle riders Most included studies were cohort studies
There were a few randomised and non-randomised trials and case-
control studies
All studies compared either mandatory or non-mandatory formal
training with no formal training except for Haworth 1997rsquos case-
control study that compared levels of training (beginner interme-
diate and advanced) to each other
We found studies that evaluated the effectiveness of manda-
tory pre-licence training in reducing crash and offence rates
(Anderson 1980 Hall 1988 Kloeden 1994 Billheimer 1998
(the basic California Motorcyclist Safety Program component of
the study) Perrino 2002) non-mandatory training that when
passed resulted in a licence (Cooper 1988 McDavid 1989
Swaddiwudhipong 1998) non-mandatory training that could be
taken by motorcycle riders with or without a licence (Raymond
1979 Osga 1980 Satten 1980 Jonah 1982 Mortimer 1984
Adams 1985 Leung 1987 Mortimer 1988 Waller 1992 Davis
1997 Billheimer 1998 (the Experienced Rider Course compo-
nent of the study)) and unspecified rider training on crash injury
and death rates (Kraus 1975 Hurt 1981 Haworth 1997 MAIDS
2004)
The studies provided conflicting evidence regarding motorcycle
rider training which is likely due to the poor design quality and
reporting of the studies Most of the studies included in the review
were likely to be adversely affected by significant selection biasalong with substantial detection bias(measurement and recall bias)
which limits our ability to make conclusive comments
Mandatory pre-licence rider training
There were two RCTs three cohort studies and one case-control
study that examined pre-licence ridertraining The two RCTs eval-
uated the Motorcycle Skills Operating Test (MOST) (Anderson
1980 Hall 1988) Anderson 1980 found a statistically significant
reduction in crashes in the intervention group at six months one
and twoyearsfollow up However when only participantswho ob-
tainedtheirlicence were included in theanalysis thereduction wasstatistically significant only at one year follow up The other RCT
(Hall 1988) found no statistically significant differences in crash
rates at any point of time neither in the total population group
nor in the licensed subgroup Neither of the two cohort stud-
ies assessing crash rates found a statistically significant difference
in crash rates between intervention and control groups ( Kloeden
1994 Billheimer1998) andthe case-control study (MAIDS 2004)
also did not find a statistically significant difference in the odds of
experiencing a crash between trained and untrained riders With
regards to offences neither of the RCTs found a statistically sig-
nificant difference in offence rate between the intervention and
control groups in the analysis of the total group of participants
entering the trial However Anderson 1980 found a statistically significant increase in offence rates in the intervention group but
only after two years of follow up andonly in the analysis of partici-
pants who obtained their licence The two cohort studies assessing
offence rates found no statistically significant difference between
the groups (Billheimer 1998 Perrino 2002)
None of the five studies (RCTs or cohort) found a reduction in
crashes or offences following mandatory pre-licence rider training
except for Anderson 1980 who observed a statistically significant
reduction in crashes in the intervention group at one year in both
analyses It is not clear why both RCTs that evaluated the same
MOST and training intervention observed different results Both
studies enrolled large samples and observed a similar uptake of
licensure following the intervention It is possible that differencesmay have arisen due to different study samples (California ver-
sus New York) or that the intervention delivery differed between
studies
In both trials of the Motorcycle Skills Operating Test including
rider training only a third of motorcycle licence applicants ob-
tained their motorcycle licence It is possible that a licensing pro-
cess including a skills test may reduce crashes through decreased
exposure
Non-mandatory pre-licence training
20Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2350
Three studies assessed the effect of non-mandatory pre-licence
training The Cooper 1988 cohort study found no difference incrash or offence rates between the trained and untrained groups
We cannot however be confident about this result as the study
sufferedfrom selectionbiaswhichcouldhave biasedthe measure of
effect toward the null The RCT conducted by Swaddiwudhipong
1998 examined injury and death rates and found a statistically
significant lower injury rate in the intervention villages after one
year of follow up but no longer at two year follow up There
was no statistically significant difference in death rates between
intervention and control villages Our confidence in the results of
this study are hindered by the inadequacy of the randomisation
process that led to significant differences between the intervention
and control districts on two important variables The intervention
districts were more likely than the control districts to have a validlicence andto wear a helmetSincethese differences were nottaken
into account in later analyses they could explain the differences
observed in injury rates between districts Furthermore with only
six districts randomised the sample size was probably too small
to detect a difference between intervention and control groups
had there been one The McDavid 1989 cohort study found a
lower percentage of injuries in the intervention group although
a statistical test was not conducted McDavid 1989 drew their
intervention and control groups from different populations but
did adjust for confounders although they only analysed 40 of
the initial sample (139 trained and 139 untrained) Because of
these substantial weaknesses we are unable to draw conclusions
from this studyAn overall effect of pre-licence training on the reductionof crashes
injuries deaths and offences was not found although one should
be aware that the research of this issue to date has had significant
flaws However a mandatory skill test (such as the MOST) in
combination with and even without pre-licence training as part
of a licensing process seems to reduce the number of individuals
completing the licensing procedure and therefore indirectly re-
duces crash rates through a reduction in exposure to motorcycle
riding Making licensure a more difficult process may therefore be
seen as advantageous if it delays licensure and therefore reduces
exposure It is possible that the group who chose not to continue
riding a motorcycle would not have been the group to experience a
crash or commit an offence if they had continued to ride anyhowfor example because they are more safety conscious and this was
the reason for their giving up riding in the first place If this is the
case deterring individuals from completing their licence and rid-
ing may not necessarily decrease overall crashes deaths or offence
rates In addition other potentially adverse effects of complicat-
ing a licensing procedure need to be considered such as increases
in unlicensed riding In Thailand non-mandatory rider training
offered as a community intervention may have decreased injury
rates but this needs to be confirmed through further research
Non-mandatory training
The majority of the studies assessed non-mandatory training
(training independent of any licensing procedure) and the results
were varied The two non-randomised trials assessing crash rates found no statistically significant difference in crash rates between
the intervention and control groups (Adams 1985 Leung 1987)
Of the nine cohort studies Satten 1980 found a lower percentage
of crashesin the intervention group although it is unclear whether
this was a statistically significant result Davis 1997 also found a
lower percentage of crashesin theinterventiongroup although it is
unclear with which control group the intervention group wascom-
pared Raymond 1979 Osga 1980 Waller 1992 and Savolainen
2007 found a statistically significant higher crash rate in the in-
tervention group and the remaining cohort studies (Jonah 1982
Mortimer 1984 Mortimer 1988 Billheimer 1998) found no sta-
tistically significant difference in crash rates between intervention
and control groups No conclusions can be drawn from these stud-
ies with confidence The Satten 1980 and Davis 1997 studies did
not adjust for any confounding factors even though important
differences existed between the intervention and control groups
Although Raymond 1979 and Waller 1992 adjusted for a num-
ber of confounders at baseline (Waller 1992) and in the analy-
sis (Raymond 1979 Waller 1992) the intervention and control
groups were drawn from such different populations that it is pos-
sible that unmeasured confounders were responsible for the sta-
tistically significant increased risk of crashes observed in the inter-
vention group The observed difference could also be attributed totheobserved differential loss to followup (Raymond 1979) which
could bias the measure of effect towards or away from the null
The Osga 1980 study found a protective effect of rider training
in the univariate analysis but statistically significant higher crash
rates in the intervention group when accounting for riding expo-
sure Sex (a significantly higher proportion of females in the in-
tervention group) and other potentially confounding factors that
could explain thedifferencebetween the groups were howeverun-
accounted for Finally Savolainen 2007 did control for potential
confounding variables yet it is not clear whether variables were
controlled for simultaneously or one at a time One case-control
study (MAIDS 2004) found no statistically significant reduction
in the odds of experiencing a crash between trained and untrainedriders but the quality of this study was poor mainly because the
analysis did not account for any confounding factors
One cohort study (Satten 1980) assessed the effect of non-manda-
tory rider training on injurious crashes and found a lower percent-
age of injuries in the intervention group although a statistical test
was not conducted As mentioned above the Satten 1980 studyrsquos
validity is reduced due to its lack of adjustment for confounding
factors
The only non-randomised trial that assessed offence rates (Leung
1987) found a statistically significant lower offence rate in the in-
tervention group and of the five cohort studies that assessed of-
fence rates (Satten 1980 Jonah 1982 Mortimer 1984 Mortimer
21Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2450
1988) one study (Jonah 1982) also found a statistically significant
lower offence rate in the intervention group However we have lit-tle confidence in the results from Leung 1987 since potential con-
founding factors were not appropriately addressed Jonah 1982
did conduct a multivariate analysis but was hindered by an initial
low response rate The other four cohort studies did control for
several potential confounders and found no difference in offence
rates following non-mandatory post-licence rider training
The non-randomised trials and cohort studies evaluating non-
mandatory rider training were at high risk of substantial selection
bias due to the non-random allocation of participants as well
as low participation rates and attrition bias and could therefore
not provide conclusive evidence with regard to the effect of the
training It should be noted that despite the biases and other issues
with this group of studies most didnot detectmajor effects relatedto rider training in either direction This may suggest that any
potential impact of rider training will be modest
Studies examining unspecified modes of ridertraining
Three case-control studies assessed whether injured or deceased
motorcyclists were more or less likely to have had rider training
compared to control motorcyclists All had conflicting results
Hurt1981 foundthe odds of havingan injuriouscrash waslower in
those who had received rider training whereas Kraus 1975 found
that the odds of having an injurious crash were higher in youngerand older trained riders Haworth 1997 who compared advanced
versus beginner training found no effect on motorcycle injuries
or deaths although they did find that those who had completed
advanced training had reduced chances of crashing compared to
those who had done a beginnersrsquo training course Similar issues
with uncontrolled confounding and biases of selection as described
above arose with the three case-control studies Hurt 1981 suffers
from inadequate control of confounding via poor matching and
both the Kraus 1975 and the Haworth 1997 findings were only
controlled for limited factors by means of stratified analysis It is
likely that these limitations are the reason for the contradictory
directions of effect observed for the three case-control studies
Methodological weaknesses of the studies
In summary most studies investigating the effectiveness of mo-
torcycle rider training on reducing crashes injuries and death suf-
fered from serious methodological weaknesses
The majority of studies were non-randomised and were there-
fore limited mainly by selection bias Although the approaches of
matching stratification or regression analyses can minimise bias
(Fleiss 2004) they were poorly employed in the reviewed studies
The inadequate adjustment for confounders in particular adds
to our lack of confidence in the studiesrsquo results since it has been
demonstrated that individuals who choose to take motorcycle rider
training differ from those not taking rider training For instance
trained riders often differ from controls in age sex and experi-ence usually ride less and may have a more positive attitude to
safety as indicated by the reported more frequent wearing of seat
belts (Mortimer 1982)
The two large randomised trials reported that a large proportion
of their study group were never exposed to the risk of offences or
crashes because they never attained licensure This made it impos-
sible to assess whether a reduction in crashes in the intervention
group was a result of training or due to delayed or never attained
licensure
In some studies the sample sizes were too small to detect any sta-
tistically significant differences Follow-up time after training dif-
fered between studies and was an impediment to observing any
real effects of training if they existed As riding experience ac-cumulates over time any differences in outcomes between those
trained and untrained would diminish in studies with lengthy fol-
low-up times This dilution of any possible effect of rider training
may have also occurred in studies that mixed both novice and ex-
perienced riders in the study groups since it is hypothesised that
novice riders serve to benefit most from rider training This would
increase the difficulty in finding any effect of rider training par-
ticularly when sample sizes are small
Detection bias may also account for some of the differences ob-
served between trained and untrained riders Detection bias would
have occurred if the self-report hospital driver or police records
used to record the outcomes of crash injury death or offences
differed between the trained and untrained riders Research hasdemonstrated that not all crashes are reported to police for ex-
ample small crashes or off-road crashes (Kraus 1975) even serious
injury crashes have been found to be underreported (Anderson
1980) However this would only introduce detection bias if the
reporting of outcome differs for trained and untrained riders This
does appear possible in that responsible riders may be more likely
to enroll in a rider training course and also may be more likely
to report crashes to police than irresponsible riders meaning that
the number of crashes from riskier riders could be underrepre-
sented on records An analysis to investigate the effectiveness of
rider training may then show that rider training is not effective in
reducing crashes
Rather than using reported outcomes some studies used self-re-ported survey data although this technique suffers from non-
response biases In fact two studies reported differences in de-
mographic characteristics in response rates to questionnaires
(Anderson 1980 Hall 1988) Self-reported data also introduce
the risk of recall bias survival bias and social desirability response
bias Recall bias may have been equally distributed among trained
and untrained riders howeverit is hypothesised that trained riders
may have felt it was lsquoexpectedrsquo to report good driving histories
hence leading the rider training to appear effective
22Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2550
Limitations of this review
Limitations of this review were that we found very few recentstudies (the two most recent ones published in 2002 and 2007)
and that we found very few studies conducted in low-and middle
income countries even though our search strategy was set up to
identify such studies Anotherlimitation wasthat forthe reviewed
studies we could not assess all outcomes of rider training that are
potentially useful Some of the included studies assessed outcomes
such as use of safety clothing including helmets Such outcomes
are valuable since the use of motorcycle helmets has been demon-
strated to be effective in preventing head injuries (Liu 2004) De-
spite this the use of safety clothing was not included as an out-
come in this review because the studies measuring this outcome
were limited by poor data quality due to very low response rates
Another outcome of interest in assessing the effectiveness of rider
training is lsquotype of injuryrsquo As Haworth 2005 state in their review
of motorcycle licensing and training rider training could be con-
sidered effective if injuries become less severe because motorcycle
riders have learned how to avoid more serious injuries through
rider training The studies included in this review did not specify
type of injury and so this outcome could not be assessed Last
changes in knowledge and attitudes following rider training are
other potentially useful outcomes We have excluded such studies
(Osga 1979 Clayton 1990 Goldenbeld 2004) as these outcomes
are proxies rather than direct measures of the effectiveness of mo-
torcycle rider training We opted to report on the outcomes of
crashes injury death and offences alone since there is little ev-
idence about how changes in knowledge and attitudes relate to
these outcomes
A U T H O R S rsquo C O N C L U S I O N S
Implications for practice
Due to the poor quality of studies included in this review we were
unable to draw any conclusions about the effectiveness of rider
training on crash rates injurious or fatal or offence rates The
findings do suggest that mandatory pre-licence training may be
an impediment to completing a motorcycle licensing process thuspossibly indirectly reducing crash injury death and offence rates
through a reduction in exposure to riding a motorcycle On the
basis of the existing evidence it is not clear if and what type of
training reduces the risk of crashes injuries deaths or offences in
motorcyclists and the selection of the best rider training practice
can therefore not be recommended The suggestion that manda-
tory pre-licence training indirectly reduces exposure to riding the
motorcycle because trainees are impeded from completing the li-
censing process does not help those motorcyclists who persist in
obtaining their licence and riding the motorcycle Some type of
rider training is likely necessary to teach motorcyclists to ride a
motorcycle safely Good policy is based on sound evidence and
evaluation but such evidence for motorcycle rider training is seri-
ously lacking Policy makers are therefore encouraged to include arigorous evaluation component to any new or existing rider train-
ing programme Such an evaluation should ideally be randomized
and assess whether the training programme is effective and if so
which component(s) of the programme work best
Implications for research
Given theresults of this review which found littlereliable evidence
about the effectiveness of motorcycle rider training it is necessary
to build up an evidence base of the effectiveness of motorcycle
rider training in the reduction of crash injury death and offence
rates To this end well designed randomised controlled trials need
to be conducted that examine the impact of rider training oncrashes and offences A randomised controlled trial is the preferred
method as it would significantly reduce most biases described in
this review To minimise the impact of participants delaying or
never attaining their licence and therefore exposure as found in
the randomised trials in this review such a trial should include
riders who are already licensed and reliable measures of riding
exposure in a reasonably lengthy follow up period It is expected
that a randomised controlled trial of sufficient sample size would
observe equal balance on possible confounders by study arm how-
ever an assessment of potentially important confounders (such as
helmet wearing) by study group would enable control for these
variables in the analyses if necessary It is imperative that the effec-
tiveness of each component of the rider training (such as theorypractice knowledge skills and modification of behaviour) and the
synergy of such components are evaluated Outcomes should be
objective ideally those that are routinely collected rather than self-
reported outcomes Even though riding the motorcycle poses dif-
ferent challenges and incurs different risks lessons may be learned
from research into the effects of car driver training
Motorcycles are a common means of transport in low- and middle
income countries and usage is increasing As the burden of mo-
torcycle crash victims is mostly borne in such countries research
should be focused on the effectiveness of types of rider training on
the reduction of crashes injuries and fatalities in such countries
and be supported by higher income countries This review found
only one study conducted in a middle income country that sug-gested decreasing injury rates It would be beneficial to conduct
additional research to confirm this suggestion and to enhance our
knowledge of the role of rider training in low- and middle-income
settings
March 2006 saw the launch of the University of North Carolina
Highway Safety Research Centre partnership with the Motorcycle
Safety Foundation (MSF) and plans were revealed to conduct a
study of the potential benefits of motorcycle rider training The
study is funded by the MSF and the National Highway Traffic Sa-
fety Administration (NHTSA) Called rsquoThe Discovery Projectrsquo it
will compare data on crash rates from riders participating in a basic
23Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2650
motorcycle safety class with riders participating in a basic motor-
cycle safety class as well as a variety of continuing education classesThe collection of pilot data was due to start in March 2007 and
the data collection portion of the study is expected to take three
years (Highway Safety Research Center) It is vital that researchers
and policy makers collaborate to develop furtherhigh quality eval-
uations of motorcycle rider training programmes to help inform
development of new and effective training programmes
A C K N O W L E D G E M E N T S
Staff from The University of Sydney library for helping us locate
and retrieve full reports of many papers
R E F E R E N C E S
References to studies included in this review
Adams 1985 published data only
Adams A Collingwood V Job RFS Evaluation of the 1982 Australia
Post motorcycle rider training scheme NSW Australia Traffic
Accident Research Unit Traffic Authority NSW 1985 [ RN 85]
Anderson 1980 published data only
Anderson J Ford JL Peck RC Improved motorcyclist licensing and testing project Washington DC National Highway Traffic Safety
Administration (NHTSA) USA 19801ndash217 [
DOTndashHSndash805ndash598]
Billheimer 1998 published data only
Billheimer JW Evaluation of the California Motorcyclist Safety
Program Transportation Research Record 19981640100ndash9 [
TRB Report Number 980652]
Cooper 1988 published data only
Cooper PJ Rothe JP Motorcyclists who they are and why they do
what they do Transportation Research Record 1988116878ndash85
Davis 1997 published data only
David CF Evaluation of community traffic safety programs and
motorcycle operator training programs Connecticut Department of Transportation 199763
Hall 1988 published data only
Hall CJ Buchanan L Engle R Motorcycle rider education evaluationproject Washington DC USA National Highway Traffic Safety
Administration (NHTSA) US Department of Transportation
1988 [ DTNH 22ndash80ndashCndash05123]
Haworth 1997 published data only
Haworth N Smith R Brumen I Pronk N In Monash University
Accident Research Centre editor(s) Case-control study of motorcycle
crashes Victoria Australia Department of Transport and Regional
Development The Federal Office of Road Safety 19971ndash149 [
ISBN 0 642 25501 6 ]
Hurt 1981 published data only
Hurt H Ouellet J Thom D Motorcycle accident cause factors and
identification of countermeasures USA Department of
Transportation 1981 [ Technical Report DOT HS 806ndash862]
Jonah 1982 published data onlylowast Jonah BA Dawson N Bragg B Are formally trained motorcyclists
safer Accident Analysis amp Prevention 198214(4)247ndash55
Jonah BA Dawson NE Bragg BW Evaluation of the Canada
Safety Councilrsquos Motorcycle Training Program Proceedings of theInternational Motorcycle Safety Conference Washington DC
USA 198059ndash93
Kloeden 1994 published data only
Kloeden CN Moore VM McLean AJ Evaluation of the pre-license
training program for motorcyclists in South Australia Walkerville
South Australia Australia NHMRC Road Accident Research Unit
The University of Adelaide 19941ndash37 [ ISBN 0 7308 0607 3]
Kraus 1975 published data only
Kraus JF Riggins RS Franti CE Some epidemiological features of
motorcycle collision injuries American Journal of Epidemiology 1975102(1)74ndash98
Leung 1987 published data only
Leung KS Reding VA Evaluation of the Wisconsin motorcycle rider course Wisconsin Bureau of Policy Planning and Analysis
Wisconsin Department of Transportation 1987
MAIDS 2004 published and unpublished data
MAIDS In-depth investigations of accidents involving powered
two wheelers ACEM (Acemacemeu) 2004 Vol Final report
20179
McDavid 1989 published data only
McDavid JC Lohrmann BA Lohrmann G Does motorcycle
training reduce accidents Evidence from a longitudinal quasi-
experimental study Journal of Safety Research 198920(2)61ndash72
Mortimer 1984 published data only
Mortimer R Evaluation of the motorcycle rider course Accident
Analysis amp Prevention 198416(1)63ndash71
24Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2750
Mortimer 1988 published data only
Mortimer RG A further evaluation of the motorcycle rider courseJournal of Safety Research 198819(4)187ndash96
Osga 1980 published data only
Osga GA An investigation of the riding experiences of MSF rider
course participants Dissertation South Dakota USA The
University of South Dakota 19801ndash224
Perrino 2002 published data only
Perrino CS Ahmed A Callender A Rozier E Cantwell A Stewart
O The role of Marylandrsquos motorcycle rider course in promoting saver
behaviors and attitudes Baltimore USA National Transportation
Center Morgan State University 20021ndash52
Raymond 1979 published data only
Raymond S Tatum S An evaluation of the effectiveness of the RAC
ACU motorcycle training scheme - final report Road Safety ResearchUnit Dept of Civil Engineering University of Salford London
19791ndash131
Satten 1980 published data only
Satten RS Analysis and evaluation of the motorcycle rider courses
in thirteen Illinois counties International Motorcycle Safety
Conference Washington DC 1980145ndash93
Savolainen 2007 published and unpublished data
Savolainen Peter Mannering Fred Additional evidence on the
effectiveness of motorcycle training and motorcyclistsrsquo risk-taking
behaviour Transportation Research Board 2007 Vol March 20
19
Swaddiwudhipong 1998 published data only
Swaddiwudhipong W Boonmak C Nguntra P Mahasakpan P
Effect of motorcycle rider education on changes in risk behavioursand motorcycle-related injuries in rural Thailand Tropical Medicine
and International Health 19983(10)767ndash70
Waller 1992 published data only
Waller FL An analysis of the impact of the Texas Department of
Public Safety Motorcycle Operator Training Program upon
motocycle related traffic accidents Texas State University Public
Administration Program Applied Research Projects 199277
References to studies excluded from this review
Anderson 1980a published data only
Anderson LC Better biking a motorcycle rider improvement
program International Motorcycle Safety Conference Washington
DC USA Motorcycle Safety Foundation 1980231ndash53Aupetit 2008 published and unpublished data
Aupetit S Riff J Buttelli O Espie S Ergonomics research program
on motorcycle learning Contributions to the development of a
motorcycle simulator for training Transport Research Arena
Ljubljana 20089
Awane 1999 published data only
Awane T Integrating simulators in motorcycle safety education
Journal of International Association of Traffic and Safety Sciences 199923(1)26ndash35
Baldi 2005 published data only
Baldi S Baer JD Cook AL Identifying best practices states in
motorcycle rider education and licensing Journal of Safety Research200536(1)19ndash32
Batchler 1988 published data only
Batchler MW Evaluation of the pre-license motorcycle training course at the New South Wales traffic education centre Federal Office of
Road Safety (FORS) 19881ndash122 [ Report No CR 77 ]
Blanchard 2006 published data only
Blanchard HT Tabloski PA Motorcycle safety educating riders at
the teachable moment Mosby (CV) Company 200632330ndash2
Bonnett 2005 published data only
Bonnett J Mature age skills training for experienced motorcycle
riders Motorcycle ridersrsquo association of the Australian Capital
Territory 2005
Braver 2007 published data only
Braver ER et alPersuasion and licensure a randomized controlled
intervention trial to increase licensure rates among Maryland
motorcycle owners Traffic Injury Prevention 20078(1)39ndash46Clayton 1980 published data only
Clayton AB Sudlow DE An evaluation of the effectiveness of a one day training programme for learner motorcyclists Berkshire UK
Transport and Road Research Laboratory 19801ndash23 [ ISSN
0266ndash7045]lowast Clayton AB Sudlow DE An evaluation of the effectiveness of a one-day training programme for learner motorcyclists Berkshire UK
Road Safety Division Berkshire 19801ndash23 [ISSN 0266ndash7045]
Council 1978 published data only
Council FM Desper LP Dutt AK Driver education for motorcycle operation University of North Carolina Highway Safety Research
Center 19781ndash40
Crowther 2005 published data onlylowast Crowther G Engaging with motorcyclists UK police and theBikeSafe Road Safety Programme Published proceedings of the
Australasian Road Safety Research Policing Education Conference
Wellington New Zealand 2005 issue ISBN 0ndash473ndash10636ndash11ndash11
Daniello 2009 published data only
Daniello A Gabler HC Mehta YA Effectiveness of motorcycle
training and licensing Transportation Research Board 88th Annual
Meeting 200913
Dewanti 2007 published data only
Dewanti M Challenging efforts in promoting young motorcyclist
safety in Indonesia Road Safety on Four Continents 14th
International Conference Swedish National Road and Transport
Research Institute 2007
Du Plooy 1990 published data onlyDu Plooy DW Motorcycle training in South Africa The Human
Element International Motorcycle Safety Conference Orlando
Florida USA Motorcycle Safety Foundation 19904-1 to 4-9
Engel 1989 published data only
Engel U Krogsgaringrd Thomsen L [Faeligrdselsundervisning og
propaganda ndash en faeligrdselssikkerhedsfremmende foranstaltning]
Dansk Vejtidsskrift 198912295ndash7
Goldenbeld 2002 published data only
Goldenbeld C Houwing S de Craen S The development of
driving skills of young moped riders [De ontwikkeling van de
rijvaardigheid van jonge bromfietsers] SWOV Institute for Road
Safety Research (Stichting Wetenschappelijk Onderzoek
Verkeersveiligheid) Leidschendam the Netherlands 200263
25Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2850
Goldenbeld 2004 published data only
Goldenbeld C Twisk D de Craen S Short and long term effects of moped training a field experiment Transportation Research Part F 20047(1)1ndash16
Haworth 2000 published data only
Haworth N Smith R Kowadlo N Motorcycle training in Victoria
evaluation of rider training curriculum in Victoria Monash
University Accident Research Centre Reports 2000 Vol 165 issue
3131
Haworth 2005 published data only
Haworth N Mulvihill C Review of motorcycle licensing and
training Monash University Accident Research Centre 2005 [
Report no240]
Haworth 2006 published data only
Haworth N Mulvihill C Rowdon P Teaching old dogs new tricksTraining and older motorcyclists Journal of the Australasian College of Road Safety 200718(4)20ndash5lowast Haworth NL Mulvihill C Rowden PJ Teaching old dogs new
tricks Training and older motorcyclists Australasian Road Safety
Research Policing and Education Conference Gold Coast
Queensland 2006
Hill 1983 published data only
Hill D Motorcycle rider training testing and licensing Motor
Cycle Council of New South Wales Sydney 1983
Imatake 1980 published data only
Imatake Y On the activities in Japan of motorcycle safe riding
promotion committee (MSRPC) - especially the education of
unskilled riders of class 1 motorbikes International Motorcycle
Safety Conference Washington DC USA Motorcycle Safety
Foundation 1980255ndash68
Ingebritsen 1990 published and unpublished data
Ingebrigtsen S [Risikofaktorer ved ferdsel med moped og
motorsykkel] TOslashI-rapport 66 Oslo Transportoslashkonomisk
institutt 1990
Jain 2009 published and unpublished data
Jain A et alTwo wheeler accidents on Indian roads a study from
Mangalore India Elsevier 200916130ndash3
Jonah 1979 published data only
Jonah BA Dawson NE Validation of the motorcycle operator skill
test Accident Analysis amp Prevention 197911(3)163ndash71
Jonah 1981 published data onlyJonah BA Dawson NE Bragg WE Predicting accident
involvement with the Motorcycle Operator Skill Test Accident Analysis amp Prevention 198113(4)307ndash418
Kadar 1990 published data only
Kadar KJ Countermeasures motorcycle rider education delivery
system The Human Element International Motorcycle Safety
Conference Motorcycle Safety Foundation 19906-27 to 6-41
Kelsey 1986 published data only
Kelsey SL Liddicoat C Ratz M Licensing novice motorcyclists a
comparison of Californiarsquos standard test and the MOST II (Motorcycle
Operator Skill Test) administered at centralized testing offices Sacramento California USA California Department of Motor
Vehicles Research and Development Office 19861ndash74
Koch 1980 published data only
Koch H Advanced course for motorcycle riders InternationalMotorcycle Safety Conference Washington DC USA Motorcycle
Safety Foundation 1980269ndash81
Koch 1990 published data only
Koch H Brendicke R How to influence risk taking behaviour
within motorcycle rider training programs The Human Element
International Motorcycle Safety Conference Orlando Florida
USA Motorcycle Safety Foundation 1990Motorcycle Safety
Foundation 1990 p 9-1 to 9-25A
Labbett 2005 published data only
Labbett S Langham M Training can make the problem worse
Road Safety Congress 20051ndash12
Law 2005 published and unpublished data
Law TH Radin Umar RS Zulkaurnain S Kulanthayan S Impactof the effect of economic crisis and the targeted motorcycle safety
programme on motorcycle-related accidents injuries and fatalities
in Malaysia International Journal of Injury Control and Safety
Promotion 200512(1)9ndash21
Lowes 1990 published data only
Lowes BF In-traffic rider training The Human Element
International Motorcycle Safety Conference Orlando Florida
USA Motorcycle Safety Foundation 19907-46 to 7-57
Mahon 1976 published data only
Mahon J Teaching accident avoidance on the motorcycle Journal of Traffic Safety Education 1976231ndash29
Maxwell 1985 published data only
Maxwell GS Evaluation of a Program for Training MotorcycleRiders Evaluation and Program Planning 19858(4)339ndash47
Mayhew 1996 published data only
Mayhew DR Simpson SH Effectiveness and role of driver educationand training in a graduated licensing system Ottowa Canada The
Traffic Injury Research Foundation 19961ndash89 [ ISBN
0ndash920071ndash12ndash0]
McKnight 1987 published data only
McKnight AJ Milburn BG An evaluation of the modified
motorcycle operator skill test exercises NHTSA Report 1987 [
Report No DOT HS 807 212 ]
McPherson 1976 published data only
McPherson K McKnight AJ The development and evaluation of a
motorcycle skill test manual and knowledge test Washington DCUSA National Highway Safety Administration 1976
Mohan 2004 published data only
Mohan D Evidence-based interventions for road traffic injuries in
South Asia Journal of the College of Physicians and SurgeonsPakistan 200414(12)746ndash7
Morgan 1980 published data only
Morgan JK Advanced motorcycle training course content and
study evaluation International Motorcycle Safety Conference
Washington DC USA Motorcycle Safety Foundation 1980
283ndash312
Motoki 1990 published data only
Motoki M Yamazaki S A study on effective motorcycle rider
education The Human Element International Motorcycle Safety
26Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2950
Conference Orlando Florida USA Motorcycle Safety
Foundation 19909-26 to 9-57A
Nairn 1992 published data onlylowast Nairn RJ Motorcycle safety research literature review 1987 to
1991 Federal Office of Road Safety Australia 1993 Vol Report
CR 117
Newland 1980 published data only
Newland R Darby J Automotive practicespre motorcycle rider
program International Motorcycle Safety Conference Washington
DC USA Motorcycle Safety Foundation 1980127ndash9
Newland 1982 published data only
Newland R Motorcycle rider training and licensing requirements
Driver training steering a course for the future Melbourne
Victoria Transport Regulation Board 1982
Newland 1990 published data only
Newland R Motorcycle training and licensing in Victoria 1980-
1990 The Human Element International Motorcycle Safety
Conference Orlando Florida USA Motorcycle Safety
Foundation 1990
Ormston 2003 published data only
Ormston R Dudleston A Pearson S Stradling Evaluation of
Bikesafe Scotland Scottish Executive Social Research Edinburgh
2003
Osga 1979 published data only
Osga GS Ellingstad VS An evaluation of the effectiveness of the 1978
South Dakota motorcycle rider course program South Dakota USA
University of South Dakota 19791ndash89
Panou 2007 published data only
Panou M Bekiaris E A new concept on the integration of driving
simulators in driving training the train-all approach Road Safety
on Four Continents 14th International Conference Swedish
National Road and Transport Research Institute 2007
Peterson 1990 published data only
Peterson A Kadar K Status report 1980-1989 Minnesota rider
education statistics The Human Element International
Motorcycle Safety Conference Orlando Florida USA 199013-
113 to 13-129
Prem 1984 published data only
Prem H Good MC Motorcycle rider skills assessment ACT
Australia Office of Road Safety Department of Transport 1984
1ndash377 [ Report No CR 34 ]
Radin Umar 2006 published data only
Radin Umar RS Motorcycle safety programmes in Malaysia how
effective are they International Journal of Injury Control and Safety Promotion 200613(2)71ndash9
Rockwell 1990 published data only
Rockwell TH Kiger S Carnot M An evaluation of the Ohio
motorcyclist enrichment program Ohio Department of Highway
Safety 1990
Rothe 1987 published data only
Rothe JP A motorcycle riding course Motorcyclists Image and reality Canada Insurance Corporation of British Columbia 1987
59ndash83
Rowden 2007 published data only
Rowden P Watson B Haworth N What can riders tell us aboutmotorcycle training A view from the other side of the fence
Australasian Road Safety Research Policing and Education
Conference Melbourne Australia Melbourne 2007
RSC Victoria 1993 published data only
Road Safety Committee Inquiry into Motorcycle Safety in Victoria
Parliament of Victoria 1993
Russam 1979 published data only
Russam K Motorcycle training and licensing in Japan UK Transport
and Road Research Laboratory (TRRL) 1979 [ Report 916 ]
Saffron 1981 published data only
Saffron DG Driver and rider education training and licensing a
brief review Traffic Accident Research Unit Traffic Authority of
New South Wales NSW Australia Traffic Accident Research UnitTraffic Authority of New South Wales 1981
Schulz 1990 published data only
Schulz U Motorcycle accidents among learner riders Motorcycle
licensing Evaluation of the effectiveness of a Graduated Licensing Scheme Forshungshefte des Institut fur Zweiradsicherheit 1990
Sexton 2004 published data only (unpublished sought but not used)
Sexton B Baughan C Elliott M Maycock G The accident risk of
motorcyclists 2004TRL Report TRL 607(ISSN 09684107)58
Shephard 1986 published data only
Shepherd R Do Motorcycle Safety Foundation programs work
Driver 198619(9)16ndash8
Simpson 1990 published data onlySimpson HM Mayhew DR The promotion of motorcycle safety
training education and awareness Health Education Research
Theory and Practice 19905(2)257ndash64
Sokolov 1990 published data only
Sokolov O Safe motorcycle riding some aspects of motorcycle
riding in the USSR The Human Element International
Motorcycle Safety Conference Motorcycle Safety Foundation
19904ndash37 - 4-48
Steffens 1988 published data only
Steffens U Gawatz R Willmes G Effectiveness of mofa courses
[Wirksamkeit von Mofakursen] Unfall- und sicherheitsforschung
strasenverkehr 198867168
Sudlow 2003 published data only
Sudlow D Scoping study on motorcycle training Road Safety
Research Report London UK The British Institute of Traffic
Education Research Department for Transport 2003 Vol June
[ Report No 36]
Tasmania 1995 published data only
Mannering FL Grodsky LL Motorcycle rider training scheme
Traffic Education Research Centre Armidale Australia Driver
Training Australia Limited 1995 Vol April1ndash56
Tetard 1985 published data only
Tetard C Motorcycle ridersrsquo training to detect indices in an
intersection situation [Lrsquoentrainement a la prise drsquoinformation dans
lrsquoapprentissage de la conduite motocycliste] Le Travail Humain198548(2)147ndash59
27Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3050
Thackray 1980 published data only
Thackray RM Prescott JC Field test of the motorcycle ridercourse International Motorcycle Safety Conference Washington
DC USA Motorcycle Safety Foundation 1980195ndash7
Thompson 1994 published data only
Thompson M Evaluation of compulsory basic training for
motorcyclists Transport Research Laboratory Crowthorne
Berkshire UK 1994 [ Report 63]
Tobler 1990 published data only
Tobler U Individual driving instruction for motorcycle driving
learners the Swiss experiment of motorcycle driving school The
Human Element International Motorcycle Safety Conference
Orlando Florida USA Motorcycle Safety Foundation 1990
Transport 1989 published data only
Transport Communications Road Crash Statistics Australia Bureauof Transport and Communications Economics 198914ndash20
Tuan 2007 published data only
Tuan VA Shimizu T Towards development and evaluation of the
motorcycle drivers re-education program in Vietnam modeling of
motorcycle driverrsquos undesired behaviors 11th World Conference
on Transport Research 200725
Victor 1980 published data only
Victor DJ Traffic safety education for motorcyclists International
Motorcycle Safety Conference Washington DC USA Motorcycle
Safety Foundation 1980221ndash8
Waterer 1990 published data only
Waterer TH Compulsory training for motorcyclists from liberty
to compulsion The Human Element International Motorcycle
Safety Conference Orlando Florida USA Motorcycle Safety
Foundation 19904-49 to 4-57
Weller 1989 published data only
Weller RB Chandler EW Motorcycle safety and motorcycle
education past research and survey results Journal of Public Policy
amp Marketing 1989893ndash108
Wells 1981 published data only
Wells P Observations of some RACACU motorcycle training
schemes Transport and Road Research Laboratory Crowthorne
Berkshire UK 1981
Winn 1986 published data only
Winn G Two important trends in motorcycle safety regulations
rider education and conspicuity improvement American
Motorcyclist Association 1986147ndash52
Wisher 1988 published data only
Wisher G Reid J Motorcycle accidents in Norfolk a three-year study of motorcycle training and the Part 1 test Norfolk UK Road Safety
Division Highways Department Norfolk County Council 1988
Wood 1987 published data only
Wood HT Bowen R Evaluation of the revised learner permit scheme
July 1983 to December 1985 Victoria Australia Road Traffic
Authority 19871ndash27 [ ISBN 0 7306 0247 8]
Yamazaki 1990 published data only
Yamazaki S Nakanishi M Survey on effect of Honda rider
education The Human Element International Motorcycle Safety
Conference Orlando Florida USA 199012-167 to 12-187
References to ongoing studies
Burgess 2010 unpublished data only
Burgess C Broughton P Fylan F Stradling Interim evaluation of
the UKs National Ride program In L Dorn editor(s) Driver
behaviour and Training Vol IV Aldershot Ashgate Publishing
Limited 2010
Additional references
Anderson 1980
Anderson J Ford JL Peck RC Improved motorcyclist licensing and
testing project Volume I and II National Highway Traffic Safety
Administration US department of Transportation Washington
DC 1980
Bjornstig 1985
Bjornstig U Bylund P Lekander T Brorsson B Motorcycle fatalitiesin Sweden Acta Chirurgica Scandinavica 1985151(7)577ndash81
Clayton 1990
Clayton A Sudlow DE An evaluation of the effectiveness of a one-
day training program for learner motorcyclists The Human
Element International Motorcycle Safety Conference Orlando
Florida USA 199013-105 to 13-111
Cochrane Collaboration 2008
Higgins JP Green S Cochrane Handbook for Systematic Reviews of
Interventions 500 The Cochrane Collaboration 2008
Elliot 2003
Elliott MA Baughan CJT Broughton J Grayson G Knowles J et
alMotorcycle Safety a Scoping Study TRL [ Report No
0968ndash4107]Fleiss 2004
Fleiss JL Levin B Paik MC Statistical methods for rates and proportions 3rd Edition John Wiley amp Sons Inc 2004 [ ISBN
978ndash0ndash471ndash52629ndash2 ]
FORS 1999
FORS Federal Office of Road Safety Road risk for sober licensed
motorcyclists Vol Monograph 27 Federal Office of Road Safety
19991ndash2
Haworth 2005
Haworth N Mulvihill C Review of motorcycle licensing and training
Victoria Australia Monash University Accident Research Centre
20051ndash83 [ Report No 240]
Huwiler-Muumlntener 2002Huwiler-Muumlntener K Juumlni P Junker C Egger M Quality of
Reporting of Randomized Trials as a Measure of Methodologic
Quality JAMA 20022872801ndash4
Johnston 2008
Johnston P Brooks C Savage H Fatal and serious road crashes
involving motorcyclists Road Safety Research and Analysis Report
ATSB Monograph 20 Canberra Department of Infrastructure
Transport Regional Development and Local Government 0205
081ndash26
Juumlni 2001
Juumlni P Altman DG Egger M Systematic reviews in health care
Assessing the quality of controlled clinical trials BMJ 2001323
42ndash6
28Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3150
Kopits 2003
Kopits E Cropper M Traffic fatalities and economic growth TheWorld Bank Development Research Group Infrastructure and
Environment Policy research working paper 3035 2003
Leung 1983
Leung P Chau BTW Motorcycle accidents in Hong Kong pattern
of injury treatment causative factors and socio-economic
implications Injury 198315(1)23ndash9
Liu 2004
Liu B Ivers R Norton R Blows S Lo SK Helmets for preventing
injury in motorcycle riders Cochrane Database of Systematic Reviews 2004 Issue 2 [DOI 10100214651858CD004333pub3]
Mohan 2002
Mohan D Road safety in less-motorized environments future
concerns International Journal of Epidemiology 200231(3)527ndash32Mortimer 1982
Mortimer RG An evaluation of the motorcycle rider course
Proceedings of the 20th conference of the American Association for
Automotive Medicine Arlington Heights IL USA American
Association for Automotive Medicine 198277ndash96
Mullin 1997
Mullin BT Risk factors for motorcycle injury the role of age
gender experience training and alcohol PhD thesis Auckland
New Zealand University of Auckland 1997
Noordzij 2001
Noordzij PC Forke E Brendicke R Chinn BP Integration of needs
of moped and motorcycle riders into safety measures review and
statistical analysis in the framework of the European research
project PROMISING SWOV Institute for Road Safety Research
Leidschendam The Netherlands 2001 [ Report No Dndash2001ndash5]
Osga 1979
Osga G Ellingstad VS An evaluation of the effectiveness of the 1978 South Dakota Motorcycle Riders Course Program South Dakota
University of South Dakota Vermillion Human Factors Lab 1979
Oxford English Dictionary Online
httpoxforddictionariescom Oxford English Dictionary Online
Oxford University Press 2010
Peden 2004
Peden M Scurfield R Sleet D World Report on Road TrafficInjury Prevention Geneva World Health Organization Available
at httpwwwwhointviolence˙injury˙preventionpublications
road˙trafficworld˙reporten 2004Chapter 3
Radin Umar 1995
Radin Umar R Macakay G Hills B Preliminary analysis of
motorcycle accidents Short-term impacts of the running
headlights campaign and regulations in Malaysia J Traffic Med 199523(1)17ndash27
Tsai 1995
Tsai YJ Wang JD et alCase-control study of the effectiveness of
different types of helmets for the prevention of head injuries among
motorcycle riders in Taipei Taiwan American Journal of
Epidemiology 1995142(9)974ndash81
Vis 1995
Vis AA In-depth study of the hazards of motorcycling [De
onveiligheid van motorrijden nader bekeken] SWOV Institute for
Road Safety Research Leidschendam the Netherlands 1995 [
Report No Rndash95ndash69]
Weaver 1990
Weaver EA Motorcycle safety education in the United States
Motorcycle Safety Foundation Orlando Florida USA 1990
Winn 1987
Winn G Two important trends in motorcycle safety regulations
rider education and conspicuity improvement No more
information available Detroit MI US Society of Automotive
Engineers 1987147ndash52
Yuan 2000
Yuan W The effectiveness of the rsquoride-brightrsquo legislation for
motorcycles in Singapore Accident Analysis amp Prevention 200032
(4)559ndash63
Zambon 2006
Zambon F Hasselberg M Factors affecting the severity of injuries
among young motorcyclists a Swedish nationwide cohort study
Traffic Injury Prevention 20067(2)143ndash9lowast Indicates the major publication for the study
29Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3250
C H A R A C T E R I S T I C S O F S T U D I E S
Characteristics of included studies [ordered by study ID]
Adams 1985
Methods Non-randomised trial
Participants Initial sample size
Total of 162 Australia Post telegram delivery riders
Not clear how many in intervention and control groups
Analysis sample size
Intervention group 63Control group 35
Total 98 (61)
Interventions No formal training vs voluntary rider training course of 12 hours over 2 consecutive days (theory 4 hours practice
8 hours)
Intervention integrity
No information given
Outcomes Australia Post recorded crashes
Length of follow up 1 year
Notes Study conducted in NSW Australia
Anderson 1980
Methods Randomised trial
Participants Initial sample size
Total 41188 motorcycle licence applicants
Analysis sample size
Intervention group with skills training 13160
Intervention group without skills training 12634
Control group 15080
Total 40874 (99)
Interventions Standard licensing skill test vs the Motorcycle Operators Skill Test (MOST) with or without a training component
MOST is not to be confused with rider training because it is a licensing test The training component was for
participants who failed the skill test on the first attempt was of 3 hours duration and consisted of instructions about
the skills needed to pass the skill test The training consisted of 30 min theory and 2 frac12 hours practice on the
motorcycle
Intervention integrity
33 in MOST with skills training 31 in MOST without skills training and 43 in control group obtained their
motorcycle licence in the end
Outcomes Police recorded crashes and offences
Length of follow up 3 years
30Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3350
Anderson 1980 (Continued)
Notes Study conducted in California USA
Billheimer 1998
Methods Cohort study
Participants Initial sample size
Intervention group Unclear numbers of CMSP graduates and ERC graduates
Control group Over 16000 untrained motorcyclists
Analysis sample size
1139 matched CMSP-pairs and 1182 ERC-pairs
Interventions No formal training vs the mandatory (for those younger than 18 years) pre-licence 16-hours basic California
Motorcyclist Safety Program (CMSP) and the 8-hour Experienced Rider Course (ERC)
Intervention integrity
No information given
Outcomes Police recorded crashes and offences
Length of follow up 2 years
Notes Study conducted in California USA
Cooper 1988
Methods Cohort study
Participants Initial sample size
Intervention group 418 graduates and 57 who failed
Control group 402
Analysis sample size
Intervention group unclear
Control group unclear
Total 863 (984)
Interventions No formaltrainingvs non-mandatory pre-licence rider trainingThe content of thetrainingwas notfurther explainedIntervention integrity
Participants had all completed the rider training
Outcomes Police recorded crashes and offences
Length of follow up 7 years
Notes Study conducted in British Columbia Canada
31Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3450
Davis 1997
Methods Retrospective cohort study
Participants Initial sample size
Intervention group 9320 graduates
Control group unclear
Analysis sample size
Intervention group 9320 graduates
Control group unclear
Interventions No formal training vs the non-mandatory Connecticut Motorcycle Operator Training Program The content of
training was not further explained
Intervention integrity Participants had all completed the rider training
Outcomes Police recorded crashes severity of crashes
Length of follow up 1 year
Notes Study conducted in Connecticut USA
Hall 1988
Methods Randomised trial
Participants Initial sample sizeIntervention group A 6248
Intervention group B 6616
Intervention group C 6996
Control group 6604
Total 26464 motorcycle licence applicants
Analysis sample size
Intervention group A 6012 (96)
Intervention group B 6428 (97)
Intervention group C 6778 (97)
Control group 6502 (98)
Interventions Standard licensing skill test vs the Motorcycle Operator Skill Test (MOST II) without training (A) MOST II with
3 hours training (B) or MOST II with 20 hours training (C) The 3 hours training comprised instructions aboutthe skills needed to pass the skill test (30 min theory and 2 frac12 hours practice on the motorcycle) The 20 hours
training consisted of the Motorcycle Safety Foundationrsquos Motorcycle Rider Course (MRC) comprising of 8 hours
theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-road practice
Intervention integrity
25 and 26 obtained their motorcycle licence in the lsquoMOST 3-hour trainingrsquo and lsquoMOST 20-hour trainingrsquo
intervention arm respectively In the lsquoMOST testrsquo intervention arm 29 obtained their licence and in the control
group 35
Outcomes Police recorded crashes and offences
Length of follow up 3 years
32Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3550
Hall 1988 (Continued)
Notes Study conducted in New York USA
Haworth 1997
Methods Population based case-control study
Participants Cases Motorcycle riders or pillion riders involved in motorcycle crash (on road) Recruited from hospital and police
records Not clear whether all cases (especially fatalities) were included
Community controls Multiple matched sample of motorcycle rider or pillion who passed the crash site at the same
time of day and day of week that the crash occurred
Initial sample size
Cases 336
Controls 1195
Analysis sample size
Cases 234 (70)
Controls 561 (47)
Interventions Basic handling skills course beginner intermediate or advanced course measured through self-report not validated
Outcomes Injuries with an Injury Severity Score of 5 or greater or admission for at least four hours and fatalities from a
motorcycle crash (from hospital and police records)
Length of follow up two years or more prior to the interview
Notes Study conducted in Victoria Australia
Hurt 1981
Methods Population based case-control study
Participants Cases Motorcycle riders who had a motorcycle crash that was reported to police (traffic accident reports) Not all
cases included
Community controls Multiple matched sample of motorcycle riders who passed the crash site at the same time of
day and day of week as the crash occurred but 2 years later
Initial sample size
Cases 4500 notifications of crashes to police (only 50 received in time research team able to respond to 1126 for
unreported reasons data collection completed for 900)
Controls Number approached unreported 2310 consented and interviewed
Analysis sample size
Cases 808 (808900 data on rider training) - 18 of initial sample
Controls 670 (6702310 data on rider training) - lt 29 of initial sample
Interventions Training experience self taught friends-family school-club formal-professional others measured through self-
report validated
Outcomes Injuries and fatalities from a motorcycle crash (from hospital records and coronerrsquos office)
33Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3650
Hurt 1981 (Continued)
Notes Study conducted in Los Angeles USA
Jonah 1982
Methods Cohort study
Participants Initial sample size
Intervention group
2310 licensed MTP graduates who are active riders
Control group
2029 licensed motorcyclists who are active riders
Non-licensed and non-active riders excluded
Analysis sample size
Intervention group
811 interviews (35) 533 driver records (23)
Control group 1080 interviews (71) 905 driver records (45)
Interventions No training vs the non-mandatory Canada Safety Council Motorcycle Training Program (MTP) comprising of 4
hours theory and 16 hours practise (6 hrs on-road) over 2 weekends
Intervention integrity
Participants had all successfully completed the MTP
Outcomes Self-reported and police reported crashes and offencesLength of follow up 4 years
Notes Study conducted in Ontario Canada
Kloeden 1994
Methods Cohort study
Participants Initial sample size
Intervention group 2533 licence applicants
Control group Unclear
Sample of riders obtaining probationary licenceIntervention group 1625 (64)
Control group 5015 (unclear)
Non-active riders excluded from control group (~66)
Analysis sample size
Intervention group 631 (39)
Control group 318 (6)
Interventions No formal training vs the mandatory pre-licence Ridersafe course consisting of two levelsthe first is a training course
in the basics of motorcycle riding where a pass entitles the rider to obtain a learnerrsquos permit The second is a skill
test where a pass entitles the rider to obtain a probationary motorcycle licence The training course consists of two 4
hour training sessions on separate days involving instruction videos and actual riding on motorcycles The skill test
consists of a 4 hour preparation course of 74 minutes of lecturediscussion 38 minutes of video and 75 minutes of
34Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3750
Kloeden 1994 (Continued)
on-course riding which is the testing phase
Intervention integrity
Unclear what percentage in the Ridersafe group and the control group obtained their probationary licence halfway
and their full motorcycle licence in the end
Outcomes Police reported crashes
Length of follow up 1 year
Notes Study conducted in SA Australia
Kraus 1975
Methods Population based case-control study
Participants Cases All motorcycle riders with a confirmed medically treated motorcycle injury (including fatalities) from all
hospitals in study base and peripheral area hospitals and all deaths due to motorcycle crashes in the study base All
serious injuries (leading to hospitalisation) and fatalities likely to be included
Community controls A random sample of registered motorcycle owners in the same temporary and geographical
study base as the cases arose from
Initial sample size Cases 1273
Controls 738
Analysis sample size Cases 741 (58)
Controls 434 (59)
Interventions Motorcycle rider training (not specified) measured through self-report
Outcomes Injuries leading to hospitalisation fatalities
Length of follow up not specified
Notes Study conducted in Sacramento county USA
Leung 1987
Methods Non-randomised trial
Participants Initial sample sizeIntervention group unknown number of MRC graduates
Control group unknown number of licensed motorcyclists
Analysis sample size
Intervention group 2941
Control group 43094
Interventions No formal training vs non-mandatory Wisconsin motorcycle rider course a 22 hours off-street programme that
teaches the basic rules of safe motorcycling
Intervention integrity
50 in the MRC group and 100 in the control group obtained their motorcycle licence in the end
35Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3850
Leung 1987 (Continued)
Outcomes Police recorded crashes and offences
Length of follow up up to 3 years
Notes Study conducted in Wisconsin USA
MAIDS 2004
Methods Population based case-control study
Participants Cases Motorcycle riders who had a motorcycle crash Unclear whether all cases included
Community controls Motorcycle riders identified at petrol stations Unclear whether matched to cases
Initial sample size
Cases not reported
Controls not reported
Analysis sample size
Cases 921 (Italy 200 Netherlands 200 Spain 121 Germany 250 France 150)
Controls 923 (Italy 200 Netherlands 200 Spain123 Germany 250 France 150)
Interventions Pre-license rider training or additional training (non-compulsory)
Outcomes Crashes
Notes Study in five sampling areas located in five European countries (France Germany Netherlands Spain and Italy)
McDavid 1989
Methods Cohort study
Participants Sample size after first match
Intervention group 346 male training graduates
Control group 346 male motorcycle riders licensed in same year
Sample size after second match
Intervention group 139 (40)
Control group 139 (40)
Analysis sample size
Intervention group 139 (40)
Control group 139 (40)
Interventions No formal training vs the non-mandatory pre-licence British Columbia Safety Councilrsquos 37-hour motorcycle safety
programme
Intervention integrity
Participants had all completed and passed the training
Outcomes Police reported crashes injurious crashes
Length of follow up 1-6 years
36Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3950
McDavid 1989 (Continued)
Notes Study conducted in British Columbia Canada
Mortimer 1984
Methods Cohort study
Participants Initial sample size
Intervention group 608 MRC graduates
Control group 360 motorcycle riders
Response rates
Intervention group 360 (59)
Control group 324 (90)
Non-active riders excluded
Analysis sample size
Intervention group 213 (35)
Control group 303 (84)
Interventions No formal training vs the non-mandatory Motorcycle Safety Foundationrsquos Motorcycle Rider Course comprising of
8 hours theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-road practice
Intervention integrity
Participants had all completed and passed the MRC
Outcomes Self-reported crashes and offencesLength of follow up 1 year
Notes Study conducted in Illinois USA
Mortimer 1988
Methods Cohort study
Participants Initial sample size
Intervention group 2307 MRC graduates
Control group 556 active licensed motorcycle riders
Response ratesIntervention group 1298 (56)
Control group 500 (90)
Non-active riders excluded
Analysis sample size
Intervention group 913 (40)
Control group 500 (90)
Interventions No formal training vs the non-mandatory Motorcycle Safety Foundationrsquos Motorcycle Rider Course comprising of
8 hours theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-road practice
Intervention integrity
Participants had all completed and passed the MRC
37Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4050
Mortimer 1988 (Continued)
Outcomes Self reported crashes and offences
Length of follow up 1 year
Notes Study conducted in Illinois USA
Osga 1980
Methods Cohort study
Participants Initial sample sizeIntervention group 1163 MSF MRC participants invited 818 (74) responded
Control group 329 motorcycle riders responded to recruitment strategies
Non-active riders excluded from analysis
Analysis sample size
Intervention group 595 (73)
Control group 329 (100)
Interventions No formal training vs the non-mandatory Motorcycle Safety Foundationrsquos Motorcycle Rider Course (MRC) com-
prising of 8 hours theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-road
practice
Intervention integrity
Participants had all successfully completed the MRC
Outcomes Self-reported crashes police recorded crash reports
Length of follow up 6 months - 3 12 years
Notes Study conducted in South Dakota USA
Perrino 2002
Methods Cohort study
Participants Initial sample size
Intervention group 2144 (unclear) MRC graduates (motorcycle licence applicants)
Control group 2601 (unclear) regular licensing test graduates (motorcycle licence applicants)Analysis sample size
Intervention group 1070 (unclear)
Control group 867 (unclear)
Interventions No formal training vs the mandatory (for those younger than 18 years) pre-licence Maryland Motorcycle Rider
Course (MRC) consisting of 8 hours theory and 12 hours practice
Intervention integrity
No information given
Outcomes Self-reported and police recorded crashes and offences
Length of follow up 1 year
38Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4150
Perrino 2002 (Continued)
Notes Study conducted in Maryland USA
Raymond 1979
Methods Cohort study
Participants Initial sample size
Intervention group
258 volunteers mainly (but not all) novice riders
Control group
1104 provisional licence applicants
Analysis sample size
Intervention group 190 (74)
Control group 627 (57)
Interventions No formal training vs the non-mandatory RACACU (Royal Automobile ClubAuto-Cycle Union) training scheme
The training program consisted of either 24 one-hour sessions or 13 two-hour sessions Twelve hours of the training
focused on theory about basic operational procedures traffic laws and highway code The practical session focused
on basic operational skills riding on a training area and riding in traffic with an instructor A proficiency test was
administered at the end of the course
Intervention integrity
No information given
Outcomes Self-reported crashes (including minor non-injury non-damage and off-road crashes)
Length of follow up 1 year
Notes Study conducted in the UK
Satten 1980
Methods Cohort study
Participants Initial sample size
Intervention group interviewers kept interviewing until they had 100 riders selected from a population of 891
enrollees in the MSF MRC course (unknown how many were asked and how many refused to participate)Control group unknown number of registered motorcycle riders
Non-active riders excluded from analysis
Analysis sample size
Intervention group 71
Control group 100
Interventions No formal training vs the non-mandatory Motorcycle Safety Foundationrsquos (MSF) Motorcycle Rider Course (MRC)
comprising of 8 hours theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-
road practice
Intervention integrity
Participants had all successfully completed the MRC
39Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4250
Satten 1980 (Continued)
Outcomes Self-reported crashes injuries and offences
Length of follow up 3 years
Notes Study conducted in Illinois USA
Savolainen 2007
Methods Cohort study
Participants Initial sample size
Intervention group 4000 riders that had conducted the MSF MRC course (passed or failed)
Control group 4000 motorcycle owners or permit or endorsement holders from the Indiana Bureau of Motor
Vehicles database
Analysis sample size
Intervention group 739 (18)
Control group 588 (15)
Interventions Intervention integrity
No information given
Outcomes Probability of crash involvement
Length of follow up up to 5 years
Notes Study conducted in Indiana USA
Swaddiwudhipong 1998
Methods Randomised trial
Participants Initial sample size
Intervention group
all 16 villages of 3 sub-districts
(participants were practising motorcyclists licensed and unlicensed)
Control group
18 villages of 3 other sub-districtsAnalysis sample size
Intervention group 3 districts (100)
Control group 3 districts (100)
Interventions No rider education vs non-mandatory community based rider education consisting of theory on motorcycle injury
risk helmet use traffic laws vehicle regulations and traffic signs and a written and skill test leading to a licence
Intervention integrity
No information given
Outcomes Injuries fatalities
Length of follow up 2 years
40Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4350
Swaddiwudhipong 1998 (Continued)
Notes Study conducted in Tak Province rural Thailand
Waller 1992
Methods Cohort study
Participants Initial sample size
Intervention group 349 riders who succesfully completed the Department of Public Safety motorcycle operator
training in 1988
Control group 349 motorcyclists who never completed the DPS motorcycle operator training
Analysis sample size
Intervention group 349
Control group 349
Interventions No DPS motorcycle operator training vs non-mandatory DPS motorcycle operator training The DPS training is
the Motorcycle Safety Foundationrsquos Motorcycle Rider Course comprising eight hours of theory (lectures discussions
films and multiple choice tests) a textbook and 12 hours off-road practice
Intervention integrity
All 349 riders succesfully completed the rider training
Outcomes Police recorded crashes
Length of follow up 3 years
Notes Study conducted in
Characteristics of excluded studies [ordered by study ID]
Study Reason for exclusion
Anderson 1980a Study type review and process evaluation of a rider course
Aupetit 2008 Outcome proposal of a rider training curriculum that combines traditional training and a motorcycle riding
simulator
Awane 1999 Study type review on use of simulators in rider and driver training
Baldi 2005 Intervention not rider training but intervention integrity of rider training availabilty of rider education
implementation of graduated licensing system etc on a state-wide level
Batchler 1988 Study type description of a training course
Blanchard 2006 Study type not a study assessing the association between rider training and outcomes
Bonnett 2005 Outcome qualitative data regarding satisfaction with the training programme
41Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4450
(Continued)
Braver 2007 Outcome licensure rates
Clayton 1980 Outcome knowledge skills attitude
Council 1978 Outcome crash reduction in ANY vehicle not motorcycles exclusively
Crowther 2005 Study type qualitative study
Daniello 2009 Study type review of the effectiveness of motorcycle training and licensing
Dewanti 2007 Intervention emotional characteristics behaviour (ie not motocycle rider training)
Du Plooy 1990 Study type review of types of rider training courses in South Africa
Engel 1989 Study type uncontrolled before-after study
Goldenbeld 2002 Outcome skills
Goldenbeld 2004 Outcome knowledge and skills
Haworth 2000 Intervention balance between attitudinal andvehicle control skills acrosscurriculla of motorcycle rider training
courses
Haworth 2005 Study type review
Haworth 2006 Study type cross-sectional survey
Hill 1983 Study type review and description of training system
Imatake 1980 Study type course descriptionreview
Ingebritsen 1990 Intervention risk factors for motorcycle crashes
Jain 2009 Study type cross-sectional descriptive study of motorcycle crashes over a 5 year period
Jonah 1979 Study type cross-sectional survey Outcome scores on motorcycle operator skill test
Jonah 1981 Intervention evaluation of a motorcycle rider skill test not motorcycle rider training
Kadar 1990 Study type review of components of rider training courses
Kelsey 1986 Intervention evaluation of a motorcycle rider skill test not motorcycle rider training
Koch 1980 Study type reviewcourse description
Koch 1990 Study type description of course content
42Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4550
(Continued)
Labbett 2005 Study type essay about possible explanations for motorcycle deaths
Law 2005 Intervention targeted motorcycle safety programme not specifically rider training
Lowes 1990 Study type review and descriptions of rider training
Mahon 1976 Study type review of motorcycle rider training methods
Maxwell 1985 Outcomesrecommendations forprogramme operatorsand otherson howthe motorcycle trainingprogramme
works and how it can be improved
Mayhew 1996 Study type review
McKnight 1987 Study type evaluation of methodological and administrative aspects of a motorcycle operator skill test
McPherson 1976 Outcome knowledge and skills
Mohan 2004 Study type review of evidence of interventions to reduce road traffic injuries
Morgan 1980 Study type reviewcourse description
Motoki 1990 Study type proposal and development of rider training curriculum
Nairn 1992 Study type literature review
Newland 1980 Study type description of development of motorcycle safety course
Newland 1982 Study type descriptionreview of a rider training course
Newland 1990 Study type uncontrolled BeforeAfter study
Ormston 2003 Outcome changes in attitude and self reported behaviour
Osga 1979 Outcome knowledge attitudes and skills
Panou 2007 Intervention new simulation tool to teach driverrider
Peterson 1990 Study factor demographic details of course participants completing the course vs those not completing the
course
Prem 1984 Study type experimental
Study factor examines association between skills and performance in tests
Radin Umar 2006 Intervention motorcycle safety programme not specifically rider training
Rockwell 1990 Study type methodologically flawed study type
43Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4650
(Continued)
Rothe 1987 Study type qualitative study of rider training in general
Rowden 2007 Study type qualitative study
RSC Victoria 1993 Study type review of motorcycle rider training and education in Victoria
Russam 1979 Study type review of training programmes in Japan
Saffron 1981 Study type literature review
Schulz 1990 Study factor predictors of motorcycle crashes
Sexton 2004 Study type cross-sectional analytic analysis
Shephard 1986 Study factor reviewdescription of programme
Simpson 1990 Study type review of motorcycle rider trainingeducation
Sokolov 1990 Study factor review of motorcycle riding courses in USSR
Steffens 1988 Outcome self-reported behaviour experiences and researcher observed riding behaviour
Sudlow 2003 Study type review of motorcycle training in the UK
Tasmania 1995 Study type cross-sectional descriptiveprocess evaluation
Tetard 1985 Study type experimental
Outcome detection of indices
Thackray 1980 Outcome instructional effectiveness user acceptance administrative feasibility
Thompson 1994 Study type cross-sectional survey
Outcome attitudes and knowledge
Tobler 1990 Study type review of course curricula
Transport 1989 Intervention syllabus for rider training
Tuan 2007 Study type cross-sectional study (survey) which does not assess the effects of an intervention but examines
motocyclistsrsquo behaviour
Victor 1980 Study type review
Waterer 1990 Study type review of UK rider training issues
Weller 1989 Study type review of studies evaluating motorcycle safety and education measures
44Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4750
(Continued)
Wells 1981 Study type description of different methods of training
Winn 1986 Study type review of rsquohistoryrsquo of rider training
Wisher 1988 Study type cross-sectional survey
Wood 1987 Study type Interrupted Time Series but did not meet the criteria for an ITS as published by the Cochrane
Effective Practice and Organisation of Care Group
Yamazaki 1990 Study type cross-sectional survey
Outcome riding skills skills retention and attitude
Characteristics of ongoing studies [ordered by study ID]
Burgess 2010
Trial name or title UKs National Ride program
Methods Cohort study
Participants Motorcycle offenders
Interventions Police led diversion program
Outcomes Self reported attitudinal and behavioural measures
Starting date
Contact information
Notes Expected to be completed in early 2010
45Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4850
D A T A A N D A N A L Y S E S
This review has no analyses
A P P E N D I C E S
Appendix 1 Search strategy
We searched the following electronic databases
bull Cochrane Injuries Group Specialised Register (searched 22 Sept 2008)
bull CENTRAL (The Cochrane Library 2008 Issue 3)
bull TRANSPORT 1988 to 200706 (latest date available)
bull ATRI Australian Transport Index (to Sept 2008)
bull MEDLINE (Ovid SP) 1950 to Sept 2008
bull EMBASE (Ovid SP) 1980 to Sept 2008
bull CINAHL (EBSCO) 1982 to Sept 2008
bull WHOLIS (World Health Organization Library Information System) (to Sept 2008)
bull PsycINFO (Ovid SP) 1806 to Sept 2008
bull LILACS (BIREME) (Latin American and Caribbean Health Sciences) (1982 to Sept 2008)
bull ISI Web of Science Social Sciences Citation Index (SSCI) 1970 to Sept 2008
bull ERIC (searched 25 Sept 2008)
bull ZETOC (searched 25 Sept 2008)
bull SIGLE (searched 25 Sept 2008)
All of the databases were searched using the following text-word search strategy The strategy was adapted using either rsquorsquo or rsquo$rsquo as atruncation symbol
1 motor-cycl or motorcycl or motor-bik or motorbik or motor-scooter or motorscooter or scooter or moped or motocycl
or motocicl or moto
2 train or educat or instruct or program or course or qualif or learn or improv
3 1 and 2
H I S T O R Y
Protocol first published Issue 2 2005
Review first published Issue 10 2010
46Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4950
C O N T R I B U T I O N S O F A U T H O R S
Katina Kardamanidis was the first author of the protocol conducted the literature search extracted full studies assessed the studies
according to inclusion criteria assessed quality of the studies compared findings with the second author discussed disagreements with
the third author wrote the drafts and the final review
Rebecca Ivers was the second author of the protocol assessed the studies according to inclusion criteria assessed quality of the studies
compared findings with the first author edited the final draft and provided overall guidance for the process
Alexandra Martiniuk assessed quality of the studies compared findings with the first author discussed disagreements with the third
author and helped write the first draft and the final review
Mark Stevenson contributed to the protocol reviewed methods of studies and contributed to the final draft
Katrina Thistlethwaite conducted a second literature search extracted full studies and contributed to the final draft
D E C L A R A T I O N S O F I N T E R E S T
None known
S O U R C E S O F S U P P O R T
Internal sources
bull No sources of support supplied
External sources
bull National Health and Medical Research Council of Australia Australia
Salary support for Martiniuk Ivers and Stevenson
bull University of Sydney Sesqui Research Grant Australia
Direct research costs and salary support for Kardamanidis
D I F F E R E N C E S B E T W E E N P R O T O C O L A N D R E V I E W
There are no differences between the protocol and the review
I N D E X T E R M S
Medical Subject Headings (MeSH)
lowastMotorcycles Accident Prevention [lowastmethods] Accidents Traffic [lowastprevention amp control] Licensure Program Evaluation Wounds
and Injuries [mortality lowastprevention amp control]
47Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 5050
MeSH check words
Humans
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 1050
Table 2 Risk of allocation bias
Anderson 1980 Allocation concealment unclear
Hall 1988 Allocation concealment unclear
Swaddiwudhipong 1998 Allocation concealment unclear
Table 3 Risk of detection bias
Anderson 1980 Blinding of outcome assessors not reported
Driver records holding police recorded crashes and offences not validated
Hall 1988 Blinding of outcome assessors not reported
Driver records holding police recorded crashes and offences not validated
Swaddiwudhipong 1998 Blinding of outcome assessors not reported
Police and hospital records Post mortem examination for deaths not validated
Adams 1985 Blinding of outcome assessors not reported
Australia Post crash records not validated but argued that underreporting is unlikely
Leung 1987 Blinding of outcome assessors not reported
Driver records holding police recorded crashes and offences
Raymond 1979 Blinding of outcome assessors not reported
Self-report through personal interviews or questionnaires - both methods validated
Osga 1980 Blinding of outcome assessors not reported
Self-report through structured questionnaires validated by cross-checking with police
recorded crash records
Satten 1980 Blinding of outcome assessors not reported
Self-report through structured interviews or questionnaires - both methods were not
validated
Intervention group were given telephone interviews but the control group mailed ques-
tionnaires
Jonah 1982 Blinding of outcome assessors not reported
Self-report through structured interviews not validated Driver records holding police
recorded crashes and offences
Mortimer 1984 Blinding of outcome assessors not reported
Self-report through questionnaires not validated
Intervention group received mailed questionnaires but the control group received ques-
tionnaires personally at the motorcycle dealership
Cooper 1988 Blinding of outcome assessors not reported
Driver records holding police recorded crashes and offences
8Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 1150
Table 3 Risk of detection bias (Continued)
Mortimer 1988 Blinding of outcome assessors not reported
Self-report through questionnaires not validated
Intervention group received mailed questionnaires but the control group received ques-
tionnaires personally at the motorcycle dealership
McDavid 1989 Blinding of outcome assessors not reported
Driver records holding police recorded crashes and offences
Kloeden 1994 Blinding of outcome assessors not reported
Driver records holding police recorded crashes and offences
Billheimer 1998 Blinding of outcome assessors not reported
Police recorded crashes and offences
Perrino 2002 Blinding of outcome assessors not reported
Self-report through mailed out questionnaires police records not validated
Kraus 1975 Blinding of intervention assessors to case or control status not reported
Intervention measured police records self-report though questionnaires not validated
Length of follow up not specified
Hurt 1981 Blinding of intervention assessors to case or control status not reported
Intervention measured face-to-face in-depth interviews (ie no structured question-
naires) Validity and reliability of data ensured by reported high level of quality controlLength of follow up not specified
Haworth 1997 Blinding of intervention assessors to case or control status not reported
Intervention measured self-report either by personal or by telephone interview for
both cases and controls not validated
Length of follow up in the 2 years prior to the interview or more than 2 years ago
Savolainen 2007 Blinding of outcome assessors not reported
Source of outcome data not reported
Length of follow up up to 5 years
Davis 1997 Blinding of outcome assessors not reported
Source of outcome data police records
Length of follow up 1 year
Waller 1992 Blinding of outcome assessors not reported
Source of outcome data Department of Public Safety driver history file and official
accident reports
Length of follow up 3 years
MAIDS 2004 Blinding of intervention assessors to case or control status not reported
Intervention measured face-to-face in-depth interviews (ie no structured question-
naires) No report on whether and how validity and reliability of data was ensured
Length of follow up 2 years
9Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 1250
Table 4 Risk of attrition bias
Anderson 1980 Percentage of participants in analysis 99 forthe total group Unclear forintervention andcontrol groups
separately
Hall 1988 Percentage of participants in analysis 96-97 forall intervention groups and 98for the control group
Swaddiwudhipong 1998 Percentage of participating sub-districts in analysis 100 for both intervention and control sub-districts
Raymond 1979 Percentage of participants in analysis 74 for the intervention group and less than 57 for the control
group
Osga 1980 Percentage of participants in analysis 73 for the intervention group and 100 for the control group
Satten 1980 Percentage of participants in analysis is unknown Initial sample size is unclear
Jonah 1982 Percentage of participants in analysis 23 for the intervention group and 45 for the control group
Mortimer 1984 Percentage of participants in analysis 35 for the intervention group and 84 for the control group
Cooper 1988 Percentage of participants in analysis 98 for the total group Unclear for intervention and control
groups separately
Mortimer 1988 Percentage of participants in analysis 40 for the intervention group and 90 in the control group
Adams 1985 Percentage of participants in analysis 61 forthe total group Unclear forintervention andcontrol groups
separately
Leung 1987 Percentage of participants in analysis unclear because initial sample size was not reported
McDavid 1989 Percentage of participants in analysis 40 of final matched cohort
Kloeden 1994 Percentage of participants in analysis 39 for the intervention group and 6 for the control group
Billheimer 1998 Percentage of participants in analysis unclear percentage of the initial sample size
Perrino 2002 Percentage of participants in analysis unclear percentage of the initial sample size
Kraus 1975 Percentage of participants in analysis 58 of cases 59 of controls
Hurt 1981 Percentage of participants in analysis 18 of cases lt 29 of controls
Haworth 1997 Percentage of participants in analysis 70 of cases 47 of controls
Savolainen 2007 Percentage of participants in analysis 18 for the intervention group and 15 for the control group
Davis 1997 Percentage of participants in analysis 100 for the intervention group and unclear for the control group
10Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 1350
Table 4 Risk of attrition bias (Continued)
Waller 1992 Percentage of participants in analysis unclear percentage for the intervention group and 100 for the
control group
MAIDS 2004 Percentrage of participants in analysis unclear for both cases and controls
R E S U L T S
Description of studies
See Characteristicsof included studies Characteristics of excluded
studies Characteristics of ongoing studies
Our search resulted in 102 studies that seemed relevant of which
we excluded 79 studies because they did not meet the criteria for
considering studies for this review (see Characteristics of excluded
studies)
Basedon ourcriteriawe includeda total of 23 studies in this review
There were three randomised trials (Anderson 1980 Hall 1988
Swaddiwudhipong 1998) two non-randomised trials (Adams
1985 Leung 1987) 14 cohort studies (Raymond 1979 Osga
1980 Satten 1980 Jonah 1982 Mortimer 1984 Cooper 1988
Mortimer 1988 McDavid 1989 Waller 1992 Kloeden 1994
Davis 1997 Billheimer 1998 Perrino 2002 Savolainen 2007)
and four case-control studies (Kraus 1975 Hurt 1981 Haworth
1997 MAIDS 2004)
Over half of the studies were conducted 20 or more years
ago and the oldest was conducted 31 years ago Only three
studies were from this decade (Perrino 2002 MAIDS 2004
Savolainen 2007) Over 60 of the studies were conducted in
the USA (Kraus 1975 Anderson 1980 Osga 1980 Satten 1980
Hurt 1981 Mortimer 1984 Leung 1987 Hall 1988 Mortimer
1988 Waller 1992 Davis 1997 Billheimer 1998 Perrino 2002
Savolainen2007) three inAustralia (Adams 1985 Haworth 1997
Kloeden 1994) three in Canada (Jonah 1982 Cooper 1988
McDavid 1989) onefromthe United Kingdom (Raymond 1979)one from Europe (with sampling areas located in France Ger-
many Netherlands Spain and Italy) (MAIDS 2004) and one
in Thailand (Swaddiwudhipong 1998) Only seven studies were
published in peer reviewed journals (Kraus 1975 Jonah 1982
Mortimer 1984 Cooper 1988 Mortimer 1988 McDavid 1989
Swaddiwudhipong 1998) The other studies were reports forhigh-
way safety and road traffic organisations
The studies by Anderson 1980 Hall 1988 Kloeden 1994
Billheimer 1998 (the basic California Motorcyclist Safety Program
component of the study) and Perrino 2002 examined manda-
tory pre-licence training Cooper 1988 McDavid 1989 and
Swaddiwudhipong 1998 assessed non-mandatory training that
when passed resulted in a licence Raymond 1979 Osga 1980
Satten 1980 Jonah 1982 Mortimer 1984 Adams 1985 Leung
1987 Mortimer 1988 Waller 1992 Davis 1997 Billheimer 1998
(the Experienced Rider Course component of the study) andSavolainen 2007 assessed non-mandatory training that could be
taken by motorcycle riders with and without a licence The case-
control studies (Kraus 1975 Hurt 1981 Haworth 1997) assessed
lsquoanyrsquo type of rider training and therefore could not be grouped by
pre- or post-licence rider training The MAIDS 2004 case-control
study did make a distinction between participants having received
compulsory pre-licence training or additional (non-compulsory)
training We did not find a study that compared one form of pre-
or post-licence training (eg on-road off-road theory or a com-
bination) versus another form of training All studies compared
eithermandatory or non-mandatory formal training versus no for-
mal training with the exception of Haworth 1997 who compared
different levels of training courses (beginner intermediate and ad-vanced)
We have presented a more detailed description of the individual
studies in Characteristics of included studies We have provided
a description below of the types of motorcycle rider training that
were evaluated in the included studies
Types of motorcycle rider training evaluated in the
studies
The types of training were varied Duration of training ranged
from two or three hours to 24 hours administered over two or
three days Most training consisted of a theoretical and a practical
component usually off-road The content ranged from provid-ing riders with manuals lectures discussions showing films and
administering knowledge and skill tests Little information was
provided about the qualifications of trainers who conducted the
training
Six studies (Osga 1980 Satten 1980 Mortimer 1984 Mortimer
1988 Waller 1992 Savolainen 2007) assessed the effectiveness of
the USA-based Motorcycle Safety Foundationrsquos Motorcycle Rider
Course (MSF MRC) This course was released in 1976 It com-
prises eight hours of theory (lectures discussions films and multi-
plechoice tests) a textbook and 12 hours of off-road practice The
Motorcycle Safety Foundation is a national not-for-profit organ-
11Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 1450
isation sponsored by manufacturers and distributors of motorcy-
cles Many USA states have adopted the MSF Basic Rider Courseand Experienced Rider Course (wwwmsf-usaorg) since its release
in 1976
The Motorcycle Operators Skill Test was also developed
by the USA Motorcycle Safety Foundation in coopera-
tion with others (httpwwwwebbikeworldcommotorcycle-
trainingmotorcycle-test) The MOST is a licensing test consist-
ing of a manual knowledge and skill test (comprising eight exer-
cises of increasing difficulty and skill requirements) and a mini-
skill test conducted with a simulator Some studies compared this
alternative licensing test to the standard licensing test However
as the MOST is not a type of rider training we have included only
the two studies that assessed the MOST in combination with one
or more training components in this review Both Anderson 1980and Hall 1988 assessed the MOST in combination with a train-
ing component of three hoursrsquo duration consisting of instructions
about the skills needed to pass the MOST (30 minutes of the-
ory and two and a half hours on the motorcycle) Hall 1988 also
assessed the MOST in combination with the Motorcycle Safety
Foundationrsquos Motorcycle Rider Course as described above
Other USA studies evaluated the Wisconsin motorcycle rider
course a 22-hour off-street programme that teachesthe basic rules
of safe motorcycling (Leung 1987) the USA Maryland Motorcy-
cle Rider Course (MRC) a mandatory (for those younger than 18
years)pre-licence 20-hour coursecomprising eight hours of theory
and12 hours of practice and created by the Maryland Motor Vehi-
cle Administration in 1985 (Perrino 2002) the California Motor-cyclist Safety Program (CMSP) a mandatory (for those younger
than 18 years) pre-licence 16-hour basic course (Billheimer 1998)
and Experienced Rider Course (ERC) of eight hoursrsquo duration
(Billheimer 1998) The CMSP still offers these courses ( http
wwwca-msporg) but they are now MSF courses as described
above Davis 1997 did not give any information about the assessed
rider training
The Canadian studies evaluated the British Colombia Safety
Councilrsquos 37-hour motorcycle safety programme (no further in-
formation on the content was provided) (McDavid 1989) and
the Canada Safety Council Motorcycle Training Program (MTP)
comprising four hours of theory and 16 hours of practice (six
hours on-road) over twoweekends (Jonah 1982) The MTPcoursewas made available in 1974 and is still being offered in Canada
(httpwwwsafety-councilorgtrainingmtpmtphtm) Cooper
1988 did not give any information about the assessed rider train-
ing However the study was conducted in British Colombia in
Canada in 1988 the same time as the McDavid 1989 study was
conducted
In the United Kingdom Raymond 1979 assessed the non-manda-
tory RACACU (Royal Automobile ClubAuto-Cycle Union)
training scheme The training programme consisted of either 24
one-hour sessions or 13 two-hour sessions Twelve hours of the
training focused on theory about basic operational procedures
traffic laws and highway codes The practical session focused on
basic operational skills riding on a training area and riding in traf-fic with an instructor A proficiency test was administered at the
end of the course
One study from rural Thailand (Swaddiwudhipong 1998) evalu-
ated a community-based rider education programme consisting of
theory on motorcycle injury risk helmet use traffic laws vehicle
regulations and traffic signs and a written and skill test leading to
a licence
TheAustralian studies evaluateda rider trainingcourse of 12 hours
over two consecutive days (four hours of theory and eight hours of
practice)administeredto AustraliaPosttelegram deliverers(Adams
1985) and a mandatory pre-licence Ridersafe course consisting
of two tiers the first tier being a training course on the basics
of motorcycle riding where a pass entitled the rider to obtain alearnerrsquos permit and the second tier being a skill test where a pass
entitled the riderto obtain a probationary motorcycle licence The
training course consisted of two four-hour training sessions on
separate days involving instruction videos and actual riding on
motorcycles The skill test consisted of a four-hour preparation
course of 74 minutes of lecturediscussion 38 minutes of video
and 75 minutes of on-course riding which was the testing phase
(Kloeden 1994) The Rider Safe course is still being provided by
Transport South Australia a state government organisation (http
wwwtransportsagovau )
The duration or content of rider training in the four case-control
studies (Kraus 1975 Hurt 1981 Haworth 1997 MAIDS 2004)
was not described
Risk of bias in included studies
We describe the methodological quality of the included studies
by study type below More detailed information can be found in
Table 1 Table 2 Table 3 and Table 4
When considering the findings from these studies it is important
to note the quality (or lack thereof) of reporting of the studies
Quality of reporting is the extent to which a report provides in-
formation about the design conduct and analysis of a study ( Juumlni2001) Huwiler-Muumlntener 2002 found that even though trials
may be well reported it does not mean that the methodological
quality is of equal standard and well-conducted trials may never-
theless be reported badly The reporting quality of most studies in
this review was poor Most studies were old and did not meet con-
temporary reporting standards This made data extraction com-
plicated It also made it difficult to contact authors for additional
information
Because of significant heterogeneity between studies as well as the
poor quality of most studiespooling of studyresults to calculatean
overall measure of effect was not appropriate Instead we provided
a description of our judgement of risk of bias in the text
12Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 1550
Risk of bias in the randomised trials
There were three randomised trials (Anderson 1980 Hall 1988Swaddiwudhipong 1998) Anderson 1980 and Hall 1988 had an
initial sample size of 41188 and 26464 motorcycle licence ap-
plicants respectively Both studies assessed the effectiveness of the
pre-licence MOST with a trainingcomponent Swaddiwudhipong
1998 evaluated the effectiveness of a community-based rider ed-
ucation programme in six districts
Allocation bias
Anderson 1980 reported using random birth date tables to allocate
participants to either the intervention or control group Hall 1988
reported the use of a computerised algorithm set in a central of-
fice but also used manual random assignment when the computer
went down Swaddiwudhipong 1998 reported lsquosimple random
samplingrsquo for the delivery of the intervention where the sampling
frame included nine districts Three districts were excluded since
they were too hard to access The remaining six districts were ran-
domly allocated to the intervention (three) and the control (three)
groups Systematic sampling wasthen used ordered by village size
for the villages within districts Allocation concealment was not
reported in any of the studies The descriptions of randomisation
in these studies suggest that there may have been limitations to
the process resulting in a lack of equivalence of the intervention
and control groups at baseline and therefore the possibility of
allocation bias Swaddiwudhipong 1998 reported that before theintervention the annual incidence rate of motorcycle-related in-
juries was slightly (but not statistically significant) higher in the
intervention areas than in the control areas Participants in the
intervention districts were also more likely to have a valid licence
and to wear a helmet more often than participants in the control
districts Anderson 1980 measured differences at baseline on the
participant and prior driving record variables and found these to
be negligible Hall 1988 did not report on baseline differences
between the intervention and control groups As the Anderson
1980 and Hall 1988 studies had sample sizes of over 40000 and
26000 respectively and used some form of randomisation it is
likely that any confounding factors at baseline would have been
equally distributed between comparison and intervention groups
Selection bias
After randomisation Anderson 1980 and Hall 1988 excluded
08 and 3 participants respectively because of assignment to
the wrong group administration of the wrong intervention or
other reasons The resulting lack of analysis by rsquointention-to-treatrsquo
can introduce biasby removing the effectivenessof randomisation
Nonetheless the percentage excluded was small and the number
of overall participants large so it is unlikely to have had a major
effect on the final results In the Swaddiwudhipong 1998 study
villages were systematically sampled into intervention or control
groups The intervention was implemented and outcomes of in-
jury and death were measured at a community level Interventionand control districts were randomly selected however there were
only three intervention and three control communities
Detection bias
None of the randomised trials reported blinding of outcome as-
sessors Data collection tools used by Anderson 1980 and Hall
1988 were driver records holding police recorded crashes and of-
fences Swaddiwudhipong 1998 used police and hospital records
at a district level and post-mortem examination for deaths None
reported on validity or reliability for any of the instruments
Attrition bias
Anderson 1980 and Hall 1988 had driver records available for
96 to 100 of participants They also collected self-reported
data but the response rate to the questionnaires was very low (low-
est response rate 42 and 30 respectively) Attrition bias in
Swaddiwudhipong 1998 was not an issue as outcome data were
available for all sub-districts included in the trial
The intervention reported in Anderson 1980 and Hall 1988 tri-
als had an unanticipated effect on the participants In both trials
an average of only 28 of licence applicants in the intervention
group eventually obtained their licence compared to 35 in the
control group This indicates that the licensing process in generalseems to have deterred applicants from obtaining a licence and
although no statistical tests of the difference were conducted this
deterrence appears even greater for those who had rider training
This is not true attrition bias as the studies did not experience
differential loss to follow up from outcome measurement Instead
differential attainment of the intermediary step of licensure pre-
cluded participants from being at risk for the outcomes of interest
they are not driving a motorcycle hence they are not at risk for
motorcycle-related crashes injury death or offences In short it is
not possible to assess whether a reduction in crashes in the inter-
vention group would be due to changes in riding habits as a result
of training or simply due to delayed or never attained licensure
Intervention integrity
Neither Anderson 1980 nor Hall 1988 reported on the consis-
tency in the way the intervention was administered the possible
lsquocontaminationrsquo of participants in the intervention group by other
types of formal or informal rider training or of participants in the
control group by the intervention Swaddiwudhipong 1998 did
not report on the number of villagers in each group that actually
attended the intervention However they did report on a number
of other potential training courses that were being implemented
at the same time
13Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 1650
Analysis
Anderson 1980 and Hall 1988 acknowledging the lsquodeterrent ef-fectrsquo described above conducted two separate analyses The first
analysis included the total study population of licence applicants
over two years from the time of their application (the population
denominator) The second analysis was performed on the licensed
(intervention and control) subgroups only two years from the
time of obtaining their licence (the licensed rider denominator)
As only a third were licensed and exposed to riding the motorcy-
cle it was impossible to determine whether reported differences
in crash rates across groups were due to the intervention or to
other lsquoconfoundingrsquo characteristics of those who were determined
in their efforts to obtain the licence When analysis was restricted
only to those who gained their licence the same difficulties re-
mained The benefits of the randomisation process were thus lost
due to this lack of licensure of riders during the process As there
were only three intervention and three control groups in the study
reported by Swaddiwudhipong 1998 analyses were hindered by
smallnumbers for outcome comparisons betweenthe intervention
and the control districts
Risk of bias in the cohort studies and non-randomised
trials
We discuss the methodological quality of cohort studies (seven
prospective studies (Raymond 1979 McDavid 1989 Waller 1992
Kloeden 1994 Billheimer 1998 Perrino 2002 Savolainen 2007)and seven retrospective studies (Osga 1980 Satten 1980 Jonah
1982 Mortimer 1984 Cooper 1988 Mortimer 1988 Davis
1997)) and non-randomised trials (Adams 1985 Leung 1987) in
the one section as the biases for the two study types are similar
The quality of these studies was in general poor
Selection bias
The confounding factors expected to have been addressed in the
studies were sex age riding experience prior driving records (of-
fences crashes) previous training and the following proxies for
risk taking behaviour helmet use speed and alcoholdrug use All
of these factors have been shown to be associated with an increasedrisk of motorcycle crash (Peden 2004 Zambon 2006) The pos-
sible source of bias in cohort studies and non-randomised trials
is that groups may differ on important characteristics at baseline
participants lsquoself-selectrsquo into either the rider training group by de-
ciding to undertake a rider training course or into the lsquocontrol
grouprsquo by deciding not to undertake a course Participants who
undertake a non-mandatory rider training course may differ from
those who choose not to on characteristics such as age sex rid-
ing experience and riding behaviour as well as other possible con-
founders For example riders who take a training course may be
more safety-conscious or on the contrary may do so because they
have previously experienced a crash or because they perceive that
they have poor riding skills Therefore ideally all participants in
the intervention group need to be included in the study as wellas a random sample of controls taken from the same population
as the intervention group If that is not possible intervention and
control groups either need to be carefully matched to make them
comparable or differences at baseline need to be adjusted for in
the analysis
Most of the studies using a non-randomised trial or cohort design
compared an intervention group to a control group that differed
on a number of important factors The studies in general failed to
adjust for important differences found at baseline and their results
cannot be meaningfully interpreted Ten studies (Raymond 1979
Osga 1980 Satten 1980 Jonah 1982 Mortimer 1984 Mortimer
1988 McDavid 1989 Waller 1992 Davis 1997 Savolainen 2007)
assessed intervention and control groups that were not drawnfrom the same population In most of these cases the intervention
group was selected from a list of training graduates who were usu-
ally young female and less experienced riders whereas controls
were recruited from a general population of motorcycle riders who
were usually older male and experienced riders For three stud-
ies it was unclear how they were selected (Leung 1987 Cooper
1988 Perrino 2002) Only Kloeden 1994 recruited groups from
the same population of motorcycle applicants and Adams 1985
from a group of professional riders Nevertheless due to non-ran-
dom selection of intervention and control groups differences at
baseline are likely even though study groups were from the same
population Three studies adjusted for some confounding fac-
tors found at baseline through matching (McDavid 1989 Waller1992 Billheimer 1998) However matching has limitations in
termsof controlling for multiple confounders especially those that
are unknown to influence the effect of rider training on the out-
comes One study matched the control and intervention groups
by excluding controls (approximately 3500 (66 of initial con-
trol group) that differed from the intervention group) (Kloeden
1994) Five studies (Osga 1980 Satten 1980 Adams 1985 Davis
1997 Perrino 2002) did not adjust for differences between study
groups at baseline either in the study design or analysis Results
from these studies are therefore not reliable The other six stud-
ies (Raymond 1979 Jonah 1982 Mortimer 1984 Cooper 1988
Mortimer 1988 Leung 1987 McDavid 1989 Billheimer 1998)
adjusted for some differences between groups at baseline by eitherbi-variate or multi-variate analyses or stratification and standardi-
sation of riding exposure (outcomemiles driven) or stratification
by age group The limitation of stratification is that it is difficult to
control for many variables simultaneously because a large number
of strata will need to be generated resulting in small numbers that
cannot be meaningfully analysed Savolainen 2007 developed sta-
tistical models to assess the probability of a crash after taking the
rider training course in an intervention and control group while
examining differences by sub-groups (eg age) Participation rates
were also very low in this study with only 5584000 (1395)
in the eligible intervention group and 5884000 (147) for the
14Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 1750
control group completing surveys
Detection bias
None of the studies reported blinding of outcome assessors to the
intervention or control status of the participants Seven studies
used driver records containing police recorded data as their main
measure of outcome (Leung 1987 Cooper 1988 McDavid 1989
Waller 1992 Kloeden 1994 Davis 1997 Billheimer 1998) An-
other four studies (Raymond 1979 Satten 1980 Mortimer 1984
Mortimer 1988) only used self-report Raymond 1979 was the
only study using self-report as the main outcome measurement in
which study instruments were validated Self-report might prevent
the underreporting that has been described for the use of police-
driver records (see rsquoRisk of biasrsquo in the randomised trials above)but has other risks of bias especially if questionnaires or inter-
view methods have not been validated and tested for reliability
Recall bias arises from mistakes in recollecting events both be-
cause of failures of memory and because of looking at things lsquowith
hindsightrsquo and possibly changed views (httpwwwcochraneorg
resourcesglossaryhtm) As many studies were retrospective risk
of recall bias is likely and may have increased with length of follow
up For studies that used self-reported outcomes only it is also
possible that social desirability response bias played a role where
participants in the training group may have felt it was lsquoexpectedrsquo
to report good driving histories Multiple methods to measure the
same outcome for both groups decreases the risk of detection bias
for example because self-reported answers can be cross checkedwith other sources of information such as driver records
In this regard three studies (Osga 1980 Jonah 1982 Perrino
2002) used both self-report and driver records Osga 1980 found
that for both intervention and control groups more crashes were
reported through self-report than on state crash records an aver-
age of 54 of self-reported crashes could be found back on state
crash records One industry-based study (Adams 1985) used oc-
cupational health records which were not validated but which are
unlikely to suffer from underreporting due to work cover claims
Savolainen 2007 used information on rider crashes for a five-year
period but did not specify from where this information was ob-
tained
Attrition bias
The majority of studies (Raymond 1979 Satten 1980 Adams
1985 Leung 1987 Kloeden 1994 Davis 1997 Billheimer 1998
Perrino 2002) inadequately reported on losses of participants to
the study We can therefore not be certain that there were no sys-
tematic differences in the characteristics between the intervention
and control group participants Savolainen 2007 reported the re-
sponse rates to their baseline survey but did not report on the per-
centage of outcome data that was available for their study sample
A similar deterrent effect of rider training as described for the ran-
domised trials (Anderson 1980 Hall 1988) occurred in two stud-
ies Jonah 1982 reported that 65(14992310) of the trained rid-
ers stopped riding compared to 47 (9492029) of the untrainedriders They excluded non-active and non-licensed riders before
the study commenced so the differential choice to continue riding
after training did not affect their analyses Leung 1987 reported
that the percentage of participants in the intervention group ob-
taining their licence was 50 (no number provided by authors)
lower than in the control group This may have biased the mea-
surement of outcomes because unlicensed participants will not be
exposed to riding the motorcycle and therefore not be at risk of
being involved in a crash or violation As explained above this
makes it difficult to assess whether the reduction in crashes in the
intervention group is due to a change in driving habits or simply
due to delayed licensure
Mortimer 1984 and Mortimer 1988 describe that about 44 and30 of course graduates respectively were excluded from analysis
because they had not ridden a motorcycle in the previous year
However because these studies were retrospective and the controls
were activeriders at the time of recruitment a decision on whether
the deterrent effect was differential between the two groups could
not be made Though not reported quantitatively Kloeden 1994
also describes the effectof rider training deterringindividuals from
becoming licensed and hence exposure to riding and risk of crash
death or offences involving a motorcycle Osga 1980 reported
that 27 of the trained riders did not ride after the course and
excluded these from the analysis
Use of different data collection strategies may have contributed to
differential loss to follow up in the cohort studies Satten 1980for example interviewed the intervention group over the tele-
phone but mailed questionnaires to the control group Mortimer
1984 and Mortimer 1988 distributed questionnaires in person to
the control group but mailed them to the intervention group
This may have played a role in the observed differential response
rates where more subjects responded in the personally approached
groups
Survival bias
For those studies that assessed offences crashes and injuries but
not death survival bias may have also played a role Motorcyclists
who were fatally or seriously injured in crashes during the follow-
up period after the intervention would have been less likely or
unable to respond to surveys This could bias the effect measures
towards or away from the null None of the studies wrote about
this issue or examined the characteristics of those lost to follow up
in order to address the possibility of survival bias
Intervention integrity
Although the interventions were intended for the entire lsquoexposedrsquo
groups most studies reported difficulty in tracking participants
whoreceivedthe interventionand those whodid not Adams 1985
15Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 1850
estimatedthat 62of thetotalstudysample received the interven-
tion The remaining studies did not report the percentage of thestudy population that received the intervention as was intended
or reported that this percentage was unknown (Raymond 1979
Kloeden 1994 Satten 1980 Jonah 1982 Mortimer 1984 Leung
1987 Cooper 1988 Mortimer 1988 Billheimer 1998 Perrino
2002 Savolainen 2007) Only Raymond 1979 reported the degree
to which their intervention was delivered as intended No studies
reported on possible lsquocontaminationrsquo of participants by other inter-
ventions than the one to which they were allocated Because Osga
1980 Satten 1980 Jonah 1982 Mortimer 1984 Cooper 1988
Mortimer 1988 Waller 1992 and Davis 1997 were retrospective
studies and participants were recruited from a list of graduates it
was known that they had completed and passed the course
Analysis
In some studies some participants who reported at baseline that
they were untrained decided to enroll in a rider training pro-
gramme after the baseline survey The non-randomised trial by
Adams 1985 experienced this move of subjects from the untrained
to the trained group but failed to conduct analyses by lsquointention-
to-treatrsquo This can bias the results in randomised and non-ran-
domised trials if participants who changed groups differ signifi-
cantly with regard to potentially confounding characteristics On
the other hand Perrino 2002 also experienced this move of sub-
jects from theuntrained to the trained groupmore than 100riders(5 of the total sample) originally stated that they were untrained
at baseline but then received training between the baseline and
follow-up surveys Yet since the Perrino 2002 study was a cohort
their analyses then proceeded by including those riders who were
lsquonewlyrsquo trained as exposed Riders who were lsquonewlyrsquo trained since
baseline but with a programme other than the one under study
were excluded from the analysis Nevertheless confounding fac-
tors were not accounted for in the analyses
McDavid 1989 matched exposed with unexposed subjects but
could not find suitable matches for the majority (60) of their
subjects They could therefore only conduct analysis on the result-
ing40 (139346) which is an inappropriate andinefficient anal-
ysis Some studies that reported on the issue of non-active licensedriders(Satten 1980 Jonah 1982 Mortimer 1984 Mortimer 1988
Kloeden 1994) excluded non-active riders from analysis intro-
ducing selection bias as it is likely that riders who continued to
ride differed with regards to potentially confounding character-
istics across intervention and control groups Kloeden 1994 for
example conducted several analyses to assess the effect of various
sub-sets of the control group Original analyses included all con-
trols and led to erroneous results since only a small proportion of
controls hadgone on to get a licence after their learnerrsquos permitand
were actually exposed to the possibility of experiencing a crash
Subsequent analyses compared only intervention and control sub-
jects who had a licence Waller 1992 reported crashes by person-
days of exposure but did not explain how they could be sure that
each study intervention and control participant was an active riderand actually exposed to the risk of experiencing a crash
Potential confounders that are unequally distributed by study arm
at baseline need to be dealt with in the analysis Satten 1980
Adams 1985 and Davis 1997 did not take into account any con-
founding factors either by design or in the analysis McDavid
1989 Waller 1992 and Billheimer 1998 accounted for potential
confounding factors at the design stage by matching the subjects
in the intervention and control groups Other studies (Raymond
1979 Mortimer 1984 Mortimer 1988 Perrino 2002 and also
Waller 1992) adjusted for one or two confounders by stratifica-
tionor standardisation (eg crashesmiles driven age group) Osga
1980 conducteda univariate analysis of crash rates in which no po-
tential confounders were accounted for but also calculated cumu-lative crash rates by mileage travelled Only Jonah 1982 Cooper
1988 and Savolainen 2007 conducted a multivariate analysis
Risk of bias in the case-control studies
A case-control design is suitable for the assessment of risk fac-
tors for outcomes such as motorcycle crashes injuries or deaths
but not appropriate for the evaluation of an intervention such
as motorcycle rider training It can potentially result in spurious
conclusions regarding the efficacy of the intervention because the
intervention has not been applied at random to an intervention
and control group Even with the control of possible confounders
using a case-control design to evaluate an intervention can lead tothese kinds of results since those who choose to undertake rider
training may differ on important prognostic factors that affect the
outcome compared to those who do not choose to participate in
rider training
Selection bias
A potential source of bias in case-control studies as with all other
study types where the intervention is not randomly allocated to
participants is that comparison groups are likely to differ on po-
tentially confounding characteristics This means that ideally all
cases need to be included in the study as well as a random sample
of controls taken from the population that gave rise to the cases If that is not possible cases and controls either need to be carefully
matched to make themcomparable or differences at baseline need
to be adjusted for in the analysis All four case-control studies were
population based Kraus 1975 included all fatalities and injured
cases leading to hospitalisation and appropriately recruited a ran-
dom sample of controls from the same population Hurt 1981 did
not manage to recruit all or a random sample of cases Only half
of the notifications of crashes to the police were received in time
by the researchers the research team for unreported reasons was
only able to respond to half of those again Data collection was
therefore completed for only about a fifth of all cases Neverthe-
less this was still a large sample of 900 cases Haworth 1997 using
16Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 1950
a similar study design was more likely to have included all cases
but this is not entirely clear from the study reported It is unclearfor example whether fatalities were included as they did not use
coronerrsquos reports for the recruitment of cases The MAIDS 2004
study was conducted in five European countries but it was unclear
how cases were recruited whether they were all cases during the
two years that the study lasted or a random sample of all cases
what the response rates were etc Controls were recruited from
randomly selected petrol stations but it was likewise not clear what
the eligibility criteria for inclusion in the study were how they
were recruited (eg randomly) and what the response rate was
Hurt 1981 and Haworth 1997 both matched controls to cases by
recruiting themat the sametime ofdayandsamedayof weekas the
crash occurred Hurt 1981 however recruited controls two years
later andit is likelythat there would have beendifferences betweenthe two comparison groups One important difference was that
during data collection for cases there was very little specialised
motorcycletrainingavailable in the studyregionwhereas two years
later when control data were collected there was some training
available
These four case-control studies also suffered from selection bias
due to potential differences between those that gave consent to
participate in the study and responded to interviews andquestion-
naires and those that did not None of the studies have assessed
such differences although Hurt 1981 did acknowledge that con-
trols giving consent to cooperate may have differed from those
motorcyclists who did not
Detection bias
Measures that can be taken to reduce detection bias in case-con-
trol studies are blinding of interviewers for case or control status
blinding of the participant for the study factor of interest validat-
ing measurement instruments and using these in an identical way
for cases and controls and cross checking self-reported informa-
tion with other more objective sources such as registers None of
the studies reported blinding of the intervention assessors to case
or control status but all used consistent measurement methods for
both case and control groups
All case-control studies used self-report as their main measure toassess whether cases and controls had received rider training and
if so the type of training The major potential source of bias in
case-control studies when self-report is used as a measure is recall
bias One way of reducing recall bias is by double-checking self-
reported data with more objective data Kraus 1975 also used po-
lice accident reports for information on rider characteristics but
such information was not available about rider training Haworth
1997 and MAIDS 2004 did not report on any measures to limit
recall bias Hurt 1981 reported further questioning of cases by
the interviewer when answers did not make sense or when par-
ticipants seemed to contradict themselves as part of the in-depth
unstructured interview technique This qualitative research tech-
nique can result in valuable information but as interviewers were
not blinded to case or control status of the participants it may have led to more thorough examination of cases compared to the
controls and thus differential detection bias
Intervention integrity
The level of detail about the rider training provided in the case-
control studies differed significantly but was minimal for each
Kraus 1975 simply asked whether lsquoanyrsquo motorcycle rider training
had lsquoeverrsquo been carried out Hurt 1981 asked participants whether
they hadbeen self taught informally taught or completed formally
taught rider training courses MAIDS 2004 asked whether partic-
ipants had had pre-license training additional (non-compulsory)
training other training or no training whereas Haworth 1997asked about the number of courses that a participant had com-
pleted whetherthey were basic intermediate or advanced courses
and how long ago the training had been done None of the case-
control studies reported collecting information about further de-
tails about any training programmes (eg type of training cur-
riculum duration of training)
Analysis
Kraus 1975 accounted for confounding factors through the exclu-
sion of females and stratification by age and frequency of motor-
cycle use Hurt 1981 matched cases and controls for time of day
and day of the week that the crash occurred and did not adjust forany confounding factors in the analysis Haworth 1997 matched
groups for time of day and day of the week that the crash occurred
and stratified by age blood alcohol level and licence status in the
analysis None of the studies used multivariate methods MAIDS
2004 did not provide any information about any participant char-
acteristics other than the study exposures of interest and did not
describe any methods to adjust for potential confounding factors
Effects of interventions
We have presented the results in a descriptive manner due to the
substantial heterogeneity in study types interventions and out-comes and the generally poor quality of the studies We have
grouped the interventions by mandatory and non-mandatory pre-
licence rider training and post-licence rider training and further
by study type and by outcome (crashes injurious and fatal crashes
and offences) We discuss results that are reported as statistically
significant (whether a P value was reported or not)
Pre-licence rider training - mandatory
Randomised trials
17Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2050
Crashes
Total group of participants
Anderson 1980 found a reduction in motorcycle crash rates in
the intervention group (MOST skill test with remedial training)
compared to the control group (who received the skill testwithout
a training component) The percent differenceranged from 143
six months after the programme (P lt 005) to 211 (P lt 0001)
after one year and 127 (P lt 001) after two years Hall 1988
found no statistically significant difference in crash rates between
intervention groups at any point in time
Subgroup of participants who obtained their licence only
Anderson 1980 found a reduction in motorcycle crash rates in the
intervention arm but this was statistically significant at one year
only (24 to 218 reduction P lt 005) Hall 1988 found no
statistically significant differences
Offences
Total group of participants
The percentage of offences in Anderson 1980 after six months
and one and two years was slightly higher in the intervention
arm compared to the control group but these differences were
not statistically significant Likewise the study by Hall 1988 did
not find a statistically significant difference in offences between
groups
Subgroup of participants who obtained their licence only
Hall 1988 found no significant difference in offence rates between
the intervention and control groups Anderson 1980 found thatthe percentage of offences increased in the intervention arms com-
pared to the control group This difference was statistically signif-
icant after two years (51 to 115 increase P lt 005)
Cohort studies
Crashes
Neitherof the cohort studies (Kloeden1994 and Billheimer1998)
found a statistically significant difference in crash rates between
the intervention and control groups
Offences
Billheimer 1998 (for the California Motorcyclist Safety Program
course) and Perrino 2002 did not find a statistically significant
difference in offence rates between the intervention and control
groups
Case-control studies
Crashes
MAIDS 2004 found that a slightly higher percentage of controls
(not involved in a crash) had had pre-license rider training than
cases (involved in a crash) This difference was not statistically significant (50 of controls vs 477 of cases OR 091 95 CI
076 to 110)
Pre-licence rider training - non-mandatory
Randomised trial
Injuries and deaths
Swaddiwudhipong 1998 found a significantly lower injury ratein the intervention villages (1051000 population 95 CI 92
to 120) compared to the control villages (1691000 population
95 CI 152 to 185) one year after the intervention This differ-
ence was not maintained two years after the intervention (127
95 CI 112 to 143 intervention 151 95 CI 135 to 167
control) There was no significant difference in death rates be-
tween groups
Cohort studies
Crashes
Cooper 1988 foundno statistically significant differencein crashes
between the intervention and control groups McDavid 1989
found a lower crash rate in the intervention group (14139) com-
pared to the control group (23139) but no test of statistical sig-
nificance was undertaken
Injuries and deaths
McDavid 1989 found that the intervention group had fewer in-
jurious crashes (814 crashes) compared to the control group (16
23 crashes) but no statistical significance test was reported
18Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2150
Offences
Cooper 1988 foundno statistically significantdifference in offencerates between the intervention and control groups McDavid 1989
foundno statistically significantdifference in offence ratesbetween
the intervention and control groups
Non-mandatory rider training
Non-randomised trials
Crashes
Neither Adams 1985 nor Leung 1987 found a statistically signif-icant effect of rider training on crash rates
Offences
Leung 1987 found a statistically significant lower offence rate in
the intervention group compared to the control group for both
males and females and for all age groups except for women 35
years or older (P values not reported)
Cohort studies
Crashes
Osga 1980 in the univariate analysis reported a lower percentage
of crashesin theintervention group (that continued to ride) (28
168595) than in the control group (34 112329) In the cu-
mulative crash rate analysis however Osga1980 found statistically
significant (at the P lt 001 level) higher crash rates in the interven-
tion group than in the control group at allmileage intervals (1000
to 29999 miles) travelled After riding more than 30000 miles
(when over 50 of the trained riders had been involved in a crash
compared to approximately 40 in untrained riders) this differ-
ence was no longer statistically significant Satten 1980 found a
lower percentage of crashes in the intervention group (13 969)
compared to the control group (25 2079) but the differencewas not assessed statistically Davis 1997 found a lower crash rate
in trained riders but it was unclear how this was calculated be-
cause the size of the control group was not reported Jonah 1982
and Billheimer 1998 (for the Experienced Rider Course) found no
statistically significant difference in crashes between the interven-
tion and control groups Although the difference between those
who completed the Experienced Rider Course and no course in
Billheimer 1998 neared significance at six months after the course
(P = 007) large P values were observed for one year after (P =
030) Raymond 1979 found a statistically significant higher crash
rate for the intervention group (1861000 miles ridden) com-
pared to the control group (1181000 miles ridden) in both males
and females (P lt 0001) especially during the first 5000 miles of
riding Waller 1992 found a statistically significant higher crashrate for the intervention group (27 crashes in three years or 070
crashes10000 person days) compared to the control group (5
crashes in three years or 013 crashes10000 person days) with a
crude relative risk of 538 (X 2=1513 plt005) Mortimer 1984
and Mortimer 1988 found no statistically significant difference in
crashes between the intervention and control groups
Savolainen 2007 found that the intervention group (those who
completed the Basic Rider Course) was 44 more likely to be
involved in crashes than the control group and those who com-
pleted the course multiple times were an additional 180 more
likely to be involved in a crash This finding was based on low
numbers and no P value or confidence level was given for either
result
Injuries and deaths
Satten 1980 also found a lower percentage of injurious crashes
(33 39) in the intervention group compared to the control
group(60 1220) although again no statistical significance test
was reported
Offences
Jonah 1982 did find a significantly lower self-reportedoffence rate
in the intervention group (15) compared to the control group(34 P lt 001) (no numbers given) Satten 1980 also found
a lower self-reported offence rate in the intervention group (1
71 versus 19100) but again this was not tested for statistical
significance Mortimer 1984 Mortimer1988 and Billheimer 1998
found no statistically significant differencein offence ratesbetween
the intervention and control groups
Case-control studies
CrashesMAIDS 2004 found that a slightly higher percentage of cases
(involved in a crash) had had additional non-mandatory rider
training than controls (not involved in a crash) This difference
was not statistically significant (17 of cases vs 12 of controls
OR 146 95 CI 064 to 337)
Rider training unspecified
Case-control studies
19Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2250
Injuries and deaths
Hurt 1981 found that the odds of injurious crashes were lower inthose who had received professional training compared to those
who had not Kraus 1975 found that in the younger age group
(less than 20 years) and the older age group (over 35 years) a larger
percentage of injured riders (12 and 21 respectively) reported
prior motorcycle rider training compared with the uninjured con-
trol group (6 and 5 respectively) (P lt 001) (no numbers
given) Haworth 1997 reported several different outcomes one
beingthat having completed at leastone trainingcoursecompared
to no training courses at all had no significant effect on the odds
of crashing (afteradjustment for age blood alcohol level or licence
status (learner permitprobationaryfull)) When comparing three
different levels of training courses (beginner intermediate and ad-
vanced) with no training no significant differences were found
in the odds of crashing However after adjustment for age and
blood alcohol level completing an advanced course as compared
to a beginners course was associated with a significant decrease in
the odds of severe motorcycle crashes resulting in serious injury or
death (OR 02 95 CI 01 to 08)
D I S C U S S I O N
The objective of this review was to quantify the effectiveness of
pre- andpost-licence motorcycle rider training on the reduction of
traffic crash involvement injuries deaths and offences committedby motorcycle riders Most included studies were cohort studies
There were a few randomised and non-randomised trials and case-
control studies
All studies compared either mandatory or non-mandatory formal
training with no formal training except for Haworth 1997rsquos case-
control study that compared levels of training (beginner interme-
diate and advanced) to each other
We found studies that evaluated the effectiveness of manda-
tory pre-licence training in reducing crash and offence rates
(Anderson 1980 Hall 1988 Kloeden 1994 Billheimer 1998
(the basic California Motorcyclist Safety Program component of
the study) Perrino 2002) non-mandatory training that when
passed resulted in a licence (Cooper 1988 McDavid 1989
Swaddiwudhipong 1998) non-mandatory training that could be
taken by motorcycle riders with or without a licence (Raymond
1979 Osga 1980 Satten 1980 Jonah 1982 Mortimer 1984
Adams 1985 Leung 1987 Mortimer 1988 Waller 1992 Davis
1997 Billheimer 1998 (the Experienced Rider Course compo-
nent of the study)) and unspecified rider training on crash injury
and death rates (Kraus 1975 Hurt 1981 Haworth 1997 MAIDS
2004)
The studies provided conflicting evidence regarding motorcycle
rider training which is likely due to the poor design quality and
reporting of the studies Most of the studies included in the review
were likely to be adversely affected by significant selection biasalong with substantial detection bias(measurement and recall bias)
which limits our ability to make conclusive comments
Mandatory pre-licence rider training
There were two RCTs three cohort studies and one case-control
study that examined pre-licence ridertraining The two RCTs eval-
uated the Motorcycle Skills Operating Test (MOST) (Anderson
1980 Hall 1988) Anderson 1980 found a statistically significant
reduction in crashes in the intervention group at six months one
and twoyearsfollow up However when only participantswho ob-
tainedtheirlicence were included in theanalysis thereduction wasstatistically significant only at one year follow up The other RCT
(Hall 1988) found no statistically significant differences in crash
rates at any point of time neither in the total population group
nor in the licensed subgroup Neither of the two cohort stud-
ies assessing crash rates found a statistically significant difference
in crash rates between intervention and control groups ( Kloeden
1994 Billheimer1998) andthe case-control study (MAIDS 2004)
also did not find a statistically significant difference in the odds of
experiencing a crash between trained and untrained riders With
regards to offences neither of the RCTs found a statistically sig-
nificant difference in offence rate between the intervention and
control groups in the analysis of the total group of participants
entering the trial However Anderson 1980 found a statistically significant increase in offence rates in the intervention group but
only after two years of follow up andonly in the analysis of partici-
pants who obtained their licence The two cohort studies assessing
offence rates found no statistically significant difference between
the groups (Billheimer 1998 Perrino 2002)
None of the five studies (RCTs or cohort) found a reduction in
crashes or offences following mandatory pre-licence rider training
except for Anderson 1980 who observed a statistically significant
reduction in crashes in the intervention group at one year in both
analyses It is not clear why both RCTs that evaluated the same
MOST and training intervention observed different results Both
studies enrolled large samples and observed a similar uptake of
licensure following the intervention It is possible that differencesmay have arisen due to different study samples (California ver-
sus New York) or that the intervention delivery differed between
studies
In both trials of the Motorcycle Skills Operating Test including
rider training only a third of motorcycle licence applicants ob-
tained their motorcycle licence It is possible that a licensing pro-
cess including a skills test may reduce crashes through decreased
exposure
Non-mandatory pre-licence training
20Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2350
Three studies assessed the effect of non-mandatory pre-licence
training The Cooper 1988 cohort study found no difference incrash or offence rates between the trained and untrained groups
We cannot however be confident about this result as the study
sufferedfrom selectionbiaswhichcouldhave biasedthe measure of
effect toward the null The RCT conducted by Swaddiwudhipong
1998 examined injury and death rates and found a statistically
significant lower injury rate in the intervention villages after one
year of follow up but no longer at two year follow up There
was no statistically significant difference in death rates between
intervention and control villages Our confidence in the results of
this study are hindered by the inadequacy of the randomisation
process that led to significant differences between the intervention
and control districts on two important variables The intervention
districts were more likely than the control districts to have a validlicence andto wear a helmetSincethese differences were nottaken
into account in later analyses they could explain the differences
observed in injury rates between districts Furthermore with only
six districts randomised the sample size was probably too small
to detect a difference between intervention and control groups
had there been one The McDavid 1989 cohort study found a
lower percentage of injuries in the intervention group although
a statistical test was not conducted McDavid 1989 drew their
intervention and control groups from different populations but
did adjust for confounders although they only analysed 40 of
the initial sample (139 trained and 139 untrained) Because of
these substantial weaknesses we are unable to draw conclusions
from this studyAn overall effect of pre-licence training on the reductionof crashes
injuries deaths and offences was not found although one should
be aware that the research of this issue to date has had significant
flaws However a mandatory skill test (such as the MOST) in
combination with and even without pre-licence training as part
of a licensing process seems to reduce the number of individuals
completing the licensing procedure and therefore indirectly re-
duces crash rates through a reduction in exposure to motorcycle
riding Making licensure a more difficult process may therefore be
seen as advantageous if it delays licensure and therefore reduces
exposure It is possible that the group who chose not to continue
riding a motorcycle would not have been the group to experience a
crash or commit an offence if they had continued to ride anyhowfor example because they are more safety conscious and this was
the reason for their giving up riding in the first place If this is the
case deterring individuals from completing their licence and rid-
ing may not necessarily decrease overall crashes deaths or offence
rates In addition other potentially adverse effects of complicat-
ing a licensing procedure need to be considered such as increases
in unlicensed riding In Thailand non-mandatory rider training
offered as a community intervention may have decreased injury
rates but this needs to be confirmed through further research
Non-mandatory training
The majority of the studies assessed non-mandatory training
(training independent of any licensing procedure) and the results
were varied The two non-randomised trials assessing crash rates found no statistically significant difference in crash rates between
the intervention and control groups (Adams 1985 Leung 1987)
Of the nine cohort studies Satten 1980 found a lower percentage
of crashesin the intervention group although it is unclear whether
this was a statistically significant result Davis 1997 also found a
lower percentage of crashesin theinterventiongroup although it is
unclear with which control group the intervention group wascom-
pared Raymond 1979 Osga 1980 Waller 1992 and Savolainen
2007 found a statistically significant higher crash rate in the in-
tervention group and the remaining cohort studies (Jonah 1982
Mortimer 1984 Mortimer 1988 Billheimer 1998) found no sta-
tistically significant difference in crash rates between intervention
and control groups No conclusions can be drawn from these stud-
ies with confidence The Satten 1980 and Davis 1997 studies did
not adjust for any confounding factors even though important
differences existed between the intervention and control groups
Although Raymond 1979 and Waller 1992 adjusted for a num-
ber of confounders at baseline (Waller 1992) and in the analy-
sis (Raymond 1979 Waller 1992) the intervention and control
groups were drawn from such different populations that it is pos-
sible that unmeasured confounders were responsible for the sta-
tistically significant increased risk of crashes observed in the inter-
vention group The observed difference could also be attributed totheobserved differential loss to followup (Raymond 1979) which
could bias the measure of effect towards or away from the null
The Osga 1980 study found a protective effect of rider training
in the univariate analysis but statistically significant higher crash
rates in the intervention group when accounting for riding expo-
sure Sex (a significantly higher proportion of females in the in-
tervention group) and other potentially confounding factors that
could explain thedifferencebetween the groups were howeverun-
accounted for Finally Savolainen 2007 did control for potential
confounding variables yet it is not clear whether variables were
controlled for simultaneously or one at a time One case-control
study (MAIDS 2004) found no statistically significant reduction
in the odds of experiencing a crash between trained and untrainedriders but the quality of this study was poor mainly because the
analysis did not account for any confounding factors
One cohort study (Satten 1980) assessed the effect of non-manda-
tory rider training on injurious crashes and found a lower percent-
age of injuries in the intervention group although a statistical test
was not conducted As mentioned above the Satten 1980 studyrsquos
validity is reduced due to its lack of adjustment for confounding
factors
The only non-randomised trial that assessed offence rates (Leung
1987) found a statistically significant lower offence rate in the in-
tervention group and of the five cohort studies that assessed of-
fence rates (Satten 1980 Jonah 1982 Mortimer 1984 Mortimer
21Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2450
1988) one study (Jonah 1982) also found a statistically significant
lower offence rate in the intervention group However we have lit-tle confidence in the results from Leung 1987 since potential con-
founding factors were not appropriately addressed Jonah 1982
did conduct a multivariate analysis but was hindered by an initial
low response rate The other four cohort studies did control for
several potential confounders and found no difference in offence
rates following non-mandatory post-licence rider training
The non-randomised trials and cohort studies evaluating non-
mandatory rider training were at high risk of substantial selection
bias due to the non-random allocation of participants as well
as low participation rates and attrition bias and could therefore
not provide conclusive evidence with regard to the effect of the
training It should be noted that despite the biases and other issues
with this group of studies most didnot detectmajor effects relatedto rider training in either direction This may suggest that any
potential impact of rider training will be modest
Studies examining unspecified modes of ridertraining
Three case-control studies assessed whether injured or deceased
motorcyclists were more or less likely to have had rider training
compared to control motorcyclists All had conflicting results
Hurt1981 foundthe odds of havingan injuriouscrash waslower in
those who had received rider training whereas Kraus 1975 found
that the odds of having an injurious crash were higher in youngerand older trained riders Haworth 1997 who compared advanced
versus beginner training found no effect on motorcycle injuries
or deaths although they did find that those who had completed
advanced training had reduced chances of crashing compared to
those who had done a beginnersrsquo training course Similar issues
with uncontrolled confounding and biases of selection as described
above arose with the three case-control studies Hurt 1981 suffers
from inadequate control of confounding via poor matching and
both the Kraus 1975 and the Haworth 1997 findings were only
controlled for limited factors by means of stratified analysis It is
likely that these limitations are the reason for the contradictory
directions of effect observed for the three case-control studies
Methodological weaknesses of the studies
In summary most studies investigating the effectiveness of mo-
torcycle rider training on reducing crashes injuries and death suf-
fered from serious methodological weaknesses
The majority of studies were non-randomised and were there-
fore limited mainly by selection bias Although the approaches of
matching stratification or regression analyses can minimise bias
(Fleiss 2004) they were poorly employed in the reviewed studies
The inadequate adjustment for confounders in particular adds
to our lack of confidence in the studiesrsquo results since it has been
demonstrated that individuals who choose to take motorcycle rider
training differ from those not taking rider training For instance
trained riders often differ from controls in age sex and experi-ence usually ride less and may have a more positive attitude to
safety as indicated by the reported more frequent wearing of seat
belts (Mortimer 1982)
The two large randomised trials reported that a large proportion
of their study group were never exposed to the risk of offences or
crashes because they never attained licensure This made it impos-
sible to assess whether a reduction in crashes in the intervention
group was a result of training or due to delayed or never attained
licensure
In some studies the sample sizes were too small to detect any sta-
tistically significant differences Follow-up time after training dif-
fered between studies and was an impediment to observing any
real effects of training if they existed As riding experience ac-cumulates over time any differences in outcomes between those
trained and untrained would diminish in studies with lengthy fol-
low-up times This dilution of any possible effect of rider training
may have also occurred in studies that mixed both novice and ex-
perienced riders in the study groups since it is hypothesised that
novice riders serve to benefit most from rider training This would
increase the difficulty in finding any effect of rider training par-
ticularly when sample sizes are small
Detection bias may also account for some of the differences ob-
served between trained and untrained riders Detection bias would
have occurred if the self-report hospital driver or police records
used to record the outcomes of crash injury death or offences
differed between the trained and untrained riders Research hasdemonstrated that not all crashes are reported to police for ex-
ample small crashes or off-road crashes (Kraus 1975) even serious
injury crashes have been found to be underreported (Anderson
1980) However this would only introduce detection bias if the
reporting of outcome differs for trained and untrained riders This
does appear possible in that responsible riders may be more likely
to enroll in a rider training course and also may be more likely
to report crashes to police than irresponsible riders meaning that
the number of crashes from riskier riders could be underrepre-
sented on records An analysis to investigate the effectiveness of
rider training may then show that rider training is not effective in
reducing crashes
Rather than using reported outcomes some studies used self-re-ported survey data although this technique suffers from non-
response biases In fact two studies reported differences in de-
mographic characteristics in response rates to questionnaires
(Anderson 1980 Hall 1988) Self-reported data also introduce
the risk of recall bias survival bias and social desirability response
bias Recall bias may have been equally distributed among trained
and untrained riders howeverit is hypothesised that trained riders
may have felt it was lsquoexpectedrsquo to report good driving histories
hence leading the rider training to appear effective
22Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2550
Limitations of this review
Limitations of this review were that we found very few recentstudies (the two most recent ones published in 2002 and 2007)
and that we found very few studies conducted in low-and middle
income countries even though our search strategy was set up to
identify such studies Anotherlimitation wasthat forthe reviewed
studies we could not assess all outcomes of rider training that are
potentially useful Some of the included studies assessed outcomes
such as use of safety clothing including helmets Such outcomes
are valuable since the use of motorcycle helmets has been demon-
strated to be effective in preventing head injuries (Liu 2004) De-
spite this the use of safety clothing was not included as an out-
come in this review because the studies measuring this outcome
were limited by poor data quality due to very low response rates
Another outcome of interest in assessing the effectiveness of rider
training is lsquotype of injuryrsquo As Haworth 2005 state in their review
of motorcycle licensing and training rider training could be con-
sidered effective if injuries become less severe because motorcycle
riders have learned how to avoid more serious injuries through
rider training The studies included in this review did not specify
type of injury and so this outcome could not be assessed Last
changes in knowledge and attitudes following rider training are
other potentially useful outcomes We have excluded such studies
(Osga 1979 Clayton 1990 Goldenbeld 2004) as these outcomes
are proxies rather than direct measures of the effectiveness of mo-
torcycle rider training We opted to report on the outcomes of
crashes injury death and offences alone since there is little ev-
idence about how changes in knowledge and attitudes relate to
these outcomes
A U T H O R S rsquo C O N C L U S I O N S
Implications for practice
Due to the poor quality of studies included in this review we were
unable to draw any conclusions about the effectiveness of rider
training on crash rates injurious or fatal or offence rates The
findings do suggest that mandatory pre-licence training may be
an impediment to completing a motorcycle licensing process thuspossibly indirectly reducing crash injury death and offence rates
through a reduction in exposure to riding a motorcycle On the
basis of the existing evidence it is not clear if and what type of
training reduces the risk of crashes injuries deaths or offences in
motorcyclists and the selection of the best rider training practice
can therefore not be recommended The suggestion that manda-
tory pre-licence training indirectly reduces exposure to riding the
motorcycle because trainees are impeded from completing the li-
censing process does not help those motorcyclists who persist in
obtaining their licence and riding the motorcycle Some type of
rider training is likely necessary to teach motorcyclists to ride a
motorcycle safely Good policy is based on sound evidence and
evaluation but such evidence for motorcycle rider training is seri-
ously lacking Policy makers are therefore encouraged to include arigorous evaluation component to any new or existing rider train-
ing programme Such an evaluation should ideally be randomized
and assess whether the training programme is effective and if so
which component(s) of the programme work best
Implications for research
Given theresults of this review which found littlereliable evidence
about the effectiveness of motorcycle rider training it is necessary
to build up an evidence base of the effectiveness of motorcycle
rider training in the reduction of crash injury death and offence
rates To this end well designed randomised controlled trials need
to be conducted that examine the impact of rider training oncrashes and offences A randomised controlled trial is the preferred
method as it would significantly reduce most biases described in
this review To minimise the impact of participants delaying or
never attaining their licence and therefore exposure as found in
the randomised trials in this review such a trial should include
riders who are already licensed and reliable measures of riding
exposure in a reasonably lengthy follow up period It is expected
that a randomised controlled trial of sufficient sample size would
observe equal balance on possible confounders by study arm how-
ever an assessment of potentially important confounders (such as
helmet wearing) by study group would enable control for these
variables in the analyses if necessary It is imperative that the effec-
tiveness of each component of the rider training (such as theorypractice knowledge skills and modification of behaviour) and the
synergy of such components are evaluated Outcomes should be
objective ideally those that are routinely collected rather than self-
reported outcomes Even though riding the motorcycle poses dif-
ferent challenges and incurs different risks lessons may be learned
from research into the effects of car driver training
Motorcycles are a common means of transport in low- and middle
income countries and usage is increasing As the burden of mo-
torcycle crash victims is mostly borne in such countries research
should be focused on the effectiveness of types of rider training on
the reduction of crashes injuries and fatalities in such countries
and be supported by higher income countries This review found
only one study conducted in a middle income country that sug-gested decreasing injury rates It would be beneficial to conduct
additional research to confirm this suggestion and to enhance our
knowledge of the role of rider training in low- and middle-income
settings
March 2006 saw the launch of the University of North Carolina
Highway Safety Research Centre partnership with the Motorcycle
Safety Foundation (MSF) and plans were revealed to conduct a
study of the potential benefits of motorcycle rider training The
study is funded by the MSF and the National Highway Traffic Sa-
fety Administration (NHTSA) Called rsquoThe Discovery Projectrsquo it
will compare data on crash rates from riders participating in a basic
23Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2650
motorcycle safety class with riders participating in a basic motor-
cycle safety class as well as a variety of continuing education classesThe collection of pilot data was due to start in March 2007 and
the data collection portion of the study is expected to take three
years (Highway Safety Research Center) It is vital that researchers
and policy makers collaborate to develop furtherhigh quality eval-
uations of motorcycle rider training programmes to help inform
development of new and effective training programmes
A C K N O W L E D G E M E N T S
Staff from The University of Sydney library for helping us locate
and retrieve full reports of many papers
R E F E R E N C E S
References to studies included in this review
Adams 1985 published data only
Adams A Collingwood V Job RFS Evaluation of the 1982 Australia
Post motorcycle rider training scheme NSW Australia Traffic
Accident Research Unit Traffic Authority NSW 1985 [ RN 85]
Anderson 1980 published data only
Anderson J Ford JL Peck RC Improved motorcyclist licensing and testing project Washington DC National Highway Traffic Safety
Administration (NHTSA) USA 19801ndash217 [
DOTndashHSndash805ndash598]
Billheimer 1998 published data only
Billheimer JW Evaluation of the California Motorcyclist Safety
Program Transportation Research Record 19981640100ndash9 [
TRB Report Number 980652]
Cooper 1988 published data only
Cooper PJ Rothe JP Motorcyclists who they are and why they do
what they do Transportation Research Record 1988116878ndash85
Davis 1997 published data only
David CF Evaluation of community traffic safety programs and
motorcycle operator training programs Connecticut Department of Transportation 199763
Hall 1988 published data only
Hall CJ Buchanan L Engle R Motorcycle rider education evaluationproject Washington DC USA National Highway Traffic Safety
Administration (NHTSA) US Department of Transportation
1988 [ DTNH 22ndash80ndashCndash05123]
Haworth 1997 published data only
Haworth N Smith R Brumen I Pronk N In Monash University
Accident Research Centre editor(s) Case-control study of motorcycle
crashes Victoria Australia Department of Transport and Regional
Development The Federal Office of Road Safety 19971ndash149 [
ISBN 0 642 25501 6 ]
Hurt 1981 published data only
Hurt H Ouellet J Thom D Motorcycle accident cause factors and
identification of countermeasures USA Department of
Transportation 1981 [ Technical Report DOT HS 806ndash862]
Jonah 1982 published data onlylowast Jonah BA Dawson N Bragg B Are formally trained motorcyclists
safer Accident Analysis amp Prevention 198214(4)247ndash55
Jonah BA Dawson NE Bragg BW Evaluation of the Canada
Safety Councilrsquos Motorcycle Training Program Proceedings of theInternational Motorcycle Safety Conference Washington DC
USA 198059ndash93
Kloeden 1994 published data only
Kloeden CN Moore VM McLean AJ Evaluation of the pre-license
training program for motorcyclists in South Australia Walkerville
South Australia Australia NHMRC Road Accident Research Unit
The University of Adelaide 19941ndash37 [ ISBN 0 7308 0607 3]
Kraus 1975 published data only
Kraus JF Riggins RS Franti CE Some epidemiological features of
motorcycle collision injuries American Journal of Epidemiology 1975102(1)74ndash98
Leung 1987 published data only
Leung KS Reding VA Evaluation of the Wisconsin motorcycle rider course Wisconsin Bureau of Policy Planning and Analysis
Wisconsin Department of Transportation 1987
MAIDS 2004 published and unpublished data
MAIDS In-depth investigations of accidents involving powered
two wheelers ACEM (Acemacemeu) 2004 Vol Final report
20179
McDavid 1989 published data only
McDavid JC Lohrmann BA Lohrmann G Does motorcycle
training reduce accidents Evidence from a longitudinal quasi-
experimental study Journal of Safety Research 198920(2)61ndash72
Mortimer 1984 published data only
Mortimer R Evaluation of the motorcycle rider course Accident
Analysis amp Prevention 198416(1)63ndash71
24Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2750
Mortimer 1988 published data only
Mortimer RG A further evaluation of the motorcycle rider courseJournal of Safety Research 198819(4)187ndash96
Osga 1980 published data only
Osga GA An investigation of the riding experiences of MSF rider
course participants Dissertation South Dakota USA The
University of South Dakota 19801ndash224
Perrino 2002 published data only
Perrino CS Ahmed A Callender A Rozier E Cantwell A Stewart
O The role of Marylandrsquos motorcycle rider course in promoting saver
behaviors and attitudes Baltimore USA National Transportation
Center Morgan State University 20021ndash52
Raymond 1979 published data only
Raymond S Tatum S An evaluation of the effectiveness of the RAC
ACU motorcycle training scheme - final report Road Safety ResearchUnit Dept of Civil Engineering University of Salford London
19791ndash131
Satten 1980 published data only
Satten RS Analysis and evaluation of the motorcycle rider courses
in thirteen Illinois counties International Motorcycle Safety
Conference Washington DC 1980145ndash93
Savolainen 2007 published and unpublished data
Savolainen Peter Mannering Fred Additional evidence on the
effectiveness of motorcycle training and motorcyclistsrsquo risk-taking
behaviour Transportation Research Board 2007 Vol March 20
19
Swaddiwudhipong 1998 published data only
Swaddiwudhipong W Boonmak C Nguntra P Mahasakpan P
Effect of motorcycle rider education on changes in risk behavioursand motorcycle-related injuries in rural Thailand Tropical Medicine
and International Health 19983(10)767ndash70
Waller 1992 published data only
Waller FL An analysis of the impact of the Texas Department of
Public Safety Motorcycle Operator Training Program upon
motocycle related traffic accidents Texas State University Public
Administration Program Applied Research Projects 199277
References to studies excluded from this review
Anderson 1980a published data only
Anderson LC Better biking a motorcycle rider improvement
program International Motorcycle Safety Conference Washington
DC USA Motorcycle Safety Foundation 1980231ndash53Aupetit 2008 published and unpublished data
Aupetit S Riff J Buttelli O Espie S Ergonomics research program
on motorcycle learning Contributions to the development of a
motorcycle simulator for training Transport Research Arena
Ljubljana 20089
Awane 1999 published data only
Awane T Integrating simulators in motorcycle safety education
Journal of International Association of Traffic and Safety Sciences 199923(1)26ndash35
Baldi 2005 published data only
Baldi S Baer JD Cook AL Identifying best practices states in
motorcycle rider education and licensing Journal of Safety Research200536(1)19ndash32
Batchler 1988 published data only
Batchler MW Evaluation of the pre-license motorcycle training course at the New South Wales traffic education centre Federal Office of
Road Safety (FORS) 19881ndash122 [ Report No CR 77 ]
Blanchard 2006 published data only
Blanchard HT Tabloski PA Motorcycle safety educating riders at
the teachable moment Mosby (CV) Company 200632330ndash2
Bonnett 2005 published data only
Bonnett J Mature age skills training for experienced motorcycle
riders Motorcycle ridersrsquo association of the Australian Capital
Territory 2005
Braver 2007 published data only
Braver ER et alPersuasion and licensure a randomized controlled
intervention trial to increase licensure rates among Maryland
motorcycle owners Traffic Injury Prevention 20078(1)39ndash46Clayton 1980 published data only
Clayton AB Sudlow DE An evaluation of the effectiveness of a one day training programme for learner motorcyclists Berkshire UK
Transport and Road Research Laboratory 19801ndash23 [ ISSN
0266ndash7045]lowast Clayton AB Sudlow DE An evaluation of the effectiveness of a one-day training programme for learner motorcyclists Berkshire UK
Road Safety Division Berkshire 19801ndash23 [ISSN 0266ndash7045]
Council 1978 published data only
Council FM Desper LP Dutt AK Driver education for motorcycle operation University of North Carolina Highway Safety Research
Center 19781ndash40
Crowther 2005 published data onlylowast Crowther G Engaging with motorcyclists UK police and theBikeSafe Road Safety Programme Published proceedings of the
Australasian Road Safety Research Policing Education Conference
Wellington New Zealand 2005 issue ISBN 0ndash473ndash10636ndash11ndash11
Daniello 2009 published data only
Daniello A Gabler HC Mehta YA Effectiveness of motorcycle
training and licensing Transportation Research Board 88th Annual
Meeting 200913
Dewanti 2007 published data only
Dewanti M Challenging efforts in promoting young motorcyclist
safety in Indonesia Road Safety on Four Continents 14th
International Conference Swedish National Road and Transport
Research Institute 2007
Du Plooy 1990 published data onlyDu Plooy DW Motorcycle training in South Africa The Human
Element International Motorcycle Safety Conference Orlando
Florida USA Motorcycle Safety Foundation 19904-1 to 4-9
Engel 1989 published data only
Engel U Krogsgaringrd Thomsen L [Faeligrdselsundervisning og
propaganda ndash en faeligrdselssikkerhedsfremmende foranstaltning]
Dansk Vejtidsskrift 198912295ndash7
Goldenbeld 2002 published data only
Goldenbeld C Houwing S de Craen S The development of
driving skills of young moped riders [De ontwikkeling van de
rijvaardigheid van jonge bromfietsers] SWOV Institute for Road
Safety Research (Stichting Wetenschappelijk Onderzoek
Verkeersveiligheid) Leidschendam the Netherlands 200263
25Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2850
Goldenbeld 2004 published data only
Goldenbeld C Twisk D de Craen S Short and long term effects of moped training a field experiment Transportation Research Part F 20047(1)1ndash16
Haworth 2000 published data only
Haworth N Smith R Kowadlo N Motorcycle training in Victoria
evaluation of rider training curriculum in Victoria Monash
University Accident Research Centre Reports 2000 Vol 165 issue
3131
Haworth 2005 published data only
Haworth N Mulvihill C Review of motorcycle licensing and
training Monash University Accident Research Centre 2005 [
Report no240]
Haworth 2006 published data only
Haworth N Mulvihill C Rowdon P Teaching old dogs new tricksTraining and older motorcyclists Journal of the Australasian College of Road Safety 200718(4)20ndash5lowast Haworth NL Mulvihill C Rowden PJ Teaching old dogs new
tricks Training and older motorcyclists Australasian Road Safety
Research Policing and Education Conference Gold Coast
Queensland 2006
Hill 1983 published data only
Hill D Motorcycle rider training testing and licensing Motor
Cycle Council of New South Wales Sydney 1983
Imatake 1980 published data only
Imatake Y On the activities in Japan of motorcycle safe riding
promotion committee (MSRPC) - especially the education of
unskilled riders of class 1 motorbikes International Motorcycle
Safety Conference Washington DC USA Motorcycle Safety
Foundation 1980255ndash68
Ingebritsen 1990 published and unpublished data
Ingebrigtsen S [Risikofaktorer ved ferdsel med moped og
motorsykkel] TOslashI-rapport 66 Oslo Transportoslashkonomisk
institutt 1990
Jain 2009 published and unpublished data
Jain A et alTwo wheeler accidents on Indian roads a study from
Mangalore India Elsevier 200916130ndash3
Jonah 1979 published data only
Jonah BA Dawson NE Validation of the motorcycle operator skill
test Accident Analysis amp Prevention 197911(3)163ndash71
Jonah 1981 published data onlyJonah BA Dawson NE Bragg WE Predicting accident
involvement with the Motorcycle Operator Skill Test Accident Analysis amp Prevention 198113(4)307ndash418
Kadar 1990 published data only
Kadar KJ Countermeasures motorcycle rider education delivery
system The Human Element International Motorcycle Safety
Conference Motorcycle Safety Foundation 19906-27 to 6-41
Kelsey 1986 published data only
Kelsey SL Liddicoat C Ratz M Licensing novice motorcyclists a
comparison of Californiarsquos standard test and the MOST II (Motorcycle
Operator Skill Test) administered at centralized testing offices Sacramento California USA California Department of Motor
Vehicles Research and Development Office 19861ndash74
Koch 1980 published data only
Koch H Advanced course for motorcycle riders InternationalMotorcycle Safety Conference Washington DC USA Motorcycle
Safety Foundation 1980269ndash81
Koch 1990 published data only
Koch H Brendicke R How to influence risk taking behaviour
within motorcycle rider training programs The Human Element
International Motorcycle Safety Conference Orlando Florida
USA Motorcycle Safety Foundation 1990Motorcycle Safety
Foundation 1990 p 9-1 to 9-25A
Labbett 2005 published data only
Labbett S Langham M Training can make the problem worse
Road Safety Congress 20051ndash12
Law 2005 published and unpublished data
Law TH Radin Umar RS Zulkaurnain S Kulanthayan S Impactof the effect of economic crisis and the targeted motorcycle safety
programme on motorcycle-related accidents injuries and fatalities
in Malaysia International Journal of Injury Control and Safety
Promotion 200512(1)9ndash21
Lowes 1990 published data only
Lowes BF In-traffic rider training The Human Element
International Motorcycle Safety Conference Orlando Florida
USA Motorcycle Safety Foundation 19907-46 to 7-57
Mahon 1976 published data only
Mahon J Teaching accident avoidance on the motorcycle Journal of Traffic Safety Education 1976231ndash29
Maxwell 1985 published data only
Maxwell GS Evaluation of a Program for Training MotorcycleRiders Evaluation and Program Planning 19858(4)339ndash47
Mayhew 1996 published data only
Mayhew DR Simpson SH Effectiveness and role of driver educationand training in a graduated licensing system Ottowa Canada The
Traffic Injury Research Foundation 19961ndash89 [ ISBN
0ndash920071ndash12ndash0]
McKnight 1987 published data only
McKnight AJ Milburn BG An evaluation of the modified
motorcycle operator skill test exercises NHTSA Report 1987 [
Report No DOT HS 807 212 ]
McPherson 1976 published data only
McPherson K McKnight AJ The development and evaluation of a
motorcycle skill test manual and knowledge test Washington DCUSA National Highway Safety Administration 1976
Mohan 2004 published data only
Mohan D Evidence-based interventions for road traffic injuries in
South Asia Journal of the College of Physicians and SurgeonsPakistan 200414(12)746ndash7
Morgan 1980 published data only
Morgan JK Advanced motorcycle training course content and
study evaluation International Motorcycle Safety Conference
Washington DC USA Motorcycle Safety Foundation 1980
283ndash312
Motoki 1990 published data only
Motoki M Yamazaki S A study on effective motorcycle rider
education The Human Element International Motorcycle Safety
26Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2950
Conference Orlando Florida USA Motorcycle Safety
Foundation 19909-26 to 9-57A
Nairn 1992 published data onlylowast Nairn RJ Motorcycle safety research literature review 1987 to
1991 Federal Office of Road Safety Australia 1993 Vol Report
CR 117
Newland 1980 published data only
Newland R Darby J Automotive practicespre motorcycle rider
program International Motorcycle Safety Conference Washington
DC USA Motorcycle Safety Foundation 1980127ndash9
Newland 1982 published data only
Newland R Motorcycle rider training and licensing requirements
Driver training steering a course for the future Melbourne
Victoria Transport Regulation Board 1982
Newland 1990 published data only
Newland R Motorcycle training and licensing in Victoria 1980-
1990 The Human Element International Motorcycle Safety
Conference Orlando Florida USA Motorcycle Safety
Foundation 1990
Ormston 2003 published data only
Ormston R Dudleston A Pearson S Stradling Evaluation of
Bikesafe Scotland Scottish Executive Social Research Edinburgh
2003
Osga 1979 published data only
Osga GS Ellingstad VS An evaluation of the effectiveness of the 1978
South Dakota motorcycle rider course program South Dakota USA
University of South Dakota 19791ndash89
Panou 2007 published data only
Panou M Bekiaris E A new concept on the integration of driving
simulators in driving training the train-all approach Road Safety
on Four Continents 14th International Conference Swedish
National Road and Transport Research Institute 2007
Peterson 1990 published data only
Peterson A Kadar K Status report 1980-1989 Minnesota rider
education statistics The Human Element International
Motorcycle Safety Conference Orlando Florida USA 199013-
113 to 13-129
Prem 1984 published data only
Prem H Good MC Motorcycle rider skills assessment ACT
Australia Office of Road Safety Department of Transport 1984
1ndash377 [ Report No CR 34 ]
Radin Umar 2006 published data only
Radin Umar RS Motorcycle safety programmes in Malaysia how
effective are they International Journal of Injury Control and Safety Promotion 200613(2)71ndash9
Rockwell 1990 published data only
Rockwell TH Kiger S Carnot M An evaluation of the Ohio
motorcyclist enrichment program Ohio Department of Highway
Safety 1990
Rothe 1987 published data only
Rothe JP A motorcycle riding course Motorcyclists Image and reality Canada Insurance Corporation of British Columbia 1987
59ndash83
Rowden 2007 published data only
Rowden P Watson B Haworth N What can riders tell us aboutmotorcycle training A view from the other side of the fence
Australasian Road Safety Research Policing and Education
Conference Melbourne Australia Melbourne 2007
RSC Victoria 1993 published data only
Road Safety Committee Inquiry into Motorcycle Safety in Victoria
Parliament of Victoria 1993
Russam 1979 published data only
Russam K Motorcycle training and licensing in Japan UK Transport
and Road Research Laboratory (TRRL) 1979 [ Report 916 ]
Saffron 1981 published data only
Saffron DG Driver and rider education training and licensing a
brief review Traffic Accident Research Unit Traffic Authority of
New South Wales NSW Australia Traffic Accident Research UnitTraffic Authority of New South Wales 1981
Schulz 1990 published data only
Schulz U Motorcycle accidents among learner riders Motorcycle
licensing Evaluation of the effectiveness of a Graduated Licensing Scheme Forshungshefte des Institut fur Zweiradsicherheit 1990
Sexton 2004 published data only (unpublished sought but not used)
Sexton B Baughan C Elliott M Maycock G The accident risk of
motorcyclists 2004TRL Report TRL 607(ISSN 09684107)58
Shephard 1986 published data only
Shepherd R Do Motorcycle Safety Foundation programs work
Driver 198619(9)16ndash8
Simpson 1990 published data onlySimpson HM Mayhew DR The promotion of motorcycle safety
training education and awareness Health Education Research
Theory and Practice 19905(2)257ndash64
Sokolov 1990 published data only
Sokolov O Safe motorcycle riding some aspects of motorcycle
riding in the USSR The Human Element International
Motorcycle Safety Conference Motorcycle Safety Foundation
19904ndash37 - 4-48
Steffens 1988 published data only
Steffens U Gawatz R Willmes G Effectiveness of mofa courses
[Wirksamkeit von Mofakursen] Unfall- und sicherheitsforschung
strasenverkehr 198867168
Sudlow 2003 published data only
Sudlow D Scoping study on motorcycle training Road Safety
Research Report London UK The British Institute of Traffic
Education Research Department for Transport 2003 Vol June
[ Report No 36]
Tasmania 1995 published data only
Mannering FL Grodsky LL Motorcycle rider training scheme
Traffic Education Research Centre Armidale Australia Driver
Training Australia Limited 1995 Vol April1ndash56
Tetard 1985 published data only
Tetard C Motorcycle ridersrsquo training to detect indices in an
intersection situation [Lrsquoentrainement a la prise drsquoinformation dans
lrsquoapprentissage de la conduite motocycliste] Le Travail Humain198548(2)147ndash59
27Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3050
Thackray 1980 published data only
Thackray RM Prescott JC Field test of the motorcycle ridercourse International Motorcycle Safety Conference Washington
DC USA Motorcycle Safety Foundation 1980195ndash7
Thompson 1994 published data only
Thompson M Evaluation of compulsory basic training for
motorcyclists Transport Research Laboratory Crowthorne
Berkshire UK 1994 [ Report 63]
Tobler 1990 published data only
Tobler U Individual driving instruction for motorcycle driving
learners the Swiss experiment of motorcycle driving school The
Human Element International Motorcycle Safety Conference
Orlando Florida USA Motorcycle Safety Foundation 1990
Transport 1989 published data only
Transport Communications Road Crash Statistics Australia Bureauof Transport and Communications Economics 198914ndash20
Tuan 2007 published data only
Tuan VA Shimizu T Towards development and evaluation of the
motorcycle drivers re-education program in Vietnam modeling of
motorcycle driverrsquos undesired behaviors 11th World Conference
on Transport Research 200725
Victor 1980 published data only
Victor DJ Traffic safety education for motorcyclists International
Motorcycle Safety Conference Washington DC USA Motorcycle
Safety Foundation 1980221ndash8
Waterer 1990 published data only
Waterer TH Compulsory training for motorcyclists from liberty
to compulsion The Human Element International Motorcycle
Safety Conference Orlando Florida USA Motorcycle Safety
Foundation 19904-49 to 4-57
Weller 1989 published data only
Weller RB Chandler EW Motorcycle safety and motorcycle
education past research and survey results Journal of Public Policy
amp Marketing 1989893ndash108
Wells 1981 published data only
Wells P Observations of some RACACU motorcycle training
schemes Transport and Road Research Laboratory Crowthorne
Berkshire UK 1981
Winn 1986 published data only
Winn G Two important trends in motorcycle safety regulations
rider education and conspicuity improvement American
Motorcyclist Association 1986147ndash52
Wisher 1988 published data only
Wisher G Reid J Motorcycle accidents in Norfolk a three-year study of motorcycle training and the Part 1 test Norfolk UK Road Safety
Division Highways Department Norfolk County Council 1988
Wood 1987 published data only
Wood HT Bowen R Evaluation of the revised learner permit scheme
July 1983 to December 1985 Victoria Australia Road Traffic
Authority 19871ndash27 [ ISBN 0 7306 0247 8]
Yamazaki 1990 published data only
Yamazaki S Nakanishi M Survey on effect of Honda rider
education The Human Element International Motorcycle Safety
Conference Orlando Florida USA 199012-167 to 12-187
References to ongoing studies
Burgess 2010 unpublished data only
Burgess C Broughton P Fylan F Stradling Interim evaluation of
the UKs National Ride program In L Dorn editor(s) Driver
behaviour and Training Vol IV Aldershot Ashgate Publishing
Limited 2010
Additional references
Anderson 1980
Anderson J Ford JL Peck RC Improved motorcyclist licensing and
testing project Volume I and II National Highway Traffic Safety
Administration US department of Transportation Washington
DC 1980
Bjornstig 1985
Bjornstig U Bylund P Lekander T Brorsson B Motorcycle fatalitiesin Sweden Acta Chirurgica Scandinavica 1985151(7)577ndash81
Clayton 1990
Clayton A Sudlow DE An evaluation of the effectiveness of a one-
day training program for learner motorcyclists The Human
Element International Motorcycle Safety Conference Orlando
Florida USA 199013-105 to 13-111
Cochrane Collaboration 2008
Higgins JP Green S Cochrane Handbook for Systematic Reviews of
Interventions 500 The Cochrane Collaboration 2008
Elliot 2003
Elliott MA Baughan CJT Broughton J Grayson G Knowles J et
alMotorcycle Safety a Scoping Study TRL [ Report No
0968ndash4107]Fleiss 2004
Fleiss JL Levin B Paik MC Statistical methods for rates and proportions 3rd Edition John Wiley amp Sons Inc 2004 [ ISBN
978ndash0ndash471ndash52629ndash2 ]
FORS 1999
FORS Federal Office of Road Safety Road risk for sober licensed
motorcyclists Vol Monograph 27 Federal Office of Road Safety
19991ndash2
Haworth 2005
Haworth N Mulvihill C Review of motorcycle licensing and training
Victoria Australia Monash University Accident Research Centre
20051ndash83 [ Report No 240]
Huwiler-Muumlntener 2002Huwiler-Muumlntener K Juumlni P Junker C Egger M Quality of
Reporting of Randomized Trials as a Measure of Methodologic
Quality JAMA 20022872801ndash4
Johnston 2008
Johnston P Brooks C Savage H Fatal and serious road crashes
involving motorcyclists Road Safety Research and Analysis Report
ATSB Monograph 20 Canberra Department of Infrastructure
Transport Regional Development and Local Government 0205
081ndash26
Juumlni 2001
Juumlni P Altman DG Egger M Systematic reviews in health care
Assessing the quality of controlled clinical trials BMJ 2001323
42ndash6
28Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3150
Kopits 2003
Kopits E Cropper M Traffic fatalities and economic growth TheWorld Bank Development Research Group Infrastructure and
Environment Policy research working paper 3035 2003
Leung 1983
Leung P Chau BTW Motorcycle accidents in Hong Kong pattern
of injury treatment causative factors and socio-economic
implications Injury 198315(1)23ndash9
Liu 2004
Liu B Ivers R Norton R Blows S Lo SK Helmets for preventing
injury in motorcycle riders Cochrane Database of Systematic Reviews 2004 Issue 2 [DOI 10100214651858CD004333pub3]
Mohan 2002
Mohan D Road safety in less-motorized environments future
concerns International Journal of Epidemiology 200231(3)527ndash32Mortimer 1982
Mortimer RG An evaluation of the motorcycle rider course
Proceedings of the 20th conference of the American Association for
Automotive Medicine Arlington Heights IL USA American
Association for Automotive Medicine 198277ndash96
Mullin 1997
Mullin BT Risk factors for motorcycle injury the role of age
gender experience training and alcohol PhD thesis Auckland
New Zealand University of Auckland 1997
Noordzij 2001
Noordzij PC Forke E Brendicke R Chinn BP Integration of needs
of moped and motorcycle riders into safety measures review and
statistical analysis in the framework of the European research
project PROMISING SWOV Institute for Road Safety Research
Leidschendam The Netherlands 2001 [ Report No Dndash2001ndash5]
Osga 1979
Osga G Ellingstad VS An evaluation of the effectiveness of the 1978 South Dakota Motorcycle Riders Course Program South Dakota
University of South Dakota Vermillion Human Factors Lab 1979
Oxford English Dictionary Online
httpoxforddictionariescom Oxford English Dictionary Online
Oxford University Press 2010
Peden 2004
Peden M Scurfield R Sleet D World Report on Road TrafficInjury Prevention Geneva World Health Organization Available
at httpwwwwhointviolence˙injury˙preventionpublications
road˙trafficworld˙reporten 2004Chapter 3
Radin Umar 1995
Radin Umar R Macakay G Hills B Preliminary analysis of
motorcycle accidents Short-term impacts of the running
headlights campaign and regulations in Malaysia J Traffic Med 199523(1)17ndash27
Tsai 1995
Tsai YJ Wang JD et alCase-control study of the effectiveness of
different types of helmets for the prevention of head injuries among
motorcycle riders in Taipei Taiwan American Journal of
Epidemiology 1995142(9)974ndash81
Vis 1995
Vis AA In-depth study of the hazards of motorcycling [De
onveiligheid van motorrijden nader bekeken] SWOV Institute for
Road Safety Research Leidschendam the Netherlands 1995 [
Report No Rndash95ndash69]
Weaver 1990
Weaver EA Motorcycle safety education in the United States
Motorcycle Safety Foundation Orlando Florida USA 1990
Winn 1987
Winn G Two important trends in motorcycle safety regulations
rider education and conspicuity improvement No more
information available Detroit MI US Society of Automotive
Engineers 1987147ndash52
Yuan 2000
Yuan W The effectiveness of the rsquoride-brightrsquo legislation for
motorcycles in Singapore Accident Analysis amp Prevention 200032
(4)559ndash63
Zambon 2006
Zambon F Hasselberg M Factors affecting the severity of injuries
among young motorcyclists a Swedish nationwide cohort study
Traffic Injury Prevention 20067(2)143ndash9lowast Indicates the major publication for the study
29Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3250
C H A R A C T E R I S T I C S O F S T U D I E S
Characteristics of included studies [ordered by study ID]
Adams 1985
Methods Non-randomised trial
Participants Initial sample size
Total of 162 Australia Post telegram delivery riders
Not clear how many in intervention and control groups
Analysis sample size
Intervention group 63Control group 35
Total 98 (61)
Interventions No formal training vs voluntary rider training course of 12 hours over 2 consecutive days (theory 4 hours practice
8 hours)
Intervention integrity
No information given
Outcomes Australia Post recorded crashes
Length of follow up 1 year
Notes Study conducted in NSW Australia
Anderson 1980
Methods Randomised trial
Participants Initial sample size
Total 41188 motorcycle licence applicants
Analysis sample size
Intervention group with skills training 13160
Intervention group without skills training 12634
Control group 15080
Total 40874 (99)
Interventions Standard licensing skill test vs the Motorcycle Operators Skill Test (MOST) with or without a training component
MOST is not to be confused with rider training because it is a licensing test The training component was for
participants who failed the skill test on the first attempt was of 3 hours duration and consisted of instructions about
the skills needed to pass the skill test The training consisted of 30 min theory and 2 frac12 hours practice on the
motorcycle
Intervention integrity
33 in MOST with skills training 31 in MOST without skills training and 43 in control group obtained their
motorcycle licence in the end
Outcomes Police recorded crashes and offences
Length of follow up 3 years
30Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3350
Anderson 1980 (Continued)
Notes Study conducted in California USA
Billheimer 1998
Methods Cohort study
Participants Initial sample size
Intervention group Unclear numbers of CMSP graduates and ERC graduates
Control group Over 16000 untrained motorcyclists
Analysis sample size
1139 matched CMSP-pairs and 1182 ERC-pairs
Interventions No formal training vs the mandatory (for those younger than 18 years) pre-licence 16-hours basic California
Motorcyclist Safety Program (CMSP) and the 8-hour Experienced Rider Course (ERC)
Intervention integrity
No information given
Outcomes Police recorded crashes and offences
Length of follow up 2 years
Notes Study conducted in California USA
Cooper 1988
Methods Cohort study
Participants Initial sample size
Intervention group 418 graduates and 57 who failed
Control group 402
Analysis sample size
Intervention group unclear
Control group unclear
Total 863 (984)
Interventions No formaltrainingvs non-mandatory pre-licence rider trainingThe content of thetrainingwas notfurther explainedIntervention integrity
Participants had all completed the rider training
Outcomes Police recorded crashes and offences
Length of follow up 7 years
Notes Study conducted in British Columbia Canada
31Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3450
Davis 1997
Methods Retrospective cohort study
Participants Initial sample size
Intervention group 9320 graduates
Control group unclear
Analysis sample size
Intervention group 9320 graduates
Control group unclear
Interventions No formal training vs the non-mandatory Connecticut Motorcycle Operator Training Program The content of
training was not further explained
Intervention integrity Participants had all completed the rider training
Outcomes Police recorded crashes severity of crashes
Length of follow up 1 year
Notes Study conducted in Connecticut USA
Hall 1988
Methods Randomised trial
Participants Initial sample sizeIntervention group A 6248
Intervention group B 6616
Intervention group C 6996
Control group 6604
Total 26464 motorcycle licence applicants
Analysis sample size
Intervention group A 6012 (96)
Intervention group B 6428 (97)
Intervention group C 6778 (97)
Control group 6502 (98)
Interventions Standard licensing skill test vs the Motorcycle Operator Skill Test (MOST II) without training (A) MOST II with
3 hours training (B) or MOST II with 20 hours training (C) The 3 hours training comprised instructions aboutthe skills needed to pass the skill test (30 min theory and 2 frac12 hours practice on the motorcycle) The 20 hours
training consisted of the Motorcycle Safety Foundationrsquos Motorcycle Rider Course (MRC) comprising of 8 hours
theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-road practice
Intervention integrity
25 and 26 obtained their motorcycle licence in the lsquoMOST 3-hour trainingrsquo and lsquoMOST 20-hour trainingrsquo
intervention arm respectively In the lsquoMOST testrsquo intervention arm 29 obtained their licence and in the control
group 35
Outcomes Police recorded crashes and offences
Length of follow up 3 years
32Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3550
Hall 1988 (Continued)
Notes Study conducted in New York USA
Haworth 1997
Methods Population based case-control study
Participants Cases Motorcycle riders or pillion riders involved in motorcycle crash (on road) Recruited from hospital and police
records Not clear whether all cases (especially fatalities) were included
Community controls Multiple matched sample of motorcycle rider or pillion who passed the crash site at the same
time of day and day of week that the crash occurred
Initial sample size
Cases 336
Controls 1195
Analysis sample size
Cases 234 (70)
Controls 561 (47)
Interventions Basic handling skills course beginner intermediate or advanced course measured through self-report not validated
Outcomes Injuries with an Injury Severity Score of 5 or greater or admission for at least four hours and fatalities from a
motorcycle crash (from hospital and police records)
Length of follow up two years or more prior to the interview
Notes Study conducted in Victoria Australia
Hurt 1981
Methods Population based case-control study
Participants Cases Motorcycle riders who had a motorcycle crash that was reported to police (traffic accident reports) Not all
cases included
Community controls Multiple matched sample of motorcycle riders who passed the crash site at the same time of
day and day of week as the crash occurred but 2 years later
Initial sample size
Cases 4500 notifications of crashes to police (only 50 received in time research team able to respond to 1126 for
unreported reasons data collection completed for 900)
Controls Number approached unreported 2310 consented and interviewed
Analysis sample size
Cases 808 (808900 data on rider training) - 18 of initial sample
Controls 670 (6702310 data on rider training) - lt 29 of initial sample
Interventions Training experience self taught friends-family school-club formal-professional others measured through self-
report validated
Outcomes Injuries and fatalities from a motorcycle crash (from hospital records and coronerrsquos office)
33Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3650
Hurt 1981 (Continued)
Notes Study conducted in Los Angeles USA
Jonah 1982
Methods Cohort study
Participants Initial sample size
Intervention group
2310 licensed MTP graduates who are active riders
Control group
2029 licensed motorcyclists who are active riders
Non-licensed and non-active riders excluded
Analysis sample size
Intervention group
811 interviews (35) 533 driver records (23)
Control group 1080 interviews (71) 905 driver records (45)
Interventions No training vs the non-mandatory Canada Safety Council Motorcycle Training Program (MTP) comprising of 4
hours theory and 16 hours practise (6 hrs on-road) over 2 weekends
Intervention integrity
Participants had all successfully completed the MTP
Outcomes Self-reported and police reported crashes and offencesLength of follow up 4 years
Notes Study conducted in Ontario Canada
Kloeden 1994
Methods Cohort study
Participants Initial sample size
Intervention group 2533 licence applicants
Control group Unclear
Sample of riders obtaining probationary licenceIntervention group 1625 (64)
Control group 5015 (unclear)
Non-active riders excluded from control group (~66)
Analysis sample size
Intervention group 631 (39)
Control group 318 (6)
Interventions No formal training vs the mandatory pre-licence Ridersafe course consisting of two levelsthe first is a training course
in the basics of motorcycle riding where a pass entitles the rider to obtain a learnerrsquos permit The second is a skill
test where a pass entitles the rider to obtain a probationary motorcycle licence The training course consists of two 4
hour training sessions on separate days involving instruction videos and actual riding on motorcycles The skill test
consists of a 4 hour preparation course of 74 minutes of lecturediscussion 38 minutes of video and 75 minutes of
34Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3750
Kloeden 1994 (Continued)
on-course riding which is the testing phase
Intervention integrity
Unclear what percentage in the Ridersafe group and the control group obtained their probationary licence halfway
and their full motorcycle licence in the end
Outcomes Police reported crashes
Length of follow up 1 year
Notes Study conducted in SA Australia
Kraus 1975
Methods Population based case-control study
Participants Cases All motorcycle riders with a confirmed medically treated motorcycle injury (including fatalities) from all
hospitals in study base and peripheral area hospitals and all deaths due to motorcycle crashes in the study base All
serious injuries (leading to hospitalisation) and fatalities likely to be included
Community controls A random sample of registered motorcycle owners in the same temporary and geographical
study base as the cases arose from
Initial sample size Cases 1273
Controls 738
Analysis sample size Cases 741 (58)
Controls 434 (59)
Interventions Motorcycle rider training (not specified) measured through self-report
Outcomes Injuries leading to hospitalisation fatalities
Length of follow up not specified
Notes Study conducted in Sacramento county USA
Leung 1987
Methods Non-randomised trial
Participants Initial sample sizeIntervention group unknown number of MRC graduates
Control group unknown number of licensed motorcyclists
Analysis sample size
Intervention group 2941
Control group 43094
Interventions No formal training vs non-mandatory Wisconsin motorcycle rider course a 22 hours off-street programme that
teaches the basic rules of safe motorcycling
Intervention integrity
50 in the MRC group and 100 in the control group obtained their motorcycle licence in the end
35Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3850
Leung 1987 (Continued)
Outcomes Police recorded crashes and offences
Length of follow up up to 3 years
Notes Study conducted in Wisconsin USA
MAIDS 2004
Methods Population based case-control study
Participants Cases Motorcycle riders who had a motorcycle crash Unclear whether all cases included
Community controls Motorcycle riders identified at petrol stations Unclear whether matched to cases
Initial sample size
Cases not reported
Controls not reported
Analysis sample size
Cases 921 (Italy 200 Netherlands 200 Spain 121 Germany 250 France 150)
Controls 923 (Italy 200 Netherlands 200 Spain123 Germany 250 France 150)
Interventions Pre-license rider training or additional training (non-compulsory)
Outcomes Crashes
Notes Study in five sampling areas located in five European countries (France Germany Netherlands Spain and Italy)
McDavid 1989
Methods Cohort study
Participants Sample size after first match
Intervention group 346 male training graduates
Control group 346 male motorcycle riders licensed in same year
Sample size after second match
Intervention group 139 (40)
Control group 139 (40)
Analysis sample size
Intervention group 139 (40)
Control group 139 (40)
Interventions No formal training vs the non-mandatory pre-licence British Columbia Safety Councilrsquos 37-hour motorcycle safety
programme
Intervention integrity
Participants had all completed and passed the training
Outcomes Police reported crashes injurious crashes
Length of follow up 1-6 years
36Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3950
McDavid 1989 (Continued)
Notes Study conducted in British Columbia Canada
Mortimer 1984
Methods Cohort study
Participants Initial sample size
Intervention group 608 MRC graduates
Control group 360 motorcycle riders
Response rates
Intervention group 360 (59)
Control group 324 (90)
Non-active riders excluded
Analysis sample size
Intervention group 213 (35)
Control group 303 (84)
Interventions No formal training vs the non-mandatory Motorcycle Safety Foundationrsquos Motorcycle Rider Course comprising of
8 hours theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-road practice
Intervention integrity
Participants had all completed and passed the MRC
Outcomes Self-reported crashes and offencesLength of follow up 1 year
Notes Study conducted in Illinois USA
Mortimer 1988
Methods Cohort study
Participants Initial sample size
Intervention group 2307 MRC graduates
Control group 556 active licensed motorcycle riders
Response ratesIntervention group 1298 (56)
Control group 500 (90)
Non-active riders excluded
Analysis sample size
Intervention group 913 (40)
Control group 500 (90)
Interventions No formal training vs the non-mandatory Motorcycle Safety Foundationrsquos Motorcycle Rider Course comprising of
8 hours theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-road practice
Intervention integrity
Participants had all completed and passed the MRC
37Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4050
Mortimer 1988 (Continued)
Outcomes Self reported crashes and offences
Length of follow up 1 year
Notes Study conducted in Illinois USA
Osga 1980
Methods Cohort study
Participants Initial sample sizeIntervention group 1163 MSF MRC participants invited 818 (74) responded
Control group 329 motorcycle riders responded to recruitment strategies
Non-active riders excluded from analysis
Analysis sample size
Intervention group 595 (73)
Control group 329 (100)
Interventions No formal training vs the non-mandatory Motorcycle Safety Foundationrsquos Motorcycle Rider Course (MRC) com-
prising of 8 hours theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-road
practice
Intervention integrity
Participants had all successfully completed the MRC
Outcomes Self-reported crashes police recorded crash reports
Length of follow up 6 months - 3 12 years
Notes Study conducted in South Dakota USA
Perrino 2002
Methods Cohort study
Participants Initial sample size
Intervention group 2144 (unclear) MRC graduates (motorcycle licence applicants)
Control group 2601 (unclear) regular licensing test graduates (motorcycle licence applicants)Analysis sample size
Intervention group 1070 (unclear)
Control group 867 (unclear)
Interventions No formal training vs the mandatory (for those younger than 18 years) pre-licence Maryland Motorcycle Rider
Course (MRC) consisting of 8 hours theory and 12 hours practice
Intervention integrity
No information given
Outcomes Self-reported and police recorded crashes and offences
Length of follow up 1 year
38Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4150
Perrino 2002 (Continued)
Notes Study conducted in Maryland USA
Raymond 1979
Methods Cohort study
Participants Initial sample size
Intervention group
258 volunteers mainly (but not all) novice riders
Control group
1104 provisional licence applicants
Analysis sample size
Intervention group 190 (74)
Control group 627 (57)
Interventions No formal training vs the non-mandatory RACACU (Royal Automobile ClubAuto-Cycle Union) training scheme
The training program consisted of either 24 one-hour sessions or 13 two-hour sessions Twelve hours of the training
focused on theory about basic operational procedures traffic laws and highway code The practical session focused
on basic operational skills riding on a training area and riding in traffic with an instructor A proficiency test was
administered at the end of the course
Intervention integrity
No information given
Outcomes Self-reported crashes (including minor non-injury non-damage and off-road crashes)
Length of follow up 1 year
Notes Study conducted in the UK
Satten 1980
Methods Cohort study
Participants Initial sample size
Intervention group interviewers kept interviewing until they had 100 riders selected from a population of 891
enrollees in the MSF MRC course (unknown how many were asked and how many refused to participate)Control group unknown number of registered motorcycle riders
Non-active riders excluded from analysis
Analysis sample size
Intervention group 71
Control group 100
Interventions No formal training vs the non-mandatory Motorcycle Safety Foundationrsquos (MSF) Motorcycle Rider Course (MRC)
comprising of 8 hours theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-
road practice
Intervention integrity
Participants had all successfully completed the MRC
39Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4250
Satten 1980 (Continued)
Outcomes Self-reported crashes injuries and offences
Length of follow up 3 years
Notes Study conducted in Illinois USA
Savolainen 2007
Methods Cohort study
Participants Initial sample size
Intervention group 4000 riders that had conducted the MSF MRC course (passed or failed)
Control group 4000 motorcycle owners or permit or endorsement holders from the Indiana Bureau of Motor
Vehicles database
Analysis sample size
Intervention group 739 (18)
Control group 588 (15)
Interventions Intervention integrity
No information given
Outcomes Probability of crash involvement
Length of follow up up to 5 years
Notes Study conducted in Indiana USA
Swaddiwudhipong 1998
Methods Randomised trial
Participants Initial sample size
Intervention group
all 16 villages of 3 sub-districts
(participants were practising motorcyclists licensed and unlicensed)
Control group
18 villages of 3 other sub-districtsAnalysis sample size
Intervention group 3 districts (100)
Control group 3 districts (100)
Interventions No rider education vs non-mandatory community based rider education consisting of theory on motorcycle injury
risk helmet use traffic laws vehicle regulations and traffic signs and a written and skill test leading to a licence
Intervention integrity
No information given
Outcomes Injuries fatalities
Length of follow up 2 years
40Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4350
Swaddiwudhipong 1998 (Continued)
Notes Study conducted in Tak Province rural Thailand
Waller 1992
Methods Cohort study
Participants Initial sample size
Intervention group 349 riders who succesfully completed the Department of Public Safety motorcycle operator
training in 1988
Control group 349 motorcyclists who never completed the DPS motorcycle operator training
Analysis sample size
Intervention group 349
Control group 349
Interventions No DPS motorcycle operator training vs non-mandatory DPS motorcycle operator training The DPS training is
the Motorcycle Safety Foundationrsquos Motorcycle Rider Course comprising eight hours of theory (lectures discussions
films and multiple choice tests) a textbook and 12 hours off-road practice
Intervention integrity
All 349 riders succesfully completed the rider training
Outcomes Police recorded crashes
Length of follow up 3 years
Notes Study conducted in
Characteristics of excluded studies [ordered by study ID]
Study Reason for exclusion
Anderson 1980a Study type review and process evaluation of a rider course
Aupetit 2008 Outcome proposal of a rider training curriculum that combines traditional training and a motorcycle riding
simulator
Awane 1999 Study type review on use of simulators in rider and driver training
Baldi 2005 Intervention not rider training but intervention integrity of rider training availabilty of rider education
implementation of graduated licensing system etc on a state-wide level
Batchler 1988 Study type description of a training course
Blanchard 2006 Study type not a study assessing the association between rider training and outcomes
Bonnett 2005 Outcome qualitative data regarding satisfaction with the training programme
41Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4450
(Continued)
Braver 2007 Outcome licensure rates
Clayton 1980 Outcome knowledge skills attitude
Council 1978 Outcome crash reduction in ANY vehicle not motorcycles exclusively
Crowther 2005 Study type qualitative study
Daniello 2009 Study type review of the effectiveness of motorcycle training and licensing
Dewanti 2007 Intervention emotional characteristics behaviour (ie not motocycle rider training)
Du Plooy 1990 Study type review of types of rider training courses in South Africa
Engel 1989 Study type uncontrolled before-after study
Goldenbeld 2002 Outcome skills
Goldenbeld 2004 Outcome knowledge and skills
Haworth 2000 Intervention balance between attitudinal andvehicle control skills acrosscurriculla of motorcycle rider training
courses
Haworth 2005 Study type review
Haworth 2006 Study type cross-sectional survey
Hill 1983 Study type review and description of training system
Imatake 1980 Study type course descriptionreview
Ingebritsen 1990 Intervention risk factors for motorcycle crashes
Jain 2009 Study type cross-sectional descriptive study of motorcycle crashes over a 5 year period
Jonah 1979 Study type cross-sectional survey Outcome scores on motorcycle operator skill test
Jonah 1981 Intervention evaluation of a motorcycle rider skill test not motorcycle rider training
Kadar 1990 Study type review of components of rider training courses
Kelsey 1986 Intervention evaluation of a motorcycle rider skill test not motorcycle rider training
Koch 1980 Study type reviewcourse description
Koch 1990 Study type description of course content
42Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4550
(Continued)
Labbett 2005 Study type essay about possible explanations for motorcycle deaths
Law 2005 Intervention targeted motorcycle safety programme not specifically rider training
Lowes 1990 Study type review and descriptions of rider training
Mahon 1976 Study type review of motorcycle rider training methods
Maxwell 1985 Outcomesrecommendations forprogramme operatorsand otherson howthe motorcycle trainingprogramme
works and how it can be improved
Mayhew 1996 Study type review
McKnight 1987 Study type evaluation of methodological and administrative aspects of a motorcycle operator skill test
McPherson 1976 Outcome knowledge and skills
Mohan 2004 Study type review of evidence of interventions to reduce road traffic injuries
Morgan 1980 Study type reviewcourse description
Motoki 1990 Study type proposal and development of rider training curriculum
Nairn 1992 Study type literature review
Newland 1980 Study type description of development of motorcycle safety course
Newland 1982 Study type descriptionreview of a rider training course
Newland 1990 Study type uncontrolled BeforeAfter study
Ormston 2003 Outcome changes in attitude and self reported behaviour
Osga 1979 Outcome knowledge attitudes and skills
Panou 2007 Intervention new simulation tool to teach driverrider
Peterson 1990 Study factor demographic details of course participants completing the course vs those not completing the
course
Prem 1984 Study type experimental
Study factor examines association between skills and performance in tests
Radin Umar 2006 Intervention motorcycle safety programme not specifically rider training
Rockwell 1990 Study type methodologically flawed study type
43Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4650
(Continued)
Rothe 1987 Study type qualitative study of rider training in general
Rowden 2007 Study type qualitative study
RSC Victoria 1993 Study type review of motorcycle rider training and education in Victoria
Russam 1979 Study type review of training programmes in Japan
Saffron 1981 Study type literature review
Schulz 1990 Study factor predictors of motorcycle crashes
Sexton 2004 Study type cross-sectional analytic analysis
Shephard 1986 Study factor reviewdescription of programme
Simpson 1990 Study type review of motorcycle rider trainingeducation
Sokolov 1990 Study factor review of motorcycle riding courses in USSR
Steffens 1988 Outcome self-reported behaviour experiences and researcher observed riding behaviour
Sudlow 2003 Study type review of motorcycle training in the UK
Tasmania 1995 Study type cross-sectional descriptiveprocess evaluation
Tetard 1985 Study type experimental
Outcome detection of indices
Thackray 1980 Outcome instructional effectiveness user acceptance administrative feasibility
Thompson 1994 Study type cross-sectional survey
Outcome attitudes and knowledge
Tobler 1990 Study type review of course curricula
Transport 1989 Intervention syllabus for rider training
Tuan 2007 Study type cross-sectional study (survey) which does not assess the effects of an intervention but examines
motocyclistsrsquo behaviour
Victor 1980 Study type review
Waterer 1990 Study type review of UK rider training issues
Weller 1989 Study type review of studies evaluating motorcycle safety and education measures
44Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4750
(Continued)
Wells 1981 Study type description of different methods of training
Winn 1986 Study type review of rsquohistoryrsquo of rider training
Wisher 1988 Study type cross-sectional survey
Wood 1987 Study type Interrupted Time Series but did not meet the criteria for an ITS as published by the Cochrane
Effective Practice and Organisation of Care Group
Yamazaki 1990 Study type cross-sectional survey
Outcome riding skills skills retention and attitude
Characteristics of ongoing studies [ordered by study ID]
Burgess 2010
Trial name or title UKs National Ride program
Methods Cohort study
Participants Motorcycle offenders
Interventions Police led diversion program
Outcomes Self reported attitudinal and behavioural measures
Starting date
Contact information
Notes Expected to be completed in early 2010
45Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4850
D A T A A N D A N A L Y S E S
This review has no analyses
A P P E N D I C E S
Appendix 1 Search strategy
We searched the following electronic databases
bull Cochrane Injuries Group Specialised Register (searched 22 Sept 2008)
bull CENTRAL (The Cochrane Library 2008 Issue 3)
bull TRANSPORT 1988 to 200706 (latest date available)
bull ATRI Australian Transport Index (to Sept 2008)
bull MEDLINE (Ovid SP) 1950 to Sept 2008
bull EMBASE (Ovid SP) 1980 to Sept 2008
bull CINAHL (EBSCO) 1982 to Sept 2008
bull WHOLIS (World Health Organization Library Information System) (to Sept 2008)
bull PsycINFO (Ovid SP) 1806 to Sept 2008
bull LILACS (BIREME) (Latin American and Caribbean Health Sciences) (1982 to Sept 2008)
bull ISI Web of Science Social Sciences Citation Index (SSCI) 1970 to Sept 2008
bull ERIC (searched 25 Sept 2008)
bull ZETOC (searched 25 Sept 2008)
bull SIGLE (searched 25 Sept 2008)
All of the databases were searched using the following text-word search strategy The strategy was adapted using either rsquorsquo or rsquo$rsquo as atruncation symbol
1 motor-cycl or motorcycl or motor-bik or motorbik or motor-scooter or motorscooter or scooter or moped or motocycl
or motocicl or moto
2 train or educat or instruct or program or course or qualif or learn or improv
3 1 and 2
H I S T O R Y
Protocol first published Issue 2 2005
Review first published Issue 10 2010
46Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4950
C O N T R I B U T I O N S O F A U T H O R S
Katina Kardamanidis was the first author of the protocol conducted the literature search extracted full studies assessed the studies
according to inclusion criteria assessed quality of the studies compared findings with the second author discussed disagreements with
the third author wrote the drafts and the final review
Rebecca Ivers was the second author of the protocol assessed the studies according to inclusion criteria assessed quality of the studies
compared findings with the first author edited the final draft and provided overall guidance for the process
Alexandra Martiniuk assessed quality of the studies compared findings with the first author discussed disagreements with the third
author and helped write the first draft and the final review
Mark Stevenson contributed to the protocol reviewed methods of studies and contributed to the final draft
Katrina Thistlethwaite conducted a second literature search extracted full studies and contributed to the final draft
D E C L A R A T I O N S O F I N T E R E S T
None known
S O U R C E S O F S U P P O R T
Internal sources
bull No sources of support supplied
External sources
bull National Health and Medical Research Council of Australia Australia
Salary support for Martiniuk Ivers and Stevenson
bull University of Sydney Sesqui Research Grant Australia
Direct research costs and salary support for Kardamanidis
D I F F E R E N C E S B E T W E E N P R O T O C O L A N D R E V I E W
There are no differences between the protocol and the review
I N D E X T E R M S
Medical Subject Headings (MeSH)
lowastMotorcycles Accident Prevention [lowastmethods] Accidents Traffic [lowastprevention amp control] Licensure Program Evaluation Wounds
and Injuries [mortality lowastprevention amp control]
47Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 5050
MeSH check words
Humans
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 1150
Table 3 Risk of detection bias (Continued)
Mortimer 1988 Blinding of outcome assessors not reported
Self-report through questionnaires not validated
Intervention group received mailed questionnaires but the control group received ques-
tionnaires personally at the motorcycle dealership
McDavid 1989 Blinding of outcome assessors not reported
Driver records holding police recorded crashes and offences
Kloeden 1994 Blinding of outcome assessors not reported
Driver records holding police recorded crashes and offences
Billheimer 1998 Blinding of outcome assessors not reported
Police recorded crashes and offences
Perrino 2002 Blinding of outcome assessors not reported
Self-report through mailed out questionnaires police records not validated
Kraus 1975 Blinding of intervention assessors to case or control status not reported
Intervention measured police records self-report though questionnaires not validated
Length of follow up not specified
Hurt 1981 Blinding of intervention assessors to case or control status not reported
Intervention measured face-to-face in-depth interviews (ie no structured question-
naires) Validity and reliability of data ensured by reported high level of quality controlLength of follow up not specified
Haworth 1997 Blinding of intervention assessors to case or control status not reported
Intervention measured self-report either by personal or by telephone interview for
both cases and controls not validated
Length of follow up in the 2 years prior to the interview or more than 2 years ago
Savolainen 2007 Blinding of outcome assessors not reported
Source of outcome data not reported
Length of follow up up to 5 years
Davis 1997 Blinding of outcome assessors not reported
Source of outcome data police records
Length of follow up 1 year
Waller 1992 Blinding of outcome assessors not reported
Source of outcome data Department of Public Safety driver history file and official
accident reports
Length of follow up 3 years
MAIDS 2004 Blinding of intervention assessors to case or control status not reported
Intervention measured face-to-face in-depth interviews (ie no structured question-
naires) No report on whether and how validity and reliability of data was ensured
Length of follow up 2 years
9Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 1250
Table 4 Risk of attrition bias
Anderson 1980 Percentage of participants in analysis 99 forthe total group Unclear forintervention andcontrol groups
separately
Hall 1988 Percentage of participants in analysis 96-97 forall intervention groups and 98for the control group
Swaddiwudhipong 1998 Percentage of participating sub-districts in analysis 100 for both intervention and control sub-districts
Raymond 1979 Percentage of participants in analysis 74 for the intervention group and less than 57 for the control
group
Osga 1980 Percentage of participants in analysis 73 for the intervention group and 100 for the control group
Satten 1980 Percentage of participants in analysis is unknown Initial sample size is unclear
Jonah 1982 Percentage of participants in analysis 23 for the intervention group and 45 for the control group
Mortimer 1984 Percentage of participants in analysis 35 for the intervention group and 84 for the control group
Cooper 1988 Percentage of participants in analysis 98 for the total group Unclear for intervention and control
groups separately
Mortimer 1988 Percentage of participants in analysis 40 for the intervention group and 90 in the control group
Adams 1985 Percentage of participants in analysis 61 forthe total group Unclear forintervention andcontrol groups
separately
Leung 1987 Percentage of participants in analysis unclear because initial sample size was not reported
McDavid 1989 Percentage of participants in analysis 40 of final matched cohort
Kloeden 1994 Percentage of participants in analysis 39 for the intervention group and 6 for the control group
Billheimer 1998 Percentage of participants in analysis unclear percentage of the initial sample size
Perrino 2002 Percentage of participants in analysis unclear percentage of the initial sample size
Kraus 1975 Percentage of participants in analysis 58 of cases 59 of controls
Hurt 1981 Percentage of participants in analysis 18 of cases lt 29 of controls
Haworth 1997 Percentage of participants in analysis 70 of cases 47 of controls
Savolainen 2007 Percentage of participants in analysis 18 for the intervention group and 15 for the control group
Davis 1997 Percentage of participants in analysis 100 for the intervention group and unclear for the control group
10Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 1350
Table 4 Risk of attrition bias (Continued)
Waller 1992 Percentage of participants in analysis unclear percentage for the intervention group and 100 for the
control group
MAIDS 2004 Percentrage of participants in analysis unclear for both cases and controls
R E S U L T S
Description of studies
See Characteristicsof included studies Characteristics of excluded
studies Characteristics of ongoing studies
Our search resulted in 102 studies that seemed relevant of which
we excluded 79 studies because they did not meet the criteria for
considering studies for this review (see Characteristics of excluded
studies)
Basedon ourcriteriawe includeda total of 23 studies in this review
There were three randomised trials (Anderson 1980 Hall 1988
Swaddiwudhipong 1998) two non-randomised trials (Adams
1985 Leung 1987) 14 cohort studies (Raymond 1979 Osga
1980 Satten 1980 Jonah 1982 Mortimer 1984 Cooper 1988
Mortimer 1988 McDavid 1989 Waller 1992 Kloeden 1994
Davis 1997 Billheimer 1998 Perrino 2002 Savolainen 2007)
and four case-control studies (Kraus 1975 Hurt 1981 Haworth
1997 MAIDS 2004)
Over half of the studies were conducted 20 or more years
ago and the oldest was conducted 31 years ago Only three
studies were from this decade (Perrino 2002 MAIDS 2004
Savolainen 2007) Over 60 of the studies were conducted in
the USA (Kraus 1975 Anderson 1980 Osga 1980 Satten 1980
Hurt 1981 Mortimer 1984 Leung 1987 Hall 1988 Mortimer
1988 Waller 1992 Davis 1997 Billheimer 1998 Perrino 2002
Savolainen2007) three inAustralia (Adams 1985 Haworth 1997
Kloeden 1994) three in Canada (Jonah 1982 Cooper 1988
McDavid 1989) onefromthe United Kingdom (Raymond 1979)one from Europe (with sampling areas located in France Ger-
many Netherlands Spain and Italy) (MAIDS 2004) and one
in Thailand (Swaddiwudhipong 1998) Only seven studies were
published in peer reviewed journals (Kraus 1975 Jonah 1982
Mortimer 1984 Cooper 1988 Mortimer 1988 McDavid 1989
Swaddiwudhipong 1998) The other studies were reports forhigh-
way safety and road traffic organisations
The studies by Anderson 1980 Hall 1988 Kloeden 1994
Billheimer 1998 (the basic California Motorcyclist Safety Program
component of the study) and Perrino 2002 examined manda-
tory pre-licence training Cooper 1988 McDavid 1989 and
Swaddiwudhipong 1998 assessed non-mandatory training that
when passed resulted in a licence Raymond 1979 Osga 1980
Satten 1980 Jonah 1982 Mortimer 1984 Adams 1985 Leung
1987 Mortimer 1988 Waller 1992 Davis 1997 Billheimer 1998
(the Experienced Rider Course component of the study) andSavolainen 2007 assessed non-mandatory training that could be
taken by motorcycle riders with and without a licence The case-
control studies (Kraus 1975 Hurt 1981 Haworth 1997) assessed
lsquoanyrsquo type of rider training and therefore could not be grouped by
pre- or post-licence rider training The MAIDS 2004 case-control
study did make a distinction between participants having received
compulsory pre-licence training or additional (non-compulsory)
training We did not find a study that compared one form of pre-
or post-licence training (eg on-road off-road theory or a com-
bination) versus another form of training All studies compared
eithermandatory or non-mandatory formal training versus no for-
mal training with the exception of Haworth 1997 who compared
different levels of training courses (beginner intermediate and ad-vanced)
We have presented a more detailed description of the individual
studies in Characteristics of included studies We have provided
a description below of the types of motorcycle rider training that
were evaluated in the included studies
Types of motorcycle rider training evaluated in the
studies
The types of training were varied Duration of training ranged
from two or three hours to 24 hours administered over two or
three days Most training consisted of a theoretical and a practical
component usually off-road The content ranged from provid-ing riders with manuals lectures discussions showing films and
administering knowledge and skill tests Little information was
provided about the qualifications of trainers who conducted the
training
Six studies (Osga 1980 Satten 1980 Mortimer 1984 Mortimer
1988 Waller 1992 Savolainen 2007) assessed the effectiveness of
the USA-based Motorcycle Safety Foundationrsquos Motorcycle Rider
Course (MSF MRC) This course was released in 1976 It com-
prises eight hours of theory (lectures discussions films and multi-
plechoice tests) a textbook and 12 hours of off-road practice The
Motorcycle Safety Foundation is a national not-for-profit organ-
11Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 1450
isation sponsored by manufacturers and distributors of motorcy-
cles Many USA states have adopted the MSF Basic Rider Courseand Experienced Rider Course (wwwmsf-usaorg) since its release
in 1976
The Motorcycle Operators Skill Test was also developed
by the USA Motorcycle Safety Foundation in coopera-
tion with others (httpwwwwebbikeworldcommotorcycle-
trainingmotorcycle-test) The MOST is a licensing test consist-
ing of a manual knowledge and skill test (comprising eight exer-
cises of increasing difficulty and skill requirements) and a mini-
skill test conducted with a simulator Some studies compared this
alternative licensing test to the standard licensing test However
as the MOST is not a type of rider training we have included only
the two studies that assessed the MOST in combination with one
or more training components in this review Both Anderson 1980and Hall 1988 assessed the MOST in combination with a train-
ing component of three hoursrsquo duration consisting of instructions
about the skills needed to pass the MOST (30 minutes of the-
ory and two and a half hours on the motorcycle) Hall 1988 also
assessed the MOST in combination with the Motorcycle Safety
Foundationrsquos Motorcycle Rider Course as described above
Other USA studies evaluated the Wisconsin motorcycle rider
course a 22-hour off-street programme that teachesthe basic rules
of safe motorcycling (Leung 1987) the USA Maryland Motorcy-
cle Rider Course (MRC) a mandatory (for those younger than 18
years)pre-licence 20-hour coursecomprising eight hours of theory
and12 hours of practice and created by the Maryland Motor Vehi-
cle Administration in 1985 (Perrino 2002) the California Motor-cyclist Safety Program (CMSP) a mandatory (for those younger
than 18 years) pre-licence 16-hour basic course (Billheimer 1998)
and Experienced Rider Course (ERC) of eight hoursrsquo duration
(Billheimer 1998) The CMSP still offers these courses ( http
wwwca-msporg) but they are now MSF courses as described
above Davis 1997 did not give any information about the assessed
rider training
The Canadian studies evaluated the British Colombia Safety
Councilrsquos 37-hour motorcycle safety programme (no further in-
formation on the content was provided) (McDavid 1989) and
the Canada Safety Council Motorcycle Training Program (MTP)
comprising four hours of theory and 16 hours of practice (six
hours on-road) over twoweekends (Jonah 1982) The MTPcoursewas made available in 1974 and is still being offered in Canada
(httpwwwsafety-councilorgtrainingmtpmtphtm) Cooper
1988 did not give any information about the assessed rider train-
ing However the study was conducted in British Colombia in
Canada in 1988 the same time as the McDavid 1989 study was
conducted
In the United Kingdom Raymond 1979 assessed the non-manda-
tory RACACU (Royal Automobile ClubAuto-Cycle Union)
training scheme The training programme consisted of either 24
one-hour sessions or 13 two-hour sessions Twelve hours of the
training focused on theory about basic operational procedures
traffic laws and highway codes The practical session focused on
basic operational skills riding on a training area and riding in traf-fic with an instructor A proficiency test was administered at the
end of the course
One study from rural Thailand (Swaddiwudhipong 1998) evalu-
ated a community-based rider education programme consisting of
theory on motorcycle injury risk helmet use traffic laws vehicle
regulations and traffic signs and a written and skill test leading to
a licence
TheAustralian studies evaluateda rider trainingcourse of 12 hours
over two consecutive days (four hours of theory and eight hours of
practice)administeredto AustraliaPosttelegram deliverers(Adams
1985) and a mandatory pre-licence Ridersafe course consisting
of two tiers the first tier being a training course on the basics
of motorcycle riding where a pass entitled the rider to obtain alearnerrsquos permit and the second tier being a skill test where a pass
entitled the riderto obtain a probationary motorcycle licence The
training course consisted of two four-hour training sessions on
separate days involving instruction videos and actual riding on
motorcycles The skill test consisted of a four-hour preparation
course of 74 minutes of lecturediscussion 38 minutes of video
and 75 minutes of on-course riding which was the testing phase
(Kloeden 1994) The Rider Safe course is still being provided by
Transport South Australia a state government organisation (http
wwwtransportsagovau )
The duration or content of rider training in the four case-control
studies (Kraus 1975 Hurt 1981 Haworth 1997 MAIDS 2004)
was not described
Risk of bias in included studies
We describe the methodological quality of the included studies
by study type below More detailed information can be found in
Table 1 Table 2 Table 3 and Table 4
When considering the findings from these studies it is important
to note the quality (or lack thereof) of reporting of the studies
Quality of reporting is the extent to which a report provides in-
formation about the design conduct and analysis of a study ( Juumlni2001) Huwiler-Muumlntener 2002 found that even though trials
may be well reported it does not mean that the methodological
quality is of equal standard and well-conducted trials may never-
theless be reported badly The reporting quality of most studies in
this review was poor Most studies were old and did not meet con-
temporary reporting standards This made data extraction com-
plicated It also made it difficult to contact authors for additional
information
Because of significant heterogeneity between studies as well as the
poor quality of most studiespooling of studyresults to calculatean
overall measure of effect was not appropriate Instead we provided
a description of our judgement of risk of bias in the text
12Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 1550
Risk of bias in the randomised trials
There were three randomised trials (Anderson 1980 Hall 1988Swaddiwudhipong 1998) Anderson 1980 and Hall 1988 had an
initial sample size of 41188 and 26464 motorcycle licence ap-
plicants respectively Both studies assessed the effectiveness of the
pre-licence MOST with a trainingcomponent Swaddiwudhipong
1998 evaluated the effectiveness of a community-based rider ed-
ucation programme in six districts
Allocation bias
Anderson 1980 reported using random birth date tables to allocate
participants to either the intervention or control group Hall 1988
reported the use of a computerised algorithm set in a central of-
fice but also used manual random assignment when the computer
went down Swaddiwudhipong 1998 reported lsquosimple random
samplingrsquo for the delivery of the intervention where the sampling
frame included nine districts Three districts were excluded since
they were too hard to access The remaining six districts were ran-
domly allocated to the intervention (three) and the control (three)
groups Systematic sampling wasthen used ordered by village size
for the villages within districts Allocation concealment was not
reported in any of the studies The descriptions of randomisation
in these studies suggest that there may have been limitations to
the process resulting in a lack of equivalence of the intervention
and control groups at baseline and therefore the possibility of
allocation bias Swaddiwudhipong 1998 reported that before theintervention the annual incidence rate of motorcycle-related in-
juries was slightly (but not statistically significant) higher in the
intervention areas than in the control areas Participants in the
intervention districts were also more likely to have a valid licence
and to wear a helmet more often than participants in the control
districts Anderson 1980 measured differences at baseline on the
participant and prior driving record variables and found these to
be negligible Hall 1988 did not report on baseline differences
between the intervention and control groups As the Anderson
1980 and Hall 1988 studies had sample sizes of over 40000 and
26000 respectively and used some form of randomisation it is
likely that any confounding factors at baseline would have been
equally distributed between comparison and intervention groups
Selection bias
After randomisation Anderson 1980 and Hall 1988 excluded
08 and 3 participants respectively because of assignment to
the wrong group administration of the wrong intervention or
other reasons The resulting lack of analysis by rsquointention-to-treatrsquo
can introduce biasby removing the effectivenessof randomisation
Nonetheless the percentage excluded was small and the number
of overall participants large so it is unlikely to have had a major
effect on the final results In the Swaddiwudhipong 1998 study
villages were systematically sampled into intervention or control
groups The intervention was implemented and outcomes of in-
jury and death were measured at a community level Interventionand control districts were randomly selected however there were
only three intervention and three control communities
Detection bias
None of the randomised trials reported blinding of outcome as-
sessors Data collection tools used by Anderson 1980 and Hall
1988 were driver records holding police recorded crashes and of-
fences Swaddiwudhipong 1998 used police and hospital records
at a district level and post-mortem examination for deaths None
reported on validity or reliability for any of the instruments
Attrition bias
Anderson 1980 and Hall 1988 had driver records available for
96 to 100 of participants They also collected self-reported
data but the response rate to the questionnaires was very low (low-
est response rate 42 and 30 respectively) Attrition bias in
Swaddiwudhipong 1998 was not an issue as outcome data were
available for all sub-districts included in the trial
The intervention reported in Anderson 1980 and Hall 1988 tri-
als had an unanticipated effect on the participants In both trials
an average of only 28 of licence applicants in the intervention
group eventually obtained their licence compared to 35 in the
control group This indicates that the licensing process in generalseems to have deterred applicants from obtaining a licence and
although no statistical tests of the difference were conducted this
deterrence appears even greater for those who had rider training
This is not true attrition bias as the studies did not experience
differential loss to follow up from outcome measurement Instead
differential attainment of the intermediary step of licensure pre-
cluded participants from being at risk for the outcomes of interest
they are not driving a motorcycle hence they are not at risk for
motorcycle-related crashes injury death or offences In short it is
not possible to assess whether a reduction in crashes in the inter-
vention group would be due to changes in riding habits as a result
of training or simply due to delayed or never attained licensure
Intervention integrity
Neither Anderson 1980 nor Hall 1988 reported on the consis-
tency in the way the intervention was administered the possible
lsquocontaminationrsquo of participants in the intervention group by other
types of formal or informal rider training or of participants in the
control group by the intervention Swaddiwudhipong 1998 did
not report on the number of villagers in each group that actually
attended the intervention However they did report on a number
of other potential training courses that were being implemented
at the same time
13Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 1650
Analysis
Anderson 1980 and Hall 1988 acknowledging the lsquodeterrent ef-fectrsquo described above conducted two separate analyses The first
analysis included the total study population of licence applicants
over two years from the time of their application (the population
denominator) The second analysis was performed on the licensed
(intervention and control) subgroups only two years from the
time of obtaining their licence (the licensed rider denominator)
As only a third were licensed and exposed to riding the motorcy-
cle it was impossible to determine whether reported differences
in crash rates across groups were due to the intervention or to
other lsquoconfoundingrsquo characteristics of those who were determined
in their efforts to obtain the licence When analysis was restricted
only to those who gained their licence the same difficulties re-
mained The benefits of the randomisation process were thus lost
due to this lack of licensure of riders during the process As there
were only three intervention and three control groups in the study
reported by Swaddiwudhipong 1998 analyses were hindered by
smallnumbers for outcome comparisons betweenthe intervention
and the control districts
Risk of bias in the cohort studies and non-randomised
trials
We discuss the methodological quality of cohort studies (seven
prospective studies (Raymond 1979 McDavid 1989 Waller 1992
Kloeden 1994 Billheimer 1998 Perrino 2002 Savolainen 2007)and seven retrospective studies (Osga 1980 Satten 1980 Jonah
1982 Mortimer 1984 Cooper 1988 Mortimer 1988 Davis
1997)) and non-randomised trials (Adams 1985 Leung 1987) in
the one section as the biases for the two study types are similar
The quality of these studies was in general poor
Selection bias
The confounding factors expected to have been addressed in the
studies were sex age riding experience prior driving records (of-
fences crashes) previous training and the following proxies for
risk taking behaviour helmet use speed and alcoholdrug use All
of these factors have been shown to be associated with an increasedrisk of motorcycle crash (Peden 2004 Zambon 2006) The pos-
sible source of bias in cohort studies and non-randomised trials
is that groups may differ on important characteristics at baseline
participants lsquoself-selectrsquo into either the rider training group by de-
ciding to undertake a rider training course or into the lsquocontrol
grouprsquo by deciding not to undertake a course Participants who
undertake a non-mandatory rider training course may differ from
those who choose not to on characteristics such as age sex rid-
ing experience and riding behaviour as well as other possible con-
founders For example riders who take a training course may be
more safety-conscious or on the contrary may do so because they
have previously experienced a crash or because they perceive that
they have poor riding skills Therefore ideally all participants in
the intervention group need to be included in the study as wellas a random sample of controls taken from the same population
as the intervention group If that is not possible intervention and
control groups either need to be carefully matched to make them
comparable or differences at baseline need to be adjusted for in
the analysis
Most of the studies using a non-randomised trial or cohort design
compared an intervention group to a control group that differed
on a number of important factors The studies in general failed to
adjust for important differences found at baseline and their results
cannot be meaningfully interpreted Ten studies (Raymond 1979
Osga 1980 Satten 1980 Jonah 1982 Mortimer 1984 Mortimer
1988 McDavid 1989 Waller 1992 Davis 1997 Savolainen 2007)
assessed intervention and control groups that were not drawnfrom the same population In most of these cases the intervention
group was selected from a list of training graduates who were usu-
ally young female and less experienced riders whereas controls
were recruited from a general population of motorcycle riders who
were usually older male and experienced riders For three stud-
ies it was unclear how they were selected (Leung 1987 Cooper
1988 Perrino 2002) Only Kloeden 1994 recruited groups from
the same population of motorcycle applicants and Adams 1985
from a group of professional riders Nevertheless due to non-ran-
dom selection of intervention and control groups differences at
baseline are likely even though study groups were from the same
population Three studies adjusted for some confounding fac-
tors found at baseline through matching (McDavid 1989 Waller1992 Billheimer 1998) However matching has limitations in
termsof controlling for multiple confounders especially those that
are unknown to influence the effect of rider training on the out-
comes One study matched the control and intervention groups
by excluding controls (approximately 3500 (66 of initial con-
trol group) that differed from the intervention group) (Kloeden
1994) Five studies (Osga 1980 Satten 1980 Adams 1985 Davis
1997 Perrino 2002) did not adjust for differences between study
groups at baseline either in the study design or analysis Results
from these studies are therefore not reliable The other six stud-
ies (Raymond 1979 Jonah 1982 Mortimer 1984 Cooper 1988
Mortimer 1988 Leung 1987 McDavid 1989 Billheimer 1998)
adjusted for some differences between groups at baseline by eitherbi-variate or multi-variate analyses or stratification and standardi-
sation of riding exposure (outcomemiles driven) or stratification
by age group The limitation of stratification is that it is difficult to
control for many variables simultaneously because a large number
of strata will need to be generated resulting in small numbers that
cannot be meaningfully analysed Savolainen 2007 developed sta-
tistical models to assess the probability of a crash after taking the
rider training course in an intervention and control group while
examining differences by sub-groups (eg age) Participation rates
were also very low in this study with only 5584000 (1395)
in the eligible intervention group and 5884000 (147) for the
14Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 1750
control group completing surveys
Detection bias
None of the studies reported blinding of outcome assessors to the
intervention or control status of the participants Seven studies
used driver records containing police recorded data as their main
measure of outcome (Leung 1987 Cooper 1988 McDavid 1989
Waller 1992 Kloeden 1994 Davis 1997 Billheimer 1998) An-
other four studies (Raymond 1979 Satten 1980 Mortimer 1984
Mortimer 1988) only used self-report Raymond 1979 was the
only study using self-report as the main outcome measurement in
which study instruments were validated Self-report might prevent
the underreporting that has been described for the use of police-
driver records (see rsquoRisk of biasrsquo in the randomised trials above)but has other risks of bias especially if questionnaires or inter-
view methods have not been validated and tested for reliability
Recall bias arises from mistakes in recollecting events both be-
cause of failures of memory and because of looking at things lsquowith
hindsightrsquo and possibly changed views (httpwwwcochraneorg
resourcesglossaryhtm) As many studies were retrospective risk
of recall bias is likely and may have increased with length of follow
up For studies that used self-reported outcomes only it is also
possible that social desirability response bias played a role where
participants in the training group may have felt it was lsquoexpectedrsquo
to report good driving histories Multiple methods to measure the
same outcome for both groups decreases the risk of detection bias
for example because self-reported answers can be cross checkedwith other sources of information such as driver records
In this regard three studies (Osga 1980 Jonah 1982 Perrino
2002) used both self-report and driver records Osga 1980 found
that for both intervention and control groups more crashes were
reported through self-report than on state crash records an aver-
age of 54 of self-reported crashes could be found back on state
crash records One industry-based study (Adams 1985) used oc-
cupational health records which were not validated but which are
unlikely to suffer from underreporting due to work cover claims
Savolainen 2007 used information on rider crashes for a five-year
period but did not specify from where this information was ob-
tained
Attrition bias
The majority of studies (Raymond 1979 Satten 1980 Adams
1985 Leung 1987 Kloeden 1994 Davis 1997 Billheimer 1998
Perrino 2002) inadequately reported on losses of participants to
the study We can therefore not be certain that there were no sys-
tematic differences in the characteristics between the intervention
and control group participants Savolainen 2007 reported the re-
sponse rates to their baseline survey but did not report on the per-
centage of outcome data that was available for their study sample
A similar deterrent effect of rider training as described for the ran-
domised trials (Anderson 1980 Hall 1988) occurred in two stud-
ies Jonah 1982 reported that 65(14992310) of the trained rid-
ers stopped riding compared to 47 (9492029) of the untrainedriders They excluded non-active and non-licensed riders before
the study commenced so the differential choice to continue riding
after training did not affect their analyses Leung 1987 reported
that the percentage of participants in the intervention group ob-
taining their licence was 50 (no number provided by authors)
lower than in the control group This may have biased the mea-
surement of outcomes because unlicensed participants will not be
exposed to riding the motorcycle and therefore not be at risk of
being involved in a crash or violation As explained above this
makes it difficult to assess whether the reduction in crashes in the
intervention group is due to a change in driving habits or simply
due to delayed licensure
Mortimer 1984 and Mortimer 1988 describe that about 44 and30 of course graduates respectively were excluded from analysis
because they had not ridden a motorcycle in the previous year
However because these studies were retrospective and the controls
were activeriders at the time of recruitment a decision on whether
the deterrent effect was differential between the two groups could
not be made Though not reported quantitatively Kloeden 1994
also describes the effectof rider training deterringindividuals from
becoming licensed and hence exposure to riding and risk of crash
death or offences involving a motorcycle Osga 1980 reported
that 27 of the trained riders did not ride after the course and
excluded these from the analysis
Use of different data collection strategies may have contributed to
differential loss to follow up in the cohort studies Satten 1980for example interviewed the intervention group over the tele-
phone but mailed questionnaires to the control group Mortimer
1984 and Mortimer 1988 distributed questionnaires in person to
the control group but mailed them to the intervention group
This may have played a role in the observed differential response
rates where more subjects responded in the personally approached
groups
Survival bias
For those studies that assessed offences crashes and injuries but
not death survival bias may have also played a role Motorcyclists
who were fatally or seriously injured in crashes during the follow-
up period after the intervention would have been less likely or
unable to respond to surveys This could bias the effect measures
towards or away from the null None of the studies wrote about
this issue or examined the characteristics of those lost to follow up
in order to address the possibility of survival bias
Intervention integrity
Although the interventions were intended for the entire lsquoexposedrsquo
groups most studies reported difficulty in tracking participants
whoreceivedthe interventionand those whodid not Adams 1985
15Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 1850
estimatedthat 62of thetotalstudysample received the interven-
tion The remaining studies did not report the percentage of thestudy population that received the intervention as was intended
or reported that this percentage was unknown (Raymond 1979
Kloeden 1994 Satten 1980 Jonah 1982 Mortimer 1984 Leung
1987 Cooper 1988 Mortimer 1988 Billheimer 1998 Perrino
2002 Savolainen 2007) Only Raymond 1979 reported the degree
to which their intervention was delivered as intended No studies
reported on possible lsquocontaminationrsquo of participants by other inter-
ventions than the one to which they were allocated Because Osga
1980 Satten 1980 Jonah 1982 Mortimer 1984 Cooper 1988
Mortimer 1988 Waller 1992 and Davis 1997 were retrospective
studies and participants were recruited from a list of graduates it
was known that they had completed and passed the course
Analysis
In some studies some participants who reported at baseline that
they were untrained decided to enroll in a rider training pro-
gramme after the baseline survey The non-randomised trial by
Adams 1985 experienced this move of subjects from the untrained
to the trained group but failed to conduct analyses by lsquointention-
to-treatrsquo This can bias the results in randomised and non-ran-
domised trials if participants who changed groups differ signifi-
cantly with regard to potentially confounding characteristics On
the other hand Perrino 2002 also experienced this move of sub-
jects from theuntrained to the trained groupmore than 100riders(5 of the total sample) originally stated that they were untrained
at baseline but then received training between the baseline and
follow-up surveys Yet since the Perrino 2002 study was a cohort
their analyses then proceeded by including those riders who were
lsquonewlyrsquo trained as exposed Riders who were lsquonewlyrsquo trained since
baseline but with a programme other than the one under study
were excluded from the analysis Nevertheless confounding fac-
tors were not accounted for in the analyses
McDavid 1989 matched exposed with unexposed subjects but
could not find suitable matches for the majority (60) of their
subjects They could therefore only conduct analysis on the result-
ing40 (139346) which is an inappropriate andinefficient anal-
ysis Some studies that reported on the issue of non-active licensedriders(Satten 1980 Jonah 1982 Mortimer 1984 Mortimer 1988
Kloeden 1994) excluded non-active riders from analysis intro-
ducing selection bias as it is likely that riders who continued to
ride differed with regards to potentially confounding character-
istics across intervention and control groups Kloeden 1994 for
example conducted several analyses to assess the effect of various
sub-sets of the control group Original analyses included all con-
trols and led to erroneous results since only a small proportion of
controls hadgone on to get a licence after their learnerrsquos permitand
were actually exposed to the possibility of experiencing a crash
Subsequent analyses compared only intervention and control sub-
jects who had a licence Waller 1992 reported crashes by person-
days of exposure but did not explain how they could be sure that
each study intervention and control participant was an active riderand actually exposed to the risk of experiencing a crash
Potential confounders that are unequally distributed by study arm
at baseline need to be dealt with in the analysis Satten 1980
Adams 1985 and Davis 1997 did not take into account any con-
founding factors either by design or in the analysis McDavid
1989 Waller 1992 and Billheimer 1998 accounted for potential
confounding factors at the design stage by matching the subjects
in the intervention and control groups Other studies (Raymond
1979 Mortimer 1984 Mortimer 1988 Perrino 2002 and also
Waller 1992) adjusted for one or two confounders by stratifica-
tionor standardisation (eg crashesmiles driven age group) Osga
1980 conducteda univariate analysis of crash rates in which no po-
tential confounders were accounted for but also calculated cumu-lative crash rates by mileage travelled Only Jonah 1982 Cooper
1988 and Savolainen 2007 conducted a multivariate analysis
Risk of bias in the case-control studies
A case-control design is suitable for the assessment of risk fac-
tors for outcomes such as motorcycle crashes injuries or deaths
but not appropriate for the evaluation of an intervention such
as motorcycle rider training It can potentially result in spurious
conclusions regarding the efficacy of the intervention because the
intervention has not been applied at random to an intervention
and control group Even with the control of possible confounders
using a case-control design to evaluate an intervention can lead tothese kinds of results since those who choose to undertake rider
training may differ on important prognostic factors that affect the
outcome compared to those who do not choose to participate in
rider training
Selection bias
A potential source of bias in case-control studies as with all other
study types where the intervention is not randomly allocated to
participants is that comparison groups are likely to differ on po-
tentially confounding characteristics This means that ideally all
cases need to be included in the study as well as a random sample
of controls taken from the population that gave rise to the cases If that is not possible cases and controls either need to be carefully
matched to make themcomparable or differences at baseline need
to be adjusted for in the analysis All four case-control studies were
population based Kraus 1975 included all fatalities and injured
cases leading to hospitalisation and appropriately recruited a ran-
dom sample of controls from the same population Hurt 1981 did
not manage to recruit all or a random sample of cases Only half
of the notifications of crashes to the police were received in time
by the researchers the research team for unreported reasons was
only able to respond to half of those again Data collection was
therefore completed for only about a fifth of all cases Neverthe-
less this was still a large sample of 900 cases Haworth 1997 using
16Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 1950
a similar study design was more likely to have included all cases
but this is not entirely clear from the study reported It is unclearfor example whether fatalities were included as they did not use
coronerrsquos reports for the recruitment of cases The MAIDS 2004
study was conducted in five European countries but it was unclear
how cases were recruited whether they were all cases during the
two years that the study lasted or a random sample of all cases
what the response rates were etc Controls were recruited from
randomly selected petrol stations but it was likewise not clear what
the eligibility criteria for inclusion in the study were how they
were recruited (eg randomly) and what the response rate was
Hurt 1981 and Haworth 1997 both matched controls to cases by
recruiting themat the sametime ofdayandsamedayof weekas the
crash occurred Hurt 1981 however recruited controls two years
later andit is likelythat there would have beendifferences betweenthe two comparison groups One important difference was that
during data collection for cases there was very little specialised
motorcycletrainingavailable in the studyregionwhereas two years
later when control data were collected there was some training
available
These four case-control studies also suffered from selection bias
due to potential differences between those that gave consent to
participate in the study and responded to interviews andquestion-
naires and those that did not None of the studies have assessed
such differences although Hurt 1981 did acknowledge that con-
trols giving consent to cooperate may have differed from those
motorcyclists who did not
Detection bias
Measures that can be taken to reduce detection bias in case-con-
trol studies are blinding of interviewers for case or control status
blinding of the participant for the study factor of interest validat-
ing measurement instruments and using these in an identical way
for cases and controls and cross checking self-reported informa-
tion with other more objective sources such as registers None of
the studies reported blinding of the intervention assessors to case
or control status but all used consistent measurement methods for
both case and control groups
All case-control studies used self-report as their main measure toassess whether cases and controls had received rider training and
if so the type of training The major potential source of bias in
case-control studies when self-report is used as a measure is recall
bias One way of reducing recall bias is by double-checking self-
reported data with more objective data Kraus 1975 also used po-
lice accident reports for information on rider characteristics but
such information was not available about rider training Haworth
1997 and MAIDS 2004 did not report on any measures to limit
recall bias Hurt 1981 reported further questioning of cases by
the interviewer when answers did not make sense or when par-
ticipants seemed to contradict themselves as part of the in-depth
unstructured interview technique This qualitative research tech-
nique can result in valuable information but as interviewers were
not blinded to case or control status of the participants it may have led to more thorough examination of cases compared to the
controls and thus differential detection bias
Intervention integrity
The level of detail about the rider training provided in the case-
control studies differed significantly but was minimal for each
Kraus 1975 simply asked whether lsquoanyrsquo motorcycle rider training
had lsquoeverrsquo been carried out Hurt 1981 asked participants whether
they hadbeen self taught informally taught or completed formally
taught rider training courses MAIDS 2004 asked whether partic-
ipants had had pre-license training additional (non-compulsory)
training other training or no training whereas Haworth 1997asked about the number of courses that a participant had com-
pleted whetherthey were basic intermediate or advanced courses
and how long ago the training had been done None of the case-
control studies reported collecting information about further de-
tails about any training programmes (eg type of training cur-
riculum duration of training)
Analysis
Kraus 1975 accounted for confounding factors through the exclu-
sion of females and stratification by age and frequency of motor-
cycle use Hurt 1981 matched cases and controls for time of day
and day of the week that the crash occurred and did not adjust forany confounding factors in the analysis Haworth 1997 matched
groups for time of day and day of the week that the crash occurred
and stratified by age blood alcohol level and licence status in the
analysis None of the studies used multivariate methods MAIDS
2004 did not provide any information about any participant char-
acteristics other than the study exposures of interest and did not
describe any methods to adjust for potential confounding factors
Effects of interventions
We have presented the results in a descriptive manner due to the
substantial heterogeneity in study types interventions and out-comes and the generally poor quality of the studies We have
grouped the interventions by mandatory and non-mandatory pre-
licence rider training and post-licence rider training and further
by study type and by outcome (crashes injurious and fatal crashes
and offences) We discuss results that are reported as statistically
significant (whether a P value was reported or not)
Pre-licence rider training - mandatory
Randomised trials
17Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2050
Crashes
Total group of participants
Anderson 1980 found a reduction in motorcycle crash rates in
the intervention group (MOST skill test with remedial training)
compared to the control group (who received the skill testwithout
a training component) The percent differenceranged from 143
six months after the programme (P lt 005) to 211 (P lt 0001)
after one year and 127 (P lt 001) after two years Hall 1988
found no statistically significant difference in crash rates between
intervention groups at any point in time
Subgroup of participants who obtained their licence only
Anderson 1980 found a reduction in motorcycle crash rates in the
intervention arm but this was statistically significant at one year
only (24 to 218 reduction P lt 005) Hall 1988 found no
statistically significant differences
Offences
Total group of participants
The percentage of offences in Anderson 1980 after six months
and one and two years was slightly higher in the intervention
arm compared to the control group but these differences were
not statistically significant Likewise the study by Hall 1988 did
not find a statistically significant difference in offences between
groups
Subgroup of participants who obtained their licence only
Hall 1988 found no significant difference in offence rates between
the intervention and control groups Anderson 1980 found thatthe percentage of offences increased in the intervention arms com-
pared to the control group This difference was statistically signif-
icant after two years (51 to 115 increase P lt 005)
Cohort studies
Crashes
Neitherof the cohort studies (Kloeden1994 and Billheimer1998)
found a statistically significant difference in crash rates between
the intervention and control groups
Offences
Billheimer 1998 (for the California Motorcyclist Safety Program
course) and Perrino 2002 did not find a statistically significant
difference in offence rates between the intervention and control
groups
Case-control studies
Crashes
MAIDS 2004 found that a slightly higher percentage of controls
(not involved in a crash) had had pre-license rider training than
cases (involved in a crash) This difference was not statistically significant (50 of controls vs 477 of cases OR 091 95 CI
076 to 110)
Pre-licence rider training - non-mandatory
Randomised trial
Injuries and deaths
Swaddiwudhipong 1998 found a significantly lower injury ratein the intervention villages (1051000 population 95 CI 92
to 120) compared to the control villages (1691000 population
95 CI 152 to 185) one year after the intervention This differ-
ence was not maintained two years after the intervention (127
95 CI 112 to 143 intervention 151 95 CI 135 to 167
control) There was no significant difference in death rates be-
tween groups
Cohort studies
Crashes
Cooper 1988 foundno statistically significant differencein crashes
between the intervention and control groups McDavid 1989
found a lower crash rate in the intervention group (14139) com-
pared to the control group (23139) but no test of statistical sig-
nificance was undertaken
Injuries and deaths
McDavid 1989 found that the intervention group had fewer in-
jurious crashes (814 crashes) compared to the control group (16
23 crashes) but no statistical significance test was reported
18Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2150
Offences
Cooper 1988 foundno statistically significantdifference in offencerates between the intervention and control groups McDavid 1989
foundno statistically significantdifference in offence ratesbetween
the intervention and control groups
Non-mandatory rider training
Non-randomised trials
Crashes
Neither Adams 1985 nor Leung 1987 found a statistically signif-icant effect of rider training on crash rates
Offences
Leung 1987 found a statistically significant lower offence rate in
the intervention group compared to the control group for both
males and females and for all age groups except for women 35
years or older (P values not reported)
Cohort studies
Crashes
Osga 1980 in the univariate analysis reported a lower percentage
of crashesin theintervention group (that continued to ride) (28
168595) than in the control group (34 112329) In the cu-
mulative crash rate analysis however Osga1980 found statistically
significant (at the P lt 001 level) higher crash rates in the interven-
tion group than in the control group at allmileage intervals (1000
to 29999 miles) travelled After riding more than 30000 miles
(when over 50 of the trained riders had been involved in a crash
compared to approximately 40 in untrained riders) this differ-
ence was no longer statistically significant Satten 1980 found a
lower percentage of crashes in the intervention group (13 969)
compared to the control group (25 2079) but the differencewas not assessed statistically Davis 1997 found a lower crash rate
in trained riders but it was unclear how this was calculated be-
cause the size of the control group was not reported Jonah 1982
and Billheimer 1998 (for the Experienced Rider Course) found no
statistically significant difference in crashes between the interven-
tion and control groups Although the difference between those
who completed the Experienced Rider Course and no course in
Billheimer 1998 neared significance at six months after the course
(P = 007) large P values were observed for one year after (P =
030) Raymond 1979 found a statistically significant higher crash
rate for the intervention group (1861000 miles ridden) com-
pared to the control group (1181000 miles ridden) in both males
and females (P lt 0001) especially during the first 5000 miles of
riding Waller 1992 found a statistically significant higher crashrate for the intervention group (27 crashes in three years or 070
crashes10000 person days) compared to the control group (5
crashes in three years or 013 crashes10000 person days) with a
crude relative risk of 538 (X 2=1513 plt005) Mortimer 1984
and Mortimer 1988 found no statistically significant difference in
crashes between the intervention and control groups
Savolainen 2007 found that the intervention group (those who
completed the Basic Rider Course) was 44 more likely to be
involved in crashes than the control group and those who com-
pleted the course multiple times were an additional 180 more
likely to be involved in a crash This finding was based on low
numbers and no P value or confidence level was given for either
result
Injuries and deaths
Satten 1980 also found a lower percentage of injurious crashes
(33 39) in the intervention group compared to the control
group(60 1220) although again no statistical significance test
was reported
Offences
Jonah 1982 did find a significantly lower self-reportedoffence rate
in the intervention group (15) compared to the control group(34 P lt 001) (no numbers given) Satten 1980 also found
a lower self-reported offence rate in the intervention group (1
71 versus 19100) but again this was not tested for statistical
significance Mortimer 1984 Mortimer1988 and Billheimer 1998
found no statistically significant differencein offence ratesbetween
the intervention and control groups
Case-control studies
CrashesMAIDS 2004 found that a slightly higher percentage of cases
(involved in a crash) had had additional non-mandatory rider
training than controls (not involved in a crash) This difference
was not statistically significant (17 of cases vs 12 of controls
OR 146 95 CI 064 to 337)
Rider training unspecified
Case-control studies
19Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2250
Injuries and deaths
Hurt 1981 found that the odds of injurious crashes were lower inthose who had received professional training compared to those
who had not Kraus 1975 found that in the younger age group
(less than 20 years) and the older age group (over 35 years) a larger
percentage of injured riders (12 and 21 respectively) reported
prior motorcycle rider training compared with the uninjured con-
trol group (6 and 5 respectively) (P lt 001) (no numbers
given) Haworth 1997 reported several different outcomes one
beingthat having completed at leastone trainingcoursecompared
to no training courses at all had no significant effect on the odds
of crashing (afteradjustment for age blood alcohol level or licence
status (learner permitprobationaryfull)) When comparing three
different levels of training courses (beginner intermediate and ad-
vanced) with no training no significant differences were found
in the odds of crashing However after adjustment for age and
blood alcohol level completing an advanced course as compared
to a beginners course was associated with a significant decrease in
the odds of severe motorcycle crashes resulting in serious injury or
death (OR 02 95 CI 01 to 08)
D I S C U S S I O N
The objective of this review was to quantify the effectiveness of
pre- andpost-licence motorcycle rider training on the reduction of
traffic crash involvement injuries deaths and offences committedby motorcycle riders Most included studies were cohort studies
There were a few randomised and non-randomised trials and case-
control studies
All studies compared either mandatory or non-mandatory formal
training with no formal training except for Haworth 1997rsquos case-
control study that compared levels of training (beginner interme-
diate and advanced) to each other
We found studies that evaluated the effectiveness of manda-
tory pre-licence training in reducing crash and offence rates
(Anderson 1980 Hall 1988 Kloeden 1994 Billheimer 1998
(the basic California Motorcyclist Safety Program component of
the study) Perrino 2002) non-mandatory training that when
passed resulted in a licence (Cooper 1988 McDavid 1989
Swaddiwudhipong 1998) non-mandatory training that could be
taken by motorcycle riders with or without a licence (Raymond
1979 Osga 1980 Satten 1980 Jonah 1982 Mortimer 1984
Adams 1985 Leung 1987 Mortimer 1988 Waller 1992 Davis
1997 Billheimer 1998 (the Experienced Rider Course compo-
nent of the study)) and unspecified rider training on crash injury
and death rates (Kraus 1975 Hurt 1981 Haworth 1997 MAIDS
2004)
The studies provided conflicting evidence regarding motorcycle
rider training which is likely due to the poor design quality and
reporting of the studies Most of the studies included in the review
were likely to be adversely affected by significant selection biasalong with substantial detection bias(measurement and recall bias)
which limits our ability to make conclusive comments
Mandatory pre-licence rider training
There were two RCTs three cohort studies and one case-control
study that examined pre-licence ridertraining The two RCTs eval-
uated the Motorcycle Skills Operating Test (MOST) (Anderson
1980 Hall 1988) Anderson 1980 found a statistically significant
reduction in crashes in the intervention group at six months one
and twoyearsfollow up However when only participantswho ob-
tainedtheirlicence were included in theanalysis thereduction wasstatistically significant only at one year follow up The other RCT
(Hall 1988) found no statistically significant differences in crash
rates at any point of time neither in the total population group
nor in the licensed subgroup Neither of the two cohort stud-
ies assessing crash rates found a statistically significant difference
in crash rates between intervention and control groups ( Kloeden
1994 Billheimer1998) andthe case-control study (MAIDS 2004)
also did not find a statistically significant difference in the odds of
experiencing a crash between trained and untrained riders With
regards to offences neither of the RCTs found a statistically sig-
nificant difference in offence rate between the intervention and
control groups in the analysis of the total group of participants
entering the trial However Anderson 1980 found a statistically significant increase in offence rates in the intervention group but
only after two years of follow up andonly in the analysis of partici-
pants who obtained their licence The two cohort studies assessing
offence rates found no statistically significant difference between
the groups (Billheimer 1998 Perrino 2002)
None of the five studies (RCTs or cohort) found a reduction in
crashes or offences following mandatory pre-licence rider training
except for Anderson 1980 who observed a statistically significant
reduction in crashes in the intervention group at one year in both
analyses It is not clear why both RCTs that evaluated the same
MOST and training intervention observed different results Both
studies enrolled large samples and observed a similar uptake of
licensure following the intervention It is possible that differencesmay have arisen due to different study samples (California ver-
sus New York) or that the intervention delivery differed between
studies
In both trials of the Motorcycle Skills Operating Test including
rider training only a third of motorcycle licence applicants ob-
tained their motorcycle licence It is possible that a licensing pro-
cess including a skills test may reduce crashes through decreased
exposure
Non-mandatory pre-licence training
20Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2350
Three studies assessed the effect of non-mandatory pre-licence
training The Cooper 1988 cohort study found no difference incrash or offence rates between the trained and untrained groups
We cannot however be confident about this result as the study
sufferedfrom selectionbiaswhichcouldhave biasedthe measure of
effect toward the null The RCT conducted by Swaddiwudhipong
1998 examined injury and death rates and found a statistically
significant lower injury rate in the intervention villages after one
year of follow up but no longer at two year follow up There
was no statistically significant difference in death rates between
intervention and control villages Our confidence in the results of
this study are hindered by the inadequacy of the randomisation
process that led to significant differences between the intervention
and control districts on two important variables The intervention
districts were more likely than the control districts to have a validlicence andto wear a helmetSincethese differences were nottaken
into account in later analyses they could explain the differences
observed in injury rates between districts Furthermore with only
six districts randomised the sample size was probably too small
to detect a difference between intervention and control groups
had there been one The McDavid 1989 cohort study found a
lower percentage of injuries in the intervention group although
a statistical test was not conducted McDavid 1989 drew their
intervention and control groups from different populations but
did adjust for confounders although they only analysed 40 of
the initial sample (139 trained and 139 untrained) Because of
these substantial weaknesses we are unable to draw conclusions
from this studyAn overall effect of pre-licence training on the reductionof crashes
injuries deaths and offences was not found although one should
be aware that the research of this issue to date has had significant
flaws However a mandatory skill test (such as the MOST) in
combination with and even without pre-licence training as part
of a licensing process seems to reduce the number of individuals
completing the licensing procedure and therefore indirectly re-
duces crash rates through a reduction in exposure to motorcycle
riding Making licensure a more difficult process may therefore be
seen as advantageous if it delays licensure and therefore reduces
exposure It is possible that the group who chose not to continue
riding a motorcycle would not have been the group to experience a
crash or commit an offence if they had continued to ride anyhowfor example because they are more safety conscious and this was
the reason for their giving up riding in the first place If this is the
case deterring individuals from completing their licence and rid-
ing may not necessarily decrease overall crashes deaths or offence
rates In addition other potentially adverse effects of complicat-
ing a licensing procedure need to be considered such as increases
in unlicensed riding In Thailand non-mandatory rider training
offered as a community intervention may have decreased injury
rates but this needs to be confirmed through further research
Non-mandatory training
The majority of the studies assessed non-mandatory training
(training independent of any licensing procedure) and the results
were varied The two non-randomised trials assessing crash rates found no statistically significant difference in crash rates between
the intervention and control groups (Adams 1985 Leung 1987)
Of the nine cohort studies Satten 1980 found a lower percentage
of crashesin the intervention group although it is unclear whether
this was a statistically significant result Davis 1997 also found a
lower percentage of crashesin theinterventiongroup although it is
unclear with which control group the intervention group wascom-
pared Raymond 1979 Osga 1980 Waller 1992 and Savolainen
2007 found a statistically significant higher crash rate in the in-
tervention group and the remaining cohort studies (Jonah 1982
Mortimer 1984 Mortimer 1988 Billheimer 1998) found no sta-
tistically significant difference in crash rates between intervention
and control groups No conclusions can be drawn from these stud-
ies with confidence The Satten 1980 and Davis 1997 studies did
not adjust for any confounding factors even though important
differences existed between the intervention and control groups
Although Raymond 1979 and Waller 1992 adjusted for a num-
ber of confounders at baseline (Waller 1992) and in the analy-
sis (Raymond 1979 Waller 1992) the intervention and control
groups were drawn from such different populations that it is pos-
sible that unmeasured confounders were responsible for the sta-
tistically significant increased risk of crashes observed in the inter-
vention group The observed difference could also be attributed totheobserved differential loss to followup (Raymond 1979) which
could bias the measure of effect towards or away from the null
The Osga 1980 study found a protective effect of rider training
in the univariate analysis but statistically significant higher crash
rates in the intervention group when accounting for riding expo-
sure Sex (a significantly higher proportion of females in the in-
tervention group) and other potentially confounding factors that
could explain thedifferencebetween the groups were howeverun-
accounted for Finally Savolainen 2007 did control for potential
confounding variables yet it is not clear whether variables were
controlled for simultaneously or one at a time One case-control
study (MAIDS 2004) found no statistically significant reduction
in the odds of experiencing a crash between trained and untrainedriders but the quality of this study was poor mainly because the
analysis did not account for any confounding factors
One cohort study (Satten 1980) assessed the effect of non-manda-
tory rider training on injurious crashes and found a lower percent-
age of injuries in the intervention group although a statistical test
was not conducted As mentioned above the Satten 1980 studyrsquos
validity is reduced due to its lack of adjustment for confounding
factors
The only non-randomised trial that assessed offence rates (Leung
1987) found a statistically significant lower offence rate in the in-
tervention group and of the five cohort studies that assessed of-
fence rates (Satten 1980 Jonah 1982 Mortimer 1984 Mortimer
21Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2450
1988) one study (Jonah 1982) also found a statistically significant
lower offence rate in the intervention group However we have lit-tle confidence in the results from Leung 1987 since potential con-
founding factors were not appropriately addressed Jonah 1982
did conduct a multivariate analysis but was hindered by an initial
low response rate The other four cohort studies did control for
several potential confounders and found no difference in offence
rates following non-mandatory post-licence rider training
The non-randomised trials and cohort studies evaluating non-
mandatory rider training were at high risk of substantial selection
bias due to the non-random allocation of participants as well
as low participation rates and attrition bias and could therefore
not provide conclusive evidence with regard to the effect of the
training It should be noted that despite the biases and other issues
with this group of studies most didnot detectmajor effects relatedto rider training in either direction This may suggest that any
potential impact of rider training will be modest
Studies examining unspecified modes of ridertraining
Three case-control studies assessed whether injured or deceased
motorcyclists were more or less likely to have had rider training
compared to control motorcyclists All had conflicting results
Hurt1981 foundthe odds of havingan injuriouscrash waslower in
those who had received rider training whereas Kraus 1975 found
that the odds of having an injurious crash were higher in youngerand older trained riders Haworth 1997 who compared advanced
versus beginner training found no effect on motorcycle injuries
or deaths although they did find that those who had completed
advanced training had reduced chances of crashing compared to
those who had done a beginnersrsquo training course Similar issues
with uncontrolled confounding and biases of selection as described
above arose with the three case-control studies Hurt 1981 suffers
from inadequate control of confounding via poor matching and
both the Kraus 1975 and the Haworth 1997 findings were only
controlled for limited factors by means of stratified analysis It is
likely that these limitations are the reason for the contradictory
directions of effect observed for the three case-control studies
Methodological weaknesses of the studies
In summary most studies investigating the effectiveness of mo-
torcycle rider training on reducing crashes injuries and death suf-
fered from serious methodological weaknesses
The majority of studies were non-randomised and were there-
fore limited mainly by selection bias Although the approaches of
matching stratification or regression analyses can minimise bias
(Fleiss 2004) they were poorly employed in the reviewed studies
The inadequate adjustment for confounders in particular adds
to our lack of confidence in the studiesrsquo results since it has been
demonstrated that individuals who choose to take motorcycle rider
training differ from those not taking rider training For instance
trained riders often differ from controls in age sex and experi-ence usually ride less and may have a more positive attitude to
safety as indicated by the reported more frequent wearing of seat
belts (Mortimer 1982)
The two large randomised trials reported that a large proportion
of their study group were never exposed to the risk of offences or
crashes because they never attained licensure This made it impos-
sible to assess whether a reduction in crashes in the intervention
group was a result of training or due to delayed or never attained
licensure
In some studies the sample sizes were too small to detect any sta-
tistically significant differences Follow-up time after training dif-
fered between studies and was an impediment to observing any
real effects of training if they existed As riding experience ac-cumulates over time any differences in outcomes between those
trained and untrained would diminish in studies with lengthy fol-
low-up times This dilution of any possible effect of rider training
may have also occurred in studies that mixed both novice and ex-
perienced riders in the study groups since it is hypothesised that
novice riders serve to benefit most from rider training This would
increase the difficulty in finding any effect of rider training par-
ticularly when sample sizes are small
Detection bias may also account for some of the differences ob-
served between trained and untrained riders Detection bias would
have occurred if the self-report hospital driver or police records
used to record the outcomes of crash injury death or offences
differed between the trained and untrained riders Research hasdemonstrated that not all crashes are reported to police for ex-
ample small crashes or off-road crashes (Kraus 1975) even serious
injury crashes have been found to be underreported (Anderson
1980) However this would only introduce detection bias if the
reporting of outcome differs for trained and untrained riders This
does appear possible in that responsible riders may be more likely
to enroll in a rider training course and also may be more likely
to report crashes to police than irresponsible riders meaning that
the number of crashes from riskier riders could be underrepre-
sented on records An analysis to investigate the effectiveness of
rider training may then show that rider training is not effective in
reducing crashes
Rather than using reported outcomes some studies used self-re-ported survey data although this technique suffers from non-
response biases In fact two studies reported differences in de-
mographic characteristics in response rates to questionnaires
(Anderson 1980 Hall 1988) Self-reported data also introduce
the risk of recall bias survival bias and social desirability response
bias Recall bias may have been equally distributed among trained
and untrained riders howeverit is hypothesised that trained riders
may have felt it was lsquoexpectedrsquo to report good driving histories
hence leading the rider training to appear effective
22Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2550
Limitations of this review
Limitations of this review were that we found very few recentstudies (the two most recent ones published in 2002 and 2007)
and that we found very few studies conducted in low-and middle
income countries even though our search strategy was set up to
identify such studies Anotherlimitation wasthat forthe reviewed
studies we could not assess all outcomes of rider training that are
potentially useful Some of the included studies assessed outcomes
such as use of safety clothing including helmets Such outcomes
are valuable since the use of motorcycle helmets has been demon-
strated to be effective in preventing head injuries (Liu 2004) De-
spite this the use of safety clothing was not included as an out-
come in this review because the studies measuring this outcome
were limited by poor data quality due to very low response rates
Another outcome of interest in assessing the effectiveness of rider
training is lsquotype of injuryrsquo As Haworth 2005 state in their review
of motorcycle licensing and training rider training could be con-
sidered effective if injuries become less severe because motorcycle
riders have learned how to avoid more serious injuries through
rider training The studies included in this review did not specify
type of injury and so this outcome could not be assessed Last
changes in knowledge and attitudes following rider training are
other potentially useful outcomes We have excluded such studies
(Osga 1979 Clayton 1990 Goldenbeld 2004) as these outcomes
are proxies rather than direct measures of the effectiveness of mo-
torcycle rider training We opted to report on the outcomes of
crashes injury death and offences alone since there is little ev-
idence about how changes in knowledge and attitudes relate to
these outcomes
A U T H O R S rsquo C O N C L U S I O N S
Implications for practice
Due to the poor quality of studies included in this review we were
unable to draw any conclusions about the effectiveness of rider
training on crash rates injurious or fatal or offence rates The
findings do suggest that mandatory pre-licence training may be
an impediment to completing a motorcycle licensing process thuspossibly indirectly reducing crash injury death and offence rates
through a reduction in exposure to riding a motorcycle On the
basis of the existing evidence it is not clear if and what type of
training reduces the risk of crashes injuries deaths or offences in
motorcyclists and the selection of the best rider training practice
can therefore not be recommended The suggestion that manda-
tory pre-licence training indirectly reduces exposure to riding the
motorcycle because trainees are impeded from completing the li-
censing process does not help those motorcyclists who persist in
obtaining their licence and riding the motorcycle Some type of
rider training is likely necessary to teach motorcyclists to ride a
motorcycle safely Good policy is based on sound evidence and
evaluation but such evidence for motorcycle rider training is seri-
ously lacking Policy makers are therefore encouraged to include arigorous evaluation component to any new or existing rider train-
ing programme Such an evaluation should ideally be randomized
and assess whether the training programme is effective and if so
which component(s) of the programme work best
Implications for research
Given theresults of this review which found littlereliable evidence
about the effectiveness of motorcycle rider training it is necessary
to build up an evidence base of the effectiveness of motorcycle
rider training in the reduction of crash injury death and offence
rates To this end well designed randomised controlled trials need
to be conducted that examine the impact of rider training oncrashes and offences A randomised controlled trial is the preferred
method as it would significantly reduce most biases described in
this review To minimise the impact of participants delaying or
never attaining their licence and therefore exposure as found in
the randomised trials in this review such a trial should include
riders who are already licensed and reliable measures of riding
exposure in a reasonably lengthy follow up period It is expected
that a randomised controlled trial of sufficient sample size would
observe equal balance on possible confounders by study arm how-
ever an assessment of potentially important confounders (such as
helmet wearing) by study group would enable control for these
variables in the analyses if necessary It is imperative that the effec-
tiveness of each component of the rider training (such as theorypractice knowledge skills and modification of behaviour) and the
synergy of such components are evaluated Outcomes should be
objective ideally those that are routinely collected rather than self-
reported outcomes Even though riding the motorcycle poses dif-
ferent challenges and incurs different risks lessons may be learned
from research into the effects of car driver training
Motorcycles are a common means of transport in low- and middle
income countries and usage is increasing As the burden of mo-
torcycle crash victims is mostly borne in such countries research
should be focused on the effectiveness of types of rider training on
the reduction of crashes injuries and fatalities in such countries
and be supported by higher income countries This review found
only one study conducted in a middle income country that sug-gested decreasing injury rates It would be beneficial to conduct
additional research to confirm this suggestion and to enhance our
knowledge of the role of rider training in low- and middle-income
settings
March 2006 saw the launch of the University of North Carolina
Highway Safety Research Centre partnership with the Motorcycle
Safety Foundation (MSF) and plans were revealed to conduct a
study of the potential benefits of motorcycle rider training The
study is funded by the MSF and the National Highway Traffic Sa-
fety Administration (NHTSA) Called rsquoThe Discovery Projectrsquo it
will compare data on crash rates from riders participating in a basic
23Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2650
motorcycle safety class with riders participating in a basic motor-
cycle safety class as well as a variety of continuing education classesThe collection of pilot data was due to start in March 2007 and
the data collection portion of the study is expected to take three
years (Highway Safety Research Center) It is vital that researchers
and policy makers collaborate to develop furtherhigh quality eval-
uations of motorcycle rider training programmes to help inform
development of new and effective training programmes
A C K N O W L E D G E M E N T S
Staff from The University of Sydney library for helping us locate
and retrieve full reports of many papers
R E F E R E N C E S
References to studies included in this review
Adams 1985 published data only
Adams A Collingwood V Job RFS Evaluation of the 1982 Australia
Post motorcycle rider training scheme NSW Australia Traffic
Accident Research Unit Traffic Authority NSW 1985 [ RN 85]
Anderson 1980 published data only
Anderson J Ford JL Peck RC Improved motorcyclist licensing and testing project Washington DC National Highway Traffic Safety
Administration (NHTSA) USA 19801ndash217 [
DOTndashHSndash805ndash598]
Billheimer 1998 published data only
Billheimer JW Evaluation of the California Motorcyclist Safety
Program Transportation Research Record 19981640100ndash9 [
TRB Report Number 980652]
Cooper 1988 published data only
Cooper PJ Rothe JP Motorcyclists who they are and why they do
what they do Transportation Research Record 1988116878ndash85
Davis 1997 published data only
David CF Evaluation of community traffic safety programs and
motorcycle operator training programs Connecticut Department of Transportation 199763
Hall 1988 published data only
Hall CJ Buchanan L Engle R Motorcycle rider education evaluationproject Washington DC USA National Highway Traffic Safety
Administration (NHTSA) US Department of Transportation
1988 [ DTNH 22ndash80ndashCndash05123]
Haworth 1997 published data only
Haworth N Smith R Brumen I Pronk N In Monash University
Accident Research Centre editor(s) Case-control study of motorcycle
crashes Victoria Australia Department of Transport and Regional
Development The Federal Office of Road Safety 19971ndash149 [
ISBN 0 642 25501 6 ]
Hurt 1981 published data only
Hurt H Ouellet J Thom D Motorcycle accident cause factors and
identification of countermeasures USA Department of
Transportation 1981 [ Technical Report DOT HS 806ndash862]
Jonah 1982 published data onlylowast Jonah BA Dawson N Bragg B Are formally trained motorcyclists
safer Accident Analysis amp Prevention 198214(4)247ndash55
Jonah BA Dawson NE Bragg BW Evaluation of the Canada
Safety Councilrsquos Motorcycle Training Program Proceedings of theInternational Motorcycle Safety Conference Washington DC
USA 198059ndash93
Kloeden 1994 published data only
Kloeden CN Moore VM McLean AJ Evaluation of the pre-license
training program for motorcyclists in South Australia Walkerville
South Australia Australia NHMRC Road Accident Research Unit
The University of Adelaide 19941ndash37 [ ISBN 0 7308 0607 3]
Kraus 1975 published data only
Kraus JF Riggins RS Franti CE Some epidemiological features of
motorcycle collision injuries American Journal of Epidemiology 1975102(1)74ndash98
Leung 1987 published data only
Leung KS Reding VA Evaluation of the Wisconsin motorcycle rider course Wisconsin Bureau of Policy Planning and Analysis
Wisconsin Department of Transportation 1987
MAIDS 2004 published and unpublished data
MAIDS In-depth investigations of accidents involving powered
two wheelers ACEM (Acemacemeu) 2004 Vol Final report
20179
McDavid 1989 published data only
McDavid JC Lohrmann BA Lohrmann G Does motorcycle
training reduce accidents Evidence from a longitudinal quasi-
experimental study Journal of Safety Research 198920(2)61ndash72
Mortimer 1984 published data only
Mortimer R Evaluation of the motorcycle rider course Accident
Analysis amp Prevention 198416(1)63ndash71
24Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2750
Mortimer 1988 published data only
Mortimer RG A further evaluation of the motorcycle rider courseJournal of Safety Research 198819(4)187ndash96
Osga 1980 published data only
Osga GA An investigation of the riding experiences of MSF rider
course participants Dissertation South Dakota USA The
University of South Dakota 19801ndash224
Perrino 2002 published data only
Perrino CS Ahmed A Callender A Rozier E Cantwell A Stewart
O The role of Marylandrsquos motorcycle rider course in promoting saver
behaviors and attitudes Baltimore USA National Transportation
Center Morgan State University 20021ndash52
Raymond 1979 published data only
Raymond S Tatum S An evaluation of the effectiveness of the RAC
ACU motorcycle training scheme - final report Road Safety ResearchUnit Dept of Civil Engineering University of Salford London
19791ndash131
Satten 1980 published data only
Satten RS Analysis and evaluation of the motorcycle rider courses
in thirteen Illinois counties International Motorcycle Safety
Conference Washington DC 1980145ndash93
Savolainen 2007 published and unpublished data
Savolainen Peter Mannering Fred Additional evidence on the
effectiveness of motorcycle training and motorcyclistsrsquo risk-taking
behaviour Transportation Research Board 2007 Vol March 20
19
Swaddiwudhipong 1998 published data only
Swaddiwudhipong W Boonmak C Nguntra P Mahasakpan P
Effect of motorcycle rider education on changes in risk behavioursand motorcycle-related injuries in rural Thailand Tropical Medicine
and International Health 19983(10)767ndash70
Waller 1992 published data only
Waller FL An analysis of the impact of the Texas Department of
Public Safety Motorcycle Operator Training Program upon
motocycle related traffic accidents Texas State University Public
Administration Program Applied Research Projects 199277
References to studies excluded from this review
Anderson 1980a published data only
Anderson LC Better biking a motorcycle rider improvement
program International Motorcycle Safety Conference Washington
DC USA Motorcycle Safety Foundation 1980231ndash53Aupetit 2008 published and unpublished data
Aupetit S Riff J Buttelli O Espie S Ergonomics research program
on motorcycle learning Contributions to the development of a
motorcycle simulator for training Transport Research Arena
Ljubljana 20089
Awane 1999 published data only
Awane T Integrating simulators in motorcycle safety education
Journal of International Association of Traffic and Safety Sciences 199923(1)26ndash35
Baldi 2005 published data only
Baldi S Baer JD Cook AL Identifying best practices states in
motorcycle rider education and licensing Journal of Safety Research200536(1)19ndash32
Batchler 1988 published data only
Batchler MW Evaluation of the pre-license motorcycle training course at the New South Wales traffic education centre Federal Office of
Road Safety (FORS) 19881ndash122 [ Report No CR 77 ]
Blanchard 2006 published data only
Blanchard HT Tabloski PA Motorcycle safety educating riders at
the teachable moment Mosby (CV) Company 200632330ndash2
Bonnett 2005 published data only
Bonnett J Mature age skills training for experienced motorcycle
riders Motorcycle ridersrsquo association of the Australian Capital
Territory 2005
Braver 2007 published data only
Braver ER et alPersuasion and licensure a randomized controlled
intervention trial to increase licensure rates among Maryland
motorcycle owners Traffic Injury Prevention 20078(1)39ndash46Clayton 1980 published data only
Clayton AB Sudlow DE An evaluation of the effectiveness of a one day training programme for learner motorcyclists Berkshire UK
Transport and Road Research Laboratory 19801ndash23 [ ISSN
0266ndash7045]lowast Clayton AB Sudlow DE An evaluation of the effectiveness of a one-day training programme for learner motorcyclists Berkshire UK
Road Safety Division Berkshire 19801ndash23 [ISSN 0266ndash7045]
Council 1978 published data only
Council FM Desper LP Dutt AK Driver education for motorcycle operation University of North Carolina Highway Safety Research
Center 19781ndash40
Crowther 2005 published data onlylowast Crowther G Engaging with motorcyclists UK police and theBikeSafe Road Safety Programme Published proceedings of the
Australasian Road Safety Research Policing Education Conference
Wellington New Zealand 2005 issue ISBN 0ndash473ndash10636ndash11ndash11
Daniello 2009 published data only
Daniello A Gabler HC Mehta YA Effectiveness of motorcycle
training and licensing Transportation Research Board 88th Annual
Meeting 200913
Dewanti 2007 published data only
Dewanti M Challenging efforts in promoting young motorcyclist
safety in Indonesia Road Safety on Four Continents 14th
International Conference Swedish National Road and Transport
Research Institute 2007
Du Plooy 1990 published data onlyDu Plooy DW Motorcycle training in South Africa The Human
Element International Motorcycle Safety Conference Orlando
Florida USA Motorcycle Safety Foundation 19904-1 to 4-9
Engel 1989 published data only
Engel U Krogsgaringrd Thomsen L [Faeligrdselsundervisning og
propaganda ndash en faeligrdselssikkerhedsfremmende foranstaltning]
Dansk Vejtidsskrift 198912295ndash7
Goldenbeld 2002 published data only
Goldenbeld C Houwing S de Craen S The development of
driving skills of young moped riders [De ontwikkeling van de
rijvaardigheid van jonge bromfietsers] SWOV Institute for Road
Safety Research (Stichting Wetenschappelijk Onderzoek
Verkeersveiligheid) Leidschendam the Netherlands 200263
25Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2850
Goldenbeld 2004 published data only
Goldenbeld C Twisk D de Craen S Short and long term effects of moped training a field experiment Transportation Research Part F 20047(1)1ndash16
Haworth 2000 published data only
Haworth N Smith R Kowadlo N Motorcycle training in Victoria
evaluation of rider training curriculum in Victoria Monash
University Accident Research Centre Reports 2000 Vol 165 issue
3131
Haworth 2005 published data only
Haworth N Mulvihill C Review of motorcycle licensing and
training Monash University Accident Research Centre 2005 [
Report no240]
Haworth 2006 published data only
Haworth N Mulvihill C Rowdon P Teaching old dogs new tricksTraining and older motorcyclists Journal of the Australasian College of Road Safety 200718(4)20ndash5lowast Haworth NL Mulvihill C Rowden PJ Teaching old dogs new
tricks Training and older motorcyclists Australasian Road Safety
Research Policing and Education Conference Gold Coast
Queensland 2006
Hill 1983 published data only
Hill D Motorcycle rider training testing and licensing Motor
Cycle Council of New South Wales Sydney 1983
Imatake 1980 published data only
Imatake Y On the activities in Japan of motorcycle safe riding
promotion committee (MSRPC) - especially the education of
unskilled riders of class 1 motorbikes International Motorcycle
Safety Conference Washington DC USA Motorcycle Safety
Foundation 1980255ndash68
Ingebritsen 1990 published and unpublished data
Ingebrigtsen S [Risikofaktorer ved ferdsel med moped og
motorsykkel] TOslashI-rapport 66 Oslo Transportoslashkonomisk
institutt 1990
Jain 2009 published and unpublished data
Jain A et alTwo wheeler accidents on Indian roads a study from
Mangalore India Elsevier 200916130ndash3
Jonah 1979 published data only
Jonah BA Dawson NE Validation of the motorcycle operator skill
test Accident Analysis amp Prevention 197911(3)163ndash71
Jonah 1981 published data onlyJonah BA Dawson NE Bragg WE Predicting accident
involvement with the Motorcycle Operator Skill Test Accident Analysis amp Prevention 198113(4)307ndash418
Kadar 1990 published data only
Kadar KJ Countermeasures motorcycle rider education delivery
system The Human Element International Motorcycle Safety
Conference Motorcycle Safety Foundation 19906-27 to 6-41
Kelsey 1986 published data only
Kelsey SL Liddicoat C Ratz M Licensing novice motorcyclists a
comparison of Californiarsquos standard test and the MOST II (Motorcycle
Operator Skill Test) administered at centralized testing offices Sacramento California USA California Department of Motor
Vehicles Research and Development Office 19861ndash74
Koch 1980 published data only
Koch H Advanced course for motorcycle riders InternationalMotorcycle Safety Conference Washington DC USA Motorcycle
Safety Foundation 1980269ndash81
Koch 1990 published data only
Koch H Brendicke R How to influence risk taking behaviour
within motorcycle rider training programs The Human Element
International Motorcycle Safety Conference Orlando Florida
USA Motorcycle Safety Foundation 1990Motorcycle Safety
Foundation 1990 p 9-1 to 9-25A
Labbett 2005 published data only
Labbett S Langham M Training can make the problem worse
Road Safety Congress 20051ndash12
Law 2005 published and unpublished data
Law TH Radin Umar RS Zulkaurnain S Kulanthayan S Impactof the effect of economic crisis and the targeted motorcycle safety
programme on motorcycle-related accidents injuries and fatalities
in Malaysia International Journal of Injury Control and Safety
Promotion 200512(1)9ndash21
Lowes 1990 published data only
Lowes BF In-traffic rider training The Human Element
International Motorcycle Safety Conference Orlando Florida
USA Motorcycle Safety Foundation 19907-46 to 7-57
Mahon 1976 published data only
Mahon J Teaching accident avoidance on the motorcycle Journal of Traffic Safety Education 1976231ndash29
Maxwell 1985 published data only
Maxwell GS Evaluation of a Program for Training MotorcycleRiders Evaluation and Program Planning 19858(4)339ndash47
Mayhew 1996 published data only
Mayhew DR Simpson SH Effectiveness and role of driver educationand training in a graduated licensing system Ottowa Canada The
Traffic Injury Research Foundation 19961ndash89 [ ISBN
0ndash920071ndash12ndash0]
McKnight 1987 published data only
McKnight AJ Milburn BG An evaluation of the modified
motorcycle operator skill test exercises NHTSA Report 1987 [
Report No DOT HS 807 212 ]
McPherson 1976 published data only
McPherson K McKnight AJ The development and evaluation of a
motorcycle skill test manual and knowledge test Washington DCUSA National Highway Safety Administration 1976
Mohan 2004 published data only
Mohan D Evidence-based interventions for road traffic injuries in
South Asia Journal of the College of Physicians and SurgeonsPakistan 200414(12)746ndash7
Morgan 1980 published data only
Morgan JK Advanced motorcycle training course content and
study evaluation International Motorcycle Safety Conference
Washington DC USA Motorcycle Safety Foundation 1980
283ndash312
Motoki 1990 published data only
Motoki M Yamazaki S A study on effective motorcycle rider
education The Human Element International Motorcycle Safety
26Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2950
Conference Orlando Florida USA Motorcycle Safety
Foundation 19909-26 to 9-57A
Nairn 1992 published data onlylowast Nairn RJ Motorcycle safety research literature review 1987 to
1991 Federal Office of Road Safety Australia 1993 Vol Report
CR 117
Newland 1980 published data only
Newland R Darby J Automotive practicespre motorcycle rider
program International Motorcycle Safety Conference Washington
DC USA Motorcycle Safety Foundation 1980127ndash9
Newland 1982 published data only
Newland R Motorcycle rider training and licensing requirements
Driver training steering a course for the future Melbourne
Victoria Transport Regulation Board 1982
Newland 1990 published data only
Newland R Motorcycle training and licensing in Victoria 1980-
1990 The Human Element International Motorcycle Safety
Conference Orlando Florida USA Motorcycle Safety
Foundation 1990
Ormston 2003 published data only
Ormston R Dudleston A Pearson S Stradling Evaluation of
Bikesafe Scotland Scottish Executive Social Research Edinburgh
2003
Osga 1979 published data only
Osga GS Ellingstad VS An evaluation of the effectiveness of the 1978
South Dakota motorcycle rider course program South Dakota USA
University of South Dakota 19791ndash89
Panou 2007 published data only
Panou M Bekiaris E A new concept on the integration of driving
simulators in driving training the train-all approach Road Safety
on Four Continents 14th International Conference Swedish
National Road and Transport Research Institute 2007
Peterson 1990 published data only
Peterson A Kadar K Status report 1980-1989 Minnesota rider
education statistics The Human Element International
Motorcycle Safety Conference Orlando Florida USA 199013-
113 to 13-129
Prem 1984 published data only
Prem H Good MC Motorcycle rider skills assessment ACT
Australia Office of Road Safety Department of Transport 1984
1ndash377 [ Report No CR 34 ]
Radin Umar 2006 published data only
Radin Umar RS Motorcycle safety programmes in Malaysia how
effective are they International Journal of Injury Control and Safety Promotion 200613(2)71ndash9
Rockwell 1990 published data only
Rockwell TH Kiger S Carnot M An evaluation of the Ohio
motorcyclist enrichment program Ohio Department of Highway
Safety 1990
Rothe 1987 published data only
Rothe JP A motorcycle riding course Motorcyclists Image and reality Canada Insurance Corporation of British Columbia 1987
59ndash83
Rowden 2007 published data only
Rowden P Watson B Haworth N What can riders tell us aboutmotorcycle training A view from the other side of the fence
Australasian Road Safety Research Policing and Education
Conference Melbourne Australia Melbourne 2007
RSC Victoria 1993 published data only
Road Safety Committee Inquiry into Motorcycle Safety in Victoria
Parliament of Victoria 1993
Russam 1979 published data only
Russam K Motorcycle training and licensing in Japan UK Transport
and Road Research Laboratory (TRRL) 1979 [ Report 916 ]
Saffron 1981 published data only
Saffron DG Driver and rider education training and licensing a
brief review Traffic Accident Research Unit Traffic Authority of
New South Wales NSW Australia Traffic Accident Research UnitTraffic Authority of New South Wales 1981
Schulz 1990 published data only
Schulz U Motorcycle accidents among learner riders Motorcycle
licensing Evaluation of the effectiveness of a Graduated Licensing Scheme Forshungshefte des Institut fur Zweiradsicherheit 1990
Sexton 2004 published data only (unpublished sought but not used)
Sexton B Baughan C Elliott M Maycock G The accident risk of
motorcyclists 2004TRL Report TRL 607(ISSN 09684107)58
Shephard 1986 published data only
Shepherd R Do Motorcycle Safety Foundation programs work
Driver 198619(9)16ndash8
Simpson 1990 published data onlySimpson HM Mayhew DR The promotion of motorcycle safety
training education and awareness Health Education Research
Theory and Practice 19905(2)257ndash64
Sokolov 1990 published data only
Sokolov O Safe motorcycle riding some aspects of motorcycle
riding in the USSR The Human Element International
Motorcycle Safety Conference Motorcycle Safety Foundation
19904ndash37 - 4-48
Steffens 1988 published data only
Steffens U Gawatz R Willmes G Effectiveness of mofa courses
[Wirksamkeit von Mofakursen] Unfall- und sicherheitsforschung
strasenverkehr 198867168
Sudlow 2003 published data only
Sudlow D Scoping study on motorcycle training Road Safety
Research Report London UK The British Institute of Traffic
Education Research Department for Transport 2003 Vol June
[ Report No 36]
Tasmania 1995 published data only
Mannering FL Grodsky LL Motorcycle rider training scheme
Traffic Education Research Centre Armidale Australia Driver
Training Australia Limited 1995 Vol April1ndash56
Tetard 1985 published data only
Tetard C Motorcycle ridersrsquo training to detect indices in an
intersection situation [Lrsquoentrainement a la prise drsquoinformation dans
lrsquoapprentissage de la conduite motocycliste] Le Travail Humain198548(2)147ndash59
27Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3050
Thackray 1980 published data only
Thackray RM Prescott JC Field test of the motorcycle ridercourse International Motorcycle Safety Conference Washington
DC USA Motorcycle Safety Foundation 1980195ndash7
Thompson 1994 published data only
Thompson M Evaluation of compulsory basic training for
motorcyclists Transport Research Laboratory Crowthorne
Berkshire UK 1994 [ Report 63]
Tobler 1990 published data only
Tobler U Individual driving instruction for motorcycle driving
learners the Swiss experiment of motorcycle driving school The
Human Element International Motorcycle Safety Conference
Orlando Florida USA Motorcycle Safety Foundation 1990
Transport 1989 published data only
Transport Communications Road Crash Statistics Australia Bureauof Transport and Communications Economics 198914ndash20
Tuan 2007 published data only
Tuan VA Shimizu T Towards development and evaluation of the
motorcycle drivers re-education program in Vietnam modeling of
motorcycle driverrsquos undesired behaviors 11th World Conference
on Transport Research 200725
Victor 1980 published data only
Victor DJ Traffic safety education for motorcyclists International
Motorcycle Safety Conference Washington DC USA Motorcycle
Safety Foundation 1980221ndash8
Waterer 1990 published data only
Waterer TH Compulsory training for motorcyclists from liberty
to compulsion The Human Element International Motorcycle
Safety Conference Orlando Florida USA Motorcycle Safety
Foundation 19904-49 to 4-57
Weller 1989 published data only
Weller RB Chandler EW Motorcycle safety and motorcycle
education past research and survey results Journal of Public Policy
amp Marketing 1989893ndash108
Wells 1981 published data only
Wells P Observations of some RACACU motorcycle training
schemes Transport and Road Research Laboratory Crowthorne
Berkshire UK 1981
Winn 1986 published data only
Winn G Two important trends in motorcycle safety regulations
rider education and conspicuity improvement American
Motorcyclist Association 1986147ndash52
Wisher 1988 published data only
Wisher G Reid J Motorcycle accidents in Norfolk a three-year study of motorcycle training and the Part 1 test Norfolk UK Road Safety
Division Highways Department Norfolk County Council 1988
Wood 1987 published data only
Wood HT Bowen R Evaluation of the revised learner permit scheme
July 1983 to December 1985 Victoria Australia Road Traffic
Authority 19871ndash27 [ ISBN 0 7306 0247 8]
Yamazaki 1990 published data only
Yamazaki S Nakanishi M Survey on effect of Honda rider
education The Human Element International Motorcycle Safety
Conference Orlando Florida USA 199012-167 to 12-187
References to ongoing studies
Burgess 2010 unpublished data only
Burgess C Broughton P Fylan F Stradling Interim evaluation of
the UKs National Ride program In L Dorn editor(s) Driver
behaviour and Training Vol IV Aldershot Ashgate Publishing
Limited 2010
Additional references
Anderson 1980
Anderson J Ford JL Peck RC Improved motorcyclist licensing and
testing project Volume I and II National Highway Traffic Safety
Administration US department of Transportation Washington
DC 1980
Bjornstig 1985
Bjornstig U Bylund P Lekander T Brorsson B Motorcycle fatalitiesin Sweden Acta Chirurgica Scandinavica 1985151(7)577ndash81
Clayton 1990
Clayton A Sudlow DE An evaluation of the effectiveness of a one-
day training program for learner motorcyclists The Human
Element International Motorcycle Safety Conference Orlando
Florida USA 199013-105 to 13-111
Cochrane Collaboration 2008
Higgins JP Green S Cochrane Handbook for Systematic Reviews of
Interventions 500 The Cochrane Collaboration 2008
Elliot 2003
Elliott MA Baughan CJT Broughton J Grayson G Knowles J et
alMotorcycle Safety a Scoping Study TRL [ Report No
0968ndash4107]Fleiss 2004
Fleiss JL Levin B Paik MC Statistical methods for rates and proportions 3rd Edition John Wiley amp Sons Inc 2004 [ ISBN
978ndash0ndash471ndash52629ndash2 ]
FORS 1999
FORS Federal Office of Road Safety Road risk for sober licensed
motorcyclists Vol Monograph 27 Federal Office of Road Safety
19991ndash2
Haworth 2005
Haworth N Mulvihill C Review of motorcycle licensing and training
Victoria Australia Monash University Accident Research Centre
20051ndash83 [ Report No 240]
Huwiler-Muumlntener 2002Huwiler-Muumlntener K Juumlni P Junker C Egger M Quality of
Reporting of Randomized Trials as a Measure of Methodologic
Quality JAMA 20022872801ndash4
Johnston 2008
Johnston P Brooks C Savage H Fatal and serious road crashes
involving motorcyclists Road Safety Research and Analysis Report
ATSB Monograph 20 Canberra Department of Infrastructure
Transport Regional Development and Local Government 0205
081ndash26
Juumlni 2001
Juumlni P Altman DG Egger M Systematic reviews in health care
Assessing the quality of controlled clinical trials BMJ 2001323
42ndash6
28Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3150
Kopits 2003
Kopits E Cropper M Traffic fatalities and economic growth TheWorld Bank Development Research Group Infrastructure and
Environment Policy research working paper 3035 2003
Leung 1983
Leung P Chau BTW Motorcycle accidents in Hong Kong pattern
of injury treatment causative factors and socio-economic
implications Injury 198315(1)23ndash9
Liu 2004
Liu B Ivers R Norton R Blows S Lo SK Helmets for preventing
injury in motorcycle riders Cochrane Database of Systematic Reviews 2004 Issue 2 [DOI 10100214651858CD004333pub3]
Mohan 2002
Mohan D Road safety in less-motorized environments future
concerns International Journal of Epidemiology 200231(3)527ndash32Mortimer 1982
Mortimer RG An evaluation of the motorcycle rider course
Proceedings of the 20th conference of the American Association for
Automotive Medicine Arlington Heights IL USA American
Association for Automotive Medicine 198277ndash96
Mullin 1997
Mullin BT Risk factors for motorcycle injury the role of age
gender experience training and alcohol PhD thesis Auckland
New Zealand University of Auckland 1997
Noordzij 2001
Noordzij PC Forke E Brendicke R Chinn BP Integration of needs
of moped and motorcycle riders into safety measures review and
statistical analysis in the framework of the European research
project PROMISING SWOV Institute for Road Safety Research
Leidschendam The Netherlands 2001 [ Report No Dndash2001ndash5]
Osga 1979
Osga G Ellingstad VS An evaluation of the effectiveness of the 1978 South Dakota Motorcycle Riders Course Program South Dakota
University of South Dakota Vermillion Human Factors Lab 1979
Oxford English Dictionary Online
httpoxforddictionariescom Oxford English Dictionary Online
Oxford University Press 2010
Peden 2004
Peden M Scurfield R Sleet D World Report on Road TrafficInjury Prevention Geneva World Health Organization Available
at httpwwwwhointviolence˙injury˙preventionpublications
road˙trafficworld˙reporten 2004Chapter 3
Radin Umar 1995
Radin Umar R Macakay G Hills B Preliminary analysis of
motorcycle accidents Short-term impacts of the running
headlights campaign and regulations in Malaysia J Traffic Med 199523(1)17ndash27
Tsai 1995
Tsai YJ Wang JD et alCase-control study of the effectiveness of
different types of helmets for the prevention of head injuries among
motorcycle riders in Taipei Taiwan American Journal of
Epidemiology 1995142(9)974ndash81
Vis 1995
Vis AA In-depth study of the hazards of motorcycling [De
onveiligheid van motorrijden nader bekeken] SWOV Institute for
Road Safety Research Leidschendam the Netherlands 1995 [
Report No Rndash95ndash69]
Weaver 1990
Weaver EA Motorcycle safety education in the United States
Motorcycle Safety Foundation Orlando Florida USA 1990
Winn 1987
Winn G Two important trends in motorcycle safety regulations
rider education and conspicuity improvement No more
information available Detroit MI US Society of Automotive
Engineers 1987147ndash52
Yuan 2000
Yuan W The effectiveness of the rsquoride-brightrsquo legislation for
motorcycles in Singapore Accident Analysis amp Prevention 200032
(4)559ndash63
Zambon 2006
Zambon F Hasselberg M Factors affecting the severity of injuries
among young motorcyclists a Swedish nationwide cohort study
Traffic Injury Prevention 20067(2)143ndash9lowast Indicates the major publication for the study
29Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3250
C H A R A C T E R I S T I C S O F S T U D I E S
Characteristics of included studies [ordered by study ID]
Adams 1985
Methods Non-randomised trial
Participants Initial sample size
Total of 162 Australia Post telegram delivery riders
Not clear how many in intervention and control groups
Analysis sample size
Intervention group 63Control group 35
Total 98 (61)
Interventions No formal training vs voluntary rider training course of 12 hours over 2 consecutive days (theory 4 hours practice
8 hours)
Intervention integrity
No information given
Outcomes Australia Post recorded crashes
Length of follow up 1 year
Notes Study conducted in NSW Australia
Anderson 1980
Methods Randomised trial
Participants Initial sample size
Total 41188 motorcycle licence applicants
Analysis sample size
Intervention group with skills training 13160
Intervention group without skills training 12634
Control group 15080
Total 40874 (99)
Interventions Standard licensing skill test vs the Motorcycle Operators Skill Test (MOST) with or without a training component
MOST is not to be confused with rider training because it is a licensing test The training component was for
participants who failed the skill test on the first attempt was of 3 hours duration and consisted of instructions about
the skills needed to pass the skill test The training consisted of 30 min theory and 2 frac12 hours practice on the
motorcycle
Intervention integrity
33 in MOST with skills training 31 in MOST without skills training and 43 in control group obtained their
motorcycle licence in the end
Outcomes Police recorded crashes and offences
Length of follow up 3 years
30Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3350
Anderson 1980 (Continued)
Notes Study conducted in California USA
Billheimer 1998
Methods Cohort study
Participants Initial sample size
Intervention group Unclear numbers of CMSP graduates and ERC graduates
Control group Over 16000 untrained motorcyclists
Analysis sample size
1139 matched CMSP-pairs and 1182 ERC-pairs
Interventions No formal training vs the mandatory (for those younger than 18 years) pre-licence 16-hours basic California
Motorcyclist Safety Program (CMSP) and the 8-hour Experienced Rider Course (ERC)
Intervention integrity
No information given
Outcomes Police recorded crashes and offences
Length of follow up 2 years
Notes Study conducted in California USA
Cooper 1988
Methods Cohort study
Participants Initial sample size
Intervention group 418 graduates and 57 who failed
Control group 402
Analysis sample size
Intervention group unclear
Control group unclear
Total 863 (984)
Interventions No formaltrainingvs non-mandatory pre-licence rider trainingThe content of thetrainingwas notfurther explainedIntervention integrity
Participants had all completed the rider training
Outcomes Police recorded crashes and offences
Length of follow up 7 years
Notes Study conducted in British Columbia Canada
31Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3450
Davis 1997
Methods Retrospective cohort study
Participants Initial sample size
Intervention group 9320 graduates
Control group unclear
Analysis sample size
Intervention group 9320 graduates
Control group unclear
Interventions No formal training vs the non-mandatory Connecticut Motorcycle Operator Training Program The content of
training was not further explained
Intervention integrity Participants had all completed the rider training
Outcomes Police recorded crashes severity of crashes
Length of follow up 1 year
Notes Study conducted in Connecticut USA
Hall 1988
Methods Randomised trial
Participants Initial sample sizeIntervention group A 6248
Intervention group B 6616
Intervention group C 6996
Control group 6604
Total 26464 motorcycle licence applicants
Analysis sample size
Intervention group A 6012 (96)
Intervention group B 6428 (97)
Intervention group C 6778 (97)
Control group 6502 (98)
Interventions Standard licensing skill test vs the Motorcycle Operator Skill Test (MOST II) without training (A) MOST II with
3 hours training (B) or MOST II with 20 hours training (C) The 3 hours training comprised instructions aboutthe skills needed to pass the skill test (30 min theory and 2 frac12 hours practice on the motorcycle) The 20 hours
training consisted of the Motorcycle Safety Foundationrsquos Motorcycle Rider Course (MRC) comprising of 8 hours
theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-road practice
Intervention integrity
25 and 26 obtained their motorcycle licence in the lsquoMOST 3-hour trainingrsquo and lsquoMOST 20-hour trainingrsquo
intervention arm respectively In the lsquoMOST testrsquo intervention arm 29 obtained their licence and in the control
group 35
Outcomes Police recorded crashes and offences
Length of follow up 3 years
32Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3550
Hall 1988 (Continued)
Notes Study conducted in New York USA
Haworth 1997
Methods Population based case-control study
Participants Cases Motorcycle riders or pillion riders involved in motorcycle crash (on road) Recruited from hospital and police
records Not clear whether all cases (especially fatalities) were included
Community controls Multiple matched sample of motorcycle rider or pillion who passed the crash site at the same
time of day and day of week that the crash occurred
Initial sample size
Cases 336
Controls 1195
Analysis sample size
Cases 234 (70)
Controls 561 (47)
Interventions Basic handling skills course beginner intermediate or advanced course measured through self-report not validated
Outcomes Injuries with an Injury Severity Score of 5 or greater or admission for at least four hours and fatalities from a
motorcycle crash (from hospital and police records)
Length of follow up two years or more prior to the interview
Notes Study conducted in Victoria Australia
Hurt 1981
Methods Population based case-control study
Participants Cases Motorcycle riders who had a motorcycle crash that was reported to police (traffic accident reports) Not all
cases included
Community controls Multiple matched sample of motorcycle riders who passed the crash site at the same time of
day and day of week as the crash occurred but 2 years later
Initial sample size
Cases 4500 notifications of crashes to police (only 50 received in time research team able to respond to 1126 for
unreported reasons data collection completed for 900)
Controls Number approached unreported 2310 consented and interviewed
Analysis sample size
Cases 808 (808900 data on rider training) - 18 of initial sample
Controls 670 (6702310 data on rider training) - lt 29 of initial sample
Interventions Training experience self taught friends-family school-club formal-professional others measured through self-
report validated
Outcomes Injuries and fatalities from a motorcycle crash (from hospital records and coronerrsquos office)
33Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3650
Hurt 1981 (Continued)
Notes Study conducted in Los Angeles USA
Jonah 1982
Methods Cohort study
Participants Initial sample size
Intervention group
2310 licensed MTP graduates who are active riders
Control group
2029 licensed motorcyclists who are active riders
Non-licensed and non-active riders excluded
Analysis sample size
Intervention group
811 interviews (35) 533 driver records (23)
Control group 1080 interviews (71) 905 driver records (45)
Interventions No training vs the non-mandatory Canada Safety Council Motorcycle Training Program (MTP) comprising of 4
hours theory and 16 hours practise (6 hrs on-road) over 2 weekends
Intervention integrity
Participants had all successfully completed the MTP
Outcomes Self-reported and police reported crashes and offencesLength of follow up 4 years
Notes Study conducted in Ontario Canada
Kloeden 1994
Methods Cohort study
Participants Initial sample size
Intervention group 2533 licence applicants
Control group Unclear
Sample of riders obtaining probationary licenceIntervention group 1625 (64)
Control group 5015 (unclear)
Non-active riders excluded from control group (~66)
Analysis sample size
Intervention group 631 (39)
Control group 318 (6)
Interventions No formal training vs the mandatory pre-licence Ridersafe course consisting of two levelsthe first is a training course
in the basics of motorcycle riding where a pass entitles the rider to obtain a learnerrsquos permit The second is a skill
test where a pass entitles the rider to obtain a probationary motorcycle licence The training course consists of two 4
hour training sessions on separate days involving instruction videos and actual riding on motorcycles The skill test
consists of a 4 hour preparation course of 74 minutes of lecturediscussion 38 minutes of video and 75 minutes of
34Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3750
Kloeden 1994 (Continued)
on-course riding which is the testing phase
Intervention integrity
Unclear what percentage in the Ridersafe group and the control group obtained their probationary licence halfway
and their full motorcycle licence in the end
Outcomes Police reported crashes
Length of follow up 1 year
Notes Study conducted in SA Australia
Kraus 1975
Methods Population based case-control study
Participants Cases All motorcycle riders with a confirmed medically treated motorcycle injury (including fatalities) from all
hospitals in study base and peripheral area hospitals and all deaths due to motorcycle crashes in the study base All
serious injuries (leading to hospitalisation) and fatalities likely to be included
Community controls A random sample of registered motorcycle owners in the same temporary and geographical
study base as the cases arose from
Initial sample size Cases 1273
Controls 738
Analysis sample size Cases 741 (58)
Controls 434 (59)
Interventions Motorcycle rider training (not specified) measured through self-report
Outcomes Injuries leading to hospitalisation fatalities
Length of follow up not specified
Notes Study conducted in Sacramento county USA
Leung 1987
Methods Non-randomised trial
Participants Initial sample sizeIntervention group unknown number of MRC graduates
Control group unknown number of licensed motorcyclists
Analysis sample size
Intervention group 2941
Control group 43094
Interventions No formal training vs non-mandatory Wisconsin motorcycle rider course a 22 hours off-street programme that
teaches the basic rules of safe motorcycling
Intervention integrity
50 in the MRC group and 100 in the control group obtained their motorcycle licence in the end
35Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3850
Leung 1987 (Continued)
Outcomes Police recorded crashes and offences
Length of follow up up to 3 years
Notes Study conducted in Wisconsin USA
MAIDS 2004
Methods Population based case-control study
Participants Cases Motorcycle riders who had a motorcycle crash Unclear whether all cases included
Community controls Motorcycle riders identified at petrol stations Unclear whether matched to cases
Initial sample size
Cases not reported
Controls not reported
Analysis sample size
Cases 921 (Italy 200 Netherlands 200 Spain 121 Germany 250 France 150)
Controls 923 (Italy 200 Netherlands 200 Spain123 Germany 250 France 150)
Interventions Pre-license rider training or additional training (non-compulsory)
Outcomes Crashes
Notes Study in five sampling areas located in five European countries (France Germany Netherlands Spain and Italy)
McDavid 1989
Methods Cohort study
Participants Sample size after first match
Intervention group 346 male training graduates
Control group 346 male motorcycle riders licensed in same year
Sample size after second match
Intervention group 139 (40)
Control group 139 (40)
Analysis sample size
Intervention group 139 (40)
Control group 139 (40)
Interventions No formal training vs the non-mandatory pre-licence British Columbia Safety Councilrsquos 37-hour motorcycle safety
programme
Intervention integrity
Participants had all completed and passed the training
Outcomes Police reported crashes injurious crashes
Length of follow up 1-6 years
36Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3950
McDavid 1989 (Continued)
Notes Study conducted in British Columbia Canada
Mortimer 1984
Methods Cohort study
Participants Initial sample size
Intervention group 608 MRC graduates
Control group 360 motorcycle riders
Response rates
Intervention group 360 (59)
Control group 324 (90)
Non-active riders excluded
Analysis sample size
Intervention group 213 (35)
Control group 303 (84)
Interventions No formal training vs the non-mandatory Motorcycle Safety Foundationrsquos Motorcycle Rider Course comprising of
8 hours theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-road practice
Intervention integrity
Participants had all completed and passed the MRC
Outcomes Self-reported crashes and offencesLength of follow up 1 year
Notes Study conducted in Illinois USA
Mortimer 1988
Methods Cohort study
Participants Initial sample size
Intervention group 2307 MRC graduates
Control group 556 active licensed motorcycle riders
Response ratesIntervention group 1298 (56)
Control group 500 (90)
Non-active riders excluded
Analysis sample size
Intervention group 913 (40)
Control group 500 (90)
Interventions No formal training vs the non-mandatory Motorcycle Safety Foundationrsquos Motorcycle Rider Course comprising of
8 hours theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-road practice
Intervention integrity
Participants had all completed and passed the MRC
37Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4050
Mortimer 1988 (Continued)
Outcomes Self reported crashes and offences
Length of follow up 1 year
Notes Study conducted in Illinois USA
Osga 1980
Methods Cohort study
Participants Initial sample sizeIntervention group 1163 MSF MRC participants invited 818 (74) responded
Control group 329 motorcycle riders responded to recruitment strategies
Non-active riders excluded from analysis
Analysis sample size
Intervention group 595 (73)
Control group 329 (100)
Interventions No formal training vs the non-mandatory Motorcycle Safety Foundationrsquos Motorcycle Rider Course (MRC) com-
prising of 8 hours theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-road
practice
Intervention integrity
Participants had all successfully completed the MRC
Outcomes Self-reported crashes police recorded crash reports
Length of follow up 6 months - 3 12 years
Notes Study conducted in South Dakota USA
Perrino 2002
Methods Cohort study
Participants Initial sample size
Intervention group 2144 (unclear) MRC graduates (motorcycle licence applicants)
Control group 2601 (unclear) regular licensing test graduates (motorcycle licence applicants)Analysis sample size
Intervention group 1070 (unclear)
Control group 867 (unclear)
Interventions No formal training vs the mandatory (for those younger than 18 years) pre-licence Maryland Motorcycle Rider
Course (MRC) consisting of 8 hours theory and 12 hours practice
Intervention integrity
No information given
Outcomes Self-reported and police recorded crashes and offences
Length of follow up 1 year
38Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4150
Perrino 2002 (Continued)
Notes Study conducted in Maryland USA
Raymond 1979
Methods Cohort study
Participants Initial sample size
Intervention group
258 volunteers mainly (but not all) novice riders
Control group
1104 provisional licence applicants
Analysis sample size
Intervention group 190 (74)
Control group 627 (57)
Interventions No formal training vs the non-mandatory RACACU (Royal Automobile ClubAuto-Cycle Union) training scheme
The training program consisted of either 24 one-hour sessions or 13 two-hour sessions Twelve hours of the training
focused on theory about basic operational procedures traffic laws and highway code The practical session focused
on basic operational skills riding on a training area and riding in traffic with an instructor A proficiency test was
administered at the end of the course
Intervention integrity
No information given
Outcomes Self-reported crashes (including minor non-injury non-damage and off-road crashes)
Length of follow up 1 year
Notes Study conducted in the UK
Satten 1980
Methods Cohort study
Participants Initial sample size
Intervention group interviewers kept interviewing until they had 100 riders selected from a population of 891
enrollees in the MSF MRC course (unknown how many were asked and how many refused to participate)Control group unknown number of registered motorcycle riders
Non-active riders excluded from analysis
Analysis sample size
Intervention group 71
Control group 100
Interventions No formal training vs the non-mandatory Motorcycle Safety Foundationrsquos (MSF) Motorcycle Rider Course (MRC)
comprising of 8 hours theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-
road practice
Intervention integrity
Participants had all successfully completed the MRC
39Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4250
Satten 1980 (Continued)
Outcomes Self-reported crashes injuries and offences
Length of follow up 3 years
Notes Study conducted in Illinois USA
Savolainen 2007
Methods Cohort study
Participants Initial sample size
Intervention group 4000 riders that had conducted the MSF MRC course (passed or failed)
Control group 4000 motorcycle owners or permit or endorsement holders from the Indiana Bureau of Motor
Vehicles database
Analysis sample size
Intervention group 739 (18)
Control group 588 (15)
Interventions Intervention integrity
No information given
Outcomes Probability of crash involvement
Length of follow up up to 5 years
Notes Study conducted in Indiana USA
Swaddiwudhipong 1998
Methods Randomised trial
Participants Initial sample size
Intervention group
all 16 villages of 3 sub-districts
(participants were practising motorcyclists licensed and unlicensed)
Control group
18 villages of 3 other sub-districtsAnalysis sample size
Intervention group 3 districts (100)
Control group 3 districts (100)
Interventions No rider education vs non-mandatory community based rider education consisting of theory on motorcycle injury
risk helmet use traffic laws vehicle regulations and traffic signs and a written and skill test leading to a licence
Intervention integrity
No information given
Outcomes Injuries fatalities
Length of follow up 2 years
40Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4350
Swaddiwudhipong 1998 (Continued)
Notes Study conducted in Tak Province rural Thailand
Waller 1992
Methods Cohort study
Participants Initial sample size
Intervention group 349 riders who succesfully completed the Department of Public Safety motorcycle operator
training in 1988
Control group 349 motorcyclists who never completed the DPS motorcycle operator training
Analysis sample size
Intervention group 349
Control group 349
Interventions No DPS motorcycle operator training vs non-mandatory DPS motorcycle operator training The DPS training is
the Motorcycle Safety Foundationrsquos Motorcycle Rider Course comprising eight hours of theory (lectures discussions
films and multiple choice tests) a textbook and 12 hours off-road practice
Intervention integrity
All 349 riders succesfully completed the rider training
Outcomes Police recorded crashes
Length of follow up 3 years
Notes Study conducted in
Characteristics of excluded studies [ordered by study ID]
Study Reason for exclusion
Anderson 1980a Study type review and process evaluation of a rider course
Aupetit 2008 Outcome proposal of a rider training curriculum that combines traditional training and a motorcycle riding
simulator
Awane 1999 Study type review on use of simulators in rider and driver training
Baldi 2005 Intervention not rider training but intervention integrity of rider training availabilty of rider education
implementation of graduated licensing system etc on a state-wide level
Batchler 1988 Study type description of a training course
Blanchard 2006 Study type not a study assessing the association between rider training and outcomes
Bonnett 2005 Outcome qualitative data regarding satisfaction with the training programme
41Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4450
(Continued)
Braver 2007 Outcome licensure rates
Clayton 1980 Outcome knowledge skills attitude
Council 1978 Outcome crash reduction in ANY vehicle not motorcycles exclusively
Crowther 2005 Study type qualitative study
Daniello 2009 Study type review of the effectiveness of motorcycle training and licensing
Dewanti 2007 Intervention emotional characteristics behaviour (ie not motocycle rider training)
Du Plooy 1990 Study type review of types of rider training courses in South Africa
Engel 1989 Study type uncontrolled before-after study
Goldenbeld 2002 Outcome skills
Goldenbeld 2004 Outcome knowledge and skills
Haworth 2000 Intervention balance between attitudinal andvehicle control skills acrosscurriculla of motorcycle rider training
courses
Haworth 2005 Study type review
Haworth 2006 Study type cross-sectional survey
Hill 1983 Study type review and description of training system
Imatake 1980 Study type course descriptionreview
Ingebritsen 1990 Intervention risk factors for motorcycle crashes
Jain 2009 Study type cross-sectional descriptive study of motorcycle crashes over a 5 year period
Jonah 1979 Study type cross-sectional survey Outcome scores on motorcycle operator skill test
Jonah 1981 Intervention evaluation of a motorcycle rider skill test not motorcycle rider training
Kadar 1990 Study type review of components of rider training courses
Kelsey 1986 Intervention evaluation of a motorcycle rider skill test not motorcycle rider training
Koch 1980 Study type reviewcourse description
Koch 1990 Study type description of course content
42Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4550
(Continued)
Labbett 2005 Study type essay about possible explanations for motorcycle deaths
Law 2005 Intervention targeted motorcycle safety programme not specifically rider training
Lowes 1990 Study type review and descriptions of rider training
Mahon 1976 Study type review of motorcycle rider training methods
Maxwell 1985 Outcomesrecommendations forprogramme operatorsand otherson howthe motorcycle trainingprogramme
works and how it can be improved
Mayhew 1996 Study type review
McKnight 1987 Study type evaluation of methodological and administrative aspects of a motorcycle operator skill test
McPherson 1976 Outcome knowledge and skills
Mohan 2004 Study type review of evidence of interventions to reduce road traffic injuries
Morgan 1980 Study type reviewcourse description
Motoki 1990 Study type proposal and development of rider training curriculum
Nairn 1992 Study type literature review
Newland 1980 Study type description of development of motorcycle safety course
Newland 1982 Study type descriptionreview of a rider training course
Newland 1990 Study type uncontrolled BeforeAfter study
Ormston 2003 Outcome changes in attitude and self reported behaviour
Osga 1979 Outcome knowledge attitudes and skills
Panou 2007 Intervention new simulation tool to teach driverrider
Peterson 1990 Study factor demographic details of course participants completing the course vs those not completing the
course
Prem 1984 Study type experimental
Study factor examines association between skills and performance in tests
Radin Umar 2006 Intervention motorcycle safety programme not specifically rider training
Rockwell 1990 Study type methodologically flawed study type
43Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4650
(Continued)
Rothe 1987 Study type qualitative study of rider training in general
Rowden 2007 Study type qualitative study
RSC Victoria 1993 Study type review of motorcycle rider training and education in Victoria
Russam 1979 Study type review of training programmes in Japan
Saffron 1981 Study type literature review
Schulz 1990 Study factor predictors of motorcycle crashes
Sexton 2004 Study type cross-sectional analytic analysis
Shephard 1986 Study factor reviewdescription of programme
Simpson 1990 Study type review of motorcycle rider trainingeducation
Sokolov 1990 Study factor review of motorcycle riding courses in USSR
Steffens 1988 Outcome self-reported behaviour experiences and researcher observed riding behaviour
Sudlow 2003 Study type review of motorcycle training in the UK
Tasmania 1995 Study type cross-sectional descriptiveprocess evaluation
Tetard 1985 Study type experimental
Outcome detection of indices
Thackray 1980 Outcome instructional effectiveness user acceptance administrative feasibility
Thompson 1994 Study type cross-sectional survey
Outcome attitudes and knowledge
Tobler 1990 Study type review of course curricula
Transport 1989 Intervention syllabus for rider training
Tuan 2007 Study type cross-sectional study (survey) which does not assess the effects of an intervention but examines
motocyclistsrsquo behaviour
Victor 1980 Study type review
Waterer 1990 Study type review of UK rider training issues
Weller 1989 Study type review of studies evaluating motorcycle safety and education measures
44Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4750
(Continued)
Wells 1981 Study type description of different methods of training
Winn 1986 Study type review of rsquohistoryrsquo of rider training
Wisher 1988 Study type cross-sectional survey
Wood 1987 Study type Interrupted Time Series but did not meet the criteria for an ITS as published by the Cochrane
Effective Practice and Organisation of Care Group
Yamazaki 1990 Study type cross-sectional survey
Outcome riding skills skills retention and attitude
Characteristics of ongoing studies [ordered by study ID]
Burgess 2010
Trial name or title UKs National Ride program
Methods Cohort study
Participants Motorcycle offenders
Interventions Police led diversion program
Outcomes Self reported attitudinal and behavioural measures
Starting date
Contact information
Notes Expected to be completed in early 2010
45Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4850
D A T A A N D A N A L Y S E S
This review has no analyses
A P P E N D I C E S
Appendix 1 Search strategy
We searched the following electronic databases
bull Cochrane Injuries Group Specialised Register (searched 22 Sept 2008)
bull CENTRAL (The Cochrane Library 2008 Issue 3)
bull TRANSPORT 1988 to 200706 (latest date available)
bull ATRI Australian Transport Index (to Sept 2008)
bull MEDLINE (Ovid SP) 1950 to Sept 2008
bull EMBASE (Ovid SP) 1980 to Sept 2008
bull CINAHL (EBSCO) 1982 to Sept 2008
bull WHOLIS (World Health Organization Library Information System) (to Sept 2008)
bull PsycINFO (Ovid SP) 1806 to Sept 2008
bull LILACS (BIREME) (Latin American and Caribbean Health Sciences) (1982 to Sept 2008)
bull ISI Web of Science Social Sciences Citation Index (SSCI) 1970 to Sept 2008
bull ERIC (searched 25 Sept 2008)
bull ZETOC (searched 25 Sept 2008)
bull SIGLE (searched 25 Sept 2008)
All of the databases were searched using the following text-word search strategy The strategy was adapted using either rsquorsquo or rsquo$rsquo as atruncation symbol
1 motor-cycl or motorcycl or motor-bik or motorbik or motor-scooter or motorscooter or scooter or moped or motocycl
or motocicl or moto
2 train or educat or instruct or program or course or qualif or learn or improv
3 1 and 2
H I S T O R Y
Protocol first published Issue 2 2005
Review first published Issue 10 2010
46Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4950
C O N T R I B U T I O N S O F A U T H O R S
Katina Kardamanidis was the first author of the protocol conducted the literature search extracted full studies assessed the studies
according to inclusion criteria assessed quality of the studies compared findings with the second author discussed disagreements with
the third author wrote the drafts and the final review
Rebecca Ivers was the second author of the protocol assessed the studies according to inclusion criteria assessed quality of the studies
compared findings with the first author edited the final draft and provided overall guidance for the process
Alexandra Martiniuk assessed quality of the studies compared findings with the first author discussed disagreements with the third
author and helped write the first draft and the final review
Mark Stevenson contributed to the protocol reviewed methods of studies and contributed to the final draft
Katrina Thistlethwaite conducted a second literature search extracted full studies and contributed to the final draft
D E C L A R A T I O N S O F I N T E R E S T
None known
S O U R C E S O F S U P P O R T
Internal sources
bull No sources of support supplied
External sources
bull National Health and Medical Research Council of Australia Australia
Salary support for Martiniuk Ivers and Stevenson
bull University of Sydney Sesqui Research Grant Australia
Direct research costs and salary support for Kardamanidis
D I F F E R E N C E S B E T W E E N P R O T O C O L A N D R E V I E W
There are no differences between the protocol and the review
I N D E X T E R M S
Medical Subject Headings (MeSH)
lowastMotorcycles Accident Prevention [lowastmethods] Accidents Traffic [lowastprevention amp control] Licensure Program Evaluation Wounds
and Injuries [mortality lowastprevention amp control]
47Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 5050
MeSH check words
Humans
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 1250
Table 4 Risk of attrition bias
Anderson 1980 Percentage of participants in analysis 99 forthe total group Unclear forintervention andcontrol groups
separately
Hall 1988 Percentage of participants in analysis 96-97 forall intervention groups and 98for the control group
Swaddiwudhipong 1998 Percentage of participating sub-districts in analysis 100 for both intervention and control sub-districts
Raymond 1979 Percentage of participants in analysis 74 for the intervention group and less than 57 for the control
group
Osga 1980 Percentage of participants in analysis 73 for the intervention group and 100 for the control group
Satten 1980 Percentage of participants in analysis is unknown Initial sample size is unclear
Jonah 1982 Percentage of participants in analysis 23 for the intervention group and 45 for the control group
Mortimer 1984 Percentage of participants in analysis 35 for the intervention group and 84 for the control group
Cooper 1988 Percentage of participants in analysis 98 for the total group Unclear for intervention and control
groups separately
Mortimer 1988 Percentage of participants in analysis 40 for the intervention group and 90 in the control group
Adams 1985 Percentage of participants in analysis 61 forthe total group Unclear forintervention andcontrol groups
separately
Leung 1987 Percentage of participants in analysis unclear because initial sample size was not reported
McDavid 1989 Percentage of participants in analysis 40 of final matched cohort
Kloeden 1994 Percentage of participants in analysis 39 for the intervention group and 6 for the control group
Billheimer 1998 Percentage of participants in analysis unclear percentage of the initial sample size
Perrino 2002 Percentage of participants in analysis unclear percentage of the initial sample size
Kraus 1975 Percentage of participants in analysis 58 of cases 59 of controls
Hurt 1981 Percentage of participants in analysis 18 of cases lt 29 of controls
Haworth 1997 Percentage of participants in analysis 70 of cases 47 of controls
Savolainen 2007 Percentage of participants in analysis 18 for the intervention group and 15 for the control group
Davis 1997 Percentage of participants in analysis 100 for the intervention group and unclear for the control group
10Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 1350
Table 4 Risk of attrition bias (Continued)
Waller 1992 Percentage of participants in analysis unclear percentage for the intervention group and 100 for the
control group
MAIDS 2004 Percentrage of participants in analysis unclear for both cases and controls
R E S U L T S
Description of studies
See Characteristicsof included studies Characteristics of excluded
studies Characteristics of ongoing studies
Our search resulted in 102 studies that seemed relevant of which
we excluded 79 studies because they did not meet the criteria for
considering studies for this review (see Characteristics of excluded
studies)
Basedon ourcriteriawe includeda total of 23 studies in this review
There were three randomised trials (Anderson 1980 Hall 1988
Swaddiwudhipong 1998) two non-randomised trials (Adams
1985 Leung 1987) 14 cohort studies (Raymond 1979 Osga
1980 Satten 1980 Jonah 1982 Mortimer 1984 Cooper 1988
Mortimer 1988 McDavid 1989 Waller 1992 Kloeden 1994
Davis 1997 Billheimer 1998 Perrino 2002 Savolainen 2007)
and four case-control studies (Kraus 1975 Hurt 1981 Haworth
1997 MAIDS 2004)
Over half of the studies were conducted 20 or more years
ago and the oldest was conducted 31 years ago Only three
studies were from this decade (Perrino 2002 MAIDS 2004
Savolainen 2007) Over 60 of the studies were conducted in
the USA (Kraus 1975 Anderson 1980 Osga 1980 Satten 1980
Hurt 1981 Mortimer 1984 Leung 1987 Hall 1988 Mortimer
1988 Waller 1992 Davis 1997 Billheimer 1998 Perrino 2002
Savolainen2007) three inAustralia (Adams 1985 Haworth 1997
Kloeden 1994) three in Canada (Jonah 1982 Cooper 1988
McDavid 1989) onefromthe United Kingdom (Raymond 1979)one from Europe (with sampling areas located in France Ger-
many Netherlands Spain and Italy) (MAIDS 2004) and one
in Thailand (Swaddiwudhipong 1998) Only seven studies were
published in peer reviewed journals (Kraus 1975 Jonah 1982
Mortimer 1984 Cooper 1988 Mortimer 1988 McDavid 1989
Swaddiwudhipong 1998) The other studies were reports forhigh-
way safety and road traffic organisations
The studies by Anderson 1980 Hall 1988 Kloeden 1994
Billheimer 1998 (the basic California Motorcyclist Safety Program
component of the study) and Perrino 2002 examined manda-
tory pre-licence training Cooper 1988 McDavid 1989 and
Swaddiwudhipong 1998 assessed non-mandatory training that
when passed resulted in a licence Raymond 1979 Osga 1980
Satten 1980 Jonah 1982 Mortimer 1984 Adams 1985 Leung
1987 Mortimer 1988 Waller 1992 Davis 1997 Billheimer 1998
(the Experienced Rider Course component of the study) andSavolainen 2007 assessed non-mandatory training that could be
taken by motorcycle riders with and without a licence The case-
control studies (Kraus 1975 Hurt 1981 Haworth 1997) assessed
lsquoanyrsquo type of rider training and therefore could not be grouped by
pre- or post-licence rider training The MAIDS 2004 case-control
study did make a distinction between participants having received
compulsory pre-licence training or additional (non-compulsory)
training We did not find a study that compared one form of pre-
or post-licence training (eg on-road off-road theory or a com-
bination) versus another form of training All studies compared
eithermandatory or non-mandatory formal training versus no for-
mal training with the exception of Haworth 1997 who compared
different levels of training courses (beginner intermediate and ad-vanced)
We have presented a more detailed description of the individual
studies in Characteristics of included studies We have provided
a description below of the types of motorcycle rider training that
were evaluated in the included studies
Types of motorcycle rider training evaluated in the
studies
The types of training were varied Duration of training ranged
from two or three hours to 24 hours administered over two or
three days Most training consisted of a theoretical and a practical
component usually off-road The content ranged from provid-ing riders with manuals lectures discussions showing films and
administering knowledge and skill tests Little information was
provided about the qualifications of trainers who conducted the
training
Six studies (Osga 1980 Satten 1980 Mortimer 1984 Mortimer
1988 Waller 1992 Savolainen 2007) assessed the effectiveness of
the USA-based Motorcycle Safety Foundationrsquos Motorcycle Rider
Course (MSF MRC) This course was released in 1976 It com-
prises eight hours of theory (lectures discussions films and multi-
plechoice tests) a textbook and 12 hours of off-road practice The
Motorcycle Safety Foundation is a national not-for-profit organ-
11Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 1450
isation sponsored by manufacturers and distributors of motorcy-
cles Many USA states have adopted the MSF Basic Rider Courseand Experienced Rider Course (wwwmsf-usaorg) since its release
in 1976
The Motorcycle Operators Skill Test was also developed
by the USA Motorcycle Safety Foundation in coopera-
tion with others (httpwwwwebbikeworldcommotorcycle-
trainingmotorcycle-test) The MOST is a licensing test consist-
ing of a manual knowledge and skill test (comprising eight exer-
cises of increasing difficulty and skill requirements) and a mini-
skill test conducted with a simulator Some studies compared this
alternative licensing test to the standard licensing test However
as the MOST is not a type of rider training we have included only
the two studies that assessed the MOST in combination with one
or more training components in this review Both Anderson 1980and Hall 1988 assessed the MOST in combination with a train-
ing component of three hoursrsquo duration consisting of instructions
about the skills needed to pass the MOST (30 minutes of the-
ory and two and a half hours on the motorcycle) Hall 1988 also
assessed the MOST in combination with the Motorcycle Safety
Foundationrsquos Motorcycle Rider Course as described above
Other USA studies evaluated the Wisconsin motorcycle rider
course a 22-hour off-street programme that teachesthe basic rules
of safe motorcycling (Leung 1987) the USA Maryland Motorcy-
cle Rider Course (MRC) a mandatory (for those younger than 18
years)pre-licence 20-hour coursecomprising eight hours of theory
and12 hours of practice and created by the Maryland Motor Vehi-
cle Administration in 1985 (Perrino 2002) the California Motor-cyclist Safety Program (CMSP) a mandatory (for those younger
than 18 years) pre-licence 16-hour basic course (Billheimer 1998)
and Experienced Rider Course (ERC) of eight hoursrsquo duration
(Billheimer 1998) The CMSP still offers these courses ( http
wwwca-msporg) but they are now MSF courses as described
above Davis 1997 did not give any information about the assessed
rider training
The Canadian studies evaluated the British Colombia Safety
Councilrsquos 37-hour motorcycle safety programme (no further in-
formation on the content was provided) (McDavid 1989) and
the Canada Safety Council Motorcycle Training Program (MTP)
comprising four hours of theory and 16 hours of practice (six
hours on-road) over twoweekends (Jonah 1982) The MTPcoursewas made available in 1974 and is still being offered in Canada
(httpwwwsafety-councilorgtrainingmtpmtphtm) Cooper
1988 did not give any information about the assessed rider train-
ing However the study was conducted in British Colombia in
Canada in 1988 the same time as the McDavid 1989 study was
conducted
In the United Kingdom Raymond 1979 assessed the non-manda-
tory RACACU (Royal Automobile ClubAuto-Cycle Union)
training scheme The training programme consisted of either 24
one-hour sessions or 13 two-hour sessions Twelve hours of the
training focused on theory about basic operational procedures
traffic laws and highway codes The practical session focused on
basic operational skills riding on a training area and riding in traf-fic with an instructor A proficiency test was administered at the
end of the course
One study from rural Thailand (Swaddiwudhipong 1998) evalu-
ated a community-based rider education programme consisting of
theory on motorcycle injury risk helmet use traffic laws vehicle
regulations and traffic signs and a written and skill test leading to
a licence
TheAustralian studies evaluateda rider trainingcourse of 12 hours
over two consecutive days (four hours of theory and eight hours of
practice)administeredto AustraliaPosttelegram deliverers(Adams
1985) and a mandatory pre-licence Ridersafe course consisting
of two tiers the first tier being a training course on the basics
of motorcycle riding where a pass entitled the rider to obtain alearnerrsquos permit and the second tier being a skill test where a pass
entitled the riderto obtain a probationary motorcycle licence The
training course consisted of two four-hour training sessions on
separate days involving instruction videos and actual riding on
motorcycles The skill test consisted of a four-hour preparation
course of 74 minutes of lecturediscussion 38 minutes of video
and 75 minutes of on-course riding which was the testing phase
(Kloeden 1994) The Rider Safe course is still being provided by
Transport South Australia a state government organisation (http
wwwtransportsagovau )
The duration or content of rider training in the four case-control
studies (Kraus 1975 Hurt 1981 Haworth 1997 MAIDS 2004)
was not described
Risk of bias in included studies
We describe the methodological quality of the included studies
by study type below More detailed information can be found in
Table 1 Table 2 Table 3 and Table 4
When considering the findings from these studies it is important
to note the quality (or lack thereof) of reporting of the studies
Quality of reporting is the extent to which a report provides in-
formation about the design conduct and analysis of a study ( Juumlni2001) Huwiler-Muumlntener 2002 found that even though trials
may be well reported it does not mean that the methodological
quality is of equal standard and well-conducted trials may never-
theless be reported badly The reporting quality of most studies in
this review was poor Most studies were old and did not meet con-
temporary reporting standards This made data extraction com-
plicated It also made it difficult to contact authors for additional
information
Because of significant heterogeneity between studies as well as the
poor quality of most studiespooling of studyresults to calculatean
overall measure of effect was not appropriate Instead we provided
a description of our judgement of risk of bias in the text
12Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 1550
Risk of bias in the randomised trials
There were three randomised trials (Anderson 1980 Hall 1988Swaddiwudhipong 1998) Anderson 1980 and Hall 1988 had an
initial sample size of 41188 and 26464 motorcycle licence ap-
plicants respectively Both studies assessed the effectiveness of the
pre-licence MOST with a trainingcomponent Swaddiwudhipong
1998 evaluated the effectiveness of a community-based rider ed-
ucation programme in six districts
Allocation bias
Anderson 1980 reported using random birth date tables to allocate
participants to either the intervention or control group Hall 1988
reported the use of a computerised algorithm set in a central of-
fice but also used manual random assignment when the computer
went down Swaddiwudhipong 1998 reported lsquosimple random
samplingrsquo for the delivery of the intervention where the sampling
frame included nine districts Three districts were excluded since
they were too hard to access The remaining six districts were ran-
domly allocated to the intervention (three) and the control (three)
groups Systematic sampling wasthen used ordered by village size
for the villages within districts Allocation concealment was not
reported in any of the studies The descriptions of randomisation
in these studies suggest that there may have been limitations to
the process resulting in a lack of equivalence of the intervention
and control groups at baseline and therefore the possibility of
allocation bias Swaddiwudhipong 1998 reported that before theintervention the annual incidence rate of motorcycle-related in-
juries was slightly (but not statistically significant) higher in the
intervention areas than in the control areas Participants in the
intervention districts were also more likely to have a valid licence
and to wear a helmet more often than participants in the control
districts Anderson 1980 measured differences at baseline on the
participant and prior driving record variables and found these to
be negligible Hall 1988 did not report on baseline differences
between the intervention and control groups As the Anderson
1980 and Hall 1988 studies had sample sizes of over 40000 and
26000 respectively and used some form of randomisation it is
likely that any confounding factors at baseline would have been
equally distributed between comparison and intervention groups
Selection bias
After randomisation Anderson 1980 and Hall 1988 excluded
08 and 3 participants respectively because of assignment to
the wrong group administration of the wrong intervention or
other reasons The resulting lack of analysis by rsquointention-to-treatrsquo
can introduce biasby removing the effectivenessof randomisation
Nonetheless the percentage excluded was small and the number
of overall participants large so it is unlikely to have had a major
effect on the final results In the Swaddiwudhipong 1998 study
villages were systematically sampled into intervention or control
groups The intervention was implemented and outcomes of in-
jury and death were measured at a community level Interventionand control districts were randomly selected however there were
only three intervention and three control communities
Detection bias
None of the randomised trials reported blinding of outcome as-
sessors Data collection tools used by Anderson 1980 and Hall
1988 were driver records holding police recorded crashes and of-
fences Swaddiwudhipong 1998 used police and hospital records
at a district level and post-mortem examination for deaths None
reported on validity or reliability for any of the instruments
Attrition bias
Anderson 1980 and Hall 1988 had driver records available for
96 to 100 of participants They also collected self-reported
data but the response rate to the questionnaires was very low (low-
est response rate 42 and 30 respectively) Attrition bias in
Swaddiwudhipong 1998 was not an issue as outcome data were
available for all sub-districts included in the trial
The intervention reported in Anderson 1980 and Hall 1988 tri-
als had an unanticipated effect on the participants In both trials
an average of only 28 of licence applicants in the intervention
group eventually obtained their licence compared to 35 in the
control group This indicates that the licensing process in generalseems to have deterred applicants from obtaining a licence and
although no statistical tests of the difference were conducted this
deterrence appears even greater for those who had rider training
This is not true attrition bias as the studies did not experience
differential loss to follow up from outcome measurement Instead
differential attainment of the intermediary step of licensure pre-
cluded participants from being at risk for the outcomes of interest
they are not driving a motorcycle hence they are not at risk for
motorcycle-related crashes injury death or offences In short it is
not possible to assess whether a reduction in crashes in the inter-
vention group would be due to changes in riding habits as a result
of training or simply due to delayed or never attained licensure
Intervention integrity
Neither Anderson 1980 nor Hall 1988 reported on the consis-
tency in the way the intervention was administered the possible
lsquocontaminationrsquo of participants in the intervention group by other
types of formal or informal rider training or of participants in the
control group by the intervention Swaddiwudhipong 1998 did
not report on the number of villagers in each group that actually
attended the intervention However they did report on a number
of other potential training courses that were being implemented
at the same time
13Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 1650
Analysis
Anderson 1980 and Hall 1988 acknowledging the lsquodeterrent ef-fectrsquo described above conducted two separate analyses The first
analysis included the total study population of licence applicants
over two years from the time of their application (the population
denominator) The second analysis was performed on the licensed
(intervention and control) subgroups only two years from the
time of obtaining their licence (the licensed rider denominator)
As only a third were licensed and exposed to riding the motorcy-
cle it was impossible to determine whether reported differences
in crash rates across groups were due to the intervention or to
other lsquoconfoundingrsquo characteristics of those who were determined
in their efforts to obtain the licence When analysis was restricted
only to those who gained their licence the same difficulties re-
mained The benefits of the randomisation process were thus lost
due to this lack of licensure of riders during the process As there
were only three intervention and three control groups in the study
reported by Swaddiwudhipong 1998 analyses were hindered by
smallnumbers for outcome comparisons betweenthe intervention
and the control districts
Risk of bias in the cohort studies and non-randomised
trials
We discuss the methodological quality of cohort studies (seven
prospective studies (Raymond 1979 McDavid 1989 Waller 1992
Kloeden 1994 Billheimer 1998 Perrino 2002 Savolainen 2007)and seven retrospective studies (Osga 1980 Satten 1980 Jonah
1982 Mortimer 1984 Cooper 1988 Mortimer 1988 Davis
1997)) and non-randomised trials (Adams 1985 Leung 1987) in
the one section as the biases for the two study types are similar
The quality of these studies was in general poor
Selection bias
The confounding factors expected to have been addressed in the
studies were sex age riding experience prior driving records (of-
fences crashes) previous training and the following proxies for
risk taking behaviour helmet use speed and alcoholdrug use All
of these factors have been shown to be associated with an increasedrisk of motorcycle crash (Peden 2004 Zambon 2006) The pos-
sible source of bias in cohort studies and non-randomised trials
is that groups may differ on important characteristics at baseline
participants lsquoself-selectrsquo into either the rider training group by de-
ciding to undertake a rider training course or into the lsquocontrol
grouprsquo by deciding not to undertake a course Participants who
undertake a non-mandatory rider training course may differ from
those who choose not to on characteristics such as age sex rid-
ing experience and riding behaviour as well as other possible con-
founders For example riders who take a training course may be
more safety-conscious or on the contrary may do so because they
have previously experienced a crash or because they perceive that
they have poor riding skills Therefore ideally all participants in
the intervention group need to be included in the study as wellas a random sample of controls taken from the same population
as the intervention group If that is not possible intervention and
control groups either need to be carefully matched to make them
comparable or differences at baseline need to be adjusted for in
the analysis
Most of the studies using a non-randomised trial or cohort design
compared an intervention group to a control group that differed
on a number of important factors The studies in general failed to
adjust for important differences found at baseline and their results
cannot be meaningfully interpreted Ten studies (Raymond 1979
Osga 1980 Satten 1980 Jonah 1982 Mortimer 1984 Mortimer
1988 McDavid 1989 Waller 1992 Davis 1997 Savolainen 2007)
assessed intervention and control groups that were not drawnfrom the same population In most of these cases the intervention
group was selected from a list of training graduates who were usu-
ally young female and less experienced riders whereas controls
were recruited from a general population of motorcycle riders who
were usually older male and experienced riders For three stud-
ies it was unclear how they were selected (Leung 1987 Cooper
1988 Perrino 2002) Only Kloeden 1994 recruited groups from
the same population of motorcycle applicants and Adams 1985
from a group of professional riders Nevertheless due to non-ran-
dom selection of intervention and control groups differences at
baseline are likely even though study groups were from the same
population Three studies adjusted for some confounding fac-
tors found at baseline through matching (McDavid 1989 Waller1992 Billheimer 1998) However matching has limitations in
termsof controlling for multiple confounders especially those that
are unknown to influence the effect of rider training on the out-
comes One study matched the control and intervention groups
by excluding controls (approximately 3500 (66 of initial con-
trol group) that differed from the intervention group) (Kloeden
1994) Five studies (Osga 1980 Satten 1980 Adams 1985 Davis
1997 Perrino 2002) did not adjust for differences between study
groups at baseline either in the study design or analysis Results
from these studies are therefore not reliable The other six stud-
ies (Raymond 1979 Jonah 1982 Mortimer 1984 Cooper 1988
Mortimer 1988 Leung 1987 McDavid 1989 Billheimer 1998)
adjusted for some differences between groups at baseline by eitherbi-variate or multi-variate analyses or stratification and standardi-
sation of riding exposure (outcomemiles driven) or stratification
by age group The limitation of stratification is that it is difficult to
control for many variables simultaneously because a large number
of strata will need to be generated resulting in small numbers that
cannot be meaningfully analysed Savolainen 2007 developed sta-
tistical models to assess the probability of a crash after taking the
rider training course in an intervention and control group while
examining differences by sub-groups (eg age) Participation rates
were also very low in this study with only 5584000 (1395)
in the eligible intervention group and 5884000 (147) for the
14Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 1750
control group completing surveys
Detection bias
None of the studies reported blinding of outcome assessors to the
intervention or control status of the participants Seven studies
used driver records containing police recorded data as their main
measure of outcome (Leung 1987 Cooper 1988 McDavid 1989
Waller 1992 Kloeden 1994 Davis 1997 Billheimer 1998) An-
other four studies (Raymond 1979 Satten 1980 Mortimer 1984
Mortimer 1988) only used self-report Raymond 1979 was the
only study using self-report as the main outcome measurement in
which study instruments were validated Self-report might prevent
the underreporting that has been described for the use of police-
driver records (see rsquoRisk of biasrsquo in the randomised trials above)but has other risks of bias especially if questionnaires or inter-
view methods have not been validated and tested for reliability
Recall bias arises from mistakes in recollecting events both be-
cause of failures of memory and because of looking at things lsquowith
hindsightrsquo and possibly changed views (httpwwwcochraneorg
resourcesglossaryhtm) As many studies were retrospective risk
of recall bias is likely and may have increased with length of follow
up For studies that used self-reported outcomes only it is also
possible that social desirability response bias played a role where
participants in the training group may have felt it was lsquoexpectedrsquo
to report good driving histories Multiple methods to measure the
same outcome for both groups decreases the risk of detection bias
for example because self-reported answers can be cross checkedwith other sources of information such as driver records
In this regard three studies (Osga 1980 Jonah 1982 Perrino
2002) used both self-report and driver records Osga 1980 found
that for both intervention and control groups more crashes were
reported through self-report than on state crash records an aver-
age of 54 of self-reported crashes could be found back on state
crash records One industry-based study (Adams 1985) used oc-
cupational health records which were not validated but which are
unlikely to suffer from underreporting due to work cover claims
Savolainen 2007 used information on rider crashes for a five-year
period but did not specify from where this information was ob-
tained
Attrition bias
The majority of studies (Raymond 1979 Satten 1980 Adams
1985 Leung 1987 Kloeden 1994 Davis 1997 Billheimer 1998
Perrino 2002) inadequately reported on losses of participants to
the study We can therefore not be certain that there were no sys-
tematic differences in the characteristics between the intervention
and control group participants Savolainen 2007 reported the re-
sponse rates to their baseline survey but did not report on the per-
centage of outcome data that was available for their study sample
A similar deterrent effect of rider training as described for the ran-
domised trials (Anderson 1980 Hall 1988) occurred in two stud-
ies Jonah 1982 reported that 65(14992310) of the trained rid-
ers stopped riding compared to 47 (9492029) of the untrainedriders They excluded non-active and non-licensed riders before
the study commenced so the differential choice to continue riding
after training did not affect their analyses Leung 1987 reported
that the percentage of participants in the intervention group ob-
taining their licence was 50 (no number provided by authors)
lower than in the control group This may have biased the mea-
surement of outcomes because unlicensed participants will not be
exposed to riding the motorcycle and therefore not be at risk of
being involved in a crash or violation As explained above this
makes it difficult to assess whether the reduction in crashes in the
intervention group is due to a change in driving habits or simply
due to delayed licensure
Mortimer 1984 and Mortimer 1988 describe that about 44 and30 of course graduates respectively were excluded from analysis
because they had not ridden a motorcycle in the previous year
However because these studies were retrospective and the controls
were activeriders at the time of recruitment a decision on whether
the deterrent effect was differential between the two groups could
not be made Though not reported quantitatively Kloeden 1994
also describes the effectof rider training deterringindividuals from
becoming licensed and hence exposure to riding and risk of crash
death or offences involving a motorcycle Osga 1980 reported
that 27 of the trained riders did not ride after the course and
excluded these from the analysis
Use of different data collection strategies may have contributed to
differential loss to follow up in the cohort studies Satten 1980for example interviewed the intervention group over the tele-
phone but mailed questionnaires to the control group Mortimer
1984 and Mortimer 1988 distributed questionnaires in person to
the control group but mailed them to the intervention group
This may have played a role in the observed differential response
rates where more subjects responded in the personally approached
groups
Survival bias
For those studies that assessed offences crashes and injuries but
not death survival bias may have also played a role Motorcyclists
who were fatally or seriously injured in crashes during the follow-
up period after the intervention would have been less likely or
unable to respond to surveys This could bias the effect measures
towards or away from the null None of the studies wrote about
this issue or examined the characteristics of those lost to follow up
in order to address the possibility of survival bias
Intervention integrity
Although the interventions were intended for the entire lsquoexposedrsquo
groups most studies reported difficulty in tracking participants
whoreceivedthe interventionand those whodid not Adams 1985
15Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 1850
estimatedthat 62of thetotalstudysample received the interven-
tion The remaining studies did not report the percentage of thestudy population that received the intervention as was intended
or reported that this percentage was unknown (Raymond 1979
Kloeden 1994 Satten 1980 Jonah 1982 Mortimer 1984 Leung
1987 Cooper 1988 Mortimer 1988 Billheimer 1998 Perrino
2002 Savolainen 2007) Only Raymond 1979 reported the degree
to which their intervention was delivered as intended No studies
reported on possible lsquocontaminationrsquo of participants by other inter-
ventions than the one to which they were allocated Because Osga
1980 Satten 1980 Jonah 1982 Mortimer 1984 Cooper 1988
Mortimer 1988 Waller 1992 and Davis 1997 were retrospective
studies and participants were recruited from a list of graduates it
was known that they had completed and passed the course
Analysis
In some studies some participants who reported at baseline that
they were untrained decided to enroll in a rider training pro-
gramme after the baseline survey The non-randomised trial by
Adams 1985 experienced this move of subjects from the untrained
to the trained group but failed to conduct analyses by lsquointention-
to-treatrsquo This can bias the results in randomised and non-ran-
domised trials if participants who changed groups differ signifi-
cantly with regard to potentially confounding characteristics On
the other hand Perrino 2002 also experienced this move of sub-
jects from theuntrained to the trained groupmore than 100riders(5 of the total sample) originally stated that they were untrained
at baseline but then received training between the baseline and
follow-up surveys Yet since the Perrino 2002 study was a cohort
their analyses then proceeded by including those riders who were
lsquonewlyrsquo trained as exposed Riders who were lsquonewlyrsquo trained since
baseline but with a programme other than the one under study
were excluded from the analysis Nevertheless confounding fac-
tors were not accounted for in the analyses
McDavid 1989 matched exposed with unexposed subjects but
could not find suitable matches for the majority (60) of their
subjects They could therefore only conduct analysis on the result-
ing40 (139346) which is an inappropriate andinefficient anal-
ysis Some studies that reported on the issue of non-active licensedriders(Satten 1980 Jonah 1982 Mortimer 1984 Mortimer 1988
Kloeden 1994) excluded non-active riders from analysis intro-
ducing selection bias as it is likely that riders who continued to
ride differed with regards to potentially confounding character-
istics across intervention and control groups Kloeden 1994 for
example conducted several analyses to assess the effect of various
sub-sets of the control group Original analyses included all con-
trols and led to erroneous results since only a small proportion of
controls hadgone on to get a licence after their learnerrsquos permitand
were actually exposed to the possibility of experiencing a crash
Subsequent analyses compared only intervention and control sub-
jects who had a licence Waller 1992 reported crashes by person-
days of exposure but did not explain how they could be sure that
each study intervention and control participant was an active riderand actually exposed to the risk of experiencing a crash
Potential confounders that are unequally distributed by study arm
at baseline need to be dealt with in the analysis Satten 1980
Adams 1985 and Davis 1997 did not take into account any con-
founding factors either by design or in the analysis McDavid
1989 Waller 1992 and Billheimer 1998 accounted for potential
confounding factors at the design stage by matching the subjects
in the intervention and control groups Other studies (Raymond
1979 Mortimer 1984 Mortimer 1988 Perrino 2002 and also
Waller 1992) adjusted for one or two confounders by stratifica-
tionor standardisation (eg crashesmiles driven age group) Osga
1980 conducteda univariate analysis of crash rates in which no po-
tential confounders were accounted for but also calculated cumu-lative crash rates by mileage travelled Only Jonah 1982 Cooper
1988 and Savolainen 2007 conducted a multivariate analysis
Risk of bias in the case-control studies
A case-control design is suitable for the assessment of risk fac-
tors for outcomes such as motorcycle crashes injuries or deaths
but not appropriate for the evaluation of an intervention such
as motorcycle rider training It can potentially result in spurious
conclusions regarding the efficacy of the intervention because the
intervention has not been applied at random to an intervention
and control group Even with the control of possible confounders
using a case-control design to evaluate an intervention can lead tothese kinds of results since those who choose to undertake rider
training may differ on important prognostic factors that affect the
outcome compared to those who do not choose to participate in
rider training
Selection bias
A potential source of bias in case-control studies as with all other
study types where the intervention is not randomly allocated to
participants is that comparison groups are likely to differ on po-
tentially confounding characteristics This means that ideally all
cases need to be included in the study as well as a random sample
of controls taken from the population that gave rise to the cases If that is not possible cases and controls either need to be carefully
matched to make themcomparable or differences at baseline need
to be adjusted for in the analysis All four case-control studies were
population based Kraus 1975 included all fatalities and injured
cases leading to hospitalisation and appropriately recruited a ran-
dom sample of controls from the same population Hurt 1981 did
not manage to recruit all or a random sample of cases Only half
of the notifications of crashes to the police were received in time
by the researchers the research team for unreported reasons was
only able to respond to half of those again Data collection was
therefore completed for only about a fifth of all cases Neverthe-
less this was still a large sample of 900 cases Haworth 1997 using
16Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 1950
a similar study design was more likely to have included all cases
but this is not entirely clear from the study reported It is unclearfor example whether fatalities were included as they did not use
coronerrsquos reports for the recruitment of cases The MAIDS 2004
study was conducted in five European countries but it was unclear
how cases were recruited whether they were all cases during the
two years that the study lasted or a random sample of all cases
what the response rates were etc Controls were recruited from
randomly selected petrol stations but it was likewise not clear what
the eligibility criteria for inclusion in the study were how they
were recruited (eg randomly) and what the response rate was
Hurt 1981 and Haworth 1997 both matched controls to cases by
recruiting themat the sametime ofdayandsamedayof weekas the
crash occurred Hurt 1981 however recruited controls two years
later andit is likelythat there would have beendifferences betweenthe two comparison groups One important difference was that
during data collection for cases there was very little specialised
motorcycletrainingavailable in the studyregionwhereas two years
later when control data were collected there was some training
available
These four case-control studies also suffered from selection bias
due to potential differences between those that gave consent to
participate in the study and responded to interviews andquestion-
naires and those that did not None of the studies have assessed
such differences although Hurt 1981 did acknowledge that con-
trols giving consent to cooperate may have differed from those
motorcyclists who did not
Detection bias
Measures that can be taken to reduce detection bias in case-con-
trol studies are blinding of interviewers for case or control status
blinding of the participant for the study factor of interest validat-
ing measurement instruments and using these in an identical way
for cases and controls and cross checking self-reported informa-
tion with other more objective sources such as registers None of
the studies reported blinding of the intervention assessors to case
or control status but all used consistent measurement methods for
both case and control groups
All case-control studies used self-report as their main measure toassess whether cases and controls had received rider training and
if so the type of training The major potential source of bias in
case-control studies when self-report is used as a measure is recall
bias One way of reducing recall bias is by double-checking self-
reported data with more objective data Kraus 1975 also used po-
lice accident reports for information on rider characteristics but
such information was not available about rider training Haworth
1997 and MAIDS 2004 did not report on any measures to limit
recall bias Hurt 1981 reported further questioning of cases by
the interviewer when answers did not make sense or when par-
ticipants seemed to contradict themselves as part of the in-depth
unstructured interview technique This qualitative research tech-
nique can result in valuable information but as interviewers were
not blinded to case or control status of the participants it may have led to more thorough examination of cases compared to the
controls and thus differential detection bias
Intervention integrity
The level of detail about the rider training provided in the case-
control studies differed significantly but was minimal for each
Kraus 1975 simply asked whether lsquoanyrsquo motorcycle rider training
had lsquoeverrsquo been carried out Hurt 1981 asked participants whether
they hadbeen self taught informally taught or completed formally
taught rider training courses MAIDS 2004 asked whether partic-
ipants had had pre-license training additional (non-compulsory)
training other training or no training whereas Haworth 1997asked about the number of courses that a participant had com-
pleted whetherthey were basic intermediate or advanced courses
and how long ago the training had been done None of the case-
control studies reported collecting information about further de-
tails about any training programmes (eg type of training cur-
riculum duration of training)
Analysis
Kraus 1975 accounted for confounding factors through the exclu-
sion of females and stratification by age and frequency of motor-
cycle use Hurt 1981 matched cases and controls for time of day
and day of the week that the crash occurred and did not adjust forany confounding factors in the analysis Haworth 1997 matched
groups for time of day and day of the week that the crash occurred
and stratified by age blood alcohol level and licence status in the
analysis None of the studies used multivariate methods MAIDS
2004 did not provide any information about any participant char-
acteristics other than the study exposures of interest and did not
describe any methods to adjust for potential confounding factors
Effects of interventions
We have presented the results in a descriptive manner due to the
substantial heterogeneity in study types interventions and out-comes and the generally poor quality of the studies We have
grouped the interventions by mandatory and non-mandatory pre-
licence rider training and post-licence rider training and further
by study type and by outcome (crashes injurious and fatal crashes
and offences) We discuss results that are reported as statistically
significant (whether a P value was reported or not)
Pre-licence rider training - mandatory
Randomised trials
17Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2050
Crashes
Total group of participants
Anderson 1980 found a reduction in motorcycle crash rates in
the intervention group (MOST skill test with remedial training)
compared to the control group (who received the skill testwithout
a training component) The percent differenceranged from 143
six months after the programme (P lt 005) to 211 (P lt 0001)
after one year and 127 (P lt 001) after two years Hall 1988
found no statistically significant difference in crash rates between
intervention groups at any point in time
Subgroup of participants who obtained their licence only
Anderson 1980 found a reduction in motorcycle crash rates in the
intervention arm but this was statistically significant at one year
only (24 to 218 reduction P lt 005) Hall 1988 found no
statistically significant differences
Offences
Total group of participants
The percentage of offences in Anderson 1980 after six months
and one and two years was slightly higher in the intervention
arm compared to the control group but these differences were
not statistically significant Likewise the study by Hall 1988 did
not find a statistically significant difference in offences between
groups
Subgroup of participants who obtained their licence only
Hall 1988 found no significant difference in offence rates between
the intervention and control groups Anderson 1980 found thatthe percentage of offences increased in the intervention arms com-
pared to the control group This difference was statistically signif-
icant after two years (51 to 115 increase P lt 005)
Cohort studies
Crashes
Neitherof the cohort studies (Kloeden1994 and Billheimer1998)
found a statistically significant difference in crash rates between
the intervention and control groups
Offences
Billheimer 1998 (for the California Motorcyclist Safety Program
course) and Perrino 2002 did not find a statistically significant
difference in offence rates between the intervention and control
groups
Case-control studies
Crashes
MAIDS 2004 found that a slightly higher percentage of controls
(not involved in a crash) had had pre-license rider training than
cases (involved in a crash) This difference was not statistically significant (50 of controls vs 477 of cases OR 091 95 CI
076 to 110)
Pre-licence rider training - non-mandatory
Randomised trial
Injuries and deaths
Swaddiwudhipong 1998 found a significantly lower injury ratein the intervention villages (1051000 population 95 CI 92
to 120) compared to the control villages (1691000 population
95 CI 152 to 185) one year after the intervention This differ-
ence was not maintained two years after the intervention (127
95 CI 112 to 143 intervention 151 95 CI 135 to 167
control) There was no significant difference in death rates be-
tween groups
Cohort studies
Crashes
Cooper 1988 foundno statistically significant differencein crashes
between the intervention and control groups McDavid 1989
found a lower crash rate in the intervention group (14139) com-
pared to the control group (23139) but no test of statistical sig-
nificance was undertaken
Injuries and deaths
McDavid 1989 found that the intervention group had fewer in-
jurious crashes (814 crashes) compared to the control group (16
23 crashes) but no statistical significance test was reported
18Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2150
Offences
Cooper 1988 foundno statistically significantdifference in offencerates between the intervention and control groups McDavid 1989
foundno statistically significantdifference in offence ratesbetween
the intervention and control groups
Non-mandatory rider training
Non-randomised trials
Crashes
Neither Adams 1985 nor Leung 1987 found a statistically signif-icant effect of rider training on crash rates
Offences
Leung 1987 found a statistically significant lower offence rate in
the intervention group compared to the control group for both
males and females and for all age groups except for women 35
years or older (P values not reported)
Cohort studies
Crashes
Osga 1980 in the univariate analysis reported a lower percentage
of crashesin theintervention group (that continued to ride) (28
168595) than in the control group (34 112329) In the cu-
mulative crash rate analysis however Osga1980 found statistically
significant (at the P lt 001 level) higher crash rates in the interven-
tion group than in the control group at allmileage intervals (1000
to 29999 miles) travelled After riding more than 30000 miles
(when over 50 of the trained riders had been involved in a crash
compared to approximately 40 in untrained riders) this differ-
ence was no longer statistically significant Satten 1980 found a
lower percentage of crashes in the intervention group (13 969)
compared to the control group (25 2079) but the differencewas not assessed statistically Davis 1997 found a lower crash rate
in trained riders but it was unclear how this was calculated be-
cause the size of the control group was not reported Jonah 1982
and Billheimer 1998 (for the Experienced Rider Course) found no
statistically significant difference in crashes between the interven-
tion and control groups Although the difference between those
who completed the Experienced Rider Course and no course in
Billheimer 1998 neared significance at six months after the course
(P = 007) large P values were observed for one year after (P =
030) Raymond 1979 found a statistically significant higher crash
rate for the intervention group (1861000 miles ridden) com-
pared to the control group (1181000 miles ridden) in both males
and females (P lt 0001) especially during the first 5000 miles of
riding Waller 1992 found a statistically significant higher crashrate for the intervention group (27 crashes in three years or 070
crashes10000 person days) compared to the control group (5
crashes in three years or 013 crashes10000 person days) with a
crude relative risk of 538 (X 2=1513 plt005) Mortimer 1984
and Mortimer 1988 found no statistically significant difference in
crashes between the intervention and control groups
Savolainen 2007 found that the intervention group (those who
completed the Basic Rider Course) was 44 more likely to be
involved in crashes than the control group and those who com-
pleted the course multiple times were an additional 180 more
likely to be involved in a crash This finding was based on low
numbers and no P value or confidence level was given for either
result
Injuries and deaths
Satten 1980 also found a lower percentage of injurious crashes
(33 39) in the intervention group compared to the control
group(60 1220) although again no statistical significance test
was reported
Offences
Jonah 1982 did find a significantly lower self-reportedoffence rate
in the intervention group (15) compared to the control group(34 P lt 001) (no numbers given) Satten 1980 also found
a lower self-reported offence rate in the intervention group (1
71 versus 19100) but again this was not tested for statistical
significance Mortimer 1984 Mortimer1988 and Billheimer 1998
found no statistically significant differencein offence ratesbetween
the intervention and control groups
Case-control studies
CrashesMAIDS 2004 found that a slightly higher percentage of cases
(involved in a crash) had had additional non-mandatory rider
training than controls (not involved in a crash) This difference
was not statistically significant (17 of cases vs 12 of controls
OR 146 95 CI 064 to 337)
Rider training unspecified
Case-control studies
19Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2250
Injuries and deaths
Hurt 1981 found that the odds of injurious crashes were lower inthose who had received professional training compared to those
who had not Kraus 1975 found that in the younger age group
(less than 20 years) and the older age group (over 35 years) a larger
percentage of injured riders (12 and 21 respectively) reported
prior motorcycle rider training compared with the uninjured con-
trol group (6 and 5 respectively) (P lt 001) (no numbers
given) Haworth 1997 reported several different outcomes one
beingthat having completed at leastone trainingcoursecompared
to no training courses at all had no significant effect on the odds
of crashing (afteradjustment for age blood alcohol level or licence
status (learner permitprobationaryfull)) When comparing three
different levels of training courses (beginner intermediate and ad-
vanced) with no training no significant differences were found
in the odds of crashing However after adjustment for age and
blood alcohol level completing an advanced course as compared
to a beginners course was associated with a significant decrease in
the odds of severe motorcycle crashes resulting in serious injury or
death (OR 02 95 CI 01 to 08)
D I S C U S S I O N
The objective of this review was to quantify the effectiveness of
pre- andpost-licence motorcycle rider training on the reduction of
traffic crash involvement injuries deaths and offences committedby motorcycle riders Most included studies were cohort studies
There were a few randomised and non-randomised trials and case-
control studies
All studies compared either mandatory or non-mandatory formal
training with no formal training except for Haworth 1997rsquos case-
control study that compared levels of training (beginner interme-
diate and advanced) to each other
We found studies that evaluated the effectiveness of manda-
tory pre-licence training in reducing crash and offence rates
(Anderson 1980 Hall 1988 Kloeden 1994 Billheimer 1998
(the basic California Motorcyclist Safety Program component of
the study) Perrino 2002) non-mandatory training that when
passed resulted in a licence (Cooper 1988 McDavid 1989
Swaddiwudhipong 1998) non-mandatory training that could be
taken by motorcycle riders with or without a licence (Raymond
1979 Osga 1980 Satten 1980 Jonah 1982 Mortimer 1984
Adams 1985 Leung 1987 Mortimer 1988 Waller 1992 Davis
1997 Billheimer 1998 (the Experienced Rider Course compo-
nent of the study)) and unspecified rider training on crash injury
and death rates (Kraus 1975 Hurt 1981 Haworth 1997 MAIDS
2004)
The studies provided conflicting evidence regarding motorcycle
rider training which is likely due to the poor design quality and
reporting of the studies Most of the studies included in the review
were likely to be adversely affected by significant selection biasalong with substantial detection bias(measurement and recall bias)
which limits our ability to make conclusive comments
Mandatory pre-licence rider training
There were two RCTs three cohort studies and one case-control
study that examined pre-licence ridertraining The two RCTs eval-
uated the Motorcycle Skills Operating Test (MOST) (Anderson
1980 Hall 1988) Anderson 1980 found a statistically significant
reduction in crashes in the intervention group at six months one
and twoyearsfollow up However when only participantswho ob-
tainedtheirlicence were included in theanalysis thereduction wasstatistically significant only at one year follow up The other RCT
(Hall 1988) found no statistically significant differences in crash
rates at any point of time neither in the total population group
nor in the licensed subgroup Neither of the two cohort stud-
ies assessing crash rates found a statistically significant difference
in crash rates between intervention and control groups ( Kloeden
1994 Billheimer1998) andthe case-control study (MAIDS 2004)
also did not find a statistically significant difference in the odds of
experiencing a crash between trained and untrained riders With
regards to offences neither of the RCTs found a statistically sig-
nificant difference in offence rate between the intervention and
control groups in the analysis of the total group of participants
entering the trial However Anderson 1980 found a statistically significant increase in offence rates in the intervention group but
only after two years of follow up andonly in the analysis of partici-
pants who obtained their licence The two cohort studies assessing
offence rates found no statistically significant difference between
the groups (Billheimer 1998 Perrino 2002)
None of the five studies (RCTs or cohort) found a reduction in
crashes or offences following mandatory pre-licence rider training
except for Anderson 1980 who observed a statistically significant
reduction in crashes in the intervention group at one year in both
analyses It is not clear why both RCTs that evaluated the same
MOST and training intervention observed different results Both
studies enrolled large samples and observed a similar uptake of
licensure following the intervention It is possible that differencesmay have arisen due to different study samples (California ver-
sus New York) or that the intervention delivery differed between
studies
In both trials of the Motorcycle Skills Operating Test including
rider training only a third of motorcycle licence applicants ob-
tained their motorcycle licence It is possible that a licensing pro-
cess including a skills test may reduce crashes through decreased
exposure
Non-mandatory pre-licence training
20Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2350
Three studies assessed the effect of non-mandatory pre-licence
training The Cooper 1988 cohort study found no difference incrash or offence rates between the trained and untrained groups
We cannot however be confident about this result as the study
sufferedfrom selectionbiaswhichcouldhave biasedthe measure of
effect toward the null The RCT conducted by Swaddiwudhipong
1998 examined injury and death rates and found a statistically
significant lower injury rate in the intervention villages after one
year of follow up but no longer at two year follow up There
was no statistically significant difference in death rates between
intervention and control villages Our confidence in the results of
this study are hindered by the inadequacy of the randomisation
process that led to significant differences between the intervention
and control districts on two important variables The intervention
districts were more likely than the control districts to have a validlicence andto wear a helmetSincethese differences were nottaken
into account in later analyses they could explain the differences
observed in injury rates between districts Furthermore with only
six districts randomised the sample size was probably too small
to detect a difference between intervention and control groups
had there been one The McDavid 1989 cohort study found a
lower percentage of injuries in the intervention group although
a statistical test was not conducted McDavid 1989 drew their
intervention and control groups from different populations but
did adjust for confounders although they only analysed 40 of
the initial sample (139 trained and 139 untrained) Because of
these substantial weaknesses we are unable to draw conclusions
from this studyAn overall effect of pre-licence training on the reductionof crashes
injuries deaths and offences was not found although one should
be aware that the research of this issue to date has had significant
flaws However a mandatory skill test (such as the MOST) in
combination with and even without pre-licence training as part
of a licensing process seems to reduce the number of individuals
completing the licensing procedure and therefore indirectly re-
duces crash rates through a reduction in exposure to motorcycle
riding Making licensure a more difficult process may therefore be
seen as advantageous if it delays licensure and therefore reduces
exposure It is possible that the group who chose not to continue
riding a motorcycle would not have been the group to experience a
crash or commit an offence if they had continued to ride anyhowfor example because they are more safety conscious and this was
the reason for their giving up riding in the first place If this is the
case deterring individuals from completing their licence and rid-
ing may not necessarily decrease overall crashes deaths or offence
rates In addition other potentially adverse effects of complicat-
ing a licensing procedure need to be considered such as increases
in unlicensed riding In Thailand non-mandatory rider training
offered as a community intervention may have decreased injury
rates but this needs to be confirmed through further research
Non-mandatory training
The majority of the studies assessed non-mandatory training
(training independent of any licensing procedure) and the results
were varied The two non-randomised trials assessing crash rates found no statistically significant difference in crash rates between
the intervention and control groups (Adams 1985 Leung 1987)
Of the nine cohort studies Satten 1980 found a lower percentage
of crashesin the intervention group although it is unclear whether
this was a statistically significant result Davis 1997 also found a
lower percentage of crashesin theinterventiongroup although it is
unclear with which control group the intervention group wascom-
pared Raymond 1979 Osga 1980 Waller 1992 and Savolainen
2007 found a statistically significant higher crash rate in the in-
tervention group and the remaining cohort studies (Jonah 1982
Mortimer 1984 Mortimer 1988 Billheimer 1998) found no sta-
tistically significant difference in crash rates between intervention
and control groups No conclusions can be drawn from these stud-
ies with confidence The Satten 1980 and Davis 1997 studies did
not adjust for any confounding factors even though important
differences existed between the intervention and control groups
Although Raymond 1979 and Waller 1992 adjusted for a num-
ber of confounders at baseline (Waller 1992) and in the analy-
sis (Raymond 1979 Waller 1992) the intervention and control
groups were drawn from such different populations that it is pos-
sible that unmeasured confounders were responsible for the sta-
tistically significant increased risk of crashes observed in the inter-
vention group The observed difference could also be attributed totheobserved differential loss to followup (Raymond 1979) which
could bias the measure of effect towards or away from the null
The Osga 1980 study found a protective effect of rider training
in the univariate analysis but statistically significant higher crash
rates in the intervention group when accounting for riding expo-
sure Sex (a significantly higher proportion of females in the in-
tervention group) and other potentially confounding factors that
could explain thedifferencebetween the groups were howeverun-
accounted for Finally Savolainen 2007 did control for potential
confounding variables yet it is not clear whether variables were
controlled for simultaneously or one at a time One case-control
study (MAIDS 2004) found no statistically significant reduction
in the odds of experiencing a crash between trained and untrainedriders but the quality of this study was poor mainly because the
analysis did not account for any confounding factors
One cohort study (Satten 1980) assessed the effect of non-manda-
tory rider training on injurious crashes and found a lower percent-
age of injuries in the intervention group although a statistical test
was not conducted As mentioned above the Satten 1980 studyrsquos
validity is reduced due to its lack of adjustment for confounding
factors
The only non-randomised trial that assessed offence rates (Leung
1987) found a statistically significant lower offence rate in the in-
tervention group and of the five cohort studies that assessed of-
fence rates (Satten 1980 Jonah 1982 Mortimer 1984 Mortimer
21Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2450
1988) one study (Jonah 1982) also found a statistically significant
lower offence rate in the intervention group However we have lit-tle confidence in the results from Leung 1987 since potential con-
founding factors were not appropriately addressed Jonah 1982
did conduct a multivariate analysis but was hindered by an initial
low response rate The other four cohort studies did control for
several potential confounders and found no difference in offence
rates following non-mandatory post-licence rider training
The non-randomised trials and cohort studies evaluating non-
mandatory rider training were at high risk of substantial selection
bias due to the non-random allocation of participants as well
as low participation rates and attrition bias and could therefore
not provide conclusive evidence with regard to the effect of the
training It should be noted that despite the biases and other issues
with this group of studies most didnot detectmajor effects relatedto rider training in either direction This may suggest that any
potential impact of rider training will be modest
Studies examining unspecified modes of ridertraining
Three case-control studies assessed whether injured or deceased
motorcyclists were more or less likely to have had rider training
compared to control motorcyclists All had conflicting results
Hurt1981 foundthe odds of havingan injuriouscrash waslower in
those who had received rider training whereas Kraus 1975 found
that the odds of having an injurious crash were higher in youngerand older trained riders Haworth 1997 who compared advanced
versus beginner training found no effect on motorcycle injuries
or deaths although they did find that those who had completed
advanced training had reduced chances of crashing compared to
those who had done a beginnersrsquo training course Similar issues
with uncontrolled confounding and biases of selection as described
above arose with the three case-control studies Hurt 1981 suffers
from inadequate control of confounding via poor matching and
both the Kraus 1975 and the Haworth 1997 findings were only
controlled for limited factors by means of stratified analysis It is
likely that these limitations are the reason for the contradictory
directions of effect observed for the three case-control studies
Methodological weaknesses of the studies
In summary most studies investigating the effectiveness of mo-
torcycle rider training on reducing crashes injuries and death suf-
fered from serious methodological weaknesses
The majority of studies were non-randomised and were there-
fore limited mainly by selection bias Although the approaches of
matching stratification or regression analyses can minimise bias
(Fleiss 2004) they were poorly employed in the reviewed studies
The inadequate adjustment for confounders in particular adds
to our lack of confidence in the studiesrsquo results since it has been
demonstrated that individuals who choose to take motorcycle rider
training differ from those not taking rider training For instance
trained riders often differ from controls in age sex and experi-ence usually ride less and may have a more positive attitude to
safety as indicated by the reported more frequent wearing of seat
belts (Mortimer 1982)
The two large randomised trials reported that a large proportion
of their study group were never exposed to the risk of offences or
crashes because they never attained licensure This made it impos-
sible to assess whether a reduction in crashes in the intervention
group was a result of training or due to delayed or never attained
licensure
In some studies the sample sizes were too small to detect any sta-
tistically significant differences Follow-up time after training dif-
fered between studies and was an impediment to observing any
real effects of training if they existed As riding experience ac-cumulates over time any differences in outcomes between those
trained and untrained would diminish in studies with lengthy fol-
low-up times This dilution of any possible effect of rider training
may have also occurred in studies that mixed both novice and ex-
perienced riders in the study groups since it is hypothesised that
novice riders serve to benefit most from rider training This would
increase the difficulty in finding any effect of rider training par-
ticularly when sample sizes are small
Detection bias may also account for some of the differences ob-
served between trained and untrained riders Detection bias would
have occurred if the self-report hospital driver or police records
used to record the outcomes of crash injury death or offences
differed between the trained and untrained riders Research hasdemonstrated that not all crashes are reported to police for ex-
ample small crashes or off-road crashes (Kraus 1975) even serious
injury crashes have been found to be underreported (Anderson
1980) However this would only introduce detection bias if the
reporting of outcome differs for trained and untrained riders This
does appear possible in that responsible riders may be more likely
to enroll in a rider training course and also may be more likely
to report crashes to police than irresponsible riders meaning that
the number of crashes from riskier riders could be underrepre-
sented on records An analysis to investigate the effectiveness of
rider training may then show that rider training is not effective in
reducing crashes
Rather than using reported outcomes some studies used self-re-ported survey data although this technique suffers from non-
response biases In fact two studies reported differences in de-
mographic characteristics in response rates to questionnaires
(Anderson 1980 Hall 1988) Self-reported data also introduce
the risk of recall bias survival bias and social desirability response
bias Recall bias may have been equally distributed among trained
and untrained riders howeverit is hypothesised that trained riders
may have felt it was lsquoexpectedrsquo to report good driving histories
hence leading the rider training to appear effective
22Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2550
Limitations of this review
Limitations of this review were that we found very few recentstudies (the two most recent ones published in 2002 and 2007)
and that we found very few studies conducted in low-and middle
income countries even though our search strategy was set up to
identify such studies Anotherlimitation wasthat forthe reviewed
studies we could not assess all outcomes of rider training that are
potentially useful Some of the included studies assessed outcomes
such as use of safety clothing including helmets Such outcomes
are valuable since the use of motorcycle helmets has been demon-
strated to be effective in preventing head injuries (Liu 2004) De-
spite this the use of safety clothing was not included as an out-
come in this review because the studies measuring this outcome
were limited by poor data quality due to very low response rates
Another outcome of interest in assessing the effectiveness of rider
training is lsquotype of injuryrsquo As Haworth 2005 state in their review
of motorcycle licensing and training rider training could be con-
sidered effective if injuries become less severe because motorcycle
riders have learned how to avoid more serious injuries through
rider training The studies included in this review did not specify
type of injury and so this outcome could not be assessed Last
changes in knowledge and attitudes following rider training are
other potentially useful outcomes We have excluded such studies
(Osga 1979 Clayton 1990 Goldenbeld 2004) as these outcomes
are proxies rather than direct measures of the effectiveness of mo-
torcycle rider training We opted to report on the outcomes of
crashes injury death and offences alone since there is little ev-
idence about how changes in knowledge and attitudes relate to
these outcomes
A U T H O R S rsquo C O N C L U S I O N S
Implications for practice
Due to the poor quality of studies included in this review we were
unable to draw any conclusions about the effectiveness of rider
training on crash rates injurious or fatal or offence rates The
findings do suggest that mandatory pre-licence training may be
an impediment to completing a motorcycle licensing process thuspossibly indirectly reducing crash injury death and offence rates
through a reduction in exposure to riding a motorcycle On the
basis of the existing evidence it is not clear if and what type of
training reduces the risk of crashes injuries deaths or offences in
motorcyclists and the selection of the best rider training practice
can therefore not be recommended The suggestion that manda-
tory pre-licence training indirectly reduces exposure to riding the
motorcycle because trainees are impeded from completing the li-
censing process does not help those motorcyclists who persist in
obtaining their licence and riding the motorcycle Some type of
rider training is likely necessary to teach motorcyclists to ride a
motorcycle safely Good policy is based on sound evidence and
evaluation but such evidence for motorcycle rider training is seri-
ously lacking Policy makers are therefore encouraged to include arigorous evaluation component to any new or existing rider train-
ing programme Such an evaluation should ideally be randomized
and assess whether the training programme is effective and if so
which component(s) of the programme work best
Implications for research
Given theresults of this review which found littlereliable evidence
about the effectiveness of motorcycle rider training it is necessary
to build up an evidence base of the effectiveness of motorcycle
rider training in the reduction of crash injury death and offence
rates To this end well designed randomised controlled trials need
to be conducted that examine the impact of rider training oncrashes and offences A randomised controlled trial is the preferred
method as it would significantly reduce most biases described in
this review To minimise the impact of participants delaying or
never attaining their licence and therefore exposure as found in
the randomised trials in this review such a trial should include
riders who are already licensed and reliable measures of riding
exposure in a reasonably lengthy follow up period It is expected
that a randomised controlled trial of sufficient sample size would
observe equal balance on possible confounders by study arm how-
ever an assessment of potentially important confounders (such as
helmet wearing) by study group would enable control for these
variables in the analyses if necessary It is imperative that the effec-
tiveness of each component of the rider training (such as theorypractice knowledge skills and modification of behaviour) and the
synergy of such components are evaluated Outcomes should be
objective ideally those that are routinely collected rather than self-
reported outcomes Even though riding the motorcycle poses dif-
ferent challenges and incurs different risks lessons may be learned
from research into the effects of car driver training
Motorcycles are a common means of transport in low- and middle
income countries and usage is increasing As the burden of mo-
torcycle crash victims is mostly borne in such countries research
should be focused on the effectiveness of types of rider training on
the reduction of crashes injuries and fatalities in such countries
and be supported by higher income countries This review found
only one study conducted in a middle income country that sug-gested decreasing injury rates It would be beneficial to conduct
additional research to confirm this suggestion and to enhance our
knowledge of the role of rider training in low- and middle-income
settings
March 2006 saw the launch of the University of North Carolina
Highway Safety Research Centre partnership with the Motorcycle
Safety Foundation (MSF) and plans were revealed to conduct a
study of the potential benefits of motorcycle rider training The
study is funded by the MSF and the National Highway Traffic Sa-
fety Administration (NHTSA) Called rsquoThe Discovery Projectrsquo it
will compare data on crash rates from riders participating in a basic
23Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2650
motorcycle safety class with riders participating in a basic motor-
cycle safety class as well as a variety of continuing education classesThe collection of pilot data was due to start in March 2007 and
the data collection portion of the study is expected to take three
years (Highway Safety Research Center) It is vital that researchers
and policy makers collaborate to develop furtherhigh quality eval-
uations of motorcycle rider training programmes to help inform
development of new and effective training programmes
A C K N O W L E D G E M E N T S
Staff from The University of Sydney library for helping us locate
and retrieve full reports of many papers
R E F E R E N C E S
References to studies included in this review
Adams 1985 published data only
Adams A Collingwood V Job RFS Evaluation of the 1982 Australia
Post motorcycle rider training scheme NSW Australia Traffic
Accident Research Unit Traffic Authority NSW 1985 [ RN 85]
Anderson 1980 published data only
Anderson J Ford JL Peck RC Improved motorcyclist licensing and testing project Washington DC National Highway Traffic Safety
Administration (NHTSA) USA 19801ndash217 [
DOTndashHSndash805ndash598]
Billheimer 1998 published data only
Billheimer JW Evaluation of the California Motorcyclist Safety
Program Transportation Research Record 19981640100ndash9 [
TRB Report Number 980652]
Cooper 1988 published data only
Cooper PJ Rothe JP Motorcyclists who they are and why they do
what they do Transportation Research Record 1988116878ndash85
Davis 1997 published data only
David CF Evaluation of community traffic safety programs and
motorcycle operator training programs Connecticut Department of Transportation 199763
Hall 1988 published data only
Hall CJ Buchanan L Engle R Motorcycle rider education evaluationproject Washington DC USA National Highway Traffic Safety
Administration (NHTSA) US Department of Transportation
1988 [ DTNH 22ndash80ndashCndash05123]
Haworth 1997 published data only
Haworth N Smith R Brumen I Pronk N In Monash University
Accident Research Centre editor(s) Case-control study of motorcycle
crashes Victoria Australia Department of Transport and Regional
Development The Federal Office of Road Safety 19971ndash149 [
ISBN 0 642 25501 6 ]
Hurt 1981 published data only
Hurt H Ouellet J Thom D Motorcycle accident cause factors and
identification of countermeasures USA Department of
Transportation 1981 [ Technical Report DOT HS 806ndash862]
Jonah 1982 published data onlylowast Jonah BA Dawson N Bragg B Are formally trained motorcyclists
safer Accident Analysis amp Prevention 198214(4)247ndash55
Jonah BA Dawson NE Bragg BW Evaluation of the Canada
Safety Councilrsquos Motorcycle Training Program Proceedings of theInternational Motorcycle Safety Conference Washington DC
USA 198059ndash93
Kloeden 1994 published data only
Kloeden CN Moore VM McLean AJ Evaluation of the pre-license
training program for motorcyclists in South Australia Walkerville
South Australia Australia NHMRC Road Accident Research Unit
The University of Adelaide 19941ndash37 [ ISBN 0 7308 0607 3]
Kraus 1975 published data only
Kraus JF Riggins RS Franti CE Some epidemiological features of
motorcycle collision injuries American Journal of Epidemiology 1975102(1)74ndash98
Leung 1987 published data only
Leung KS Reding VA Evaluation of the Wisconsin motorcycle rider course Wisconsin Bureau of Policy Planning and Analysis
Wisconsin Department of Transportation 1987
MAIDS 2004 published and unpublished data
MAIDS In-depth investigations of accidents involving powered
two wheelers ACEM (Acemacemeu) 2004 Vol Final report
20179
McDavid 1989 published data only
McDavid JC Lohrmann BA Lohrmann G Does motorcycle
training reduce accidents Evidence from a longitudinal quasi-
experimental study Journal of Safety Research 198920(2)61ndash72
Mortimer 1984 published data only
Mortimer R Evaluation of the motorcycle rider course Accident
Analysis amp Prevention 198416(1)63ndash71
24Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2750
Mortimer 1988 published data only
Mortimer RG A further evaluation of the motorcycle rider courseJournal of Safety Research 198819(4)187ndash96
Osga 1980 published data only
Osga GA An investigation of the riding experiences of MSF rider
course participants Dissertation South Dakota USA The
University of South Dakota 19801ndash224
Perrino 2002 published data only
Perrino CS Ahmed A Callender A Rozier E Cantwell A Stewart
O The role of Marylandrsquos motorcycle rider course in promoting saver
behaviors and attitudes Baltimore USA National Transportation
Center Morgan State University 20021ndash52
Raymond 1979 published data only
Raymond S Tatum S An evaluation of the effectiveness of the RAC
ACU motorcycle training scheme - final report Road Safety ResearchUnit Dept of Civil Engineering University of Salford London
19791ndash131
Satten 1980 published data only
Satten RS Analysis and evaluation of the motorcycle rider courses
in thirteen Illinois counties International Motorcycle Safety
Conference Washington DC 1980145ndash93
Savolainen 2007 published and unpublished data
Savolainen Peter Mannering Fred Additional evidence on the
effectiveness of motorcycle training and motorcyclistsrsquo risk-taking
behaviour Transportation Research Board 2007 Vol March 20
19
Swaddiwudhipong 1998 published data only
Swaddiwudhipong W Boonmak C Nguntra P Mahasakpan P
Effect of motorcycle rider education on changes in risk behavioursand motorcycle-related injuries in rural Thailand Tropical Medicine
and International Health 19983(10)767ndash70
Waller 1992 published data only
Waller FL An analysis of the impact of the Texas Department of
Public Safety Motorcycle Operator Training Program upon
motocycle related traffic accidents Texas State University Public
Administration Program Applied Research Projects 199277
References to studies excluded from this review
Anderson 1980a published data only
Anderson LC Better biking a motorcycle rider improvement
program International Motorcycle Safety Conference Washington
DC USA Motorcycle Safety Foundation 1980231ndash53Aupetit 2008 published and unpublished data
Aupetit S Riff J Buttelli O Espie S Ergonomics research program
on motorcycle learning Contributions to the development of a
motorcycle simulator for training Transport Research Arena
Ljubljana 20089
Awane 1999 published data only
Awane T Integrating simulators in motorcycle safety education
Journal of International Association of Traffic and Safety Sciences 199923(1)26ndash35
Baldi 2005 published data only
Baldi S Baer JD Cook AL Identifying best practices states in
motorcycle rider education and licensing Journal of Safety Research200536(1)19ndash32
Batchler 1988 published data only
Batchler MW Evaluation of the pre-license motorcycle training course at the New South Wales traffic education centre Federal Office of
Road Safety (FORS) 19881ndash122 [ Report No CR 77 ]
Blanchard 2006 published data only
Blanchard HT Tabloski PA Motorcycle safety educating riders at
the teachable moment Mosby (CV) Company 200632330ndash2
Bonnett 2005 published data only
Bonnett J Mature age skills training for experienced motorcycle
riders Motorcycle ridersrsquo association of the Australian Capital
Territory 2005
Braver 2007 published data only
Braver ER et alPersuasion and licensure a randomized controlled
intervention trial to increase licensure rates among Maryland
motorcycle owners Traffic Injury Prevention 20078(1)39ndash46Clayton 1980 published data only
Clayton AB Sudlow DE An evaluation of the effectiveness of a one day training programme for learner motorcyclists Berkshire UK
Transport and Road Research Laboratory 19801ndash23 [ ISSN
0266ndash7045]lowast Clayton AB Sudlow DE An evaluation of the effectiveness of a one-day training programme for learner motorcyclists Berkshire UK
Road Safety Division Berkshire 19801ndash23 [ISSN 0266ndash7045]
Council 1978 published data only
Council FM Desper LP Dutt AK Driver education for motorcycle operation University of North Carolina Highway Safety Research
Center 19781ndash40
Crowther 2005 published data onlylowast Crowther G Engaging with motorcyclists UK police and theBikeSafe Road Safety Programme Published proceedings of the
Australasian Road Safety Research Policing Education Conference
Wellington New Zealand 2005 issue ISBN 0ndash473ndash10636ndash11ndash11
Daniello 2009 published data only
Daniello A Gabler HC Mehta YA Effectiveness of motorcycle
training and licensing Transportation Research Board 88th Annual
Meeting 200913
Dewanti 2007 published data only
Dewanti M Challenging efforts in promoting young motorcyclist
safety in Indonesia Road Safety on Four Continents 14th
International Conference Swedish National Road and Transport
Research Institute 2007
Du Plooy 1990 published data onlyDu Plooy DW Motorcycle training in South Africa The Human
Element International Motorcycle Safety Conference Orlando
Florida USA Motorcycle Safety Foundation 19904-1 to 4-9
Engel 1989 published data only
Engel U Krogsgaringrd Thomsen L [Faeligrdselsundervisning og
propaganda ndash en faeligrdselssikkerhedsfremmende foranstaltning]
Dansk Vejtidsskrift 198912295ndash7
Goldenbeld 2002 published data only
Goldenbeld C Houwing S de Craen S The development of
driving skills of young moped riders [De ontwikkeling van de
rijvaardigheid van jonge bromfietsers] SWOV Institute for Road
Safety Research (Stichting Wetenschappelijk Onderzoek
Verkeersveiligheid) Leidschendam the Netherlands 200263
25Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2850
Goldenbeld 2004 published data only
Goldenbeld C Twisk D de Craen S Short and long term effects of moped training a field experiment Transportation Research Part F 20047(1)1ndash16
Haworth 2000 published data only
Haworth N Smith R Kowadlo N Motorcycle training in Victoria
evaluation of rider training curriculum in Victoria Monash
University Accident Research Centre Reports 2000 Vol 165 issue
3131
Haworth 2005 published data only
Haworth N Mulvihill C Review of motorcycle licensing and
training Monash University Accident Research Centre 2005 [
Report no240]
Haworth 2006 published data only
Haworth N Mulvihill C Rowdon P Teaching old dogs new tricksTraining and older motorcyclists Journal of the Australasian College of Road Safety 200718(4)20ndash5lowast Haworth NL Mulvihill C Rowden PJ Teaching old dogs new
tricks Training and older motorcyclists Australasian Road Safety
Research Policing and Education Conference Gold Coast
Queensland 2006
Hill 1983 published data only
Hill D Motorcycle rider training testing and licensing Motor
Cycle Council of New South Wales Sydney 1983
Imatake 1980 published data only
Imatake Y On the activities in Japan of motorcycle safe riding
promotion committee (MSRPC) - especially the education of
unskilled riders of class 1 motorbikes International Motorcycle
Safety Conference Washington DC USA Motorcycle Safety
Foundation 1980255ndash68
Ingebritsen 1990 published and unpublished data
Ingebrigtsen S [Risikofaktorer ved ferdsel med moped og
motorsykkel] TOslashI-rapport 66 Oslo Transportoslashkonomisk
institutt 1990
Jain 2009 published and unpublished data
Jain A et alTwo wheeler accidents on Indian roads a study from
Mangalore India Elsevier 200916130ndash3
Jonah 1979 published data only
Jonah BA Dawson NE Validation of the motorcycle operator skill
test Accident Analysis amp Prevention 197911(3)163ndash71
Jonah 1981 published data onlyJonah BA Dawson NE Bragg WE Predicting accident
involvement with the Motorcycle Operator Skill Test Accident Analysis amp Prevention 198113(4)307ndash418
Kadar 1990 published data only
Kadar KJ Countermeasures motorcycle rider education delivery
system The Human Element International Motorcycle Safety
Conference Motorcycle Safety Foundation 19906-27 to 6-41
Kelsey 1986 published data only
Kelsey SL Liddicoat C Ratz M Licensing novice motorcyclists a
comparison of Californiarsquos standard test and the MOST II (Motorcycle
Operator Skill Test) administered at centralized testing offices Sacramento California USA California Department of Motor
Vehicles Research and Development Office 19861ndash74
Koch 1980 published data only
Koch H Advanced course for motorcycle riders InternationalMotorcycle Safety Conference Washington DC USA Motorcycle
Safety Foundation 1980269ndash81
Koch 1990 published data only
Koch H Brendicke R How to influence risk taking behaviour
within motorcycle rider training programs The Human Element
International Motorcycle Safety Conference Orlando Florida
USA Motorcycle Safety Foundation 1990Motorcycle Safety
Foundation 1990 p 9-1 to 9-25A
Labbett 2005 published data only
Labbett S Langham M Training can make the problem worse
Road Safety Congress 20051ndash12
Law 2005 published and unpublished data
Law TH Radin Umar RS Zulkaurnain S Kulanthayan S Impactof the effect of economic crisis and the targeted motorcycle safety
programme on motorcycle-related accidents injuries and fatalities
in Malaysia International Journal of Injury Control and Safety
Promotion 200512(1)9ndash21
Lowes 1990 published data only
Lowes BF In-traffic rider training The Human Element
International Motorcycle Safety Conference Orlando Florida
USA Motorcycle Safety Foundation 19907-46 to 7-57
Mahon 1976 published data only
Mahon J Teaching accident avoidance on the motorcycle Journal of Traffic Safety Education 1976231ndash29
Maxwell 1985 published data only
Maxwell GS Evaluation of a Program for Training MotorcycleRiders Evaluation and Program Planning 19858(4)339ndash47
Mayhew 1996 published data only
Mayhew DR Simpson SH Effectiveness and role of driver educationand training in a graduated licensing system Ottowa Canada The
Traffic Injury Research Foundation 19961ndash89 [ ISBN
0ndash920071ndash12ndash0]
McKnight 1987 published data only
McKnight AJ Milburn BG An evaluation of the modified
motorcycle operator skill test exercises NHTSA Report 1987 [
Report No DOT HS 807 212 ]
McPherson 1976 published data only
McPherson K McKnight AJ The development and evaluation of a
motorcycle skill test manual and knowledge test Washington DCUSA National Highway Safety Administration 1976
Mohan 2004 published data only
Mohan D Evidence-based interventions for road traffic injuries in
South Asia Journal of the College of Physicians and SurgeonsPakistan 200414(12)746ndash7
Morgan 1980 published data only
Morgan JK Advanced motorcycle training course content and
study evaluation International Motorcycle Safety Conference
Washington DC USA Motorcycle Safety Foundation 1980
283ndash312
Motoki 1990 published data only
Motoki M Yamazaki S A study on effective motorcycle rider
education The Human Element International Motorcycle Safety
26Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2950
Conference Orlando Florida USA Motorcycle Safety
Foundation 19909-26 to 9-57A
Nairn 1992 published data onlylowast Nairn RJ Motorcycle safety research literature review 1987 to
1991 Federal Office of Road Safety Australia 1993 Vol Report
CR 117
Newland 1980 published data only
Newland R Darby J Automotive practicespre motorcycle rider
program International Motorcycle Safety Conference Washington
DC USA Motorcycle Safety Foundation 1980127ndash9
Newland 1982 published data only
Newland R Motorcycle rider training and licensing requirements
Driver training steering a course for the future Melbourne
Victoria Transport Regulation Board 1982
Newland 1990 published data only
Newland R Motorcycle training and licensing in Victoria 1980-
1990 The Human Element International Motorcycle Safety
Conference Orlando Florida USA Motorcycle Safety
Foundation 1990
Ormston 2003 published data only
Ormston R Dudleston A Pearson S Stradling Evaluation of
Bikesafe Scotland Scottish Executive Social Research Edinburgh
2003
Osga 1979 published data only
Osga GS Ellingstad VS An evaluation of the effectiveness of the 1978
South Dakota motorcycle rider course program South Dakota USA
University of South Dakota 19791ndash89
Panou 2007 published data only
Panou M Bekiaris E A new concept on the integration of driving
simulators in driving training the train-all approach Road Safety
on Four Continents 14th International Conference Swedish
National Road and Transport Research Institute 2007
Peterson 1990 published data only
Peterson A Kadar K Status report 1980-1989 Minnesota rider
education statistics The Human Element International
Motorcycle Safety Conference Orlando Florida USA 199013-
113 to 13-129
Prem 1984 published data only
Prem H Good MC Motorcycle rider skills assessment ACT
Australia Office of Road Safety Department of Transport 1984
1ndash377 [ Report No CR 34 ]
Radin Umar 2006 published data only
Radin Umar RS Motorcycle safety programmes in Malaysia how
effective are they International Journal of Injury Control and Safety Promotion 200613(2)71ndash9
Rockwell 1990 published data only
Rockwell TH Kiger S Carnot M An evaluation of the Ohio
motorcyclist enrichment program Ohio Department of Highway
Safety 1990
Rothe 1987 published data only
Rothe JP A motorcycle riding course Motorcyclists Image and reality Canada Insurance Corporation of British Columbia 1987
59ndash83
Rowden 2007 published data only
Rowden P Watson B Haworth N What can riders tell us aboutmotorcycle training A view from the other side of the fence
Australasian Road Safety Research Policing and Education
Conference Melbourne Australia Melbourne 2007
RSC Victoria 1993 published data only
Road Safety Committee Inquiry into Motorcycle Safety in Victoria
Parliament of Victoria 1993
Russam 1979 published data only
Russam K Motorcycle training and licensing in Japan UK Transport
and Road Research Laboratory (TRRL) 1979 [ Report 916 ]
Saffron 1981 published data only
Saffron DG Driver and rider education training and licensing a
brief review Traffic Accident Research Unit Traffic Authority of
New South Wales NSW Australia Traffic Accident Research UnitTraffic Authority of New South Wales 1981
Schulz 1990 published data only
Schulz U Motorcycle accidents among learner riders Motorcycle
licensing Evaluation of the effectiveness of a Graduated Licensing Scheme Forshungshefte des Institut fur Zweiradsicherheit 1990
Sexton 2004 published data only (unpublished sought but not used)
Sexton B Baughan C Elliott M Maycock G The accident risk of
motorcyclists 2004TRL Report TRL 607(ISSN 09684107)58
Shephard 1986 published data only
Shepherd R Do Motorcycle Safety Foundation programs work
Driver 198619(9)16ndash8
Simpson 1990 published data onlySimpson HM Mayhew DR The promotion of motorcycle safety
training education and awareness Health Education Research
Theory and Practice 19905(2)257ndash64
Sokolov 1990 published data only
Sokolov O Safe motorcycle riding some aspects of motorcycle
riding in the USSR The Human Element International
Motorcycle Safety Conference Motorcycle Safety Foundation
19904ndash37 - 4-48
Steffens 1988 published data only
Steffens U Gawatz R Willmes G Effectiveness of mofa courses
[Wirksamkeit von Mofakursen] Unfall- und sicherheitsforschung
strasenverkehr 198867168
Sudlow 2003 published data only
Sudlow D Scoping study on motorcycle training Road Safety
Research Report London UK The British Institute of Traffic
Education Research Department for Transport 2003 Vol June
[ Report No 36]
Tasmania 1995 published data only
Mannering FL Grodsky LL Motorcycle rider training scheme
Traffic Education Research Centre Armidale Australia Driver
Training Australia Limited 1995 Vol April1ndash56
Tetard 1985 published data only
Tetard C Motorcycle ridersrsquo training to detect indices in an
intersection situation [Lrsquoentrainement a la prise drsquoinformation dans
lrsquoapprentissage de la conduite motocycliste] Le Travail Humain198548(2)147ndash59
27Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3050
Thackray 1980 published data only
Thackray RM Prescott JC Field test of the motorcycle ridercourse International Motorcycle Safety Conference Washington
DC USA Motorcycle Safety Foundation 1980195ndash7
Thompson 1994 published data only
Thompson M Evaluation of compulsory basic training for
motorcyclists Transport Research Laboratory Crowthorne
Berkshire UK 1994 [ Report 63]
Tobler 1990 published data only
Tobler U Individual driving instruction for motorcycle driving
learners the Swiss experiment of motorcycle driving school The
Human Element International Motorcycle Safety Conference
Orlando Florida USA Motorcycle Safety Foundation 1990
Transport 1989 published data only
Transport Communications Road Crash Statistics Australia Bureauof Transport and Communications Economics 198914ndash20
Tuan 2007 published data only
Tuan VA Shimizu T Towards development and evaluation of the
motorcycle drivers re-education program in Vietnam modeling of
motorcycle driverrsquos undesired behaviors 11th World Conference
on Transport Research 200725
Victor 1980 published data only
Victor DJ Traffic safety education for motorcyclists International
Motorcycle Safety Conference Washington DC USA Motorcycle
Safety Foundation 1980221ndash8
Waterer 1990 published data only
Waterer TH Compulsory training for motorcyclists from liberty
to compulsion The Human Element International Motorcycle
Safety Conference Orlando Florida USA Motorcycle Safety
Foundation 19904-49 to 4-57
Weller 1989 published data only
Weller RB Chandler EW Motorcycle safety and motorcycle
education past research and survey results Journal of Public Policy
amp Marketing 1989893ndash108
Wells 1981 published data only
Wells P Observations of some RACACU motorcycle training
schemes Transport and Road Research Laboratory Crowthorne
Berkshire UK 1981
Winn 1986 published data only
Winn G Two important trends in motorcycle safety regulations
rider education and conspicuity improvement American
Motorcyclist Association 1986147ndash52
Wisher 1988 published data only
Wisher G Reid J Motorcycle accidents in Norfolk a three-year study of motorcycle training and the Part 1 test Norfolk UK Road Safety
Division Highways Department Norfolk County Council 1988
Wood 1987 published data only
Wood HT Bowen R Evaluation of the revised learner permit scheme
July 1983 to December 1985 Victoria Australia Road Traffic
Authority 19871ndash27 [ ISBN 0 7306 0247 8]
Yamazaki 1990 published data only
Yamazaki S Nakanishi M Survey on effect of Honda rider
education The Human Element International Motorcycle Safety
Conference Orlando Florida USA 199012-167 to 12-187
References to ongoing studies
Burgess 2010 unpublished data only
Burgess C Broughton P Fylan F Stradling Interim evaluation of
the UKs National Ride program In L Dorn editor(s) Driver
behaviour and Training Vol IV Aldershot Ashgate Publishing
Limited 2010
Additional references
Anderson 1980
Anderson J Ford JL Peck RC Improved motorcyclist licensing and
testing project Volume I and II National Highway Traffic Safety
Administration US department of Transportation Washington
DC 1980
Bjornstig 1985
Bjornstig U Bylund P Lekander T Brorsson B Motorcycle fatalitiesin Sweden Acta Chirurgica Scandinavica 1985151(7)577ndash81
Clayton 1990
Clayton A Sudlow DE An evaluation of the effectiveness of a one-
day training program for learner motorcyclists The Human
Element International Motorcycle Safety Conference Orlando
Florida USA 199013-105 to 13-111
Cochrane Collaboration 2008
Higgins JP Green S Cochrane Handbook for Systematic Reviews of
Interventions 500 The Cochrane Collaboration 2008
Elliot 2003
Elliott MA Baughan CJT Broughton J Grayson G Knowles J et
alMotorcycle Safety a Scoping Study TRL [ Report No
0968ndash4107]Fleiss 2004
Fleiss JL Levin B Paik MC Statistical methods for rates and proportions 3rd Edition John Wiley amp Sons Inc 2004 [ ISBN
978ndash0ndash471ndash52629ndash2 ]
FORS 1999
FORS Federal Office of Road Safety Road risk for sober licensed
motorcyclists Vol Monograph 27 Federal Office of Road Safety
19991ndash2
Haworth 2005
Haworth N Mulvihill C Review of motorcycle licensing and training
Victoria Australia Monash University Accident Research Centre
20051ndash83 [ Report No 240]
Huwiler-Muumlntener 2002Huwiler-Muumlntener K Juumlni P Junker C Egger M Quality of
Reporting of Randomized Trials as a Measure of Methodologic
Quality JAMA 20022872801ndash4
Johnston 2008
Johnston P Brooks C Savage H Fatal and serious road crashes
involving motorcyclists Road Safety Research and Analysis Report
ATSB Monograph 20 Canberra Department of Infrastructure
Transport Regional Development and Local Government 0205
081ndash26
Juumlni 2001
Juumlni P Altman DG Egger M Systematic reviews in health care
Assessing the quality of controlled clinical trials BMJ 2001323
42ndash6
28Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3150
Kopits 2003
Kopits E Cropper M Traffic fatalities and economic growth TheWorld Bank Development Research Group Infrastructure and
Environment Policy research working paper 3035 2003
Leung 1983
Leung P Chau BTW Motorcycle accidents in Hong Kong pattern
of injury treatment causative factors and socio-economic
implications Injury 198315(1)23ndash9
Liu 2004
Liu B Ivers R Norton R Blows S Lo SK Helmets for preventing
injury in motorcycle riders Cochrane Database of Systematic Reviews 2004 Issue 2 [DOI 10100214651858CD004333pub3]
Mohan 2002
Mohan D Road safety in less-motorized environments future
concerns International Journal of Epidemiology 200231(3)527ndash32Mortimer 1982
Mortimer RG An evaluation of the motorcycle rider course
Proceedings of the 20th conference of the American Association for
Automotive Medicine Arlington Heights IL USA American
Association for Automotive Medicine 198277ndash96
Mullin 1997
Mullin BT Risk factors for motorcycle injury the role of age
gender experience training and alcohol PhD thesis Auckland
New Zealand University of Auckland 1997
Noordzij 2001
Noordzij PC Forke E Brendicke R Chinn BP Integration of needs
of moped and motorcycle riders into safety measures review and
statistical analysis in the framework of the European research
project PROMISING SWOV Institute for Road Safety Research
Leidschendam The Netherlands 2001 [ Report No Dndash2001ndash5]
Osga 1979
Osga G Ellingstad VS An evaluation of the effectiveness of the 1978 South Dakota Motorcycle Riders Course Program South Dakota
University of South Dakota Vermillion Human Factors Lab 1979
Oxford English Dictionary Online
httpoxforddictionariescom Oxford English Dictionary Online
Oxford University Press 2010
Peden 2004
Peden M Scurfield R Sleet D World Report on Road TrafficInjury Prevention Geneva World Health Organization Available
at httpwwwwhointviolence˙injury˙preventionpublications
road˙trafficworld˙reporten 2004Chapter 3
Radin Umar 1995
Radin Umar R Macakay G Hills B Preliminary analysis of
motorcycle accidents Short-term impacts of the running
headlights campaign and regulations in Malaysia J Traffic Med 199523(1)17ndash27
Tsai 1995
Tsai YJ Wang JD et alCase-control study of the effectiveness of
different types of helmets for the prevention of head injuries among
motorcycle riders in Taipei Taiwan American Journal of
Epidemiology 1995142(9)974ndash81
Vis 1995
Vis AA In-depth study of the hazards of motorcycling [De
onveiligheid van motorrijden nader bekeken] SWOV Institute for
Road Safety Research Leidschendam the Netherlands 1995 [
Report No Rndash95ndash69]
Weaver 1990
Weaver EA Motorcycle safety education in the United States
Motorcycle Safety Foundation Orlando Florida USA 1990
Winn 1987
Winn G Two important trends in motorcycle safety regulations
rider education and conspicuity improvement No more
information available Detroit MI US Society of Automotive
Engineers 1987147ndash52
Yuan 2000
Yuan W The effectiveness of the rsquoride-brightrsquo legislation for
motorcycles in Singapore Accident Analysis amp Prevention 200032
(4)559ndash63
Zambon 2006
Zambon F Hasselberg M Factors affecting the severity of injuries
among young motorcyclists a Swedish nationwide cohort study
Traffic Injury Prevention 20067(2)143ndash9lowast Indicates the major publication for the study
29Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3250
C H A R A C T E R I S T I C S O F S T U D I E S
Characteristics of included studies [ordered by study ID]
Adams 1985
Methods Non-randomised trial
Participants Initial sample size
Total of 162 Australia Post telegram delivery riders
Not clear how many in intervention and control groups
Analysis sample size
Intervention group 63Control group 35
Total 98 (61)
Interventions No formal training vs voluntary rider training course of 12 hours over 2 consecutive days (theory 4 hours practice
8 hours)
Intervention integrity
No information given
Outcomes Australia Post recorded crashes
Length of follow up 1 year
Notes Study conducted in NSW Australia
Anderson 1980
Methods Randomised trial
Participants Initial sample size
Total 41188 motorcycle licence applicants
Analysis sample size
Intervention group with skills training 13160
Intervention group without skills training 12634
Control group 15080
Total 40874 (99)
Interventions Standard licensing skill test vs the Motorcycle Operators Skill Test (MOST) with or without a training component
MOST is not to be confused with rider training because it is a licensing test The training component was for
participants who failed the skill test on the first attempt was of 3 hours duration and consisted of instructions about
the skills needed to pass the skill test The training consisted of 30 min theory and 2 frac12 hours practice on the
motorcycle
Intervention integrity
33 in MOST with skills training 31 in MOST without skills training and 43 in control group obtained their
motorcycle licence in the end
Outcomes Police recorded crashes and offences
Length of follow up 3 years
30Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3350
Anderson 1980 (Continued)
Notes Study conducted in California USA
Billheimer 1998
Methods Cohort study
Participants Initial sample size
Intervention group Unclear numbers of CMSP graduates and ERC graduates
Control group Over 16000 untrained motorcyclists
Analysis sample size
1139 matched CMSP-pairs and 1182 ERC-pairs
Interventions No formal training vs the mandatory (for those younger than 18 years) pre-licence 16-hours basic California
Motorcyclist Safety Program (CMSP) and the 8-hour Experienced Rider Course (ERC)
Intervention integrity
No information given
Outcomes Police recorded crashes and offences
Length of follow up 2 years
Notes Study conducted in California USA
Cooper 1988
Methods Cohort study
Participants Initial sample size
Intervention group 418 graduates and 57 who failed
Control group 402
Analysis sample size
Intervention group unclear
Control group unclear
Total 863 (984)
Interventions No formaltrainingvs non-mandatory pre-licence rider trainingThe content of thetrainingwas notfurther explainedIntervention integrity
Participants had all completed the rider training
Outcomes Police recorded crashes and offences
Length of follow up 7 years
Notes Study conducted in British Columbia Canada
31Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3450
Davis 1997
Methods Retrospective cohort study
Participants Initial sample size
Intervention group 9320 graduates
Control group unclear
Analysis sample size
Intervention group 9320 graduates
Control group unclear
Interventions No formal training vs the non-mandatory Connecticut Motorcycle Operator Training Program The content of
training was not further explained
Intervention integrity Participants had all completed the rider training
Outcomes Police recorded crashes severity of crashes
Length of follow up 1 year
Notes Study conducted in Connecticut USA
Hall 1988
Methods Randomised trial
Participants Initial sample sizeIntervention group A 6248
Intervention group B 6616
Intervention group C 6996
Control group 6604
Total 26464 motorcycle licence applicants
Analysis sample size
Intervention group A 6012 (96)
Intervention group B 6428 (97)
Intervention group C 6778 (97)
Control group 6502 (98)
Interventions Standard licensing skill test vs the Motorcycle Operator Skill Test (MOST II) without training (A) MOST II with
3 hours training (B) or MOST II with 20 hours training (C) The 3 hours training comprised instructions aboutthe skills needed to pass the skill test (30 min theory and 2 frac12 hours practice on the motorcycle) The 20 hours
training consisted of the Motorcycle Safety Foundationrsquos Motorcycle Rider Course (MRC) comprising of 8 hours
theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-road practice
Intervention integrity
25 and 26 obtained their motorcycle licence in the lsquoMOST 3-hour trainingrsquo and lsquoMOST 20-hour trainingrsquo
intervention arm respectively In the lsquoMOST testrsquo intervention arm 29 obtained their licence and in the control
group 35
Outcomes Police recorded crashes and offences
Length of follow up 3 years
32Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3550
Hall 1988 (Continued)
Notes Study conducted in New York USA
Haworth 1997
Methods Population based case-control study
Participants Cases Motorcycle riders or pillion riders involved in motorcycle crash (on road) Recruited from hospital and police
records Not clear whether all cases (especially fatalities) were included
Community controls Multiple matched sample of motorcycle rider or pillion who passed the crash site at the same
time of day and day of week that the crash occurred
Initial sample size
Cases 336
Controls 1195
Analysis sample size
Cases 234 (70)
Controls 561 (47)
Interventions Basic handling skills course beginner intermediate or advanced course measured through self-report not validated
Outcomes Injuries with an Injury Severity Score of 5 or greater or admission for at least four hours and fatalities from a
motorcycle crash (from hospital and police records)
Length of follow up two years or more prior to the interview
Notes Study conducted in Victoria Australia
Hurt 1981
Methods Population based case-control study
Participants Cases Motorcycle riders who had a motorcycle crash that was reported to police (traffic accident reports) Not all
cases included
Community controls Multiple matched sample of motorcycle riders who passed the crash site at the same time of
day and day of week as the crash occurred but 2 years later
Initial sample size
Cases 4500 notifications of crashes to police (only 50 received in time research team able to respond to 1126 for
unreported reasons data collection completed for 900)
Controls Number approached unreported 2310 consented and interviewed
Analysis sample size
Cases 808 (808900 data on rider training) - 18 of initial sample
Controls 670 (6702310 data on rider training) - lt 29 of initial sample
Interventions Training experience self taught friends-family school-club formal-professional others measured through self-
report validated
Outcomes Injuries and fatalities from a motorcycle crash (from hospital records and coronerrsquos office)
33Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3650
Hurt 1981 (Continued)
Notes Study conducted in Los Angeles USA
Jonah 1982
Methods Cohort study
Participants Initial sample size
Intervention group
2310 licensed MTP graduates who are active riders
Control group
2029 licensed motorcyclists who are active riders
Non-licensed and non-active riders excluded
Analysis sample size
Intervention group
811 interviews (35) 533 driver records (23)
Control group 1080 interviews (71) 905 driver records (45)
Interventions No training vs the non-mandatory Canada Safety Council Motorcycle Training Program (MTP) comprising of 4
hours theory and 16 hours practise (6 hrs on-road) over 2 weekends
Intervention integrity
Participants had all successfully completed the MTP
Outcomes Self-reported and police reported crashes and offencesLength of follow up 4 years
Notes Study conducted in Ontario Canada
Kloeden 1994
Methods Cohort study
Participants Initial sample size
Intervention group 2533 licence applicants
Control group Unclear
Sample of riders obtaining probationary licenceIntervention group 1625 (64)
Control group 5015 (unclear)
Non-active riders excluded from control group (~66)
Analysis sample size
Intervention group 631 (39)
Control group 318 (6)
Interventions No formal training vs the mandatory pre-licence Ridersafe course consisting of two levelsthe first is a training course
in the basics of motorcycle riding where a pass entitles the rider to obtain a learnerrsquos permit The second is a skill
test where a pass entitles the rider to obtain a probationary motorcycle licence The training course consists of two 4
hour training sessions on separate days involving instruction videos and actual riding on motorcycles The skill test
consists of a 4 hour preparation course of 74 minutes of lecturediscussion 38 minutes of video and 75 minutes of
34Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3750
Kloeden 1994 (Continued)
on-course riding which is the testing phase
Intervention integrity
Unclear what percentage in the Ridersafe group and the control group obtained their probationary licence halfway
and their full motorcycle licence in the end
Outcomes Police reported crashes
Length of follow up 1 year
Notes Study conducted in SA Australia
Kraus 1975
Methods Population based case-control study
Participants Cases All motorcycle riders with a confirmed medically treated motorcycle injury (including fatalities) from all
hospitals in study base and peripheral area hospitals and all deaths due to motorcycle crashes in the study base All
serious injuries (leading to hospitalisation) and fatalities likely to be included
Community controls A random sample of registered motorcycle owners in the same temporary and geographical
study base as the cases arose from
Initial sample size Cases 1273
Controls 738
Analysis sample size Cases 741 (58)
Controls 434 (59)
Interventions Motorcycle rider training (not specified) measured through self-report
Outcomes Injuries leading to hospitalisation fatalities
Length of follow up not specified
Notes Study conducted in Sacramento county USA
Leung 1987
Methods Non-randomised trial
Participants Initial sample sizeIntervention group unknown number of MRC graduates
Control group unknown number of licensed motorcyclists
Analysis sample size
Intervention group 2941
Control group 43094
Interventions No formal training vs non-mandatory Wisconsin motorcycle rider course a 22 hours off-street programme that
teaches the basic rules of safe motorcycling
Intervention integrity
50 in the MRC group and 100 in the control group obtained their motorcycle licence in the end
35Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3850
Leung 1987 (Continued)
Outcomes Police recorded crashes and offences
Length of follow up up to 3 years
Notes Study conducted in Wisconsin USA
MAIDS 2004
Methods Population based case-control study
Participants Cases Motorcycle riders who had a motorcycle crash Unclear whether all cases included
Community controls Motorcycle riders identified at petrol stations Unclear whether matched to cases
Initial sample size
Cases not reported
Controls not reported
Analysis sample size
Cases 921 (Italy 200 Netherlands 200 Spain 121 Germany 250 France 150)
Controls 923 (Italy 200 Netherlands 200 Spain123 Germany 250 France 150)
Interventions Pre-license rider training or additional training (non-compulsory)
Outcomes Crashes
Notes Study in five sampling areas located in five European countries (France Germany Netherlands Spain and Italy)
McDavid 1989
Methods Cohort study
Participants Sample size after first match
Intervention group 346 male training graduates
Control group 346 male motorcycle riders licensed in same year
Sample size after second match
Intervention group 139 (40)
Control group 139 (40)
Analysis sample size
Intervention group 139 (40)
Control group 139 (40)
Interventions No formal training vs the non-mandatory pre-licence British Columbia Safety Councilrsquos 37-hour motorcycle safety
programme
Intervention integrity
Participants had all completed and passed the training
Outcomes Police reported crashes injurious crashes
Length of follow up 1-6 years
36Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3950
McDavid 1989 (Continued)
Notes Study conducted in British Columbia Canada
Mortimer 1984
Methods Cohort study
Participants Initial sample size
Intervention group 608 MRC graduates
Control group 360 motorcycle riders
Response rates
Intervention group 360 (59)
Control group 324 (90)
Non-active riders excluded
Analysis sample size
Intervention group 213 (35)
Control group 303 (84)
Interventions No formal training vs the non-mandatory Motorcycle Safety Foundationrsquos Motorcycle Rider Course comprising of
8 hours theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-road practice
Intervention integrity
Participants had all completed and passed the MRC
Outcomes Self-reported crashes and offencesLength of follow up 1 year
Notes Study conducted in Illinois USA
Mortimer 1988
Methods Cohort study
Participants Initial sample size
Intervention group 2307 MRC graduates
Control group 556 active licensed motorcycle riders
Response ratesIntervention group 1298 (56)
Control group 500 (90)
Non-active riders excluded
Analysis sample size
Intervention group 913 (40)
Control group 500 (90)
Interventions No formal training vs the non-mandatory Motorcycle Safety Foundationrsquos Motorcycle Rider Course comprising of
8 hours theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-road practice
Intervention integrity
Participants had all completed and passed the MRC
37Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4050
Mortimer 1988 (Continued)
Outcomes Self reported crashes and offences
Length of follow up 1 year
Notes Study conducted in Illinois USA
Osga 1980
Methods Cohort study
Participants Initial sample sizeIntervention group 1163 MSF MRC participants invited 818 (74) responded
Control group 329 motorcycle riders responded to recruitment strategies
Non-active riders excluded from analysis
Analysis sample size
Intervention group 595 (73)
Control group 329 (100)
Interventions No formal training vs the non-mandatory Motorcycle Safety Foundationrsquos Motorcycle Rider Course (MRC) com-
prising of 8 hours theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-road
practice
Intervention integrity
Participants had all successfully completed the MRC
Outcomes Self-reported crashes police recorded crash reports
Length of follow up 6 months - 3 12 years
Notes Study conducted in South Dakota USA
Perrino 2002
Methods Cohort study
Participants Initial sample size
Intervention group 2144 (unclear) MRC graduates (motorcycle licence applicants)
Control group 2601 (unclear) regular licensing test graduates (motorcycle licence applicants)Analysis sample size
Intervention group 1070 (unclear)
Control group 867 (unclear)
Interventions No formal training vs the mandatory (for those younger than 18 years) pre-licence Maryland Motorcycle Rider
Course (MRC) consisting of 8 hours theory and 12 hours practice
Intervention integrity
No information given
Outcomes Self-reported and police recorded crashes and offences
Length of follow up 1 year
38Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4150
Perrino 2002 (Continued)
Notes Study conducted in Maryland USA
Raymond 1979
Methods Cohort study
Participants Initial sample size
Intervention group
258 volunteers mainly (but not all) novice riders
Control group
1104 provisional licence applicants
Analysis sample size
Intervention group 190 (74)
Control group 627 (57)
Interventions No formal training vs the non-mandatory RACACU (Royal Automobile ClubAuto-Cycle Union) training scheme
The training program consisted of either 24 one-hour sessions or 13 two-hour sessions Twelve hours of the training
focused on theory about basic operational procedures traffic laws and highway code The practical session focused
on basic operational skills riding on a training area and riding in traffic with an instructor A proficiency test was
administered at the end of the course
Intervention integrity
No information given
Outcomes Self-reported crashes (including minor non-injury non-damage and off-road crashes)
Length of follow up 1 year
Notes Study conducted in the UK
Satten 1980
Methods Cohort study
Participants Initial sample size
Intervention group interviewers kept interviewing until they had 100 riders selected from a population of 891
enrollees in the MSF MRC course (unknown how many were asked and how many refused to participate)Control group unknown number of registered motorcycle riders
Non-active riders excluded from analysis
Analysis sample size
Intervention group 71
Control group 100
Interventions No formal training vs the non-mandatory Motorcycle Safety Foundationrsquos (MSF) Motorcycle Rider Course (MRC)
comprising of 8 hours theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-
road practice
Intervention integrity
Participants had all successfully completed the MRC
39Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4250
Satten 1980 (Continued)
Outcomes Self-reported crashes injuries and offences
Length of follow up 3 years
Notes Study conducted in Illinois USA
Savolainen 2007
Methods Cohort study
Participants Initial sample size
Intervention group 4000 riders that had conducted the MSF MRC course (passed or failed)
Control group 4000 motorcycle owners or permit or endorsement holders from the Indiana Bureau of Motor
Vehicles database
Analysis sample size
Intervention group 739 (18)
Control group 588 (15)
Interventions Intervention integrity
No information given
Outcomes Probability of crash involvement
Length of follow up up to 5 years
Notes Study conducted in Indiana USA
Swaddiwudhipong 1998
Methods Randomised trial
Participants Initial sample size
Intervention group
all 16 villages of 3 sub-districts
(participants were practising motorcyclists licensed and unlicensed)
Control group
18 villages of 3 other sub-districtsAnalysis sample size
Intervention group 3 districts (100)
Control group 3 districts (100)
Interventions No rider education vs non-mandatory community based rider education consisting of theory on motorcycle injury
risk helmet use traffic laws vehicle regulations and traffic signs and a written and skill test leading to a licence
Intervention integrity
No information given
Outcomes Injuries fatalities
Length of follow up 2 years
40Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4350
Swaddiwudhipong 1998 (Continued)
Notes Study conducted in Tak Province rural Thailand
Waller 1992
Methods Cohort study
Participants Initial sample size
Intervention group 349 riders who succesfully completed the Department of Public Safety motorcycle operator
training in 1988
Control group 349 motorcyclists who never completed the DPS motorcycle operator training
Analysis sample size
Intervention group 349
Control group 349
Interventions No DPS motorcycle operator training vs non-mandatory DPS motorcycle operator training The DPS training is
the Motorcycle Safety Foundationrsquos Motorcycle Rider Course comprising eight hours of theory (lectures discussions
films and multiple choice tests) a textbook and 12 hours off-road practice
Intervention integrity
All 349 riders succesfully completed the rider training
Outcomes Police recorded crashes
Length of follow up 3 years
Notes Study conducted in
Characteristics of excluded studies [ordered by study ID]
Study Reason for exclusion
Anderson 1980a Study type review and process evaluation of a rider course
Aupetit 2008 Outcome proposal of a rider training curriculum that combines traditional training and a motorcycle riding
simulator
Awane 1999 Study type review on use of simulators in rider and driver training
Baldi 2005 Intervention not rider training but intervention integrity of rider training availabilty of rider education
implementation of graduated licensing system etc on a state-wide level
Batchler 1988 Study type description of a training course
Blanchard 2006 Study type not a study assessing the association between rider training and outcomes
Bonnett 2005 Outcome qualitative data regarding satisfaction with the training programme
41Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4450
(Continued)
Braver 2007 Outcome licensure rates
Clayton 1980 Outcome knowledge skills attitude
Council 1978 Outcome crash reduction in ANY vehicle not motorcycles exclusively
Crowther 2005 Study type qualitative study
Daniello 2009 Study type review of the effectiveness of motorcycle training and licensing
Dewanti 2007 Intervention emotional characteristics behaviour (ie not motocycle rider training)
Du Plooy 1990 Study type review of types of rider training courses in South Africa
Engel 1989 Study type uncontrolled before-after study
Goldenbeld 2002 Outcome skills
Goldenbeld 2004 Outcome knowledge and skills
Haworth 2000 Intervention balance between attitudinal andvehicle control skills acrosscurriculla of motorcycle rider training
courses
Haworth 2005 Study type review
Haworth 2006 Study type cross-sectional survey
Hill 1983 Study type review and description of training system
Imatake 1980 Study type course descriptionreview
Ingebritsen 1990 Intervention risk factors for motorcycle crashes
Jain 2009 Study type cross-sectional descriptive study of motorcycle crashes over a 5 year period
Jonah 1979 Study type cross-sectional survey Outcome scores on motorcycle operator skill test
Jonah 1981 Intervention evaluation of a motorcycle rider skill test not motorcycle rider training
Kadar 1990 Study type review of components of rider training courses
Kelsey 1986 Intervention evaluation of a motorcycle rider skill test not motorcycle rider training
Koch 1980 Study type reviewcourse description
Koch 1990 Study type description of course content
42Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4550
(Continued)
Labbett 2005 Study type essay about possible explanations for motorcycle deaths
Law 2005 Intervention targeted motorcycle safety programme not specifically rider training
Lowes 1990 Study type review and descriptions of rider training
Mahon 1976 Study type review of motorcycle rider training methods
Maxwell 1985 Outcomesrecommendations forprogramme operatorsand otherson howthe motorcycle trainingprogramme
works and how it can be improved
Mayhew 1996 Study type review
McKnight 1987 Study type evaluation of methodological and administrative aspects of a motorcycle operator skill test
McPherson 1976 Outcome knowledge and skills
Mohan 2004 Study type review of evidence of interventions to reduce road traffic injuries
Morgan 1980 Study type reviewcourse description
Motoki 1990 Study type proposal and development of rider training curriculum
Nairn 1992 Study type literature review
Newland 1980 Study type description of development of motorcycle safety course
Newland 1982 Study type descriptionreview of a rider training course
Newland 1990 Study type uncontrolled BeforeAfter study
Ormston 2003 Outcome changes in attitude and self reported behaviour
Osga 1979 Outcome knowledge attitudes and skills
Panou 2007 Intervention new simulation tool to teach driverrider
Peterson 1990 Study factor demographic details of course participants completing the course vs those not completing the
course
Prem 1984 Study type experimental
Study factor examines association between skills and performance in tests
Radin Umar 2006 Intervention motorcycle safety programme not specifically rider training
Rockwell 1990 Study type methodologically flawed study type
43Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4650
(Continued)
Rothe 1987 Study type qualitative study of rider training in general
Rowden 2007 Study type qualitative study
RSC Victoria 1993 Study type review of motorcycle rider training and education in Victoria
Russam 1979 Study type review of training programmes in Japan
Saffron 1981 Study type literature review
Schulz 1990 Study factor predictors of motorcycle crashes
Sexton 2004 Study type cross-sectional analytic analysis
Shephard 1986 Study factor reviewdescription of programme
Simpson 1990 Study type review of motorcycle rider trainingeducation
Sokolov 1990 Study factor review of motorcycle riding courses in USSR
Steffens 1988 Outcome self-reported behaviour experiences and researcher observed riding behaviour
Sudlow 2003 Study type review of motorcycle training in the UK
Tasmania 1995 Study type cross-sectional descriptiveprocess evaluation
Tetard 1985 Study type experimental
Outcome detection of indices
Thackray 1980 Outcome instructional effectiveness user acceptance administrative feasibility
Thompson 1994 Study type cross-sectional survey
Outcome attitudes and knowledge
Tobler 1990 Study type review of course curricula
Transport 1989 Intervention syllabus for rider training
Tuan 2007 Study type cross-sectional study (survey) which does not assess the effects of an intervention but examines
motocyclistsrsquo behaviour
Victor 1980 Study type review
Waterer 1990 Study type review of UK rider training issues
Weller 1989 Study type review of studies evaluating motorcycle safety and education measures
44Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4750
(Continued)
Wells 1981 Study type description of different methods of training
Winn 1986 Study type review of rsquohistoryrsquo of rider training
Wisher 1988 Study type cross-sectional survey
Wood 1987 Study type Interrupted Time Series but did not meet the criteria for an ITS as published by the Cochrane
Effective Practice and Organisation of Care Group
Yamazaki 1990 Study type cross-sectional survey
Outcome riding skills skills retention and attitude
Characteristics of ongoing studies [ordered by study ID]
Burgess 2010
Trial name or title UKs National Ride program
Methods Cohort study
Participants Motorcycle offenders
Interventions Police led diversion program
Outcomes Self reported attitudinal and behavioural measures
Starting date
Contact information
Notes Expected to be completed in early 2010
45Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4850
D A T A A N D A N A L Y S E S
This review has no analyses
A P P E N D I C E S
Appendix 1 Search strategy
We searched the following electronic databases
bull Cochrane Injuries Group Specialised Register (searched 22 Sept 2008)
bull CENTRAL (The Cochrane Library 2008 Issue 3)
bull TRANSPORT 1988 to 200706 (latest date available)
bull ATRI Australian Transport Index (to Sept 2008)
bull MEDLINE (Ovid SP) 1950 to Sept 2008
bull EMBASE (Ovid SP) 1980 to Sept 2008
bull CINAHL (EBSCO) 1982 to Sept 2008
bull WHOLIS (World Health Organization Library Information System) (to Sept 2008)
bull PsycINFO (Ovid SP) 1806 to Sept 2008
bull LILACS (BIREME) (Latin American and Caribbean Health Sciences) (1982 to Sept 2008)
bull ISI Web of Science Social Sciences Citation Index (SSCI) 1970 to Sept 2008
bull ERIC (searched 25 Sept 2008)
bull ZETOC (searched 25 Sept 2008)
bull SIGLE (searched 25 Sept 2008)
All of the databases were searched using the following text-word search strategy The strategy was adapted using either rsquorsquo or rsquo$rsquo as atruncation symbol
1 motor-cycl or motorcycl or motor-bik or motorbik or motor-scooter or motorscooter or scooter or moped or motocycl
or motocicl or moto
2 train or educat or instruct or program or course or qualif or learn or improv
3 1 and 2
H I S T O R Y
Protocol first published Issue 2 2005
Review first published Issue 10 2010
46Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4950
C O N T R I B U T I O N S O F A U T H O R S
Katina Kardamanidis was the first author of the protocol conducted the literature search extracted full studies assessed the studies
according to inclusion criteria assessed quality of the studies compared findings with the second author discussed disagreements with
the third author wrote the drafts and the final review
Rebecca Ivers was the second author of the protocol assessed the studies according to inclusion criteria assessed quality of the studies
compared findings with the first author edited the final draft and provided overall guidance for the process
Alexandra Martiniuk assessed quality of the studies compared findings with the first author discussed disagreements with the third
author and helped write the first draft and the final review
Mark Stevenson contributed to the protocol reviewed methods of studies and contributed to the final draft
Katrina Thistlethwaite conducted a second literature search extracted full studies and contributed to the final draft
D E C L A R A T I O N S O F I N T E R E S T
None known
S O U R C E S O F S U P P O R T
Internal sources
bull No sources of support supplied
External sources
bull National Health and Medical Research Council of Australia Australia
Salary support for Martiniuk Ivers and Stevenson
bull University of Sydney Sesqui Research Grant Australia
Direct research costs and salary support for Kardamanidis
D I F F E R E N C E S B E T W E E N P R O T O C O L A N D R E V I E W
There are no differences between the protocol and the review
I N D E X T E R M S
Medical Subject Headings (MeSH)
lowastMotorcycles Accident Prevention [lowastmethods] Accidents Traffic [lowastprevention amp control] Licensure Program Evaluation Wounds
and Injuries [mortality lowastprevention amp control]
47Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 5050
MeSH check words
Humans
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 1350
Table 4 Risk of attrition bias (Continued)
Waller 1992 Percentage of participants in analysis unclear percentage for the intervention group and 100 for the
control group
MAIDS 2004 Percentrage of participants in analysis unclear for both cases and controls
R E S U L T S
Description of studies
See Characteristicsof included studies Characteristics of excluded
studies Characteristics of ongoing studies
Our search resulted in 102 studies that seemed relevant of which
we excluded 79 studies because they did not meet the criteria for
considering studies for this review (see Characteristics of excluded
studies)
Basedon ourcriteriawe includeda total of 23 studies in this review
There were three randomised trials (Anderson 1980 Hall 1988
Swaddiwudhipong 1998) two non-randomised trials (Adams
1985 Leung 1987) 14 cohort studies (Raymond 1979 Osga
1980 Satten 1980 Jonah 1982 Mortimer 1984 Cooper 1988
Mortimer 1988 McDavid 1989 Waller 1992 Kloeden 1994
Davis 1997 Billheimer 1998 Perrino 2002 Savolainen 2007)
and four case-control studies (Kraus 1975 Hurt 1981 Haworth
1997 MAIDS 2004)
Over half of the studies were conducted 20 or more years
ago and the oldest was conducted 31 years ago Only three
studies were from this decade (Perrino 2002 MAIDS 2004
Savolainen 2007) Over 60 of the studies were conducted in
the USA (Kraus 1975 Anderson 1980 Osga 1980 Satten 1980
Hurt 1981 Mortimer 1984 Leung 1987 Hall 1988 Mortimer
1988 Waller 1992 Davis 1997 Billheimer 1998 Perrino 2002
Savolainen2007) three inAustralia (Adams 1985 Haworth 1997
Kloeden 1994) three in Canada (Jonah 1982 Cooper 1988
McDavid 1989) onefromthe United Kingdom (Raymond 1979)one from Europe (with sampling areas located in France Ger-
many Netherlands Spain and Italy) (MAIDS 2004) and one
in Thailand (Swaddiwudhipong 1998) Only seven studies were
published in peer reviewed journals (Kraus 1975 Jonah 1982
Mortimer 1984 Cooper 1988 Mortimer 1988 McDavid 1989
Swaddiwudhipong 1998) The other studies were reports forhigh-
way safety and road traffic organisations
The studies by Anderson 1980 Hall 1988 Kloeden 1994
Billheimer 1998 (the basic California Motorcyclist Safety Program
component of the study) and Perrino 2002 examined manda-
tory pre-licence training Cooper 1988 McDavid 1989 and
Swaddiwudhipong 1998 assessed non-mandatory training that
when passed resulted in a licence Raymond 1979 Osga 1980
Satten 1980 Jonah 1982 Mortimer 1984 Adams 1985 Leung
1987 Mortimer 1988 Waller 1992 Davis 1997 Billheimer 1998
(the Experienced Rider Course component of the study) andSavolainen 2007 assessed non-mandatory training that could be
taken by motorcycle riders with and without a licence The case-
control studies (Kraus 1975 Hurt 1981 Haworth 1997) assessed
lsquoanyrsquo type of rider training and therefore could not be grouped by
pre- or post-licence rider training The MAIDS 2004 case-control
study did make a distinction between participants having received
compulsory pre-licence training or additional (non-compulsory)
training We did not find a study that compared one form of pre-
or post-licence training (eg on-road off-road theory or a com-
bination) versus another form of training All studies compared
eithermandatory or non-mandatory formal training versus no for-
mal training with the exception of Haworth 1997 who compared
different levels of training courses (beginner intermediate and ad-vanced)
We have presented a more detailed description of the individual
studies in Characteristics of included studies We have provided
a description below of the types of motorcycle rider training that
were evaluated in the included studies
Types of motorcycle rider training evaluated in the
studies
The types of training were varied Duration of training ranged
from two or three hours to 24 hours administered over two or
three days Most training consisted of a theoretical and a practical
component usually off-road The content ranged from provid-ing riders with manuals lectures discussions showing films and
administering knowledge and skill tests Little information was
provided about the qualifications of trainers who conducted the
training
Six studies (Osga 1980 Satten 1980 Mortimer 1984 Mortimer
1988 Waller 1992 Savolainen 2007) assessed the effectiveness of
the USA-based Motorcycle Safety Foundationrsquos Motorcycle Rider
Course (MSF MRC) This course was released in 1976 It com-
prises eight hours of theory (lectures discussions films and multi-
plechoice tests) a textbook and 12 hours of off-road practice The
Motorcycle Safety Foundation is a national not-for-profit organ-
11Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 1450
isation sponsored by manufacturers and distributors of motorcy-
cles Many USA states have adopted the MSF Basic Rider Courseand Experienced Rider Course (wwwmsf-usaorg) since its release
in 1976
The Motorcycle Operators Skill Test was also developed
by the USA Motorcycle Safety Foundation in coopera-
tion with others (httpwwwwebbikeworldcommotorcycle-
trainingmotorcycle-test) The MOST is a licensing test consist-
ing of a manual knowledge and skill test (comprising eight exer-
cises of increasing difficulty and skill requirements) and a mini-
skill test conducted with a simulator Some studies compared this
alternative licensing test to the standard licensing test However
as the MOST is not a type of rider training we have included only
the two studies that assessed the MOST in combination with one
or more training components in this review Both Anderson 1980and Hall 1988 assessed the MOST in combination with a train-
ing component of three hoursrsquo duration consisting of instructions
about the skills needed to pass the MOST (30 minutes of the-
ory and two and a half hours on the motorcycle) Hall 1988 also
assessed the MOST in combination with the Motorcycle Safety
Foundationrsquos Motorcycle Rider Course as described above
Other USA studies evaluated the Wisconsin motorcycle rider
course a 22-hour off-street programme that teachesthe basic rules
of safe motorcycling (Leung 1987) the USA Maryland Motorcy-
cle Rider Course (MRC) a mandatory (for those younger than 18
years)pre-licence 20-hour coursecomprising eight hours of theory
and12 hours of practice and created by the Maryland Motor Vehi-
cle Administration in 1985 (Perrino 2002) the California Motor-cyclist Safety Program (CMSP) a mandatory (for those younger
than 18 years) pre-licence 16-hour basic course (Billheimer 1998)
and Experienced Rider Course (ERC) of eight hoursrsquo duration
(Billheimer 1998) The CMSP still offers these courses ( http
wwwca-msporg) but they are now MSF courses as described
above Davis 1997 did not give any information about the assessed
rider training
The Canadian studies evaluated the British Colombia Safety
Councilrsquos 37-hour motorcycle safety programme (no further in-
formation on the content was provided) (McDavid 1989) and
the Canada Safety Council Motorcycle Training Program (MTP)
comprising four hours of theory and 16 hours of practice (six
hours on-road) over twoweekends (Jonah 1982) The MTPcoursewas made available in 1974 and is still being offered in Canada
(httpwwwsafety-councilorgtrainingmtpmtphtm) Cooper
1988 did not give any information about the assessed rider train-
ing However the study was conducted in British Colombia in
Canada in 1988 the same time as the McDavid 1989 study was
conducted
In the United Kingdom Raymond 1979 assessed the non-manda-
tory RACACU (Royal Automobile ClubAuto-Cycle Union)
training scheme The training programme consisted of either 24
one-hour sessions or 13 two-hour sessions Twelve hours of the
training focused on theory about basic operational procedures
traffic laws and highway codes The practical session focused on
basic operational skills riding on a training area and riding in traf-fic with an instructor A proficiency test was administered at the
end of the course
One study from rural Thailand (Swaddiwudhipong 1998) evalu-
ated a community-based rider education programme consisting of
theory on motorcycle injury risk helmet use traffic laws vehicle
regulations and traffic signs and a written and skill test leading to
a licence
TheAustralian studies evaluateda rider trainingcourse of 12 hours
over two consecutive days (four hours of theory and eight hours of
practice)administeredto AustraliaPosttelegram deliverers(Adams
1985) and a mandatory pre-licence Ridersafe course consisting
of two tiers the first tier being a training course on the basics
of motorcycle riding where a pass entitled the rider to obtain alearnerrsquos permit and the second tier being a skill test where a pass
entitled the riderto obtain a probationary motorcycle licence The
training course consisted of two four-hour training sessions on
separate days involving instruction videos and actual riding on
motorcycles The skill test consisted of a four-hour preparation
course of 74 minutes of lecturediscussion 38 minutes of video
and 75 minutes of on-course riding which was the testing phase
(Kloeden 1994) The Rider Safe course is still being provided by
Transport South Australia a state government organisation (http
wwwtransportsagovau )
The duration or content of rider training in the four case-control
studies (Kraus 1975 Hurt 1981 Haworth 1997 MAIDS 2004)
was not described
Risk of bias in included studies
We describe the methodological quality of the included studies
by study type below More detailed information can be found in
Table 1 Table 2 Table 3 and Table 4
When considering the findings from these studies it is important
to note the quality (or lack thereof) of reporting of the studies
Quality of reporting is the extent to which a report provides in-
formation about the design conduct and analysis of a study ( Juumlni2001) Huwiler-Muumlntener 2002 found that even though trials
may be well reported it does not mean that the methodological
quality is of equal standard and well-conducted trials may never-
theless be reported badly The reporting quality of most studies in
this review was poor Most studies were old and did not meet con-
temporary reporting standards This made data extraction com-
plicated It also made it difficult to contact authors for additional
information
Because of significant heterogeneity between studies as well as the
poor quality of most studiespooling of studyresults to calculatean
overall measure of effect was not appropriate Instead we provided
a description of our judgement of risk of bias in the text
12Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 1550
Risk of bias in the randomised trials
There were three randomised trials (Anderson 1980 Hall 1988Swaddiwudhipong 1998) Anderson 1980 and Hall 1988 had an
initial sample size of 41188 and 26464 motorcycle licence ap-
plicants respectively Both studies assessed the effectiveness of the
pre-licence MOST with a trainingcomponent Swaddiwudhipong
1998 evaluated the effectiveness of a community-based rider ed-
ucation programme in six districts
Allocation bias
Anderson 1980 reported using random birth date tables to allocate
participants to either the intervention or control group Hall 1988
reported the use of a computerised algorithm set in a central of-
fice but also used manual random assignment when the computer
went down Swaddiwudhipong 1998 reported lsquosimple random
samplingrsquo for the delivery of the intervention where the sampling
frame included nine districts Three districts were excluded since
they were too hard to access The remaining six districts were ran-
domly allocated to the intervention (three) and the control (three)
groups Systematic sampling wasthen used ordered by village size
for the villages within districts Allocation concealment was not
reported in any of the studies The descriptions of randomisation
in these studies suggest that there may have been limitations to
the process resulting in a lack of equivalence of the intervention
and control groups at baseline and therefore the possibility of
allocation bias Swaddiwudhipong 1998 reported that before theintervention the annual incidence rate of motorcycle-related in-
juries was slightly (but not statistically significant) higher in the
intervention areas than in the control areas Participants in the
intervention districts were also more likely to have a valid licence
and to wear a helmet more often than participants in the control
districts Anderson 1980 measured differences at baseline on the
participant and prior driving record variables and found these to
be negligible Hall 1988 did not report on baseline differences
between the intervention and control groups As the Anderson
1980 and Hall 1988 studies had sample sizes of over 40000 and
26000 respectively and used some form of randomisation it is
likely that any confounding factors at baseline would have been
equally distributed between comparison and intervention groups
Selection bias
After randomisation Anderson 1980 and Hall 1988 excluded
08 and 3 participants respectively because of assignment to
the wrong group administration of the wrong intervention or
other reasons The resulting lack of analysis by rsquointention-to-treatrsquo
can introduce biasby removing the effectivenessof randomisation
Nonetheless the percentage excluded was small and the number
of overall participants large so it is unlikely to have had a major
effect on the final results In the Swaddiwudhipong 1998 study
villages were systematically sampled into intervention or control
groups The intervention was implemented and outcomes of in-
jury and death were measured at a community level Interventionand control districts were randomly selected however there were
only three intervention and three control communities
Detection bias
None of the randomised trials reported blinding of outcome as-
sessors Data collection tools used by Anderson 1980 and Hall
1988 were driver records holding police recorded crashes and of-
fences Swaddiwudhipong 1998 used police and hospital records
at a district level and post-mortem examination for deaths None
reported on validity or reliability for any of the instruments
Attrition bias
Anderson 1980 and Hall 1988 had driver records available for
96 to 100 of participants They also collected self-reported
data but the response rate to the questionnaires was very low (low-
est response rate 42 and 30 respectively) Attrition bias in
Swaddiwudhipong 1998 was not an issue as outcome data were
available for all sub-districts included in the trial
The intervention reported in Anderson 1980 and Hall 1988 tri-
als had an unanticipated effect on the participants In both trials
an average of only 28 of licence applicants in the intervention
group eventually obtained their licence compared to 35 in the
control group This indicates that the licensing process in generalseems to have deterred applicants from obtaining a licence and
although no statistical tests of the difference were conducted this
deterrence appears even greater for those who had rider training
This is not true attrition bias as the studies did not experience
differential loss to follow up from outcome measurement Instead
differential attainment of the intermediary step of licensure pre-
cluded participants from being at risk for the outcomes of interest
they are not driving a motorcycle hence they are not at risk for
motorcycle-related crashes injury death or offences In short it is
not possible to assess whether a reduction in crashes in the inter-
vention group would be due to changes in riding habits as a result
of training or simply due to delayed or never attained licensure
Intervention integrity
Neither Anderson 1980 nor Hall 1988 reported on the consis-
tency in the way the intervention was administered the possible
lsquocontaminationrsquo of participants in the intervention group by other
types of formal or informal rider training or of participants in the
control group by the intervention Swaddiwudhipong 1998 did
not report on the number of villagers in each group that actually
attended the intervention However they did report on a number
of other potential training courses that were being implemented
at the same time
13Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 1650
Analysis
Anderson 1980 and Hall 1988 acknowledging the lsquodeterrent ef-fectrsquo described above conducted two separate analyses The first
analysis included the total study population of licence applicants
over two years from the time of their application (the population
denominator) The second analysis was performed on the licensed
(intervention and control) subgroups only two years from the
time of obtaining their licence (the licensed rider denominator)
As only a third were licensed and exposed to riding the motorcy-
cle it was impossible to determine whether reported differences
in crash rates across groups were due to the intervention or to
other lsquoconfoundingrsquo characteristics of those who were determined
in their efforts to obtain the licence When analysis was restricted
only to those who gained their licence the same difficulties re-
mained The benefits of the randomisation process were thus lost
due to this lack of licensure of riders during the process As there
were only three intervention and three control groups in the study
reported by Swaddiwudhipong 1998 analyses were hindered by
smallnumbers for outcome comparisons betweenthe intervention
and the control districts
Risk of bias in the cohort studies and non-randomised
trials
We discuss the methodological quality of cohort studies (seven
prospective studies (Raymond 1979 McDavid 1989 Waller 1992
Kloeden 1994 Billheimer 1998 Perrino 2002 Savolainen 2007)and seven retrospective studies (Osga 1980 Satten 1980 Jonah
1982 Mortimer 1984 Cooper 1988 Mortimer 1988 Davis
1997)) and non-randomised trials (Adams 1985 Leung 1987) in
the one section as the biases for the two study types are similar
The quality of these studies was in general poor
Selection bias
The confounding factors expected to have been addressed in the
studies were sex age riding experience prior driving records (of-
fences crashes) previous training and the following proxies for
risk taking behaviour helmet use speed and alcoholdrug use All
of these factors have been shown to be associated with an increasedrisk of motorcycle crash (Peden 2004 Zambon 2006) The pos-
sible source of bias in cohort studies and non-randomised trials
is that groups may differ on important characteristics at baseline
participants lsquoself-selectrsquo into either the rider training group by de-
ciding to undertake a rider training course or into the lsquocontrol
grouprsquo by deciding not to undertake a course Participants who
undertake a non-mandatory rider training course may differ from
those who choose not to on characteristics such as age sex rid-
ing experience and riding behaviour as well as other possible con-
founders For example riders who take a training course may be
more safety-conscious or on the contrary may do so because they
have previously experienced a crash or because they perceive that
they have poor riding skills Therefore ideally all participants in
the intervention group need to be included in the study as wellas a random sample of controls taken from the same population
as the intervention group If that is not possible intervention and
control groups either need to be carefully matched to make them
comparable or differences at baseline need to be adjusted for in
the analysis
Most of the studies using a non-randomised trial or cohort design
compared an intervention group to a control group that differed
on a number of important factors The studies in general failed to
adjust for important differences found at baseline and their results
cannot be meaningfully interpreted Ten studies (Raymond 1979
Osga 1980 Satten 1980 Jonah 1982 Mortimer 1984 Mortimer
1988 McDavid 1989 Waller 1992 Davis 1997 Savolainen 2007)
assessed intervention and control groups that were not drawnfrom the same population In most of these cases the intervention
group was selected from a list of training graduates who were usu-
ally young female and less experienced riders whereas controls
were recruited from a general population of motorcycle riders who
were usually older male and experienced riders For three stud-
ies it was unclear how they were selected (Leung 1987 Cooper
1988 Perrino 2002) Only Kloeden 1994 recruited groups from
the same population of motorcycle applicants and Adams 1985
from a group of professional riders Nevertheless due to non-ran-
dom selection of intervention and control groups differences at
baseline are likely even though study groups were from the same
population Three studies adjusted for some confounding fac-
tors found at baseline through matching (McDavid 1989 Waller1992 Billheimer 1998) However matching has limitations in
termsof controlling for multiple confounders especially those that
are unknown to influence the effect of rider training on the out-
comes One study matched the control and intervention groups
by excluding controls (approximately 3500 (66 of initial con-
trol group) that differed from the intervention group) (Kloeden
1994) Five studies (Osga 1980 Satten 1980 Adams 1985 Davis
1997 Perrino 2002) did not adjust for differences between study
groups at baseline either in the study design or analysis Results
from these studies are therefore not reliable The other six stud-
ies (Raymond 1979 Jonah 1982 Mortimer 1984 Cooper 1988
Mortimer 1988 Leung 1987 McDavid 1989 Billheimer 1998)
adjusted for some differences between groups at baseline by eitherbi-variate or multi-variate analyses or stratification and standardi-
sation of riding exposure (outcomemiles driven) or stratification
by age group The limitation of stratification is that it is difficult to
control for many variables simultaneously because a large number
of strata will need to be generated resulting in small numbers that
cannot be meaningfully analysed Savolainen 2007 developed sta-
tistical models to assess the probability of a crash after taking the
rider training course in an intervention and control group while
examining differences by sub-groups (eg age) Participation rates
were also very low in this study with only 5584000 (1395)
in the eligible intervention group and 5884000 (147) for the
14Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 1750
control group completing surveys
Detection bias
None of the studies reported blinding of outcome assessors to the
intervention or control status of the participants Seven studies
used driver records containing police recorded data as their main
measure of outcome (Leung 1987 Cooper 1988 McDavid 1989
Waller 1992 Kloeden 1994 Davis 1997 Billheimer 1998) An-
other four studies (Raymond 1979 Satten 1980 Mortimer 1984
Mortimer 1988) only used self-report Raymond 1979 was the
only study using self-report as the main outcome measurement in
which study instruments were validated Self-report might prevent
the underreporting that has been described for the use of police-
driver records (see rsquoRisk of biasrsquo in the randomised trials above)but has other risks of bias especially if questionnaires or inter-
view methods have not been validated and tested for reliability
Recall bias arises from mistakes in recollecting events both be-
cause of failures of memory and because of looking at things lsquowith
hindsightrsquo and possibly changed views (httpwwwcochraneorg
resourcesglossaryhtm) As many studies were retrospective risk
of recall bias is likely and may have increased with length of follow
up For studies that used self-reported outcomes only it is also
possible that social desirability response bias played a role where
participants in the training group may have felt it was lsquoexpectedrsquo
to report good driving histories Multiple methods to measure the
same outcome for both groups decreases the risk of detection bias
for example because self-reported answers can be cross checkedwith other sources of information such as driver records
In this regard three studies (Osga 1980 Jonah 1982 Perrino
2002) used both self-report and driver records Osga 1980 found
that for both intervention and control groups more crashes were
reported through self-report than on state crash records an aver-
age of 54 of self-reported crashes could be found back on state
crash records One industry-based study (Adams 1985) used oc-
cupational health records which were not validated but which are
unlikely to suffer from underreporting due to work cover claims
Savolainen 2007 used information on rider crashes for a five-year
period but did not specify from where this information was ob-
tained
Attrition bias
The majority of studies (Raymond 1979 Satten 1980 Adams
1985 Leung 1987 Kloeden 1994 Davis 1997 Billheimer 1998
Perrino 2002) inadequately reported on losses of participants to
the study We can therefore not be certain that there were no sys-
tematic differences in the characteristics between the intervention
and control group participants Savolainen 2007 reported the re-
sponse rates to their baseline survey but did not report on the per-
centage of outcome data that was available for their study sample
A similar deterrent effect of rider training as described for the ran-
domised trials (Anderson 1980 Hall 1988) occurred in two stud-
ies Jonah 1982 reported that 65(14992310) of the trained rid-
ers stopped riding compared to 47 (9492029) of the untrainedriders They excluded non-active and non-licensed riders before
the study commenced so the differential choice to continue riding
after training did not affect their analyses Leung 1987 reported
that the percentage of participants in the intervention group ob-
taining their licence was 50 (no number provided by authors)
lower than in the control group This may have biased the mea-
surement of outcomes because unlicensed participants will not be
exposed to riding the motorcycle and therefore not be at risk of
being involved in a crash or violation As explained above this
makes it difficult to assess whether the reduction in crashes in the
intervention group is due to a change in driving habits or simply
due to delayed licensure
Mortimer 1984 and Mortimer 1988 describe that about 44 and30 of course graduates respectively were excluded from analysis
because they had not ridden a motorcycle in the previous year
However because these studies were retrospective and the controls
were activeriders at the time of recruitment a decision on whether
the deterrent effect was differential between the two groups could
not be made Though not reported quantitatively Kloeden 1994
also describes the effectof rider training deterringindividuals from
becoming licensed and hence exposure to riding and risk of crash
death or offences involving a motorcycle Osga 1980 reported
that 27 of the trained riders did not ride after the course and
excluded these from the analysis
Use of different data collection strategies may have contributed to
differential loss to follow up in the cohort studies Satten 1980for example interviewed the intervention group over the tele-
phone but mailed questionnaires to the control group Mortimer
1984 and Mortimer 1988 distributed questionnaires in person to
the control group but mailed them to the intervention group
This may have played a role in the observed differential response
rates where more subjects responded in the personally approached
groups
Survival bias
For those studies that assessed offences crashes and injuries but
not death survival bias may have also played a role Motorcyclists
who were fatally or seriously injured in crashes during the follow-
up period after the intervention would have been less likely or
unable to respond to surveys This could bias the effect measures
towards or away from the null None of the studies wrote about
this issue or examined the characteristics of those lost to follow up
in order to address the possibility of survival bias
Intervention integrity
Although the interventions were intended for the entire lsquoexposedrsquo
groups most studies reported difficulty in tracking participants
whoreceivedthe interventionand those whodid not Adams 1985
15Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 1850
estimatedthat 62of thetotalstudysample received the interven-
tion The remaining studies did not report the percentage of thestudy population that received the intervention as was intended
or reported that this percentage was unknown (Raymond 1979
Kloeden 1994 Satten 1980 Jonah 1982 Mortimer 1984 Leung
1987 Cooper 1988 Mortimer 1988 Billheimer 1998 Perrino
2002 Savolainen 2007) Only Raymond 1979 reported the degree
to which their intervention was delivered as intended No studies
reported on possible lsquocontaminationrsquo of participants by other inter-
ventions than the one to which they were allocated Because Osga
1980 Satten 1980 Jonah 1982 Mortimer 1984 Cooper 1988
Mortimer 1988 Waller 1992 and Davis 1997 were retrospective
studies and participants were recruited from a list of graduates it
was known that they had completed and passed the course
Analysis
In some studies some participants who reported at baseline that
they were untrained decided to enroll in a rider training pro-
gramme after the baseline survey The non-randomised trial by
Adams 1985 experienced this move of subjects from the untrained
to the trained group but failed to conduct analyses by lsquointention-
to-treatrsquo This can bias the results in randomised and non-ran-
domised trials if participants who changed groups differ signifi-
cantly with regard to potentially confounding characteristics On
the other hand Perrino 2002 also experienced this move of sub-
jects from theuntrained to the trained groupmore than 100riders(5 of the total sample) originally stated that they were untrained
at baseline but then received training between the baseline and
follow-up surveys Yet since the Perrino 2002 study was a cohort
their analyses then proceeded by including those riders who were
lsquonewlyrsquo trained as exposed Riders who were lsquonewlyrsquo trained since
baseline but with a programme other than the one under study
were excluded from the analysis Nevertheless confounding fac-
tors were not accounted for in the analyses
McDavid 1989 matched exposed with unexposed subjects but
could not find suitable matches for the majority (60) of their
subjects They could therefore only conduct analysis on the result-
ing40 (139346) which is an inappropriate andinefficient anal-
ysis Some studies that reported on the issue of non-active licensedriders(Satten 1980 Jonah 1982 Mortimer 1984 Mortimer 1988
Kloeden 1994) excluded non-active riders from analysis intro-
ducing selection bias as it is likely that riders who continued to
ride differed with regards to potentially confounding character-
istics across intervention and control groups Kloeden 1994 for
example conducted several analyses to assess the effect of various
sub-sets of the control group Original analyses included all con-
trols and led to erroneous results since only a small proportion of
controls hadgone on to get a licence after their learnerrsquos permitand
were actually exposed to the possibility of experiencing a crash
Subsequent analyses compared only intervention and control sub-
jects who had a licence Waller 1992 reported crashes by person-
days of exposure but did not explain how they could be sure that
each study intervention and control participant was an active riderand actually exposed to the risk of experiencing a crash
Potential confounders that are unequally distributed by study arm
at baseline need to be dealt with in the analysis Satten 1980
Adams 1985 and Davis 1997 did not take into account any con-
founding factors either by design or in the analysis McDavid
1989 Waller 1992 and Billheimer 1998 accounted for potential
confounding factors at the design stage by matching the subjects
in the intervention and control groups Other studies (Raymond
1979 Mortimer 1984 Mortimer 1988 Perrino 2002 and also
Waller 1992) adjusted for one or two confounders by stratifica-
tionor standardisation (eg crashesmiles driven age group) Osga
1980 conducteda univariate analysis of crash rates in which no po-
tential confounders were accounted for but also calculated cumu-lative crash rates by mileage travelled Only Jonah 1982 Cooper
1988 and Savolainen 2007 conducted a multivariate analysis
Risk of bias in the case-control studies
A case-control design is suitable for the assessment of risk fac-
tors for outcomes such as motorcycle crashes injuries or deaths
but not appropriate for the evaluation of an intervention such
as motorcycle rider training It can potentially result in spurious
conclusions regarding the efficacy of the intervention because the
intervention has not been applied at random to an intervention
and control group Even with the control of possible confounders
using a case-control design to evaluate an intervention can lead tothese kinds of results since those who choose to undertake rider
training may differ on important prognostic factors that affect the
outcome compared to those who do not choose to participate in
rider training
Selection bias
A potential source of bias in case-control studies as with all other
study types where the intervention is not randomly allocated to
participants is that comparison groups are likely to differ on po-
tentially confounding characteristics This means that ideally all
cases need to be included in the study as well as a random sample
of controls taken from the population that gave rise to the cases If that is not possible cases and controls either need to be carefully
matched to make themcomparable or differences at baseline need
to be adjusted for in the analysis All four case-control studies were
population based Kraus 1975 included all fatalities and injured
cases leading to hospitalisation and appropriately recruited a ran-
dom sample of controls from the same population Hurt 1981 did
not manage to recruit all or a random sample of cases Only half
of the notifications of crashes to the police were received in time
by the researchers the research team for unreported reasons was
only able to respond to half of those again Data collection was
therefore completed for only about a fifth of all cases Neverthe-
less this was still a large sample of 900 cases Haworth 1997 using
16Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 1950
a similar study design was more likely to have included all cases
but this is not entirely clear from the study reported It is unclearfor example whether fatalities were included as they did not use
coronerrsquos reports for the recruitment of cases The MAIDS 2004
study was conducted in five European countries but it was unclear
how cases were recruited whether they were all cases during the
two years that the study lasted or a random sample of all cases
what the response rates were etc Controls were recruited from
randomly selected petrol stations but it was likewise not clear what
the eligibility criteria for inclusion in the study were how they
were recruited (eg randomly) and what the response rate was
Hurt 1981 and Haworth 1997 both matched controls to cases by
recruiting themat the sametime ofdayandsamedayof weekas the
crash occurred Hurt 1981 however recruited controls two years
later andit is likelythat there would have beendifferences betweenthe two comparison groups One important difference was that
during data collection for cases there was very little specialised
motorcycletrainingavailable in the studyregionwhereas two years
later when control data were collected there was some training
available
These four case-control studies also suffered from selection bias
due to potential differences between those that gave consent to
participate in the study and responded to interviews andquestion-
naires and those that did not None of the studies have assessed
such differences although Hurt 1981 did acknowledge that con-
trols giving consent to cooperate may have differed from those
motorcyclists who did not
Detection bias
Measures that can be taken to reduce detection bias in case-con-
trol studies are blinding of interviewers for case or control status
blinding of the participant for the study factor of interest validat-
ing measurement instruments and using these in an identical way
for cases and controls and cross checking self-reported informa-
tion with other more objective sources such as registers None of
the studies reported blinding of the intervention assessors to case
or control status but all used consistent measurement methods for
both case and control groups
All case-control studies used self-report as their main measure toassess whether cases and controls had received rider training and
if so the type of training The major potential source of bias in
case-control studies when self-report is used as a measure is recall
bias One way of reducing recall bias is by double-checking self-
reported data with more objective data Kraus 1975 also used po-
lice accident reports for information on rider characteristics but
such information was not available about rider training Haworth
1997 and MAIDS 2004 did not report on any measures to limit
recall bias Hurt 1981 reported further questioning of cases by
the interviewer when answers did not make sense or when par-
ticipants seemed to contradict themselves as part of the in-depth
unstructured interview technique This qualitative research tech-
nique can result in valuable information but as interviewers were
not blinded to case or control status of the participants it may have led to more thorough examination of cases compared to the
controls and thus differential detection bias
Intervention integrity
The level of detail about the rider training provided in the case-
control studies differed significantly but was minimal for each
Kraus 1975 simply asked whether lsquoanyrsquo motorcycle rider training
had lsquoeverrsquo been carried out Hurt 1981 asked participants whether
they hadbeen self taught informally taught or completed formally
taught rider training courses MAIDS 2004 asked whether partic-
ipants had had pre-license training additional (non-compulsory)
training other training or no training whereas Haworth 1997asked about the number of courses that a participant had com-
pleted whetherthey were basic intermediate or advanced courses
and how long ago the training had been done None of the case-
control studies reported collecting information about further de-
tails about any training programmes (eg type of training cur-
riculum duration of training)
Analysis
Kraus 1975 accounted for confounding factors through the exclu-
sion of females and stratification by age and frequency of motor-
cycle use Hurt 1981 matched cases and controls for time of day
and day of the week that the crash occurred and did not adjust forany confounding factors in the analysis Haworth 1997 matched
groups for time of day and day of the week that the crash occurred
and stratified by age blood alcohol level and licence status in the
analysis None of the studies used multivariate methods MAIDS
2004 did not provide any information about any participant char-
acteristics other than the study exposures of interest and did not
describe any methods to adjust for potential confounding factors
Effects of interventions
We have presented the results in a descriptive manner due to the
substantial heterogeneity in study types interventions and out-comes and the generally poor quality of the studies We have
grouped the interventions by mandatory and non-mandatory pre-
licence rider training and post-licence rider training and further
by study type and by outcome (crashes injurious and fatal crashes
and offences) We discuss results that are reported as statistically
significant (whether a P value was reported or not)
Pre-licence rider training - mandatory
Randomised trials
17Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2050
Crashes
Total group of participants
Anderson 1980 found a reduction in motorcycle crash rates in
the intervention group (MOST skill test with remedial training)
compared to the control group (who received the skill testwithout
a training component) The percent differenceranged from 143
six months after the programme (P lt 005) to 211 (P lt 0001)
after one year and 127 (P lt 001) after two years Hall 1988
found no statistically significant difference in crash rates between
intervention groups at any point in time
Subgroup of participants who obtained their licence only
Anderson 1980 found a reduction in motorcycle crash rates in the
intervention arm but this was statistically significant at one year
only (24 to 218 reduction P lt 005) Hall 1988 found no
statistically significant differences
Offences
Total group of participants
The percentage of offences in Anderson 1980 after six months
and one and two years was slightly higher in the intervention
arm compared to the control group but these differences were
not statistically significant Likewise the study by Hall 1988 did
not find a statistically significant difference in offences between
groups
Subgroup of participants who obtained their licence only
Hall 1988 found no significant difference in offence rates between
the intervention and control groups Anderson 1980 found thatthe percentage of offences increased in the intervention arms com-
pared to the control group This difference was statistically signif-
icant after two years (51 to 115 increase P lt 005)
Cohort studies
Crashes
Neitherof the cohort studies (Kloeden1994 and Billheimer1998)
found a statistically significant difference in crash rates between
the intervention and control groups
Offences
Billheimer 1998 (for the California Motorcyclist Safety Program
course) and Perrino 2002 did not find a statistically significant
difference in offence rates between the intervention and control
groups
Case-control studies
Crashes
MAIDS 2004 found that a slightly higher percentage of controls
(not involved in a crash) had had pre-license rider training than
cases (involved in a crash) This difference was not statistically significant (50 of controls vs 477 of cases OR 091 95 CI
076 to 110)
Pre-licence rider training - non-mandatory
Randomised trial
Injuries and deaths
Swaddiwudhipong 1998 found a significantly lower injury ratein the intervention villages (1051000 population 95 CI 92
to 120) compared to the control villages (1691000 population
95 CI 152 to 185) one year after the intervention This differ-
ence was not maintained two years after the intervention (127
95 CI 112 to 143 intervention 151 95 CI 135 to 167
control) There was no significant difference in death rates be-
tween groups
Cohort studies
Crashes
Cooper 1988 foundno statistically significant differencein crashes
between the intervention and control groups McDavid 1989
found a lower crash rate in the intervention group (14139) com-
pared to the control group (23139) but no test of statistical sig-
nificance was undertaken
Injuries and deaths
McDavid 1989 found that the intervention group had fewer in-
jurious crashes (814 crashes) compared to the control group (16
23 crashes) but no statistical significance test was reported
18Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2150
Offences
Cooper 1988 foundno statistically significantdifference in offencerates between the intervention and control groups McDavid 1989
foundno statistically significantdifference in offence ratesbetween
the intervention and control groups
Non-mandatory rider training
Non-randomised trials
Crashes
Neither Adams 1985 nor Leung 1987 found a statistically signif-icant effect of rider training on crash rates
Offences
Leung 1987 found a statistically significant lower offence rate in
the intervention group compared to the control group for both
males and females and for all age groups except for women 35
years or older (P values not reported)
Cohort studies
Crashes
Osga 1980 in the univariate analysis reported a lower percentage
of crashesin theintervention group (that continued to ride) (28
168595) than in the control group (34 112329) In the cu-
mulative crash rate analysis however Osga1980 found statistically
significant (at the P lt 001 level) higher crash rates in the interven-
tion group than in the control group at allmileage intervals (1000
to 29999 miles) travelled After riding more than 30000 miles
(when over 50 of the trained riders had been involved in a crash
compared to approximately 40 in untrained riders) this differ-
ence was no longer statistically significant Satten 1980 found a
lower percentage of crashes in the intervention group (13 969)
compared to the control group (25 2079) but the differencewas not assessed statistically Davis 1997 found a lower crash rate
in trained riders but it was unclear how this was calculated be-
cause the size of the control group was not reported Jonah 1982
and Billheimer 1998 (for the Experienced Rider Course) found no
statistically significant difference in crashes between the interven-
tion and control groups Although the difference between those
who completed the Experienced Rider Course and no course in
Billheimer 1998 neared significance at six months after the course
(P = 007) large P values were observed for one year after (P =
030) Raymond 1979 found a statistically significant higher crash
rate for the intervention group (1861000 miles ridden) com-
pared to the control group (1181000 miles ridden) in both males
and females (P lt 0001) especially during the first 5000 miles of
riding Waller 1992 found a statistically significant higher crashrate for the intervention group (27 crashes in three years or 070
crashes10000 person days) compared to the control group (5
crashes in three years or 013 crashes10000 person days) with a
crude relative risk of 538 (X 2=1513 plt005) Mortimer 1984
and Mortimer 1988 found no statistically significant difference in
crashes between the intervention and control groups
Savolainen 2007 found that the intervention group (those who
completed the Basic Rider Course) was 44 more likely to be
involved in crashes than the control group and those who com-
pleted the course multiple times were an additional 180 more
likely to be involved in a crash This finding was based on low
numbers and no P value or confidence level was given for either
result
Injuries and deaths
Satten 1980 also found a lower percentage of injurious crashes
(33 39) in the intervention group compared to the control
group(60 1220) although again no statistical significance test
was reported
Offences
Jonah 1982 did find a significantly lower self-reportedoffence rate
in the intervention group (15) compared to the control group(34 P lt 001) (no numbers given) Satten 1980 also found
a lower self-reported offence rate in the intervention group (1
71 versus 19100) but again this was not tested for statistical
significance Mortimer 1984 Mortimer1988 and Billheimer 1998
found no statistically significant differencein offence ratesbetween
the intervention and control groups
Case-control studies
CrashesMAIDS 2004 found that a slightly higher percentage of cases
(involved in a crash) had had additional non-mandatory rider
training than controls (not involved in a crash) This difference
was not statistically significant (17 of cases vs 12 of controls
OR 146 95 CI 064 to 337)
Rider training unspecified
Case-control studies
19Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2250
Injuries and deaths
Hurt 1981 found that the odds of injurious crashes were lower inthose who had received professional training compared to those
who had not Kraus 1975 found that in the younger age group
(less than 20 years) and the older age group (over 35 years) a larger
percentage of injured riders (12 and 21 respectively) reported
prior motorcycle rider training compared with the uninjured con-
trol group (6 and 5 respectively) (P lt 001) (no numbers
given) Haworth 1997 reported several different outcomes one
beingthat having completed at leastone trainingcoursecompared
to no training courses at all had no significant effect on the odds
of crashing (afteradjustment for age blood alcohol level or licence
status (learner permitprobationaryfull)) When comparing three
different levels of training courses (beginner intermediate and ad-
vanced) with no training no significant differences were found
in the odds of crashing However after adjustment for age and
blood alcohol level completing an advanced course as compared
to a beginners course was associated with a significant decrease in
the odds of severe motorcycle crashes resulting in serious injury or
death (OR 02 95 CI 01 to 08)
D I S C U S S I O N
The objective of this review was to quantify the effectiveness of
pre- andpost-licence motorcycle rider training on the reduction of
traffic crash involvement injuries deaths and offences committedby motorcycle riders Most included studies were cohort studies
There were a few randomised and non-randomised trials and case-
control studies
All studies compared either mandatory or non-mandatory formal
training with no formal training except for Haworth 1997rsquos case-
control study that compared levels of training (beginner interme-
diate and advanced) to each other
We found studies that evaluated the effectiveness of manda-
tory pre-licence training in reducing crash and offence rates
(Anderson 1980 Hall 1988 Kloeden 1994 Billheimer 1998
(the basic California Motorcyclist Safety Program component of
the study) Perrino 2002) non-mandatory training that when
passed resulted in a licence (Cooper 1988 McDavid 1989
Swaddiwudhipong 1998) non-mandatory training that could be
taken by motorcycle riders with or without a licence (Raymond
1979 Osga 1980 Satten 1980 Jonah 1982 Mortimer 1984
Adams 1985 Leung 1987 Mortimer 1988 Waller 1992 Davis
1997 Billheimer 1998 (the Experienced Rider Course compo-
nent of the study)) and unspecified rider training on crash injury
and death rates (Kraus 1975 Hurt 1981 Haworth 1997 MAIDS
2004)
The studies provided conflicting evidence regarding motorcycle
rider training which is likely due to the poor design quality and
reporting of the studies Most of the studies included in the review
were likely to be adversely affected by significant selection biasalong with substantial detection bias(measurement and recall bias)
which limits our ability to make conclusive comments
Mandatory pre-licence rider training
There were two RCTs three cohort studies and one case-control
study that examined pre-licence ridertraining The two RCTs eval-
uated the Motorcycle Skills Operating Test (MOST) (Anderson
1980 Hall 1988) Anderson 1980 found a statistically significant
reduction in crashes in the intervention group at six months one
and twoyearsfollow up However when only participantswho ob-
tainedtheirlicence were included in theanalysis thereduction wasstatistically significant only at one year follow up The other RCT
(Hall 1988) found no statistically significant differences in crash
rates at any point of time neither in the total population group
nor in the licensed subgroup Neither of the two cohort stud-
ies assessing crash rates found a statistically significant difference
in crash rates between intervention and control groups ( Kloeden
1994 Billheimer1998) andthe case-control study (MAIDS 2004)
also did not find a statistically significant difference in the odds of
experiencing a crash between trained and untrained riders With
regards to offences neither of the RCTs found a statistically sig-
nificant difference in offence rate between the intervention and
control groups in the analysis of the total group of participants
entering the trial However Anderson 1980 found a statistically significant increase in offence rates in the intervention group but
only after two years of follow up andonly in the analysis of partici-
pants who obtained their licence The two cohort studies assessing
offence rates found no statistically significant difference between
the groups (Billheimer 1998 Perrino 2002)
None of the five studies (RCTs or cohort) found a reduction in
crashes or offences following mandatory pre-licence rider training
except for Anderson 1980 who observed a statistically significant
reduction in crashes in the intervention group at one year in both
analyses It is not clear why both RCTs that evaluated the same
MOST and training intervention observed different results Both
studies enrolled large samples and observed a similar uptake of
licensure following the intervention It is possible that differencesmay have arisen due to different study samples (California ver-
sus New York) or that the intervention delivery differed between
studies
In both trials of the Motorcycle Skills Operating Test including
rider training only a third of motorcycle licence applicants ob-
tained their motorcycle licence It is possible that a licensing pro-
cess including a skills test may reduce crashes through decreased
exposure
Non-mandatory pre-licence training
20Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2350
Three studies assessed the effect of non-mandatory pre-licence
training The Cooper 1988 cohort study found no difference incrash or offence rates between the trained and untrained groups
We cannot however be confident about this result as the study
sufferedfrom selectionbiaswhichcouldhave biasedthe measure of
effect toward the null The RCT conducted by Swaddiwudhipong
1998 examined injury and death rates and found a statistically
significant lower injury rate in the intervention villages after one
year of follow up but no longer at two year follow up There
was no statistically significant difference in death rates between
intervention and control villages Our confidence in the results of
this study are hindered by the inadequacy of the randomisation
process that led to significant differences between the intervention
and control districts on two important variables The intervention
districts were more likely than the control districts to have a validlicence andto wear a helmetSincethese differences were nottaken
into account in later analyses they could explain the differences
observed in injury rates between districts Furthermore with only
six districts randomised the sample size was probably too small
to detect a difference between intervention and control groups
had there been one The McDavid 1989 cohort study found a
lower percentage of injuries in the intervention group although
a statistical test was not conducted McDavid 1989 drew their
intervention and control groups from different populations but
did adjust for confounders although they only analysed 40 of
the initial sample (139 trained and 139 untrained) Because of
these substantial weaknesses we are unable to draw conclusions
from this studyAn overall effect of pre-licence training on the reductionof crashes
injuries deaths and offences was not found although one should
be aware that the research of this issue to date has had significant
flaws However a mandatory skill test (such as the MOST) in
combination with and even without pre-licence training as part
of a licensing process seems to reduce the number of individuals
completing the licensing procedure and therefore indirectly re-
duces crash rates through a reduction in exposure to motorcycle
riding Making licensure a more difficult process may therefore be
seen as advantageous if it delays licensure and therefore reduces
exposure It is possible that the group who chose not to continue
riding a motorcycle would not have been the group to experience a
crash or commit an offence if they had continued to ride anyhowfor example because they are more safety conscious and this was
the reason for their giving up riding in the first place If this is the
case deterring individuals from completing their licence and rid-
ing may not necessarily decrease overall crashes deaths or offence
rates In addition other potentially adverse effects of complicat-
ing a licensing procedure need to be considered such as increases
in unlicensed riding In Thailand non-mandatory rider training
offered as a community intervention may have decreased injury
rates but this needs to be confirmed through further research
Non-mandatory training
The majority of the studies assessed non-mandatory training
(training independent of any licensing procedure) and the results
were varied The two non-randomised trials assessing crash rates found no statistically significant difference in crash rates between
the intervention and control groups (Adams 1985 Leung 1987)
Of the nine cohort studies Satten 1980 found a lower percentage
of crashesin the intervention group although it is unclear whether
this was a statistically significant result Davis 1997 also found a
lower percentage of crashesin theinterventiongroup although it is
unclear with which control group the intervention group wascom-
pared Raymond 1979 Osga 1980 Waller 1992 and Savolainen
2007 found a statistically significant higher crash rate in the in-
tervention group and the remaining cohort studies (Jonah 1982
Mortimer 1984 Mortimer 1988 Billheimer 1998) found no sta-
tistically significant difference in crash rates between intervention
and control groups No conclusions can be drawn from these stud-
ies with confidence The Satten 1980 and Davis 1997 studies did
not adjust for any confounding factors even though important
differences existed between the intervention and control groups
Although Raymond 1979 and Waller 1992 adjusted for a num-
ber of confounders at baseline (Waller 1992) and in the analy-
sis (Raymond 1979 Waller 1992) the intervention and control
groups were drawn from such different populations that it is pos-
sible that unmeasured confounders were responsible for the sta-
tistically significant increased risk of crashes observed in the inter-
vention group The observed difference could also be attributed totheobserved differential loss to followup (Raymond 1979) which
could bias the measure of effect towards or away from the null
The Osga 1980 study found a protective effect of rider training
in the univariate analysis but statistically significant higher crash
rates in the intervention group when accounting for riding expo-
sure Sex (a significantly higher proportion of females in the in-
tervention group) and other potentially confounding factors that
could explain thedifferencebetween the groups were howeverun-
accounted for Finally Savolainen 2007 did control for potential
confounding variables yet it is not clear whether variables were
controlled for simultaneously or one at a time One case-control
study (MAIDS 2004) found no statistically significant reduction
in the odds of experiencing a crash between trained and untrainedriders but the quality of this study was poor mainly because the
analysis did not account for any confounding factors
One cohort study (Satten 1980) assessed the effect of non-manda-
tory rider training on injurious crashes and found a lower percent-
age of injuries in the intervention group although a statistical test
was not conducted As mentioned above the Satten 1980 studyrsquos
validity is reduced due to its lack of adjustment for confounding
factors
The only non-randomised trial that assessed offence rates (Leung
1987) found a statistically significant lower offence rate in the in-
tervention group and of the five cohort studies that assessed of-
fence rates (Satten 1980 Jonah 1982 Mortimer 1984 Mortimer
21Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2450
1988) one study (Jonah 1982) also found a statistically significant
lower offence rate in the intervention group However we have lit-tle confidence in the results from Leung 1987 since potential con-
founding factors were not appropriately addressed Jonah 1982
did conduct a multivariate analysis but was hindered by an initial
low response rate The other four cohort studies did control for
several potential confounders and found no difference in offence
rates following non-mandatory post-licence rider training
The non-randomised trials and cohort studies evaluating non-
mandatory rider training were at high risk of substantial selection
bias due to the non-random allocation of participants as well
as low participation rates and attrition bias and could therefore
not provide conclusive evidence with regard to the effect of the
training It should be noted that despite the biases and other issues
with this group of studies most didnot detectmajor effects relatedto rider training in either direction This may suggest that any
potential impact of rider training will be modest
Studies examining unspecified modes of ridertraining
Three case-control studies assessed whether injured or deceased
motorcyclists were more or less likely to have had rider training
compared to control motorcyclists All had conflicting results
Hurt1981 foundthe odds of havingan injuriouscrash waslower in
those who had received rider training whereas Kraus 1975 found
that the odds of having an injurious crash were higher in youngerand older trained riders Haworth 1997 who compared advanced
versus beginner training found no effect on motorcycle injuries
or deaths although they did find that those who had completed
advanced training had reduced chances of crashing compared to
those who had done a beginnersrsquo training course Similar issues
with uncontrolled confounding and biases of selection as described
above arose with the three case-control studies Hurt 1981 suffers
from inadequate control of confounding via poor matching and
both the Kraus 1975 and the Haworth 1997 findings were only
controlled for limited factors by means of stratified analysis It is
likely that these limitations are the reason for the contradictory
directions of effect observed for the three case-control studies
Methodological weaknesses of the studies
In summary most studies investigating the effectiveness of mo-
torcycle rider training on reducing crashes injuries and death suf-
fered from serious methodological weaknesses
The majority of studies were non-randomised and were there-
fore limited mainly by selection bias Although the approaches of
matching stratification or regression analyses can minimise bias
(Fleiss 2004) they were poorly employed in the reviewed studies
The inadequate adjustment for confounders in particular adds
to our lack of confidence in the studiesrsquo results since it has been
demonstrated that individuals who choose to take motorcycle rider
training differ from those not taking rider training For instance
trained riders often differ from controls in age sex and experi-ence usually ride less and may have a more positive attitude to
safety as indicated by the reported more frequent wearing of seat
belts (Mortimer 1982)
The two large randomised trials reported that a large proportion
of their study group were never exposed to the risk of offences or
crashes because they never attained licensure This made it impos-
sible to assess whether a reduction in crashes in the intervention
group was a result of training or due to delayed or never attained
licensure
In some studies the sample sizes were too small to detect any sta-
tistically significant differences Follow-up time after training dif-
fered between studies and was an impediment to observing any
real effects of training if they existed As riding experience ac-cumulates over time any differences in outcomes between those
trained and untrained would diminish in studies with lengthy fol-
low-up times This dilution of any possible effect of rider training
may have also occurred in studies that mixed both novice and ex-
perienced riders in the study groups since it is hypothesised that
novice riders serve to benefit most from rider training This would
increase the difficulty in finding any effect of rider training par-
ticularly when sample sizes are small
Detection bias may also account for some of the differences ob-
served between trained and untrained riders Detection bias would
have occurred if the self-report hospital driver or police records
used to record the outcomes of crash injury death or offences
differed between the trained and untrained riders Research hasdemonstrated that not all crashes are reported to police for ex-
ample small crashes or off-road crashes (Kraus 1975) even serious
injury crashes have been found to be underreported (Anderson
1980) However this would only introduce detection bias if the
reporting of outcome differs for trained and untrained riders This
does appear possible in that responsible riders may be more likely
to enroll in a rider training course and also may be more likely
to report crashes to police than irresponsible riders meaning that
the number of crashes from riskier riders could be underrepre-
sented on records An analysis to investigate the effectiveness of
rider training may then show that rider training is not effective in
reducing crashes
Rather than using reported outcomes some studies used self-re-ported survey data although this technique suffers from non-
response biases In fact two studies reported differences in de-
mographic characteristics in response rates to questionnaires
(Anderson 1980 Hall 1988) Self-reported data also introduce
the risk of recall bias survival bias and social desirability response
bias Recall bias may have been equally distributed among trained
and untrained riders howeverit is hypothesised that trained riders
may have felt it was lsquoexpectedrsquo to report good driving histories
hence leading the rider training to appear effective
22Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2550
Limitations of this review
Limitations of this review were that we found very few recentstudies (the two most recent ones published in 2002 and 2007)
and that we found very few studies conducted in low-and middle
income countries even though our search strategy was set up to
identify such studies Anotherlimitation wasthat forthe reviewed
studies we could not assess all outcomes of rider training that are
potentially useful Some of the included studies assessed outcomes
such as use of safety clothing including helmets Such outcomes
are valuable since the use of motorcycle helmets has been demon-
strated to be effective in preventing head injuries (Liu 2004) De-
spite this the use of safety clothing was not included as an out-
come in this review because the studies measuring this outcome
were limited by poor data quality due to very low response rates
Another outcome of interest in assessing the effectiveness of rider
training is lsquotype of injuryrsquo As Haworth 2005 state in their review
of motorcycle licensing and training rider training could be con-
sidered effective if injuries become less severe because motorcycle
riders have learned how to avoid more serious injuries through
rider training The studies included in this review did not specify
type of injury and so this outcome could not be assessed Last
changes in knowledge and attitudes following rider training are
other potentially useful outcomes We have excluded such studies
(Osga 1979 Clayton 1990 Goldenbeld 2004) as these outcomes
are proxies rather than direct measures of the effectiveness of mo-
torcycle rider training We opted to report on the outcomes of
crashes injury death and offences alone since there is little ev-
idence about how changes in knowledge and attitudes relate to
these outcomes
A U T H O R S rsquo C O N C L U S I O N S
Implications for practice
Due to the poor quality of studies included in this review we were
unable to draw any conclusions about the effectiveness of rider
training on crash rates injurious or fatal or offence rates The
findings do suggest that mandatory pre-licence training may be
an impediment to completing a motorcycle licensing process thuspossibly indirectly reducing crash injury death and offence rates
through a reduction in exposure to riding a motorcycle On the
basis of the existing evidence it is not clear if and what type of
training reduces the risk of crashes injuries deaths or offences in
motorcyclists and the selection of the best rider training practice
can therefore not be recommended The suggestion that manda-
tory pre-licence training indirectly reduces exposure to riding the
motorcycle because trainees are impeded from completing the li-
censing process does not help those motorcyclists who persist in
obtaining their licence and riding the motorcycle Some type of
rider training is likely necessary to teach motorcyclists to ride a
motorcycle safely Good policy is based on sound evidence and
evaluation but such evidence for motorcycle rider training is seri-
ously lacking Policy makers are therefore encouraged to include arigorous evaluation component to any new or existing rider train-
ing programme Such an evaluation should ideally be randomized
and assess whether the training programme is effective and if so
which component(s) of the programme work best
Implications for research
Given theresults of this review which found littlereliable evidence
about the effectiveness of motorcycle rider training it is necessary
to build up an evidence base of the effectiveness of motorcycle
rider training in the reduction of crash injury death and offence
rates To this end well designed randomised controlled trials need
to be conducted that examine the impact of rider training oncrashes and offences A randomised controlled trial is the preferred
method as it would significantly reduce most biases described in
this review To minimise the impact of participants delaying or
never attaining their licence and therefore exposure as found in
the randomised trials in this review such a trial should include
riders who are already licensed and reliable measures of riding
exposure in a reasonably lengthy follow up period It is expected
that a randomised controlled trial of sufficient sample size would
observe equal balance on possible confounders by study arm how-
ever an assessment of potentially important confounders (such as
helmet wearing) by study group would enable control for these
variables in the analyses if necessary It is imperative that the effec-
tiveness of each component of the rider training (such as theorypractice knowledge skills and modification of behaviour) and the
synergy of such components are evaluated Outcomes should be
objective ideally those that are routinely collected rather than self-
reported outcomes Even though riding the motorcycle poses dif-
ferent challenges and incurs different risks lessons may be learned
from research into the effects of car driver training
Motorcycles are a common means of transport in low- and middle
income countries and usage is increasing As the burden of mo-
torcycle crash victims is mostly borne in such countries research
should be focused on the effectiveness of types of rider training on
the reduction of crashes injuries and fatalities in such countries
and be supported by higher income countries This review found
only one study conducted in a middle income country that sug-gested decreasing injury rates It would be beneficial to conduct
additional research to confirm this suggestion and to enhance our
knowledge of the role of rider training in low- and middle-income
settings
March 2006 saw the launch of the University of North Carolina
Highway Safety Research Centre partnership with the Motorcycle
Safety Foundation (MSF) and plans were revealed to conduct a
study of the potential benefits of motorcycle rider training The
study is funded by the MSF and the National Highway Traffic Sa-
fety Administration (NHTSA) Called rsquoThe Discovery Projectrsquo it
will compare data on crash rates from riders participating in a basic
23Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2650
motorcycle safety class with riders participating in a basic motor-
cycle safety class as well as a variety of continuing education classesThe collection of pilot data was due to start in March 2007 and
the data collection portion of the study is expected to take three
years (Highway Safety Research Center) It is vital that researchers
and policy makers collaborate to develop furtherhigh quality eval-
uations of motorcycle rider training programmes to help inform
development of new and effective training programmes
A C K N O W L E D G E M E N T S
Staff from The University of Sydney library for helping us locate
and retrieve full reports of many papers
R E F E R E N C E S
References to studies included in this review
Adams 1985 published data only
Adams A Collingwood V Job RFS Evaluation of the 1982 Australia
Post motorcycle rider training scheme NSW Australia Traffic
Accident Research Unit Traffic Authority NSW 1985 [ RN 85]
Anderson 1980 published data only
Anderson J Ford JL Peck RC Improved motorcyclist licensing and testing project Washington DC National Highway Traffic Safety
Administration (NHTSA) USA 19801ndash217 [
DOTndashHSndash805ndash598]
Billheimer 1998 published data only
Billheimer JW Evaluation of the California Motorcyclist Safety
Program Transportation Research Record 19981640100ndash9 [
TRB Report Number 980652]
Cooper 1988 published data only
Cooper PJ Rothe JP Motorcyclists who they are and why they do
what they do Transportation Research Record 1988116878ndash85
Davis 1997 published data only
David CF Evaluation of community traffic safety programs and
motorcycle operator training programs Connecticut Department of Transportation 199763
Hall 1988 published data only
Hall CJ Buchanan L Engle R Motorcycle rider education evaluationproject Washington DC USA National Highway Traffic Safety
Administration (NHTSA) US Department of Transportation
1988 [ DTNH 22ndash80ndashCndash05123]
Haworth 1997 published data only
Haworth N Smith R Brumen I Pronk N In Monash University
Accident Research Centre editor(s) Case-control study of motorcycle
crashes Victoria Australia Department of Transport and Regional
Development The Federal Office of Road Safety 19971ndash149 [
ISBN 0 642 25501 6 ]
Hurt 1981 published data only
Hurt H Ouellet J Thom D Motorcycle accident cause factors and
identification of countermeasures USA Department of
Transportation 1981 [ Technical Report DOT HS 806ndash862]
Jonah 1982 published data onlylowast Jonah BA Dawson N Bragg B Are formally trained motorcyclists
safer Accident Analysis amp Prevention 198214(4)247ndash55
Jonah BA Dawson NE Bragg BW Evaluation of the Canada
Safety Councilrsquos Motorcycle Training Program Proceedings of theInternational Motorcycle Safety Conference Washington DC
USA 198059ndash93
Kloeden 1994 published data only
Kloeden CN Moore VM McLean AJ Evaluation of the pre-license
training program for motorcyclists in South Australia Walkerville
South Australia Australia NHMRC Road Accident Research Unit
The University of Adelaide 19941ndash37 [ ISBN 0 7308 0607 3]
Kraus 1975 published data only
Kraus JF Riggins RS Franti CE Some epidemiological features of
motorcycle collision injuries American Journal of Epidemiology 1975102(1)74ndash98
Leung 1987 published data only
Leung KS Reding VA Evaluation of the Wisconsin motorcycle rider course Wisconsin Bureau of Policy Planning and Analysis
Wisconsin Department of Transportation 1987
MAIDS 2004 published and unpublished data
MAIDS In-depth investigations of accidents involving powered
two wheelers ACEM (Acemacemeu) 2004 Vol Final report
20179
McDavid 1989 published data only
McDavid JC Lohrmann BA Lohrmann G Does motorcycle
training reduce accidents Evidence from a longitudinal quasi-
experimental study Journal of Safety Research 198920(2)61ndash72
Mortimer 1984 published data only
Mortimer R Evaluation of the motorcycle rider course Accident
Analysis amp Prevention 198416(1)63ndash71
24Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2750
Mortimer 1988 published data only
Mortimer RG A further evaluation of the motorcycle rider courseJournal of Safety Research 198819(4)187ndash96
Osga 1980 published data only
Osga GA An investigation of the riding experiences of MSF rider
course participants Dissertation South Dakota USA The
University of South Dakota 19801ndash224
Perrino 2002 published data only
Perrino CS Ahmed A Callender A Rozier E Cantwell A Stewart
O The role of Marylandrsquos motorcycle rider course in promoting saver
behaviors and attitudes Baltimore USA National Transportation
Center Morgan State University 20021ndash52
Raymond 1979 published data only
Raymond S Tatum S An evaluation of the effectiveness of the RAC
ACU motorcycle training scheme - final report Road Safety ResearchUnit Dept of Civil Engineering University of Salford London
19791ndash131
Satten 1980 published data only
Satten RS Analysis and evaluation of the motorcycle rider courses
in thirteen Illinois counties International Motorcycle Safety
Conference Washington DC 1980145ndash93
Savolainen 2007 published and unpublished data
Savolainen Peter Mannering Fred Additional evidence on the
effectiveness of motorcycle training and motorcyclistsrsquo risk-taking
behaviour Transportation Research Board 2007 Vol March 20
19
Swaddiwudhipong 1998 published data only
Swaddiwudhipong W Boonmak C Nguntra P Mahasakpan P
Effect of motorcycle rider education on changes in risk behavioursand motorcycle-related injuries in rural Thailand Tropical Medicine
and International Health 19983(10)767ndash70
Waller 1992 published data only
Waller FL An analysis of the impact of the Texas Department of
Public Safety Motorcycle Operator Training Program upon
motocycle related traffic accidents Texas State University Public
Administration Program Applied Research Projects 199277
References to studies excluded from this review
Anderson 1980a published data only
Anderson LC Better biking a motorcycle rider improvement
program International Motorcycle Safety Conference Washington
DC USA Motorcycle Safety Foundation 1980231ndash53Aupetit 2008 published and unpublished data
Aupetit S Riff J Buttelli O Espie S Ergonomics research program
on motorcycle learning Contributions to the development of a
motorcycle simulator for training Transport Research Arena
Ljubljana 20089
Awane 1999 published data only
Awane T Integrating simulators in motorcycle safety education
Journal of International Association of Traffic and Safety Sciences 199923(1)26ndash35
Baldi 2005 published data only
Baldi S Baer JD Cook AL Identifying best practices states in
motorcycle rider education and licensing Journal of Safety Research200536(1)19ndash32
Batchler 1988 published data only
Batchler MW Evaluation of the pre-license motorcycle training course at the New South Wales traffic education centre Federal Office of
Road Safety (FORS) 19881ndash122 [ Report No CR 77 ]
Blanchard 2006 published data only
Blanchard HT Tabloski PA Motorcycle safety educating riders at
the teachable moment Mosby (CV) Company 200632330ndash2
Bonnett 2005 published data only
Bonnett J Mature age skills training for experienced motorcycle
riders Motorcycle ridersrsquo association of the Australian Capital
Territory 2005
Braver 2007 published data only
Braver ER et alPersuasion and licensure a randomized controlled
intervention trial to increase licensure rates among Maryland
motorcycle owners Traffic Injury Prevention 20078(1)39ndash46Clayton 1980 published data only
Clayton AB Sudlow DE An evaluation of the effectiveness of a one day training programme for learner motorcyclists Berkshire UK
Transport and Road Research Laboratory 19801ndash23 [ ISSN
0266ndash7045]lowast Clayton AB Sudlow DE An evaluation of the effectiveness of a one-day training programme for learner motorcyclists Berkshire UK
Road Safety Division Berkshire 19801ndash23 [ISSN 0266ndash7045]
Council 1978 published data only
Council FM Desper LP Dutt AK Driver education for motorcycle operation University of North Carolina Highway Safety Research
Center 19781ndash40
Crowther 2005 published data onlylowast Crowther G Engaging with motorcyclists UK police and theBikeSafe Road Safety Programme Published proceedings of the
Australasian Road Safety Research Policing Education Conference
Wellington New Zealand 2005 issue ISBN 0ndash473ndash10636ndash11ndash11
Daniello 2009 published data only
Daniello A Gabler HC Mehta YA Effectiveness of motorcycle
training and licensing Transportation Research Board 88th Annual
Meeting 200913
Dewanti 2007 published data only
Dewanti M Challenging efforts in promoting young motorcyclist
safety in Indonesia Road Safety on Four Continents 14th
International Conference Swedish National Road and Transport
Research Institute 2007
Du Plooy 1990 published data onlyDu Plooy DW Motorcycle training in South Africa The Human
Element International Motorcycle Safety Conference Orlando
Florida USA Motorcycle Safety Foundation 19904-1 to 4-9
Engel 1989 published data only
Engel U Krogsgaringrd Thomsen L [Faeligrdselsundervisning og
propaganda ndash en faeligrdselssikkerhedsfremmende foranstaltning]
Dansk Vejtidsskrift 198912295ndash7
Goldenbeld 2002 published data only
Goldenbeld C Houwing S de Craen S The development of
driving skills of young moped riders [De ontwikkeling van de
rijvaardigheid van jonge bromfietsers] SWOV Institute for Road
Safety Research (Stichting Wetenschappelijk Onderzoek
Verkeersveiligheid) Leidschendam the Netherlands 200263
25Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2850
Goldenbeld 2004 published data only
Goldenbeld C Twisk D de Craen S Short and long term effects of moped training a field experiment Transportation Research Part F 20047(1)1ndash16
Haworth 2000 published data only
Haworth N Smith R Kowadlo N Motorcycle training in Victoria
evaluation of rider training curriculum in Victoria Monash
University Accident Research Centre Reports 2000 Vol 165 issue
3131
Haworth 2005 published data only
Haworth N Mulvihill C Review of motorcycle licensing and
training Monash University Accident Research Centre 2005 [
Report no240]
Haworth 2006 published data only
Haworth N Mulvihill C Rowdon P Teaching old dogs new tricksTraining and older motorcyclists Journal of the Australasian College of Road Safety 200718(4)20ndash5lowast Haworth NL Mulvihill C Rowden PJ Teaching old dogs new
tricks Training and older motorcyclists Australasian Road Safety
Research Policing and Education Conference Gold Coast
Queensland 2006
Hill 1983 published data only
Hill D Motorcycle rider training testing and licensing Motor
Cycle Council of New South Wales Sydney 1983
Imatake 1980 published data only
Imatake Y On the activities in Japan of motorcycle safe riding
promotion committee (MSRPC) - especially the education of
unskilled riders of class 1 motorbikes International Motorcycle
Safety Conference Washington DC USA Motorcycle Safety
Foundation 1980255ndash68
Ingebritsen 1990 published and unpublished data
Ingebrigtsen S [Risikofaktorer ved ferdsel med moped og
motorsykkel] TOslashI-rapport 66 Oslo Transportoslashkonomisk
institutt 1990
Jain 2009 published and unpublished data
Jain A et alTwo wheeler accidents on Indian roads a study from
Mangalore India Elsevier 200916130ndash3
Jonah 1979 published data only
Jonah BA Dawson NE Validation of the motorcycle operator skill
test Accident Analysis amp Prevention 197911(3)163ndash71
Jonah 1981 published data onlyJonah BA Dawson NE Bragg WE Predicting accident
involvement with the Motorcycle Operator Skill Test Accident Analysis amp Prevention 198113(4)307ndash418
Kadar 1990 published data only
Kadar KJ Countermeasures motorcycle rider education delivery
system The Human Element International Motorcycle Safety
Conference Motorcycle Safety Foundation 19906-27 to 6-41
Kelsey 1986 published data only
Kelsey SL Liddicoat C Ratz M Licensing novice motorcyclists a
comparison of Californiarsquos standard test and the MOST II (Motorcycle
Operator Skill Test) administered at centralized testing offices Sacramento California USA California Department of Motor
Vehicles Research and Development Office 19861ndash74
Koch 1980 published data only
Koch H Advanced course for motorcycle riders InternationalMotorcycle Safety Conference Washington DC USA Motorcycle
Safety Foundation 1980269ndash81
Koch 1990 published data only
Koch H Brendicke R How to influence risk taking behaviour
within motorcycle rider training programs The Human Element
International Motorcycle Safety Conference Orlando Florida
USA Motorcycle Safety Foundation 1990Motorcycle Safety
Foundation 1990 p 9-1 to 9-25A
Labbett 2005 published data only
Labbett S Langham M Training can make the problem worse
Road Safety Congress 20051ndash12
Law 2005 published and unpublished data
Law TH Radin Umar RS Zulkaurnain S Kulanthayan S Impactof the effect of economic crisis and the targeted motorcycle safety
programme on motorcycle-related accidents injuries and fatalities
in Malaysia International Journal of Injury Control and Safety
Promotion 200512(1)9ndash21
Lowes 1990 published data only
Lowes BF In-traffic rider training The Human Element
International Motorcycle Safety Conference Orlando Florida
USA Motorcycle Safety Foundation 19907-46 to 7-57
Mahon 1976 published data only
Mahon J Teaching accident avoidance on the motorcycle Journal of Traffic Safety Education 1976231ndash29
Maxwell 1985 published data only
Maxwell GS Evaluation of a Program for Training MotorcycleRiders Evaluation and Program Planning 19858(4)339ndash47
Mayhew 1996 published data only
Mayhew DR Simpson SH Effectiveness and role of driver educationand training in a graduated licensing system Ottowa Canada The
Traffic Injury Research Foundation 19961ndash89 [ ISBN
0ndash920071ndash12ndash0]
McKnight 1987 published data only
McKnight AJ Milburn BG An evaluation of the modified
motorcycle operator skill test exercises NHTSA Report 1987 [
Report No DOT HS 807 212 ]
McPherson 1976 published data only
McPherson K McKnight AJ The development and evaluation of a
motorcycle skill test manual and knowledge test Washington DCUSA National Highway Safety Administration 1976
Mohan 2004 published data only
Mohan D Evidence-based interventions for road traffic injuries in
South Asia Journal of the College of Physicians and SurgeonsPakistan 200414(12)746ndash7
Morgan 1980 published data only
Morgan JK Advanced motorcycle training course content and
study evaluation International Motorcycle Safety Conference
Washington DC USA Motorcycle Safety Foundation 1980
283ndash312
Motoki 1990 published data only
Motoki M Yamazaki S A study on effective motorcycle rider
education The Human Element International Motorcycle Safety
26Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2950
Conference Orlando Florida USA Motorcycle Safety
Foundation 19909-26 to 9-57A
Nairn 1992 published data onlylowast Nairn RJ Motorcycle safety research literature review 1987 to
1991 Federal Office of Road Safety Australia 1993 Vol Report
CR 117
Newland 1980 published data only
Newland R Darby J Automotive practicespre motorcycle rider
program International Motorcycle Safety Conference Washington
DC USA Motorcycle Safety Foundation 1980127ndash9
Newland 1982 published data only
Newland R Motorcycle rider training and licensing requirements
Driver training steering a course for the future Melbourne
Victoria Transport Regulation Board 1982
Newland 1990 published data only
Newland R Motorcycle training and licensing in Victoria 1980-
1990 The Human Element International Motorcycle Safety
Conference Orlando Florida USA Motorcycle Safety
Foundation 1990
Ormston 2003 published data only
Ormston R Dudleston A Pearson S Stradling Evaluation of
Bikesafe Scotland Scottish Executive Social Research Edinburgh
2003
Osga 1979 published data only
Osga GS Ellingstad VS An evaluation of the effectiveness of the 1978
South Dakota motorcycle rider course program South Dakota USA
University of South Dakota 19791ndash89
Panou 2007 published data only
Panou M Bekiaris E A new concept on the integration of driving
simulators in driving training the train-all approach Road Safety
on Four Continents 14th International Conference Swedish
National Road and Transport Research Institute 2007
Peterson 1990 published data only
Peterson A Kadar K Status report 1980-1989 Minnesota rider
education statistics The Human Element International
Motorcycle Safety Conference Orlando Florida USA 199013-
113 to 13-129
Prem 1984 published data only
Prem H Good MC Motorcycle rider skills assessment ACT
Australia Office of Road Safety Department of Transport 1984
1ndash377 [ Report No CR 34 ]
Radin Umar 2006 published data only
Radin Umar RS Motorcycle safety programmes in Malaysia how
effective are they International Journal of Injury Control and Safety Promotion 200613(2)71ndash9
Rockwell 1990 published data only
Rockwell TH Kiger S Carnot M An evaluation of the Ohio
motorcyclist enrichment program Ohio Department of Highway
Safety 1990
Rothe 1987 published data only
Rothe JP A motorcycle riding course Motorcyclists Image and reality Canada Insurance Corporation of British Columbia 1987
59ndash83
Rowden 2007 published data only
Rowden P Watson B Haworth N What can riders tell us aboutmotorcycle training A view from the other side of the fence
Australasian Road Safety Research Policing and Education
Conference Melbourne Australia Melbourne 2007
RSC Victoria 1993 published data only
Road Safety Committee Inquiry into Motorcycle Safety in Victoria
Parliament of Victoria 1993
Russam 1979 published data only
Russam K Motorcycle training and licensing in Japan UK Transport
and Road Research Laboratory (TRRL) 1979 [ Report 916 ]
Saffron 1981 published data only
Saffron DG Driver and rider education training and licensing a
brief review Traffic Accident Research Unit Traffic Authority of
New South Wales NSW Australia Traffic Accident Research UnitTraffic Authority of New South Wales 1981
Schulz 1990 published data only
Schulz U Motorcycle accidents among learner riders Motorcycle
licensing Evaluation of the effectiveness of a Graduated Licensing Scheme Forshungshefte des Institut fur Zweiradsicherheit 1990
Sexton 2004 published data only (unpublished sought but not used)
Sexton B Baughan C Elliott M Maycock G The accident risk of
motorcyclists 2004TRL Report TRL 607(ISSN 09684107)58
Shephard 1986 published data only
Shepherd R Do Motorcycle Safety Foundation programs work
Driver 198619(9)16ndash8
Simpson 1990 published data onlySimpson HM Mayhew DR The promotion of motorcycle safety
training education and awareness Health Education Research
Theory and Practice 19905(2)257ndash64
Sokolov 1990 published data only
Sokolov O Safe motorcycle riding some aspects of motorcycle
riding in the USSR The Human Element International
Motorcycle Safety Conference Motorcycle Safety Foundation
19904ndash37 - 4-48
Steffens 1988 published data only
Steffens U Gawatz R Willmes G Effectiveness of mofa courses
[Wirksamkeit von Mofakursen] Unfall- und sicherheitsforschung
strasenverkehr 198867168
Sudlow 2003 published data only
Sudlow D Scoping study on motorcycle training Road Safety
Research Report London UK The British Institute of Traffic
Education Research Department for Transport 2003 Vol June
[ Report No 36]
Tasmania 1995 published data only
Mannering FL Grodsky LL Motorcycle rider training scheme
Traffic Education Research Centre Armidale Australia Driver
Training Australia Limited 1995 Vol April1ndash56
Tetard 1985 published data only
Tetard C Motorcycle ridersrsquo training to detect indices in an
intersection situation [Lrsquoentrainement a la prise drsquoinformation dans
lrsquoapprentissage de la conduite motocycliste] Le Travail Humain198548(2)147ndash59
27Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3050
Thackray 1980 published data only
Thackray RM Prescott JC Field test of the motorcycle ridercourse International Motorcycle Safety Conference Washington
DC USA Motorcycle Safety Foundation 1980195ndash7
Thompson 1994 published data only
Thompson M Evaluation of compulsory basic training for
motorcyclists Transport Research Laboratory Crowthorne
Berkshire UK 1994 [ Report 63]
Tobler 1990 published data only
Tobler U Individual driving instruction for motorcycle driving
learners the Swiss experiment of motorcycle driving school The
Human Element International Motorcycle Safety Conference
Orlando Florida USA Motorcycle Safety Foundation 1990
Transport 1989 published data only
Transport Communications Road Crash Statistics Australia Bureauof Transport and Communications Economics 198914ndash20
Tuan 2007 published data only
Tuan VA Shimizu T Towards development and evaluation of the
motorcycle drivers re-education program in Vietnam modeling of
motorcycle driverrsquos undesired behaviors 11th World Conference
on Transport Research 200725
Victor 1980 published data only
Victor DJ Traffic safety education for motorcyclists International
Motorcycle Safety Conference Washington DC USA Motorcycle
Safety Foundation 1980221ndash8
Waterer 1990 published data only
Waterer TH Compulsory training for motorcyclists from liberty
to compulsion The Human Element International Motorcycle
Safety Conference Orlando Florida USA Motorcycle Safety
Foundation 19904-49 to 4-57
Weller 1989 published data only
Weller RB Chandler EW Motorcycle safety and motorcycle
education past research and survey results Journal of Public Policy
amp Marketing 1989893ndash108
Wells 1981 published data only
Wells P Observations of some RACACU motorcycle training
schemes Transport and Road Research Laboratory Crowthorne
Berkshire UK 1981
Winn 1986 published data only
Winn G Two important trends in motorcycle safety regulations
rider education and conspicuity improvement American
Motorcyclist Association 1986147ndash52
Wisher 1988 published data only
Wisher G Reid J Motorcycle accidents in Norfolk a three-year study of motorcycle training and the Part 1 test Norfolk UK Road Safety
Division Highways Department Norfolk County Council 1988
Wood 1987 published data only
Wood HT Bowen R Evaluation of the revised learner permit scheme
July 1983 to December 1985 Victoria Australia Road Traffic
Authority 19871ndash27 [ ISBN 0 7306 0247 8]
Yamazaki 1990 published data only
Yamazaki S Nakanishi M Survey on effect of Honda rider
education The Human Element International Motorcycle Safety
Conference Orlando Florida USA 199012-167 to 12-187
References to ongoing studies
Burgess 2010 unpublished data only
Burgess C Broughton P Fylan F Stradling Interim evaluation of
the UKs National Ride program In L Dorn editor(s) Driver
behaviour and Training Vol IV Aldershot Ashgate Publishing
Limited 2010
Additional references
Anderson 1980
Anderson J Ford JL Peck RC Improved motorcyclist licensing and
testing project Volume I and II National Highway Traffic Safety
Administration US department of Transportation Washington
DC 1980
Bjornstig 1985
Bjornstig U Bylund P Lekander T Brorsson B Motorcycle fatalitiesin Sweden Acta Chirurgica Scandinavica 1985151(7)577ndash81
Clayton 1990
Clayton A Sudlow DE An evaluation of the effectiveness of a one-
day training program for learner motorcyclists The Human
Element International Motorcycle Safety Conference Orlando
Florida USA 199013-105 to 13-111
Cochrane Collaboration 2008
Higgins JP Green S Cochrane Handbook for Systematic Reviews of
Interventions 500 The Cochrane Collaboration 2008
Elliot 2003
Elliott MA Baughan CJT Broughton J Grayson G Knowles J et
alMotorcycle Safety a Scoping Study TRL [ Report No
0968ndash4107]Fleiss 2004
Fleiss JL Levin B Paik MC Statistical methods for rates and proportions 3rd Edition John Wiley amp Sons Inc 2004 [ ISBN
978ndash0ndash471ndash52629ndash2 ]
FORS 1999
FORS Federal Office of Road Safety Road risk for sober licensed
motorcyclists Vol Monograph 27 Federal Office of Road Safety
19991ndash2
Haworth 2005
Haworth N Mulvihill C Review of motorcycle licensing and training
Victoria Australia Monash University Accident Research Centre
20051ndash83 [ Report No 240]
Huwiler-Muumlntener 2002Huwiler-Muumlntener K Juumlni P Junker C Egger M Quality of
Reporting of Randomized Trials as a Measure of Methodologic
Quality JAMA 20022872801ndash4
Johnston 2008
Johnston P Brooks C Savage H Fatal and serious road crashes
involving motorcyclists Road Safety Research and Analysis Report
ATSB Monograph 20 Canberra Department of Infrastructure
Transport Regional Development and Local Government 0205
081ndash26
Juumlni 2001
Juumlni P Altman DG Egger M Systematic reviews in health care
Assessing the quality of controlled clinical trials BMJ 2001323
42ndash6
28Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3150
Kopits 2003
Kopits E Cropper M Traffic fatalities and economic growth TheWorld Bank Development Research Group Infrastructure and
Environment Policy research working paper 3035 2003
Leung 1983
Leung P Chau BTW Motorcycle accidents in Hong Kong pattern
of injury treatment causative factors and socio-economic
implications Injury 198315(1)23ndash9
Liu 2004
Liu B Ivers R Norton R Blows S Lo SK Helmets for preventing
injury in motorcycle riders Cochrane Database of Systematic Reviews 2004 Issue 2 [DOI 10100214651858CD004333pub3]
Mohan 2002
Mohan D Road safety in less-motorized environments future
concerns International Journal of Epidemiology 200231(3)527ndash32Mortimer 1982
Mortimer RG An evaluation of the motorcycle rider course
Proceedings of the 20th conference of the American Association for
Automotive Medicine Arlington Heights IL USA American
Association for Automotive Medicine 198277ndash96
Mullin 1997
Mullin BT Risk factors for motorcycle injury the role of age
gender experience training and alcohol PhD thesis Auckland
New Zealand University of Auckland 1997
Noordzij 2001
Noordzij PC Forke E Brendicke R Chinn BP Integration of needs
of moped and motorcycle riders into safety measures review and
statistical analysis in the framework of the European research
project PROMISING SWOV Institute for Road Safety Research
Leidschendam The Netherlands 2001 [ Report No Dndash2001ndash5]
Osga 1979
Osga G Ellingstad VS An evaluation of the effectiveness of the 1978 South Dakota Motorcycle Riders Course Program South Dakota
University of South Dakota Vermillion Human Factors Lab 1979
Oxford English Dictionary Online
httpoxforddictionariescom Oxford English Dictionary Online
Oxford University Press 2010
Peden 2004
Peden M Scurfield R Sleet D World Report on Road TrafficInjury Prevention Geneva World Health Organization Available
at httpwwwwhointviolence˙injury˙preventionpublications
road˙trafficworld˙reporten 2004Chapter 3
Radin Umar 1995
Radin Umar R Macakay G Hills B Preliminary analysis of
motorcycle accidents Short-term impacts of the running
headlights campaign and regulations in Malaysia J Traffic Med 199523(1)17ndash27
Tsai 1995
Tsai YJ Wang JD et alCase-control study of the effectiveness of
different types of helmets for the prevention of head injuries among
motorcycle riders in Taipei Taiwan American Journal of
Epidemiology 1995142(9)974ndash81
Vis 1995
Vis AA In-depth study of the hazards of motorcycling [De
onveiligheid van motorrijden nader bekeken] SWOV Institute for
Road Safety Research Leidschendam the Netherlands 1995 [
Report No Rndash95ndash69]
Weaver 1990
Weaver EA Motorcycle safety education in the United States
Motorcycle Safety Foundation Orlando Florida USA 1990
Winn 1987
Winn G Two important trends in motorcycle safety regulations
rider education and conspicuity improvement No more
information available Detroit MI US Society of Automotive
Engineers 1987147ndash52
Yuan 2000
Yuan W The effectiveness of the rsquoride-brightrsquo legislation for
motorcycles in Singapore Accident Analysis amp Prevention 200032
(4)559ndash63
Zambon 2006
Zambon F Hasselberg M Factors affecting the severity of injuries
among young motorcyclists a Swedish nationwide cohort study
Traffic Injury Prevention 20067(2)143ndash9lowast Indicates the major publication for the study
29Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3250
C H A R A C T E R I S T I C S O F S T U D I E S
Characteristics of included studies [ordered by study ID]
Adams 1985
Methods Non-randomised trial
Participants Initial sample size
Total of 162 Australia Post telegram delivery riders
Not clear how many in intervention and control groups
Analysis sample size
Intervention group 63Control group 35
Total 98 (61)
Interventions No formal training vs voluntary rider training course of 12 hours over 2 consecutive days (theory 4 hours practice
8 hours)
Intervention integrity
No information given
Outcomes Australia Post recorded crashes
Length of follow up 1 year
Notes Study conducted in NSW Australia
Anderson 1980
Methods Randomised trial
Participants Initial sample size
Total 41188 motorcycle licence applicants
Analysis sample size
Intervention group with skills training 13160
Intervention group without skills training 12634
Control group 15080
Total 40874 (99)
Interventions Standard licensing skill test vs the Motorcycle Operators Skill Test (MOST) with or without a training component
MOST is not to be confused with rider training because it is a licensing test The training component was for
participants who failed the skill test on the first attempt was of 3 hours duration and consisted of instructions about
the skills needed to pass the skill test The training consisted of 30 min theory and 2 frac12 hours practice on the
motorcycle
Intervention integrity
33 in MOST with skills training 31 in MOST without skills training and 43 in control group obtained their
motorcycle licence in the end
Outcomes Police recorded crashes and offences
Length of follow up 3 years
30Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3350
Anderson 1980 (Continued)
Notes Study conducted in California USA
Billheimer 1998
Methods Cohort study
Participants Initial sample size
Intervention group Unclear numbers of CMSP graduates and ERC graduates
Control group Over 16000 untrained motorcyclists
Analysis sample size
1139 matched CMSP-pairs and 1182 ERC-pairs
Interventions No formal training vs the mandatory (for those younger than 18 years) pre-licence 16-hours basic California
Motorcyclist Safety Program (CMSP) and the 8-hour Experienced Rider Course (ERC)
Intervention integrity
No information given
Outcomes Police recorded crashes and offences
Length of follow up 2 years
Notes Study conducted in California USA
Cooper 1988
Methods Cohort study
Participants Initial sample size
Intervention group 418 graduates and 57 who failed
Control group 402
Analysis sample size
Intervention group unclear
Control group unclear
Total 863 (984)
Interventions No formaltrainingvs non-mandatory pre-licence rider trainingThe content of thetrainingwas notfurther explainedIntervention integrity
Participants had all completed the rider training
Outcomes Police recorded crashes and offences
Length of follow up 7 years
Notes Study conducted in British Columbia Canada
31Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3450
Davis 1997
Methods Retrospective cohort study
Participants Initial sample size
Intervention group 9320 graduates
Control group unclear
Analysis sample size
Intervention group 9320 graduates
Control group unclear
Interventions No formal training vs the non-mandatory Connecticut Motorcycle Operator Training Program The content of
training was not further explained
Intervention integrity Participants had all completed the rider training
Outcomes Police recorded crashes severity of crashes
Length of follow up 1 year
Notes Study conducted in Connecticut USA
Hall 1988
Methods Randomised trial
Participants Initial sample sizeIntervention group A 6248
Intervention group B 6616
Intervention group C 6996
Control group 6604
Total 26464 motorcycle licence applicants
Analysis sample size
Intervention group A 6012 (96)
Intervention group B 6428 (97)
Intervention group C 6778 (97)
Control group 6502 (98)
Interventions Standard licensing skill test vs the Motorcycle Operator Skill Test (MOST II) without training (A) MOST II with
3 hours training (B) or MOST II with 20 hours training (C) The 3 hours training comprised instructions aboutthe skills needed to pass the skill test (30 min theory and 2 frac12 hours practice on the motorcycle) The 20 hours
training consisted of the Motorcycle Safety Foundationrsquos Motorcycle Rider Course (MRC) comprising of 8 hours
theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-road practice
Intervention integrity
25 and 26 obtained their motorcycle licence in the lsquoMOST 3-hour trainingrsquo and lsquoMOST 20-hour trainingrsquo
intervention arm respectively In the lsquoMOST testrsquo intervention arm 29 obtained their licence and in the control
group 35
Outcomes Police recorded crashes and offences
Length of follow up 3 years
32Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3550
Hall 1988 (Continued)
Notes Study conducted in New York USA
Haworth 1997
Methods Population based case-control study
Participants Cases Motorcycle riders or pillion riders involved in motorcycle crash (on road) Recruited from hospital and police
records Not clear whether all cases (especially fatalities) were included
Community controls Multiple matched sample of motorcycle rider or pillion who passed the crash site at the same
time of day and day of week that the crash occurred
Initial sample size
Cases 336
Controls 1195
Analysis sample size
Cases 234 (70)
Controls 561 (47)
Interventions Basic handling skills course beginner intermediate or advanced course measured through self-report not validated
Outcomes Injuries with an Injury Severity Score of 5 or greater or admission for at least four hours and fatalities from a
motorcycle crash (from hospital and police records)
Length of follow up two years or more prior to the interview
Notes Study conducted in Victoria Australia
Hurt 1981
Methods Population based case-control study
Participants Cases Motorcycle riders who had a motorcycle crash that was reported to police (traffic accident reports) Not all
cases included
Community controls Multiple matched sample of motorcycle riders who passed the crash site at the same time of
day and day of week as the crash occurred but 2 years later
Initial sample size
Cases 4500 notifications of crashes to police (only 50 received in time research team able to respond to 1126 for
unreported reasons data collection completed for 900)
Controls Number approached unreported 2310 consented and interviewed
Analysis sample size
Cases 808 (808900 data on rider training) - 18 of initial sample
Controls 670 (6702310 data on rider training) - lt 29 of initial sample
Interventions Training experience self taught friends-family school-club formal-professional others measured through self-
report validated
Outcomes Injuries and fatalities from a motorcycle crash (from hospital records and coronerrsquos office)
33Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3650
Hurt 1981 (Continued)
Notes Study conducted in Los Angeles USA
Jonah 1982
Methods Cohort study
Participants Initial sample size
Intervention group
2310 licensed MTP graduates who are active riders
Control group
2029 licensed motorcyclists who are active riders
Non-licensed and non-active riders excluded
Analysis sample size
Intervention group
811 interviews (35) 533 driver records (23)
Control group 1080 interviews (71) 905 driver records (45)
Interventions No training vs the non-mandatory Canada Safety Council Motorcycle Training Program (MTP) comprising of 4
hours theory and 16 hours practise (6 hrs on-road) over 2 weekends
Intervention integrity
Participants had all successfully completed the MTP
Outcomes Self-reported and police reported crashes and offencesLength of follow up 4 years
Notes Study conducted in Ontario Canada
Kloeden 1994
Methods Cohort study
Participants Initial sample size
Intervention group 2533 licence applicants
Control group Unclear
Sample of riders obtaining probationary licenceIntervention group 1625 (64)
Control group 5015 (unclear)
Non-active riders excluded from control group (~66)
Analysis sample size
Intervention group 631 (39)
Control group 318 (6)
Interventions No formal training vs the mandatory pre-licence Ridersafe course consisting of two levelsthe first is a training course
in the basics of motorcycle riding where a pass entitles the rider to obtain a learnerrsquos permit The second is a skill
test where a pass entitles the rider to obtain a probationary motorcycle licence The training course consists of two 4
hour training sessions on separate days involving instruction videos and actual riding on motorcycles The skill test
consists of a 4 hour preparation course of 74 minutes of lecturediscussion 38 minutes of video and 75 minutes of
34Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3750
Kloeden 1994 (Continued)
on-course riding which is the testing phase
Intervention integrity
Unclear what percentage in the Ridersafe group and the control group obtained their probationary licence halfway
and their full motorcycle licence in the end
Outcomes Police reported crashes
Length of follow up 1 year
Notes Study conducted in SA Australia
Kraus 1975
Methods Population based case-control study
Participants Cases All motorcycle riders with a confirmed medically treated motorcycle injury (including fatalities) from all
hospitals in study base and peripheral area hospitals and all deaths due to motorcycle crashes in the study base All
serious injuries (leading to hospitalisation) and fatalities likely to be included
Community controls A random sample of registered motorcycle owners in the same temporary and geographical
study base as the cases arose from
Initial sample size Cases 1273
Controls 738
Analysis sample size Cases 741 (58)
Controls 434 (59)
Interventions Motorcycle rider training (not specified) measured through self-report
Outcomes Injuries leading to hospitalisation fatalities
Length of follow up not specified
Notes Study conducted in Sacramento county USA
Leung 1987
Methods Non-randomised trial
Participants Initial sample sizeIntervention group unknown number of MRC graduates
Control group unknown number of licensed motorcyclists
Analysis sample size
Intervention group 2941
Control group 43094
Interventions No formal training vs non-mandatory Wisconsin motorcycle rider course a 22 hours off-street programme that
teaches the basic rules of safe motorcycling
Intervention integrity
50 in the MRC group and 100 in the control group obtained their motorcycle licence in the end
35Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3850
Leung 1987 (Continued)
Outcomes Police recorded crashes and offences
Length of follow up up to 3 years
Notes Study conducted in Wisconsin USA
MAIDS 2004
Methods Population based case-control study
Participants Cases Motorcycle riders who had a motorcycle crash Unclear whether all cases included
Community controls Motorcycle riders identified at petrol stations Unclear whether matched to cases
Initial sample size
Cases not reported
Controls not reported
Analysis sample size
Cases 921 (Italy 200 Netherlands 200 Spain 121 Germany 250 France 150)
Controls 923 (Italy 200 Netherlands 200 Spain123 Germany 250 France 150)
Interventions Pre-license rider training or additional training (non-compulsory)
Outcomes Crashes
Notes Study in five sampling areas located in five European countries (France Germany Netherlands Spain and Italy)
McDavid 1989
Methods Cohort study
Participants Sample size after first match
Intervention group 346 male training graduates
Control group 346 male motorcycle riders licensed in same year
Sample size after second match
Intervention group 139 (40)
Control group 139 (40)
Analysis sample size
Intervention group 139 (40)
Control group 139 (40)
Interventions No formal training vs the non-mandatory pre-licence British Columbia Safety Councilrsquos 37-hour motorcycle safety
programme
Intervention integrity
Participants had all completed and passed the training
Outcomes Police reported crashes injurious crashes
Length of follow up 1-6 years
36Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3950
McDavid 1989 (Continued)
Notes Study conducted in British Columbia Canada
Mortimer 1984
Methods Cohort study
Participants Initial sample size
Intervention group 608 MRC graduates
Control group 360 motorcycle riders
Response rates
Intervention group 360 (59)
Control group 324 (90)
Non-active riders excluded
Analysis sample size
Intervention group 213 (35)
Control group 303 (84)
Interventions No formal training vs the non-mandatory Motorcycle Safety Foundationrsquos Motorcycle Rider Course comprising of
8 hours theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-road practice
Intervention integrity
Participants had all completed and passed the MRC
Outcomes Self-reported crashes and offencesLength of follow up 1 year
Notes Study conducted in Illinois USA
Mortimer 1988
Methods Cohort study
Participants Initial sample size
Intervention group 2307 MRC graduates
Control group 556 active licensed motorcycle riders
Response ratesIntervention group 1298 (56)
Control group 500 (90)
Non-active riders excluded
Analysis sample size
Intervention group 913 (40)
Control group 500 (90)
Interventions No formal training vs the non-mandatory Motorcycle Safety Foundationrsquos Motorcycle Rider Course comprising of
8 hours theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-road practice
Intervention integrity
Participants had all completed and passed the MRC
37Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4050
Mortimer 1988 (Continued)
Outcomes Self reported crashes and offences
Length of follow up 1 year
Notes Study conducted in Illinois USA
Osga 1980
Methods Cohort study
Participants Initial sample sizeIntervention group 1163 MSF MRC participants invited 818 (74) responded
Control group 329 motorcycle riders responded to recruitment strategies
Non-active riders excluded from analysis
Analysis sample size
Intervention group 595 (73)
Control group 329 (100)
Interventions No formal training vs the non-mandatory Motorcycle Safety Foundationrsquos Motorcycle Rider Course (MRC) com-
prising of 8 hours theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-road
practice
Intervention integrity
Participants had all successfully completed the MRC
Outcomes Self-reported crashes police recorded crash reports
Length of follow up 6 months - 3 12 years
Notes Study conducted in South Dakota USA
Perrino 2002
Methods Cohort study
Participants Initial sample size
Intervention group 2144 (unclear) MRC graduates (motorcycle licence applicants)
Control group 2601 (unclear) regular licensing test graduates (motorcycle licence applicants)Analysis sample size
Intervention group 1070 (unclear)
Control group 867 (unclear)
Interventions No formal training vs the mandatory (for those younger than 18 years) pre-licence Maryland Motorcycle Rider
Course (MRC) consisting of 8 hours theory and 12 hours practice
Intervention integrity
No information given
Outcomes Self-reported and police recorded crashes and offences
Length of follow up 1 year
38Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4150
Perrino 2002 (Continued)
Notes Study conducted in Maryland USA
Raymond 1979
Methods Cohort study
Participants Initial sample size
Intervention group
258 volunteers mainly (but not all) novice riders
Control group
1104 provisional licence applicants
Analysis sample size
Intervention group 190 (74)
Control group 627 (57)
Interventions No formal training vs the non-mandatory RACACU (Royal Automobile ClubAuto-Cycle Union) training scheme
The training program consisted of either 24 one-hour sessions or 13 two-hour sessions Twelve hours of the training
focused on theory about basic operational procedures traffic laws and highway code The practical session focused
on basic operational skills riding on a training area and riding in traffic with an instructor A proficiency test was
administered at the end of the course
Intervention integrity
No information given
Outcomes Self-reported crashes (including minor non-injury non-damage and off-road crashes)
Length of follow up 1 year
Notes Study conducted in the UK
Satten 1980
Methods Cohort study
Participants Initial sample size
Intervention group interviewers kept interviewing until they had 100 riders selected from a population of 891
enrollees in the MSF MRC course (unknown how many were asked and how many refused to participate)Control group unknown number of registered motorcycle riders
Non-active riders excluded from analysis
Analysis sample size
Intervention group 71
Control group 100
Interventions No formal training vs the non-mandatory Motorcycle Safety Foundationrsquos (MSF) Motorcycle Rider Course (MRC)
comprising of 8 hours theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-
road practice
Intervention integrity
Participants had all successfully completed the MRC
39Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4250
Satten 1980 (Continued)
Outcomes Self-reported crashes injuries and offences
Length of follow up 3 years
Notes Study conducted in Illinois USA
Savolainen 2007
Methods Cohort study
Participants Initial sample size
Intervention group 4000 riders that had conducted the MSF MRC course (passed or failed)
Control group 4000 motorcycle owners or permit or endorsement holders from the Indiana Bureau of Motor
Vehicles database
Analysis sample size
Intervention group 739 (18)
Control group 588 (15)
Interventions Intervention integrity
No information given
Outcomes Probability of crash involvement
Length of follow up up to 5 years
Notes Study conducted in Indiana USA
Swaddiwudhipong 1998
Methods Randomised trial
Participants Initial sample size
Intervention group
all 16 villages of 3 sub-districts
(participants were practising motorcyclists licensed and unlicensed)
Control group
18 villages of 3 other sub-districtsAnalysis sample size
Intervention group 3 districts (100)
Control group 3 districts (100)
Interventions No rider education vs non-mandatory community based rider education consisting of theory on motorcycle injury
risk helmet use traffic laws vehicle regulations and traffic signs and a written and skill test leading to a licence
Intervention integrity
No information given
Outcomes Injuries fatalities
Length of follow up 2 years
40Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4350
Swaddiwudhipong 1998 (Continued)
Notes Study conducted in Tak Province rural Thailand
Waller 1992
Methods Cohort study
Participants Initial sample size
Intervention group 349 riders who succesfully completed the Department of Public Safety motorcycle operator
training in 1988
Control group 349 motorcyclists who never completed the DPS motorcycle operator training
Analysis sample size
Intervention group 349
Control group 349
Interventions No DPS motorcycle operator training vs non-mandatory DPS motorcycle operator training The DPS training is
the Motorcycle Safety Foundationrsquos Motorcycle Rider Course comprising eight hours of theory (lectures discussions
films and multiple choice tests) a textbook and 12 hours off-road practice
Intervention integrity
All 349 riders succesfully completed the rider training
Outcomes Police recorded crashes
Length of follow up 3 years
Notes Study conducted in
Characteristics of excluded studies [ordered by study ID]
Study Reason for exclusion
Anderson 1980a Study type review and process evaluation of a rider course
Aupetit 2008 Outcome proposal of a rider training curriculum that combines traditional training and a motorcycle riding
simulator
Awane 1999 Study type review on use of simulators in rider and driver training
Baldi 2005 Intervention not rider training but intervention integrity of rider training availabilty of rider education
implementation of graduated licensing system etc on a state-wide level
Batchler 1988 Study type description of a training course
Blanchard 2006 Study type not a study assessing the association between rider training and outcomes
Bonnett 2005 Outcome qualitative data regarding satisfaction with the training programme
41Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4450
(Continued)
Braver 2007 Outcome licensure rates
Clayton 1980 Outcome knowledge skills attitude
Council 1978 Outcome crash reduction in ANY vehicle not motorcycles exclusively
Crowther 2005 Study type qualitative study
Daniello 2009 Study type review of the effectiveness of motorcycle training and licensing
Dewanti 2007 Intervention emotional characteristics behaviour (ie not motocycle rider training)
Du Plooy 1990 Study type review of types of rider training courses in South Africa
Engel 1989 Study type uncontrolled before-after study
Goldenbeld 2002 Outcome skills
Goldenbeld 2004 Outcome knowledge and skills
Haworth 2000 Intervention balance between attitudinal andvehicle control skills acrosscurriculla of motorcycle rider training
courses
Haworth 2005 Study type review
Haworth 2006 Study type cross-sectional survey
Hill 1983 Study type review and description of training system
Imatake 1980 Study type course descriptionreview
Ingebritsen 1990 Intervention risk factors for motorcycle crashes
Jain 2009 Study type cross-sectional descriptive study of motorcycle crashes over a 5 year period
Jonah 1979 Study type cross-sectional survey Outcome scores on motorcycle operator skill test
Jonah 1981 Intervention evaluation of a motorcycle rider skill test not motorcycle rider training
Kadar 1990 Study type review of components of rider training courses
Kelsey 1986 Intervention evaluation of a motorcycle rider skill test not motorcycle rider training
Koch 1980 Study type reviewcourse description
Koch 1990 Study type description of course content
42Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4550
(Continued)
Labbett 2005 Study type essay about possible explanations for motorcycle deaths
Law 2005 Intervention targeted motorcycle safety programme not specifically rider training
Lowes 1990 Study type review and descriptions of rider training
Mahon 1976 Study type review of motorcycle rider training methods
Maxwell 1985 Outcomesrecommendations forprogramme operatorsand otherson howthe motorcycle trainingprogramme
works and how it can be improved
Mayhew 1996 Study type review
McKnight 1987 Study type evaluation of methodological and administrative aspects of a motorcycle operator skill test
McPherson 1976 Outcome knowledge and skills
Mohan 2004 Study type review of evidence of interventions to reduce road traffic injuries
Morgan 1980 Study type reviewcourse description
Motoki 1990 Study type proposal and development of rider training curriculum
Nairn 1992 Study type literature review
Newland 1980 Study type description of development of motorcycle safety course
Newland 1982 Study type descriptionreview of a rider training course
Newland 1990 Study type uncontrolled BeforeAfter study
Ormston 2003 Outcome changes in attitude and self reported behaviour
Osga 1979 Outcome knowledge attitudes and skills
Panou 2007 Intervention new simulation tool to teach driverrider
Peterson 1990 Study factor demographic details of course participants completing the course vs those not completing the
course
Prem 1984 Study type experimental
Study factor examines association between skills and performance in tests
Radin Umar 2006 Intervention motorcycle safety programme not specifically rider training
Rockwell 1990 Study type methodologically flawed study type
43Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4650
(Continued)
Rothe 1987 Study type qualitative study of rider training in general
Rowden 2007 Study type qualitative study
RSC Victoria 1993 Study type review of motorcycle rider training and education in Victoria
Russam 1979 Study type review of training programmes in Japan
Saffron 1981 Study type literature review
Schulz 1990 Study factor predictors of motorcycle crashes
Sexton 2004 Study type cross-sectional analytic analysis
Shephard 1986 Study factor reviewdescription of programme
Simpson 1990 Study type review of motorcycle rider trainingeducation
Sokolov 1990 Study factor review of motorcycle riding courses in USSR
Steffens 1988 Outcome self-reported behaviour experiences and researcher observed riding behaviour
Sudlow 2003 Study type review of motorcycle training in the UK
Tasmania 1995 Study type cross-sectional descriptiveprocess evaluation
Tetard 1985 Study type experimental
Outcome detection of indices
Thackray 1980 Outcome instructional effectiveness user acceptance administrative feasibility
Thompson 1994 Study type cross-sectional survey
Outcome attitudes and knowledge
Tobler 1990 Study type review of course curricula
Transport 1989 Intervention syllabus for rider training
Tuan 2007 Study type cross-sectional study (survey) which does not assess the effects of an intervention but examines
motocyclistsrsquo behaviour
Victor 1980 Study type review
Waterer 1990 Study type review of UK rider training issues
Weller 1989 Study type review of studies evaluating motorcycle safety and education measures
44Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4750
(Continued)
Wells 1981 Study type description of different methods of training
Winn 1986 Study type review of rsquohistoryrsquo of rider training
Wisher 1988 Study type cross-sectional survey
Wood 1987 Study type Interrupted Time Series but did not meet the criteria for an ITS as published by the Cochrane
Effective Practice and Organisation of Care Group
Yamazaki 1990 Study type cross-sectional survey
Outcome riding skills skills retention and attitude
Characteristics of ongoing studies [ordered by study ID]
Burgess 2010
Trial name or title UKs National Ride program
Methods Cohort study
Participants Motorcycle offenders
Interventions Police led diversion program
Outcomes Self reported attitudinal and behavioural measures
Starting date
Contact information
Notes Expected to be completed in early 2010
45Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4850
D A T A A N D A N A L Y S E S
This review has no analyses
A P P E N D I C E S
Appendix 1 Search strategy
We searched the following electronic databases
bull Cochrane Injuries Group Specialised Register (searched 22 Sept 2008)
bull CENTRAL (The Cochrane Library 2008 Issue 3)
bull TRANSPORT 1988 to 200706 (latest date available)
bull ATRI Australian Transport Index (to Sept 2008)
bull MEDLINE (Ovid SP) 1950 to Sept 2008
bull EMBASE (Ovid SP) 1980 to Sept 2008
bull CINAHL (EBSCO) 1982 to Sept 2008
bull WHOLIS (World Health Organization Library Information System) (to Sept 2008)
bull PsycINFO (Ovid SP) 1806 to Sept 2008
bull LILACS (BIREME) (Latin American and Caribbean Health Sciences) (1982 to Sept 2008)
bull ISI Web of Science Social Sciences Citation Index (SSCI) 1970 to Sept 2008
bull ERIC (searched 25 Sept 2008)
bull ZETOC (searched 25 Sept 2008)
bull SIGLE (searched 25 Sept 2008)
All of the databases were searched using the following text-word search strategy The strategy was adapted using either rsquorsquo or rsquo$rsquo as atruncation symbol
1 motor-cycl or motorcycl or motor-bik or motorbik or motor-scooter or motorscooter or scooter or moped or motocycl
or motocicl or moto
2 train or educat or instruct or program or course or qualif or learn or improv
3 1 and 2
H I S T O R Y
Protocol first published Issue 2 2005
Review first published Issue 10 2010
46Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4950
C O N T R I B U T I O N S O F A U T H O R S
Katina Kardamanidis was the first author of the protocol conducted the literature search extracted full studies assessed the studies
according to inclusion criteria assessed quality of the studies compared findings with the second author discussed disagreements with
the third author wrote the drafts and the final review
Rebecca Ivers was the second author of the protocol assessed the studies according to inclusion criteria assessed quality of the studies
compared findings with the first author edited the final draft and provided overall guidance for the process
Alexandra Martiniuk assessed quality of the studies compared findings with the first author discussed disagreements with the third
author and helped write the first draft and the final review
Mark Stevenson contributed to the protocol reviewed methods of studies and contributed to the final draft
Katrina Thistlethwaite conducted a second literature search extracted full studies and contributed to the final draft
D E C L A R A T I O N S O F I N T E R E S T
None known
S O U R C E S O F S U P P O R T
Internal sources
bull No sources of support supplied
External sources
bull National Health and Medical Research Council of Australia Australia
Salary support for Martiniuk Ivers and Stevenson
bull University of Sydney Sesqui Research Grant Australia
Direct research costs and salary support for Kardamanidis
D I F F E R E N C E S B E T W E E N P R O T O C O L A N D R E V I E W
There are no differences between the protocol and the review
I N D E X T E R M S
Medical Subject Headings (MeSH)
lowastMotorcycles Accident Prevention [lowastmethods] Accidents Traffic [lowastprevention amp control] Licensure Program Evaluation Wounds
and Injuries [mortality lowastprevention amp control]
47Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 5050
MeSH check words
Humans
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 1450
isation sponsored by manufacturers and distributors of motorcy-
cles Many USA states have adopted the MSF Basic Rider Courseand Experienced Rider Course (wwwmsf-usaorg) since its release
in 1976
The Motorcycle Operators Skill Test was also developed
by the USA Motorcycle Safety Foundation in coopera-
tion with others (httpwwwwebbikeworldcommotorcycle-
trainingmotorcycle-test) The MOST is a licensing test consist-
ing of a manual knowledge and skill test (comprising eight exer-
cises of increasing difficulty and skill requirements) and a mini-
skill test conducted with a simulator Some studies compared this
alternative licensing test to the standard licensing test However
as the MOST is not a type of rider training we have included only
the two studies that assessed the MOST in combination with one
or more training components in this review Both Anderson 1980and Hall 1988 assessed the MOST in combination with a train-
ing component of three hoursrsquo duration consisting of instructions
about the skills needed to pass the MOST (30 minutes of the-
ory and two and a half hours on the motorcycle) Hall 1988 also
assessed the MOST in combination with the Motorcycle Safety
Foundationrsquos Motorcycle Rider Course as described above
Other USA studies evaluated the Wisconsin motorcycle rider
course a 22-hour off-street programme that teachesthe basic rules
of safe motorcycling (Leung 1987) the USA Maryland Motorcy-
cle Rider Course (MRC) a mandatory (for those younger than 18
years)pre-licence 20-hour coursecomprising eight hours of theory
and12 hours of practice and created by the Maryland Motor Vehi-
cle Administration in 1985 (Perrino 2002) the California Motor-cyclist Safety Program (CMSP) a mandatory (for those younger
than 18 years) pre-licence 16-hour basic course (Billheimer 1998)
and Experienced Rider Course (ERC) of eight hoursrsquo duration
(Billheimer 1998) The CMSP still offers these courses ( http
wwwca-msporg) but they are now MSF courses as described
above Davis 1997 did not give any information about the assessed
rider training
The Canadian studies evaluated the British Colombia Safety
Councilrsquos 37-hour motorcycle safety programme (no further in-
formation on the content was provided) (McDavid 1989) and
the Canada Safety Council Motorcycle Training Program (MTP)
comprising four hours of theory and 16 hours of practice (six
hours on-road) over twoweekends (Jonah 1982) The MTPcoursewas made available in 1974 and is still being offered in Canada
(httpwwwsafety-councilorgtrainingmtpmtphtm) Cooper
1988 did not give any information about the assessed rider train-
ing However the study was conducted in British Colombia in
Canada in 1988 the same time as the McDavid 1989 study was
conducted
In the United Kingdom Raymond 1979 assessed the non-manda-
tory RACACU (Royal Automobile ClubAuto-Cycle Union)
training scheme The training programme consisted of either 24
one-hour sessions or 13 two-hour sessions Twelve hours of the
training focused on theory about basic operational procedures
traffic laws and highway codes The practical session focused on
basic operational skills riding on a training area and riding in traf-fic with an instructor A proficiency test was administered at the
end of the course
One study from rural Thailand (Swaddiwudhipong 1998) evalu-
ated a community-based rider education programme consisting of
theory on motorcycle injury risk helmet use traffic laws vehicle
regulations and traffic signs and a written and skill test leading to
a licence
TheAustralian studies evaluateda rider trainingcourse of 12 hours
over two consecutive days (four hours of theory and eight hours of
practice)administeredto AustraliaPosttelegram deliverers(Adams
1985) and a mandatory pre-licence Ridersafe course consisting
of two tiers the first tier being a training course on the basics
of motorcycle riding where a pass entitled the rider to obtain alearnerrsquos permit and the second tier being a skill test where a pass
entitled the riderto obtain a probationary motorcycle licence The
training course consisted of two four-hour training sessions on
separate days involving instruction videos and actual riding on
motorcycles The skill test consisted of a four-hour preparation
course of 74 minutes of lecturediscussion 38 minutes of video
and 75 minutes of on-course riding which was the testing phase
(Kloeden 1994) The Rider Safe course is still being provided by
Transport South Australia a state government organisation (http
wwwtransportsagovau )
The duration or content of rider training in the four case-control
studies (Kraus 1975 Hurt 1981 Haworth 1997 MAIDS 2004)
was not described
Risk of bias in included studies
We describe the methodological quality of the included studies
by study type below More detailed information can be found in
Table 1 Table 2 Table 3 and Table 4
When considering the findings from these studies it is important
to note the quality (or lack thereof) of reporting of the studies
Quality of reporting is the extent to which a report provides in-
formation about the design conduct and analysis of a study ( Juumlni2001) Huwiler-Muumlntener 2002 found that even though trials
may be well reported it does not mean that the methodological
quality is of equal standard and well-conducted trials may never-
theless be reported badly The reporting quality of most studies in
this review was poor Most studies were old and did not meet con-
temporary reporting standards This made data extraction com-
plicated It also made it difficult to contact authors for additional
information
Because of significant heterogeneity between studies as well as the
poor quality of most studiespooling of studyresults to calculatean
overall measure of effect was not appropriate Instead we provided
a description of our judgement of risk of bias in the text
12Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 1550
Risk of bias in the randomised trials
There were three randomised trials (Anderson 1980 Hall 1988Swaddiwudhipong 1998) Anderson 1980 and Hall 1988 had an
initial sample size of 41188 and 26464 motorcycle licence ap-
plicants respectively Both studies assessed the effectiveness of the
pre-licence MOST with a trainingcomponent Swaddiwudhipong
1998 evaluated the effectiveness of a community-based rider ed-
ucation programme in six districts
Allocation bias
Anderson 1980 reported using random birth date tables to allocate
participants to either the intervention or control group Hall 1988
reported the use of a computerised algorithm set in a central of-
fice but also used manual random assignment when the computer
went down Swaddiwudhipong 1998 reported lsquosimple random
samplingrsquo for the delivery of the intervention where the sampling
frame included nine districts Three districts were excluded since
they were too hard to access The remaining six districts were ran-
domly allocated to the intervention (three) and the control (three)
groups Systematic sampling wasthen used ordered by village size
for the villages within districts Allocation concealment was not
reported in any of the studies The descriptions of randomisation
in these studies suggest that there may have been limitations to
the process resulting in a lack of equivalence of the intervention
and control groups at baseline and therefore the possibility of
allocation bias Swaddiwudhipong 1998 reported that before theintervention the annual incidence rate of motorcycle-related in-
juries was slightly (but not statistically significant) higher in the
intervention areas than in the control areas Participants in the
intervention districts were also more likely to have a valid licence
and to wear a helmet more often than participants in the control
districts Anderson 1980 measured differences at baseline on the
participant and prior driving record variables and found these to
be negligible Hall 1988 did not report on baseline differences
between the intervention and control groups As the Anderson
1980 and Hall 1988 studies had sample sizes of over 40000 and
26000 respectively and used some form of randomisation it is
likely that any confounding factors at baseline would have been
equally distributed between comparison and intervention groups
Selection bias
After randomisation Anderson 1980 and Hall 1988 excluded
08 and 3 participants respectively because of assignment to
the wrong group administration of the wrong intervention or
other reasons The resulting lack of analysis by rsquointention-to-treatrsquo
can introduce biasby removing the effectivenessof randomisation
Nonetheless the percentage excluded was small and the number
of overall participants large so it is unlikely to have had a major
effect on the final results In the Swaddiwudhipong 1998 study
villages were systematically sampled into intervention or control
groups The intervention was implemented and outcomes of in-
jury and death were measured at a community level Interventionand control districts were randomly selected however there were
only three intervention and three control communities
Detection bias
None of the randomised trials reported blinding of outcome as-
sessors Data collection tools used by Anderson 1980 and Hall
1988 were driver records holding police recorded crashes and of-
fences Swaddiwudhipong 1998 used police and hospital records
at a district level and post-mortem examination for deaths None
reported on validity or reliability for any of the instruments
Attrition bias
Anderson 1980 and Hall 1988 had driver records available for
96 to 100 of participants They also collected self-reported
data but the response rate to the questionnaires was very low (low-
est response rate 42 and 30 respectively) Attrition bias in
Swaddiwudhipong 1998 was not an issue as outcome data were
available for all sub-districts included in the trial
The intervention reported in Anderson 1980 and Hall 1988 tri-
als had an unanticipated effect on the participants In both trials
an average of only 28 of licence applicants in the intervention
group eventually obtained their licence compared to 35 in the
control group This indicates that the licensing process in generalseems to have deterred applicants from obtaining a licence and
although no statistical tests of the difference were conducted this
deterrence appears even greater for those who had rider training
This is not true attrition bias as the studies did not experience
differential loss to follow up from outcome measurement Instead
differential attainment of the intermediary step of licensure pre-
cluded participants from being at risk for the outcomes of interest
they are not driving a motorcycle hence they are not at risk for
motorcycle-related crashes injury death or offences In short it is
not possible to assess whether a reduction in crashes in the inter-
vention group would be due to changes in riding habits as a result
of training or simply due to delayed or never attained licensure
Intervention integrity
Neither Anderson 1980 nor Hall 1988 reported on the consis-
tency in the way the intervention was administered the possible
lsquocontaminationrsquo of participants in the intervention group by other
types of formal or informal rider training or of participants in the
control group by the intervention Swaddiwudhipong 1998 did
not report on the number of villagers in each group that actually
attended the intervention However they did report on a number
of other potential training courses that were being implemented
at the same time
13Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 1650
Analysis
Anderson 1980 and Hall 1988 acknowledging the lsquodeterrent ef-fectrsquo described above conducted two separate analyses The first
analysis included the total study population of licence applicants
over two years from the time of their application (the population
denominator) The second analysis was performed on the licensed
(intervention and control) subgroups only two years from the
time of obtaining their licence (the licensed rider denominator)
As only a third were licensed and exposed to riding the motorcy-
cle it was impossible to determine whether reported differences
in crash rates across groups were due to the intervention or to
other lsquoconfoundingrsquo characteristics of those who were determined
in their efforts to obtain the licence When analysis was restricted
only to those who gained their licence the same difficulties re-
mained The benefits of the randomisation process were thus lost
due to this lack of licensure of riders during the process As there
were only three intervention and three control groups in the study
reported by Swaddiwudhipong 1998 analyses were hindered by
smallnumbers for outcome comparisons betweenthe intervention
and the control districts
Risk of bias in the cohort studies and non-randomised
trials
We discuss the methodological quality of cohort studies (seven
prospective studies (Raymond 1979 McDavid 1989 Waller 1992
Kloeden 1994 Billheimer 1998 Perrino 2002 Savolainen 2007)and seven retrospective studies (Osga 1980 Satten 1980 Jonah
1982 Mortimer 1984 Cooper 1988 Mortimer 1988 Davis
1997)) and non-randomised trials (Adams 1985 Leung 1987) in
the one section as the biases for the two study types are similar
The quality of these studies was in general poor
Selection bias
The confounding factors expected to have been addressed in the
studies were sex age riding experience prior driving records (of-
fences crashes) previous training and the following proxies for
risk taking behaviour helmet use speed and alcoholdrug use All
of these factors have been shown to be associated with an increasedrisk of motorcycle crash (Peden 2004 Zambon 2006) The pos-
sible source of bias in cohort studies and non-randomised trials
is that groups may differ on important characteristics at baseline
participants lsquoself-selectrsquo into either the rider training group by de-
ciding to undertake a rider training course or into the lsquocontrol
grouprsquo by deciding not to undertake a course Participants who
undertake a non-mandatory rider training course may differ from
those who choose not to on characteristics such as age sex rid-
ing experience and riding behaviour as well as other possible con-
founders For example riders who take a training course may be
more safety-conscious or on the contrary may do so because they
have previously experienced a crash or because they perceive that
they have poor riding skills Therefore ideally all participants in
the intervention group need to be included in the study as wellas a random sample of controls taken from the same population
as the intervention group If that is not possible intervention and
control groups either need to be carefully matched to make them
comparable or differences at baseline need to be adjusted for in
the analysis
Most of the studies using a non-randomised trial or cohort design
compared an intervention group to a control group that differed
on a number of important factors The studies in general failed to
adjust for important differences found at baseline and their results
cannot be meaningfully interpreted Ten studies (Raymond 1979
Osga 1980 Satten 1980 Jonah 1982 Mortimer 1984 Mortimer
1988 McDavid 1989 Waller 1992 Davis 1997 Savolainen 2007)
assessed intervention and control groups that were not drawnfrom the same population In most of these cases the intervention
group was selected from a list of training graduates who were usu-
ally young female and less experienced riders whereas controls
were recruited from a general population of motorcycle riders who
were usually older male and experienced riders For three stud-
ies it was unclear how they were selected (Leung 1987 Cooper
1988 Perrino 2002) Only Kloeden 1994 recruited groups from
the same population of motorcycle applicants and Adams 1985
from a group of professional riders Nevertheless due to non-ran-
dom selection of intervention and control groups differences at
baseline are likely even though study groups were from the same
population Three studies adjusted for some confounding fac-
tors found at baseline through matching (McDavid 1989 Waller1992 Billheimer 1998) However matching has limitations in
termsof controlling for multiple confounders especially those that
are unknown to influence the effect of rider training on the out-
comes One study matched the control and intervention groups
by excluding controls (approximately 3500 (66 of initial con-
trol group) that differed from the intervention group) (Kloeden
1994) Five studies (Osga 1980 Satten 1980 Adams 1985 Davis
1997 Perrino 2002) did not adjust for differences between study
groups at baseline either in the study design or analysis Results
from these studies are therefore not reliable The other six stud-
ies (Raymond 1979 Jonah 1982 Mortimer 1984 Cooper 1988
Mortimer 1988 Leung 1987 McDavid 1989 Billheimer 1998)
adjusted for some differences between groups at baseline by eitherbi-variate or multi-variate analyses or stratification and standardi-
sation of riding exposure (outcomemiles driven) or stratification
by age group The limitation of stratification is that it is difficult to
control for many variables simultaneously because a large number
of strata will need to be generated resulting in small numbers that
cannot be meaningfully analysed Savolainen 2007 developed sta-
tistical models to assess the probability of a crash after taking the
rider training course in an intervention and control group while
examining differences by sub-groups (eg age) Participation rates
were also very low in this study with only 5584000 (1395)
in the eligible intervention group and 5884000 (147) for the
14Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 1750
control group completing surveys
Detection bias
None of the studies reported blinding of outcome assessors to the
intervention or control status of the participants Seven studies
used driver records containing police recorded data as their main
measure of outcome (Leung 1987 Cooper 1988 McDavid 1989
Waller 1992 Kloeden 1994 Davis 1997 Billheimer 1998) An-
other four studies (Raymond 1979 Satten 1980 Mortimer 1984
Mortimer 1988) only used self-report Raymond 1979 was the
only study using self-report as the main outcome measurement in
which study instruments were validated Self-report might prevent
the underreporting that has been described for the use of police-
driver records (see rsquoRisk of biasrsquo in the randomised trials above)but has other risks of bias especially if questionnaires or inter-
view methods have not been validated and tested for reliability
Recall bias arises from mistakes in recollecting events both be-
cause of failures of memory and because of looking at things lsquowith
hindsightrsquo and possibly changed views (httpwwwcochraneorg
resourcesglossaryhtm) As many studies were retrospective risk
of recall bias is likely and may have increased with length of follow
up For studies that used self-reported outcomes only it is also
possible that social desirability response bias played a role where
participants in the training group may have felt it was lsquoexpectedrsquo
to report good driving histories Multiple methods to measure the
same outcome for both groups decreases the risk of detection bias
for example because self-reported answers can be cross checkedwith other sources of information such as driver records
In this regard three studies (Osga 1980 Jonah 1982 Perrino
2002) used both self-report and driver records Osga 1980 found
that for both intervention and control groups more crashes were
reported through self-report than on state crash records an aver-
age of 54 of self-reported crashes could be found back on state
crash records One industry-based study (Adams 1985) used oc-
cupational health records which were not validated but which are
unlikely to suffer from underreporting due to work cover claims
Savolainen 2007 used information on rider crashes for a five-year
period but did not specify from where this information was ob-
tained
Attrition bias
The majority of studies (Raymond 1979 Satten 1980 Adams
1985 Leung 1987 Kloeden 1994 Davis 1997 Billheimer 1998
Perrino 2002) inadequately reported on losses of participants to
the study We can therefore not be certain that there were no sys-
tematic differences in the characteristics between the intervention
and control group participants Savolainen 2007 reported the re-
sponse rates to their baseline survey but did not report on the per-
centage of outcome data that was available for their study sample
A similar deterrent effect of rider training as described for the ran-
domised trials (Anderson 1980 Hall 1988) occurred in two stud-
ies Jonah 1982 reported that 65(14992310) of the trained rid-
ers stopped riding compared to 47 (9492029) of the untrainedriders They excluded non-active and non-licensed riders before
the study commenced so the differential choice to continue riding
after training did not affect their analyses Leung 1987 reported
that the percentage of participants in the intervention group ob-
taining their licence was 50 (no number provided by authors)
lower than in the control group This may have biased the mea-
surement of outcomes because unlicensed participants will not be
exposed to riding the motorcycle and therefore not be at risk of
being involved in a crash or violation As explained above this
makes it difficult to assess whether the reduction in crashes in the
intervention group is due to a change in driving habits or simply
due to delayed licensure
Mortimer 1984 and Mortimer 1988 describe that about 44 and30 of course graduates respectively were excluded from analysis
because they had not ridden a motorcycle in the previous year
However because these studies were retrospective and the controls
were activeriders at the time of recruitment a decision on whether
the deterrent effect was differential between the two groups could
not be made Though not reported quantitatively Kloeden 1994
also describes the effectof rider training deterringindividuals from
becoming licensed and hence exposure to riding and risk of crash
death or offences involving a motorcycle Osga 1980 reported
that 27 of the trained riders did not ride after the course and
excluded these from the analysis
Use of different data collection strategies may have contributed to
differential loss to follow up in the cohort studies Satten 1980for example interviewed the intervention group over the tele-
phone but mailed questionnaires to the control group Mortimer
1984 and Mortimer 1988 distributed questionnaires in person to
the control group but mailed them to the intervention group
This may have played a role in the observed differential response
rates where more subjects responded in the personally approached
groups
Survival bias
For those studies that assessed offences crashes and injuries but
not death survival bias may have also played a role Motorcyclists
who were fatally or seriously injured in crashes during the follow-
up period after the intervention would have been less likely or
unable to respond to surveys This could bias the effect measures
towards or away from the null None of the studies wrote about
this issue or examined the characteristics of those lost to follow up
in order to address the possibility of survival bias
Intervention integrity
Although the interventions were intended for the entire lsquoexposedrsquo
groups most studies reported difficulty in tracking participants
whoreceivedthe interventionand those whodid not Adams 1985
15Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 1850
estimatedthat 62of thetotalstudysample received the interven-
tion The remaining studies did not report the percentage of thestudy population that received the intervention as was intended
or reported that this percentage was unknown (Raymond 1979
Kloeden 1994 Satten 1980 Jonah 1982 Mortimer 1984 Leung
1987 Cooper 1988 Mortimer 1988 Billheimer 1998 Perrino
2002 Savolainen 2007) Only Raymond 1979 reported the degree
to which their intervention was delivered as intended No studies
reported on possible lsquocontaminationrsquo of participants by other inter-
ventions than the one to which they were allocated Because Osga
1980 Satten 1980 Jonah 1982 Mortimer 1984 Cooper 1988
Mortimer 1988 Waller 1992 and Davis 1997 were retrospective
studies and participants were recruited from a list of graduates it
was known that they had completed and passed the course
Analysis
In some studies some participants who reported at baseline that
they were untrained decided to enroll in a rider training pro-
gramme after the baseline survey The non-randomised trial by
Adams 1985 experienced this move of subjects from the untrained
to the trained group but failed to conduct analyses by lsquointention-
to-treatrsquo This can bias the results in randomised and non-ran-
domised trials if participants who changed groups differ signifi-
cantly with regard to potentially confounding characteristics On
the other hand Perrino 2002 also experienced this move of sub-
jects from theuntrained to the trained groupmore than 100riders(5 of the total sample) originally stated that they were untrained
at baseline but then received training between the baseline and
follow-up surveys Yet since the Perrino 2002 study was a cohort
their analyses then proceeded by including those riders who were
lsquonewlyrsquo trained as exposed Riders who were lsquonewlyrsquo trained since
baseline but with a programme other than the one under study
were excluded from the analysis Nevertheless confounding fac-
tors were not accounted for in the analyses
McDavid 1989 matched exposed with unexposed subjects but
could not find suitable matches for the majority (60) of their
subjects They could therefore only conduct analysis on the result-
ing40 (139346) which is an inappropriate andinefficient anal-
ysis Some studies that reported on the issue of non-active licensedriders(Satten 1980 Jonah 1982 Mortimer 1984 Mortimer 1988
Kloeden 1994) excluded non-active riders from analysis intro-
ducing selection bias as it is likely that riders who continued to
ride differed with regards to potentially confounding character-
istics across intervention and control groups Kloeden 1994 for
example conducted several analyses to assess the effect of various
sub-sets of the control group Original analyses included all con-
trols and led to erroneous results since only a small proportion of
controls hadgone on to get a licence after their learnerrsquos permitand
were actually exposed to the possibility of experiencing a crash
Subsequent analyses compared only intervention and control sub-
jects who had a licence Waller 1992 reported crashes by person-
days of exposure but did not explain how they could be sure that
each study intervention and control participant was an active riderand actually exposed to the risk of experiencing a crash
Potential confounders that are unequally distributed by study arm
at baseline need to be dealt with in the analysis Satten 1980
Adams 1985 and Davis 1997 did not take into account any con-
founding factors either by design or in the analysis McDavid
1989 Waller 1992 and Billheimer 1998 accounted for potential
confounding factors at the design stage by matching the subjects
in the intervention and control groups Other studies (Raymond
1979 Mortimer 1984 Mortimer 1988 Perrino 2002 and also
Waller 1992) adjusted for one or two confounders by stratifica-
tionor standardisation (eg crashesmiles driven age group) Osga
1980 conducteda univariate analysis of crash rates in which no po-
tential confounders were accounted for but also calculated cumu-lative crash rates by mileage travelled Only Jonah 1982 Cooper
1988 and Savolainen 2007 conducted a multivariate analysis
Risk of bias in the case-control studies
A case-control design is suitable for the assessment of risk fac-
tors for outcomes such as motorcycle crashes injuries or deaths
but not appropriate for the evaluation of an intervention such
as motorcycle rider training It can potentially result in spurious
conclusions regarding the efficacy of the intervention because the
intervention has not been applied at random to an intervention
and control group Even with the control of possible confounders
using a case-control design to evaluate an intervention can lead tothese kinds of results since those who choose to undertake rider
training may differ on important prognostic factors that affect the
outcome compared to those who do not choose to participate in
rider training
Selection bias
A potential source of bias in case-control studies as with all other
study types where the intervention is not randomly allocated to
participants is that comparison groups are likely to differ on po-
tentially confounding characteristics This means that ideally all
cases need to be included in the study as well as a random sample
of controls taken from the population that gave rise to the cases If that is not possible cases and controls either need to be carefully
matched to make themcomparable or differences at baseline need
to be adjusted for in the analysis All four case-control studies were
population based Kraus 1975 included all fatalities and injured
cases leading to hospitalisation and appropriately recruited a ran-
dom sample of controls from the same population Hurt 1981 did
not manage to recruit all or a random sample of cases Only half
of the notifications of crashes to the police were received in time
by the researchers the research team for unreported reasons was
only able to respond to half of those again Data collection was
therefore completed for only about a fifth of all cases Neverthe-
less this was still a large sample of 900 cases Haworth 1997 using
16Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 1950
a similar study design was more likely to have included all cases
but this is not entirely clear from the study reported It is unclearfor example whether fatalities were included as they did not use
coronerrsquos reports for the recruitment of cases The MAIDS 2004
study was conducted in five European countries but it was unclear
how cases were recruited whether they were all cases during the
two years that the study lasted or a random sample of all cases
what the response rates were etc Controls were recruited from
randomly selected petrol stations but it was likewise not clear what
the eligibility criteria for inclusion in the study were how they
were recruited (eg randomly) and what the response rate was
Hurt 1981 and Haworth 1997 both matched controls to cases by
recruiting themat the sametime ofdayandsamedayof weekas the
crash occurred Hurt 1981 however recruited controls two years
later andit is likelythat there would have beendifferences betweenthe two comparison groups One important difference was that
during data collection for cases there was very little specialised
motorcycletrainingavailable in the studyregionwhereas two years
later when control data were collected there was some training
available
These four case-control studies also suffered from selection bias
due to potential differences between those that gave consent to
participate in the study and responded to interviews andquestion-
naires and those that did not None of the studies have assessed
such differences although Hurt 1981 did acknowledge that con-
trols giving consent to cooperate may have differed from those
motorcyclists who did not
Detection bias
Measures that can be taken to reduce detection bias in case-con-
trol studies are blinding of interviewers for case or control status
blinding of the participant for the study factor of interest validat-
ing measurement instruments and using these in an identical way
for cases and controls and cross checking self-reported informa-
tion with other more objective sources such as registers None of
the studies reported blinding of the intervention assessors to case
or control status but all used consistent measurement methods for
both case and control groups
All case-control studies used self-report as their main measure toassess whether cases and controls had received rider training and
if so the type of training The major potential source of bias in
case-control studies when self-report is used as a measure is recall
bias One way of reducing recall bias is by double-checking self-
reported data with more objective data Kraus 1975 also used po-
lice accident reports for information on rider characteristics but
such information was not available about rider training Haworth
1997 and MAIDS 2004 did not report on any measures to limit
recall bias Hurt 1981 reported further questioning of cases by
the interviewer when answers did not make sense or when par-
ticipants seemed to contradict themselves as part of the in-depth
unstructured interview technique This qualitative research tech-
nique can result in valuable information but as interviewers were
not blinded to case or control status of the participants it may have led to more thorough examination of cases compared to the
controls and thus differential detection bias
Intervention integrity
The level of detail about the rider training provided in the case-
control studies differed significantly but was minimal for each
Kraus 1975 simply asked whether lsquoanyrsquo motorcycle rider training
had lsquoeverrsquo been carried out Hurt 1981 asked participants whether
they hadbeen self taught informally taught or completed formally
taught rider training courses MAIDS 2004 asked whether partic-
ipants had had pre-license training additional (non-compulsory)
training other training or no training whereas Haworth 1997asked about the number of courses that a participant had com-
pleted whetherthey were basic intermediate or advanced courses
and how long ago the training had been done None of the case-
control studies reported collecting information about further de-
tails about any training programmes (eg type of training cur-
riculum duration of training)
Analysis
Kraus 1975 accounted for confounding factors through the exclu-
sion of females and stratification by age and frequency of motor-
cycle use Hurt 1981 matched cases and controls for time of day
and day of the week that the crash occurred and did not adjust forany confounding factors in the analysis Haworth 1997 matched
groups for time of day and day of the week that the crash occurred
and stratified by age blood alcohol level and licence status in the
analysis None of the studies used multivariate methods MAIDS
2004 did not provide any information about any participant char-
acteristics other than the study exposures of interest and did not
describe any methods to adjust for potential confounding factors
Effects of interventions
We have presented the results in a descriptive manner due to the
substantial heterogeneity in study types interventions and out-comes and the generally poor quality of the studies We have
grouped the interventions by mandatory and non-mandatory pre-
licence rider training and post-licence rider training and further
by study type and by outcome (crashes injurious and fatal crashes
and offences) We discuss results that are reported as statistically
significant (whether a P value was reported or not)
Pre-licence rider training - mandatory
Randomised trials
17Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2050
Crashes
Total group of participants
Anderson 1980 found a reduction in motorcycle crash rates in
the intervention group (MOST skill test with remedial training)
compared to the control group (who received the skill testwithout
a training component) The percent differenceranged from 143
six months after the programme (P lt 005) to 211 (P lt 0001)
after one year and 127 (P lt 001) after two years Hall 1988
found no statistically significant difference in crash rates between
intervention groups at any point in time
Subgroup of participants who obtained their licence only
Anderson 1980 found a reduction in motorcycle crash rates in the
intervention arm but this was statistically significant at one year
only (24 to 218 reduction P lt 005) Hall 1988 found no
statistically significant differences
Offences
Total group of participants
The percentage of offences in Anderson 1980 after six months
and one and two years was slightly higher in the intervention
arm compared to the control group but these differences were
not statistically significant Likewise the study by Hall 1988 did
not find a statistically significant difference in offences between
groups
Subgroup of participants who obtained their licence only
Hall 1988 found no significant difference in offence rates between
the intervention and control groups Anderson 1980 found thatthe percentage of offences increased in the intervention arms com-
pared to the control group This difference was statistically signif-
icant after two years (51 to 115 increase P lt 005)
Cohort studies
Crashes
Neitherof the cohort studies (Kloeden1994 and Billheimer1998)
found a statistically significant difference in crash rates between
the intervention and control groups
Offences
Billheimer 1998 (for the California Motorcyclist Safety Program
course) and Perrino 2002 did not find a statistically significant
difference in offence rates between the intervention and control
groups
Case-control studies
Crashes
MAIDS 2004 found that a slightly higher percentage of controls
(not involved in a crash) had had pre-license rider training than
cases (involved in a crash) This difference was not statistically significant (50 of controls vs 477 of cases OR 091 95 CI
076 to 110)
Pre-licence rider training - non-mandatory
Randomised trial
Injuries and deaths
Swaddiwudhipong 1998 found a significantly lower injury ratein the intervention villages (1051000 population 95 CI 92
to 120) compared to the control villages (1691000 population
95 CI 152 to 185) one year after the intervention This differ-
ence was not maintained two years after the intervention (127
95 CI 112 to 143 intervention 151 95 CI 135 to 167
control) There was no significant difference in death rates be-
tween groups
Cohort studies
Crashes
Cooper 1988 foundno statistically significant differencein crashes
between the intervention and control groups McDavid 1989
found a lower crash rate in the intervention group (14139) com-
pared to the control group (23139) but no test of statistical sig-
nificance was undertaken
Injuries and deaths
McDavid 1989 found that the intervention group had fewer in-
jurious crashes (814 crashes) compared to the control group (16
23 crashes) but no statistical significance test was reported
18Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2150
Offences
Cooper 1988 foundno statistically significantdifference in offencerates between the intervention and control groups McDavid 1989
foundno statistically significantdifference in offence ratesbetween
the intervention and control groups
Non-mandatory rider training
Non-randomised trials
Crashes
Neither Adams 1985 nor Leung 1987 found a statistically signif-icant effect of rider training on crash rates
Offences
Leung 1987 found a statistically significant lower offence rate in
the intervention group compared to the control group for both
males and females and for all age groups except for women 35
years or older (P values not reported)
Cohort studies
Crashes
Osga 1980 in the univariate analysis reported a lower percentage
of crashesin theintervention group (that continued to ride) (28
168595) than in the control group (34 112329) In the cu-
mulative crash rate analysis however Osga1980 found statistically
significant (at the P lt 001 level) higher crash rates in the interven-
tion group than in the control group at allmileage intervals (1000
to 29999 miles) travelled After riding more than 30000 miles
(when over 50 of the trained riders had been involved in a crash
compared to approximately 40 in untrained riders) this differ-
ence was no longer statistically significant Satten 1980 found a
lower percentage of crashes in the intervention group (13 969)
compared to the control group (25 2079) but the differencewas not assessed statistically Davis 1997 found a lower crash rate
in trained riders but it was unclear how this was calculated be-
cause the size of the control group was not reported Jonah 1982
and Billheimer 1998 (for the Experienced Rider Course) found no
statistically significant difference in crashes between the interven-
tion and control groups Although the difference between those
who completed the Experienced Rider Course and no course in
Billheimer 1998 neared significance at six months after the course
(P = 007) large P values were observed for one year after (P =
030) Raymond 1979 found a statistically significant higher crash
rate for the intervention group (1861000 miles ridden) com-
pared to the control group (1181000 miles ridden) in both males
and females (P lt 0001) especially during the first 5000 miles of
riding Waller 1992 found a statistically significant higher crashrate for the intervention group (27 crashes in three years or 070
crashes10000 person days) compared to the control group (5
crashes in three years or 013 crashes10000 person days) with a
crude relative risk of 538 (X 2=1513 plt005) Mortimer 1984
and Mortimer 1988 found no statistically significant difference in
crashes between the intervention and control groups
Savolainen 2007 found that the intervention group (those who
completed the Basic Rider Course) was 44 more likely to be
involved in crashes than the control group and those who com-
pleted the course multiple times were an additional 180 more
likely to be involved in a crash This finding was based on low
numbers and no P value or confidence level was given for either
result
Injuries and deaths
Satten 1980 also found a lower percentage of injurious crashes
(33 39) in the intervention group compared to the control
group(60 1220) although again no statistical significance test
was reported
Offences
Jonah 1982 did find a significantly lower self-reportedoffence rate
in the intervention group (15) compared to the control group(34 P lt 001) (no numbers given) Satten 1980 also found
a lower self-reported offence rate in the intervention group (1
71 versus 19100) but again this was not tested for statistical
significance Mortimer 1984 Mortimer1988 and Billheimer 1998
found no statistically significant differencein offence ratesbetween
the intervention and control groups
Case-control studies
CrashesMAIDS 2004 found that a slightly higher percentage of cases
(involved in a crash) had had additional non-mandatory rider
training than controls (not involved in a crash) This difference
was not statistically significant (17 of cases vs 12 of controls
OR 146 95 CI 064 to 337)
Rider training unspecified
Case-control studies
19Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2250
Injuries and deaths
Hurt 1981 found that the odds of injurious crashes were lower inthose who had received professional training compared to those
who had not Kraus 1975 found that in the younger age group
(less than 20 years) and the older age group (over 35 years) a larger
percentage of injured riders (12 and 21 respectively) reported
prior motorcycle rider training compared with the uninjured con-
trol group (6 and 5 respectively) (P lt 001) (no numbers
given) Haworth 1997 reported several different outcomes one
beingthat having completed at leastone trainingcoursecompared
to no training courses at all had no significant effect on the odds
of crashing (afteradjustment for age blood alcohol level or licence
status (learner permitprobationaryfull)) When comparing three
different levels of training courses (beginner intermediate and ad-
vanced) with no training no significant differences were found
in the odds of crashing However after adjustment for age and
blood alcohol level completing an advanced course as compared
to a beginners course was associated with a significant decrease in
the odds of severe motorcycle crashes resulting in serious injury or
death (OR 02 95 CI 01 to 08)
D I S C U S S I O N
The objective of this review was to quantify the effectiveness of
pre- andpost-licence motorcycle rider training on the reduction of
traffic crash involvement injuries deaths and offences committedby motorcycle riders Most included studies were cohort studies
There were a few randomised and non-randomised trials and case-
control studies
All studies compared either mandatory or non-mandatory formal
training with no formal training except for Haworth 1997rsquos case-
control study that compared levels of training (beginner interme-
diate and advanced) to each other
We found studies that evaluated the effectiveness of manda-
tory pre-licence training in reducing crash and offence rates
(Anderson 1980 Hall 1988 Kloeden 1994 Billheimer 1998
(the basic California Motorcyclist Safety Program component of
the study) Perrino 2002) non-mandatory training that when
passed resulted in a licence (Cooper 1988 McDavid 1989
Swaddiwudhipong 1998) non-mandatory training that could be
taken by motorcycle riders with or without a licence (Raymond
1979 Osga 1980 Satten 1980 Jonah 1982 Mortimer 1984
Adams 1985 Leung 1987 Mortimer 1988 Waller 1992 Davis
1997 Billheimer 1998 (the Experienced Rider Course compo-
nent of the study)) and unspecified rider training on crash injury
and death rates (Kraus 1975 Hurt 1981 Haworth 1997 MAIDS
2004)
The studies provided conflicting evidence regarding motorcycle
rider training which is likely due to the poor design quality and
reporting of the studies Most of the studies included in the review
were likely to be adversely affected by significant selection biasalong with substantial detection bias(measurement and recall bias)
which limits our ability to make conclusive comments
Mandatory pre-licence rider training
There were two RCTs three cohort studies and one case-control
study that examined pre-licence ridertraining The two RCTs eval-
uated the Motorcycle Skills Operating Test (MOST) (Anderson
1980 Hall 1988) Anderson 1980 found a statistically significant
reduction in crashes in the intervention group at six months one
and twoyearsfollow up However when only participantswho ob-
tainedtheirlicence were included in theanalysis thereduction wasstatistically significant only at one year follow up The other RCT
(Hall 1988) found no statistically significant differences in crash
rates at any point of time neither in the total population group
nor in the licensed subgroup Neither of the two cohort stud-
ies assessing crash rates found a statistically significant difference
in crash rates between intervention and control groups ( Kloeden
1994 Billheimer1998) andthe case-control study (MAIDS 2004)
also did not find a statistically significant difference in the odds of
experiencing a crash between trained and untrained riders With
regards to offences neither of the RCTs found a statistically sig-
nificant difference in offence rate between the intervention and
control groups in the analysis of the total group of participants
entering the trial However Anderson 1980 found a statistically significant increase in offence rates in the intervention group but
only after two years of follow up andonly in the analysis of partici-
pants who obtained their licence The two cohort studies assessing
offence rates found no statistically significant difference between
the groups (Billheimer 1998 Perrino 2002)
None of the five studies (RCTs or cohort) found a reduction in
crashes or offences following mandatory pre-licence rider training
except for Anderson 1980 who observed a statistically significant
reduction in crashes in the intervention group at one year in both
analyses It is not clear why both RCTs that evaluated the same
MOST and training intervention observed different results Both
studies enrolled large samples and observed a similar uptake of
licensure following the intervention It is possible that differencesmay have arisen due to different study samples (California ver-
sus New York) or that the intervention delivery differed between
studies
In both trials of the Motorcycle Skills Operating Test including
rider training only a third of motorcycle licence applicants ob-
tained their motorcycle licence It is possible that a licensing pro-
cess including a skills test may reduce crashes through decreased
exposure
Non-mandatory pre-licence training
20Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2350
Three studies assessed the effect of non-mandatory pre-licence
training The Cooper 1988 cohort study found no difference incrash or offence rates between the trained and untrained groups
We cannot however be confident about this result as the study
sufferedfrom selectionbiaswhichcouldhave biasedthe measure of
effect toward the null The RCT conducted by Swaddiwudhipong
1998 examined injury and death rates and found a statistically
significant lower injury rate in the intervention villages after one
year of follow up but no longer at two year follow up There
was no statistically significant difference in death rates between
intervention and control villages Our confidence in the results of
this study are hindered by the inadequacy of the randomisation
process that led to significant differences between the intervention
and control districts on two important variables The intervention
districts were more likely than the control districts to have a validlicence andto wear a helmetSincethese differences were nottaken
into account in later analyses they could explain the differences
observed in injury rates between districts Furthermore with only
six districts randomised the sample size was probably too small
to detect a difference between intervention and control groups
had there been one The McDavid 1989 cohort study found a
lower percentage of injuries in the intervention group although
a statistical test was not conducted McDavid 1989 drew their
intervention and control groups from different populations but
did adjust for confounders although they only analysed 40 of
the initial sample (139 trained and 139 untrained) Because of
these substantial weaknesses we are unable to draw conclusions
from this studyAn overall effect of pre-licence training on the reductionof crashes
injuries deaths and offences was not found although one should
be aware that the research of this issue to date has had significant
flaws However a mandatory skill test (such as the MOST) in
combination with and even without pre-licence training as part
of a licensing process seems to reduce the number of individuals
completing the licensing procedure and therefore indirectly re-
duces crash rates through a reduction in exposure to motorcycle
riding Making licensure a more difficult process may therefore be
seen as advantageous if it delays licensure and therefore reduces
exposure It is possible that the group who chose not to continue
riding a motorcycle would not have been the group to experience a
crash or commit an offence if they had continued to ride anyhowfor example because they are more safety conscious and this was
the reason for their giving up riding in the first place If this is the
case deterring individuals from completing their licence and rid-
ing may not necessarily decrease overall crashes deaths or offence
rates In addition other potentially adverse effects of complicat-
ing a licensing procedure need to be considered such as increases
in unlicensed riding In Thailand non-mandatory rider training
offered as a community intervention may have decreased injury
rates but this needs to be confirmed through further research
Non-mandatory training
The majority of the studies assessed non-mandatory training
(training independent of any licensing procedure) and the results
were varied The two non-randomised trials assessing crash rates found no statistically significant difference in crash rates between
the intervention and control groups (Adams 1985 Leung 1987)
Of the nine cohort studies Satten 1980 found a lower percentage
of crashesin the intervention group although it is unclear whether
this was a statistically significant result Davis 1997 also found a
lower percentage of crashesin theinterventiongroup although it is
unclear with which control group the intervention group wascom-
pared Raymond 1979 Osga 1980 Waller 1992 and Savolainen
2007 found a statistically significant higher crash rate in the in-
tervention group and the remaining cohort studies (Jonah 1982
Mortimer 1984 Mortimer 1988 Billheimer 1998) found no sta-
tistically significant difference in crash rates between intervention
and control groups No conclusions can be drawn from these stud-
ies with confidence The Satten 1980 and Davis 1997 studies did
not adjust for any confounding factors even though important
differences existed between the intervention and control groups
Although Raymond 1979 and Waller 1992 adjusted for a num-
ber of confounders at baseline (Waller 1992) and in the analy-
sis (Raymond 1979 Waller 1992) the intervention and control
groups were drawn from such different populations that it is pos-
sible that unmeasured confounders were responsible for the sta-
tistically significant increased risk of crashes observed in the inter-
vention group The observed difference could also be attributed totheobserved differential loss to followup (Raymond 1979) which
could bias the measure of effect towards or away from the null
The Osga 1980 study found a protective effect of rider training
in the univariate analysis but statistically significant higher crash
rates in the intervention group when accounting for riding expo-
sure Sex (a significantly higher proportion of females in the in-
tervention group) and other potentially confounding factors that
could explain thedifferencebetween the groups were howeverun-
accounted for Finally Savolainen 2007 did control for potential
confounding variables yet it is not clear whether variables were
controlled for simultaneously or one at a time One case-control
study (MAIDS 2004) found no statistically significant reduction
in the odds of experiencing a crash between trained and untrainedriders but the quality of this study was poor mainly because the
analysis did not account for any confounding factors
One cohort study (Satten 1980) assessed the effect of non-manda-
tory rider training on injurious crashes and found a lower percent-
age of injuries in the intervention group although a statistical test
was not conducted As mentioned above the Satten 1980 studyrsquos
validity is reduced due to its lack of adjustment for confounding
factors
The only non-randomised trial that assessed offence rates (Leung
1987) found a statistically significant lower offence rate in the in-
tervention group and of the five cohort studies that assessed of-
fence rates (Satten 1980 Jonah 1982 Mortimer 1984 Mortimer
21Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2450
1988) one study (Jonah 1982) also found a statistically significant
lower offence rate in the intervention group However we have lit-tle confidence in the results from Leung 1987 since potential con-
founding factors were not appropriately addressed Jonah 1982
did conduct a multivariate analysis but was hindered by an initial
low response rate The other four cohort studies did control for
several potential confounders and found no difference in offence
rates following non-mandatory post-licence rider training
The non-randomised trials and cohort studies evaluating non-
mandatory rider training were at high risk of substantial selection
bias due to the non-random allocation of participants as well
as low participation rates and attrition bias and could therefore
not provide conclusive evidence with regard to the effect of the
training It should be noted that despite the biases and other issues
with this group of studies most didnot detectmajor effects relatedto rider training in either direction This may suggest that any
potential impact of rider training will be modest
Studies examining unspecified modes of ridertraining
Three case-control studies assessed whether injured or deceased
motorcyclists were more or less likely to have had rider training
compared to control motorcyclists All had conflicting results
Hurt1981 foundthe odds of havingan injuriouscrash waslower in
those who had received rider training whereas Kraus 1975 found
that the odds of having an injurious crash were higher in youngerand older trained riders Haworth 1997 who compared advanced
versus beginner training found no effect on motorcycle injuries
or deaths although they did find that those who had completed
advanced training had reduced chances of crashing compared to
those who had done a beginnersrsquo training course Similar issues
with uncontrolled confounding and biases of selection as described
above arose with the three case-control studies Hurt 1981 suffers
from inadequate control of confounding via poor matching and
both the Kraus 1975 and the Haworth 1997 findings were only
controlled for limited factors by means of stratified analysis It is
likely that these limitations are the reason for the contradictory
directions of effect observed for the three case-control studies
Methodological weaknesses of the studies
In summary most studies investigating the effectiveness of mo-
torcycle rider training on reducing crashes injuries and death suf-
fered from serious methodological weaknesses
The majority of studies were non-randomised and were there-
fore limited mainly by selection bias Although the approaches of
matching stratification or regression analyses can minimise bias
(Fleiss 2004) they were poorly employed in the reviewed studies
The inadequate adjustment for confounders in particular adds
to our lack of confidence in the studiesrsquo results since it has been
demonstrated that individuals who choose to take motorcycle rider
training differ from those not taking rider training For instance
trained riders often differ from controls in age sex and experi-ence usually ride less and may have a more positive attitude to
safety as indicated by the reported more frequent wearing of seat
belts (Mortimer 1982)
The two large randomised trials reported that a large proportion
of their study group were never exposed to the risk of offences or
crashes because they never attained licensure This made it impos-
sible to assess whether a reduction in crashes in the intervention
group was a result of training or due to delayed or never attained
licensure
In some studies the sample sizes were too small to detect any sta-
tistically significant differences Follow-up time after training dif-
fered between studies and was an impediment to observing any
real effects of training if they existed As riding experience ac-cumulates over time any differences in outcomes between those
trained and untrained would diminish in studies with lengthy fol-
low-up times This dilution of any possible effect of rider training
may have also occurred in studies that mixed both novice and ex-
perienced riders in the study groups since it is hypothesised that
novice riders serve to benefit most from rider training This would
increase the difficulty in finding any effect of rider training par-
ticularly when sample sizes are small
Detection bias may also account for some of the differences ob-
served between trained and untrained riders Detection bias would
have occurred if the self-report hospital driver or police records
used to record the outcomes of crash injury death or offences
differed between the trained and untrained riders Research hasdemonstrated that not all crashes are reported to police for ex-
ample small crashes or off-road crashes (Kraus 1975) even serious
injury crashes have been found to be underreported (Anderson
1980) However this would only introduce detection bias if the
reporting of outcome differs for trained and untrained riders This
does appear possible in that responsible riders may be more likely
to enroll in a rider training course and also may be more likely
to report crashes to police than irresponsible riders meaning that
the number of crashes from riskier riders could be underrepre-
sented on records An analysis to investigate the effectiveness of
rider training may then show that rider training is not effective in
reducing crashes
Rather than using reported outcomes some studies used self-re-ported survey data although this technique suffers from non-
response biases In fact two studies reported differences in de-
mographic characteristics in response rates to questionnaires
(Anderson 1980 Hall 1988) Self-reported data also introduce
the risk of recall bias survival bias and social desirability response
bias Recall bias may have been equally distributed among trained
and untrained riders howeverit is hypothesised that trained riders
may have felt it was lsquoexpectedrsquo to report good driving histories
hence leading the rider training to appear effective
22Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2550
Limitations of this review
Limitations of this review were that we found very few recentstudies (the two most recent ones published in 2002 and 2007)
and that we found very few studies conducted in low-and middle
income countries even though our search strategy was set up to
identify such studies Anotherlimitation wasthat forthe reviewed
studies we could not assess all outcomes of rider training that are
potentially useful Some of the included studies assessed outcomes
such as use of safety clothing including helmets Such outcomes
are valuable since the use of motorcycle helmets has been demon-
strated to be effective in preventing head injuries (Liu 2004) De-
spite this the use of safety clothing was not included as an out-
come in this review because the studies measuring this outcome
were limited by poor data quality due to very low response rates
Another outcome of interest in assessing the effectiveness of rider
training is lsquotype of injuryrsquo As Haworth 2005 state in their review
of motorcycle licensing and training rider training could be con-
sidered effective if injuries become less severe because motorcycle
riders have learned how to avoid more serious injuries through
rider training The studies included in this review did not specify
type of injury and so this outcome could not be assessed Last
changes in knowledge and attitudes following rider training are
other potentially useful outcomes We have excluded such studies
(Osga 1979 Clayton 1990 Goldenbeld 2004) as these outcomes
are proxies rather than direct measures of the effectiveness of mo-
torcycle rider training We opted to report on the outcomes of
crashes injury death and offences alone since there is little ev-
idence about how changes in knowledge and attitudes relate to
these outcomes
A U T H O R S rsquo C O N C L U S I O N S
Implications for practice
Due to the poor quality of studies included in this review we were
unable to draw any conclusions about the effectiveness of rider
training on crash rates injurious or fatal or offence rates The
findings do suggest that mandatory pre-licence training may be
an impediment to completing a motorcycle licensing process thuspossibly indirectly reducing crash injury death and offence rates
through a reduction in exposure to riding a motorcycle On the
basis of the existing evidence it is not clear if and what type of
training reduces the risk of crashes injuries deaths or offences in
motorcyclists and the selection of the best rider training practice
can therefore not be recommended The suggestion that manda-
tory pre-licence training indirectly reduces exposure to riding the
motorcycle because trainees are impeded from completing the li-
censing process does not help those motorcyclists who persist in
obtaining their licence and riding the motorcycle Some type of
rider training is likely necessary to teach motorcyclists to ride a
motorcycle safely Good policy is based on sound evidence and
evaluation but such evidence for motorcycle rider training is seri-
ously lacking Policy makers are therefore encouraged to include arigorous evaluation component to any new or existing rider train-
ing programme Such an evaluation should ideally be randomized
and assess whether the training programme is effective and if so
which component(s) of the programme work best
Implications for research
Given theresults of this review which found littlereliable evidence
about the effectiveness of motorcycle rider training it is necessary
to build up an evidence base of the effectiveness of motorcycle
rider training in the reduction of crash injury death and offence
rates To this end well designed randomised controlled trials need
to be conducted that examine the impact of rider training oncrashes and offences A randomised controlled trial is the preferred
method as it would significantly reduce most biases described in
this review To minimise the impact of participants delaying or
never attaining their licence and therefore exposure as found in
the randomised trials in this review such a trial should include
riders who are already licensed and reliable measures of riding
exposure in a reasonably lengthy follow up period It is expected
that a randomised controlled trial of sufficient sample size would
observe equal balance on possible confounders by study arm how-
ever an assessment of potentially important confounders (such as
helmet wearing) by study group would enable control for these
variables in the analyses if necessary It is imperative that the effec-
tiveness of each component of the rider training (such as theorypractice knowledge skills and modification of behaviour) and the
synergy of such components are evaluated Outcomes should be
objective ideally those that are routinely collected rather than self-
reported outcomes Even though riding the motorcycle poses dif-
ferent challenges and incurs different risks lessons may be learned
from research into the effects of car driver training
Motorcycles are a common means of transport in low- and middle
income countries and usage is increasing As the burden of mo-
torcycle crash victims is mostly borne in such countries research
should be focused on the effectiveness of types of rider training on
the reduction of crashes injuries and fatalities in such countries
and be supported by higher income countries This review found
only one study conducted in a middle income country that sug-gested decreasing injury rates It would be beneficial to conduct
additional research to confirm this suggestion and to enhance our
knowledge of the role of rider training in low- and middle-income
settings
March 2006 saw the launch of the University of North Carolina
Highway Safety Research Centre partnership with the Motorcycle
Safety Foundation (MSF) and plans were revealed to conduct a
study of the potential benefits of motorcycle rider training The
study is funded by the MSF and the National Highway Traffic Sa-
fety Administration (NHTSA) Called rsquoThe Discovery Projectrsquo it
will compare data on crash rates from riders participating in a basic
23Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2650
motorcycle safety class with riders participating in a basic motor-
cycle safety class as well as a variety of continuing education classesThe collection of pilot data was due to start in March 2007 and
the data collection portion of the study is expected to take three
years (Highway Safety Research Center) It is vital that researchers
and policy makers collaborate to develop furtherhigh quality eval-
uations of motorcycle rider training programmes to help inform
development of new and effective training programmes
A C K N O W L E D G E M E N T S
Staff from The University of Sydney library for helping us locate
and retrieve full reports of many papers
R E F E R E N C E S
References to studies included in this review
Adams 1985 published data only
Adams A Collingwood V Job RFS Evaluation of the 1982 Australia
Post motorcycle rider training scheme NSW Australia Traffic
Accident Research Unit Traffic Authority NSW 1985 [ RN 85]
Anderson 1980 published data only
Anderson J Ford JL Peck RC Improved motorcyclist licensing and testing project Washington DC National Highway Traffic Safety
Administration (NHTSA) USA 19801ndash217 [
DOTndashHSndash805ndash598]
Billheimer 1998 published data only
Billheimer JW Evaluation of the California Motorcyclist Safety
Program Transportation Research Record 19981640100ndash9 [
TRB Report Number 980652]
Cooper 1988 published data only
Cooper PJ Rothe JP Motorcyclists who they are and why they do
what they do Transportation Research Record 1988116878ndash85
Davis 1997 published data only
David CF Evaluation of community traffic safety programs and
motorcycle operator training programs Connecticut Department of Transportation 199763
Hall 1988 published data only
Hall CJ Buchanan L Engle R Motorcycle rider education evaluationproject Washington DC USA National Highway Traffic Safety
Administration (NHTSA) US Department of Transportation
1988 [ DTNH 22ndash80ndashCndash05123]
Haworth 1997 published data only
Haworth N Smith R Brumen I Pronk N In Monash University
Accident Research Centre editor(s) Case-control study of motorcycle
crashes Victoria Australia Department of Transport and Regional
Development The Federal Office of Road Safety 19971ndash149 [
ISBN 0 642 25501 6 ]
Hurt 1981 published data only
Hurt H Ouellet J Thom D Motorcycle accident cause factors and
identification of countermeasures USA Department of
Transportation 1981 [ Technical Report DOT HS 806ndash862]
Jonah 1982 published data onlylowast Jonah BA Dawson N Bragg B Are formally trained motorcyclists
safer Accident Analysis amp Prevention 198214(4)247ndash55
Jonah BA Dawson NE Bragg BW Evaluation of the Canada
Safety Councilrsquos Motorcycle Training Program Proceedings of theInternational Motorcycle Safety Conference Washington DC
USA 198059ndash93
Kloeden 1994 published data only
Kloeden CN Moore VM McLean AJ Evaluation of the pre-license
training program for motorcyclists in South Australia Walkerville
South Australia Australia NHMRC Road Accident Research Unit
The University of Adelaide 19941ndash37 [ ISBN 0 7308 0607 3]
Kraus 1975 published data only
Kraus JF Riggins RS Franti CE Some epidemiological features of
motorcycle collision injuries American Journal of Epidemiology 1975102(1)74ndash98
Leung 1987 published data only
Leung KS Reding VA Evaluation of the Wisconsin motorcycle rider course Wisconsin Bureau of Policy Planning and Analysis
Wisconsin Department of Transportation 1987
MAIDS 2004 published and unpublished data
MAIDS In-depth investigations of accidents involving powered
two wheelers ACEM (Acemacemeu) 2004 Vol Final report
20179
McDavid 1989 published data only
McDavid JC Lohrmann BA Lohrmann G Does motorcycle
training reduce accidents Evidence from a longitudinal quasi-
experimental study Journal of Safety Research 198920(2)61ndash72
Mortimer 1984 published data only
Mortimer R Evaluation of the motorcycle rider course Accident
Analysis amp Prevention 198416(1)63ndash71
24Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2750
Mortimer 1988 published data only
Mortimer RG A further evaluation of the motorcycle rider courseJournal of Safety Research 198819(4)187ndash96
Osga 1980 published data only
Osga GA An investigation of the riding experiences of MSF rider
course participants Dissertation South Dakota USA The
University of South Dakota 19801ndash224
Perrino 2002 published data only
Perrino CS Ahmed A Callender A Rozier E Cantwell A Stewart
O The role of Marylandrsquos motorcycle rider course in promoting saver
behaviors and attitudes Baltimore USA National Transportation
Center Morgan State University 20021ndash52
Raymond 1979 published data only
Raymond S Tatum S An evaluation of the effectiveness of the RAC
ACU motorcycle training scheme - final report Road Safety ResearchUnit Dept of Civil Engineering University of Salford London
19791ndash131
Satten 1980 published data only
Satten RS Analysis and evaluation of the motorcycle rider courses
in thirteen Illinois counties International Motorcycle Safety
Conference Washington DC 1980145ndash93
Savolainen 2007 published and unpublished data
Savolainen Peter Mannering Fred Additional evidence on the
effectiveness of motorcycle training and motorcyclistsrsquo risk-taking
behaviour Transportation Research Board 2007 Vol March 20
19
Swaddiwudhipong 1998 published data only
Swaddiwudhipong W Boonmak C Nguntra P Mahasakpan P
Effect of motorcycle rider education on changes in risk behavioursand motorcycle-related injuries in rural Thailand Tropical Medicine
and International Health 19983(10)767ndash70
Waller 1992 published data only
Waller FL An analysis of the impact of the Texas Department of
Public Safety Motorcycle Operator Training Program upon
motocycle related traffic accidents Texas State University Public
Administration Program Applied Research Projects 199277
References to studies excluded from this review
Anderson 1980a published data only
Anderson LC Better biking a motorcycle rider improvement
program International Motorcycle Safety Conference Washington
DC USA Motorcycle Safety Foundation 1980231ndash53Aupetit 2008 published and unpublished data
Aupetit S Riff J Buttelli O Espie S Ergonomics research program
on motorcycle learning Contributions to the development of a
motorcycle simulator for training Transport Research Arena
Ljubljana 20089
Awane 1999 published data only
Awane T Integrating simulators in motorcycle safety education
Journal of International Association of Traffic and Safety Sciences 199923(1)26ndash35
Baldi 2005 published data only
Baldi S Baer JD Cook AL Identifying best practices states in
motorcycle rider education and licensing Journal of Safety Research200536(1)19ndash32
Batchler 1988 published data only
Batchler MW Evaluation of the pre-license motorcycle training course at the New South Wales traffic education centre Federal Office of
Road Safety (FORS) 19881ndash122 [ Report No CR 77 ]
Blanchard 2006 published data only
Blanchard HT Tabloski PA Motorcycle safety educating riders at
the teachable moment Mosby (CV) Company 200632330ndash2
Bonnett 2005 published data only
Bonnett J Mature age skills training for experienced motorcycle
riders Motorcycle ridersrsquo association of the Australian Capital
Territory 2005
Braver 2007 published data only
Braver ER et alPersuasion and licensure a randomized controlled
intervention trial to increase licensure rates among Maryland
motorcycle owners Traffic Injury Prevention 20078(1)39ndash46Clayton 1980 published data only
Clayton AB Sudlow DE An evaluation of the effectiveness of a one day training programme for learner motorcyclists Berkshire UK
Transport and Road Research Laboratory 19801ndash23 [ ISSN
0266ndash7045]lowast Clayton AB Sudlow DE An evaluation of the effectiveness of a one-day training programme for learner motorcyclists Berkshire UK
Road Safety Division Berkshire 19801ndash23 [ISSN 0266ndash7045]
Council 1978 published data only
Council FM Desper LP Dutt AK Driver education for motorcycle operation University of North Carolina Highway Safety Research
Center 19781ndash40
Crowther 2005 published data onlylowast Crowther G Engaging with motorcyclists UK police and theBikeSafe Road Safety Programme Published proceedings of the
Australasian Road Safety Research Policing Education Conference
Wellington New Zealand 2005 issue ISBN 0ndash473ndash10636ndash11ndash11
Daniello 2009 published data only
Daniello A Gabler HC Mehta YA Effectiveness of motorcycle
training and licensing Transportation Research Board 88th Annual
Meeting 200913
Dewanti 2007 published data only
Dewanti M Challenging efforts in promoting young motorcyclist
safety in Indonesia Road Safety on Four Continents 14th
International Conference Swedish National Road and Transport
Research Institute 2007
Du Plooy 1990 published data onlyDu Plooy DW Motorcycle training in South Africa The Human
Element International Motorcycle Safety Conference Orlando
Florida USA Motorcycle Safety Foundation 19904-1 to 4-9
Engel 1989 published data only
Engel U Krogsgaringrd Thomsen L [Faeligrdselsundervisning og
propaganda ndash en faeligrdselssikkerhedsfremmende foranstaltning]
Dansk Vejtidsskrift 198912295ndash7
Goldenbeld 2002 published data only
Goldenbeld C Houwing S de Craen S The development of
driving skills of young moped riders [De ontwikkeling van de
rijvaardigheid van jonge bromfietsers] SWOV Institute for Road
Safety Research (Stichting Wetenschappelijk Onderzoek
Verkeersveiligheid) Leidschendam the Netherlands 200263
25Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2850
Goldenbeld 2004 published data only
Goldenbeld C Twisk D de Craen S Short and long term effects of moped training a field experiment Transportation Research Part F 20047(1)1ndash16
Haworth 2000 published data only
Haworth N Smith R Kowadlo N Motorcycle training in Victoria
evaluation of rider training curriculum in Victoria Monash
University Accident Research Centre Reports 2000 Vol 165 issue
3131
Haworth 2005 published data only
Haworth N Mulvihill C Review of motorcycle licensing and
training Monash University Accident Research Centre 2005 [
Report no240]
Haworth 2006 published data only
Haworth N Mulvihill C Rowdon P Teaching old dogs new tricksTraining and older motorcyclists Journal of the Australasian College of Road Safety 200718(4)20ndash5lowast Haworth NL Mulvihill C Rowden PJ Teaching old dogs new
tricks Training and older motorcyclists Australasian Road Safety
Research Policing and Education Conference Gold Coast
Queensland 2006
Hill 1983 published data only
Hill D Motorcycle rider training testing and licensing Motor
Cycle Council of New South Wales Sydney 1983
Imatake 1980 published data only
Imatake Y On the activities in Japan of motorcycle safe riding
promotion committee (MSRPC) - especially the education of
unskilled riders of class 1 motorbikes International Motorcycle
Safety Conference Washington DC USA Motorcycle Safety
Foundation 1980255ndash68
Ingebritsen 1990 published and unpublished data
Ingebrigtsen S [Risikofaktorer ved ferdsel med moped og
motorsykkel] TOslashI-rapport 66 Oslo Transportoslashkonomisk
institutt 1990
Jain 2009 published and unpublished data
Jain A et alTwo wheeler accidents on Indian roads a study from
Mangalore India Elsevier 200916130ndash3
Jonah 1979 published data only
Jonah BA Dawson NE Validation of the motorcycle operator skill
test Accident Analysis amp Prevention 197911(3)163ndash71
Jonah 1981 published data onlyJonah BA Dawson NE Bragg WE Predicting accident
involvement with the Motorcycle Operator Skill Test Accident Analysis amp Prevention 198113(4)307ndash418
Kadar 1990 published data only
Kadar KJ Countermeasures motorcycle rider education delivery
system The Human Element International Motorcycle Safety
Conference Motorcycle Safety Foundation 19906-27 to 6-41
Kelsey 1986 published data only
Kelsey SL Liddicoat C Ratz M Licensing novice motorcyclists a
comparison of Californiarsquos standard test and the MOST II (Motorcycle
Operator Skill Test) administered at centralized testing offices Sacramento California USA California Department of Motor
Vehicles Research and Development Office 19861ndash74
Koch 1980 published data only
Koch H Advanced course for motorcycle riders InternationalMotorcycle Safety Conference Washington DC USA Motorcycle
Safety Foundation 1980269ndash81
Koch 1990 published data only
Koch H Brendicke R How to influence risk taking behaviour
within motorcycle rider training programs The Human Element
International Motorcycle Safety Conference Orlando Florida
USA Motorcycle Safety Foundation 1990Motorcycle Safety
Foundation 1990 p 9-1 to 9-25A
Labbett 2005 published data only
Labbett S Langham M Training can make the problem worse
Road Safety Congress 20051ndash12
Law 2005 published and unpublished data
Law TH Radin Umar RS Zulkaurnain S Kulanthayan S Impactof the effect of economic crisis and the targeted motorcycle safety
programme on motorcycle-related accidents injuries and fatalities
in Malaysia International Journal of Injury Control and Safety
Promotion 200512(1)9ndash21
Lowes 1990 published data only
Lowes BF In-traffic rider training The Human Element
International Motorcycle Safety Conference Orlando Florida
USA Motorcycle Safety Foundation 19907-46 to 7-57
Mahon 1976 published data only
Mahon J Teaching accident avoidance on the motorcycle Journal of Traffic Safety Education 1976231ndash29
Maxwell 1985 published data only
Maxwell GS Evaluation of a Program for Training MotorcycleRiders Evaluation and Program Planning 19858(4)339ndash47
Mayhew 1996 published data only
Mayhew DR Simpson SH Effectiveness and role of driver educationand training in a graduated licensing system Ottowa Canada The
Traffic Injury Research Foundation 19961ndash89 [ ISBN
0ndash920071ndash12ndash0]
McKnight 1987 published data only
McKnight AJ Milburn BG An evaluation of the modified
motorcycle operator skill test exercises NHTSA Report 1987 [
Report No DOT HS 807 212 ]
McPherson 1976 published data only
McPherson K McKnight AJ The development and evaluation of a
motorcycle skill test manual and knowledge test Washington DCUSA National Highway Safety Administration 1976
Mohan 2004 published data only
Mohan D Evidence-based interventions for road traffic injuries in
South Asia Journal of the College of Physicians and SurgeonsPakistan 200414(12)746ndash7
Morgan 1980 published data only
Morgan JK Advanced motorcycle training course content and
study evaluation International Motorcycle Safety Conference
Washington DC USA Motorcycle Safety Foundation 1980
283ndash312
Motoki 1990 published data only
Motoki M Yamazaki S A study on effective motorcycle rider
education The Human Element International Motorcycle Safety
26Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2950
Conference Orlando Florida USA Motorcycle Safety
Foundation 19909-26 to 9-57A
Nairn 1992 published data onlylowast Nairn RJ Motorcycle safety research literature review 1987 to
1991 Federal Office of Road Safety Australia 1993 Vol Report
CR 117
Newland 1980 published data only
Newland R Darby J Automotive practicespre motorcycle rider
program International Motorcycle Safety Conference Washington
DC USA Motorcycle Safety Foundation 1980127ndash9
Newland 1982 published data only
Newland R Motorcycle rider training and licensing requirements
Driver training steering a course for the future Melbourne
Victoria Transport Regulation Board 1982
Newland 1990 published data only
Newland R Motorcycle training and licensing in Victoria 1980-
1990 The Human Element International Motorcycle Safety
Conference Orlando Florida USA Motorcycle Safety
Foundation 1990
Ormston 2003 published data only
Ormston R Dudleston A Pearson S Stradling Evaluation of
Bikesafe Scotland Scottish Executive Social Research Edinburgh
2003
Osga 1979 published data only
Osga GS Ellingstad VS An evaluation of the effectiveness of the 1978
South Dakota motorcycle rider course program South Dakota USA
University of South Dakota 19791ndash89
Panou 2007 published data only
Panou M Bekiaris E A new concept on the integration of driving
simulators in driving training the train-all approach Road Safety
on Four Continents 14th International Conference Swedish
National Road and Transport Research Institute 2007
Peterson 1990 published data only
Peterson A Kadar K Status report 1980-1989 Minnesota rider
education statistics The Human Element International
Motorcycle Safety Conference Orlando Florida USA 199013-
113 to 13-129
Prem 1984 published data only
Prem H Good MC Motorcycle rider skills assessment ACT
Australia Office of Road Safety Department of Transport 1984
1ndash377 [ Report No CR 34 ]
Radin Umar 2006 published data only
Radin Umar RS Motorcycle safety programmes in Malaysia how
effective are they International Journal of Injury Control and Safety Promotion 200613(2)71ndash9
Rockwell 1990 published data only
Rockwell TH Kiger S Carnot M An evaluation of the Ohio
motorcyclist enrichment program Ohio Department of Highway
Safety 1990
Rothe 1987 published data only
Rothe JP A motorcycle riding course Motorcyclists Image and reality Canada Insurance Corporation of British Columbia 1987
59ndash83
Rowden 2007 published data only
Rowden P Watson B Haworth N What can riders tell us aboutmotorcycle training A view from the other side of the fence
Australasian Road Safety Research Policing and Education
Conference Melbourne Australia Melbourne 2007
RSC Victoria 1993 published data only
Road Safety Committee Inquiry into Motorcycle Safety in Victoria
Parliament of Victoria 1993
Russam 1979 published data only
Russam K Motorcycle training and licensing in Japan UK Transport
and Road Research Laboratory (TRRL) 1979 [ Report 916 ]
Saffron 1981 published data only
Saffron DG Driver and rider education training and licensing a
brief review Traffic Accident Research Unit Traffic Authority of
New South Wales NSW Australia Traffic Accident Research UnitTraffic Authority of New South Wales 1981
Schulz 1990 published data only
Schulz U Motorcycle accidents among learner riders Motorcycle
licensing Evaluation of the effectiveness of a Graduated Licensing Scheme Forshungshefte des Institut fur Zweiradsicherheit 1990
Sexton 2004 published data only (unpublished sought but not used)
Sexton B Baughan C Elliott M Maycock G The accident risk of
motorcyclists 2004TRL Report TRL 607(ISSN 09684107)58
Shephard 1986 published data only
Shepherd R Do Motorcycle Safety Foundation programs work
Driver 198619(9)16ndash8
Simpson 1990 published data onlySimpson HM Mayhew DR The promotion of motorcycle safety
training education and awareness Health Education Research
Theory and Practice 19905(2)257ndash64
Sokolov 1990 published data only
Sokolov O Safe motorcycle riding some aspects of motorcycle
riding in the USSR The Human Element International
Motorcycle Safety Conference Motorcycle Safety Foundation
19904ndash37 - 4-48
Steffens 1988 published data only
Steffens U Gawatz R Willmes G Effectiveness of mofa courses
[Wirksamkeit von Mofakursen] Unfall- und sicherheitsforschung
strasenverkehr 198867168
Sudlow 2003 published data only
Sudlow D Scoping study on motorcycle training Road Safety
Research Report London UK The British Institute of Traffic
Education Research Department for Transport 2003 Vol June
[ Report No 36]
Tasmania 1995 published data only
Mannering FL Grodsky LL Motorcycle rider training scheme
Traffic Education Research Centre Armidale Australia Driver
Training Australia Limited 1995 Vol April1ndash56
Tetard 1985 published data only
Tetard C Motorcycle ridersrsquo training to detect indices in an
intersection situation [Lrsquoentrainement a la prise drsquoinformation dans
lrsquoapprentissage de la conduite motocycliste] Le Travail Humain198548(2)147ndash59
27Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3050
Thackray 1980 published data only
Thackray RM Prescott JC Field test of the motorcycle ridercourse International Motorcycle Safety Conference Washington
DC USA Motorcycle Safety Foundation 1980195ndash7
Thompson 1994 published data only
Thompson M Evaluation of compulsory basic training for
motorcyclists Transport Research Laboratory Crowthorne
Berkshire UK 1994 [ Report 63]
Tobler 1990 published data only
Tobler U Individual driving instruction for motorcycle driving
learners the Swiss experiment of motorcycle driving school The
Human Element International Motorcycle Safety Conference
Orlando Florida USA Motorcycle Safety Foundation 1990
Transport 1989 published data only
Transport Communications Road Crash Statistics Australia Bureauof Transport and Communications Economics 198914ndash20
Tuan 2007 published data only
Tuan VA Shimizu T Towards development and evaluation of the
motorcycle drivers re-education program in Vietnam modeling of
motorcycle driverrsquos undesired behaviors 11th World Conference
on Transport Research 200725
Victor 1980 published data only
Victor DJ Traffic safety education for motorcyclists International
Motorcycle Safety Conference Washington DC USA Motorcycle
Safety Foundation 1980221ndash8
Waterer 1990 published data only
Waterer TH Compulsory training for motorcyclists from liberty
to compulsion The Human Element International Motorcycle
Safety Conference Orlando Florida USA Motorcycle Safety
Foundation 19904-49 to 4-57
Weller 1989 published data only
Weller RB Chandler EW Motorcycle safety and motorcycle
education past research and survey results Journal of Public Policy
amp Marketing 1989893ndash108
Wells 1981 published data only
Wells P Observations of some RACACU motorcycle training
schemes Transport and Road Research Laboratory Crowthorne
Berkshire UK 1981
Winn 1986 published data only
Winn G Two important trends in motorcycle safety regulations
rider education and conspicuity improvement American
Motorcyclist Association 1986147ndash52
Wisher 1988 published data only
Wisher G Reid J Motorcycle accidents in Norfolk a three-year study of motorcycle training and the Part 1 test Norfolk UK Road Safety
Division Highways Department Norfolk County Council 1988
Wood 1987 published data only
Wood HT Bowen R Evaluation of the revised learner permit scheme
July 1983 to December 1985 Victoria Australia Road Traffic
Authority 19871ndash27 [ ISBN 0 7306 0247 8]
Yamazaki 1990 published data only
Yamazaki S Nakanishi M Survey on effect of Honda rider
education The Human Element International Motorcycle Safety
Conference Orlando Florida USA 199012-167 to 12-187
References to ongoing studies
Burgess 2010 unpublished data only
Burgess C Broughton P Fylan F Stradling Interim evaluation of
the UKs National Ride program In L Dorn editor(s) Driver
behaviour and Training Vol IV Aldershot Ashgate Publishing
Limited 2010
Additional references
Anderson 1980
Anderson J Ford JL Peck RC Improved motorcyclist licensing and
testing project Volume I and II National Highway Traffic Safety
Administration US department of Transportation Washington
DC 1980
Bjornstig 1985
Bjornstig U Bylund P Lekander T Brorsson B Motorcycle fatalitiesin Sweden Acta Chirurgica Scandinavica 1985151(7)577ndash81
Clayton 1990
Clayton A Sudlow DE An evaluation of the effectiveness of a one-
day training program for learner motorcyclists The Human
Element International Motorcycle Safety Conference Orlando
Florida USA 199013-105 to 13-111
Cochrane Collaboration 2008
Higgins JP Green S Cochrane Handbook for Systematic Reviews of
Interventions 500 The Cochrane Collaboration 2008
Elliot 2003
Elliott MA Baughan CJT Broughton J Grayson G Knowles J et
alMotorcycle Safety a Scoping Study TRL [ Report No
0968ndash4107]Fleiss 2004
Fleiss JL Levin B Paik MC Statistical methods for rates and proportions 3rd Edition John Wiley amp Sons Inc 2004 [ ISBN
978ndash0ndash471ndash52629ndash2 ]
FORS 1999
FORS Federal Office of Road Safety Road risk for sober licensed
motorcyclists Vol Monograph 27 Federal Office of Road Safety
19991ndash2
Haworth 2005
Haworth N Mulvihill C Review of motorcycle licensing and training
Victoria Australia Monash University Accident Research Centre
20051ndash83 [ Report No 240]
Huwiler-Muumlntener 2002Huwiler-Muumlntener K Juumlni P Junker C Egger M Quality of
Reporting of Randomized Trials as a Measure of Methodologic
Quality JAMA 20022872801ndash4
Johnston 2008
Johnston P Brooks C Savage H Fatal and serious road crashes
involving motorcyclists Road Safety Research and Analysis Report
ATSB Monograph 20 Canberra Department of Infrastructure
Transport Regional Development and Local Government 0205
081ndash26
Juumlni 2001
Juumlni P Altman DG Egger M Systematic reviews in health care
Assessing the quality of controlled clinical trials BMJ 2001323
42ndash6
28Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3150
Kopits 2003
Kopits E Cropper M Traffic fatalities and economic growth TheWorld Bank Development Research Group Infrastructure and
Environment Policy research working paper 3035 2003
Leung 1983
Leung P Chau BTW Motorcycle accidents in Hong Kong pattern
of injury treatment causative factors and socio-economic
implications Injury 198315(1)23ndash9
Liu 2004
Liu B Ivers R Norton R Blows S Lo SK Helmets for preventing
injury in motorcycle riders Cochrane Database of Systematic Reviews 2004 Issue 2 [DOI 10100214651858CD004333pub3]
Mohan 2002
Mohan D Road safety in less-motorized environments future
concerns International Journal of Epidemiology 200231(3)527ndash32Mortimer 1982
Mortimer RG An evaluation of the motorcycle rider course
Proceedings of the 20th conference of the American Association for
Automotive Medicine Arlington Heights IL USA American
Association for Automotive Medicine 198277ndash96
Mullin 1997
Mullin BT Risk factors for motorcycle injury the role of age
gender experience training and alcohol PhD thesis Auckland
New Zealand University of Auckland 1997
Noordzij 2001
Noordzij PC Forke E Brendicke R Chinn BP Integration of needs
of moped and motorcycle riders into safety measures review and
statistical analysis in the framework of the European research
project PROMISING SWOV Institute for Road Safety Research
Leidschendam The Netherlands 2001 [ Report No Dndash2001ndash5]
Osga 1979
Osga G Ellingstad VS An evaluation of the effectiveness of the 1978 South Dakota Motorcycle Riders Course Program South Dakota
University of South Dakota Vermillion Human Factors Lab 1979
Oxford English Dictionary Online
httpoxforddictionariescom Oxford English Dictionary Online
Oxford University Press 2010
Peden 2004
Peden M Scurfield R Sleet D World Report on Road TrafficInjury Prevention Geneva World Health Organization Available
at httpwwwwhointviolence˙injury˙preventionpublications
road˙trafficworld˙reporten 2004Chapter 3
Radin Umar 1995
Radin Umar R Macakay G Hills B Preliminary analysis of
motorcycle accidents Short-term impacts of the running
headlights campaign and regulations in Malaysia J Traffic Med 199523(1)17ndash27
Tsai 1995
Tsai YJ Wang JD et alCase-control study of the effectiveness of
different types of helmets for the prevention of head injuries among
motorcycle riders in Taipei Taiwan American Journal of
Epidemiology 1995142(9)974ndash81
Vis 1995
Vis AA In-depth study of the hazards of motorcycling [De
onveiligheid van motorrijden nader bekeken] SWOV Institute for
Road Safety Research Leidschendam the Netherlands 1995 [
Report No Rndash95ndash69]
Weaver 1990
Weaver EA Motorcycle safety education in the United States
Motorcycle Safety Foundation Orlando Florida USA 1990
Winn 1987
Winn G Two important trends in motorcycle safety regulations
rider education and conspicuity improvement No more
information available Detroit MI US Society of Automotive
Engineers 1987147ndash52
Yuan 2000
Yuan W The effectiveness of the rsquoride-brightrsquo legislation for
motorcycles in Singapore Accident Analysis amp Prevention 200032
(4)559ndash63
Zambon 2006
Zambon F Hasselberg M Factors affecting the severity of injuries
among young motorcyclists a Swedish nationwide cohort study
Traffic Injury Prevention 20067(2)143ndash9lowast Indicates the major publication for the study
29Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3250
C H A R A C T E R I S T I C S O F S T U D I E S
Characteristics of included studies [ordered by study ID]
Adams 1985
Methods Non-randomised trial
Participants Initial sample size
Total of 162 Australia Post telegram delivery riders
Not clear how many in intervention and control groups
Analysis sample size
Intervention group 63Control group 35
Total 98 (61)
Interventions No formal training vs voluntary rider training course of 12 hours over 2 consecutive days (theory 4 hours practice
8 hours)
Intervention integrity
No information given
Outcomes Australia Post recorded crashes
Length of follow up 1 year
Notes Study conducted in NSW Australia
Anderson 1980
Methods Randomised trial
Participants Initial sample size
Total 41188 motorcycle licence applicants
Analysis sample size
Intervention group with skills training 13160
Intervention group without skills training 12634
Control group 15080
Total 40874 (99)
Interventions Standard licensing skill test vs the Motorcycle Operators Skill Test (MOST) with or without a training component
MOST is not to be confused with rider training because it is a licensing test The training component was for
participants who failed the skill test on the first attempt was of 3 hours duration and consisted of instructions about
the skills needed to pass the skill test The training consisted of 30 min theory and 2 frac12 hours practice on the
motorcycle
Intervention integrity
33 in MOST with skills training 31 in MOST without skills training and 43 in control group obtained their
motorcycle licence in the end
Outcomes Police recorded crashes and offences
Length of follow up 3 years
30Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3350
Anderson 1980 (Continued)
Notes Study conducted in California USA
Billheimer 1998
Methods Cohort study
Participants Initial sample size
Intervention group Unclear numbers of CMSP graduates and ERC graduates
Control group Over 16000 untrained motorcyclists
Analysis sample size
1139 matched CMSP-pairs and 1182 ERC-pairs
Interventions No formal training vs the mandatory (for those younger than 18 years) pre-licence 16-hours basic California
Motorcyclist Safety Program (CMSP) and the 8-hour Experienced Rider Course (ERC)
Intervention integrity
No information given
Outcomes Police recorded crashes and offences
Length of follow up 2 years
Notes Study conducted in California USA
Cooper 1988
Methods Cohort study
Participants Initial sample size
Intervention group 418 graduates and 57 who failed
Control group 402
Analysis sample size
Intervention group unclear
Control group unclear
Total 863 (984)
Interventions No formaltrainingvs non-mandatory pre-licence rider trainingThe content of thetrainingwas notfurther explainedIntervention integrity
Participants had all completed the rider training
Outcomes Police recorded crashes and offences
Length of follow up 7 years
Notes Study conducted in British Columbia Canada
31Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3450
Davis 1997
Methods Retrospective cohort study
Participants Initial sample size
Intervention group 9320 graduates
Control group unclear
Analysis sample size
Intervention group 9320 graduates
Control group unclear
Interventions No formal training vs the non-mandatory Connecticut Motorcycle Operator Training Program The content of
training was not further explained
Intervention integrity Participants had all completed the rider training
Outcomes Police recorded crashes severity of crashes
Length of follow up 1 year
Notes Study conducted in Connecticut USA
Hall 1988
Methods Randomised trial
Participants Initial sample sizeIntervention group A 6248
Intervention group B 6616
Intervention group C 6996
Control group 6604
Total 26464 motorcycle licence applicants
Analysis sample size
Intervention group A 6012 (96)
Intervention group B 6428 (97)
Intervention group C 6778 (97)
Control group 6502 (98)
Interventions Standard licensing skill test vs the Motorcycle Operator Skill Test (MOST II) without training (A) MOST II with
3 hours training (B) or MOST II with 20 hours training (C) The 3 hours training comprised instructions aboutthe skills needed to pass the skill test (30 min theory and 2 frac12 hours practice on the motorcycle) The 20 hours
training consisted of the Motorcycle Safety Foundationrsquos Motorcycle Rider Course (MRC) comprising of 8 hours
theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-road practice
Intervention integrity
25 and 26 obtained their motorcycle licence in the lsquoMOST 3-hour trainingrsquo and lsquoMOST 20-hour trainingrsquo
intervention arm respectively In the lsquoMOST testrsquo intervention arm 29 obtained their licence and in the control
group 35
Outcomes Police recorded crashes and offences
Length of follow up 3 years
32Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3550
Hall 1988 (Continued)
Notes Study conducted in New York USA
Haworth 1997
Methods Population based case-control study
Participants Cases Motorcycle riders or pillion riders involved in motorcycle crash (on road) Recruited from hospital and police
records Not clear whether all cases (especially fatalities) were included
Community controls Multiple matched sample of motorcycle rider or pillion who passed the crash site at the same
time of day and day of week that the crash occurred
Initial sample size
Cases 336
Controls 1195
Analysis sample size
Cases 234 (70)
Controls 561 (47)
Interventions Basic handling skills course beginner intermediate or advanced course measured through self-report not validated
Outcomes Injuries with an Injury Severity Score of 5 or greater or admission for at least four hours and fatalities from a
motorcycle crash (from hospital and police records)
Length of follow up two years or more prior to the interview
Notes Study conducted in Victoria Australia
Hurt 1981
Methods Population based case-control study
Participants Cases Motorcycle riders who had a motorcycle crash that was reported to police (traffic accident reports) Not all
cases included
Community controls Multiple matched sample of motorcycle riders who passed the crash site at the same time of
day and day of week as the crash occurred but 2 years later
Initial sample size
Cases 4500 notifications of crashes to police (only 50 received in time research team able to respond to 1126 for
unreported reasons data collection completed for 900)
Controls Number approached unreported 2310 consented and interviewed
Analysis sample size
Cases 808 (808900 data on rider training) - 18 of initial sample
Controls 670 (6702310 data on rider training) - lt 29 of initial sample
Interventions Training experience self taught friends-family school-club formal-professional others measured through self-
report validated
Outcomes Injuries and fatalities from a motorcycle crash (from hospital records and coronerrsquos office)
33Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3650
Hurt 1981 (Continued)
Notes Study conducted in Los Angeles USA
Jonah 1982
Methods Cohort study
Participants Initial sample size
Intervention group
2310 licensed MTP graduates who are active riders
Control group
2029 licensed motorcyclists who are active riders
Non-licensed and non-active riders excluded
Analysis sample size
Intervention group
811 interviews (35) 533 driver records (23)
Control group 1080 interviews (71) 905 driver records (45)
Interventions No training vs the non-mandatory Canada Safety Council Motorcycle Training Program (MTP) comprising of 4
hours theory and 16 hours practise (6 hrs on-road) over 2 weekends
Intervention integrity
Participants had all successfully completed the MTP
Outcomes Self-reported and police reported crashes and offencesLength of follow up 4 years
Notes Study conducted in Ontario Canada
Kloeden 1994
Methods Cohort study
Participants Initial sample size
Intervention group 2533 licence applicants
Control group Unclear
Sample of riders obtaining probationary licenceIntervention group 1625 (64)
Control group 5015 (unclear)
Non-active riders excluded from control group (~66)
Analysis sample size
Intervention group 631 (39)
Control group 318 (6)
Interventions No formal training vs the mandatory pre-licence Ridersafe course consisting of two levelsthe first is a training course
in the basics of motorcycle riding where a pass entitles the rider to obtain a learnerrsquos permit The second is a skill
test where a pass entitles the rider to obtain a probationary motorcycle licence The training course consists of two 4
hour training sessions on separate days involving instruction videos and actual riding on motorcycles The skill test
consists of a 4 hour preparation course of 74 minutes of lecturediscussion 38 minutes of video and 75 minutes of
34Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3750
Kloeden 1994 (Continued)
on-course riding which is the testing phase
Intervention integrity
Unclear what percentage in the Ridersafe group and the control group obtained their probationary licence halfway
and their full motorcycle licence in the end
Outcomes Police reported crashes
Length of follow up 1 year
Notes Study conducted in SA Australia
Kraus 1975
Methods Population based case-control study
Participants Cases All motorcycle riders with a confirmed medically treated motorcycle injury (including fatalities) from all
hospitals in study base and peripheral area hospitals and all deaths due to motorcycle crashes in the study base All
serious injuries (leading to hospitalisation) and fatalities likely to be included
Community controls A random sample of registered motorcycle owners in the same temporary and geographical
study base as the cases arose from
Initial sample size Cases 1273
Controls 738
Analysis sample size Cases 741 (58)
Controls 434 (59)
Interventions Motorcycle rider training (not specified) measured through self-report
Outcomes Injuries leading to hospitalisation fatalities
Length of follow up not specified
Notes Study conducted in Sacramento county USA
Leung 1987
Methods Non-randomised trial
Participants Initial sample sizeIntervention group unknown number of MRC graduates
Control group unknown number of licensed motorcyclists
Analysis sample size
Intervention group 2941
Control group 43094
Interventions No formal training vs non-mandatory Wisconsin motorcycle rider course a 22 hours off-street programme that
teaches the basic rules of safe motorcycling
Intervention integrity
50 in the MRC group and 100 in the control group obtained their motorcycle licence in the end
35Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3850
Leung 1987 (Continued)
Outcomes Police recorded crashes and offences
Length of follow up up to 3 years
Notes Study conducted in Wisconsin USA
MAIDS 2004
Methods Population based case-control study
Participants Cases Motorcycle riders who had a motorcycle crash Unclear whether all cases included
Community controls Motorcycle riders identified at petrol stations Unclear whether matched to cases
Initial sample size
Cases not reported
Controls not reported
Analysis sample size
Cases 921 (Italy 200 Netherlands 200 Spain 121 Germany 250 France 150)
Controls 923 (Italy 200 Netherlands 200 Spain123 Germany 250 France 150)
Interventions Pre-license rider training or additional training (non-compulsory)
Outcomes Crashes
Notes Study in five sampling areas located in five European countries (France Germany Netherlands Spain and Italy)
McDavid 1989
Methods Cohort study
Participants Sample size after first match
Intervention group 346 male training graduates
Control group 346 male motorcycle riders licensed in same year
Sample size after second match
Intervention group 139 (40)
Control group 139 (40)
Analysis sample size
Intervention group 139 (40)
Control group 139 (40)
Interventions No formal training vs the non-mandatory pre-licence British Columbia Safety Councilrsquos 37-hour motorcycle safety
programme
Intervention integrity
Participants had all completed and passed the training
Outcomes Police reported crashes injurious crashes
Length of follow up 1-6 years
36Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3950
McDavid 1989 (Continued)
Notes Study conducted in British Columbia Canada
Mortimer 1984
Methods Cohort study
Participants Initial sample size
Intervention group 608 MRC graduates
Control group 360 motorcycle riders
Response rates
Intervention group 360 (59)
Control group 324 (90)
Non-active riders excluded
Analysis sample size
Intervention group 213 (35)
Control group 303 (84)
Interventions No formal training vs the non-mandatory Motorcycle Safety Foundationrsquos Motorcycle Rider Course comprising of
8 hours theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-road practice
Intervention integrity
Participants had all completed and passed the MRC
Outcomes Self-reported crashes and offencesLength of follow up 1 year
Notes Study conducted in Illinois USA
Mortimer 1988
Methods Cohort study
Participants Initial sample size
Intervention group 2307 MRC graduates
Control group 556 active licensed motorcycle riders
Response ratesIntervention group 1298 (56)
Control group 500 (90)
Non-active riders excluded
Analysis sample size
Intervention group 913 (40)
Control group 500 (90)
Interventions No formal training vs the non-mandatory Motorcycle Safety Foundationrsquos Motorcycle Rider Course comprising of
8 hours theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-road practice
Intervention integrity
Participants had all completed and passed the MRC
37Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4050
Mortimer 1988 (Continued)
Outcomes Self reported crashes and offences
Length of follow up 1 year
Notes Study conducted in Illinois USA
Osga 1980
Methods Cohort study
Participants Initial sample sizeIntervention group 1163 MSF MRC participants invited 818 (74) responded
Control group 329 motorcycle riders responded to recruitment strategies
Non-active riders excluded from analysis
Analysis sample size
Intervention group 595 (73)
Control group 329 (100)
Interventions No formal training vs the non-mandatory Motorcycle Safety Foundationrsquos Motorcycle Rider Course (MRC) com-
prising of 8 hours theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-road
practice
Intervention integrity
Participants had all successfully completed the MRC
Outcomes Self-reported crashes police recorded crash reports
Length of follow up 6 months - 3 12 years
Notes Study conducted in South Dakota USA
Perrino 2002
Methods Cohort study
Participants Initial sample size
Intervention group 2144 (unclear) MRC graduates (motorcycle licence applicants)
Control group 2601 (unclear) regular licensing test graduates (motorcycle licence applicants)Analysis sample size
Intervention group 1070 (unclear)
Control group 867 (unclear)
Interventions No formal training vs the mandatory (for those younger than 18 years) pre-licence Maryland Motorcycle Rider
Course (MRC) consisting of 8 hours theory and 12 hours practice
Intervention integrity
No information given
Outcomes Self-reported and police recorded crashes and offences
Length of follow up 1 year
38Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4150
Perrino 2002 (Continued)
Notes Study conducted in Maryland USA
Raymond 1979
Methods Cohort study
Participants Initial sample size
Intervention group
258 volunteers mainly (but not all) novice riders
Control group
1104 provisional licence applicants
Analysis sample size
Intervention group 190 (74)
Control group 627 (57)
Interventions No formal training vs the non-mandatory RACACU (Royal Automobile ClubAuto-Cycle Union) training scheme
The training program consisted of either 24 one-hour sessions or 13 two-hour sessions Twelve hours of the training
focused on theory about basic operational procedures traffic laws and highway code The practical session focused
on basic operational skills riding on a training area and riding in traffic with an instructor A proficiency test was
administered at the end of the course
Intervention integrity
No information given
Outcomes Self-reported crashes (including minor non-injury non-damage and off-road crashes)
Length of follow up 1 year
Notes Study conducted in the UK
Satten 1980
Methods Cohort study
Participants Initial sample size
Intervention group interviewers kept interviewing until they had 100 riders selected from a population of 891
enrollees in the MSF MRC course (unknown how many were asked and how many refused to participate)Control group unknown number of registered motorcycle riders
Non-active riders excluded from analysis
Analysis sample size
Intervention group 71
Control group 100
Interventions No formal training vs the non-mandatory Motorcycle Safety Foundationrsquos (MSF) Motorcycle Rider Course (MRC)
comprising of 8 hours theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-
road practice
Intervention integrity
Participants had all successfully completed the MRC
39Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4250
Satten 1980 (Continued)
Outcomes Self-reported crashes injuries and offences
Length of follow up 3 years
Notes Study conducted in Illinois USA
Savolainen 2007
Methods Cohort study
Participants Initial sample size
Intervention group 4000 riders that had conducted the MSF MRC course (passed or failed)
Control group 4000 motorcycle owners or permit or endorsement holders from the Indiana Bureau of Motor
Vehicles database
Analysis sample size
Intervention group 739 (18)
Control group 588 (15)
Interventions Intervention integrity
No information given
Outcomes Probability of crash involvement
Length of follow up up to 5 years
Notes Study conducted in Indiana USA
Swaddiwudhipong 1998
Methods Randomised trial
Participants Initial sample size
Intervention group
all 16 villages of 3 sub-districts
(participants were practising motorcyclists licensed and unlicensed)
Control group
18 villages of 3 other sub-districtsAnalysis sample size
Intervention group 3 districts (100)
Control group 3 districts (100)
Interventions No rider education vs non-mandatory community based rider education consisting of theory on motorcycle injury
risk helmet use traffic laws vehicle regulations and traffic signs and a written and skill test leading to a licence
Intervention integrity
No information given
Outcomes Injuries fatalities
Length of follow up 2 years
40Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4350
Swaddiwudhipong 1998 (Continued)
Notes Study conducted in Tak Province rural Thailand
Waller 1992
Methods Cohort study
Participants Initial sample size
Intervention group 349 riders who succesfully completed the Department of Public Safety motorcycle operator
training in 1988
Control group 349 motorcyclists who never completed the DPS motorcycle operator training
Analysis sample size
Intervention group 349
Control group 349
Interventions No DPS motorcycle operator training vs non-mandatory DPS motorcycle operator training The DPS training is
the Motorcycle Safety Foundationrsquos Motorcycle Rider Course comprising eight hours of theory (lectures discussions
films and multiple choice tests) a textbook and 12 hours off-road practice
Intervention integrity
All 349 riders succesfully completed the rider training
Outcomes Police recorded crashes
Length of follow up 3 years
Notes Study conducted in
Characteristics of excluded studies [ordered by study ID]
Study Reason for exclusion
Anderson 1980a Study type review and process evaluation of a rider course
Aupetit 2008 Outcome proposal of a rider training curriculum that combines traditional training and a motorcycle riding
simulator
Awane 1999 Study type review on use of simulators in rider and driver training
Baldi 2005 Intervention not rider training but intervention integrity of rider training availabilty of rider education
implementation of graduated licensing system etc on a state-wide level
Batchler 1988 Study type description of a training course
Blanchard 2006 Study type not a study assessing the association between rider training and outcomes
Bonnett 2005 Outcome qualitative data regarding satisfaction with the training programme
41Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4450
(Continued)
Braver 2007 Outcome licensure rates
Clayton 1980 Outcome knowledge skills attitude
Council 1978 Outcome crash reduction in ANY vehicle not motorcycles exclusively
Crowther 2005 Study type qualitative study
Daniello 2009 Study type review of the effectiveness of motorcycle training and licensing
Dewanti 2007 Intervention emotional characteristics behaviour (ie not motocycle rider training)
Du Plooy 1990 Study type review of types of rider training courses in South Africa
Engel 1989 Study type uncontrolled before-after study
Goldenbeld 2002 Outcome skills
Goldenbeld 2004 Outcome knowledge and skills
Haworth 2000 Intervention balance between attitudinal andvehicle control skills acrosscurriculla of motorcycle rider training
courses
Haworth 2005 Study type review
Haworth 2006 Study type cross-sectional survey
Hill 1983 Study type review and description of training system
Imatake 1980 Study type course descriptionreview
Ingebritsen 1990 Intervention risk factors for motorcycle crashes
Jain 2009 Study type cross-sectional descriptive study of motorcycle crashes over a 5 year period
Jonah 1979 Study type cross-sectional survey Outcome scores on motorcycle operator skill test
Jonah 1981 Intervention evaluation of a motorcycle rider skill test not motorcycle rider training
Kadar 1990 Study type review of components of rider training courses
Kelsey 1986 Intervention evaluation of a motorcycle rider skill test not motorcycle rider training
Koch 1980 Study type reviewcourse description
Koch 1990 Study type description of course content
42Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4550
(Continued)
Labbett 2005 Study type essay about possible explanations for motorcycle deaths
Law 2005 Intervention targeted motorcycle safety programme not specifically rider training
Lowes 1990 Study type review and descriptions of rider training
Mahon 1976 Study type review of motorcycle rider training methods
Maxwell 1985 Outcomesrecommendations forprogramme operatorsand otherson howthe motorcycle trainingprogramme
works and how it can be improved
Mayhew 1996 Study type review
McKnight 1987 Study type evaluation of methodological and administrative aspects of a motorcycle operator skill test
McPherson 1976 Outcome knowledge and skills
Mohan 2004 Study type review of evidence of interventions to reduce road traffic injuries
Morgan 1980 Study type reviewcourse description
Motoki 1990 Study type proposal and development of rider training curriculum
Nairn 1992 Study type literature review
Newland 1980 Study type description of development of motorcycle safety course
Newland 1982 Study type descriptionreview of a rider training course
Newland 1990 Study type uncontrolled BeforeAfter study
Ormston 2003 Outcome changes in attitude and self reported behaviour
Osga 1979 Outcome knowledge attitudes and skills
Panou 2007 Intervention new simulation tool to teach driverrider
Peterson 1990 Study factor demographic details of course participants completing the course vs those not completing the
course
Prem 1984 Study type experimental
Study factor examines association between skills and performance in tests
Radin Umar 2006 Intervention motorcycle safety programme not specifically rider training
Rockwell 1990 Study type methodologically flawed study type
43Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4650
(Continued)
Rothe 1987 Study type qualitative study of rider training in general
Rowden 2007 Study type qualitative study
RSC Victoria 1993 Study type review of motorcycle rider training and education in Victoria
Russam 1979 Study type review of training programmes in Japan
Saffron 1981 Study type literature review
Schulz 1990 Study factor predictors of motorcycle crashes
Sexton 2004 Study type cross-sectional analytic analysis
Shephard 1986 Study factor reviewdescription of programme
Simpson 1990 Study type review of motorcycle rider trainingeducation
Sokolov 1990 Study factor review of motorcycle riding courses in USSR
Steffens 1988 Outcome self-reported behaviour experiences and researcher observed riding behaviour
Sudlow 2003 Study type review of motorcycle training in the UK
Tasmania 1995 Study type cross-sectional descriptiveprocess evaluation
Tetard 1985 Study type experimental
Outcome detection of indices
Thackray 1980 Outcome instructional effectiveness user acceptance administrative feasibility
Thompson 1994 Study type cross-sectional survey
Outcome attitudes and knowledge
Tobler 1990 Study type review of course curricula
Transport 1989 Intervention syllabus for rider training
Tuan 2007 Study type cross-sectional study (survey) which does not assess the effects of an intervention but examines
motocyclistsrsquo behaviour
Victor 1980 Study type review
Waterer 1990 Study type review of UK rider training issues
Weller 1989 Study type review of studies evaluating motorcycle safety and education measures
44Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4750
(Continued)
Wells 1981 Study type description of different methods of training
Winn 1986 Study type review of rsquohistoryrsquo of rider training
Wisher 1988 Study type cross-sectional survey
Wood 1987 Study type Interrupted Time Series but did not meet the criteria for an ITS as published by the Cochrane
Effective Practice and Organisation of Care Group
Yamazaki 1990 Study type cross-sectional survey
Outcome riding skills skills retention and attitude
Characteristics of ongoing studies [ordered by study ID]
Burgess 2010
Trial name or title UKs National Ride program
Methods Cohort study
Participants Motorcycle offenders
Interventions Police led diversion program
Outcomes Self reported attitudinal and behavioural measures
Starting date
Contact information
Notes Expected to be completed in early 2010
45Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4850
D A T A A N D A N A L Y S E S
This review has no analyses
A P P E N D I C E S
Appendix 1 Search strategy
We searched the following electronic databases
bull Cochrane Injuries Group Specialised Register (searched 22 Sept 2008)
bull CENTRAL (The Cochrane Library 2008 Issue 3)
bull TRANSPORT 1988 to 200706 (latest date available)
bull ATRI Australian Transport Index (to Sept 2008)
bull MEDLINE (Ovid SP) 1950 to Sept 2008
bull EMBASE (Ovid SP) 1980 to Sept 2008
bull CINAHL (EBSCO) 1982 to Sept 2008
bull WHOLIS (World Health Organization Library Information System) (to Sept 2008)
bull PsycINFO (Ovid SP) 1806 to Sept 2008
bull LILACS (BIREME) (Latin American and Caribbean Health Sciences) (1982 to Sept 2008)
bull ISI Web of Science Social Sciences Citation Index (SSCI) 1970 to Sept 2008
bull ERIC (searched 25 Sept 2008)
bull ZETOC (searched 25 Sept 2008)
bull SIGLE (searched 25 Sept 2008)
All of the databases were searched using the following text-word search strategy The strategy was adapted using either rsquorsquo or rsquo$rsquo as atruncation symbol
1 motor-cycl or motorcycl or motor-bik or motorbik or motor-scooter or motorscooter or scooter or moped or motocycl
or motocicl or moto
2 train or educat or instruct or program or course or qualif or learn or improv
3 1 and 2
H I S T O R Y
Protocol first published Issue 2 2005
Review first published Issue 10 2010
46Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4950
C O N T R I B U T I O N S O F A U T H O R S
Katina Kardamanidis was the first author of the protocol conducted the literature search extracted full studies assessed the studies
according to inclusion criteria assessed quality of the studies compared findings with the second author discussed disagreements with
the third author wrote the drafts and the final review
Rebecca Ivers was the second author of the protocol assessed the studies according to inclusion criteria assessed quality of the studies
compared findings with the first author edited the final draft and provided overall guidance for the process
Alexandra Martiniuk assessed quality of the studies compared findings with the first author discussed disagreements with the third
author and helped write the first draft and the final review
Mark Stevenson contributed to the protocol reviewed methods of studies and contributed to the final draft
Katrina Thistlethwaite conducted a second literature search extracted full studies and contributed to the final draft
D E C L A R A T I O N S O F I N T E R E S T
None known
S O U R C E S O F S U P P O R T
Internal sources
bull No sources of support supplied
External sources
bull National Health and Medical Research Council of Australia Australia
Salary support for Martiniuk Ivers and Stevenson
bull University of Sydney Sesqui Research Grant Australia
Direct research costs and salary support for Kardamanidis
D I F F E R E N C E S B E T W E E N P R O T O C O L A N D R E V I E W
There are no differences between the protocol and the review
I N D E X T E R M S
Medical Subject Headings (MeSH)
lowastMotorcycles Accident Prevention [lowastmethods] Accidents Traffic [lowastprevention amp control] Licensure Program Evaluation Wounds
and Injuries [mortality lowastprevention amp control]
47Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 5050
MeSH check words
Humans
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 1550
Risk of bias in the randomised trials
There were three randomised trials (Anderson 1980 Hall 1988Swaddiwudhipong 1998) Anderson 1980 and Hall 1988 had an
initial sample size of 41188 and 26464 motorcycle licence ap-
plicants respectively Both studies assessed the effectiveness of the
pre-licence MOST with a trainingcomponent Swaddiwudhipong
1998 evaluated the effectiveness of a community-based rider ed-
ucation programme in six districts
Allocation bias
Anderson 1980 reported using random birth date tables to allocate
participants to either the intervention or control group Hall 1988
reported the use of a computerised algorithm set in a central of-
fice but also used manual random assignment when the computer
went down Swaddiwudhipong 1998 reported lsquosimple random
samplingrsquo for the delivery of the intervention where the sampling
frame included nine districts Three districts were excluded since
they were too hard to access The remaining six districts were ran-
domly allocated to the intervention (three) and the control (three)
groups Systematic sampling wasthen used ordered by village size
for the villages within districts Allocation concealment was not
reported in any of the studies The descriptions of randomisation
in these studies suggest that there may have been limitations to
the process resulting in a lack of equivalence of the intervention
and control groups at baseline and therefore the possibility of
allocation bias Swaddiwudhipong 1998 reported that before theintervention the annual incidence rate of motorcycle-related in-
juries was slightly (but not statistically significant) higher in the
intervention areas than in the control areas Participants in the
intervention districts were also more likely to have a valid licence
and to wear a helmet more often than participants in the control
districts Anderson 1980 measured differences at baseline on the
participant and prior driving record variables and found these to
be negligible Hall 1988 did not report on baseline differences
between the intervention and control groups As the Anderson
1980 and Hall 1988 studies had sample sizes of over 40000 and
26000 respectively and used some form of randomisation it is
likely that any confounding factors at baseline would have been
equally distributed between comparison and intervention groups
Selection bias
After randomisation Anderson 1980 and Hall 1988 excluded
08 and 3 participants respectively because of assignment to
the wrong group administration of the wrong intervention or
other reasons The resulting lack of analysis by rsquointention-to-treatrsquo
can introduce biasby removing the effectivenessof randomisation
Nonetheless the percentage excluded was small and the number
of overall participants large so it is unlikely to have had a major
effect on the final results In the Swaddiwudhipong 1998 study
villages were systematically sampled into intervention or control
groups The intervention was implemented and outcomes of in-
jury and death were measured at a community level Interventionand control districts were randomly selected however there were
only three intervention and three control communities
Detection bias
None of the randomised trials reported blinding of outcome as-
sessors Data collection tools used by Anderson 1980 and Hall
1988 were driver records holding police recorded crashes and of-
fences Swaddiwudhipong 1998 used police and hospital records
at a district level and post-mortem examination for deaths None
reported on validity or reliability for any of the instruments
Attrition bias
Anderson 1980 and Hall 1988 had driver records available for
96 to 100 of participants They also collected self-reported
data but the response rate to the questionnaires was very low (low-
est response rate 42 and 30 respectively) Attrition bias in
Swaddiwudhipong 1998 was not an issue as outcome data were
available for all sub-districts included in the trial
The intervention reported in Anderson 1980 and Hall 1988 tri-
als had an unanticipated effect on the participants In both trials
an average of only 28 of licence applicants in the intervention
group eventually obtained their licence compared to 35 in the
control group This indicates that the licensing process in generalseems to have deterred applicants from obtaining a licence and
although no statistical tests of the difference were conducted this
deterrence appears even greater for those who had rider training
This is not true attrition bias as the studies did not experience
differential loss to follow up from outcome measurement Instead
differential attainment of the intermediary step of licensure pre-
cluded participants from being at risk for the outcomes of interest
they are not driving a motorcycle hence they are not at risk for
motorcycle-related crashes injury death or offences In short it is
not possible to assess whether a reduction in crashes in the inter-
vention group would be due to changes in riding habits as a result
of training or simply due to delayed or never attained licensure
Intervention integrity
Neither Anderson 1980 nor Hall 1988 reported on the consis-
tency in the way the intervention was administered the possible
lsquocontaminationrsquo of participants in the intervention group by other
types of formal or informal rider training or of participants in the
control group by the intervention Swaddiwudhipong 1998 did
not report on the number of villagers in each group that actually
attended the intervention However they did report on a number
of other potential training courses that were being implemented
at the same time
13Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 1650
Analysis
Anderson 1980 and Hall 1988 acknowledging the lsquodeterrent ef-fectrsquo described above conducted two separate analyses The first
analysis included the total study population of licence applicants
over two years from the time of their application (the population
denominator) The second analysis was performed on the licensed
(intervention and control) subgroups only two years from the
time of obtaining their licence (the licensed rider denominator)
As only a third were licensed and exposed to riding the motorcy-
cle it was impossible to determine whether reported differences
in crash rates across groups were due to the intervention or to
other lsquoconfoundingrsquo characteristics of those who were determined
in their efforts to obtain the licence When analysis was restricted
only to those who gained their licence the same difficulties re-
mained The benefits of the randomisation process were thus lost
due to this lack of licensure of riders during the process As there
were only three intervention and three control groups in the study
reported by Swaddiwudhipong 1998 analyses were hindered by
smallnumbers for outcome comparisons betweenthe intervention
and the control districts
Risk of bias in the cohort studies and non-randomised
trials
We discuss the methodological quality of cohort studies (seven
prospective studies (Raymond 1979 McDavid 1989 Waller 1992
Kloeden 1994 Billheimer 1998 Perrino 2002 Savolainen 2007)and seven retrospective studies (Osga 1980 Satten 1980 Jonah
1982 Mortimer 1984 Cooper 1988 Mortimer 1988 Davis
1997)) and non-randomised trials (Adams 1985 Leung 1987) in
the one section as the biases for the two study types are similar
The quality of these studies was in general poor
Selection bias
The confounding factors expected to have been addressed in the
studies were sex age riding experience prior driving records (of-
fences crashes) previous training and the following proxies for
risk taking behaviour helmet use speed and alcoholdrug use All
of these factors have been shown to be associated with an increasedrisk of motorcycle crash (Peden 2004 Zambon 2006) The pos-
sible source of bias in cohort studies and non-randomised trials
is that groups may differ on important characteristics at baseline
participants lsquoself-selectrsquo into either the rider training group by de-
ciding to undertake a rider training course or into the lsquocontrol
grouprsquo by deciding not to undertake a course Participants who
undertake a non-mandatory rider training course may differ from
those who choose not to on characteristics such as age sex rid-
ing experience and riding behaviour as well as other possible con-
founders For example riders who take a training course may be
more safety-conscious or on the contrary may do so because they
have previously experienced a crash or because they perceive that
they have poor riding skills Therefore ideally all participants in
the intervention group need to be included in the study as wellas a random sample of controls taken from the same population
as the intervention group If that is not possible intervention and
control groups either need to be carefully matched to make them
comparable or differences at baseline need to be adjusted for in
the analysis
Most of the studies using a non-randomised trial or cohort design
compared an intervention group to a control group that differed
on a number of important factors The studies in general failed to
adjust for important differences found at baseline and their results
cannot be meaningfully interpreted Ten studies (Raymond 1979
Osga 1980 Satten 1980 Jonah 1982 Mortimer 1984 Mortimer
1988 McDavid 1989 Waller 1992 Davis 1997 Savolainen 2007)
assessed intervention and control groups that were not drawnfrom the same population In most of these cases the intervention
group was selected from a list of training graduates who were usu-
ally young female and less experienced riders whereas controls
were recruited from a general population of motorcycle riders who
were usually older male and experienced riders For three stud-
ies it was unclear how they were selected (Leung 1987 Cooper
1988 Perrino 2002) Only Kloeden 1994 recruited groups from
the same population of motorcycle applicants and Adams 1985
from a group of professional riders Nevertheless due to non-ran-
dom selection of intervention and control groups differences at
baseline are likely even though study groups were from the same
population Three studies adjusted for some confounding fac-
tors found at baseline through matching (McDavid 1989 Waller1992 Billheimer 1998) However matching has limitations in
termsof controlling for multiple confounders especially those that
are unknown to influence the effect of rider training on the out-
comes One study matched the control and intervention groups
by excluding controls (approximately 3500 (66 of initial con-
trol group) that differed from the intervention group) (Kloeden
1994) Five studies (Osga 1980 Satten 1980 Adams 1985 Davis
1997 Perrino 2002) did not adjust for differences between study
groups at baseline either in the study design or analysis Results
from these studies are therefore not reliable The other six stud-
ies (Raymond 1979 Jonah 1982 Mortimer 1984 Cooper 1988
Mortimer 1988 Leung 1987 McDavid 1989 Billheimer 1998)
adjusted for some differences between groups at baseline by eitherbi-variate or multi-variate analyses or stratification and standardi-
sation of riding exposure (outcomemiles driven) or stratification
by age group The limitation of stratification is that it is difficult to
control for many variables simultaneously because a large number
of strata will need to be generated resulting in small numbers that
cannot be meaningfully analysed Savolainen 2007 developed sta-
tistical models to assess the probability of a crash after taking the
rider training course in an intervention and control group while
examining differences by sub-groups (eg age) Participation rates
were also very low in this study with only 5584000 (1395)
in the eligible intervention group and 5884000 (147) for the
14Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 1750
control group completing surveys
Detection bias
None of the studies reported blinding of outcome assessors to the
intervention or control status of the participants Seven studies
used driver records containing police recorded data as their main
measure of outcome (Leung 1987 Cooper 1988 McDavid 1989
Waller 1992 Kloeden 1994 Davis 1997 Billheimer 1998) An-
other four studies (Raymond 1979 Satten 1980 Mortimer 1984
Mortimer 1988) only used self-report Raymond 1979 was the
only study using self-report as the main outcome measurement in
which study instruments were validated Self-report might prevent
the underreporting that has been described for the use of police-
driver records (see rsquoRisk of biasrsquo in the randomised trials above)but has other risks of bias especially if questionnaires or inter-
view methods have not been validated and tested for reliability
Recall bias arises from mistakes in recollecting events both be-
cause of failures of memory and because of looking at things lsquowith
hindsightrsquo and possibly changed views (httpwwwcochraneorg
resourcesglossaryhtm) As many studies were retrospective risk
of recall bias is likely and may have increased with length of follow
up For studies that used self-reported outcomes only it is also
possible that social desirability response bias played a role where
participants in the training group may have felt it was lsquoexpectedrsquo
to report good driving histories Multiple methods to measure the
same outcome for both groups decreases the risk of detection bias
for example because self-reported answers can be cross checkedwith other sources of information such as driver records
In this regard three studies (Osga 1980 Jonah 1982 Perrino
2002) used both self-report and driver records Osga 1980 found
that for both intervention and control groups more crashes were
reported through self-report than on state crash records an aver-
age of 54 of self-reported crashes could be found back on state
crash records One industry-based study (Adams 1985) used oc-
cupational health records which were not validated but which are
unlikely to suffer from underreporting due to work cover claims
Savolainen 2007 used information on rider crashes for a five-year
period but did not specify from where this information was ob-
tained
Attrition bias
The majority of studies (Raymond 1979 Satten 1980 Adams
1985 Leung 1987 Kloeden 1994 Davis 1997 Billheimer 1998
Perrino 2002) inadequately reported on losses of participants to
the study We can therefore not be certain that there were no sys-
tematic differences in the characteristics between the intervention
and control group participants Savolainen 2007 reported the re-
sponse rates to their baseline survey but did not report on the per-
centage of outcome data that was available for their study sample
A similar deterrent effect of rider training as described for the ran-
domised trials (Anderson 1980 Hall 1988) occurred in two stud-
ies Jonah 1982 reported that 65(14992310) of the trained rid-
ers stopped riding compared to 47 (9492029) of the untrainedriders They excluded non-active and non-licensed riders before
the study commenced so the differential choice to continue riding
after training did not affect their analyses Leung 1987 reported
that the percentage of participants in the intervention group ob-
taining their licence was 50 (no number provided by authors)
lower than in the control group This may have biased the mea-
surement of outcomes because unlicensed participants will not be
exposed to riding the motorcycle and therefore not be at risk of
being involved in a crash or violation As explained above this
makes it difficult to assess whether the reduction in crashes in the
intervention group is due to a change in driving habits or simply
due to delayed licensure
Mortimer 1984 and Mortimer 1988 describe that about 44 and30 of course graduates respectively were excluded from analysis
because they had not ridden a motorcycle in the previous year
However because these studies were retrospective and the controls
were activeriders at the time of recruitment a decision on whether
the deterrent effect was differential between the two groups could
not be made Though not reported quantitatively Kloeden 1994
also describes the effectof rider training deterringindividuals from
becoming licensed and hence exposure to riding and risk of crash
death or offences involving a motorcycle Osga 1980 reported
that 27 of the trained riders did not ride after the course and
excluded these from the analysis
Use of different data collection strategies may have contributed to
differential loss to follow up in the cohort studies Satten 1980for example interviewed the intervention group over the tele-
phone but mailed questionnaires to the control group Mortimer
1984 and Mortimer 1988 distributed questionnaires in person to
the control group but mailed them to the intervention group
This may have played a role in the observed differential response
rates where more subjects responded in the personally approached
groups
Survival bias
For those studies that assessed offences crashes and injuries but
not death survival bias may have also played a role Motorcyclists
who were fatally or seriously injured in crashes during the follow-
up period after the intervention would have been less likely or
unable to respond to surveys This could bias the effect measures
towards or away from the null None of the studies wrote about
this issue or examined the characteristics of those lost to follow up
in order to address the possibility of survival bias
Intervention integrity
Although the interventions were intended for the entire lsquoexposedrsquo
groups most studies reported difficulty in tracking participants
whoreceivedthe interventionand those whodid not Adams 1985
15Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 1850
estimatedthat 62of thetotalstudysample received the interven-
tion The remaining studies did not report the percentage of thestudy population that received the intervention as was intended
or reported that this percentage was unknown (Raymond 1979
Kloeden 1994 Satten 1980 Jonah 1982 Mortimer 1984 Leung
1987 Cooper 1988 Mortimer 1988 Billheimer 1998 Perrino
2002 Savolainen 2007) Only Raymond 1979 reported the degree
to which their intervention was delivered as intended No studies
reported on possible lsquocontaminationrsquo of participants by other inter-
ventions than the one to which they were allocated Because Osga
1980 Satten 1980 Jonah 1982 Mortimer 1984 Cooper 1988
Mortimer 1988 Waller 1992 and Davis 1997 were retrospective
studies and participants were recruited from a list of graduates it
was known that they had completed and passed the course
Analysis
In some studies some participants who reported at baseline that
they were untrained decided to enroll in a rider training pro-
gramme after the baseline survey The non-randomised trial by
Adams 1985 experienced this move of subjects from the untrained
to the trained group but failed to conduct analyses by lsquointention-
to-treatrsquo This can bias the results in randomised and non-ran-
domised trials if participants who changed groups differ signifi-
cantly with regard to potentially confounding characteristics On
the other hand Perrino 2002 also experienced this move of sub-
jects from theuntrained to the trained groupmore than 100riders(5 of the total sample) originally stated that they were untrained
at baseline but then received training between the baseline and
follow-up surveys Yet since the Perrino 2002 study was a cohort
their analyses then proceeded by including those riders who were
lsquonewlyrsquo trained as exposed Riders who were lsquonewlyrsquo trained since
baseline but with a programme other than the one under study
were excluded from the analysis Nevertheless confounding fac-
tors were not accounted for in the analyses
McDavid 1989 matched exposed with unexposed subjects but
could not find suitable matches for the majority (60) of their
subjects They could therefore only conduct analysis on the result-
ing40 (139346) which is an inappropriate andinefficient anal-
ysis Some studies that reported on the issue of non-active licensedriders(Satten 1980 Jonah 1982 Mortimer 1984 Mortimer 1988
Kloeden 1994) excluded non-active riders from analysis intro-
ducing selection bias as it is likely that riders who continued to
ride differed with regards to potentially confounding character-
istics across intervention and control groups Kloeden 1994 for
example conducted several analyses to assess the effect of various
sub-sets of the control group Original analyses included all con-
trols and led to erroneous results since only a small proportion of
controls hadgone on to get a licence after their learnerrsquos permitand
were actually exposed to the possibility of experiencing a crash
Subsequent analyses compared only intervention and control sub-
jects who had a licence Waller 1992 reported crashes by person-
days of exposure but did not explain how they could be sure that
each study intervention and control participant was an active riderand actually exposed to the risk of experiencing a crash
Potential confounders that are unequally distributed by study arm
at baseline need to be dealt with in the analysis Satten 1980
Adams 1985 and Davis 1997 did not take into account any con-
founding factors either by design or in the analysis McDavid
1989 Waller 1992 and Billheimer 1998 accounted for potential
confounding factors at the design stage by matching the subjects
in the intervention and control groups Other studies (Raymond
1979 Mortimer 1984 Mortimer 1988 Perrino 2002 and also
Waller 1992) adjusted for one or two confounders by stratifica-
tionor standardisation (eg crashesmiles driven age group) Osga
1980 conducteda univariate analysis of crash rates in which no po-
tential confounders were accounted for but also calculated cumu-lative crash rates by mileage travelled Only Jonah 1982 Cooper
1988 and Savolainen 2007 conducted a multivariate analysis
Risk of bias in the case-control studies
A case-control design is suitable for the assessment of risk fac-
tors for outcomes such as motorcycle crashes injuries or deaths
but not appropriate for the evaluation of an intervention such
as motorcycle rider training It can potentially result in spurious
conclusions regarding the efficacy of the intervention because the
intervention has not been applied at random to an intervention
and control group Even with the control of possible confounders
using a case-control design to evaluate an intervention can lead tothese kinds of results since those who choose to undertake rider
training may differ on important prognostic factors that affect the
outcome compared to those who do not choose to participate in
rider training
Selection bias
A potential source of bias in case-control studies as with all other
study types where the intervention is not randomly allocated to
participants is that comparison groups are likely to differ on po-
tentially confounding characteristics This means that ideally all
cases need to be included in the study as well as a random sample
of controls taken from the population that gave rise to the cases If that is not possible cases and controls either need to be carefully
matched to make themcomparable or differences at baseline need
to be adjusted for in the analysis All four case-control studies were
population based Kraus 1975 included all fatalities and injured
cases leading to hospitalisation and appropriately recruited a ran-
dom sample of controls from the same population Hurt 1981 did
not manage to recruit all or a random sample of cases Only half
of the notifications of crashes to the police were received in time
by the researchers the research team for unreported reasons was
only able to respond to half of those again Data collection was
therefore completed for only about a fifth of all cases Neverthe-
less this was still a large sample of 900 cases Haworth 1997 using
16Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 1950
a similar study design was more likely to have included all cases
but this is not entirely clear from the study reported It is unclearfor example whether fatalities were included as they did not use
coronerrsquos reports for the recruitment of cases The MAIDS 2004
study was conducted in five European countries but it was unclear
how cases were recruited whether they were all cases during the
two years that the study lasted or a random sample of all cases
what the response rates were etc Controls were recruited from
randomly selected petrol stations but it was likewise not clear what
the eligibility criteria for inclusion in the study were how they
were recruited (eg randomly) and what the response rate was
Hurt 1981 and Haworth 1997 both matched controls to cases by
recruiting themat the sametime ofdayandsamedayof weekas the
crash occurred Hurt 1981 however recruited controls two years
later andit is likelythat there would have beendifferences betweenthe two comparison groups One important difference was that
during data collection for cases there was very little specialised
motorcycletrainingavailable in the studyregionwhereas two years
later when control data were collected there was some training
available
These four case-control studies also suffered from selection bias
due to potential differences between those that gave consent to
participate in the study and responded to interviews andquestion-
naires and those that did not None of the studies have assessed
such differences although Hurt 1981 did acknowledge that con-
trols giving consent to cooperate may have differed from those
motorcyclists who did not
Detection bias
Measures that can be taken to reduce detection bias in case-con-
trol studies are blinding of interviewers for case or control status
blinding of the participant for the study factor of interest validat-
ing measurement instruments and using these in an identical way
for cases and controls and cross checking self-reported informa-
tion with other more objective sources such as registers None of
the studies reported blinding of the intervention assessors to case
or control status but all used consistent measurement methods for
both case and control groups
All case-control studies used self-report as their main measure toassess whether cases and controls had received rider training and
if so the type of training The major potential source of bias in
case-control studies when self-report is used as a measure is recall
bias One way of reducing recall bias is by double-checking self-
reported data with more objective data Kraus 1975 also used po-
lice accident reports for information on rider characteristics but
such information was not available about rider training Haworth
1997 and MAIDS 2004 did not report on any measures to limit
recall bias Hurt 1981 reported further questioning of cases by
the interviewer when answers did not make sense or when par-
ticipants seemed to contradict themselves as part of the in-depth
unstructured interview technique This qualitative research tech-
nique can result in valuable information but as interviewers were
not blinded to case or control status of the participants it may have led to more thorough examination of cases compared to the
controls and thus differential detection bias
Intervention integrity
The level of detail about the rider training provided in the case-
control studies differed significantly but was minimal for each
Kraus 1975 simply asked whether lsquoanyrsquo motorcycle rider training
had lsquoeverrsquo been carried out Hurt 1981 asked participants whether
they hadbeen self taught informally taught or completed formally
taught rider training courses MAIDS 2004 asked whether partic-
ipants had had pre-license training additional (non-compulsory)
training other training or no training whereas Haworth 1997asked about the number of courses that a participant had com-
pleted whetherthey were basic intermediate or advanced courses
and how long ago the training had been done None of the case-
control studies reported collecting information about further de-
tails about any training programmes (eg type of training cur-
riculum duration of training)
Analysis
Kraus 1975 accounted for confounding factors through the exclu-
sion of females and stratification by age and frequency of motor-
cycle use Hurt 1981 matched cases and controls for time of day
and day of the week that the crash occurred and did not adjust forany confounding factors in the analysis Haworth 1997 matched
groups for time of day and day of the week that the crash occurred
and stratified by age blood alcohol level and licence status in the
analysis None of the studies used multivariate methods MAIDS
2004 did not provide any information about any participant char-
acteristics other than the study exposures of interest and did not
describe any methods to adjust for potential confounding factors
Effects of interventions
We have presented the results in a descriptive manner due to the
substantial heterogeneity in study types interventions and out-comes and the generally poor quality of the studies We have
grouped the interventions by mandatory and non-mandatory pre-
licence rider training and post-licence rider training and further
by study type and by outcome (crashes injurious and fatal crashes
and offences) We discuss results that are reported as statistically
significant (whether a P value was reported or not)
Pre-licence rider training - mandatory
Randomised trials
17Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2050
Crashes
Total group of participants
Anderson 1980 found a reduction in motorcycle crash rates in
the intervention group (MOST skill test with remedial training)
compared to the control group (who received the skill testwithout
a training component) The percent differenceranged from 143
six months after the programme (P lt 005) to 211 (P lt 0001)
after one year and 127 (P lt 001) after two years Hall 1988
found no statistically significant difference in crash rates between
intervention groups at any point in time
Subgroup of participants who obtained their licence only
Anderson 1980 found a reduction in motorcycle crash rates in the
intervention arm but this was statistically significant at one year
only (24 to 218 reduction P lt 005) Hall 1988 found no
statistically significant differences
Offences
Total group of participants
The percentage of offences in Anderson 1980 after six months
and one and two years was slightly higher in the intervention
arm compared to the control group but these differences were
not statistically significant Likewise the study by Hall 1988 did
not find a statistically significant difference in offences between
groups
Subgroup of participants who obtained their licence only
Hall 1988 found no significant difference in offence rates between
the intervention and control groups Anderson 1980 found thatthe percentage of offences increased in the intervention arms com-
pared to the control group This difference was statistically signif-
icant after two years (51 to 115 increase P lt 005)
Cohort studies
Crashes
Neitherof the cohort studies (Kloeden1994 and Billheimer1998)
found a statistically significant difference in crash rates between
the intervention and control groups
Offences
Billheimer 1998 (for the California Motorcyclist Safety Program
course) and Perrino 2002 did not find a statistically significant
difference in offence rates between the intervention and control
groups
Case-control studies
Crashes
MAIDS 2004 found that a slightly higher percentage of controls
(not involved in a crash) had had pre-license rider training than
cases (involved in a crash) This difference was not statistically significant (50 of controls vs 477 of cases OR 091 95 CI
076 to 110)
Pre-licence rider training - non-mandatory
Randomised trial
Injuries and deaths
Swaddiwudhipong 1998 found a significantly lower injury ratein the intervention villages (1051000 population 95 CI 92
to 120) compared to the control villages (1691000 population
95 CI 152 to 185) one year after the intervention This differ-
ence was not maintained two years after the intervention (127
95 CI 112 to 143 intervention 151 95 CI 135 to 167
control) There was no significant difference in death rates be-
tween groups
Cohort studies
Crashes
Cooper 1988 foundno statistically significant differencein crashes
between the intervention and control groups McDavid 1989
found a lower crash rate in the intervention group (14139) com-
pared to the control group (23139) but no test of statistical sig-
nificance was undertaken
Injuries and deaths
McDavid 1989 found that the intervention group had fewer in-
jurious crashes (814 crashes) compared to the control group (16
23 crashes) but no statistical significance test was reported
18Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2150
Offences
Cooper 1988 foundno statistically significantdifference in offencerates between the intervention and control groups McDavid 1989
foundno statistically significantdifference in offence ratesbetween
the intervention and control groups
Non-mandatory rider training
Non-randomised trials
Crashes
Neither Adams 1985 nor Leung 1987 found a statistically signif-icant effect of rider training on crash rates
Offences
Leung 1987 found a statistically significant lower offence rate in
the intervention group compared to the control group for both
males and females and for all age groups except for women 35
years or older (P values not reported)
Cohort studies
Crashes
Osga 1980 in the univariate analysis reported a lower percentage
of crashesin theintervention group (that continued to ride) (28
168595) than in the control group (34 112329) In the cu-
mulative crash rate analysis however Osga1980 found statistically
significant (at the P lt 001 level) higher crash rates in the interven-
tion group than in the control group at allmileage intervals (1000
to 29999 miles) travelled After riding more than 30000 miles
(when over 50 of the trained riders had been involved in a crash
compared to approximately 40 in untrained riders) this differ-
ence was no longer statistically significant Satten 1980 found a
lower percentage of crashes in the intervention group (13 969)
compared to the control group (25 2079) but the differencewas not assessed statistically Davis 1997 found a lower crash rate
in trained riders but it was unclear how this was calculated be-
cause the size of the control group was not reported Jonah 1982
and Billheimer 1998 (for the Experienced Rider Course) found no
statistically significant difference in crashes between the interven-
tion and control groups Although the difference between those
who completed the Experienced Rider Course and no course in
Billheimer 1998 neared significance at six months after the course
(P = 007) large P values were observed for one year after (P =
030) Raymond 1979 found a statistically significant higher crash
rate for the intervention group (1861000 miles ridden) com-
pared to the control group (1181000 miles ridden) in both males
and females (P lt 0001) especially during the first 5000 miles of
riding Waller 1992 found a statistically significant higher crashrate for the intervention group (27 crashes in three years or 070
crashes10000 person days) compared to the control group (5
crashes in three years or 013 crashes10000 person days) with a
crude relative risk of 538 (X 2=1513 plt005) Mortimer 1984
and Mortimer 1988 found no statistically significant difference in
crashes between the intervention and control groups
Savolainen 2007 found that the intervention group (those who
completed the Basic Rider Course) was 44 more likely to be
involved in crashes than the control group and those who com-
pleted the course multiple times were an additional 180 more
likely to be involved in a crash This finding was based on low
numbers and no P value or confidence level was given for either
result
Injuries and deaths
Satten 1980 also found a lower percentage of injurious crashes
(33 39) in the intervention group compared to the control
group(60 1220) although again no statistical significance test
was reported
Offences
Jonah 1982 did find a significantly lower self-reportedoffence rate
in the intervention group (15) compared to the control group(34 P lt 001) (no numbers given) Satten 1980 also found
a lower self-reported offence rate in the intervention group (1
71 versus 19100) but again this was not tested for statistical
significance Mortimer 1984 Mortimer1988 and Billheimer 1998
found no statistically significant differencein offence ratesbetween
the intervention and control groups
Case-control studies
CrashesMAIDS 2004 found that a slightly higher percentage of cases
(involved in a crash) had had additional non-mandatory rider
training than controls (not involved in a crash) This difference
was not statistically significant (17 of cases vs 12 of controls
OR 146 95 CI 064 to 337)
Rider training unspecified
Case-control studies
19Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2250
Injuries and deaths
Hurt 1981 found that the odds of injurious crashes were lower inthose who had received professional training compared to those
who had not Kraus 1975 found that in the younger age group
(less than 20 years) and the older age group (over 35 years) a larger
percentage of injured riders (12 and 21 respectively) reported
prior motorcycle rider training compared with the uninjured con-
trol group (6 and 5 respectively) (P lt 001) (no numbers
given) Haworth 1997 reported several different outcomes one
beingthat having completed at leastone trainingcoursecompared
to no training courses at all had no significant effect on the odds
of crashing (afteradjustment for age blood alcohol level or licence
status (learner permitprobationaryfull)) When comparing three
different levels of training courses (beginner intermediate and ad-
vanced) with no training no significant differences were found
in the odds of crashing However after adjustment for age and
blood alcohol level completing an advanced course as compared
to a beginners course was associated with a significant decrease in
the odds of severe motorcycle crashes resulting in serious injury or
death (OR 02 95 CI 01 to 08)
D I S C U S S I O N
The objective of this review was to quantify the effectiveness of
pre- andpost-licence motorcycle rider training on the reduction of
traffic crash involvement injuries deaths and offences committedby motorcycle riders Most included studies were cohort studies
There were a few randomised and non-randomised trials and case-
control studies
All studies compared either mandatory or non-mandatory formal
training with no formal training except for Haworth 1997rsquos case-
control study that compared levels of training (beginner interme-
diate and advanced) to each other
We found studies that evaluated the effectiveness of manda-
tory pre-licence training in reducing crash and offence rates
(Anderson 1980 Hall 1988 Kloeden 1994 Billheimer 1998
(the basic California Motorcyclist Safety Program component of
the study) Perrino 2002) non-mandatory training that when
passed resulted in a licence (Cooper 1988 McDavid 1989
Swaddiwudhipong 1998) non-mandatory training that could be
taken by motorcycle riders with or without a licence (Raymond
1979 Osga 1980 Satten 1980 Jonah 1982 Mortimer 1984
Adams 1985 Leung 1987 Mortimer 1988 Waller 1992 Davis
1997 Billheimer 1998 (the Experienced Rider Course compo-
nent of the study)) and unspecified rider training on crash injury
and death rates (Kraus 1975 Hurt 1981 Haworth 1997 MAIDS
2004)
The studies provided conflicting evidence regarding motorcycle
rider training which is likely due to the poor design quality and
reporting of the studies Most of the studies included in the review
were likely to be adversely affected by significant selection biasalong with substantial detection bias(measurement and recall bias)
which limits our ability to make conclusive comments
Mandatory pre-licence rider training
There were two RCTs three cohort studies and one case-control
study that examined pre-licence ridertraining The two RCTs eval-
uated the Motorcycle Skills Operating Test (MOST) (Anderson
1980 Hall 1988) Anderson 1980 found a statistically significant
reduction in crashes in the intervention group at six months one
and twoyearsfollow up However when only participantswho ob-
tainedtheirlicence were included in theanalysis thereduction wasstatistically significant only at one year follow up The other RCT
(Hall 1988) found no statistically significant differences in crash
rates at any point of time neither in the total population group
nor in the licensed subgroup Neither of the two cohort stud-
ies assessing crash rates found a statistically significant difference
in crash rates between intervention and control groups ( Kloeden
1994 Billheimer1998) andthe case-control study (MAIDS 2004)
also did not find a statistically significant difference in the odds of
experiencing a crash between trained and untrained riders With
regards to offences neither of the RCTs found a statistically sig-
nificant difference in offence rate between the intervention and
control groups in the analysis of the total group of participants
entering the trial However Anderson 1980 found a statistically significant increase in offence rates in the intervention group but
only after two years of follow up andonly in the analysis of partici-
pants who obtained their licence The two cohort studies assessing
offence rates found no statistically significant difference between
the groups (Billheimer 1998 Perrino 2002)
None of the five studies (RCTs or cohort) found a reduction in
crashes or offences following mandatory pre-licence rider training
except for Anderson 1980 who observed a statistically significant
reduction in crashes in the intervention group at one year in both
analyses It is not clear why both RCTs that evaluated the same
MOST and training intervention observed different results Both
studies enrolled large samples and observed a similar uptake of
licensure following the intervention It is possible that differencesmay have arisen due to different study samples (California ver-
sus New York) or that the intervention delivery differed between
studies
In both trials of the Motorcycle Skills Operating Test including
rider training only a third of motorcycle licence applicants ob-
tained their motorcycle licence It is possible that a licensing pro-
cess including a skills test may reduce crashes through decreased
exposure
Non-mandatory pre-licence training
20Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2350
Three studies assessed the effect of non-mandatory pre-licence
training The Cooper 1988 cohort study found no difference incrash or offence rates between the trained and untrained groups
We cannot however be confident about this result as the study
sufferedfrom selectionbiaswhichcouldhave biasedthe measure of
effect toward the null The RCT conducted by Swaddiwudhipong
1998 examined injury and death rates and found a statistically
significant lower injury rate in the intervention villages after one
year of follow up but no longer at two year follow up There
was no statistically significant difference in death rates between
intervention and control villages Our confidence in the results of
this study are hindered by the inadequacy of the randomisation
process that led to significant differences between the intervention
and control districts on two important variables The intervention
districts were more likely than the control districts to have a validlicence andto wear a helmetSincethese differences were nottaken
into account in later analyses they could explain the differences
observed in injury rates between districts Furthermore with only
six districts randomised the sample size was probably too small
to detect a difference between intervention and control groups
had there been one The McDavid 1989 cohort study found a
lower percentage of injuries in the intervention group although
a statistical test was not conducted McDavid 1989 drew their
intervention and control groups from different populations but
did adjust for confounders although they only analysed 40 of
the initial sample (139 trained and 139 untrained) Because of
these substantial weaknesses we are unable to draw conclusions
from this studyAn overall effect of pre-licence training on the reductionof crashes
injuries deaths and offences was not found although one should
be aware that the research of this issue to date has had significant
flaws However a mandatory skill test (such as the MOST) in
combination with and even without pre-licence training as part
of a licensing process seems to reduce the number of individuals
completing the licensing procedure and therefore indirectly re-
duces crash rates through a reduction in exposure to motorcycle
riding Making licensure a more difficult process may therefore be
seen as advantageous if it delays licensure and therefore reduces
exposure It is possible that the group who chose not to continue
riding a motorcycle would not have been the group to experience a
crash or commit an offence if they had continued to ride anyhowfor example because they are more safety conscious and this was
the reason for their giving up riding in the first place If this is the
case deterring individuals from completing their licence and rid-
ing may not necessarily decrease overall crashes deaths or offence
rates In addition other potentially adverse effects of complicat-
ing a licensing procedure need to be considered such as increases
in unlicensed riding In Thailand non-mandatory rider training
offered as a community intervention may have decreased injury
rates but this needs to be confirmed through further research
Non-mandatory training
The majority of the studies assessed non-mandatory training
(training independent of any licensing procedure) and the results
were varied The two non-randomised trials assessing crash rates found no statistically significant difference in crash rates between
the intervention and control groups (Adams 1985 Leung 1987)
Of the nine cohort studies Satten 1980 found a lower percentage
of crashesin the intervention group although it is unclear whether
this was a statistically significant result Davis 1997 also found a
lower percentage of crashesin theinterventiongroup although it is
unclear with which control group the intervention group wascom-
pared Raymond 1979 Osga 1980 Waller 1992 and Savolainen
2007 found a statistically significant higher crash rate in the in-
tervention group and the remaining cohort studies (Jonah 1982
Mortimer 1984 Mortimer 1988 Billheimer 1998) found no sta-
tistically significant difference in crash rates between intervention
and control groups No conclusions can be drawn from these stud-
ies with confidence The Satten 1980 and Davis 1997 studies did
not adjust for any confounding factors even though important
differences existed between the intervention and control groups
Although Raymond 1979 and Waller 1992 adjusted for a num-
ber of confounders at baseline (Waller 1992) and in the analy-
sis (Raymond 1979 Waller 1992) the intervention and control
groups were drawn from such different populations that it is pos-
sible that unmeasured confounders were responsible for the sta-
tistically significant increased risk of crashes observed in the inter-
vention group The observed difference could also be attributed totheobserved differential loss to followup (Raymond 1979) which
could bias the measure of effect towards or away from the null
The Osga 1980 study found a protective effect of rider training
in the univariate analysis but statistically significant higher crash
rates in the intervention group when accounting for riding expo-
sure Sex (a significantly higher proportion of females in the in-
tervention group) and other potentially confounding factors that
could explain thedifferencebetween the groups were howeverun-
accounted for Finally Savolainen 2007 did control for potential
confounding variables yet it is not clear whether variables were
controlled for simultaneously or one at a time One case-control
study (MAIDS 2004) found no statistically significant reduction
in the odds of experiencing a crash between trained and untrainedriders but the quality of this study was poor mainly because the
analysis did not account for any confounding factors
One cohort study (Satten 1980) assessed the effect of non-manda-
tory rider training on injurious crashes and found a lower percent-
age of injuries in the intervention group although a statistical test
was not conducted As mentioned above the Satten 1980 studyrsquos
validity is reduced due to its lack of adjustment for confounding
factors
The only non-randomised trial that assessed offence rates (Leung
1987) found a statistically significant lower offence rate in the in-
tervention group and of the five cohort studies that assessed of-
fence rates (Satten 1980 Jonah 1982 Mortimer 1984 Mortimer
21Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2450
1988) one study (Jonah 1982) also found a statistically significant
lower offence rate in the intervention group However we have lit-tle confidence in the results from Leung 1987 since potential con-
founding factors were not appropriately addressed Jonah 1982
did conduct a multivariate analysis but was hindered by an initial
low response rate The other four cohort studies did control for
several potential confounders and found no difference in offence
rates following non-mandatory post-licence rider training
The non-randomised trials and cohort studies evaluating non-
mandatory rider training were at high risk of substantial selection
bias due to the non-random allocation of participants as well
as low participation rates and attrition bias and could therefore
not provide conclusive evidence with regard to the effect of the
training It should be noted that despite the biases and other issues
with this group of studies most didnot detectmajor effects relatedto rider training in either direction This may suggest that any
potential impact of rider training will be modest
Studies examining unspecified modes of ridertraining
Three case-control studies assessed whether injured or deceased
motorcyclists were more or less likely to have had rider training
compared to control motorcyclists All had conflicting results
Hurt1981 foundthe odds of havingan injuriouscrash waslower in
those who had received rider training whereas Kraus 1975 found
that the odds of having an injurious crash were higher in youngerand older trained riders Haworth 1997 who compared advanced
versus beginner training found no effect on motorcycle injuries
or deaths although they did find that those who had completed
advanced training had reduced chances of crashing compared to
those who had done a beginnersrsquo training course Similar issues
with uncontrolled confounding and biases of selection as described
above arose with the three case-control studies Hurt 1981 suffers
from inadequate control of confounding via poor matching and
both the Kraus 1975 and the Haworth 1997 findings were only
controlled for limited factors by means of stratified analysis It is
likely that these limitations are the reason for the contradictory
directions of effect observed for the three case-control studies
Methodological weaknesses of the studies
In summary most studies investigating the effectiveness of mo-
torcycle rider training on reducing crashes injuries and death suf-
fered from serious methodological weaknesses
The majority of studies were non-randomised and were there-
fore limited mainly by selection bias Although the approaches of
matching stratification or regression analyses can minimise bias
(Fleiss 2004) they were poorly employed in the reviewed studies
The inadequate adjustment for confounders in particular adds
to our lack of confidence in the studiesrsquo results since it has been
demonstrated that individuals who choose to take motorcycle rider
training differ from those not taking rider training For instance
trained riders often differ from controls in age sex and experi-ence usually ride less and may have a more positive attitude to
safety as indicated by the reported more frequent wearing of seat
belts (Mortimer 1982)
The two large randomised trials reported that a large proportion
of their study group were never exposed to the risk of offences or
crashes because they never attained licensure This made it impos-
sible to assess whether a reduction in crashes in the intervention
group was a result of training or due to delayed or never attained
licensure
In some studies the sample sizes were too small to detect any sta-
tistically significant differences Follow-up time after training dif-
fered between studies and was an impediment to observing any
real effects of training if they existed As riding experience ac-cumulates over time any differences in outcomes between those
trained and untrained would diminish in studies with lengthy fol-
low-up times This dilution of any possible effect of rider training
may have also occurred in studies that mixed both novice and ex-
perienced riders in the study groups since it is hypothesised that
novice riders serve to benefit most from rider training This would
increase the difficulty in finding any effect of rider training par-
ticularly when sample sizes are small
Detection bias may also account for some of the differences ob-
served between trained and untrained riders Detection bias would
have occurred if the self-report hospital driver or police records
used to record the outcomes of crash injury death or offences
differed between the trained and untrained riders Research hasdemonstrated that not all crashes are reported to police for ex-
ample small crashes or off-road crashes (Kraus 1975) even serious
injury crashes have been found to be underreported (Anderson
1980) However this would only introduce detection bias if the
reporting of outcome differs for trained and untrained riders This
does appear possible in that responsible riders may be more likely
to enroll in a rider training course and also may be more likely
to report crashes to police than irresponsible riders meaning that
the number of crashes from riskier riders could be underrepre-
sented on records An analysis to investigate the effectiveness of
rider training may then show that rider training is not effective in
reducing crashes
Rather than using reported outcomes some studies used self-re-ported survey data although this technique suffers from non-
response biases In fact two studies reported differences in de-
mographic characteristics in response rates to questionnaires
(Anderson 1980 Hall 1988) Self-reported data also introduce
the risk of recall bias survival bias and social desirability response
bias Recall bias may have been equally distributed among trained
and untrained riders howeverit is hypothesised that trained riders
may have felt it was lsquoexpectedrsquo to report good driving histories
hence leading the rider training to appear effective
22Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2550
Limitations of this review
Limitations of this review were that we found very few recentstudies (the two most recent ones published in 2002 and 2007)
and that we found very few studies conducted in low-and middle
income countries even though our search strategy was set up to
identify such studies Anotherlimitation wasthat forthe reviewed
studies we could not assess all outcomes of rider training that are
potentially useful Some of the included studies assessed outcomes
such as use of safety clothing including helmets Such outcomes
are valuable since the use of motorcycle helmets has been demon-
strated to be effective in preventing head injuries (Liu 2004) De-
spite this the use of safety clothing was not included as an out-
come in this review because the studies measuring this outcome
were limited by poor data quality due to very low response rates
Another outcome of interest in assessing the effectiveness of rider
training is lsquotype of injuryrsquo As Haworth 2005 state in their review
of motorcycle licensing and training rider training could be con-
sidered effective if injuries become less severe because motorcycle
riders have learned how to avoid more serious injuries through
rider training The studies included in this review did not specify
type of injury and so this outcome could not be assessed Last
changes in knowledge and attitudes following rider training are
other potentially useful outcomes We have excluded such studies
(Osga 1979 Clayton 1990 Goldenbeld 2004) as these outcomes
are proxies rather than direct measures of the effectiveness of mo-
torcycle rider training We opted to report on the outcomes of
crashes injury death and offences alone since there is little ev-
idence about how changes in knowledge and attitudes relate to
these outcomes
A U T H O R S rsquo C O N C L U S I O N S
Implications for practice
Due to the poor quality of studies included in this review we were
unable to draw any conclusions about the effectiveness of rider
training on crash rates injurious or fatal or offence rates The
findings do suggest that mandatory pre-licence training may be
an impediment to completing a motorcycle licensing process thuspossibly indirectly reducing crash injury death and offence rates
through a reduction in exposure to riding a motorcycle On the
basis of the existing evidence it is not clear if and what type of
training reduces the risk of crashes injuries deaths or offences in
motorcyclists and the selection of the best rider training practice
can therefore not be recommended The suggestion that manda-
tory pre-licence training indirectly reduces exposure to riding the
motorcycle because trainees are impeded from completing the li-
censing process does not help those motorcyclists who persist in
obtaining their licence and riding the motorcycle Some type of
rider training is likely necessary to teach motorcyclists to ride a
motorcycle safely Good policy is based on sound evidence and
evaluation but such evidence for motorcycle rider training is seri-
ously lacking Policy makers are therefore encouraged to include arigorous evaluation component to any new or existing rider train-
ing programme Such an evaluation should ideally be randomized
and assess whether the training programme is effective and if so
which component(s) of the programme work best
Implications for research
Given theresults of this review which found littlereliable evidence
about the effectiveness of motorcycle rider training it is necessary
to build up an evidence base of the effectiveness of motorcycle
rider training in the reduction of crash injury death and offence
rates To this end well designed randomised controlled trials need
to be conducted that examine the impact of rider training oncrashes and offences A randomised controlled trial is the preferred
method as it would significantly reduce most biases described in
this review To minimise the impact of participants delaying or
never attaining their licence and therefore exposure as found in
the randomised trials in this review such a trial should include
riders who are already licensed and reliable measures of riding
exposure in a reasonably lengthy follow up period It is expected
that a randomised controlled trial of sufficient sample size would
observe equal balance on possible confounders by study arm how-
ever an assessment of potentially important confounders (such as
helmet wearing) by study group would enable control for these
variables in the analyses if necessary It is imperative that the effec-
tiveness of each component of the rider training (such as theorypractice knowledge skills and modification of behaviour) and the
synergy of such components are evaluated Outcomes should be
objective ideally those that are routinely collected rather than self-
reported outcomes Even though riding the motorcycle poses dif-
ferent challenges and incurs different risks lessons may be learned
from research into the effects of car driver training
Motorcycles are a common means of transport in low- and middle
income countries and usage is increasing As the burden of mo-
torcycle crash victims is mostly borne in such countries research
should be focused on the effectiveness of types of rider training on
the reduction of crashes injuries and fatalities in such countries
and be supported by higher income countries This review found
only one study conducted in a middle income country that sug-gested decreasing injury rates It would be beneficial to conduct
additional research to confirm this suggestion and to enhance our
knowledge of the role of rider training in low- and middle-income
settings
March 2006 saw the launch of the University of North Carolina
Highway Safety Research Centre partnership with the Motorcycle
Safety Foundation (MSF) and plans were revealed to conduct a
study of the potential benefits of motorcycle rider training The
study is funded by the MSF and the National Highway Traffic Sa-
fety Administration (NHTSA) Called rsquoThe Discovery Projectrsquo it
will compare data on crash rates from riders participating in a basic
23Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2650
motorcycle safety class with riders participating in a basic motor-
cycle safety class as well as a variety of continuing education classesThe collection of pilot data was due to start in March 2007 and
the data collection portion of the study is expected to take three
years (Highway Safety Research Center) It is vital that researchers
and policy makers collaborate to develop furtherhigh quality eval-
uations of motorcycle rider training programmes to help inform
development of new and effective training programmes
A C K N O W L E D G E M E N T S
Staff from The University of Sydney library for helping us locate
and retrieve full reports of many papers
R E F E R E N C E S
References to studies included in this review
Adams 1985 published data only
Adams A Collingwood V Job RFS Evaluation of the 1982 Australia
Post motorcycle rider training scheme NSW Australia Traffic
Accident Research Unit Traffic Authority NSW 1985 [ RN 85]
Anderson 1980 published data only
Anderson J Ford JL Peck RC Improved motorcyclist licensing and testing project Washington DC National Highway Traffic Safety
Administration (NHTSA) USA 19801ndash217 [
DOTndashHSndash805ndash598]
Billheimer 1998 published data only
Billheimer JW Evaluation of the California Motorcyclist Safety
Program Transportation Research Record 19981640100ndash9 [
TRB Report Number 980652]
Cooper 1988 published data only
Cooper PJ Rothe JP Motorcyclists who they are and why they do
what they do Transportation Research Record 1988116878ndash85
Davis 1997 published data only
David CF Evaluation of community traffic safety programs and
motorcycle operator training programs Connecticut Department of Transportation 199763
Hall 1988 published data only
Hall CJ Buchanan L Engle R Motorcycle rider education evaluationproject Washington DC USA National Highway Traffic Safety
Administration (NHTSA) US Department of Transportation
1988 [ DTNH 22ndash80ndashCndash05123]
Haworth 1997 published data only
Haworth N Smith R Brumen I Pronk N In Monash University
Accident Research Centre editor(s) Case-control study of motorcycle
crashes Victoria Australia Department of Transport and Regional
Development The Federal Office of Road Safety 19971ndash149 [
ISBN 0 642 25501 6 ]
Hurt 1981 published data only
Hurt H Ouellet J Thom D Motorcycle accident cause factors and
identification of countermeasures USA Department of
Transportation 1981 [ Technical Report DOT HS 806ndash862]
Jonah 1982 published data onlylowast Jonah BA Dawson N Bragg B Are formally trained motorcyclists
safer Accident Analysis amp Prevention 198214(4)247ndash55
Jonah BA Dawson NE Bragg BW Evaluation of the Canada
Safety Councilrsquos Motorcycle Training Program Proceedings of theInternational Motorcycle Safety Conference Washington DC
USA 198059ndash93
Kloeden 1994 published data only
Kloeden CN Moore VM McLean AJ Evaluation of the pre-license
training program for motorcyclists in South Australia Walkerville
South Australia Australia NHMRC Road Accident Research Unit
The University of Adelaide 19941ndash37 [ ISBN 0 7308 0607 3]
Kraus 1975 published data only
Kraus JF Riggins RS Franti CE Some epidemiological features of
motorcycle collision injuries American Journal of Epidemiology 1975102(1)74ndash98
Leung 1987 published data only
Leung KS Reding VA Evaluation of the Wisconsin motorcycle rider course Wisconsin Bureau of Policy Planning and Analysis
Wisconsin Department of Transportation 1987
MAIDS 2004 published and unpublished data
MAIDS In-depth investigations of accidents involving powered
two wheelers ACEM (Acemacemeu) 2004 Vol Final report
20179
McDavid 1989 published data only
McDavid JC Lohrmann BA Lohrmann G Does motorcycle
training reduce accidents Evidence from a longitudinal quasi-
experimental study Journal of Safety Research 198920(2)61ndash72
Mortimer 1984 published data only
Mortimer R Evaluation of the motorcycle rider course Accident
Analysis amp Prevention 198416(1)63ndash71
24Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2750
Mortimer 1988 published data only
Mortimer RG A further evaluation of the motorcycle rider courseJournal of Safety Research 198819(4)187ndash96
Osga 1980 published data only
Osga GA An investigation of the riding experiences of MSF rider
course participants Dissertation South Dakota USA The
University of South Dakota 19801ndash224
Perrino 2002 published data only
Perrino CS Ahmed A Callender A Rozier E Cantwell A Stewart
O The role of Marylandrsquos motorcycle rider course in promoting saver
behaviors and attitudes Baltimore USA National Transportation
Center Morgan State University 20021ndash52
Raymond 1979 published data only
Raymond S Tatum S An evaluation of the effectiveness of the RAC
ACU motorcycle training scheme - final report Road Safety ResearchUnit Dept of Civil Engineering University of Salford London
19791ndash131
Satten 1980 published data only
Satten RS Analysis and evaluation of the motorcycle rider courses
in thirteen Illinois counties International Motorcycle Safety
Conference Washington DC 1980145ndash93
Savolainen 2007 published and unpublished data
Savolainen Peter Mannering Fred Additional evidence on the
effectiveness of motorcycle training and motorcyclistsrsquo risk-taking
behaviour Transportation Research Board 2007 Vol March 20
19
Swaddiwudhipong 1998 published data only
Swaddiwudhipong W Boonmak C Nguntra P Mahasakpan P
Effect of motorcycle rider education on changes in risk behavioursand motorcycle-related injuries in rural Thailand Tropical Medicine
and International Health 19983(10)767ndash70
Waller 1992 published data only
Waller FL An analysis of the impact of the Texas Department of
Public Safety Motorcycle Operator Training Program upon
motocycle related traffic accidents Texas State University Public
Administration Program Applied Research Projects 199277
References to studies excluded from this review
Anderson 1980a published data only
Anderson LC Better biking a motorcycle rider improvement
program International Motorcycle Safety Conference Washington
DC USA Motorcycle Safety Foundation 1980231ndash53Aupetit 2008 published and unpublished data
Aupetit S Riff J Buttelli O Espie S Ergonomics research program
on motorcycle learning Contributions to the development of a
motorcycle simulator for training Transport Research Arena
Ljubljana 20089
Awane 1999 published data only
Awane T Integrating simulators in motorcycle safety education
Journal of International Association of Traffic and Safety Sciences 199923(1)26ndash35
Baldi 2005 published data only
Baldi S Baer JD Cook AL Identifying best practices states in
motorcycle rider education and licensing Journal of Safety Research200536(1)19ndash32
Batchler 1988 published data only
Batchler MW Evaluation of the pre-license motorcycle training course at the New South Wales traffic education centre Federal Office of
Road Safety (FORS) 19881ndash122 [ Report No CR 77 ]
Blanchard 2006 published data only
Blanchard HT Tabloski PA Motorcycle safety educating riders at
the teachable moment Mosby (CV) Company 200632330ndash2
Bonnett 2005 published data only
Bonnett J Mature age skills training for experienced motorcycle
riders Motorcycle ridersrsquo association of the Australian Capital
Territory 2005
Braver 2007 published data only
Braver ER et alPersuasion and licensure a randomized controlled
intervention trial to increase licensure rates among Maryland
motorcycle owners Traffic Injury Prevention 20078(1)39ndash46Clayton 1980 published data only
Clayton AB Sudlow DE An evaluation of the effectiveness of a one day training programme for learner motorcyclists Berkshire UK
Transport and Road Research Laboratory 19801ndash23 [ ISSN
0266ndash7045]lowast Clayton AB Sudlow DE An evaluation of the effectiveness of a one-day training programme for learner motorcyclists Berkshire UK
Road Safety Division Berkshire 19801ndash23 [ISSN 0266ndash7045]
Council 1978 published data only
Council FM Desper LP Dutt AK Driver education for motorcycle operation University of North Carolina Highway Safety Research
Center 19781ndash40
Crowther 2005 published data onlylowast Crowther G Engaging with motorcyclists UK police and theBikeSafe Road Safety Programme Published proceedings of the
Australasian Road Safety Research Policing Education Conference
Wellington New Zealand 2005 issue ISBN 0ndash473ndash10636ndash11ndash11
Daniello 2009 published data only
Daniello A Gabler HC Mehta YA Effectiveness of motorcycle
training and licensing Transportation Research Board 88th Annual
Meeting 200913
Dewanti 2007 published data only
Dewanti M Challenging efforts in promoting young motorcyclist
safety in Indonesia Road Safety on Four Continents 14th
International Conference Swedish National Road and Transport
Research Institute 2007
Du Plooy 1990 published data onlyDu Plooy DW Motorcycle training in South Africa The Human
Element International Motorcycle Safety Conference Orlando
Florida USA Motorcycle Safety Foundation 19904-1 to 4-9
Engel 1989 published data only
Engel U Krogsgaringrd Thomsen L [Faeligrdselsundervisning og
propaganda ndash en faeligrdselssikkerhedsfremmende foranstaltning]
Dansk Vejtidsskrift 198912295ndash7
Goldenbeld 2002 published data only
Goldenbeld C Houwing S de Craen S The development of
driving skills of young moped riders [De ontwikkeling van de
rijvaardigheid van jonge bromfietsers] SWOV Institute for Road
Safety Research (Stichting Wetenschappelijk Onderzoek
Verkeersveiligheid) Leidschendam the Netherlands 200263
25Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2850
Goldenbeld 2004 published data only
Goldenbeld C Twisk D de Craen S Short and long term effects of moped training a field experiment Transportation Research Part F 20047(1)1ndash16
Haworth 2000 published data only
Haworth N Smith R Kowadlo N Motorcycle training in Victoria
evaluation of rider training curriculum in Victoria Monash
University Accident Research Centre Reports 2000 Vol 165 issue
3131
Haworth 2005 published data only
Haworth N Mulvihill C Review of motorcycle licensing and
training Monash University Accident Research Centre 2005 [
Report no240]
Haworth 2006 published data only
Haworth N Mulvihill C Rowdon P Teaching old dogs new tricksTraining and older motorcyclists Journal of the Australasian College of Road Safety 200718(4)20ndash5lowast Haworth NL Mulvihill C Rowden PJ Teaching old dogs new
tricks Training and older motorcyclists Australasian Road Safety
Research Policing and Education Conference Gold Coast
Queensland 2006
Hill 1983 published data only
Hill D Motorcycle rider training testing and licensing Motor
Cycle Council of New South Wales Sydney 1983
Imatake 1980 published data only
Imatake Y On the activities in Japan of motorcycle safe riding
promotion committee (MSRPC) - especially the education of
unskilled riders of class 1 motorbikes International Motorcycle
Safety Conference Washington DC USA Motorcycle Safety
Foundation 1980255ndash68
Ingebritsen 1990 published and unpublished data
Ingebrigtsen S [Risikofaktorer ved ferdsel med moped og
motorsykkel] TOslashI-rapport 66 Oslo Transportoslashkonomisk
institutt 1990
Jain 2009 published and unpublished data
Jain A et alTwo wheeler accidents on Indian roads a study from
Mangalore India Elsevier 200916130ndash3
Jonah 1979 published data only
Jonah BA Dawson NE Validation of the motorcycle operator skill
test Accident Analysis amp Prevention 197911(3)163ndash71
Jonah 1981 published data onlyJonah BA Dawson NE Bragg WE Predicting accident
involvement with the Motorcycle Operator Skill Test Accident Analysis amp Prevention 198113(4)307ndash418
Kadar 1990 published data only
Kadar KJ Countermeasures motorcycle rider education delivery
system The Human Element International Motorcycle Safety
Conference Motorcycle Safety Foundation 19906-27 to 6-41
Kelsey 1986 published data only
Kelsey SL Liddicoat C Ratz M Licensing novice motorcyclists a
comparison of Californiarsquos standard test and the MOST II (Motorcycle
Operator Skill Test) administered at centralized testing offices Sacramento California USA California Department of Motor
Vehicles Research and Development Office 19861ndash74
Koch 1980 published data only
Koch H Advanced course for motorcycle riders InternationalMotorcycle Safety Conference Washington DC USA Motorcycle
Safety Foundation 1980269ndash81
Koch 1990 published data only
Koch H Brendicke R How to influence risk taking behaviour
within motorcycle rider training programs The Human Element
International Motorcycle Safety Conference Orlando Florida
USA Motorcycle Safety Foundation 1990Motorcycle Safety
Foundation 1990 p 9-1 to 9-25A
Labbett 2005 published data only
Labbett S Langham M Training can make the problem worse
Road Safety Congress 20051ndash12
Law 2005 published and unpublished data
Law TH Radin Umar RS Zulkaurnain S Kulanthayan S Impactof the effect of economic crisis and the targeted motorcycle safety
programme on motorcycle-related accidents injuries and fatalities
in Malaysia International Journal of Injury Control and Safety
Promotion 200512(1)9ndash21
Lowes 1990 published data only
Lowes BF In-traffic rider training The Human Element
International Motorcycle Safety Conference Orlando Florida
USA Motorcycle Safety Foundation 19907-46 to 7-57
Mahon 1976 published data only
Mahon J Teaching accident avoidance on the motorcycle Journal of Traffic Safety Education 1976231ndash29
Maxwell 1985 published data only
Maxwell GS Evaluation of a Program for Training MotorcycleRiders Evaluation and Program Planning 19858(4)339ndash47
Mayhew 1996 published data only
Mayhew DR Simpson SH Effectiveness and role of driver educationand training in a graduated licensing system Ottowa Canada The
Traffic Injury Research Foundation 19961ndash89 [ ISBN
0ndash920071ndash12ndash0]
McKnight 1987 published data only
McKnight AJ Milburn BG An evaluation of the modified
motorcycle operator skill test exercises NHTSA Report 1987 [
Report No DOT HS 807 212 ]
McPherson 1976 published data only
McPherson K McKnight AJ The development and evaluation of a
motorcycle skill test manual and knowledge test Washington DCUSA National Highway Safety Administration 1976
Mohan 2004 published data only
Mohan D Evidence-based interventions for road traffic injuries in
South Asia Journal of the College of Physicians and SurgeonsPakistan 200414(12)746ndash7
Morgan 1980 published data only
Morgan JK Advanced motorcycle training course content and
study evaluation International Motorcycle Safety Conference
Washington DC USA Motorcycle Safety Foundation 1980
283ndash312
Motoki 1990 published data only
Motoki M Yamazaki S A study on effective motorcycle rider
education The Human Element International Motorcycle Safety
26Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2950
Conference Orlando Florida USA Motorcycle Safety
Foundation 19909-26 to 9-57A
Nairn 1992 published data onlylowast Nairn RJ Motorcycle safety research literature review 1987 to
1991 Federal Office of Road Safety Australia 1993 Vol Report
CR 117
Newland 1980 published data only
Newland R Darby J Automotive practicespre motorcycle rider
program International Motorcycle Safety Conference Washington
DC USA Motorcycle Safety Foundation 1980127ndash9
Newland 1982 published data only
Newland R Motorcycle rider training and licensing requirements
Driver training steering a course for the future Melbourne
Victoria Transport Regulation Board 1982
Newland 1990 published data only
Newland R Motorcycle training and licensing in Victoria 1980-
1990 The Human Element International Motorcycle Safety
Conference Orlando Florida USA Motorcycle Safety
Foundation 1990
Ormston 2003 published data only
Ormston R Dudleston A Pearson S Stradling Evaluation of
Bikesafe Scotland Scottish Executive Social Research Edinburgh
2003
Osga 1979 published data only
Osga GS Ellingstad VS An evaluation of the effectiveness of the 1978
South Dakota motorcycle rider course program South Dakota USA
University of South Dakota 19791ndash89
Panou 2007 published data only
Panou M Bekiaris E A new concept on the integration of driving
simulators in driving training the train-all approach Road Safety
on Four Continents 14th International Conference Swedish
National Road and Transport Research Institute 2007
Peterson 1990 published data only
Peterson A Kadar K Status report 1980-1989 Minnesota rider
education statistics The Human Element International
Motorcycle Safety Conference Orlando Florida USA 199013-
113 to 13-129
Prem 1984 published data only
Prem H Good MC Motorcycle rider skills assessment ACT
Australia Office of Road Safety Department of Transport 1984
1ndash377 [ Report No CR 34 ]
Radin Umar 2006 published data only
Radin Umar RS Motorcycle safety programmes in Malaysia how
effective are they International Journal of Injury Control and Safety Promotion 200613(2)71ndash9
Rockwell 1990 published data only
Rockwell TH Kiger S Carnot M An evaluation of the Ohio
motorcyclist enrichment program Ohio Department of Highway
Safety 1990
Rothe 1987 published data only
Rothe JP A motorcycle riding course Motorcyclists Image and reality Canada Insurance Corporation of British Columbia 1987
59ndash83
Rowden 2007 published data only
Rowden P Watson B Haworth N What can riders tell us aboutmotorcycle training A view from the other side of the fence
Australasian Road Safety Research Policing and Education
Conference Melbourne Australia Melbourne 2007
RSC Victoria 1993 published data only
Road Safety Committee Inquiry into Motorcycle Safety in Victoria
Parliament of Victoria 1993
Russam 1979 published data only
Russam K Motorcycle training and licensing in Japan UK Transport
and Road Research Laboratory (TRRL) 1979 [ Report 916 ]
Saffron 1981 published data only
Saffron DG Driver and rider education training and licensing a
brief review Traffic Accident Research Unit Traffic Authority of
New South Wales NSW Australia Traffic Accident Research UnitTraffic Authority of New South Wales 1981
Schulz 1990 published data only
Schulz U Motorcycle accidents among learner riders Motorcycle
licensing Evaluation of the effectiveness of a Graduated Licensing Scheme Forshungshefte des Institut fur Zweiradsicherheit 1990
Sexton 2004 published data only (unpublished sought but not used)
Sexton B Baughan C Elliott M Maycock G The accident risk of
motorcyclists 2004TRL Report TRL 607(ISSN 09684107)58
Shephard 1986 published data only
Shepherd R Do Motorcycle Safety Foundation programs work
Driver 198619(9)16ndash8
Simpson 1990 published data onlySimpson HM Mayhew DR The promotion of motorcycle safety
training education and awareness Health Education Research
Theory and Practice 19905(2)257ndash64
Sokolov 1990 published data only
Sokolov O Safe motorcycle riding some aspects of motorcycle
riding in the USSR The Human Element International
Motorcycle Safety Conference Motorcycle Safety Foundation
19904ndash37 - 4-48
Steffens 1988 published data only
Steffens U Gawatz R Willmes G Effectiveness of mofa courses
[Wirksamkeit von Mofakursen] Unfall- und sicherheitsforschung
strasenverkehr 198867168
Sudlow 2003 published data only
Sudlow D Scoping study on motorcycle training Road Safety
Research Report London UK The British Institute of Traffic
Education Research Department for Transport 2003 Vol June
[ Report No 36]
Tasmania 1995 published data only
Mannering FL Grodsky LL Motorcycle rider training scheme
Traffic Education Research Centre Armidale Australia Driver
Training Australia Limited 1995 Vol April1ndash56
Tetard 1985 published data only
Tetard C Motorcycle ridersrsquo training to detect indices in an
intersection situation [Lrsquoentrainement a la prise drsquoinformation dans
lrsquoapprentissage de la conduite motocycliste] Le Travail Humain198548(2)147ndash59
27Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3050
Thackray 1980 published data only
Thackray RM Prescott JC Field test of the motorcycle ridercourse International Motorcycle Safety Conference Washington
DC USA Motorcycle Safety Foundation 1980195ndash7
Thompson 1994 published data only
Thompson M Evaluation of compulsory basic training for
motorcyclists Transport Research Laboratory Crowthorne
Berkshire UK 1994 [ Report 63]
Tobler 1990 published data only
Tobler U Individual driving instruction for motorcycle driving
learners the Swiss experiment of motorcycle driving school The
Human Element International Motorcycle Safety Conference
Orlando Florida USA Motorcycle Safety Foundation 1990
Transport 1989 published data only
Transport Communications Road Crash Statistics Australia Bureauof Transport and Communications Economics 198914ndash20
Tuan 2007 published data only
Tuan VA Shimizu T Towards development and evaluation of the
motorcycle drivers re-education program in Vietnam modeling of
motorcycle driverrsquos undesired behaviors 11th World Conference
on Transport Research 200725
Victor 1980 published data only
Victor DJ Traffic safety education for motorcyclists International
Motorcycle Safety Conference Washington DC USA Motorcycle
Safety Foundation 1980221ndash8
Waterer 1990 published data only
Waterer TH Compulsory training for motorcyclists from liberty
to compulsion The Human Element International Motorcycle
Safety Conference Orlando Florida USA Motorcycle Safety
Foundation 19904-49 to 4-57
Weller 1989 published data only
Weller RB Chandler EW Motorcycle safety and motorcycle
education past research and survey results Journal of Public Policy
amp Marketing 1989893ndash108
Wells 1981 published data only
Wells P Observations of some RACACU motorcycle training
schemes Transport and Road Research Laboratory Crowthorne
Berkshire UK 1981
Winn 1986 published data only
Winn G Two important trends in motorcycle safety regulations
rider education and conspicuity improvement American
Motorcyclist Association 1986147ndash52
Wisher 1988 published data only
Wisher G Reid J Motorcycle accidents in Norfolk a three-year study of motorcycle training and the Part 1 test Norfolk UK Road Safety
Division Highways Department Norfolk County Council 1988
Wood 1987 published data only
Wood HT Bowen R Evaluation of the revised learner permit scheme
July 1983 to December 1985 Victoria Australia Road Traffic
Authority 19871ndash27 [ ISBN 0 7306 0247 8]
Yamazaki 1990 published data only
Yamazaki S Nakanishi M Survey on effect of Honda rider
education The Human Element International Motorcycle Safety
Conference Orlando Florida USA 199012-167 to 12-187
References to ongoing studies
Burgess 2010 unpublished data only
Burgess C Broughton P Fylan F Stradling Interim evaluation of
the UKs National Ride program In L Dorn editor(s) Driver
behaviour and Training Vol IV Aldershot Ashgate Publishing
Limited 2010
Additional references
Anderson 1980
Anderson J Ford JL Peck RC Improved motorcyclist licensing and
testing project Volume I and II National Highway Traffic Safety
Administration US department of Transportation Washington
DC 1980
Bjornstig 1985
Bjornstig U Bylund P Lekander T Brorsson B Motorcycle fatalitiesin Sweden Acta Chirurgica Scandinavica 1985151(7)577ndash81
Clayton 1990
Clayton A Sudlow DE An evaluation of the effectiveness of a one-
day training program for learner motorcyclists The Human
Element International Motorcycle Safety Conference Orlando
Florida USA 199013-105 to 13-111
Cochrane Collaboration 2008
Higgins JP Green S Cochrane Handbook for Systematic Reviews of
Interventions 500 The Cochrane Collaboration 2008
Elliot 2003
Elliott MA Baughan CJT Broughton J Grayson G Knowles J et
alMotorcycle Safety a Scoping Study TRL [ Report No
0968ndash4107]Fleiss 2004
Fleiss JL Levin B Paik MC Statistical methods for rates and proportions 3rd Edition John Wiley amp Sons Inc 2004 [ ISBN
978ndash0ndash471ndash52629ndash2 ]
FORS 1999
FORS Federal Office of Road Safety Road risk for sober licensed
motorcyclists Vol Monograph 27 Federal Office of Road Safety
19991ndash2
Haworth 2005
Haworth N Mulvihill C Review of motorcycle licensing and training
Victoria Australia Monash University Accident Research Centre
20051ndash83 [ Report No 240]
Huwiler-Muumlntener 2002Huwiler-Muumlntener K Juumlni P Junker C Egger M Quality of
Reporting of Randomized Trials as a Measure of Methodologic
Quality JAMA 20022872801ndash4
Johnston 2008
Johnston P Brooks C Savage H Fatal and serious road crashes
involving motorcyclists Road Safety Research and Analysis Report
ATSB Monograph 20 Canberra Department of Infrastructure
Transport Regional Development and Local Government 0205
081ndash26
Juumlni 2001
Juumlni P Altman DG Egger M Systematic reviews in health care
Assessing the quality of controlled clinical trials BMJ 2001323
42ndash6
28Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3150
Kopits 2003
Kopits E Cropper M Traffic fatalities and economic growth TheWorld Bank Development Research Group Infrastructure and
Environment Policy research working paper 3035 2003
Leung 1983
Leung P Chau BTW Motorcycle accidents in Hong Kong pattern
of injury treatment causative factors and socio-economic
implications Injury 198315(1)23ndash9
Liu 2004
Liu B Ivers R Norton R Blows S Lo SK Helmets for preventing
injury in motorcycle riders Cochrane Database of Systematic Reviews 2004 Issue 2 [DOI 10100214651858CD004333pub3]
Mohan 2002
Mohan D Road safety in less-motorized environments future
concerns International Journal of Epidemiology 200231(3)527ndash32Mortimer 1982
Mortimer RG An evaluation of the motorcycle rider course
Proceedings of the 20th conference of the American Association for
Automotive Medicine Arlington Heights IL USA American
Association for Automotive Medicine 198277ndash96
Mullin 1997
Mullin BT Risk factors for motorcycle injury the role of age
gender experience training and alcohol PhD thesis Auckland
New Zealand University of Auckland 1997
Noordzij 2001
Noordzij PC Forke E Brendicke R Chinn BP Integration of needs
of moped and motorcycle riders into safety measures review and
statistical analysis in the framework of the European research
project PROMISING SWOV Institute for Road Safety Research
Leidschendam The Netherlands 2001 [ Report No Dndash2001ndash5]
Osga 1979
Osga G Ellingstad VS An evaluation of the effectiveness of the 1978 South Dakota Motorcycle Riders Course Program South Dakota
University of South Dakota Vermillion Human Factors Lab 1979
Oxford English Dictionary Online
httpoxforddictionariescom Oxford English Dictionary Online
Oxford University Press 2010
Peden 2004
Peden M Scurfield R Sleet D World Report on Road TrafficInjury Prevention Geneva World Health Organization Available
at httpwwwwhointviolence˙injury˙preventionpublications
road˙trafficworld˙reporten 2004Chapter 3
Radin Umar 1995
Radin Umar R Macakay G Hills B Preliminary analysis of
motorcycle accidents Short-term impacts of the running
headlights campaign and regulations in Malaysia J Traffic Med 199523(1)17ndash27
Tsai 1995
Tsai YJ Wang JD et alCase-control study of the effectiveness of
different types of helmets for the prevention of head injuries among
motorcycle riders in Taipei Taiwan American Journal of
Epidemiology 1995142(9)974ndash81
Vis 1995
Vis AA In-depth study of the hazards of motorcycling [De
onveiligheid van motorrijden nader bekeken] SWOV Institute for
Road Safety Research Leidschendam the Netherlands 1995 [
Report No Rndash95ndash69]
Weaver 1990
Weaver EA Motorcycle safety education in the United States
Motorcycle Safety Foundation Orlando Florida USA 1990
Winn 1987
Winn G Two important trends in motorcycle safety regulations
rider education and conspicuity improvement No more
information available Detroit MI US Society of Automotive
Engineers 1987147ndash52
Yuan 2000
Yuan W The effectiveness of the rsquoride-brightrsquo legislation for
motorcycles in Singapore Accident Analysis amp Prevention 200032
(4)559ndash63
Zambon 2006
Zambon F Hasselberg M Factors affecting the severity of injuries
among young motorcyclists a Swedish nationwide cohort study
Traffic Injury Prevention 20067(2)143ndash9lowast Indicates the major publication for the study
29Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3250
C H A R A C T E R I S T I C S O F S T U D I E S
Characteristics of included studies [ordered by study ID]
Adams 1985
Methods Non-randomised trial
Participants Initial sample size
Total of 162 Australia Post telegram delivery riders
Not clear how many in intervention and control groups
Analysis sample size
Intervention group 63Control group 35
Total 98 (61)
Interventions No formal training vs voluntary rider training course of 12 hours over 2 consecutive days (theory 4 hours practice
8 hours)
Intervention integrity
No information given
Outcomes Australia Post recorded crashes
Length of follow up 1 year
Notes Study conducted in NSW Australia
Anderson 1980
Methods Randomised trial
Participants Initial sample size
Total 41188 motorcycle licence applicants
Analysis sample size
Intervention group with skills training 13160
Intervention group without skills training 12634
Control group 15080
Total 40874 (99)
Interventions Standard licensing skill test vs the Motorcycle Operators Skill Test (MOST) with or without a training component
MOST is not to be confused with rider training because it is a licensing test The training component was for
participants who failed the skill test on the first attempt was of 3 hours duration and consisted of instructions about
the skills needed to pass the skill test The training consisted of 30 min theory and 2 frac12 hours practice on the
motorcycle
Intervention integrity
33 in MOST with skills training 31 in MOST without skills training and 43 in control group obtained their
motorcycle licence in the end
Outcomes Police recorded crashes and offences
Length of follow up 3 years
30Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3350
Anderson 1980 (Continued)
Notes Study conducted in California USA
Billheimer 1998
Methods Cohort study
Participants Initial sample size
Intervention group Unclear numbers of CMSP graduates and ERC graduates
Control group Over 16000 untrained motorcyclists
Analysis sample size
1139 matched CMSP-pairs and 1182 ERC-pairs
Interventions No formal training vs the mandatory (for those younger than 18 years) pre-licence 16-hours basic California
Motorcyclist Safety Program (CMSP) and the 8-hour Experienced Rider Course (ERC)
Intervention integrity
No information given
Outcomes Police recorded crashes and offences
Length of follow up 2 years
Notes Study conducted in California USA
Cooper 1988
Methods Cohort study
Participants Initial sample size
Intervention group 418 graduates and 57 who failed
Control group 402
Analysis sample size
Intervention group unclear
Control group unclear
Total 863 (984)
Interventions No formaltrainingvs non-mandatory pre-licence rider trainingThe content of thetrainingwas notfurther explainedIntervention integrity
Participants had all completed the rider training
Outcomes Police recorded crashes and offences
Length of follow up 7 years
Notes Study conducted in British Columbia Canada
31Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3450
Davis 1997
Methods Retrospective cohort study
Participants Initial sample size
Intervention group 9320 graduates
Control group unclear
Analysis sample size
Intervention group 9320 graduates
Control group unclear
Interventions No formal training vs the non-mandatory Connecticut Motorcycle Operator Training Program The content of
training was not further explained
Intervention integrity Participants had all completed the rider training
Outcomes Police recorded crashes severity of crashes
Length of follow up 1 year
Notes Study conducted in Connecticut USA
Hall 1988
Methods Randomised trial
Participants Initial sample sizeIntervention group A 6248
Intervention group B 6616
Intervention group C 6996
Control group 6604
Total 26464 motorcycle licence applicants
Analysis sample size
Intervention group A 6012 (96)
Intervention group B 6428 (97)
Intervention group C 6778 (97)
Control group 6502 (98)
Interventions Standard licensing skill test vs the Motorcycle Operator Skill Test (MOST II) without training (A) MOST II with
3 hours training (B) or MOST II with 20 hours training (C) The 3 hours training comprised instructions aboutthe skills needed to pass the skill test (30 min theory and 2 frac12 hours practice on the motorcycle) The 20 hours
training consisted of the Motorcycle Safety Foundationrsquos Motorcycle Rider Course (MRC) comprising of 8 hours
theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-road practice
Intervention integrity
25 and 26 obtained their motorcycle licence in the lsquoMOST 3-hour trainingrsquo and lsquoMOST 20-hour trainingrsquo
intervention arm respectively In the lsquoMOST testrsquo intervention arm 29 obtained their licence and in the control
group 35
Outcomes Police recorded crashes and offences
Length of follow up 3 years
32Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3550
Hall 1988 (Continued)
Notes Study conducted in New York USA
Haworth 1997
Methods Population based case-control study
Participants Cases Motorcycle riders or pillion riders involved in motorcycle crash (on road) Recruited from hospital and police
records Not clear whether all cases (especially fatalities) were included
Community controls Multiple matched sample of motorcycle rider or pillion who passed the crash site at the same
time of day and day of week that the crash occurred
Initial sample size
Cases 336
Controls 1195
Analysis sample size
Cases 234 (70)
Controls 561 (47)
Interventions Basic handling skills course beginner intermediate or advanced course measured through self-report not validated
Outcomes Injuries with an Injury Severity Score of 5 or greater or admission for at least four hours and fatalities from a
motorcycle crash (from hospital and police records)
Length of follow up two years or more prior to the interview
Notes Study conducted in Victoria Australia
Hurt 1981
Methods Population based case-control study
Participants Cases Motorcycle riders who had a motorcycle crash that was reported to police (traffic accident reports) Not all
cases included
Community controls Multiple matched sample of motorcycle riders who passed the crash site at the same time of
day and day of week as the crash occurred but 2 years later
Initial sample size
Cases 4500 notifications of crashes to police (only 50 received in time research team able to respond to 1126 for
unreported reasons data collection completed for 900)
Controls Number approached unreported 2310 consented and interviewed
Analysis sample size
Cases 808 (808900 data on rider training) - 18 of initial sample
Controls 670 (6702310 data on rider training) - lt 29 of initial sample
Interventions Training experience self taught friends-family school-club formal-professional others measured through self-
report validated
Outcomes Injuries and fatalities from a motorcycle crash (from hospital records and coronerrsquos office)
33Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3650
Hurt 1981 (Continued)
Notes Study conducted in Los Angeles USA
Jonah 1982
Methods Cohort study
Participants Initial sample size
Intervention group
2310 licensed MTP graduates who are active riders
Control group
2029 licensed motorcyclists who are active riders
Non-licensed and non-active riders excluded
Analysis sample size
Intervention group
811 interviews (35) 533 driver records (23)
Control group 1080 interviews (71) 905 driver records (45)
Interventions No training vs the non-mandatory Canada Safety Council Motorcycle Training Program (MTP) comprising of 4
hours theory and 16 hours practise (6 hrs on-road) over 2 weekends
Intervention integrity
Participants had all successfully completed the MTP
Outcomes Self-reported and police reported crashes and offencesLength of follow up 4 years
Notes Study conducted in Ontario Canada
Kloeden 1994
Methods Cohort study
Participants Initial sample size
Intervention group 2533 licence applicants
Control group Unclear
Sample of riders obtaining probationary licenceIntervention group 1625 (64)
Control group 5015 (unclear)
Non-active riders excluded from control group (~66)
Analysis sample size
Intervention group 631 (39)
Control group 318 (6)
Interventions No formal training vs the mandatory pre-licence Ridersafe course consisting of two levelsthe first is a training course
in the basics of motorcycle riding where a pass entitles the rider to obtain a learnerrsquos permit The second is a skill
test where a pass entitles the rider to obtain a probationary motorcycle licence The training course consists of two 4
hour training sessions on separate days involving instruction videos and actual riding on motorcycles The skill test
consists of a 4 hour preparation course of 74 minutes of lecturediscussion 38 minutes of video and 75 minutes of
34Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3750
Kloeden 1994 (Continued)
on-course riding which is the testing phase
Intervention integrity
Unclear what percentage in the Ridersafe group and the control group obtained their probationary licence halfway
and their full motorcycle licence in the end
Outcomes Police reported crashes
Length of follow up 1 year
Notes Study conducted in SA Australia
Kraus 1975
Methods Population based case-control study
Participants Cases All motorcycle riders with a confirmed medically treated motorcycle injury (including fatalities) from all
hospitals in study base and peripheral area hospitals and all deaths due to motorcycle crashes in the study base All
serious injuries (leading to hospitalisation) and fatalities likely to be included
Community controls A random sample of registered motorcycle owners in the same temporary and geographical
study base as the cases arose from
Initial sample size Cases 1273
Controls 738
Analysis sample size Cases 741 (58)
Controls 434 (59)
Interventions Motorcycle rider training (not specified) measured through self-report
Outcomes Injuries leading to hospitalisation fatalities
Length of follow up not specified
Notes Study conducted in Sacramento county USA
Leung 1987
Methods Non-randomised trial
Participants Initial sample sizeIntervention group unknown number of MRC graduates
Control group unknown number of licensed motorcyclists
Analysis sample size
Intervention group 2941
Control group 43094
Interventions No formal training vs non-mandatory Wisconsin motorcycle rider course a 22 hours off-street programme that
teaches the basic rules of safe motorcycling
Intervention integrity
50 in the MRC group and 100 in the control group obtained their motorcycle licence in the end
35Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3850
Leung 1987 (Continued)
Outcomes Police recorded crashes and offences
Length of follow up up to 3 years
Notes Study conducted in Wisconsin USA
MAIDS 2004
Methods Population based case-control study
Participants Cases Motorcycle riders who had a motorcycle crash Unclear whether all cases included
Community controls Motorcycle riders identified at petrol stations Unclear whether matched to cases
Initial sample size
Cases not reported
Controls not reported
Analysis sample size
Cases 921 (Italy 200 Netherlands 200 Spain 121 Germany 250 France 150)
Controls 923 (Italy 200 Netherlands 200 Spain123 Germany 250 France 150)
Interventions Pre-license rider training or additional training (non-compulsory)
Outcomes Crashes
Notes Study in five sampling areas located in five European countries (France Germany Netherlands Spain and Italy)
McDavid 1989
Methods Cohort study
Participants Sample size after first match
Intervention group 346 male training graduates
Control group 346 male motorcycle riders licensed in same year
Sample size after second match
Intervention group 139 (40)
Control group 139 (40)
Analysis sample size
Intervention group 139 (40)
Control group 139 (40)
Interventions No formal training vs the non-mandatory pre-licence British Columbia Safety Councilrsquos 37-hour motorcycle safety
programme
Intervention integrity
Participants had all completed and passed the training
Outcomes Police reported crashes injurious crashes
Length of follow up 1-6 years
36Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3950
McDavid 1989 (Continued)
Notes Study conducted in British Columbia Canada
Mortimer 1984
Methods Cohort study
Participants Initial sample size
Intervention group 608 MRC graduates
Control group 360 motorcycle riders
Response rates
Intervention group 360 (59)
Control group 324 (90)
Non-active riders excluded
Analysis sample size
Intervention group 213 (35)
Control group 303 (84)
Interventions No formal training vs the non-mandatory Motorcycle Safety Foundationrsquos Motorcycle Rider Course comprising of
8 hours theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-road practice
Intervention integrity
Participants had all completed and passed the MRC
Outcomes Self-reported crashes and offencesLength of follow up 1 year
Notes Study conducted in Illinois USA
Mortimer 1988
Methods Cohort study
Participants Initial sample size
Intervention group 2307 MRC graduates
Control group 556 active licensed motorcycle riders
Response ratesIntervention group 1298 (56)
Control group 500 (90)
Non-active riders excluded
Analysis sample size
Intervention group 913 (40)
Control group 500 (90)
Interventions No formal training vs the non-mandatory Motorcycle Safety Foundationrsquos Motorcycle Rider Course comprising of
8 hours theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-road practice
Intervention integrity
Participants had all completed and passed the MRC
37Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4050
Mortimer 1988 (Continued)
Outcomes Self reported crashes and offences
Length of follow up 1 year
Notes Study conducted in Illinois USA
Osga 1980
Methods Cohort study
Participants Initial sample sizeIntervention group 1163 MSF MRC participants invited 818 (74) responded
Control group 329 motorcycle riders responded to recruitment strategies
Non-active riders excluded from analysis
Analysis sample size
Intervention group 595 (73)
Control group 329 (100)
Interventions No formal training vs the non-mandatory Motorcycle Safety Foundationrsquos Motorcycle Rider Course (MRC) com-
prising of 8 hours theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-road
practice
Intervention integrity
Participants had all successfully completed the MRC
Outcomes Self-reported crashes police recorded crash reports
Length of follow up 6 months - 3 12 years
Notes Study conducted in South Dakota USA
Perrino 2002
Methods Cohort study
Participants Initial sample size
Intervention group 2144 (unclear) MRC graduates (motorcycle licence applicants)
Control group 2601 (unclear) regular licensing test graduates (motorcycle licence applicants)Analysis sample size
Intervention group 1070 (unclear)
Control group 867 (unclear)
Interventions No formal training vs the mandatory (for those younger than 18 years) pre-licence Maryland Motorcycle Rider
Course (MRC) consisting of 8 hours theory and 12 hours practice
Intervention integrity
No information given
Outcomes Self-reported and police recorded crashes and offences
Length of follow up 1 year
38Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4150
Perrino 2002 (Continued)
Notes Study conducted in Maryland USA
Raymond 1979
Methods Cohort study
Participants Initial sample size
Intervention group
258 volunteers mainly (but not all) novice riders
Control group
1104 provisional licence applicants
Analysis sample size
Intervention group 190 (74)
Control group 627 (57)
Interventions No formal training vs the non-mandatory RACACU (Royal Automobile ClubAuto-Cycle Union) training scheme
The training program consisted of either 24 one-hour sessions or 13 two-hour sessions Twelve hours of the training
focused on theory about basic operational procedures traffic laws and highway code The practical session focused
on basic operational skills riding on a training area and riding in traffic with an instructor A proficiency test was
administered at the end of the course
Intervention integrity
No information given
Outcomes Self-reported crashes (including minor non-injury non-damage and off-road crashes)
Length of follow up 1 year
Notes Study conducted in the UK
Satten 1980
Methods Cohort study
Participants Initial sample size
Intervention group interviewers kept interviewing until they had 100 riders selected from a population of 891
enrollees in the MSF MRC course (unknown how many were asked and how many refused to participate)Control group unknown number of registered motorcycle riders
Non-active riders excluded from analysis
Analysis sample size
Intervention group 71
Control group 100
Interventions No formal training vs the non-mandatory Motorcycle Safety Foundationrsquos (MSF) Motorcycle Rider Course (MRC)
comprising of 8 hours theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-
road practice
Intervention integrity
Participants had all successfully completed the MRC
39Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4250
Satten 1980 (Continued)
Outcomes Self-reported crashes injuries and offences
Length of follow up 3 years
Notes Study conducted in Illinois USA
Savolainen 2007
Methods Cohort study
Participants Initial sample size
Intervention group 4000 riders that had conducted the MSF MRC course (passed or failed)
Control group 4000 motorcycle owners or permit or endorsement holders from the Indiana Bureau of Motor
Vehicles database
Analysis sample size
Intervention group 739 (18)
Control group 588 (15)
Interventions Intervention integrity
No information given
Outcomes Probability of crash involvement
Length of follow up up to 5 years
Notes Study conducted in Indiana USA
Swaddiwudhipong 1998
Methods Randomised trial
Participants Initial sample size
Intervention group
all 16 villages of 3 sub-districts
(participants were practising motorcyclists licensed and unlicensed)
Control group
18 villages of 3 other sub-districtsAnalysis sample size
Intervention group 3 districts (100)
Control group 3 districts (100)
Interventions No rider education vs non-mandatory community based rider education consisting of theory on motorcycle injury
risk helmet use traffic laws vehicle regulations and traffic signs and a written and skill test leading to a licence
Intervention integrity
No information given
Outcomes Injuries fatalities
Length of follow up 2 years
40Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4350
Swaddiwudhipong 1998 (Continued)
Notes Study conducted in Tak Province rural Thailand
Waller 1992
Methods Cohort study
Participants Initial sample size
Intervention group 349 riders who succesfully completed the Department of Public Safety motorcycle operator
training in 1988
Control group 349 motorcyclists who never completed the DPS motorcycle operator training
Analysis sample size
Intervention group 349
Control group 349
Interventions No DPS motorcycle operator training vs non-mandatory DPS motorcycle operator training The DPS training is
the Motorcycle Safety Foundationrsquos Motorcycle Rider Course comprising eight hours of theory (lectures discussions
films and multiple choice tests) a textbook and 12 hours off-road practice
Intervention integrity
All 349 riders succesfully completed the rider training
Outcomes Police recorded crashes
Length of follow up 3 years
Notes Study conducted in
Characteristics of excluded studies [ordered by study ID]
Study Reason for exclusion
Anderson 1980a Study type review and process evaluation of a rider course
Aupetit 2008 Outcome proposal of a rider training curriculum that combines traditional training and a motorcycle riding
simulator
Awane 1999 Study type review on use of simulators in rider and driver training
Baldi 2005 Intervention not rider training but intervention integrity of rider training availabilty of rider education
implementation of graduated licensing system etc on a state-wide level
Batchler 1988 Study type description of a training course
Blanchard 2006 Study type not a study assessing the association between rider training and outcomes
Bonnett 2005 Outcome qualitative data regarding satisfaction with the training programme
41Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4450
(Continued)
Braver 2007 Outcome licensure rates
Clayton 1980 Outcome knowledge skills attitude
Council 1978 Outcome crash reduction in ANY vehicle not motorcycles exclusively
Crowther 2005 Study type qualitative study
Daniello 2009 Study type review of the effectiveness of motorcycle training and licensing
Dewanti 2007 Intervention emotional characteristics behaviour (ie not motocycle rider training)
Du Plooy 1990 Study type review of types of rider training courses in South Africa
Engel 1989 Study type uncontrolled before-after study
Goldenbeld 2002 Outcome skills
Goldenbeld 2004 Outcome knowledge and skills
Haworth 2000 Intervention balance between attitudinal andvehicle control skills acrosscurriculla of motorcycle rider training
courses
Haworth 2005 Study type review
Haworth 2006 Study type cross-sectional survey
Hill 1983 Study type review and description of training system
Imatake 1980 Study type course descriptionreview
Ingebritsen 1990 Intervention risk factors for motorcycle crashes
Jain 2009 Study type cross-sectional descriptive study of motorcycle crashes over a 5 year period
Jonah 1979 Study type cross-sectional survey Outcome scores on motorcycle operator skill test
Jonah 1981 Intervention evaluation of a motorcycle rider skill test not motorcycle rider training
Kadar 1990 Study type review of components of rider training courses
Kelsey 1986 Intervention evaluation of a motorcycle rider skill test not motorcycle rider training
Koch 1980 Study type reviewcourse description
Koch 1990 Study type description of course content
42Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4550
(Continued)
Labbett 2005 Study type essay about possible explanations for motorcycle deaths
Law 2005 Intervention targeted motorcycle safety programme not specifically rider training
Lowes 1990 Study type review and descriptions of rider training
Mahon 1976 Study type review of motorcycle rider training methods
Maxwell 1985 Outcomesrecommendations forprogramme operatorsand otherson howthe motorcycle trainingprogramme
works and how it can be improved
Mayhew 1996 Study type review
McKnight 1987 Study type evaluation of methodological and administrative aspects of a motorcycle operator skill test
McPherson 1976 Outcome knowledge and skills
Mohan 2004 Study type review of evidence of interventions to reduce road traffic injuries
Morgan 1980 Study type reviewcourse description
Motoki 1990 Study type proposal and development of rider training curriculum
Nairn 1992 Study type literature review
Newland 1980 Study type description of development of motorcycle safety course
Newland 1982 Study type descriptionreview of a rider training course
Newland 1990 Study type uncontrolled BeforeAfter study
Ormston 2003 Outcome changes in attitude and self reported behaviour
Osga 1979 Outcome knowledge attitudes and skills
Panou 2007 Intervention new simulation tool to teach driverrider
Peterson 1990 Study factor demographic details of course participants completing the course vs those not completing the
course
Prem 1984 Study type experimental
Study factor examines association between skills and performance in tests
Radin Umar 2006 Intervention motorcycle safety programme not specifically rider training
Rockwell 1990 Study type methodologically flawed study type
43Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4650
(Continued)
Rothe 1987 Study type qualitative study of rider training in general
Rowden 2007 Study type qualitative study
RSC Victoria 1993 Study type review of motorcycle rider training and education in Victoria
Russam 1979 Study type review of training programmes in Japan
Saffron 1981 Study type literature review
Schulz 1990 Study factor predictors of motorcycle crashes
Sexton 2004 Study type cross-sectional analytic analysis
Shephard 1986 Study factor reviewdescription of programme
Simpson 1990 Study type review of motorcycle rider trainingeducation
Sokolov 1990 Study factor review of motorcycle riding courses in USSR
Steffens 1988 Outcome self-reported behaviour experiences and researcher observed riding behaviour
Sudlow 2003 Study type review of motorcycle training in the UK
Tasmania 1995 Study type cross-sectional descriptiveprocess evaluation
Tetard 1985 Study type experimental
Outcome detection of indices
Thackray 1980 Outcome instructional effectiveness user acceptance administrative feasibility
Thompson 1994 Study type cross-sectional survey
Outcome attitudes and knowledge
Tobler 1990 Study type review of course curricula
Transport 1989 Intervention syllabus for rider training
Tuan 2007 Study type cross-sectional study (survey) which does not assess the effects of an intervention but examines
motocyclistsrsquo behaviour
Victor 1980 Study type review
Waterer 1990 Study type review of UK rider training issues
Weller 1989 Study type review of studies evaluating motorcycle safety and education measures
44Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4750
(Continued)
Wells 1981 Study type description of different methods of training
Winn 1986 Study type review of rsquohistoryrsquo of rider training
Wisher 1988 Study type cross-sectional survey
Wood 1987 Study type Interrupted Time Series but did not meet the criteria for an ITS as published by the Cochrane
Effective Practice and Organisation of Care Group
Yamazaki 1990 Study type cross-sectional survey
Outcome riding skills skills retention and attitude
Characteristics of ongoing studies [ordered by study ID]
Burgess 2010
Trial name or title UKs National Ride program
Methods Cohort study
Participants Motorcycle offenders
Interventions Police led diversion program
Outcomes Self reported attitudinal and behavioural measures
Starting date
Contact information
Notes Expected to be completed in early 2010
45Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4850
D A T A A N D A N A L Y S E S
This review has no analyses
A P P E N D I C E S
Appendix 1 Search strategy
We searched the following electronic databases
bull Cochrane Injuries Group Specialised Register (searched 22 Sept 2008)
bull CENTRAL (The Cochrane Library 2008 Issue 3)
bull TRANSPORT 1988 to 200706 (latest date available)
bull ATRI Australian Transport Index (to Sept 2008)
bull MEDLINE (Ovid SP) 1950 to Sept 2008
bull EMBASE (Ovid SP) 1980 to Sept 2008
bull CINAHL (EBSCO) 1982 to Sept 2008
bull WHOLIS (World Health Organization Library Information System) (to Sept 2008)
bull PsycINFO (Ovid SP) 1806 to Sept 2008
bull LILACS (BIREME) (Latin American and Caribbean Health Sciences) (1982 to Sept 2008)
bull ISI Web of Science Social Sciences Citation Index (SSCI) 1970 to Sept 2008
bull ERIC (searched 25 Sept 2008)
bull ZETOC (searched 25 Sept 2008)
bull SIGLE (searched 25 Sept 2008)
All of the databases were searched using the following text-word search strategy The strategy was adapted using either rsquorsquo or rsquo$rsquo as atruncation symbol
1 motor-cycl or motorcycl or motor-bik or motorbik or motor-scooter or motorscooter or scooter or moped or motocycl
or motocicl or moto
2 train or educat or instruct or program or course or qualif or learn or improv
3 1 and 2
H I S T O R Y
Protocol first published Issue 2 2005
Review first published Issue 10 2010
46Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4950
C O N T R I B U T I O N S O F A U T H O R S
Katina Kardamanidis was the first author of the protocol conducted the literature search extracted full studies assessed the studies
according to inclusion criteria assessed quality of the studies compared findings with the second author discussed disagreements with
the third author wrote the drafts and the final review
Rebecca Ivers was the second author of the protocol assessed the studies according to inclusion criteria assessed quality of the studies
compared findings with the first author edited the final draft and provided overall guidance for the process
Alexandra Martiniuk assessed quality of the studies compared findings with the first author discussed disagreements with the third
author and helped write the first draft and the final review
Mark Stevenson contributed to the protocol reviewed methods of studies and contributed to the final draft
Katrina Thistlethwaite conducted a second literature search extracted full studies and contributed to the final draft
D E C L A R A T I O N S O F I N T E R E S T
None known
S O U R C E S O F S U P P O R T
Internal sources
bull No sources of support supplied
External sources
bull National Health and Medical Research Council of Australia Australia
Salary support for Martiniuk Ivers and Stevenson
bull University of Sydney Sesqui Research Grant Australia
Direct research costs and salary support for Kardamanidis
D I F F E R E N C E S B E T W E E N P R O T O C O L A N D R E V I E W
There are no differences between the protocol and the review
I N D E X T E R M S
Medical Subject Headings (MeSH)
lowastMotorcycles Accident Prevention [lowastmethods] Accidents Traffic [lowastprevention amp control] Licensure Program Evaluation Wounds
and Injuries [mortality lowastprevention amp control]
47Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 5050
MeSH check words
Humans
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 1650
Analysis
Anderson 1980 and Hall 1988 acknowledging the lsquodeterrent ef-fectrsquo described above conducted two separate analyses The first
analysis included the total study population of licence applicants
over two years from the time of their application (the population
denominator) The second analysis was performed on the licensed
(intervention and control) subgroups only two years from the
time of obtaining their licence (the licensed rider denominator)
As only a third were licensed and exposed to riding the motorcy-
cle it was impossible to determine whether reported differences
in crash rates across groups were due to the intervention or to
other lsquoconfoundingrsquo characteristics of those who were determined
in their efforts to obtain the licence When analysis was restricted
only to those who gained their licence the same difficulties re-
mained The benefits of the randomisation process were thus lost
due to this lack of licensure of riders during the process As there
were only three intervention and three control groups in the study
reported by Swaddiwudhipong 1998 analyses were hindered by
smallnumbers for outcome comparisons betweenthe intervention
and the control districts
Risk of bias in the cohort studies and non-randomised
trials
We discuss the methodological quality of cohort studies (seven
prospective studies (Raymond 1979 McDavid 1989 Waller 1992
Kloeden 1994 Billheimer 1998 Perrino 2002 Savolainen 2007)and seven retrospective studies (Osga 1980 Satten 1980 Jonah
1982 Mortimer 1984 Cooper 1988 Mortimer 1988 Davis
1997)) and non-randomised trials (Adams 1985 Leung 1987) in
the one section as the biases for the two study types are similar
The quality of these studies was in general poor
Selection bias
The confounding factors expected to have been addressed in the
studies were sex age riding experience prior driving records (of-
fences crashes) previous training and the following proxies for
risk taking behaviour helmet use speed and alcoholdrug use All
of these factors have been shown to be associated with an increasedrisk of motorcycle crash (Peden 2004 Zambon 2006) The pos-
sible source of bias in cohort studies and non-randomised trials
is that groups may differ on important characteristics at baseline
participants lsquoself-selectrsquo into either the rider training group by de-
ciding to undertake a rider training course or into the lsquocontrol
grouprsquo by deciding not to undertake a course Participants who
undertake a non-mandatory rider training course may differ from
those who choose not to on characteristics such as age sex rid-
ing experience and riding behaviour as well as other possible con-
founders For example riders who take a training course may be
more safety-conscious or on the contrary may do so because they
have previously experienced a crash or because they perceive that
they have poor riding skills Therefore ideally all participants in
the intervention group need to be included in the study as wellas a random sample of controls taken from the same population
as the intervention group If that is not possible intervention and
control groups either need to be carefully matched to make them
comparable or differences at baseline need to be adjusted for in
the analysis
Most of the studies using a non-randomised trial or cohort design
compared an intervention group to a control group that differed
on a number of important factors The studies in general failed to
adjust for important differences found at baseline and their results
cannot be meaningfully interpreted Ten studies (Raymond 1979
Osga 1980 Satten 1980 Jonah 1982 Mortimer 1984 Mortimer
1988 McDavid 1989 Waller 1992 Davis 1997 Savolainen 2007)
assessed intervention and control groups that were not drawnfrom the same population In most of these cases the intervention
group was selected from a list of training graduates who were usu-
ally young female and less experienced riders whereas controls
were recruited from a general population of motorcycle riders who
were usually older male and experienced riders For three stud-
ies it was unclear how they were selected (Leung 1987 Cooper
1988 Perrino 2002) Only Kloeden 1994 recruited groups from
the same population of motorcycle applicants and Adams 1985
from a group of professional riders Nevertheless due to non-ran-
dom selection of intervention and control groups differences at
baseline are likely even though study groups were from the same
population Three studies adjusted for some confounding fac-
tors found at baseline through matching (McDavid 1989 Waller1992 Billheimer 1998) However matching has limitations in
termsof controlling for multiple confounders especially those that
are unknown to influence the effect of rider training on the out-
comes One study matched the control and intervention groups
by excluding controls (approximately 3500 (66 of initial con-
trol group) that differed from the intervention group) (Kloeden
1994) Five studies (Osga 1980 Satten 1980 Adams 1985 Davis
1997 Perrino 2002) did not adjust for differences between study
groups at baseline either in the study design or analysis Results
from these studies are therefore not reliable The other six stud-
ies (Raymond 1979 Jonah 1982 Mortimer 1984 Cooper 1988
Mortimer 1988 Leung 1987 McDavid 1989 Billheimer 1998)
adjusted for some differences between groups at baseline by eitherbi-variate or multi-variate analyses or stratification and standardi-
sation of riding exposure (outcomemiles driven) or stratification
by age group The limitation of stratification is that it is difficult to
control for many variables simultaneously because a large number
of strata will need to be generated resulting in small numbers that
cannot be meaningfully analysed Savolainen 2007 developed sta-
tistical models to assess the probability of a crash after taking the
rider training course in an intervention and control group while
examining differences by sub-groups (eg age) Participation rates
were also very low in this study with only 5584000 (1395)
in the eligible intervention group and 5884000 (147) for the
14Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 1750
control group completing surveys
Detection bias
None of the studies reported blinding of outcome assessors to the
intervention or control status of the participants Seven studies
used driver records containing police recorded data as their main
measure of outcome (Leung 1987 Cooper 1988 McDavid 1989
Waller 1992 Kloeden 1994 Davis 1997 Billheimer 1998) An-
other four studies (Raymond 1979 Satten 1980 Mortimer 1984
Mortimer 1988) only used self-report Raymond 1979 was the
only study using self-report as the main outcome measurement in
which study instruments were validated Self-report might prevent
the underreporting that has been described for the use of police-
driver records (see rsquoRisk of biasrsquo in the randomised trials above)but has other risks of bias especially if questionnaires or inter-
view methods have not been validated and tested for reliability
Recall bias arises from mistakes in recollecting events both be-
cause of failures of memory and because of looking at things lsquowith
hindsightrsquo and possibly changed views (httpwwwcochraneorg
resourcesglossaryhtm) As many studies were retrospective risk
of recall bias is likely and may have increased with length of follow
up For studies that used self-reported outcomes only it is also
possible that social desirability response bias played a role where
participants in the training group may have felt it was lsquoexpectedrsquo
to report good driving histories Multiple methods to measure the
same outcome for both groups decreases the risk of detection bias
for example because self-reported answers can be cross checkedwith other sources of information such as driver records
In this regard three studies (Osga 1980 Jonah 1982 Perrino
2002) used both self-report and driver records Osga 1980 found
that for both intervention and control groups more crashes were
reported through self-report than on state crash records an aver-
age of 54 of self-reported crashes could be found back on state
crash records One industry-based study (Adams 1985) used oc-
cupational health records which were not validated but which are
unlikely to suffer from underreporting due to work cover claims
Savolainen 2007 used information on rider crashes for a five-year
period but did not specify from where this information was ob-
tained
Attrition bias
The majority of studies (Raymond 1979 Satten 1980 Adams
1985 Leung 1987 Kloeden 1994 Davis 1997 Billheimer 1998
Perrino 2002) inadequately reported on losses of participants to
the study We can therefore not be certain that there were no sys-
tematic differences in the characteristics between the intervention
and control group participants Savolainen 2007 reported the re-
sponse rates to their baseline survey but did not report on the per-
centage of outcome data that was available for their study sample
A similar deterrent effect of rider training as described for the ran-
domised trials (Anderson 1980 Hall 1988) occurred in two stud-
ies Jonah 1982 reported that 65(14992310) of the trained rid-
ers stopped riding compared to 47 (9492029) of the untrainedriders They excluded non-active and non-licensed riders before
the study commenced so the differential choice to continue riding
after training did not affect their analyses Leung 1987 reported
that the percentage of participants in the intervention group ob-
taining their licence was 50 (no number provided by authors)
lower than in the control group This may have biased the mea-
surement of outcomes because unlicensed participants will not be
exposed to riding the motorcycle and therefore not be at risk of
being involved in a crash or violation As explained above this
makes it difficult to assess whether the reduction in crashes in the
intervention group is due to a change in driving habits or simply
due to delayed licensure
Mortimer 1984 and Mortimer 1988 describe that about 44 and30 of course graduates respectively were excluded from analysis
because they had not ridden a motorcycle in the previous year
However because these studies were retrospective and the controls
were activeriders at the time of recruitment a decision on whether
the deterrent effect was differential between the two groups could
not be made Though not reported quantitatively Kloeden 1994
also describes the effectof rider training deterringindividuals from
becoming licensed and hence exposure to riding and risk of crash
death or offences involving a motorcycle Osga 1980 reported
that 27 of the trained riders did not ride after the course and
excluded these from the analysis
Use of different data collection strategies may have contributed to
differential loss to follow up in the cohort studies Satten 1980for example interviewed the intervention group over the tele-
phone but mailed questionnaires to the control group Mortimer
1984 and Mortimer 1988 distributed questionnaires in person to
the control group but mailed them to the intervention group
This may have played a role in the observed differential response
rates where more subjects responded in the personally approached
groups
Survival bias
For those studies that assessed offences crashes and injuries but
not death survival bias may have also played a role Motorcyclists
who were fatally or seriously injured in crashes during the follow-
up period after the intervention would have been less likely or
unable to respond to surveys This could bias the effect measures
towards or away from the null None of the studies wrote about
this issue or examined the characteristics of those lost to follow up
in order to address the possibility of survival bias
Intervention integrity
Although the interventions were intended for the entire lsquoexposedrsquo
groups most studies reported difficulty in tracking participants
whoreceivedthe interventionand those whodid not Adams 1985
15Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 1850
estimatedthat 62of thetotalstudysample received the interven-
tion The remaining studies did not report the percentage of thestudy population that received the intervention as was intended
or reported that this percentage was unknown (Raymond 1979
Kloeden 1994 Satten 1980 Jonah 1982 Mortimer 1984 Leung
1987 Cooper 1988 Mortimer 1988 Billheimer 1998 Perrino
2002 Savolainen 2007) Only Raymond 1979 reported the degree
to which their intervention was delivered as intended No studies
reported on possible lsquocontaminationrsquo of participants by other inter-
ventions than the one to which they were allocated Because Osga
1980 Satten 1980 Jonah 1982 Mortimer 1984 Cooper 1988
Mortimer 1988 Waller 1992 and Davis 1997 were retrospective
studies and participants were recruited from a list of graduates it
was known that they had completed and passed the course
Analysis
In some studies some participants who reported at baseline that
they were untrained decided to enroll in a rider training pro-
gramme after the baseline survey The non-randomised trial by
Adams 1985 experienced this move of subjects from the untrained
to the trained group but failed to conduct analyses by lsquointention-
to-treatrsquo This can bias the results in randomised and non-ran-
domised trials if participants who changed groups differ signifi-
cantly with regard to potentially confounding characteristics On
the other hand Perrino 2002 also experienced this move of sub-
jects from theuntrained to the trained groupmore than 100riders(5 of the total sample) originally stated that they were untrained
at baseline but then received training between the baseline and
follow-up surveys Yet since the Perrino 2002 study was a cohort
their analyses then proceeded by including those riders who were
lsquonewlyrsquo trained as exposed Riders who were lsquonewlyrsquo trained since
baseline but with a programme other than the one under study
were excluded from the analysis Nevertheless confounding fac-
tors were not accounted for in the analyses
McDavid 1989 matched exposed with unexposed subjects but
could not find suitable matches for the majority (60) of their
subjects They could therefore only conduct analysis on the result-
ing40 (139346) which is an inappropriate andinefficient anal-
ysis Some studies that reported on the issue of non-active licensedriders(Satten 1980 Jonah 1982 Mortimer 1984 Mortimer 1988
Kloeden 1994) excluded non-active riders from analysis intro-
ducing selection bias as it is likely that riders who continued to
ride differed with regards to potentially confounding character-
istics across intervention and control groups Kloeden 1994 for
example conducted several analyses to assess the effect of various
sub-sets of the control group Original analyses included all con-
trols and led to erroneous results since only a small proportion of
controls hadgone on to get a licence after their learnerrsquos permitand
were actually exposed to the possibility of experiencing a crash
Subsequent analyses compared only intervention and control sub-
jects who had a licence Waller 1992 reported crashes by person-
days of exposure but did not explain how they could be sure that
each study intervention and control participant was an active riderand actually exposed to the risk of experiencing a crash
Potential confounders that are unequally distributed by study arm
at baseline need to be dealt with in the analysis Satten 1980
Adams 1985 and Davis 1997 did not take into account any con-
founding factors either by design or in the analysis McDavid
1989 Waller 1992 and Billheimer 1998 accounted for potential
confounding factors at the design stage by matching the subjects
in the intervention and control groups Other studies (Raymond
1979 Mortimer 1984 Mortimer 1988 Perrino 2002 and also
Waller 1992) adjusted for one or two confounders by stratifica-
tionor standardisation (eg crashesmiles driven age group) Osga
1980 conducteda univariate analysis of crash rates in which no po-
tential confounders were accounted for but also calculated cumu-lative crash rates by mileage travelled Only Jonah 1982 Cooper
1988 and Savolainen 2007 conducted a multivariate analysis
Risk of bias in the case-control studies
A case-control design is suitable for the assessment of risk fac-
tors for outcomes such as motorcycle crashes injuries or deaths
but not appropriate for the evaluation of an intervention such
as motorcycle rider training It can potentially result in spurious
conclusions regarding the efficacy of the intervention because the
intervention has not been applied at random to an intervention
and control group Even with the control of possible confounders
using a case-control design to evaluate an intervention can lead tothese kinds of results since those who choose to undertake rider
training may differ on important prognostic factors that affect the
outcome compared to those who do not choose to participate in
rider training
Selection bias
A potential source of bias in case-control studies as with all other
study types where the intervention is not randomly allocated to
participants is that comparison groups are likely to differ on po-
tentially confounding characteristics This means that ideally all
cases need to be included in the study as well as a random sample
of controls taken from the population that gave rise to the cases If that is not possible cases and controls either need to be carefully
matched to make themcomparable or differences at baseline need
to be adjusted for in the analysis All four case-control studies were
population based Kraus 1975 included all fatalities and injured
cases leading to hospitalisation and appropriately recruited a ran-
dom sample of controls from the same population Hurt 1981 did
not manage to recruit all or a random sample of cases Only half
of the notifications of crashes to the police were received in time
by the researchers the research team for unreported reasons was
only able to respond to half of those again Data collection was
therefore completed for only about a fifth of all cases Neverthe-
less this was still a large sample of 900 cases Haworth 1997 using
16Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 1950
a similar study design was more likely to have included all cases
but this is not entirely clear from the study reported It is unclearfor example whether fatalities were included as they did not use
coronerrsquos reports for the recruitment of cases The MAIDS 2004
study was conducted in five European countries but it was unclear
how cases were recruited whether they were all cases during the
two years that the study lasted or a random sample of all cases
what the response rates were etc Controls were recruited from
randomly selected petrol stations but it was likewise not clear what
the eligibility criteria for inclusion in the study were how they
were recruited (eg randomly) and what the response rate was
Hurt 1981 and Haworth 1997 both matched controls to cases by
recruiting themat the sametime ofdayandsamedayof weekas the
crash occurred Hurt 1981 however recruited controls two years
later andit is likelythat there would have beendifferences betweenthe two comparison groups One important difference was that
during data collection for cases there was very little specialised
motorcycletrainingavailable in the studyregionwhereas two years
later when control data were collected there was some training
available
These four case-control studies also suffered from selection bias
due to potential differences between those that gave consent to
participate in the study and responded to interviews andquestion-
naires and those that did not None of the studies have assessed
such differences although Hurt 1981 did acknowledge that con-
trols giving consent to cooperate may have differed from those
motorcyclists who did not
Detection bias
Measures that can be taken to reduce detection bias in case-con-
trol studies are blinding of interviewers for case or control status
blinding of the participant for the study factor of interest validat-
ing measurement instruments and using these in an identical way
for cases and controls and cross checking self-reported informa-
tion with other more objective sources such as registers None of
the studies reported blinding of the intervention assessors to case
or control status but all used consistent measurement methods for
both case and control groups
All case-control studies used self-report as their main measure toassess whether cases and controls had received rider training and
if so the type of training The major potential source of bias in
case-control studies when self-report is used as a measure is recall
bias One way of reducing recall bias is by double-checking self-
reported data with more objective data Kraus 1975 also used po-
lice accident reports for information on rider characteristics but
such information was not available about rider training Haworth
1997 and MAIDS 2004 did not report on any measures to limit
recall bias Hurt 1981 reported further questioning of cases by
the interviewer when answers did not make sense or when par-
ticipants seemed to contradict themselves as part of the in-depth
unstructured interview technique This qualitative research tech-
nique can result in valuable information but as interviewers were
not blinded to case or control status of the participants it may have led to more thorough examination of cases compared to the
controls and thus differential detection bias
Intervention integrity
The level of detail about the rider training provided in the case-
control studies differed significantly but was minimal for each
Kraus 1975 simply asked whether lsquoanyrsquo motorcycle rider training
had lsquoeverrsquo been carried out Hurt 1981 asked participants whether
they hadbeen self taught informally taught or completed formally
taught rider training courses MAIDS 2004 asked whether partic-
ipants had had pre-license training additional (non-compulsory)
training other training or no training whereas Haworth 1997asked about the number of courses that a participant had com-
pleted whetherthey were basic intermediate or advanced courses
and how long ago the training had been done None of the case-
control studies reported collecting information about further de-
tails about any training programmes (eg type of training cur-
riculum duration of training)
Analysis
Kraus 1975 accounted for confounding factors through the exclu-
sion of females and stratification by age and frequency of motor-
cycle use Hurt 1981 matched cases and controls for time of day
and day of the week that the crash occurred and did not adjust forany confounding factors in the analysis Haworth 1997 matched
groups for time of day and day of the week that the crash occurred
and stratified by age blood alcohol level and licence status in the
analysis None of the studies used multivariate methods MAIDS
2004 did not provide any information about any participant char-
acteristics other than the study exposures of interest and did not
describe any methods to adjust for potential confounding factors
Effects of interventions
We have presented the results in a descriptive manner due to the
substantial heterogeneity in study types interventions and out-comes and the generally poor quality of the studies We have
grouped the interventions by mandatory and non-mandatory pre-
licence rider training and post-licence rider training and further
by study type and by outcome (crashes injurious and fatal crashes
and offences) We discuss results that are reported as statistically
significant (whether a P value was reported or not)
Pre-licence rider training - mandatory
Randomised trials
17Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2050
Crashes
Total group of participants
Anderson 1980 found a reduction in motorcycle crash rates in
the intervention group (MOST skill test with remedial training)
compared to the control group (who received the skill testwithout
a training component) The percent differenceranged from 143
six months after the programme (P lt 005) to 211 (P lt 0001)
after one year and 127 (P lt 001) after two years Hall 1988
found no statistically significant difference in crash rates between
intervention groups at any point in time
Subgroup of participants who obtained their licence only
Anderson 1980 found a reduction in motorcycle crash rates in the
intervention arm but this was statistically significant at one year
only (24 to 218 reduction P lt 005) Hall 1988 found no
statistically significant differences
Offences
Total group of participants
The percentage of offences in Anderson 1980 after six months
and one and two years was slightly higher in the intervention
arm compared to the control group but these differences were
not statistically significant Likewise the study by Hall 1988 did
not find a statistically significant difference in offences between
groups
Subgroup of participants who obtained their licence only
Hall 1988 found no significant difference in offence rates between
the intervention and control groups Anderson 1980 found thatthe percentage of offences increased in the intervention arms com-
pared to the control group This difference was statistically signif-
icant after two years (51 to 115 increase P lt 005)
Cohort studies
Crashes
Neitherof the cohort studies (Kloeden1994 and Billheimer1998)
found a statistically significant difference in crash rates between
the intervention and control groups
Offences
Billheimer 1998 (for the California Motorcyclist Safety Program
course) and Perrino 2002 did not find a statistically significant
difference in offence rates between the intervention and control
groups
Case-control studies
Crashes
MAIDS 2004 found that a slightly higher percentage of controls
(not involved in a crash) had had pre-license rider training than
cases (involved in a crash) This difference was not statistically significant (50 of controls vs 477 of cases OR 091 95 CI
076 to 110)
Pre-licence rider training - non-mandatory
Randomised trial
Injuries and deaths
Swaddiwudhipong 1998 found a significantly lower injury ratein the intervention villages (1051000 population 95 CI 92
to 120) compared to the control villages (1691000 population
95 CI 152 to 185) one year after the intervention This differ-
ence was not maintained two years after the intervention (127
95 CI 112 to 143 intervention 151 95 CI 135 to 167
control) There was no significant difference in death rates be-
tween groups
Cohort studies
Crashes
Cooper 1988 foundno statistically significant differencein crashes
between the intervention and control groups McDavid 1989
found a lower crash rate in the intervention group (14139) com-
pared to the control group (23139) but no test of statistical sig-
nificance was undertaken
Injuries and deaths
McDavid 1989 found that the intervention group had fewer in-
jurious crashes (814 crashes) compared to the control group (16
23 crashes) but no statistical significance test was reported
18Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2150
Offences
Cooper 1988 foundno statistically significantdifference in offencerates between the intervention and control groups McDavid 1989
foundno statistically significantdifference in offence ratesbetween
the intervention and control groups
Non-mandatory rider training
Non-randomised trials
Crashes
Neither Adams 1985 nor Leung 1987 found a statistically signif-icant effect of rider training on crash rates
Offences
Leung 1987 found a statistically significant lower offence rate in
the intervention group compared to the control group for both
males and females and for all age groups except for women 35
years or older (P values not reported)
Cohort studies
Crashes
Osga 1980 in the univariate analysis reported a lower percentage
of crashesin theintervention group (that continued to ride) (28
168595) than in the control group (34 112329) In the cu-
mulative crash rate analysis however Osga1980 found statistically
significant (at the P lt 001 level) higher crash rates in the interven-
tion group than in the control group at allmileage intervals (1000
to 29999 miles) travelled After riding more than 30000 miles
(when over 50 of the trained riders had been involved in a crash
compared to approximately 40 in untrained riders) this differ-
ence was no longer statistically significant Satten 1980 found a
lower percentage of crashes in the intervention group (13 969)
compared to the control group (25 2079) but the differencewas not assessed statistically Davis 1997 found a lower crash rate
in trained riders but it was unclear how this was calculated be-
cause the size of the control group was not reported Jonah 1982
and Billheimer 1998 (for the Experienced Rider Course) found no
statistically significant difference in crashes between the interven-
tion and control groups Although the difference between those
who completed the Experienced Rider Course and no course in
Billheimer 1998 neared significance at six months after the course
(P = 007) large P values were observed for one year after (P =
030) Raymond 1979 found a statistically significant higher crash
rate for the intervention group (1861000 miles ridden) com-
pared to the control group (1181000 miles ridden) in both males
and females (P lt 0001) especially during the first 5000 miles of
riding Waller 1992 found a statistically significant higher crashrate for the intervention group (27 crashes in three years or 070
crashes10000 person days) compared to the control group (5
crashes in three years or 013 crashes10000 person days) with a
crude relative risk of 538 (X 2=1513 plt005) Mortimer 1984
and Mortimer 1988 found no statistically significant difference in
crashes between the intervention and control groups
Savolainen 2007 found that the intervention group (those who
completed the Basic Rider Course) was 44 more likely to be
involved in crashes than the control group and those who com-
pleted the course multiple times were an additional 180 more
likely to be involved in a crash This finding was based on low
numbers and no P value or confidence level was given for either
result
Injuries and deaths
Satten 1980 also found a lower percentage of injurious crashes
(33 39) in the intervention group compared to the control
group(60 1220) although again no statistical significance test
was reported
Offences
Jonah 1982 did find a significantly lower self-reportedoffence rate
in the intervention group (15) compared to the control group(34 P lt 001) (no numbers given) Satten 1980 also found
a lower self-reported offence rate in the intervention group (1
71 versus 19100) but again this was not tested for statistical
significance Mortimer 1984 Mortimer1988 and Billheimer 1998
found no statistically significant differencein offence ratesbetween
the intervention and control groups
Case-control studies
CrashesMAIDS 2004 found that a slightly higher percentage of cases
(involved in a crash) had had additional non-mandatory rider
training than controls (not involved in a crash) This difference
was not statistically significant (17 of cases vs 12 of controls
OR 146 95 CI 064 to 337)
Rider training unspecified
Case-control studies
19Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2250
Injuries and deaths
Hurt 1981 found that the odds of injurious crashes were lower inthose who had received professional training compared to those
who had not Kraus 1975 found that in the younger age group
(less than 20 years) and the older age group (over 35 years) a larger
percentage of injured riders (12 and 21 respectively) reported
prior motorcycle rider training compared with the uninjured con-
trol group (6 and 5 respectively) (P lt 001) (no numbers
given) Haworth 1997 reported several different outcomes one
beingthat having completed at leastone trainingcoursecompared
to no training courses at all had no significant effect on the odds
of crashing (afteradjustment for age blood alcohol level or licence
status (learner permitprobationaryfull)) When comparing three
different levels of training courses (beginner intermediate and ad-
vanced) with no training no significant differences were found
in the odds of crashing However after adjustment for age and
blood alcohol level completing an advanced course as compared
to a beginners course was associated with a significant decrease in
the odds of severe motorcycle crashes resulting in serious injury or
death (OR 02 95 CI 01 to 08)
D I S C U S S I O N
The objective of this review was to quantify the effectiveness of
pre- andpost-licence motorcycle rider training on the reduction of
traffic crash involvement injuries deaths and offences committedby motorcycle riders Most included studies were cohort studies
There were a few randomised and non-randomised trials and case-
control studies
All studies compared either mandatory or non-mandatory formal
training with no formal training except for Haworth 1997rsquos case-
control study that compared levels of training (beginner interme-
diate and advanced) to each other
We found studies that evaluated the effectiveness of manda-
tory pre-licence training in reducing crash and offence rates
(Anderson 1980 Hall 1988 Kloeden 1994 Billheimer 1998
(the basic California Motorcyclist Safety Program component of
the study) Perrino 2002) non-mandatory training that when
passed resulted in a licence (Cooper 1988 McDavid 1989
Swaddiwudhipong 1998) non-mandatory training that could be
taken by motorcycle riders with or without a licence (Raymond
1979 Osga 1980 Satten 1980 Jonah 1982 Mortimer 1984
Adams 1985 Leung 1987 Mortimer 1988 Waller 1992 Davis
1997 Billheimer 1998 (the Experienced Rider Course compo-
nent of the study)) and unspecified rider training on crash injury
and death rates (Kraus 1975 Hurt 1981 Haworth 1997 MAIDS
2004)
The studies provided conflicting evidence regarding motorcycle
rider training which is likely due to the poor design quality and
reporting of the studies Most of the studies included in the review
were likely to be adversely affected by significant selection biasalong with substantial detection bias(measurement and recall bias)
which limits our ability to make conclusive comments
Mandatory pre-licence rider training
There were two RCTs three cohort studies and one case-control
study that examined pre-licence ridertraining The two RCTs eval-
uated the Motorcycle Skills Operating Test (MOST) (Anderson
1980 Hall 1988) Anderson 1980 found a statistically significant
reduction in crashes in the intervention group at six months one
and twoyearsfollow up However when only participantswho ob-
tainedtheirlicence were included in theanalysis thereduction wasstatistically significant only at one year follow up The other RCT
(Hall 1988) found no statistically significant differences in crash
rates at any point of time neither in the total population group
nor in the licensed subgroup Neither of the two cohort stud-
ies assessing crash rates found a statistically significant difference
in crash rates between intervention and control groups ( Kloeden
1994 Billheimer1998) andthe case-control study (MAIDS 2004)
also did not find a statistically significant difference in the odds of
experiencing a crash between trained and untrained riders With
regards to offences neither of the RCTs found a statistically sig-
nificant difference in offence rate between the intervention and
control groups in the analysis of the total group of participants
entering the trial However Anderson 1980 found a statistically significant increase in offence rates in the intervention group but
only after two years of follow up andonly in the analysis of partici-
pants who obtained their licence The two cohort studies assessing
offence rates found no statistically significant difference between
the groups (Billheimer 1998 Perrino 2002)
None of the five studies (RCTs or cohort) found a reduction in
crashes or offences following mandatory pre-licence rider training
except for Anderson 1980 who observed a statistically significant
reduction in crashes in the intervention group at one year in both
analyses It is not clear why both RCTs that evaluated the same
MOST and training intervention observed different results Both
studies enrolled large samples and observed a similar uptake of
licensure following the intervention It is possible that differencesmay have arisen due to different study samples (California ver-
sus New York) or that the intervention delivery differed between
studies
In both trials of the Motorcycle Skills Operating Test including
rider training only a third of motorcycle licence applicants ob-
tained their motorcycle licence It is possible that a licensing pro-
cess including a skills test may reduce crashes through decreased
exposure
Non-mandatory pre-licence training
20Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2350
Three studies assessed the effect of non-mandatory pre-licence
training The Cooper 1988 cohort study found no difference incrash or offence rates between the trained and untrained groups
We cannot however be confident about this result as the study
sufferedfrom selectionbiaswhichcouldhave biasedthe measure of
effect toward the null The RCT conducted by Swaddiwudhipong
1998 examined injury and death rates and found a statistically
significant lower injury rate in the intervention villages after one
year of follow up but no longer at two year follow up There
was no statistically significant difference in death rates between
intervention and control villages Our confidence in the results of
this study are hindered by the inadequacy of the randomisation
process that led to significant differences between the intervention
and control districts on two important variables The intervention
districts were more likely than the control districts to have a validlicence andto wear a helmetSincethese differences were nottaken
into account in later analyses they could explain the differences
observed in injury rates between districts Furthermore with only
six districts randomised the sample size was probably too small
to detect a difference between intervention and control groups
had there been one The McDavid 1989 cohort study found a
lower percentage of injuries in the intervention group although
a statistical test was not conducted McDavid 1989 drew their
intervention and control groups from different populations but
did adjust for confounders although they only analysed 40 of
the initial sample (139 trained and 139 untrained) Because of
these substantial weaknesses we are unable to draw conclusions
from this studyAn overall effect of pre-licence training on the reductionof crashes
injuries deaths and offences was not found although one should
be aware that the research of this issue to date has had significant
flaws However a mandatory skill test (such as the MOST) in
combination with and even without pre-licence training as part
of a licensing process seems to reduce the number of individuals
completing the licensing procedure and therefore indirectly re-
duces crash rates through a reduction in exposure to motorcycle
riding Making licensure a more difficult process may therefore be
seen as advantageous if it delays licensure and therefore reduces
exposure It is possible that the group who chose not to continue
riding a motorcycle would not have been the group to experience a
crash or commit an offence if they had continued to ride anyhowfor example because they are more safety conscious and this was
the reason for their giving up riding in the first place If this is the
case deterring individuals from completing their licence and rid-
ing may not necessarily decrease overall crashes deaths or offence
rates In addition other potentially adverse effects of complicat-
ing a licensing procedure need to be considered such as increases
in unlicensed riding In Thailand non-mandatory rider training
offered as a community intervention may have decreased injury
rates but this needs to be confirmed through further research
Non-mandatory training
The majority of the studies assessed non-mandatory training
(training independent of any licensing procedure) and the results
were varied The two non-randomised trials assessing crash rates found no statistically significant difference in crash rates between
the intervention and control groups (Adams 1985 Leung 1987)
Of the nine cohort studies Satten 1980 found a lower percentage
of crashesin the intervention group although it is unclear whether
this was a statistically significant result Davis 1997 also found a
lower percentage of crashesin theinterventiongroup although it is
unclear with which control group the intervention group wascom-
pared Raymond 1979 Osga 1980 Waller 1992 and Savolainen
2007 found a statistically significant higher crash rate in the in-
tervention group and the remaining cohort studies (Jonah 1982
Mortimer 1984 Mortimer 1988 Billheimer 1998) found no sta-
tistically significant difference in crash rates between intervention
and control groups No conclusions can be drawn from these stud-
ies with confidence The Satten 1980 and Davis 1997 studies did
not adjust for any confounding factors even though important
differences existed between the intervention and control groups
Although Raymond 1979 and Waller 1992 adjusted for a num-
ber of confounders at baseline (Waller 1992) and in the analy-
sis (Raymond 1979 Waller 1992) the intervention and control
groups were drawn from such different populations that it is pos-
sible that unmeasured confounders were responsible for the sta-
tistically significant increased risk of crashes observed in the inter-
vention group The observed difference could also be attributed totheobserved differential loss to followup (Raymond 1979) which
could bias the measure of effect towards or away from the null
The Osga 1980 study found a protective effect of rider training
in the univariate analysis but statistically significant higher crash
rates in the intervention group when accounting for riding expo-
sure Sex (a significantly higher proportion of females in the in-
tervention group) and other potentially confounding factors that
could explain thedifferencebetween the groups were howeverun-
accounted for Finally Savolainen 2007 did control for potential
confounding variables yet it is not clear whether variables were
controlled for simultaneously or one at a time One case-control
study (MAIDS 2004) found no statistically significant reduction
in the odds of experiencing a crash between trained and untrainedriders but the quality of this study was poor mainly because the
analysis did not account for any confounding factors
One cohort study (Satten 1980) assessed the effect of non-manda-
tory rider training on injurious crashes and found a lower percent-
age of injuries in the intervention group although a statistical test
was not conducted As mentioned above the Satten 1980 studyrsquos
validity is reduced due to its lack of adjustment for confounding
factors
The only non-randomised trial that assessed offence rates (Leung
1987) found a statistically significant lower offence rate in the in-
tervention group and of the five cohort studies that assessed of-
fence rates (Satten 1980 Jonah 1982 Mortimer 1984 Mortimer
21Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2450
1988) one study (Jonah 1982) also found a statistically significant
lower offence rate in the intervention group However we have lit-tle confidence in the results from Leung 1987 since potential con-
founding factors were not appropriately addressed Jonah 1982
did conduct a multivariate analysis but was hindered by an initial
low response rate The other four cohort studies did control for
several potential confounders and found no difference in offence
rates following non-mandatory post-licence rider training
The non-randomised trials and cohort studies evaluating non-
mandatory rider training were at high risk of substantial selection
bias due to the non-random allocation of participants as well
as low participation rates and attrition bias and could therefore
not provide conclusive evidence with regard to the effect of the
training It should be noted that despite the biases and other issues
with this group of studies most didnot detectmajor effects relatedto rider training in either direction This may suggest that any
potential impact of rider training will be modest
Studies examining unspecified modes of ridertraining
Three case-control studies assessed whether injured or deceased
motorcyclists were more or less likely to have had rider training
compared to control motorcyclists All had conflicting results
Hurt1981 foundthe odds of havingan injuriouscrash waslower in
those who had received rider training whereas Kraus 1975 found
that the odds of having an injurious crash were higher in youngerand older trained riders Haworth 1997 who compared advanced
versus beginner training found no effect on motorcycle injuries
or deaths although they did find that those who had completed
advanced training had reduced chances of crashing compared to
those who had done a beginnersrsquo training course Similar issues
with uncontrolled confounding and biases of selection as described
above arose with the three case-control studies Hurt 1981 suffers
from inadequate control of confounding via poor matching and
both the Kraus 1975 and the Haworth 1997 findings were only
controlled for limited factors by means of stratified analysis It is
likely that these limitations are the reason for the contradictory
directions of effect observed for the three case-control studies
Methodological weaknesses of the studies
In summary most studies investigating the effectiveness of mo-
torcycle rider training on reducing crashes injuries and death suf-
fered from serious methodological weaknesses
The majority of studies were non-randomised and were there-
fore limited mainly by selection bias Although the approaches of
matching stratification or regression analyses can minimise bias
(Fleiss 2004) they were poorly employed in the reviewed studies
The inadequate adjustment for confounders in particular adds
to our lack of confidence in the studiesrsquo results since it has been
demonstrated that individuals who choose to take motorcycle rider
training differ from those not taking rider training For instance
trained riders often differ from controls in age sex and experi-ence usually ride less and may have a more positive attitude to
safety as indicated by the reported more frequent wearing of seat
belts (Mortimer 1982)
The two large randomised trials reported that a large proportion
of their study group were never exposed to the risk of offences or
crashes because they never attained licensure This made it impos-
sible to assess whether a reduction in crashes in the intervention
group was a result of training or due to delayed or never attained
licensure
In some studies the sample sizes were too small to detect any sta-
tistically significant differences Follow-up time after training dif-
fered between studies and was an impediment to observing any
real effects of training if they existed As riding experience ac-cumulates over time any differences in outcomes between those
trained and untrained would diminish in studies with lengthy fol-
low-up times This dilution of any possible effect of rider training
may have also occurred in studies that mixed both novice and ex-
perienced riders in the study groups since it is hypothesised that
novice riders serve to benefit most from rider training This would
increase the difficulty in finding any effect of rider training par-
ticularly when sample sizes are small
Detection bias may also account for some of the differences ob-
served between trained and untrained riders Detection bias would
have occurred if the self-report hospital driver or police records
used to record the outcomes of crash injury death or offences
differed between the trained and untrained riders Research hasdemonstrated that not all crashes are reported to police for ex-
ample small crashes or off-road crashes (Kraus 1975) even serious
injury crashes have been found to be underreported (Anderson
1980) However this would only introduce detection bias if the
reporting of outcome differs for trained and untrained riders This
does appear possible in that responsible riders may be more likely
to enroll in a rider training course and also may be more likely
to report crashes to police than irresponsible riders meaning that
the number of crashes from riskier riders could be underrepre-
sented on records An analysis to investigate the effectiveness of
rider training may then show that rider training is not effective in
reducing crashes
Rather than using reported outcomes some studies used self-re-ported survey data although this technique suffers from non-
response biases In fact two studies reported differences in de-
mographic characteristics in response rates to questionnaires
(Anderson 1980 Hall 1988) Self-reported data also introduce
the risk of recall bias survival bias and social desirability response
bias Recall bias may have been equally distributed among trained
and untrained riders howeverit is hypothesised that trained riders
may have felt it was lsquoexpectedrsquo to report good driving histories
hence leading the rider training to appear effective
22Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2550
Limitations of this review
Limitations of this review were that we found very few recentstudies (the two most recent ones published in 2002 and 2007)
and that we found very few studies conducted in low-and middle
income countries even though our search strategy was set up to
identify such studies Anotherlimitation wasthat forthe reviewed
studies we could not assess all outcomes of rider training that are
potentially useful Some of the included studies assessed outcomes
such as use of safety clothing including helmets Such outcomes
are valuable since the use of motorcycle helmets has been demon-
strated to be effective in preventing head injuries (Liu 2004) De-
spite this the use of safety clothing was not included as an out-
come in this review because the studies measuring this outcome
were limited by poor data quality due to very low response rates
Another outcome of interest in assessing the effectiveness of rider
training is lsquotype of injuryrsquo As Haworth 2005 state in their review
of motorcycle licensing and training rider training could be con-
sidered effective if injuries become less severe because motorcycle
riders have learned how to avoid more serious injuries through
rider training The studies included in this review did not specify
type of injury and so this outcome could not be assessed Last
changes in knowledge and attitudes following rider training are
other potentially useful outcomes We have excluded such studies
(Osga 1979 Clayton 1990 Goldenbeld 2004) as these outcomes
are proxies rather than direct measures of the effectiveness of mo-
torcycle rider training We opted to report on the outcomes of
crashes injury death and offences alone since there is little ev-
idence about how changes in knowledge and attitudes relate to
these outcomes
A U T H O R S rsquo C O N C L U S I O N S
Implications for practice
Due to the poor quality of studies included in this review we were
unable to draw any conclusions about the effectiveness of rider
training on crash rates injurious or fatal or offence rates The
findings do suggest that mandatory pre-licence training may be
an impediment to completing a motorcycle licensing process thuspossibly indirectly reducing crash injury death and offence rates
through a reduction in exposure to riding a motorcycle On the
basis of the existing evidence it is not clear if and what type of
training reduces the risk of crashes injuries deaths or offences in
motorcyclists and the selection of the best rider training practice
can therefore not be recommended The suggestion that manda-
tory pre-licence training indirectly reduces exposure to riding the
motorcycle because trainees are impeded from completing the li-
censing process does not help those motorcyclists who persist in
obtaining their licence and riding the motorcycle Some type of
rider training is likely necessary to teach motorcyclists to ride a
motorcycle safely Good policy is based on sound evidence and
evaluation but such evidence for motorcycle rider training is seri-
ously lacking Policy makers are therefore encouraged to include arigorous evaluation component to any new or existing rider train-
ing programme Such an evaluation should ideally be randomized
and assess whether the training programme is effective and if so
which component(s) of the programme work best
Implications for research
Given theresults of this review which found littlereliable evidence
about the effectiveness of motorcycle rider training it is necessary
to build up an evidence base of the effectiveness of motorcycle
rider training in the reduction of crash injury death and offence
rates To this end well designed randomised controlled trials need
to be conducted that examine the impact of rider training oncrashes and offences A randomised controlled trial is the preferred
method as it would significantly reduce most biases described in
this review To minimise the impact of participants delaying or
never attaining their licence and therefore exposure as found in
the randomised trials in this review such a trial should include
riders who are already licensed and reliable measures of riding
exposure in a reasonably lengthy follow up period It is expected
that a randomised controlled trial of sufficient sample size would
observe equal balance on possible confounders by study arm how-
ever an assessment of potentially important confounders (such as
helmet wearing) by study group would enable control for these
variables in the analyses if necessary It is imperative that the effec-
tiveness of each component of the rider training (such as theorypractice knowledge skills and modification of behaviour) and the
synergy of such components are evaluated Outcomes should be
objective ideally those that are routinely collected rather than self-
reported outcomes Even though riding the motorcycle poses dif-
ferent challenges and incurs different risks lessons may be learned
from research into the effects of car driver training
Motorcycles are a common means of transport in low- and middle
income countries and usage is increasing As the burden of mo-
torcycle crash victims is mostly borne in such countries research
should be focused on the effectiveness of types of rider training on
the reduction of crashes injuries and fatalities in such countries
and be supported by higher income countries This review found
only one study conducted in a middle income country that sug-gested decreasing injury rates It would be beneficial to conduct
additional research to confirm this suggestion and to enhance our
knowledge of the role of rider training in low- and middle-income
settings
March 2006 saw the launch of the University of North Carolina
Highway Safety Research Centre partnership with the Motorcycle
Safety Foundation (MSF) and plans were revealed to conduct a
study of the potential benefits of motorcycle rider training The
study is funded by the MSF and the National Highway Traffic Sa-
fety Administration (NHTSA) Called rsquoThe Discovery Projectrsquo it
will compare data on crash rates from riders participating in a basic
23Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2650
motorcycle safety class with riders participating in a basic motor-
cycle safety class as well as a variety of continuing education classesThe collection of pilot data was due to start in March 2007 and
the data collection portion of the study is expected to take three
years (Highway Safety Research Center) It is vital that researchers
and policy makers collaborate to develop furtherhigh quality eval-
uations of motorcycle rider training programmes to help inform
development of new and effective training programmes
A C K N O W L E D G E M E N T S
Staff from The University of Sydney library for helping us locate
and retrieve full reports of many papers
R E F E R E N C E S
References to studies included in this review
Adams 1985 published data only
Adams A Collingwood V Job RFS Evaluation of the 1982 Australia
Post motorcycle rider training scheme NSW Australia Traffic
Accident Research Unit Traffic Authority NSW 1985 [ RN 85]
Anderson 1980 published data only
Anderson J Ford JL Peck RC Improved motorcyclist licensing and testing project Washington DC National Highway Traffic Safety
Administration (NHTSA) USA 19801ndash217 [
DOTndashHSndash805ndash598]
Billheimer 1998 published data only
Billheimer JW Evaluation of the California Motorcyclist Safety
Program Transportation Research Record 19981640100ndash9 [
TRB Report Number 980652]
Cooper 1988 published data only
Cooper PJ Rothe JP Motorcyclists who they are and why they do
what they do Transportation Research Record 1988116878ndash85
Davis 1997 published data only
David CF Evaluation of community traffic safety programs and
motorcycle operator training programs Connecticut Department of Transportation 199763
Hall 1988 published data only
Hall CJ Buchanan L Engle R Motorcycle rider education evaluationproject Washington DC USA National Highway Traffic Safety
Administration (NHTSA) US Department of Transportation
1988 [ DTNH 22ndash80ndashCndash05123]
Haworth 1997 published data only
Haworth N Smith R Brumen I Pronk N In Monash University
Accident Research Centre editor(s) Case-control study of motorcycle
crashes Victoria Australia Department of Transport and Regional
Development The Federal Office of Road Safety 19971ndash149 [
ISBN 0 642 25501 6 ]
Hurt 1981 published data only
Hurt H Ouellet J Thom D Motorcycle accident cause factors and
identification of countermeasures USA Department of
Transportation 1981 [ Technical Report DOT HS 806ndash862]
Jonah 1982 published data onlylowast Jonah BA Dawson N Bragg B Are formally trained motorcyclists
safer Accident Analysis amp Prevention 198214(4)247ndash55
Jonah BA Dawson NE Bragg BW Evaluation of the Canada
Safety Councilrsquos Motorcycle Training Program Proceedings of theInternational Motorcycle Safety Conference Washington DC
USA 198059ndash93
Kloeden 1994 published data only
Kloeden CN Moore VM McLean AJ Evaluation of the pre-license
training program for motorcyclists in South Australia Walkerville
South Australia Australia NHMRC Road Accident Research Unit
The University of Adelaide 19941ndash37 [ ISBN 0 7308 0607 3]
Kraus 1975 published data only
Kraus JF Riggins RS Franti CE Some epidemiological features of
motorcycle collision injuries American Journal of Epidemiology 1975102(1)74ndash98
Leung 1987 published data only
Leung KS Reding VA Evaluation of the Wisconsin motorcycle rider course Wisconsin Bureau of Policy Planning and Analysis
Wisconsin Department of Transportation 1987
MAIDS 2004 published and unpublished data
MAIDS In-depth investigations of accidents involving powered
two wheelers ACEM (Acemacemeu) 2004 Vol Final report
20179
McDavid 1989 published data only
McDavid JC Lohrmann BA Lohrmann G Does motorcycle
training reduce accidents Evidence from a longitudinal quasi-
experimental study Journal of Safety Research 198920(2)61ndash72
Mortimer 1984 published data only
Mortimer R Evaluation of the motorcycle rider course Accident
Analysis amp Prevention 198416(1)63ndash71
24Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2750
Mortimer 1988 published data only
Mortimer RG A further evaluation of the motorcycle rider courseJournal of Safety Research 198819(4)187ndash96
Osga 1980 published data only
Osga GA An investigation of the riding experiences of MSF rider
course participants Dissertation South Dakota USA The
University of South Dakota 19801ndash224
Perrino 2002 published data only
Perrino CS Ahmed A Callender A Rozier E Cantwell A Stewart
O The role of Marylandrsquos motorcycle rider course in promoting saver
behaviors and attitudes Baltimore USA National Transportation
Center Morgan State University 20021ndash52
Raymond 1979 published data only
Raymond S Tatum S An evaluation of the effectiveness of the RAC
ACU motorcycle training scheme - final report Road Safety ResearchUnit Dept of Civil Engineering University of Salford London
19791ndash131
Satten 1980 published data only
Satten RS Analysis and evaluation of the motorcycle rider courses
in thirteen Illinois counties International Motorcycle Safety
Conference Washington DC 1980145ndash93
Savolainen 2007 published and unpublished data
Savolainen Peter Mannering Fred Additional evidence on the
effectiveness of motorcycle training and motorcyclistsrsquo risk-taking
behaviour Transportation Research Board 2007 Vol March 20
19
Swaddiwudhipong 1998 published data only
Swaddiwudhipong W Boonmak C Nguntra P Mahasakpan P
Effect of motorcycle rider education on changes in risk behavioursand motorcycle-related injuries in rural Thailand Tropical Medicine
and International Health 19983(10)767ndash70
Waller 1992 published data only
Waller FL An analysis of the impact of the Texas Department of
Public Safety Motorcycle Operator Training Program upon
motocycle related traffic accidents Texas State University Public
Administration Program Applied Research Projects 199277
References to studies excluded from this review
Anderson 1980a published data only
Anderson LC Better biking a motorcycle rider improvement
program International Motorcycle Safety Conference Washington
DC USA Motorcycle Safety Foundation 1980231ndash53Aupetit 2008 published and unpublished data
Aupetit S Riff J Buttelli O Espie S Ergonomics research program
on motorcycle learning Contributions to the development of a
motorcycle simulator for training Transport Research Arena
Ljubljana 20089
Awane 1999 published data only
Awane T Integrating simulators in motorcycle safety education
Journal of International Association of Traffic and Safety Sciences 199923(1)26ndash35
Baldi 2005 published data only
Baldi S Baer JD Cook AL Identifying best practices states in
motorcycle rider education and licensing Journal of Safety Research200536(1)19ndash32
Batchler 1988 published data only
Batchler MW Evaluation of the pre-license motorcycle training course at the New South Wales traffic education centre Federal Office of
Road Safety (FORS) 19881ndash122 [ Report No CR 77 ]
Blanchard 2006 published data only
Blanchard HT Tabloski PA Motorcycle safety educating riders at
the teachable moment Mosby (CV) Company 200632330ndash2
Bonnett 2005 published data only
Bonnett J Mature age skills training for experienced motorcycle
riders Motorcycle ridersrsquo association of the Australian Capital
Territory 2005
Braver 2007 published data only
Braver ER et alPersuasion and licensure a randomized controlled
intervention trial to increase licensure rates among Maryland
motorcycle owners Traffic Injury Prevention 20078(1)39ndash46Clayton 1980 published data only
Clayton AB Sudlow DE An evaluation of the effectiveness of a one day training programme for learner motorcyclists Berkshire UK
Transport and Road Research Laboratory 19801ndash23 [ ISSN
0266ndash7045]lowast Clayton AB Sudlow DE An evaluation of the effectiveness of a one-day training programme for learner motorcyclists Berkshire UK
Road Safety Division Berkshire 19801ndash23 [ISSN 0266ndash7045]
Council 1978 published data only
Council FM Desper LP Dutt AK Driver education for motorcycle operation University of North Carolina Highway Safety Research
Center 19781ndash40
Crowther 2005 published data onlylowast Crowther G Engaging with motorcyclists UK police and theBikeSafe Road Safety Programme Published proceedings of the
Australasian Road Safety Research Policing Education Conference
Wellington New Zealand 2005 issue ISBN 0ndash473ndash10636ndash11ndash11
Daniello 2009 published data only
Daniello A Gabler HC Mehta YA Effectiveness of motorcycle
training and licensing Transportation Research Board 88th Annual
Meeting 200913
Dewanti 2007 published data only
Dewanti M Challenging efforts in promoting young motorcyclist
safety in Indonesia Road Safety on Four Continents 14th
International Conference Swedish National Road and Transport
Research Institute 2007
Du Plooy 1990 published data onlyDu Plooy DW Motorcycle training in South Africa The Human
Element International Motorcycle Safety Conference Orlando
Florida USA Motorcycle Safety Foundation 19904-1 to 4-9
Engel 1989 published data only
Engel U Krogsgaringrd Thomsen L [Faeligrdselsundervisning og
propaganda ndash en faeligrdselssikkerhedsfremmende foranstaltning]
Dansk Vejtidsskrift 198912295ndash7
Goldenbeld 2002 published data only
Goldenbeld C Houwing S de Craen S The development of
driving skills of young moped riders [De ontwikkeling van de
rijvaardigheid van jonge bromfietsers] SWOV Institute for Road
Safety Research (Stichting Wetenschappelijk Onderzoek
Verkeersveiligheid) Leidschendam the Netherlands 200263
25Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2850
Goldenbeld 2004 published data only
Goldenbeld C Twisk D de Craen S Short and long term effects of moped training a field experiment Transportation Research Part F 20047(1)1ndash16
Haworth 2000 published data only
Haworth N Smith R Kowadlo N Motorcycle training in Victoria
evaluation of rider training curriculum in Victoria Monash
University Accident Research Centre Reports 2000 Vol 165 issue
3131
Haworth 2005 published data only
Haworth N Mulvihill C Review of motorcycle licensing and
training Monash University Accident Research Centre 2005 [
Report no240]
Haworth 2006 published data only
Haworth N Mulvihill C Rowdon P Teaching old dogs new tricksTraining and older motorcyclists Journal of the Australasian College of Road Safety 200718(4)20ndash5lowast Haworth NL Mulvihill C Rowden PJ Teaching old dogs new
tricks Training and older motorcyclists Australasian Road Safety
Research Policing and Education Conference Gold Coast
Queensland 2006
Hill 1983 published data only
Hill D Motorcycle rider training testing and licensing Motor
Cycle Council of New South Wales Sydney 1983
Imatake 1980 published data only
Imatake Y On the activities in Japan of motorcycle safe riding
promotion committee (MSRPC) - especially the education of
unskilled riders of class 1 motorbikes International Motorcycle
Safety Conference Washington DC USA Motorcycle Safety
Foundation 1980255ndash68
Ingebritsen 1990 published and unpublished data
Ingebrigtsen S [Risikofaktorer ved ferdsel med moped og
motorsykkel] TOslashI-rapport 66 Oslo Transportoslashkonomisk
institutt 1990
Jain 2009 published and unpublished data
Jain A et alTwo wheeler accidents on Indian roads a study from
Mangalore India Elsevier 200916130ndash3
Jonah 1979 published data only
Jonah BA Dawson NE Validation of the motorcycle operator skill
test Accident Analysis amp Prevention 197911(3)163ndash71
Jonah 1981 published data onlyJonah BA Dawson NE Bragg WE Predicting accident
involvement with the Motorcycle Operator Skill Test Accident Analysis amp Prevention 198113(4)307ndash418
Kadar 1990 published data only
Kadar KJ Countermeasures motorcycle rider education delivery
system The Human Element International Motorcycle Safety
Conference Motorcycle Safety Foundation 19906-27 to 6-41
Kelsey 1986 published data only
Kelsey SL Liddicoat C Ratz M Licensing novice motorcyclists a
comparison of Californiarsquos standard test and the MOST II (Motorcycle
Operator Skill Test) administered at centralized testing offices Sacramento California USA California Department of Motor
Vehicles Research and Development Office 19861ndash74
Koch 1980 published data only
Koch H Advanced course for motorcycle riders InternationalMotorcycle Safety Conference Washington DC USA Motorcycle
Safety Foundation 1980269ndash81
Koch 1990 published data only
Koch H Brendicke R How to influence risk taking behaviour
within motorcycle rider training programs The Human Element
International Motorcycle Safety Conference Orlando Florida
USA Motorcycle Safety Foundation 1990Motorcycle Safety
Foundation 1990 p 9-1 to 9-25A
Labbett 2005 published data only
Labbett S Langham M Training can make the problem worse
Road Safety Congress 20051ndash12
Law 2005 published and unpublished data
Law TH Radin Umar RS Zulkaurnain S Kulanthayan S Impactof the effect of economic crisis and the targeted motorcycle safety
programme on motorcycle-related accidents injuries and fatalities
in Malaysia International Journal of Injury Control and Safety
Promotion 200512(1)9ndash21
Lowes 1990 published data only
Lowes BF In-traffic rider training The Human Element
International Motorcycle Safety Conference Orlando Florida
USA Motorcycle Safety Foundation 19907-46 to 7-57
Mahon 1976 published data only
Mahon J Teaching accident avoidance on the motorcycle Journal of Traffic Safety Education 1976231ndash29
Maxwell 1985 published data only
Maxwell GS Evaluation of a Program for Training MotorcycleRiders Evaluation and Program Planning 19858(4)339ndash47
Mayhew 1996 published data only
Mayhew DR Simpson SH Effectiveness and role of driver educationand training in a graduated licensing system Ottowa Canada The
Traffic Injury Research Foundation 19961ndash89 [ ISBN
0ndash920071ndash12ndash0]
McKnight 1987 published data only
McKnight AJ Milburn BG An evaluation of the modified
motorcycle operator skill test exercises NHTSA Report 1987 [
Report No DOT HS 807 212 ]
McPherson 1976 published data only
McPherson K McKnight AJ The development and evaluation of a
motorcycle skill test manual and knowledge test Washington DCUSA National Highway Safety Administration 1976
Mohan 2004 published data only
Mohan D Evidence-based interventions for road traffic injuries in
South Asia Journal of the College of Physicians and SurgeonsPakistan 200414(12)746ndash7
Morgan 1980 published data only
Morgan JK Advanced motorcycle training course content and
study evaluation International Motorcycle Safety Conference
Washington DC USA Motorcycle Safety Foundation 1980
283ndash312
Motoki 1990 published data only
Motoki M Yamazaki S A study on effective motorcycle rider
education The Human Element International Motorcycle Safety
26Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2950
Conference Orlando Florida USA Motorcycle Safety
Foundation 19909-26 to 9-57A
Nairn 1992 published data onlylowast Nairn RJ Motorcycle safety research literature review 1987 to
1991 Federal Office of Road Safety Australia 1993 Vol Report
CR 117
Newland 1980 published data only
Newland R Darby J Automotive practicespre motorcycle rider
program International Motorcycle Safety Conference Washington
DC USA Motorcycle Safety Foundation 1980127ndash9
Newland 1982 published data only
Newland R Motorcycle rider training and licensing requirements
Driver training steering a course for the future Melbourne
Victoria Transport Regulation Board 1982
Newland 1990 published data only
Newland R Motorcycle training and licensing in Victoria 1980-
1990 The Human Element International Motorcycle Safety
Conference Orlando Florida USA Motorcycle Safety
Foundation 1990
Ormston 2003 published data only
Ormston R Dudleston A Pearson S Stradling Evaluation of
Bikesafe Scotland Scottish Executive Social Research Edinburgh
2003
Osga 1979 published data only
Osga GS Ellingstad VS An evaluation of the effectiveness of the 1978
South Dakota motorcycle rider course program South Dakota USA
University of South Dakota 19791ndash89
Panou 2007 published data only
Panou M Bekiaris E A new concept on the integration of driving
simulators in driving training the train-all approach Road Safety
on Four Continents 14th International Conference Swedish
National Road and Transport Research Institute 2007
Peterson 1990 published data only
Peterson A Kadar K Status report 1980-1989 Minnesota rider
education statistics The Human Element International
Motorcycle Safety Conference Orlando Florida USA 199013-
113 to 13-129
Prem 1984 published data only
Prem H Good MC Motorcycle rider skills assessment ACT
Australia Office of Road Safety Department of Transport 1984
1ndash377 [ Report No CR 34 ]
Radin Umar 2006 published data only
Radin Umar RS Motorcycle safety programmes in Malaysia how
effective are they International Journal of Injury Control and Safety Promotion 200613(2)71ndash9
Rockwell 1990 published data only
Rockwell TH Kiger S Carnot M An evaluation of the Ohio
motorcyclist enrichment program Ohio Department of Highway
Safety 1990
Rothe 1987 published data only
Rothe JP A motorcycle riding course Motorcyclists Image and reality Canada Insurance Corporation of British Columbia 1987
59ndash83
Rowden 2007 published data only
Rowden P Watson B Haworth N What can riders tell us aboutmotorcycle training A view from the other side of the fence
Australasian Road Safety Research Policing and Education
Conference Melbourne Australia Melbourne 2007
RSC Victoria 1993 published data only
Road Safety Committee Inquiry into Motorcycle Safety in Victoria
Parliament of Victoria 1993
Russam 1979 published data only
Russam K Motorcycle training and licensing in Japan UK Transport
and Road Research Laboratory (TRRL) 1979 [ Report 916 ]
Saffron 1981 published data only
Saffron DG Driver and rider education training and licensing a
brief review Traffic Accident Research Unit Traffic Authority of
New South Wales NSW Australia Traffic Accident Research UnitTraffic Authority of New South Wales 1981
Schulz 1990 published data only
Schulz U Motorcycle accidents among learner riders Motorcycle
licensing Evaluation of the effectiveness of a Graduated Licensing Scheme Forshungshefte des Institut fur Zweiradsicherheit 1990
Sexton 2004 published data only (unpublished sought but not used)
Sexton B Baughan C Elliott M Maycock G The accident risk of
motorcyclists 2004TRL Report TRL 607(ISSN 09684107)58
Shephard 1986 published data only
Shepherd R Do Motorcycle Safety Foundation programs work
Driver 198619(9)16ndash8
Simpson 1990 published data onlySimpson HM Mayhew DR The promotion of motorcycle safety
training education and awareness Health Education Research
Theory and Practice 19905(2)257ndash64
Sokolov 1990 published data only
Sokolov O Safe motorcycle riding some aspects of motorcycle
riding in the USSR The Human Element International
Motorcycle Safety Conference Motorcycle Safety Foundation
19904ndash37 - 4-48
Steffens 1988 published data only
Steffens U Gawatz R Willmes G Effectiveness of mofa courses
[Wirksamkeit von Mofakursen] Unfall- und sicherheitsforschung
strasenverkehr 198867168
Sudlow 2003 published data only
Sudlow D Scoping study on motorcycle training Road Safety
Research Report London UK The British Institute of Traffic
Education Research Department for Transport 2003 Vol June
[ Report No 36]
Tasmania 1995 published data only
Mannering FL Grodsky LL Motorcycle rider training scheme
Traffic Education Research Centre Armidale Australia Driver
Training Australia Limited 1995 Vol April1ndash56
Tetard 1985 published data only
Tetard C Motorcycle ridersrsquo training to detect indices in an
intersection situation [Lrsquoentrainement a la prise drsquoinformation dans
lrsquoapprentissage de la conduite motocycliste] Le Travail Humain198548(2)147ndash59
27Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3050
Thackray 1980 published data only
Thackray RM Prescott JC Field test of the motorcycle ridercourse International Motorcycle Safety Conference Washington
DC USA Motorcycle Safety Foundation 1980195ndash7
Thompson 1994 published data only
Thompson M Evaluation of compulsory basic training for
motorcyclists Transport Research Laboratory Crowthorne
Berkshire UK 1994 [ Report 63]
Tobler 1990 published data only
Tobler U Individual driving instruction for motorcycle driving
learners the Swiss experiment of motorcycle driving school The
Human Element International Motorcycle Safety Conference
Orlando Florida USA Motorcycle Safety Foundation 1990
Transport 1989 published data only
Transport Communications Road Crash Statistics Australia Bureauof Transport and Communications Economics 198914ndash20
Tuan 2007 published data only
Tuan VA Shimizu T Towards development and evaluation of the
motorcycle drivers re-education program in Vietnam modeling of
motorcycle driverrsquos undesired behaviors 11th World Conference
on Transport Research 200725
Victor 1980 published data only
Victor DJ Traffic safety education for motorcyclists International
Motorcycle Safety Conference Washington DC USA Motorcycle
Safety Foundation 1980221ndash8
Waterer 1990 published data only
Waterer TH Compulsory training for motorcyclists from liberty
to compulsion The Human Element International Motorcycle
Safety Conference Orlando Florida USA Motorcycle Safety
Foundation 19904-49 to 4-57
Weller 1989 published data only
Weller RB Chandler EW Motorcycle safety and motorcycle
education past research and survey results Journal of Public Policy
amp Marketing 1989893ndash108
Wells 1981 published data only
Wells P Observations of some RACACU motorcycle training
schemes Transport and Road Research Laboratory Crowthorne
Berkshire UK 1981
Winn 1986 published data only
Winn G Two important trends in motorcycle safety regulations
rider education and conspicuity improvement American
Motorcyclist Association 1986147ndash52
Wisher 1988 published data only
Wisher G Reid J Motorcycle accidents in Norfolk a three-year study of motorcycle training and the Part 1 test Norfolk UK Road Safety
Division Highways Department Norfolk County Council 1988
Wood 1987 published data only
Wood HT Bowen R Evaluation of the revised learner permit scheme
July 1983 to December 1985 Victoria Australia Road Traffic
Authority 19871ndash27 [ ISBN 0 7306 0247 8]
Yamazaki 1990 published data only
Yamazaki S Nakanishi M Survey on effect of Honda rider
education The Human Element International Motorcycle Safety
Conference Orlando Florida USA 199012-167 to 12-187
References to ongoing studies
Burgess 2010 unpublished data only
Burgess C Broughton P Fylan F Stradling Interim evaluation of
the UKs National Ride program In L Dorn editor(s) Driver
behaviour and Training Vol IV Aldershot Ashgate Publishing
Limited 2010
Additional references
Anderson 1980
Anderson J Ford JL Peck RC Improved motorcyclist licensing and
testing project Volume I and II National Highway Traffic Safety
Administration US department of Transportation Washington
DC 1980
Bjornstig 1985
Bjornstig U Bylund P Lekander T Brorsson B Motorcycle fatalitiesin Sweden Acta Chirurgica Scandinavica 1985151(7)577ndash81
Clayton 1990
Clayton A Sudlow DE An evaluation of the effectiveness of a one-
day training program for learner motorcyclists The Human
Element International Motorcycle Safety Conference Orlando
Florida USA 199013-105 to 13-111
Cochrane Collaboration 2008
Higgins JP Green S Cochrane Handbook for Systematic Reviews of
Interventions 500 The Cochrane Collaboration 2008
Elliot 2003
Elliott MA Baughan CJT Broughton J Grayson G Knowles J et
alMotorcycle Safety a Scoping Study TRL [ Report No
0968ndash4107]Fleiss 2004
Fleiss JL Levin B Paik MC Statistical methods for rates and proportions 3rd Edition John Wiley amp Sons Inc 2004 [ ISBN
978ndash0ndash471ndash52629ndash2 ]
FORS 1999
FORS Federal Office of Road Safety Road risk for sober licensed
motorcyclists Vol Monograph 27 Federal Office of Road Safety
19991ndash2
Haworth 2005
Haworth N Mulvihill C Review of motorcycle licensing and training
Victoria Australia Monash University Accident Research Centre
20051ndash83 [ Report No 240]
Huwiler-Muumlntener 2002Huwiler-Muumlntener K Juumlni P Junker C Egger M Quality of
Reporting of Randomized Trials as a Measure of Methodologic
Quality JAMA 20022872801ndash4
Johnston 2008
Johnston P Brooks C Savage H Fatal and serious road crashes
involving motorcyclists Road Safety Research and Analysis Report
ATSB Monograph 20 Canberra Department of Infrastructure
Transport Regional Development and Local Government 0205
081ndash26
Juumlni 2001
Juumlni P Altman DG Egger M Systematic reviews in health care
Assessing the quality of controlled clinical trials BMJ 2001323
42ndash6
28Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3150
Kopits 2003
Kopits E Cropper M Traffic fatalities and economic growth TheWorld Bank Development Research Group Infrastructure and
Environment Policy research working paper 3035 2003
Leung 1983
Leung P Chau BTW Motorcycle accidents in Hong Kong pattern
of injury treatment causative factors and socio-economic
implications Injury 198315(1)23ndash9
Liu 2004
Liu B Ivers R Norton R Blows S Lo SK Helmets for preventing
injury in motorcycle riders Cochrane Database of Systematic Reviews 2004 Issue 2 [DOI 10100214651858CD004333pub3]
Mohan 2002
Mohan D Road safety in less-motorized environments future
concerns International Journal of Epidemiology 200231(3)527ndash32Mortimer 1982
Mortimer RG An evaluation of the motorcycle rider course
Proceedings of the 20th conference of the American Association for
Automotive Medicine Arlington Heights IL USA American
Association for Automotive Medicine 198277ndash96
Mullin 1997
Mullin BT Risk factors for motorcycle injury the role of age
gender experience training and alcohol PhD thesis Auckland
New Zealand University of Auckland 1997
Noordzij 2001
Noordzij PC Forke E Brendicke R Chinn BP Integration of needs
of moped and motorcycle riders into safety measures review and
statistical analysis in the framework of the European research
project PROMISING SWOV Institute for Road Safety Research
Leidschendam The Netherlands 2001 [ Report No Dndash2001ndash5]
Osga 1979
Osga G Ellingstad VS An evaluation of the effectiveness of the 1978 South Dakota Motorcycle Riders Course Program South Dakota
University of South Dakota Vermillion Human Factors Lab 1979
Oxford English Dictionary Online
httpoxforddictionariescom Oxford English Dictionary Online
Oxford University Press 2010
Peden 2004
Peden M Scurfield R Sleet D World Report on Road TrafficInjury Prevention Geneva World Health Organization Available
at httpwwwwhointviolence˙injury˙preventionpublications
road˙trafficworld˙reporten 2004Chapter 3
Radin Umar 1995
Radin Umar R Macakay G Hills B Preliminary analysis of
motorcycle accidents Short-term impacts of the running
headlights campaign and regulations in Malaysia J Traffic Med 199523(1)17ndash27
Tsai 1995
Tsai YJ Wang JD et alCase-control study of the effectiveness of
different types of helmets for the prevention of head injuries among
motorcycle riders in Taipei Taiwan American Journal of
Epidemiology 1995142(9)974ndash81
Vis 1995
Vis AA In-depth study of the hazards of motorcycling [De
onveiligheid van motorrijden nader bekeken] SWOV Institute for
Road Safety Research Leidschendam the Netherlands 1995 [
Report No Rndash95ndash69]
Weaver 1990
Weaver EA Motorcycle safety education in the United States
Motorcycle Safety Foundation Orlando Florida USA 1990
Winn 1987
Winn G Two important trends in motorcycle safety regulations
rider education and conspicuity improvement No more
information available Detroit MI US Society of Automotive
Engineers 1987147ndash52
Yuan 2000
Yuan W The effectiveness of the rsquoride-brightrsquo legislation for
motorcycles in Singapore Accident Analysis amp Prevention 200032
(4)559ndash63
Zambon 2006
Zambon F Hasselberg M Factors affecting the severity of injuries
among young motorcyclists a Swedish nationwide cohort study
Traffic Injury Prevention 20067(2)143ndash9lowast Indicates the major publication for the study
29Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3250
C H A R A C T E R I S T I C S O F S T U D I E S
Characteristics of included studies [ordered by study ID]
Adams 1985
Methods Non-randomised trial
Participants Initial sample size
Total of 162 Australia Post telegram delivery riders
Not clear how many in intervention and control groups
Analysis sample size
Intervention group 63Control group 35
Total 98 (61)
Interventions No formal training vs voluntary rider training course of 12 hours over 2 consecutive days (theory 4 hours practice
8 hours)
Intervention integrity
No information given
Outcomes Australia Post recorded crashes
Length of follow up 1 year
Notes Study conducted in NSW Australia
Anderson 1980
Methods Randomised trial
Participants Initial sample size
Total 41188 motorcycle licence applicants
Analysis sample size
Intervention group with skills training 13160
Intervention group without skills training 12634
Control group 15080
Total 40874 (99)
Interventions Standard licensing skill test vs the Motorcycle Operators Skill Test (MOST) with or without a training component
MOST is not to be confused with rider training because it is a licensing test The training component was for
participants who failed the skill test on the first attempt was of 3 hours duration and consisted of instructions about
the skills needed to pass the skill test The training consisted of 30 min theory and 2 frac12 hours practice on the
motorcycle
Intervention integrity
33 in MOST with skills training 31 in MOST without skills training and 43 in control group obtained their
motorcycle licence in the end
Outcomes Police recorded crashes and offences
Length of follow up 3 years
30Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3350
Anderson 1980 (Continued)
Notes Study conducted in California USA
Billheimer 1998
Methods Cohort study
Participants Initial sample size
Intervention group Unclear numbers of CMSP graduates and ERC graduates
Control group Over 16000 untrained motorcyclists
Analysis sample size
1139 matched CMSP-pairs and 1182 ERC-pairs
Interventions No formal training vs the mandatory (for those younger than 18 years) pre-licence 16-hours basic California
Motorcyclist Safety Program (CMSP) and the 8-hour Experienced Rider Course (ERC)
Intervention integrity
No information given
Outcomes Police recorded crashes and offences
Length of follow up 2 years
Notes Study conducted in California USA
Cooper 1988
Methods Cohort study
Participants Initial sample size
Intervention group 418 graduates and 57 who failed
Control group 402
Analysis sample size
Intervention group unclear
Control group unclear
Total 863 (984)
Interventions No formaltrainingvs non-mandatory pre-licence rider trainingThe content of thetrainingwas notfurther explainedIntervention integrity
Participants had all completed the rider training
Outcomes Police recorded crashes and offences
Length of follow up 7 years
Notes Study conducted in British Columbia Canada
31Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3450
Davis 1997
Methods Retrospective cohort study
Participants Initial sample size
Intervention group 9320 graduates
Control group unclear
Analysis sample size
Intervention group 9320 graduates
Control group unclear
Interventions No formal training vs the non-mandatory Connecticut Motorcycle Operator Training Program The content of
training was not further explained
Intervention integrity Participants had all completed the rider training
Outcomes Police recorded crashes severity of crashes
Length of follow up 1 year
Notes Study conducted in Connecticut USA
Hall 1988
Methods Randomised trial
Participants Initial sample sizeIntervention group A 6248
Intervention group B 6616
Intervention group C 6996
Control group 6604
Total 26464 motorcycle licence applicants
Analysis sample size
Intervention group A 6012 (96)
Intervention group B 6428 (97)
Intervention group C 6778 (97)
Control group 6502 (98)
Interventions Standard licensing skill test vs the Motorcycle Operator Skill Test (MOST II) without training (A) MOST II with
3 hours training (B) or MOST II with 20 hours training (C) The 3 hours training comprised instructions aboutthe skills needed to pass the skill test (30 min theory and 2 frac12 hours practice on the motorcycle) The 20 hours
training consisted of the Motorcycle Safety Foundationrsquos Motorcycle Rider Course (MRC) comprising of 8 hours
theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-road practice
Intervention integrity
25 and 26 obtained their motorcycle licence in the lsquoMOST 3-hour trainingrsquo and lsquoMOST 20-hour trainingrsquo
intervention arm respectively In the lsquoMOST testrsquo intervention arm 29 obtained their licence and in the control
group 35
Outcomes Police recorded crashes and offences
Length of follow up 3 years
32Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3550
Hall 1988 (Continued)
Notes Study conducted in New York USA
Haworth 1997
Methods Population based case-control study
Participants Cases Motorcycle riders or pillion riders involved in motorcycle crash (on road) Recruited from hospital and police
records Not clear whether all cases (especially fatalities) were included
Community controls Multiple matched sample of motorcycle rider or pillion who passed the crash site at the same
time of day and day of week that the crash occurred
Initial sample size
Cases 336
Controls 1195
Analysis sample size
Cases 234 (70)
Controls 561 (47)
Interventions Basic handling skills course beginner intermediate or advanced course measured through self-report not validated
Outcomes Injuries with an Injury Severity Score of 5 or greater or admission for at least four hours and fatalities from a
motorcycle crash (from hospital and police records)
Length of follow up two years or more prior to the interview
Notes Study conducted in Victoria Australia
Hurt 1981
Methods Population based case-control study
Participants Cases Motorcycle riders who had a motorcycle crash that was reported to police (traffic accident reports) Not all
cases included
Community controls Multiple matched sample of motorcycle riders who passed the crash site at the same time of
day and day of week as the crash occurred but 2 years later
Initial sample size
Cases 4500 notifications of crashes to police (only 50 received in time research team able to respond to 1126 for
unreported reasons data collection completed for 900)
Controls Number approached unreported 2310 consented and interviewed
Analysis sample size
Cases 808 (808900 data on rider training) - 18 of initial sample
Controls 670 (6702310 data on rider training) - lt 29 of initial sample
Interventions Training experience self taught friends-family school-club formal-professional others measured through self-
report validated
Outcomes Injuries and fatalities from a motorcycle crash (from hospital records and coronerrsquos office)
33Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3650
Hurt 1981 (Continued)
Notes Study conducted in Los Angeles USA
Jonah 1982
Methods Cohort study
Participants Initial sample size
Intervention group
2310 licensed MTP graduates who are active riders
Control group
2029 licensed motorcyclists who are active riders
Non-licensed and non-active riders excluded
Analysis sample size
Intervention group
811 interviews (35) 533 driver records (23)
Control group 1080 interviews (71) 905 driver records (45)
Interventions No training vs the non-mandatory Canada Safety Council Motorcycle Training Program (MTP) comprising of 4
hours theory and 16 hours practise (6 hrs on-road) over 2 weekends
Intervention integrity
Participants had all successfully completed the MTP
Outcomes Self-reported and police reported crashes and offencesLength of follow up 4 years
Notes Study conducted in Ontario Canada
Kloeden 1994
Methods Cohort study
Participants Initial sample size
Intervention group 2533 licence applicants
Control group Unclear
Sample of riders obtaining probationary licenceIntervention group 1625 (64)
Control group 5015 (unclear)
Non-active riders excluded from control group (~66)
Analysis sample size
Intervention group 631 (39)
Control group 318 (6)
Interventions No formal training vs the mandatory pre-licence Ridersafe course consisting of two levelsthe first is a training course
in the basics of motorcycle riding where a pass entitles the rider to obtain a learnerrsquos permit The second is a skill
test where a pass entitles the rider to obtain a probationary motorcycle licence The training course consists of two 4
hour training sessions on separate days involving instruction videos and actual riding on motorcycles The skill test
consists of a 4 hour preparation course of 74 minutes of lecturediscussion 38 minutes of video and 75 minutes of
34Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3750
Kloeden 1994 (Continued)
on-course riding which is the testing phase
Intervention integrity
Unclear what percentage in the Ridersafe group and the control group obtained their probationary licence halfway
and their full motorcycle licence in the end
Outcomes Police reported crashes
Length of follow up 1 year
Notes Study conducted in SA Australia
Kraus 1975
Methods Population based case-control study
Participants Cases All motorcycle riders with a confirmed medically treated motorcycle injury (including fatalities) from all
hospitals in study base and peripheral area hospitals and all deaths due to motorcycle crashes in the study base All
serious injuries (leading to hospitalisation) and fatalities likely to be included
Community controls A random sample of registered motorcycle owners in the same temporary and geographical
study base as the cases arose from
Initial sample size Cases 1273
Controls 738
Analysis sample size Cases 741 (58)
Controls 434 (59)
Interventions Motorcycle rider training (not specified) measured through self-report
Outcomes Injuries leading to hospitalisation fatalities
Length of follow up not specified
Notes Study conducted in Sacramento county USA
Leung 1987
Methods Non-randomised trial
Participants Initial sample sizeIntervention group unknown number of MRC graduates
Control group unknown number of licensed motorcyclists
Analysis sample size
Intervention group 2941
Control group 43094
Interventions No formal training vs non-mandatory Wisconsin motorcycle rider course a 22 hours off-street programme that
teaches the basic rules of safe motorcycling
Intervention integrity
50 in the MRC group and 100 in the control group obtained their motorcycle licence in the end
35Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3850
Leung 1987 (Continued)
Outcomes Police recorded crashes and offences
Length of follow up up to 3 years
Notes Study conducted in Wisconsin USA
MAIDS 2004
Methods Population based case-control study
Participants Cases Motorcycle riders who had a motorcycle crash Unclear whether all cases included
Community controls Motorcycle riders identified at petrol stations Unclear whether matched to cases
Initial sample size
Cases not reported
Controls not reported
Analysis sample size
Cases 921 (Italy 200 Netherlands 200 Spain 121 Germany 250 France 150)
Controls 923 (Italy 200 Netherlands 200 Spain123 Germany 250 France 150)
Interventions Pre-license rider training or additional training (non-compulsory)
Outcomes Crashes
Notes Study in five sampling areas located in five European countries (France Germany Netherlands Spain and Italy)
McDavid 1989
Methods Cohort study
Participants Sample size after first match
Intervention group 346 male training graduates
Control group 346 male motorcycle riders licensed in same year
Sample size after second match
Intervention group 139 (40)
Control group 139 (40)
Analysis sample size
Intervention group 139 (40)
Control group 139 (40)
Interventions No formal training vs the non-mandatory pre-licence British Columbia Safety Councilrsquos 37-hour motorcycle safety
programme
Intervention integrity
Participants had all completed and passed the training
Outcomes Police reported crashes injurious crashes
Length of follow up 1-6 years
36Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3950
McDavid 1989 (Continued)
Notes Study conducted in British Columbia Canada
Mortimer 1984
Methods Cohort study
Participants Initial sample size
Intervention group 608 MRC graduates
Control group 360 motorcycle riders
Response rates
Intervention group 360 (59)
Control group 324 (90)
Non-active riders excluded
Analysis sample size
Intervention group 213 (35)
Control group 303 (84)
Interventions No formal training vs the non-mandatory Motorcycle Safety Foundationrsquos Motorcycle Rider Course comprising of
8 hours theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-road practice
Intervention integrity
Participants had all completed and passed the MRC
Outcomes Self-reported crashes and offencesLength of follow up 1 year
Notes Study conducted in Illinois USA
Mortimer 1988
Methods Cohort study
Participants Initial sample size
Intervention group 2307 MRC graduates
Control group 556 active licensed motorcycle riders
Response ratesIntervention group 1298 (56)
Control group 500 (90)
Non-active riders excluded
Analysis sample size
Intervention group 913 (40)
Control group 500 (90)
Interventions No formal training vs the non-mandatory Motorcycle Safety Foundationrsquos Motorcycle Rider Course comprising of
8 hours theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-road practice
Intervention integrity
Participants had all completed and passed the MRC
37Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4050
Mortimer 1988 (Continued)
Outcomes Self reported crashes and offences
Length of follow up 1 year
Notes Study conducted in Illinois USA
Osga 1980
Methods Cohort study
Participants Initial sample sizeIntervention group 1163 MSF MRC participants invited 818 (74) responded
Control group 329 motorcycle riders responded to recruitment strategies
Non-active riders excluded from analysis
Analysis sample size
Intervention group 595 (73)
Control group 329 (100)
Interventions No formal training vs the non-mandatory Motorcycle Safety Foundationrsquos Motorcycle Rider Course (MRC) com-
prising of 8 hours theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-road
practice
Intervention integrity
Participants had all successfully completed the MRC
Outcomes Self-reported crashes police recorded crash reports
Length of follow up 6 months - 3 12 years
Notes Study conducted in South Dakota USA
Perrino 2002
Methods Cohort study
Participants Initial sample size
Intervention group 2144 (unclear) MRC graduates (motorcycle licence applicants)
Control group 2601 (unclear) regular licensing test graduates (motorcycle licence applicants)Analysis sample size
Intervention group 1070 (unclear)
Control group 867 (unclear)
Interventions No formal training vs the mandatory (for those younger than 18 years) pre-licence Maryland Motorcycle Rider
Course (MRC) consisting of 8 hours theory and 12 hours practice
Intervention integrity
No information given
Outcomes Self-reported and police recorded crashes and offences
Length of follow up 1 year
38Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4150
Perrino 2002 (Continued)
Notes Study conducted in Maryland USA
Raymond 1979
Methods Cohort study
Participants Initial sample size
Intervention group
258 volunteers mainly (but not all) novice riders
Control group
1104 provisional licence applicants
Analysis sample size
Intervention group 190 (74)
Control group 627 (57)
Interventions No formal training vs the non-mandatory RACACU (Royal Automobile ClubAuto-Cycle Union) training scheme
The training program consisted of either 24 one-hour sessions or 13 two-hour sessions Twelve hours of the training
focused on theory about basic operational procedures traffic laws and highway code The practical session focused
on basic operational skills riding on a training area and riding in traffic with an instructor A proficiency test was
administered at the end of the course
Intervention integrity
No information given
Outcomes Self-reported crashes (including minor non-injury non-damage and off-road crashes)
Length of follow up 1 year
Notes Study conducted in the UK
Satten 1980
Methods Cohort study
Participants Initial sample size
Intervention group interviewers kept interviewing until they had 100 riders selected from a population of 891
enrollees in the MSF MRC course (unknown how many were asked and how many refused to participate)Control group unknown number of registered motorcycle riders
Non-active riders excluded from analysis
Analysis sample size
Intervention group 71
Control group 100
Interventions No formal training vs the non-mandatory Motorcycle Safety Foundationrsquos (MSF) Motorcycle Rider Course (MRC)
comprising of 8 hours theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-
road practice
Intervention integrity
Participants had all successfully completed the MRC
39Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4250
Satten 1980 (Continued)
Outcomes Self-reported crashes injuries and offences
Length of follow up 3 years
Notes Study conducted in Illinois USA
Savolainen 2007
Methods Cohort study
Participants Initial sample size
Intervention group 4000 riders that had conducted the MSF MRC course (passed or failed)
Control group 4000 motorcycle owners or permit or endorsement holders from the Indiana Bureau of Motor
Vehicles database
Analysis sample size
Intervention group 739 (18)
Control group 588 (15)
Interventions Intervention integrity
No information given
Outcomes Probability of crash involvement
Length of follow up up to 5 years
Notes Study conducted in Indiana USA
Swaddiwudhipong 1998
Methods Randomised trial
Participants Initial sample size
Intervention group
all 16 villages of 3 sub-districts
(participants were practising motorcyclists licensed and unlicensed)
Control group
18 villages of 3 other sub-districtsAnalysis sample size
Intervention group 3 districts (100)
Control group 3 districts (100)
Interventions No rider education vs non-mandatory community based rider education consisting of theory on motorcycle injury
risk helmet use traffic laws vehicle regulations and traffic signs and a written and skill test leading to a licence
Intervention integrity
No information given
Outcomes Injuries fatalities
Length of follow up 2 years
40Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4350
Swaddiwudhipong 1998 (Continued)
Notes Study conducted in Tak Province rural Thailand
Waller 1992
Methods Cohort study
Participants Initial sample size
Intervention group 349 riders who succesfully completed the Department of Public Safety motorcycle operator
training in 1988
Control group 349 motorcyclists who never completed the DPS motorcycle operator training
Analysis sample size
Intervention group 349
Control group 349
Interventions No DPS motorcycle operator training vs non-mandatory DPS motorcycle operator training The DPS training is
the Motorcycle Safety Foundationrsquos Motorcycle Rider Course comprising eight hours of theory (lectures discussions
films and multiple choice tests) a textbook and 12 hours off-road practice
Intervention integrity
All 349 riders succesfully completed the rider training
Outcomes Police recorded crashes
Length of follow up 3 years
Notes Study conducted in
Characteristics of excluded studies [ordered by study ID]
Study Reason for exclusion
Anderson 1980a Study type review and process evaluation of a rider course
Aupetit 2008 Outcome proposal of a rider training curriculum that combines traditional training and a motorcycle riding
simulator
Awane 1999 Study type review on use of simulators in rider and driver training
Baldi 2005 Intervention not rider training but intervention integrity of rider training availabilty of rider education
implementation of graduated licensing system etc on a state-wide level
Batchler 1988 Study type description of a training course
Blanchard 2006 Study type not a study assessing the association between rider training and outcomes
Bonnett 2005 Outcome qualitative data regarding satisfaction with the training programme
41Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4450
(Continued)
Braver 2007 Outcome licensure rates
Clayton 1980 Outcome knowledge skills attitude
Council 1978 Outcome crash reduction in ANY vehicle not motorcycles exclusively
Crowther 2005 Study type qualitative study
Daniello 2009 Study type review of the effectiveness of motorcycle training and licensing
Dewanti 2007 Intervention emotional characteristics behaviour (ie not motocycle rider training)
Du Plooy 1990 Study type review of types of rider training courses in South Africa
Engel 1989 Study type uncontrolled before-after study
Goldenbeld 2002 Outcome skills
Goldenbeld 2004 Outcome knowledge and skills
Haworth 2000 Intervention balance between attitudinal andvehicle control skills acrosscurriculla of motorcycle rider training
courses
Haworth 2005 Study type review
Haworth 2006 Study type cross-sectional survey
Hill 1983 Study type review and description of training system
Imatake 1980 Study type course descriptionreview
Ingebritsen 1990 Intervention risk factors for motorcycle crashes
Jain 2009 Study type cross-sectional descriptive study of motorcycle crashes over a 5 year period
Jonah 1979 Study type cross-sectional survey Outcome scores on motorcycle operator skill test
Jonah 1981 Intervention evaluation of a motorcycle rider skill test not motorcycle rider training
Kadar 1990 Study type review of components of rider training courses
Kelsey 1986 Intervention evaluation of a motorcycle rider skill test not motorcycle rider training
Koch 1980 Study type reviewcourse description
Koch 1990 Study type description of course content
42Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4550
(Continued)
Labbett 2005 Study type essay about possible explanations for motorcycle deaths
Law 2005 Intervention targeted motorcycle safety programme not specifically rider training
Lowes 1990 Study type review and descriptions of rider training
Mahon 1976 Study type review of motorcycle rider training methods
Maxwell 1985 Outcomesrecommendations forprogramme operatorsand otherson howthe motorcycle trainingprogramme
works and how it can be improved
Mayhew 1996 Study type review
McKnight 1987 Study type evaluation of methodological and administrative aspects of a motorcycle operator skill test
McPherson 1976 Outcome knowledge and skills
Mohan 2004 Study type review of evidence of interventions to reduce road traffic injuries
Morgan 1980 Study type reviewcourse description
Motoki 1990 Study type proposal and development of rider training curriculum
Nairn 1992 Study type literature review
Newland 1980 Study type description of development of motorcycle safety course
Newland 1982 Study type descriptionreview of a rider training course
Newland 1990 Study type uncontrolled BeforeAfter study
Ormston 2003 Outcome changes in attitude and self reported behaviour
Osga 1979 Outcome knowledge attitudes and skills
Panou 2007 Intervention new simulation tool to teach driverrider
Peterson 1990 Study factor demographic details of course participants completing the course vs those not completing the
course
Prem 1984 Study type experimental
Study factor examines association between skills and performance in tests
Radin Umar 2006 Intervention motorcycle safety programme not specifically rider training
Rockwell 1990 Study type methodologically flawed study type
43Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4650
(Continued)
Rothe 1987 Study type qualitative study of rider training in general
Rowden 2007 Study type qualitative study
RSC Victoria 1993 Study type review of motorcycle rider training and education in Victoria
Russam 1979 Study type review of training programmes in Japan
Saffron 1981 Study type literature review
Schulz 1990 Study factor predictors of motorcycle crashes
Sexton 2004 Study type cross-sectional analytic analysis
Shephard 1986 Study factor reviewdescription of programme
Simpson 1990 Study type review of motorcycle rider trainingeducation
Sokolov 1990 Study factor review of motorcycle riding courses in USSR
Steffens 1988 Outcome self-reported behaviour experiences and researcher observed riding behaviour
Sudlow 2003 Study type review of motorcycle training in the UK
Tasmania 1995 Study type cross-sectional descriptiveprocess evaluation
Tetard 1985 Study type experimental
Outcome detection of indices
Thackray 1980 Outcome instructional effectiveness user acceptance administrative feasibility
Thompson 1994 Study type cross-sectional survey
Outcome attitudes and knowledge
Tobler 1990 Study type review of course curricula
Transport 1989 Intervention syllabus for rider training
Tuan 2007 Study type cross-sectional study (survey) which does not assess the effects of an intervention but examines
motocyclistsrsquo behaviour
Victor 1980 Study type review
Waterer 1990 Study type review of UK rider training issues
Weller 1989 Study type review of studies evaluating motorcycle safety and education measures
44Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4750
(Continued)
Wells 1981 Study type description of different methods of training
Winn 1986 Study type review of rsquohistoryrsquo of rider training
Wisher 1988 Study type cross-sectional survey
Wood 1987 Study type Interrupted Time Series but did not meet the criteria for an ITS as published by the Cochrane
Effective Practice and Organisation of Care Group
Yamazaki 1990 Study type cross-sectional survey
Outcome riding skills skills retention and attitude
Characteristics of ongoing studies [ordered by study ID]
Burgess 2010
Trial name or title UKs National Ride program
Methods Cohort study
Participants Motorcycle offenders
Interventions Police led diversion program
Outcomes Self reported attitudinal and behavioural measures
Starting date
Contact information
Notes Expected to be completed in early 2010
45Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4850
D A T A A N D A N A L Y S E S
This review has no analyses
A P P E N D I C E S
Appendix 1 Search strategy
We searched the following electronic databases
bull Cochrane Injuries Group Specialised Register (searched 22 Sept 2008)
bull CENTRAL (The Cochrane Library 2008 Issue 3)
bull TRANSPORT 1988 to 200706 (latest date available)
bull ATRI Australian Transport Index (to Sept 2008)
bull MEDLINE (Ovid SP) 1950 to Sept 2008
bull EMBASE (Ovid SP) 1980 to Sept 2008
bull CINAHL (EBSCO) 1982 to Sept 2008
bull WHOLIS (World Health Organization Library Information System) (to Sept 2008)
bull PsycINFO (Ovid SP) 1806 to Sept 2008
bull LILACS (BIREME) (Latin American and Caribbean Health Sciences) (1982 to Sept 2008)
bull ISI Web of Science Social Sciences Citation Index (SSCI) 1970 to Sept 2008
bull ERIC (searched 25 Sept 2008)
bull ZETOC (searched 25 Sept 2008)
bull SIGLE (searched 25 Sept 2008)
All of the databases were searched using the following text-word search strategy The strategy was adapted using either rsquorsquo or rsquo$rsquo as atruncation symbol
1 motor-cycl or motorcycl or motor-bik or motorbik or motor-scooter or motorscooter or scooter or moped or motocycl
or motocicl or moto
2 train or educat or instruct or program or course or qualif or learn or improv
3 1 and 2
H I S T O R Y
Protocol first published Issue 2 2005
Review first published Issue 10 2010
46Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4950
C O N T R I B U T I O N S O F A U T H O R S
Katina Kardamanidis was the first author of the protocol conducted the literature search extracted full studies assessed the studies
according to inclusion criteria assessed quality of the studies compared findings with the second author discussed disagreements with
the third author wrote the drafts and the final review
Rebecca Ivers was the second author of the protocol assessed the studies according to inclusion criteria assessed quality of the studies
compared findings with the first author edited the final draft and provided overall guidance for the process
Alexandra Martiniuk assessed quality of the studies compared findings with the first author discussed disagreements with the third
author and helped write the first draft and the final review
Mark Stevenson contributed to the protocol reviewed methods of studies and contributed to the final draft
Katrina Thistlethwaite conducted a second literature search extracted full studies and contributed to the final draft
D E C L A R A T I O N S O F I N T E R E S T
None known
S O U R C E S O F S U P P O R T
Internal sources
bull No sources of support supplied
External sources
bull National Health and Medical Research Council of Australia Australia
Salary support for Martiniuk Ivers and Stevenson
bull University of Sydney Sesqui Research Grant Australia
Direct research costs and salary support for Kardamanidis
D I F F E R E N C E S B E T W E E N P R O T O C O L A N D R E V I E W
There are no differences between the protocol and the review
I N D E X T E R M S
Medical Subject Headings (MeSH)
lowastMotorcycles Accident Prevention [lowastmethods] Accidents Traffic [lowastprevention amp control] Licensure Program Evaluation Wounds
and Injuries [mortality lowastprevention amp control]
47Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 5050
MeSH check words
Humans
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 1750
control group completing surveys
Detection bias
None of the studies reported blinding of outcome assessors to the
intervention or control status of the participants Seven studies
used driver records containing police recorded data as their main
measure of outcome (Leung 1987 Cooper 1988 McDavid 1989
Waller 1992 Kloeden 1994 Davis 1997 Billheimer 1998) An-
other four studies (Raymond 1979 Satten 1980 Mortimer 1984
Mortimer 1988) only used self-report Raymond 1979 was the
only study using self-report as the main outcome measurement in
which study instruments were validated Self-report might prevent
the underreporting that has been described for the use of police-
driver records (see rsquoRisk of biasrsquo in the randomised trials above)but has other risks of bias especially if questionnaires or inter-
view methods have not been validated and tested for reliability
Recall bias arises from mistakes in recollecting events both be-
cause of failures of memory and because of looking at things lsquowith
hindsightrsquo and possibly changed views (httpwwwcochraneorg
resourcesglossaryhtm) As many studies were retrospective risk
of recall bias is likely and may have increased with length of follow
up For studies that used self-reported outcomes only it is also
possible that social desirability response bias played a role where
participants in the training group may have felt it was lsquoexpectedrsquo
to report good driving histories Multiple methods to measure the
same outcome for both groups decreases the risk of detection bias
for example because self-reported answers can be cross checkedwith other sources of information such as driver records
In this regard three studies (Osga 1980 Jonah 1982 Perrino
2002) used both self-report and driver records Osga 1980 found
that for both intervention and control groups more crashes were
reported through self-report than on state crash records an aver-
age of 54 of self-reported crashes could be found back on state
crash records One industry-based study (Adams 1985) used oc-
cupational health records which were not validated but which are
unlikely to suffer from underreporting due to work cover claims
Savolainen 2007 used information on rider crashes for a five-year
period but did not specify from where this information was ob-
tained
Attrition bias
The majority of studies (Raymond 1979 Satten 1980 Adams
1985 Leung 1987 Kloeden 1994 Davis 1997 Billheimer 1998
Perrino 2002) inadequately reported on losses of participants to
the study We can therefore not be certain that there were no sys-
tematic differences in the characteristics between the intervention
and control group participants Savolainen 2007 reported the re-
sponse rates to their baseline survey but did not report on the per-
centage of outcome data that was available for their study sample
A similar deterrent effect of rider training as described for the ran-
domised trials (Anderson 1980 Hall 1988) occurred in two stud-
ies Jonah 1982 reported that 65(14992310) of the trained rid-
ers stopped riding compared to 47 (9492029) of the untrainedriders They excluded non-active and non-licensed riders before
the study commenced so the differential choice to continue riding
after training did not affect their analyses Leung 1987 reported
that the percentage of participants in the intervention group ob-
taining their licence was 50 (no number provided by authors)
lower than in the control group This may have biased the mea-
surement of outcomes because unlicensed participants will not be
exposed to riding the motorcycle and therefore not be at risk of
being involved in a crash or violation As explained above this
makes it difficult to assess whether the reduction in crashes in the
intervention group is due to a change in driving habits or simply
due to delayed licensure
Mortimer 1984 and Mortimer 1988 describe that about 44 and30 of course graduates respectively were excluded from analysis
because they had not ridden a motorcycle in the previous year
However because these studies were retrospective and the controls
were activeriders at the time of recruitment a decision on whether
the deterrent effect was differential between the two groups could
not be made Though not reported quantitatively Kloeden 1994
also describes the effectof rider training deterringindividuals from
becoming licensed and hence exposure to riding and risk of crash
death or offences involving a motorcycle Osga 1980 reported
that 27 of the trained riders did not ride after the course and
excluded these from the analysis
Use of different data collection strategies may have contributed to
differential loss to follow up in the cohort studies Satten 1980for example interviewed the intervention group over the tele-
phone but mailed questionnaires to the control group Mortimer
1984 and Mortimer 1988 distributed questionnaires in person to
the control group but mailed them to the intervention group
This may have played a role in the observed differential response
rates where more subjects responded in the personally approached
groups
Survival bias
For those studies that assessed offences crashes and injuries but
not death survival bias may have also played a role Motorcyclists
who were fatally or seriously injured in crashes during the follow-
up period after the intervention would have been less likely or
unable to respond to surveys This could bias the effect measures
towards or away from the null None of the studies wrote about
this issue or examined the characteristics of those lost to follow up
in order to address the possibility of survival bias
Intervention integrity
Although the interventions were intended for the entire lsquoexposedrsquo
groups most studies reported difficulty in tracking participants
whoreceivedthe interventionand those whodid not Adams 1985
15Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 1850
estimatedthat 62of thetotalstudysample received the interven-
tion The remaining studies did not report the percentage of thestudy population that received the intervention as was intended
or reported that this percentage was unknown (Raymond 1979
Kloeden 1994 Satten 1980 Jonah 1982 Mortimer 1984 Leung
1987 Cooper 1988 Mortimer 1988 Billheimer 1998 Perrino
2002 Savolainen 2007) Only Raymond 1979 reported the degree
to which their intervention was delivered as intended No studies
reported on possible lsquocontaminationrsquo of participants by other inter-
ventions than the one to which they were allocated Because Osga
1980 Satten 1980 Jonah 1982 Mortimer 1984 Cooper 1988
Mortimer 1988 Waller 1992 and Davis 1997 were retrospective
studies and participants were recruited from a list of graduates it
was known that they had completed and passed the course
Analysis
In some studies some participants who reported at baseline that
they were untrained decided to enroll in a rider training pro-
gramme after the baseline survey The non-randomised trial by
Adams 1985 experienced this move of subjects from the untrained
to the trained group but failed to conduct analyses by lsquointention-
to-treatrsquo This can bias the results in randomised and non-ran-
domised trials if participants who changed groups differ signifi-
cantly with regard to potentially confounding characteristics On
the other hand Perrino 2002 also experienced this move of sub-
jects from theuntrained to the trained groupmore than 100riders(5 of the total sample) originally stated that they were untrained
at baseline but then received training between the baseline and
follow-up surveys Yet since the Perrino 2002 study was a cohort
their analyses then proceeded by including those riders who were
lsquonewlyrsquo trained as exposed Riders who were lsquonewlyrsquo trained since
baseline but with a programme other than the one under study
were excluded from the analysis Nevertheless confounding fac-
tors were not accounted for in the analyses
McDavid 1989 matched exposed with unexposed subjects but
could not find suitable matches for the majority (60) of their
subjects They could therefore only conduct analysis on the result-
ing40 (139346) which is an inappropriate andinefficient anal-
ysis Some studies that reported on the issue of non-active licensedriders(Satten 1980 Jonah 1982 Mortimer 1984 Mortimer 1988
Kloeden 1994) excluded non-active riders from analysis intro-
ducing selection bias as it is likely that riders who continued to
ride differed with regards to potentially confounding character-
istics across intervention and control groups Kloeden 1994 for
example conducted several analyses to assess the effect of various
sub-sets of the control group Original analyses included all con-
trols and led to erroneous results since only a small proportion of
controls hadgone on to get a licence after their learnerrsquos permitand
were actually exposed to the possibility of experiencing a crash
Subsequent analyses compared only intervention and control sub-
jects who had a licence Waller 1992 reported crashes by person-
days of exposure but did not explain how they could be sure that
each study intervention and control participant was an active riderand actually exposed to the risk of experiencing a crash
Potential confounders that are unequally distributed by study arm
at baseline need to be dealt with in the analysis Satten 1980
Adams 1985 and Davis 1997 did not take into account any con-
founding factors either by design or in the analysis McDavid
1989 Waller 1992 and Billheimer 1998 accounted for potential
confounding factors at the design stage by matching the subjects
in the intervention and control groups Other studies (Raymond
1979 Mortimer 1984 Mortimer 1988 Perrino 2002 and also
Waller 1992) adjusted for one or two confounders by stratifica-
tionor standardisation (eg crashesmiles driven age group) Osga
1980 conducteda univariate analysis of crash rates in which no po-
tential confounders were accounted for but also calculated cumu-lative crash rates by mileage travelled Only Jonah 1982 Cooper
1988 and Savolainen 2007 conducted a multivariate analysis
Risk of bias in the case-control studies
A case-control design is suitable for the assessment of risk fac-
tors for outcomes such as motorcycle crashes injuries or deaths
but not appropriate for the evaluation of an intervention such
as motorcycle rider training It can potentially result in spurious
conclusions regarding the efficacy of the intervention because the
intervention has not been applied at random to an intervention
and control group Even with the control of possible confounders
using a case-control design to evaluate an intervention can lead tothese kinds of results since those who choose to undertake rider
training may differ on important prognostic factors that affect the
outcome compared to those who do not choose to participate in
rider training
Selection bias
A potential source of bias in case-control studies as with all other
study types where the intervention is not randomly allocated to
participants is that comparison groups are likely to differ on po-
tentially confounding characteristics This means that ideally all
cases need to be included in the study as well as a random sample
of controls taken from the population that gave rise to the cases If that is not possible cases and controls either need to be carefully
matched to make themcomparable or differences at baseline need
to be adjusted for in the analysis All four case-control studies were
population based Kraus 1975 included all fatalities and injured
cases leading to hospitalisation and appropriately recruited a ran-
dom sample of controls from the same population Hurt 1981 did
not manage to recruit all or a random sample of cases Only half
of the notifications of crashes to the police were received in time
by the researchers the research team for unreported reasons was
only able to respond to half of those again Data collection was
therefore completed for only about a fifth of all cases Neverthe-
less this was still a large sample of 900 cases Haworth 1997 using
16Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 1950
a similar study design was more likely to have included all cases
but this is not entirely clear from the study reported It is unclearfor example whether fatalities were included as they did not use
coronerrsquos reports for the recruitment of cases The MAIDS 2004
study was conducted in five European countries but it was unclear
how cases were recruited whether they were all cases during the
two years that the study lasted or a random sample of all cases
what the response rates were etc Controls were recruited from
randomly selected petrol stations but it was likewise not clear what
the eligibility criteria for inclusion in the study were how they
were recruited (eg randomly) and what the response rate was
Hurt 1981 and Haworth 1997 both matched controls to cases by
recruiting themat the sametime ofdayandsamedayof weekas the
crash occurred Hurt 1981 however recruited controls two years
later andit is likelythat there would have beendifferences betweenthe two comparison groups One important difference was that
during data collection for cases there was very little specialised
motorcycletrainingavailable in the studyregionwhereas two years
later when control data were collected there was some training
available
These four case-control studies also suffered from selection bias
due to potential differences between those that gave consent to
participate in the study and responded to interviews andquestion-
naires and those that did not None of the studies have assessed
such differences although Hurt 1981 did acknowledge that con-
trols giving consent to cooperate may have differed from those
motorcyclists who did not
Detection bias
Measures that can be taken to reduce detection bias in case-con-
trol studies are blinding of interviewers for case or control status
blinding of the participant for the study factor of interest validat-
ing measurement instruments and using these in an identical way
for cases and controls and cross checking self-reported informa-
tion with other more objective sources such as registers None of
the studies reported blinding of the intervention assessors to case
or control status but all used consistent measurement methods for
both case and control groups
All case-control studies used self-report as their main measure toassess whether cases and controls had received rider training and
if so the type of training The major potential source of bias in
case-control studies when self-report is used as a measure is recall
bias One way of reducing recall bias is by double-checking self-
reported data with more objective data Kraus 1975 also used po-
lice accident reports for information on rider characteristics but
such information was not available about rider training Haworth
1997 and MAIDS 2004 did not report on any measures to limit
recall bias Hurt 1981 reported further questioning of cases by
the interviewer when answers did not make sense or when par-
ticipants seemed to contradict themselves as part of the in-depth
unstructured interview technique This qualitative research tech-
nique can result in valuable information but as interviewers were
not blinded to case or control status of the participants it may have led to more thorough examination of cases compared to the
controls and thus differential detection bias
Intervention integrity
The level of detail about the rider training provided in the case-
control studies differed significantly but was minimal for each
Kraus 1975 simply asked whether lsquoanyrsquo motorcycle rider training
had lsquoeverrsquo been carried out Hurt 1981 asked participants whether
they hadbeen self taught informally taught or completed formally
taught rider training courses MAIDS 2004 asked whether partic-
ipants had had pre-license training additional (non-compulsory)
training other training or no training whereas Haworth 1997asked about the number of courses that a participant had com-
pleted whetherthey were basic intermediate or advanced courses
and how long ago the training had been done None of the case-
control studies reported collecting information about further de-
tails about any training programmes (eg type of training cur-
riculum duration of training)
Analysis
Kraus 1975 accounted for confounding factors through the exclu-
sion of females and stratification by age and frequency of motor-
cycle use Hurt 1981 matched cases and controls for time of day
and day of the week that the crash occurred and did not adjust forany confounding factors in the analysis Haworth 1997 matched
groups for time of day and day of the week that the crash occurred
and stratified by age blood alcohol level and licence status in the
analysis None of the studies used multivariate methods MAIDS
2004 did not provide any information about any participant char-
acteristics other than the study exposures of interest and did not
describe any methods to adjust for potential confounding factors
Effects of interventions
We have presented the results in a descriptive manner due to the
substantial heterogeneity in study types interventions and out-comes and the generally poor quality of the studies We have
grouped the interventions by mandatory and non-mandatory pre-
licence rider training and post-licence rider training and further
by study type and by outcome (crashes injurious and fatal crashes
and offences) We discuss results that are reported as statistically
significant (whether a P value was reported or not)
Pre-licence rider training - mandatory
Randomised trials
17Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2050
Crashes
Total group of participants
Anderson 1980 found a reduction in motorcycle crash rates in
the intervention group (MOST skill test with remedial training)
compared to the control group (who received the skill testwithout
a training component) The percent differenceranged from 143
six months after the programme (P lt 005) to 211 (P lt 0001)
after one year and 127 (P lt 001) after two years Hall 1988
found no statistically significant difference in crash rates between
intervention groups at any point in time
Subgroup of participants who obtained their licence only
Anderson 1980 found a reduction in motorcycle crash rates in the
intervention arm but this was statistically significant at one year
only (24 to 218 reduction P lt 005) Hall 1988 found no
statistically significant differences
Offences
Total group of participants
The percentage of offences in Anderson 1980 after six months
and one and two years was slightly higher in the intervention
arm compared to the control group but these differences were
not statistically significant Likewise the study by Hall 1988 did
not find a statistically significant difference in offences between
groups
Subgroup of participants who obtained their licence only
Hall 1988 found no significant difference in offence rates between
the intervention and control groups Anderson 1980 found thatthe percentage of offences increased in the intervention arms com-
pared to the control group This difference was statistically signif-
icant after two years (51 to 115 increase P lt 005)
Cohort studies
Crashes
Neitherof the cohort studies (Kloeden1994 and Billheimer1998)
found a statistically significant difference in crash rates between
the intervention and control groups
Offences
Billheimer 1998 (for the California Motorcyclist Safety Program
course) and Perrino 2002 did not find a statistically significant
difference in offence rates between the intervention and control
groups
Case-control studies
Crashes
MAIDS 2004 found that a slightly higher percentage of controls
(not involved in a crash) had had pre-license rider training than
cases (involved in a crash) This difference was not statistically significant (50 of controls vs 477 of cases OR 091 95 CI
076 to 110)
Pre-licence rider training - non-mandatory
Randomised trial
Injuries and deaths
Swaddiwudhipong 1998 found a significantly lower injury ratein the intervention villages (1051000 population 95 CI 92
to 120) compared to the control villages (1691000 population
95 CI 152 to 185) one year after the intervention This differ-
ence was not maintained two years after the intervention (127
95 CI 112 to 143 intervention 151 95 CI 135 to 167
control) There was no significant difference in death rates be-
tween groups
Cohort studies
Crashes
Cooper 1988 foundno statistically significant differencein crashes
between the intervention and control groups McDavid 1989
found a lower crash rate in the intervention group (14139) com-
pared to the control group (23139) but no test of statistical sig-
nificance was undertaken
Injuries and deaths
McDavid 1989 found that the intervention group had fewer in-
jurious crashes (814 crashes) compared to the control group (16
23 crashes) but no statistical significance test was reported
18Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2150
Offences
Cooper 1988 foundno statistically significantdifference in offencerates between the intervention and control groups McDavid 1989
foundno statistically significantdifference in offence ratesbetween
the intervention and control groups
Non-mandatory rider training
Non-randomised trials
Crashes
Neither Adams 1985 nor Leung 1987 found a statistically signif-icant effect of rider training on crash rates
Offences
Leung 1987 found a statistically significant lower offence rate in
the intervention group compared to the control group for both
males and females and for all age groups except for women 35
years or older (P values not reported)
Cohort studies
Crashes
Osga 1980 in the univariate analysis reported a lower percentage
of crashesin theintervention group (that continued to ride) (28
168595) than in the control group (34 112329) In the cu-
mulative crash rate analysis however Osga1980 found statistically
significant (at the P lt 001 level) higher crash rates in the interven-
tion group than in the control group at allmileage intervals (1000
to 29999 miles) travelled After riding more than 30000 miles
(when over 50 of the trained riders had been involved in a crash
compared to approximately 40 in untrained riders) this differ-
ence was no longer statistically significant Satten 1980 found a
lower percentage of crashes in the intervention group (13 969)
compared to the control group (25 2079) but the differencewas not assessed statistically Davis 1997 found a lower crash rate
in trained riders but it was unclear how this was calculated be-
cause the size of the control group was not reported Jonah 1982
and Billheimer 1998 (for the Experienced Rider Course) found no
statistically significant difference in crashes between the interven-
tion and control groups Although the difference between those
who completed the Experienced Rider Course and no course in
Billheimer 1998 neared significance at six months after the course
(P = 007) large P values were observed for one year after (P =
030) Raymond 1979 found a statistically significant higher crash
rate for the intervention group (1861000 miles ridden) com-
pared to the control group (1181000 miles ridden) in both males
and females (P lt 0001) especially during the first 5000 miles of
riding Waller 1992 found a statistically significant higher crashrate for the intervention group (27 crashes in three years or 070
crashes10000 person days) compared to the control group (5
crashes in three years or 013 crashes10000 person days) with a
crude relative risk of 538 (X 2=1513 plt005) Mortimer 1984
and Mortimer 1988 found no statistically significant difference in
crashes between the intervention and control groups
Savolainen 2007 found that the intervention group (those who
completed the Basic Rider Course) was 44 more likely to be
involved in crashes than the control group and those who com-
pleted the course multiple times were an additional 180 more
likely to be involved in a crash This finding was based on low
numbers and no P value or confidence level was given for either
result
Injuries and deaths
Satten 1980 also found a lower percentage of injurious crashes
(33 39) in the intervention group compared to the control
group(60 1220) although again no statistical significance test
was reported
Offences
Jonah 1982 did find a significantly lower self-reportedoffence rate
in the intervention group (15) compared to the control group(34 P lt 001) (no numbers given) Satten 1980 also found
a lower self-reported offence rate in the intervention group (1
71 versus 19100) but again this was not tested for statistical
significance Mortimer 1984 Mortimer1988 and Billheimer 1998
found no statistically significant differencein offence ratesbetween
the intervention and control groups
Case-control studies
CrashesMAIDS 2004 found that a slightly higher percentage of cases
(involved in a crash) had had additional non-mandatory rider
training than controls (not involved in a crash) This difference
was not statistically significant (17 of cases vs 12 of controls
OR 146 95 CI 064 to 337)
Rider training unspecified
Case-control studies
19Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2250
Injuries and deaths
Hurt 1981 found that the odds of injurious crashes were lower inthose who had received professional training compared to those
who had not Kraus 1975 found that in the younger age group
(less than 20 years) and the older age group (over 35 years) a larger
percentage of injured riders (12 and 21 respectively) reported
prior motorcycle rider training compared with the uninjured con-
trol group (6 and 5 respectively) (P lt 001) (no numbers
given) Haworth 1997 reported several different outcomes one
beingthat having completed at leastone trainingcoursecompared
to no training courses at all had no significant effect on the odds
of crashing (afteradjustment for age blood alcohol level or licence
status (learner permitprobationaryfull)) When comparing three
different levels of training courses (beginner intermediate and ad-
vanced) with no training no significant differences were found
in the odds of crashing However after adjustment for age and
blood alcohol level completing an advanced course as compared
to a beginners course was associated with a significant decrease in
the odds of severe motorcycle crashes resulting in serious injury or
death (OR 02 95 CI 01 to 08)
D I S C U S S I O N
The objective of this review was to quantify the effectiveness of
pre- andpost-licence motorcycle rider training on the reduction of
traffic crash involvement injuries deaths and offences committedby motorcycle riders Most included studies were cohort studies
There were a few randomised and non-randomised trials and case-
control studies
All studies compared either mandatory or non-mandatory formal
training with no formal training except for Haworth 1997rsquos case-
control study that compared levels of training (beginner interme-
diate and advanced) to each other
We found studies that evaluated the effectiveness of manda-
tory pre-licence training in reducing crash and offence rates
(Anderson 1980 Hall 1988 Kloeden 1994 Billheimer 1998
(the basic California Motorcyclist Safety Program component of
the study) Perrino 2002) non-mandatory training that when
passed resulted in a licence (Cooper 1988 McDavid 1989
Swaddiwudhipong 1998) non-mandatory training that could be
taken by motorcycle riders with or without a licence (Raymond
1979 Osga 1980 Satten 1980 Jonah 1982 Mortimer 1984
Adams 1985 Leung 1987 Mortimer 1988 Waller 1992 Davis
1997 Billheimer 1998 (the Experienced Rider Course compo-
nent of the study)) and unspecified rider training on crash injury
and death rates (Kraus 1975 Hurt 1981 Haworth 1997 MAIDS
2004)
The studies provided conflicting evidence regarding motorcycle
rider training which is likely due to the poor design quality and
reporting of the studies Most of the studies included in the review
were likely to be adversely affected by significant selection biasalong with substantial detection bias(measurement and recall bias)
which limits our ability to make conclusive comments
Mandatory pre-licence rider training
There were two RCTs three cohort studies and one case-control
study that examined pre-licence ridertraining The two RCTs eval-
uated the Motorcycle Skills Operating Test (MOST) (Anderson
1980 Hall 1988) Anderson 1980 found a statistically significant
reduction in crashes in the intervention group at six months one
and twoyearsfollow up However when only participantswho ob-
tainedtheirlicence were included in theanalysis thereduction wasstatistically significant only at one year follow up The other RCT
(Hall 1988) found no statistically significant differences in crash
rates at any point of time neither in the total population group
nor in the licensed subgroup Neither of the two cohort stud-
ies assessing crash rates found a statistically significant difference
in crash rates between intervention and control groups ( Kloeden
1994 Billheimer1998) andthe case-control study (MAIDS 2004)
also did not find a statistically significant difference in the odds of
experiencing a crash between trained and untrained riders With
regards to offences neither of the RCTs found a statistically sig-
nificant difference in offence rate between the intervention and
control groups in the analysis of the total group of participants
entering the trial However Anderson 1980 found a statistically significant increase in offence rates in the intervention group but
only after two years of follow up andonly in the analysis of partici-
pants who obtained their licence The two cohort studies assessing
offence rates found no statistically significant difference between
the groups (Billheimer 1998 Perrino 2002)
None of the five studies (RCTs or cohort) found a reduction in
crashes or offences following mandatory pre-licence rider training
except for Anderson 1980 who observed a statistically significant
reduction in crashes in the intervention group at one year in both
analyses It is not clear why both RCTs that evaluated the same
MOST and training intervention observed different results Both
studies enrolled large samples and observed a similar uptake of
licensure following the intervention It is possible that differencesmay have arisen due to different study samples (California ver-
sus New York) or that the intervention delivery differed between
studies
In both trials of the Motorcycle Skills Operating Test including
rider training only a third of motorcycle licence applicants ob-
tained their motorcycle licence It is possible that a licensing pro-
cess including a skills test may reduce crashes through decreased
exposure
Non-mandatory pre-licence training
20Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2350
Three studies assessed the effect of non-mandatory pre-licence
training The Cooper 1988 cohort study found no difference incrash or offence rates between the trained and untrained groups
We cannot however be confident about this result as the study
sufferedfrom selectionbiaswhichcouldhave biasedthe measure of
effect toward the null The RCT conducted by Swaddiwudhipong
1998 examined injury and death rates and found a statistically
significant lower injury rate in the intervention villages after one
year of follow up but no longer at two year follow up There
was no statistically significant difference in death rates between
intervention and control villages Our confidence in the results of
this study are hindered by the inadequacy of the randomisation
process that led to significant differences between the intervention
and control districts on two important variables The intervention
districts were more likely than the control districts to have a validlicence andto wear a helmetSincethese differences were nottaken
into account in later analyses they could explain the differences
observed in injury rates between districts Furthermore with only
six districts randomised the sample size was probably too small
to detect a difference between intervention and control groups
had there been one The McDavid 1989 cohort study found a
lower percentage of injuries in the intervention group although
a statistical test was not conducted McDavid 1989 drew their
intervention and control groups from different populations but
did adjust for confounders although they only analysed 40 of
the initial sample (139 trained and 139 untrained) Because of
these substantial weaknesses we are unable to draw conclusions
from this studyAn overall effect of pre-licence training on the reductionof crashes
injuries deaths and offences was not found although one should
be aware that the research of this issue to date has had significant
flaws However a mandatory skill test (such as the MOST) in
combination with and even without pre-licence training as part
of a licensing process seems to reduce the number of individuals
completing the licensing procedure and therefore indirectly re-
duces crash rates through a reduction in exposure to motorcycle
riding Making licensure a more difficult process may therefore be
seen as advantageous if it delays licensure and therefore reduces
exposure It is possible that the group who chose not to continue
riding a motorcycle would not have been the group to experience a
crash or commit an offence if they had continued to ride anyhowfor example because they are more safety conscious and this was
the reason for their giving up riding in the first place If this is the
case deterring individuals from completing their licence and rid-
ing may not necessarily decrease overall crashes deaths or offence
rates In addition other potentially adverse effects of complicat-
ing a licensing procedure need to be considered such as increases
in unlicensed riding In Thailand non-mandatory rider training
offered as a community intervention may have decreased injury
rates but this needs to be confirmed through further research
Non-mandatory training
The majority of the studies assessed non-mandatory training
(training independent of any licensing procedure) and the results
were varied The two non-randomised trials assessing crash rates found no statistically significant difference in crash rates between
the intervention and control groups (Adams 1985 Leung 1987)
Of the nine cohort studies Satten 1980 found a lower percentage
of crashesin the intervention group although it is unclear whether
this was a statistically significant result Davis 1997 also found a
lower percentage of crashesin theinterventiongroup although it is
unclear with which control group the intervention group wascom-
pared Raymond 1979 Osga 1980 Waller 1992 and Savolainen
2007 found a statistically significant higher crash rate in the in-
tervention group and the remaining cohort studies (Jonah 1982
Mortimer 1984 Mortimer 1988 Billheimer 1998) found no sta-
tistically significant difference in crash rates between intervention
and control groups No conclusions can be drawn from these stud-
ies with confidence The Satten 1980 and Davis 1997 studies did
not adjust for any confounding factors even though important
differences existed between the intervention and control groups
Although Raymond 1979 and Waller 1992 adjusted for a num-
ber of confounders at baseline (Waller 1992) and in the analy-
sis (Raymond 1979 Waller 1992) the intervention and control
groups were drawn from such different populations that it is pos-
sible that unmeasured confounders were responsible for the sta-
tistically significant increased risk of crashes observed in the inter-
vention group The observed difference could also be attributed totheobserved differential loss to followup (Raymond 1979) which
could bias the measure of effect towards or away from the null
The Osga 1980 study found a protective effect of rider training
in the univariate analysis but statistically significant higher crash
rates in the intervention group when accounting for riding expo-
sure Sex (a significantly higher proportion of females in the in-
tervention group) and other potentially confounding factors that
could explain thedifferencebetween the groups were howeverun-
accounted for Finally Savolainen 2007 did control for potential
confounding variables yet it is not clear whether variables were
controlled for simultaneously or one at a time One case-control
study (MAIDS 2004) found no statistically significant reduction
in the odds of experiencing a crash between trained and untrainedriders but the quality of this study was poor mainly because the
analysis did not account for any confounding factors
One cohort study (Satten 1980) assessed the effect of non-manda-
tory rider training on injurious crashes and found a lower percent-
age of injuries in the intervention group although a statistical test
was not conducted As mentioned above the Satten 1980 studyrsquos
validity is reduced due to its lack of adjustment for confounding
factors
The only non-randomised trial that assessed offence rates (Leung
1987) found a statistically significant lower offence rate in the in-
tervention group and of the five cohort studies that assessed of-
fence rates (Satten 1980 Jonah 1982 Mortimer 1984 Mortimer
21Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2450
1988) one study (Jonah 1982) also found a statistically significant
lower offence rate in the intervention group However we have lit-tle confidence in the results from Leung 1987 since potential con-
founding factors were not appropriately addressed Jonah 1982
did conduct a multivariate analysis but was hindered by an initial
low response rate The other four cohort studies did control for
several potential confounders and found no difference in offence
rates following non-mandatory post-licence rider training
The non-randomised trials and cohort studies evaluating non-
mandatory rider training were at high risk of substantial selection
bias due to the non-random allocation of participants as well
as low participation rates and attrition bias and could therefore
not provide conclusive evidence with regard to the effect of the
training It should be noted that despite the biases and other issues
with this group of studies most didnot detectmajor effects relatedto rider training in either direction This may suggest that any
potential impact of rider training will be modest
Studies examining unspecified modes of ridertraining
Three case-control studies assessed whether injured or deceased
motorcyclists were more or less likely to have had rider training
compared to control motorcyclists All had conflicting results
Hurt1981 foundthe odds of havingan injuriouscrash waslower in
those who had received rider training whereas Kraus 1975 found
that the odds of having an injurious crash were higher in youngerand older trained riders Haworth 1997 who compared advanced
versus beginner training found no effect on motorcycle injuries
or deaths although they did find that those who had completed
advanced training had reduced chances of crashing compared to
those who had done a beginnersrsquo training course Similar issues
with uncontrolled confounding and biases of selection as described
above arose with the three case-control studies Hurt 1981 suffers
from inadequate control of confounding via poor matching and
both the Kraus 1975 and the Haworth 1997 findings were only
controlled for limited factors by means of stratified analysis It is
likely that these limitations are the reason for the contradictory
directions of effect observed for the three case-control studies
Methodological weaknesses of the studies
In summary most studies investigating the effectiveness of mo-
torcycle rider training on reducing crashes injuries and death suf-
fered from serious methodological weaknesses
The majority of studies were non-randomised and were there-
fore limited mainly by selection bias Although the approaches of
matching stratification or regression analyses can minimise bias
(Fleiss 2004) they were poorly employed in the reviewed studies
The inadequate adjustment for confounders in particular adds
to our lack of confidence in the studiesrsquo results since it has been
demonstrated that individuals who choose to take motorcycle rider
training differ from those not taking rider training For instance
trained riders often differ from controls in age sex and experi-ence usually ride less and may have a more positive attitude to
safety as indicated by the reported more frequent wearing of seat
belts (Mortimer 1982)
The two large randomised trials reported that a large proportion
of their study group were never exposed to the risk of offences or
crashes because they never attained licensure This made it impos-
sible to assess whether a reduction in crashes in the intervention
group was a result of training or due to delayed or never attained
licensure
In some studies the sample sizes were too small to detect any sta-
tistically significant differences Follow-up time after training dif-
fered between studies and was an impediment to observing any
real effects of training if they existed As riding experience ac-cumulates over time any differences in outcomes between those
trained and untrained would diminish in studies with lengthy fol-
low-up times This dilution of any possible effect of rider training
may have also occurred in studies that mixed both novice and ex-
perienced riders in the study groups since it is hypothesised that
novice riders serve to benefit most from rider training This would
increase the difficulty in finding any effect of rider training par-
ticularly when sample sizes are small
Detection bias may also account for some of the differences ob-
served between trained and untrained riders Detection bias would
have occurred if the self-report hospital driver or police records
used to record the outcomes of crash injury death or offences
differed between the trained and untrained riders Research hasdemonstrated that not all crashes are reported to police for ex-
ample small crashes or off-road crashes (Kraus 1975) even serious
injury crashes have been found to be underreported (Anderson
1980) However this would only introduce detection bias if the
reporting of outcome differs for trained and untrained riders This
does appear possible in that responsible riders may be more likely
to enroll in a rider training course and also may be more likely
to report crashes to police than irresponsible riders meaning that
the number of crashes from riskier riders could be underrepre-
sented on records An analysis to investigate the effectiveness of
rider training may then show that rider training is not effective in
reducing crashes
Rather than using reported outcomes some studies used self-re-ported survey data although this technique suffers from non-
response biases In fact two studies reported differences in de-
mographic characteristics in response rates to questionnaires
(Anderson 1980 Hall 1988) Self-reported data also introduce
the risk of recall bias survival bias and social desirability response
bias Recall bias may have been equally distributed among trained
and untrained riders howeverit is hypothesised that trained riders
may have felt it was lsquoexpectedrsquo to report good driving histories
hence leading the rider training to appear effective
22Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2550
Limitations of this review
Limitations of this review were that we found very few recentstudies (the two most recent ones published in 2002 and 2007)
and that we found very few studies conducted in low-and middle
income countries even though our search strategy was set up to
identify such studies Anotherlimitation wasthat forthe reviewed
studies we could not assess all outcomes of rider training that are
potentially useful Some of the included studies assessed outcomes
such as use of safety clothing including helmets Such outcomes
are valuable since the use of motorcycle helmets has been demon-
strated to be effective in preventing head injuries (Liu 2004) De-
spite this the use of safety clothing was not included as an out-
come in this review because the studies measuring this outcome
were limited by poor data quality due to very low response rates
Another outcome of interest in assessing the effectiveness of rider
training is lsquotype of injuryrsquo As Haworth 2005 state in their review
of motorcycle licensing and training rider training could be con-
sidered effective if injuries become less severe because motorcycle
riders have learned how to avoid more serious injuries through
rider training The studies included in this review did not specify
type of injury and so this outcome could not be assessed Last
changes in knowledge and attitudes following rider training are
other potentially useful outcomes We have excluded such studies
(Osga 1979 Clayton 1990 Goldenbeld 2004) as these outcomes
are proxies rather than direct measures of the effectiveness of mo-
torcycle rider training We opted to report on the outcomes of
crashes injury death and offences alone since there is little ev-
idence about how changes in knowledge and attitudes relate to
these outcomes
A U T H O R S rsquo C O N C L U S I O N S
Implications for practice
Due to the poor quality of studies included in this review we were
unable to draw any conclusions about the effectiveness of rider
training on crash rates injurious or fatal or offence rates The
findings do suggest that mandatory pre-licence training may be
an impediment to completing a motorcycle licensing process thuspossibly indirectly reducing crash injury death and offence rates
through a reduction in exposure to riding a motorcycle On the
basis of the existing evidence it is not clear if and what type of
training reduces the risk of crashes injuries deaths or offences in
motorcyclists and the selection of the best rider training practice
can therefore not be recommended The suggestion that manda-
tory pre-licence training indirectly reduces exposure to riding the
motorcycle because trainees are impeded from completing the li-
censing process does not help those motorcyclists who persist in
obtaining their licence and riding the motorcycle Some type of
rider training is likely necessary to teach motorcyclists to ride a
motorcycle safely Good policy is based on sound evidence and
evaluation but such evidence for motorcycle rider training is seri-
ously lacking Policy makers are therefore encouraged to include arigorous evaluation component to any new or existing rider train-
ing programme Such an evaluation should ideally be randomized
and assess whether the training programme is effective and if so
which component(s) of the programme work best
Implications for research
Given theresults of this review which found littlereliable evidence
about the effectiveness of motorcycle rider training it is necessary
to build up an evidence base of the effectiveness of motorcycle
rider training in the reduction of crash injury death and offence
rates To this end well designed randomised controlled trials need
to be conducted that examine the impact of rider training oncrashes and offences A randomised controlled trial is the preferred
method as it would significantly reduce most biases described in
this review To minimise the impact of participants delaying or
never attaining their licence and therefore exposure as found in
the randomised trials in this review such a trial should include
riders who are already licensed and reliable measures of riding
exposure in a reasonably lengthy follow up period It is expected
that a randomised controlled trial of sufficient sample size would
observe equal balance on possible confounders by study arm how-
ever an assessment of potentially important confounders (such as
helmet wearing) by study group would enable control for these
variables in the analyses if necessary It is imperative that the effec-
tiveness of each component of the rider training (such as theorypractice knowledge skills and modification of behaviour) and the
synergy of such components are evaluated Outcomes should be
objective ideally those that are routinely collected rather than self-
reported outcomes Even though riding the motorcycle poses dif-
ferent challenges and incurs different risks lessons may be learned
from research into the effects of car driver training
Motorcycles are a common means of transport in low- and middle
income countries and usage is increasing As the burden of mo-
torcycle crash victims is mostly borne in such countries research
should be focused on the effectiveness of types of rider training on
the reduction of crashes injuries and fatalities in such countries
and be supported by higher income countries This review found
only one study conducted in a middle income country that sug-gested decreasing injury rates It would be beneficial to conduct
additional research to confirm this suggestion and to enhance our
knowledge of the role of rider training in low- and middle-income
settings
March 2006 saw the launch of the University of North Carolina
Highway Safety Research Centre partnership with the Motorcycle
Safety Foundation (MSF) and plans were revealed to conduct a
study of the potential benefits of motorcycle rider training The
study is funded by the MSF and the National Highway Traffic Sa-
fety Administration (NHTSA) Called rsquoThe Discovery Projectrsquo it
will compare data on crash rates from riders participating in a basic
23Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2650
motorcycle safety class with riders participating in a basic motor-
cycle safety class as well as a variety of continuing education classesThe collection of pilot data was due to start in March 2007 and
the data collection portion of the study is expected to take three
years (Highway Safety Research Center) It is vital that researchers
and policy makers collaborate to develop furtherhigh quality eval-
uations of motorcycle rider training programmes to help inform
development of new and effective training programmes
A C K N O W L E D G E M E N T S
Staff from The University of Sydney library for helping us locate
and retrieve full reports of many papers
R E F E R E N C E S
References to studies included in this review
Adams 1985 published data only
Adams A Collingwood V Job RFS Evaluation of the 1982 Australia
Post motorcycle rider training scheme NSW Australia Traffic
Accident Research Unit Traffic Authority NSW 1985 [ RN 85]
Anderson 1980 published data only
Anderson J Ford JL Peck RC Improved motorcyclist licensing and testing project Washington DC National Highway Traffic Safety
Administration (NHTSA) USA 19801ndash217 [
DOTndashHSndash805ndash598]
Billheimer 1998 published data only
Billheimer JW Evaluation of the California Motorcyclist Safety
Program Transportation Research Record 19981640100ndash9 [
TRB Report Number 980652]
Cooper 1988 published data only
Cooper PJ Rothe JP Motorcyclists who they are and why they do
what they do Transportation Research Record 1988116878ndash85
Davis 1997 published data only
David CF Evaluation of community traffic safety programs and
motorcycle operator training programs Connecticut Department of Transportation 199763
Hall 1988 published data only
Hall CJ Buchanan L Engle R Motorcycle rider education evaluationproject Washington DC USA National Highway Traffic Safety
Administration (NHTSA) US Department of Transportation
1988 [ DTNH 22ndash80ndashCndash05123]
Haworth 1997 published data only
Haworth N Smith R Brumen I Pronk N In Monash University
Accident Research Centre editor(s) Case-control study of motorcycle
crashes Victoria Australia Department of Transport and Regional
Development The Federal Office of Road Safety 19971ndash149 [
ISBN 0 642 25501 6 ]
Hurt 1981 published data only
Hurt H Ouellet J Thom D Motorcycle accident cause factors and
identification of countermeasures USA Department of
Transportation 1981 [ Technical Report DOT HS 806ndash862]
Jonah 1982 published data onlylowast Jonah BA Dawson N Bragg B Are formally trained motorcyclists
safer Accident Analysis amp Prevention 198214(4)247ndash55
Jonah BA Dawson NE Bragg BW Evaluation of the Canada
Safety Councilrsquos Motorcycle Training Program Proceedings of theInternational Motorcycle Safety Conference Washington DC
USA 198059ndash93
Kloeden 1994 published data only
Kloeden CN Moore VM McLean AJ Evaluation of the pre-license
training program for motorcyclists in South Australia Walkerville
South Australia Australia NHMRC Road Accident Research Unit
The University of Adelaide 19941ndash37 [ ISBN 0 7308 0607 3]
Kraus 1975 published data only
Kraus JF Riggins RS Franti CE Some epidemiological features of
motorcycle collision injuries American Journal of Epidemiology 1975102(1)74ndash98
Leung 1987 published data only
Leung KS Reding VA Evaluation of the Wisconsin motorcycle rider course Wisconsin Bureau of Policy Planning and Analysis
Wisconsin Department of Transportation 1987
MAIDS 2004 published and unpublished data
MAIDS In-depth investigations of accidents involving powered
two wheelers ACEM (Acemacemeu) 2004 Vol Final report
20179
McDavid 1989 published data only
McDavid JC Lohrmann BA Lohrmann G Does motorcycle
training reduce accidents Evidence from a longitudinal quasi-
experimental study Journal of Safety Research 198920(2)61ndash72
Mortimer 1984 published data only
Mortimer R Evaluation of the motorcycle rider course Accident
Analysis amp Prevention 198416(1)63ndash71
24Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2750
Mortimer 1988 published data only
Mortimer RG A further evaluation of the motorcycle rider courseJournal of Safety Research 198819(4)187ndash96
Osga 1980 published data only
Osga GA An investigation of the riding experiences of MSF rider
course participants Dissertation South Dakota USA The
University of South Dakota 19801ndash224
Perrino 2002 published data only
Perrino CS Ahmed A Callender A Rozier E Cantwell A Stewart
O The role of Marylandrsquos motorcycle rider course in promoting saver
behaviors and attitudes Baltimore USA National Transportation
Center Morgan State University 20021ndash52
Raymond 1979 published data only
Raymond S Tatum S An evaluation of the effectiveness of the RAC
ACU motorcycle training scheme - final report Road Safety ResearchUnit Dept of Civil Engineering University of Salford London
19791ndash131
Satten 1980 published data only
Satten RS Analysis and evaluation of the motorcycle rider courses
in thirteen Illinois counties International Motorcycle Safety
Conference Washington DC 1980145ndash93
Savolainen 2007 published and unpublished data
Savolainen Peter Mannering Fred Additional evidence on the
effectiveness of motorcycle training and motorcyclistsrsquo risk-taking
behaviour Transportation Research Board 2007 Vol March 20
19
Swaddiwudhipong 1998 published data only
Swaddiwudhipong W Boonmak C Nguntra P Mahasakpan P
Effect of motorcycle rider education on changes in risk behavioursand motorcycle-related injuries in rural Thailand Tropical Medicine
and International Health 19983(10)767ndash70
Waller 1992 published data only
Waller FL An analysis of the impact of the Texas Department of
Public Safety Motorcycle Operator Training Program upon
motocycle related traffic accidents Texas State University Public
Administration Program Applied Research Projects 199277
References to studies excluded from this review
Anderson 1980a published data only
Anderson LC Better biking a motorcycle rider improvement
program International Motorcycle Safety Conference Washington
DC USA Motorcycle Safety Foundation 1980231ndash53Aupetit 2008 published and unpublished data
Aupetit S Riff J Buttelli O Espie S Ergonomics research program
on motorcycle learning Contributions to the development of a
motorcycle simulator for training Transport Research Arena
Ljubljana 20089
Awane 1999 published data only
Awane T Integrating simulators in motorcycle safety education
Journal of International Association of Traffic and Safety Sciences 199923(1)26ndash35
Baldi 2005 published data only
Baldi S Baer JD Cook AL Identifying best practices states in
motorcycle rider education and licensing Journal of Safety Research200536(1)19ndash32
Batchler 1988 published data only
Batchler MW Evaluation of the pre-license motorcycle training course at the New South Wales traffic education centre Federal Office of
Road Safety (FORS) 19881ndash122 [ Report No CR 77 ]
Blanchard 2006 published data only
Blanchard HT Tabloski PA Motorcycle safety educating riders at
the teachable moment Mosby (CV) Company 200632330ndash2
Bonnett 2005 published data only
Bonnett J Mature age skills training for experienced motorcycle
riders Motorcycle ridersrsquo association of the Australian Capital
Territory 2005
Braver 2007 published data only
Braver ER et alPersuasion and licensure a randomized controlled
intervention trial to increase licensure rates among Maryland
motorcycle owners Traffic Injury Prevention 20078(1)39ndash46Clayton 1980 published data only
Clayton AB Sudlow DE An evaluation of the effectiveness of a one day training programme for learner motorcyclists Berkshire UK
Transport and Road Research Laboratory 19801ndash23 [ ISSN
0266ndash7045]lowast Clayton AB Sudlow DE An evaluation of the effectiveness of a one-day training programme for learner motorcyclists Berkshire UK
Road Safety Division Berkshire 19801ndash23 [ISSN 0266ndash7045]
Council 1978 published data only
Council FM Desper LP Dutt AK Driver education for motorcycle operation University of North Carolina Highway Safety Research
Center 19781ndash40
Crowther 2005 published data onlylowast Crowther G Engaging with motorcyclists UK police and theBikeSafe Road Safety Programme Published proceedings of the
Australasian Road Safety Research Policing Education Conference
Wellington New Zealand 2005 issue ISBN 0ndash473ndash10636ndash11ndash11
Daniello 2009 published data only
Daniello A Gabler HC Mehta YA Effectiveness of motorcycle
training and licensing Transportation Research Board 88th Annual
Meeting 200913
Dewanti 2007 published data only
Dewanti M Challenging efforts in promoting young motorcyclist
safety in Indonesia Road Safety on Four Continents 14th
International Conference Swedish National Road and Transport
Research Institute 2007
Du Plooy 1990 published data onlyDu Plooy DW Motorcycle training in South Africa The Human
Element International Motorcycle Safety Conference Orlando
Florida USA Motorcycle Safety Foundation 19904-1 to 4-9
Engel 1989 published data only
Engel U Krogsgaringrd Thomsen L [Faeligrdselsundervisning og
propaganda ndash en faeligrdselssikkerhedsfremmende foranstaltning]
Dansk Vejtidsskrift 198912295ndash7
Goldenbeld 2002 published data only
Goldenbeld C Houwing S de Craen S The development of
driving skills of young moped riders [De ontwikkeling van de
rijvaardigheid van jonge bromfietsers] SWOV Institute for Road
Safety Research (Stichting Wetenschappelijk Onderzoek
Verkeersveiligheid) Leidschendam the Netherlands 200263
25Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2850
Goldenbeld 2004 published data only
Goldenbeld C Twisk D de Craen S Short and long term effects of moped training a field experiment Transportation Research Part F 20047(1)1ndash16
Haworth 2000 published data only
Haworth N Smith R Kowadlo N Motorcycle training in Victoria
evaluation of rider training curriculum in Victoria Monash
University Accident Research Centre Reports 2000 Vol 165 issue
3131
Haworth 2005 published data only
Haworth N Mulvihill C Review of motorcycle licensing and
training Monash University Accident Research Centre 2005 [
Report no240]
Haworth 2006 published data only
Haworth N Mulvihill C Rowdon P Teaching old dogs new tricksTraining and older motorcyclists Journal of the Australasian College of Road Safety 200718(4)20ndash5lowast Haworth NL Mulvihill C Rowden PJ Teaching old dogs new
tricks Training and older motorcyclists Australasian Road Safety
Research Policing and Education Conference Gold Coast
Queensland 2006
Hill 1983 published data only
Hill D Motorcycle rider training testing and licensing Motor
Cycle Council of New South Wales Sydney 1983
Imatake 1980 published data only
Imatake Y On the activities in Japan of motorcycle safe riding
promotion committee (MSRPC) - especially the education of
unskilled riders of class 1 motorbikes International Motorcycle
Safety Conference Washington DC USA Motorcycle Safety
Foundation 1980255ndash68
Ingebritsen 1990 published and unpublished data
Ingebrigtsen S [Risikofaktorer ved ferdsel med moped og
motorsykkel] TOslashI-rapport 66 Oslo Transportoslashkonomisk
institutt 1990
Jain 2009 published and unpublished data
Jain A et alTwo wheeler accidents on Indian roads a study from
Mangalore India Elsevier 200916130ndash3
Jonah 1979 published data only
Jonah BA Dawson NE Validation of the motorcycle operator skill
test Accident Analysis amp Prevention 197911(3)163ndash71
Jonah 1981 published data onlyJonah BA Dawson NE Bragg WE Predicting accident
involvement with the Motorcycle Operator Skill Test Accident Analysis amp Prevention 198113(4)307ndash418
Kadar 1990 published data only
Kadar KJ Countermeasures motorcycle rider education delivery
system The Human Element International Motorcycle Safety
Conference Motorcycle Safety Foundation 19906-27 to 6-41
Kelsey 1986 published data only
Kelsey SL Liddicoat C Ratz M Licensing novice motorcyclists a
comparison of Californiarsquos standard test and the MOST II (Motorcycle
Operator Skill Test) administered at centralized testing offices Sacramento California USA California Department of Motor
Vehicles Research and Development Office 19861ndash74
Koch 1980 published data only
Koch H Advanced course for motorcycle riders InternationalMotorcycle Safety Conference Washington DC USA Motorcycle
Safety Foundation 1980269ndash81
Koch 1990 published data only
Koch H Brendicke R How to influence risk taking behaviour
within motorcycle rider training programs The Human Element
International Motorcycle Safety Conference Orlando Florida
USA Motorcycle Safety Foundation 1990Motorcycle Safety
Foundation 1990 p 9-1 to 9-25A
Labbett 2005 published data only
Labbett S Langham M Training can make the problem worse
Road Safety Congress 20051ndash12
Law 2005 published and unpublished data
Law TH Radin Umar RS Zulkaurnain S Kulanthayan S Impactof the effect of economic crisis and the targeted motorcycle safety
programme on motorcycle-related accidents injuries and fatalities
in Malaysia International Journal of Injury Control and Safety
Promotion 200512(1)9ndash21
Lowes 1990 published data only
Lowes BF In-traffic rider training The Human Element
International Motorcycle Safety Conference Orlando Florida
USA Motorcycle Safety Foundation 19907-46 to 7-57
Mahon 1976 published data only
Mahon J Teaching accident avoidance on the motorcycle Journal of Traffic Safety Education 1976231ndash29
Maxwell 1985 published data only
Maxwell GS Evaluation of a Program for Training MotorcycleRiders Evaluation and Program Planning 19858(4)339ndash47
Mayhew 1996 published data only
Mayhew DR Simpson SH Effectiveness and role of driver educationand training in a graduated licensing system Ottowa Canada The
Traffic Injury Research Foundation 19961ndash89 [ ISBN
0ndash920071ndash12ndash0]
McKnight 1987 published data only
McKnight AJ Milburn BG An evaluation of the modified
motorcycle operator skill test exercises NHTSA Report 1987 [
Report No DOT HS 807 212 ]
McPherson 1976 published data only
McPherson K McKnight AJ The development and evaluation of a
motorcycle skill test manual and knowledge test Washington DCUSA National Highway Safety Administration 1976
Mohan 2004 published data only
Mohan D Evidence-based interventions for road traffic injuries in
South Asia Journal of the College of Physicians and SurgeonsPakistan 200414(12)746ndash7
Morgan 1980 published data only
Morgan JK Advanced motorcycle training course content and
study evaluation International Motorcycle Safety Conference
Washington DC USA Motorcycle Safety Foundation 1980
283ndash312
Motoki 1990 published data only
Motoki M Yamazaki S A study on effective motorcycle rider
education The Human Element International Motorcycle Safety
26Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2950
Conference Orlando Florida USA Motorcycle Safety
Foundation 19909-26 to 9-57A
Nairn 1992 published data onlylowast Nairn RJ Motorcycle safety research literature review 1987 to
1991 Federal Office of Road Safety Australia 1993 Vol Report
CR 117
Newland 1980 published data only
Newland R Darby J Automotive practicespre motorcycle rider
program International Motorcycle Safety Conference Washington
DC USA Motorcycle Safety Foundation 1980127ndash9
Newland 1982 published data only
Newland R Motorcycle rider training and licensing requirements
Driver training steering a course for the future Melbourne
Victoria Transport Regulation Board 1982
Newland 1990 published data only
Newland R Motorcycle training and licensing in Victoria 1980-
1990 The Human Element International Motorcycle Safety
Conference Orlando Florida USA Motorcycle Safety
Foundation 1990
Ormston 2003 published data only
Ormston R Dudleston A Pearson S Stradling Evaluation of
Bikesafe Scotland Scottish Executive Social Research Edinburgh
2003
Osga 1979 published data only
Osga GS Ellingstad VS An evaluation of the effectiveness of the 1978
South Dakota motorcycle rider course program South Dakota USA
University of South Dakota 19791ndash89
Panou 2007 published data only
Panou M Bekiaris E A new concept on the integration of driving
simulators in driving training the train-all approach Road Safety
on Four Continents 14th International Conference Swedish
National Road and Transport Research Institute 2007
Peterson 1990 published data only
Peterson A Kadar K Status report 1980-1989 Minnesota rider
education statistics The Human Element International
Motorcycle Safety Conference Orlando Florida USA 199013-
113 to 13-129
Prem 1984 published data only
Prem H Good MC Motorcycle rider skills assessment ACT
Australia Office of Road Safety Department of Transport 1984
1ndash377 [ Report No CR 34 ]
Radin Umar 2006 published data only
Radin Umar RS Motorcycle safety programmes in Malaysia how
effective are they International Journal of Injury Control and Safety Promotion 200613(2)71ndash9
Rockwell 1990 published data only
Rockwell TH Kiger S Carnot M An evaluation of the Ohio
motorcyclist enrichment program Ohio Department of Highway
Safety 1990
Rothe 1987 published data only
Rothe JP A motorcycle riding course Motorcyclists Image and reality Canada Insurance Corporation of British Columbia 1987
59ndash83
Rowden 2007 published data only
Rowden P Watson B Haworth N What can riders tell us aboutmotorcycle training A view from the other side of the fence
Australasian Road Safety Research Policing and Education
Conference Melbourne Australia Melbourne 2007
RSC Victoria 1993 published data only
Road Safety Committee Inquiry into Motorcycle Safety in Victoria
Parliament of Victoria 1993
Russam 1979 published data only
Russam K Motorcycle training and licensing in Japan UK Transport
and Road Research Laboratory (TRRL) 1979 [ Report 916 ]
Saffron 1981 published data only
Saffron DG Driver and rider education training and licensing a
brief review Traffic Accident Research Unit Traffic Authority of
New South Wales NSW Australia Traffic Accident Research UnitTraffic Authority of New South Wales 1981
Schulz 1990 published data only
Schulz U Motorcycle accidents among learner riders Motorcycle
licensing Evaluation of the effectiveness of a Graduated Licensing Scheme Forshungshefte des Institut fur Zweiradsicherheit 1990
Sexton 2004 published data only (unpublished sought but not used)
Sexton B Baughan C Elliott M Maycock G The accident risk of
motorcyclists 2004TRL Report TRL 607(ISSN 09684107)58
Shephard 1986 published data only
Shepherd R Do Motorcycle Safety Foundation programs work
Driver 198619(9)16ndash8
Simpson 1990 published data onlySimpson HM Mayhew DR The promotion of motorcycle safety
training education and awareness Health Education Research
Theory and Practice 19905(2)257ndash64
Sokolov 1990 published data only
Sokolov O Safe motorcycle riding some aspects of motorcycle
riding in the USSR The Human Element International
Motorcycle Safety Conference Motorcycle Safety Foundation
19904ndash37 - 4-48
Steffens 1988 published data only
Steffens U Gawatz R Willmes G Effectiveness of mofa courses
[Wirksamkeit von Mofakursen] Unfall- und sicherheitsforschung
strasenverkehr 198867168
Sudlow 2003 published data only
Sudlow D Scoping study on motorcycle training Road Safety
Research Report London UK The British Institute of Traffic
Education Research Department for Transport 2003 Vol June
[ Report No 36]
Tasmania 1995 published data only
Mannering FL Grodsky LL Motorcycle rider training scheme
Traffic Education Research Centre Armidale Australia Driver
Training Australia Limited 1995 Vol April1ndash56
Tetard 1985 published data only
Tetard C Motorcycle ridersrsquo training to detect indices in an
intersection situation [Lrsquoentrainement a la prise drsquoinformation dans
lrsquoapprentissage de la conduite motocycliste] Le Travail Humain198548(2)147ndash59
27Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3050
Thackray 1980 published data only
Thackray RM Prescott JC Field test of the motorcycle ridercourse International Motorcycle Safety Conference Washington
DC USA Motorcycle Safety Foundation 1980195ndash7
Thompson 1994 published data only
Thompson M Evaluation of compulsory basic training for
motorcyclists Transport Research Laboratory Crowthorne
Berkshire UK 1994 [ Report 63]
Tobler 1990 published data only
Tobler U Individual driving instruction for motorcycle driving
learners the Swiss experiment of motorcycle driving school The
Human Element International Motorcycle Safety Conference
Orlando Florida USA Motorcycle Safety Foundation 1990
Transport 1989 published data only
Transport Communications Road Crash Statistics Australia Bureauof Transport and Communications Economics 198914ndash20
Tuan 2007 published data only
Tuan VA Shimizu T Towards development and evaluation of the
motorcycle drivers re-education program in Vietnam modeling of
motorcycle driverrsquos undesired behaviors 11th World Conference
on Transport Research 200725
Victor 1980 published data only
Victor DJ Traffic safety education for motorcyclists International
Motorcycle Safety Conference Washington DC USA Motorcycle
Safety Foundation 1980221ndash8
Waterer 1990 published data only
Waterer TH Compulsory training for motorcyclists from liberty
to compulsion The Human Element International Motorcycle
Safety Conference Orlando Florida USA Motorcycle Safety
Foundation 19904-49 to 4-57
Weller 1989 published data only
Weller RB Chandler EW Motorcycle safety and motorcycle
education past research and survey results Journal of Public Policy
amp Marketing 1989893ndash108
Wells 1981 published data only
Wells P Observations of some RACACU motorcycle training
schemes Transport and Road Research Laboratory Crowthorne
Berkshire UK 1981
Winn 1986 published data only
Winn G Two important trends in motorcycle safety regulations
rider education and conspicuity improvement American
Motorcyclist Association 1986147ndash52
Wisher 1988 published data only
Wisher G Reid J Motorcycle accidents in Norfolk a three-year study of motorcycle training and the Part 1 test Norfolk UK Road Safety
Division Highways Department Norfolk County Council 1988
Wood 1987 published data only
Wood HT Bowen R Evaluation of the revised learner permit scheme
July 1983 to December 1985 Victoria Australia Road Traffic
Authority 19871ndash27 [ ISBN 0 7306 0247 8]
Yamazaki 1990 published data only
Yamazaki S Nakanishi M Survey on effect of Honda rider
education The Human Element International Motorcycle Safety
Conference Orlando Florida USA 199012-167 to 12-187
References to ongoing studies
Burgess 2010 unpublished data only
Burgess C Broughton P Fylan F Stradling Interim evaluation of
the UKs National Ride program In L Dorn editor(s) Driver
behaviour and Training Vol IV Aldershot Ashgate Publishing
Limited 2010
Additional references
Anderson 1980
Anderson J Ford JL Peck RC Improved motorcyclist licensing and
testing project Volume I and II National Highway Traffic Safety
Administration US department of Transportation Washington
DC 1980
Bjornstig 1985
Bjornstig U Bylund P Lekander T Brorsson B Motorcycle fatalitiesin Sweden Acta Chirurgica Scandinavica 1985151(7)577ndash81
Clayton 1990
Clayton A Sudlow DE An evaluation of the effectiveness of a one-
day training program for learner motorcyclists The Human
Element International Motorcycle Safety Conference Orlando
Florida USA 199013-105 to 13-111
Cochrane Collaboration 2008
Higgins JP Green S Cochrane Handbook for Systematic Reviews of
Interventions 500 The Cochrane Collaboration 2008
Elliot 2003
Elliott MA Baughan CJT Broughton J Grayson G Knowles J et
alMotorcycle Safety a Scoping Study TRL [ Report No
0968ndash4107]Fleiss 2004
Fleiss JL Levin B Paik MC Statistical methods for rates and proportions 3rd Edition John Wiley amp Sons Inc 2004 [ ISBN
978ndash0ndash471ndash52629ndash2 ]
FORS 1999
FORS Federal Office of Road Safety Road risk for sober licensed
motorcyclists Vol Monograph 27 Federal Office of Road Safety
19991ndash2
Haworth 2005
Haworth N Mulvihill C Review of motorcycle licensing and training
Victoria Australia Monash University Accident Research Centre
20051ndash83 [ Report No 240]
Huwiler-Muumlntener 2002Huwiler-Muumlntener K Juumlni P Junker C Egger M Quality of
Reporting of Randomized Trials as a Measure of Methodologic
Quality JAMA 20022872801ndash4
Johnston 2008
Johnston P Brooks C Savage H Fatal and serious road crashes
involving motorcyclists Road Safety Research and Analysis Report
ATSB Monograph 20 Canberra Department of Infrastructure
Transport Regional Development and Local Government 0205
081ndash26
Juumlni 2001
Juumlni P Altman DG Egger M Systematic reviews in health care
Assessing the quality of controlled clinical trials BMJ 2001323
42ndash6
28Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3150
Kopits 2003
Kopits E Cropper M Traffic fatalities and economic growth TheWorld Bank Development Research Group Infrastructure and
Environment Policy research working paper 3035 2003
Leung 1983
Leung P Chau BTW Motorcycle accidents in Hong Kong pattern
of injury treatment causative factors and socio-economic
implications Injury 198315(1)23ndash9
Liu 2004
Liu B Ivers R Norton R Blows S Lo SK Helmets for preventing
injury in motorcycle riders Cochrane Database of Systematic Reviews 2004 Issue 2 [DOI 10100214651858CD004333pub3]
Mohan 2002
Mohan D Road safety in less-motorized environments future
concerns International Journal of Epidemiology 200231(3)527ndash32Mortimer 1982
Mortimer RG An evaluation of the motorcycle rider course
Proceedings of the 20th conference of the American Association for
Automotive Medicine Arlington Heights IL USA American
Association for Automotive Medicine 198277ndash96
Mullin 1997
Mullin BT Risk factors for motorcycle injury the role of age
gender experience training and alcohol PhD thesis Auckland
New Zealand University of Auckland 1997
Noordzij 2001
Noordzij PC Forke E Brendicke R Chinn BP Integration of needs
of moped and motorcycle riders into safety measures review and
statistical analysis in the framework of the European research
project PROMISING SWOV Institute for Road Safety Research
Leidschendam The Netherlands 2001 [ Report No Dndash2001ndash5]
Osga 1979
Osga G Ellingstad VS An evaluation of the effectiveness of the 1978 South Dakota Motorcycle Riders Course Program South Dakota
University of South Dakota Vermillion Human Factors Lab 1979
Oxford English Dictionary Online
httpoxforddictionariescom Oxford English Dictionary Online
Oxford University Press 2010
Peden 2004
Peden M Scurfield R Sleet D World Report on Road TrafficInjury Prevention Geneva World Health Organization Available
at httpwwwwhointviolence˙injury˙preventionpublications
road˙trafficworld˙reporten 2004Chapter 3
Radin Umar 1995
Radin Umar R Macakay G Hills B Preliminary analysis of
motorcycle accidents Short-term impacts of the running
headlights campaign and regulations in Malaysia J Traffic Med 199523(1)17ndash27
Tsai 1995
Tsai YJ Wang JD et alCase-control study of the effectiveness of
different types of helmets for the prevention of head injuries among
motorcycle riders in Taipei Taiwan American Journal of
Epidemiology 1995142(9)974ndash81
Vis 1995
Vis AA In-depth study of the hazards of motorcycling [De
onveiligheid van motorrijden nader bekeken] SWOV Institute for
Road Safety Research Leidschendam the Netherlands 1995 [
Report No Rndash95ndash69]
Weaver 1990
Weaver EA Motorcycle safety education in the United States
Motorcycle Safety Foundation Orlando Florida USA 1990
Winn 1987
Winn G Two important trends in motorcycle safety regulations
rider education and conspicuity improvement No more
information available Detroit MI US Society of Automotive
Engineers 1987147ndash52
Yuan 2000
Yuan W The effectiveness of the rsquoride-brightrsquo legislation for
motorcycles in Singapore Accident Analysis amp Prevention 200032
(4)559ndash63
Zambon 2006
Zambon F Hasselberg M Factors affecting the severity of injuries
among young motorcyclists a Swedish nationwide cohort study
Traffic Injury Prevention 20067(2)143ndash9lowast Indicates the major publication for the study
29Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3250
C H A R A C T E R I S T I C S O F S T U D I E S
Characteristics of included studies [ordered by study ID]
Adams 1985
Methods Non-randomised trial
Participants Initial sample size
Total of 162 Australia Post telegram delivery riders
Not clear how many in intervention and control groups
Analysis sample size
Intervention group 63Control group 35
Total 98 (61)
Interventions No formal training vs voluntary rider training course of 12 hours over 2 consecutive days (theory 4 hours practice
8 hours)
Intervention integrity
No information given
Outcomes Australia Post recorded crashes
Length of follow up 1 year
Notes Study conducted in NSW Australia
Anderson 1980
Methods Randomised trial
Participants Initial sample size
Total 41188 motorcycle licence applicants
Analysis sample size
Intervention group with skills training 13160
Intervention group without skills training 12634
Control group 15080
Total 40874 (99)
Interventions Standard licensing skill test vs the Motorcycle Operators Skill Test (MOST) with or without a training component
MOST is not to be confused with rider training because it is a licensing test The training component was for
participants who failed the skill test on the first attempt was of 3 hours duration and consisted of instructions about
the skills needed to pass the skill test The training consisted of 30 min theory and 2 frac12 hours practice on the
motorcycle
Intervention integrity
33 in MOST with skills training 31 in MOST without skills training and 43 in control group obtained their
motorcycle licence in the end
Outcomes Police recorded crashes and offences
Length of follow up 3 years
30Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3350
Anderson 1980 (Continued)
Notes Study conducted in California USA
Billheimer 1998
Methods Cohort study
Participants Initial sample size
Intervention group Unclear numbers of CMSP graduates and ERC graduates
Control group Over 16000 untrained motorcyclists
Analysis sample size
1139 matched CMSP-pairs and 1182 ERC-pairs
Interventions No formal training vs the mandatory (for those younger than 18 years) pre-licence 16-hours basic California
Motorcyclist Safety Program (CMSP) and the 8-hour Experienced Rider Course (ERC)
Intervention integrity
No information given
Outcomes Police recorded crashes and offences
Length of follow up 2 years
Notes Study conducted in California USA
Cooper 1988
Methods Cohort study
Participants Initial sample size
Intervention group 418 graduates and 57 who failed
Control group 402
Analysis sample size
Intervention group unclear
Control group unclear
Total 863 (984)
Interventions No formaltrainingvs non-mandatory pre-licence rider trainingThe content of thetrainingwas notfurther explainedIntervention integrity
Participants had all completed the rider training
Outcomes Police recorded crashes and offences
Length of follow up 7 years
Notes Study conducted in British Columbia Canada
31Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3450
Davis 1997
Methods Retrospective cohort study
Participants Initial sample size
Intervention group 9320 graduates
Control group unclear
Analysis sample size
Intervention group 9320 graduates
Control group unclear
Interventions No formal training vs the non-mandatory Connecticut Motorcycle Operator Training Program The content of
training was not further explained
Intervention integrity Participants had all completed the rider training
Outcomes Police recorded crashes severity of crashes
Length of follow up 1 year
Notes Study conducted in Connecticut USA
Hall 1988
Methods Randomised trial
Participants Initial sample sizeIntervention group A 6248
Intervention group B 6616
Intervention group C 6996
Control group 6604
Total 26464 motorcycle licence applicants
Analysis sample size
Intervention group A 6012 (96)
Intervention group B 6428 (97)
Intervention group C 6778 (97)
Control group 6502 (98)
Interventions Standard licensing skill test vs the Motorcycle Operator Skill Test (MOST II) without training (A) MOST II with
3 hours training (B) or MOST II with 20 hours training (C) The 3 hours training comprised instructions aboutthe skills needed to pass the skill test (30 min theory and 2 frac12 hours practice on the motorcycle) The 20 hours
training consisted of the Motorcycle Safety Foundationrsquos Motorcycle Rider Course (MRC) comprising of 8 hours
theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-road practice
Intervention integrity
25 and 26 obtained their motorcycle licence in the lsquoMOST 3-hour trainingrsquo and lsquoMOST 20-hour trainingrsquo
intervention arm respectively In the lsquoMOST testrsquo intervention arm 29 obtained their licence and in the control
group 35
Outcomes Police recorded crashes and offences
Length of follow up 3 years
32Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3550
Hall 1988 (Continued)
Notes Study conducted in New York USA
Haworth 1997
Methods Population based case-control study
Participants Cases Motorcycle riders or pillion riders involved in motorcycle crash (on road) Recruited from hospital and police
records Not clear whether all cases (especially fatalities) were included
Community controls Multiple matched sample of motorcycle rider or pillion who passed the crash site at the same
time of day and day of week that the crash occurred
Initial sample size
Cases 336
Controls 1195
Analysis sample size
Cases 234 (70)
Controls 561 (47)
Interventions Basic handling skills course beginner intermediate or advanced course measured through self-report not validated
Outcomes Injuries with an Injury Severity Score of 5 or greater or admission for at least four hours and fatalities from a
motorcycle crash (from hospital and police records)
Length of follow up two years or more prior to the interview
Notes Study conducted in Victoria Australia
Hurt 1981
Methods Population based case-control study
Participants Cases Motorcycle riders who had a motorcycle crash that was reported to police (traffic accident reports) Not all
cases included
Community controls Multiple matched sample of motorcycle riders who passed the crash site at the same time of
day and day of week as the crash occurred but 2 years later
Initial sample size
Cases 4500 notifications of crashes to police (only 50 received in time research team able to respond to 1126 for
unreported reasons data collection completed for 900)
Controls Number approached unreported 2310 consented and interviewed
Analysis sample size
Cases 808 (808900 data on rider training) - 18 of initial sample
Controls 670 (6702310 data on rider training) - lt 29 of initial sample
Interventions Training experience self taught friends-family school-club formal-professional others measured through self-
report validated
Outcomes Injuries and fatalities from a motorcycle crash (from hospital records and coronerrsquos office)
33Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3650
Hurt 1981 (Continued)
Notes Study conducted in Los Angeles USA
Jonah 1982
Methods Cohort study
Participants Initial sample size
Intervention group
2310 licensed MTP graduates who are active riders
Control group
2029 licensed motorcyclists who are active riders
Non-licensed and non-active riders excluded
Analysis sample size
Intervention group
811 interviews (35) 533 driver records (23)
Control group 1080 interviews (71) 905 driver records (45)
Interventions No training vs the non-mandatory Canada Safety Council Motorcycle Training Program (MTP) comprising of 4
hours theory and 16 hours practise (6 hrs on-road) over 2 weekends
Intervention integrity
Participants had all successfully completed the MTP
Outcomes Self-reported and police reported crashes and offencesLength of follow up 4 years
Notes Study conducted in Ontario Canada
Kloeden 1994
Methods Cohort study
Participants Initial sample size
Intervention group 2533 licence applicants
Control group Unclear
Sample of riders obtaining probationary licenceIntervention group 1625 (64)
Control group 5015 (unclear)
Non-active riders excluded from control group (~66)
Analysis sample size
Intervention group 631 (39)
Control group 318 (6)
Interventions No formal training vs the mandatory pre-licence Ridersafe course consisting of two levelsthe first is a training course
in the basics of motorcycle riding where a pass entitles the rider to obtain a learnerrsquos permit The second is a skill
test where a pass entitles the rider to obtain a probationary motorcycle licence The training course consists of two 4
hour training sessions on separate days involving instruction videos and actual riding on motorcycles The skill test
consists of a 4 hour preparation course of 74 minutes of lecturediscussion 38 minutes of video and 75 minutes of
34Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3750
Kloeden 1994 (Continued)
on-course riding which is the testing phase
Intervention integrity
Unclear what percentage in the Ridersafe group and the control group obtained their probationary licence halfway
and their full motorcycle licence in the end
Outcomes Police reported crashes
Length of follow up 1 year
Notes Study conducted in SA Australia
Kraus 1975
Methods Population based case-control study
Participants Cases All motorcycle riders with a confirmed medically treated motorcycle injury (including fatalities) from all
hospitals in study base and peripheral area hospitals and all deaths due to motorcycle crashes in the study base All
serious injuries (leading to hospitalisation) and fatalities likely to be included
Community controls A random sample of registered motorcycle owners in the same temporary and geographical
study base as the cases arose from
Initial sample size Cases 1273
Controls 738
Analysis sample size Cases 741 (58)
Controls 434 (59)
Interventions Motorcycle rider training (not specified) measured through self-report
Outcomes Injuries leading to hospitalisation fatalities
Length of follow up not specified
Notes Study conducted in Sacramento county USA
Leung 1987
Methods Non-randomised trial
Participants Initial sample sizeIntervention group unknown number of MRC graduates
Control group unknown number of licensed motorcyclists
Analysis sample size
Intervention group 2941
Control group 43094
Interventions No formal training vs non-mandatory Wisconsin motorcycle rider course a 22 hours off-street programme that
teaches the basic rules of safe motorcycling
Intervention integrity
50 in the MRC group and 100 in the control group obtained their motorcycle licence in the end
35Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3850
Leung 1987 (Continued)
Outcomes Police recorded crashes and offences
Length of follow up up to 3 years
Notes Study conducted in Wisconsin USA
MAIDS 2004
Methods Population based case-control study
Participants Cases Motorcycle riders who had a motorcycle crash Unclear whether all cases included
Community controls Motorcycle riders identified at petrol stations Unclear whether matched to cases
Initial sample size
Cases not reported
Controls not reported
Analysis sample size
Cases 921 (Italy 200 Netherlands 200 Spain 121 Germany 250 France 150)
Controls 923 (Italy 200 Netherlands 200 Spain123 Germany 250 France 150)
Interventions Pre-license rider training or additional training (non-compulsory)
Outcomes Crashes
Notes Study in five sampling areas located in five European countries (France Germany Netherlands Spain and Italy)
McDavid 1989
Methods Cohort study
Participants Sample size after first match
Intervention group 346 male training graduates
Control group 346 male motorcycle riders licensed in same year
Sample size after second match
Intervention group 139 (40)
Control group 139 (40)
Analysis sample size
Intervention group 139 (40)
Control group 139 (40)
Interventions No formal training vs the non-mandatory pre-licence British Columbia Safety Councilrsquos 37-hour motorcycle safety
programme
Intervention integrity
Participants had all completed and passed the training
Outcomes Police reported crashes injurious crashes
Length of follow up 1-6 years
36Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3950
McDavid 1989 (Continued)
Notes Study conducted in British Columbia Canada
Mortimer 1984
Methods Cohort study
Participants Initial sample size
Intervention group 608 MRC graduates
Control group 360 motorcycle riders
Response rates
Intervention group 360 (59)
Control group 324 (90)
Non-active riders excluded
Analysis sample size
Intervention group 213 (35)
Control group 303 (84)
Interventions No formal training vs the non-mandatory Motorcycle Safety Foundationrsquos Motorcycle Rider Course comprising of
8 hours theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-road practice
Intervention integrity
Participants had all completed and passed the MRC
Outcomes Self-reported crashes and offencesLength of follow up 1 year
Notes Study conducted in Illinois USA
Mortimer 1988
Methods Cohort study
Participants Initial sample size
Intervention group 2307 MRC graduates
Control group 556 active licensed motorcycle riders
Response ratesIntervention group 1298 (56)
Control group 500 (90)
Non-active riders excluded
Analysis sample size
Intervention group 913 (40)
Control group 500 (90)
Interventions No formal training vs the non-mandatory Motorcycle Safety Foundationrsquos Motorcycle Rider Course comprising of
8 hours theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-road practice
Intervention integrity
Participants had all completed and passed the MRC
37Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4050
Mortimer 1988 (Continued)
Outcomes Self reported crashes and offences
Length of follow up 1 year
Notes Study conducted in Illinois USA
Osga 1980
Methods Cohort study
Participants Initial sample sizeIntervention group 1163 MSF MRC participants invited 818 (74) responded
Control group 329 motorcycle riders responded to recruitment strategies
Non-active riders excluded from analysis
Analysis sample size
Intervention group 595 (73)
Control group 329 (100)
Interventions No formal training vs the non-mandatory Motorcycle Safety Foundationrsquos Motorcycle Rider Course (MRC) com-
prising of 8 hours theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-road
practice
Intervention integrity
Participants had all successfully completed the MRC
Outcomes Self-reported crashes police recorded crash reports
Length of follow up 6 months - 3 12 years
Notes Study conducted in South Dakota USA
Perrino 2002
Methods Cohort study
Participants Initial sample size
Intervention group 2144 (unclear) MRC graduates (motorcycle licence applicants)
Control group 2601 (unclear) regular licensing test graduates (motorcycle licence applicants)Analysis sample size
Intervention group 1070 (unclear)
Control group 867 (unclear)
Interventions No formal training vs the mandatory (for those younger than 18 years) pre-licence Maryland Motorcycle Rider
Course (MRC) consisting of 8 hours theory and 12 hours practice
Intervention integrity
No information given
Outcomes Self-reported and police recorded crashes and offences
Length of follow up 1 year
38Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4150
Perrino 2002 (Continued)
Notes Study conducted in Maryland USA
Raymond 1979
Methods Cohort study
Participants Initial sample size
Intervention group
258 volunteers mainly (but not all) novice riders
Control group
1104 provisional licence applicants
Analysis sample size
Intervention group 190 (74)
Control group 627 (57)
Interventions No formal training vs the non-mandatory RACACU (Royal Automobile ClubAuto-Cycle Union) training scheme
The training program consisted of either 24 one-hour sessions or 13 two-hour sessions Twelve hours of the training
focused on theory about basic operational procedures traffic laws and highway code The practical session focused
on basic operational skills riding on a training area and riding in traffic with an instructor A proficiency test was
administered at the end of the course
Intervention integrity
No information given
Outcomes Self-reported crashes (including minor non-injury non-damage and off-road crashes)
Length of follow up 1 year
Notes Study conducted in the UK
Satten 1980
Methods Cohort study
Participants Initial sample size
Intervention group interviewers kept interviewing until they had 100 riders selected from a population of 891
enrollees in the MSF MRC course (unknown how many were asked and how many refused to participate)Control group unknown number of registered motorcycle riders
Non-active riders excluded from analysis
Analysis sample size
Intervention group 71
Control group 100
Interventions No formal training vs the non-mandatory Motorcycle Safety Foundationrsquos (MSF) Motorcycle Rider Course (MRC)
comprising of 8 hours theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-
road practice
Intervention integrity
Participants had all successfully completed the MRC
39Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4250
Satten 1980 (Continued)
Outcomes Self-reported crashes injuries and offences
Length of follow up 3 years
Notes Study conducted in Illinois USA
Savolainen 2007
Methods Cohort study
Participants Initial sample size
Intervention group 4000 riders that had conducted the MSF MRC course (passed or failed)
Control group 4000 motorcycle owners or permit or endorsement holders from the Indiana Bureau of Motor
Vehicles database
Analysis sample size
Intervention group 739 (18)
Control group 588 (15)
Interventions Intervention integrity
No information given
Outcomes Probability of crash involvement
Length of follow up up to 5 years
Notes Study conducted in Indiana USA
Swaddiwudhipong 1998
Methods Randomised trial
Participants Initial sample size
Intervention group
all 16 villages of 3 sub-districts
(participants were practising motorcyclists licensed and unlicensed)
Control group
18 villages of 3 other sub-districtsAnalysis sample size
Intervention group 3 districts (100)
Control group 3 districts (100)
Interventions No rider education vs non-mandatory community based rider education consisting of theory on motorcycle injury
risk helmet use traffic laws vehicle regulations and traffic signs and a written and skill test leading to a licence
Intervention integrity
No information given
Outcomes Injuries fatalities
Length of follow up 2 years
40Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4350
Swaddiwudhipong 1998 (Continued)
Notes Study conducted in Tak Province rural Thailand
Waller 1992
Methods Cohort study
Participants Initial sample size
Intervention group 349 riders who succesfully completed the Department of Public Safety motorcycle operator
training in 1988
Control group 349 motorcyclists who never completed the DPS motorcycle operator training
Analysis sample size
Intervention group 349
Control group 349
Interventions No DPS motorcycle operator training vs non-mandatory DPS motorcycle operator training The DPS training is
the Motorcycle Safety Foundationrsquos Motorcycle Rider Course comprising eight hours of theory (lectures discussions
films and multiple choice tests) a textbook and 12 hours off-road practice
Intervention integrity
All 349 riders succesfully completed the rider training
Outcomes Police recorded crashes
Length of follow up 3 years
Notes Study conducted in
Characteristics of excluded studies [ordered by study ID]
Study Reason for exclusion
Anderson 1980a Study type review and process evaluation of a rider course
Aupetit 2008 Outcome proposal of a rider training curriculum that combines traditional training and a motorcycle riding
simulator
Awane 1999 Study type review on use of simulators in rider and driver training
Baldi 2005 Intervention not rider training but intervention integrity of rider training availabilty of rider education
implementation of graduated licensing system etc on a state-wide level
Batchler 1988 Study type description of a training course
Blanchard 2006 Study type not a study assessing the association between rider training and outcomes
Bonnett 2005 Outcome qualitative data regarding satisfaction with the training programme
41Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4450
(Continued)
Braver 2007 Outcome licensure rates
Clayton 1980 Outcome knowledge skills attitude
Council 1978 Outcome crash reduction in ANY vehicle not motorcycles exclusively
Crowther 2005 Study type qualitative study
Daniello 2009 Study type review of the effectiveness of motorcycle training and licensing
Dewanti 2007 Intervention emotional characteristics behaviour (ie not motocycle rider training)
Du Plooy 1990 Study type review of types of rider training courses in South Africa
Engel 1989 Study type uncontrolled before-after study
Goldenbeld 2002 Outcome skills
Goldenbeld 2004 Outcome knowledge and skills
Haworth 2000 Intervention balance between attitudinal andvehicle control skills acrosscurriculla of motorcycle rider training
courses
Haworth 2005 Study type review
Haworth 2006 Study type cross-sectional survey
Hill 1983 Study type review and description of training system
Imatake 1980 Study type course descriptionreview
Ingebritsen 1990 Intervention risk factors for motorcycle crashes
Jain 2009 Study type cross-sectional descriptive study of motorcycle crashes over a 5 year period
Jonah 1979 Study type cross-sectional survey Outcome scores on motorcycle operator skill test
Jonah 1981 Intervention evaluation of a motorcycle rider skill test not motorcycle rider training
Kadar 1990 Study type review of components of rider training courses
Kelsey 1986 Intervention evaluation of a motorcycle rider skill test not motorcycle rider training
Koch 1980 Study type reviewcourse description
Koch 1990 Study type description of course content
42Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4550
(Continued)
Labbett 2005 Study type essay about possible explanations for motorcycle deaths
Law 2005 Intervention targeted motorcycle safety programme not specifically rider training
Lowes 1990 Study type review and descriptions of rider training
Mahon 1976 Study type review of motorcycle rider training methods
Maxwell 1985 Outcomesrecommendations forprogramme operatorsand otherson howthe motorcycle trainingprogramme
works and how it can be improved
Mayhew 1996 Study type review
McKnight 1987 Study type evaluation of methodological and administrative aspects of a motorcycle operator skill test
McPherson 1976 Outcome knowledge and skills
Mohan 2004 Study type review of evidence of interventions to reduce road traffic injuries
Morgan 1980 Study type reviewcourse description
Motoki 1990 Study type proposal and development of rider training curriculum
Nairn 1992 Study type literature review
Newland 1980 Study type description of development of motorcycle safety course
Newland 1982 Study type descriptionreview of a rider training course
Newland 1990 Study type uncontrolled BeforeAfter study
Ormston 2003 Outcome changes in attitude and self reported behaviour
Osga 1979 Outcome knowledge attitudes and skills
Panou 2007 Intervention new simulation tool to teach driverrider
Peterson 1990 Study factor demographic details of course participants completing the course vs those not completing the
course
Prem 1984 Study type experimental
Study factor examines association between skills and performance in tests
Radin Umar 2006 Intervention motorcycle safety programme not specifically rider training
Rockwell 1990 Study type methodologically flawed study type
43Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4650
(Continued)
Rothe 1987 Study type qualitative study of rider training in general
Rowden 2007 Study type qualitative study
RSC Victoria 1993 Study type review of motorcycle rider training and education in Victoria
Russam 1979 Study type review of training programmes in Japan
Saffron 1981 Study type literature review
Schulz 1990 Study factor predictors of motorcycle crashes
Sexton 2004 Study type cross-sectional analytic analysis
Shephard 1986 Study factor reviewdescription of programme
Simpson 1990 Study type review of motorcycle rider trainingeducation
Sokolov 1990 Study factor review of motorcycle riding courses in USSR
Steffens 1988 Outcome self-reported behaviour experiences and researcher observed riding behaviour
Sudlow 2003 Study type review of motorcycle training in the UK
Tasmania 1995 Study type cross-sectional descriptiveprocess evaluation
Tetard 1985 Study type experimental
Outcome detection of indices
Thackray 1980 Outcome instructional effectiveness user acceptance administrative feasibility
Thompson 1994 Study type cross-sectional survey
Outcome attitudes and knowledge
Tobler 1990 Study type review of course curricula
Transport 1989 Intervention syllabus for rider training
Tuan 2007 Study type cross-sectional study (survey) which does not assess the effects of an intervention but examines
motocyclistsrsquo behaviour
Victor 1980 Study type review
Waterer 1990 Study type review of UK rider training issues
Weller 1989 Study type review of studies evaluating motorcycle safety and education measures
44Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4750
(Continued)
Wells 1981 Study type description of different methods of training
Winn 1986 Study type review of rsquohistoryrsquo of rider training
Wisher 1988 Study type cross-sectional survey
Wood 1987 Study type Interrupted Time Series but did not meet the criteria for an ITS as published by the Cochrane
Effective Practice and Organisation of Care Group
Yamazaki 1990 Study type cross-sectional survey
Outcome riding skills skills retention and attitude
Characteristics of ongoing studies [ordered by study ID]
Burgess 2010
Trial name or title UKs National Ride program
Methods Cohort study
Participants Motorcycle offenders
Interventions Police led diversion program
Outcomes Self reported attitudinal and behavioural measures
Starting date
Contact information
Notes Expected to be completed in early 2010
45Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4850
D A T A A N D A N A L Y S E S
This review has no analyses
A P P E N D I C E S
Appendix 1 Search strategy
We searched the following electronic databases
bull Cochrane Injuries Group Specialised Register (searched 22 Sept 2008)
bull CENTRAL (The Cochrane Library 2008 Issue 3)
bull TRANSPORT 1988 to 200706 (latest date available)
bull ATRI Australian Transport Index (to Sept 2008)
bull MEDLINE (Ovid SP) 1950 to Sept 2008
bull EMBASE (Ovid SP) 1980 to Sept 2008
bull CINAHL (EBSCO) 1982 to Sept 2008
bull WHOLIS (World Health Organization Library Information System) (to Sept 2008)
bull PsycINFO (Ovid SP) 1806 to Sept 2008
bull LILACS (BIREME) (Latin American and Caribbean Health Sciences) (1982 to Sept 2008)
bull ISI Web of Science Social Sciences Citation Index (SSCI) 1970 to Sept 2008
bull ERIC (searched 25 Sept 2008)
bull ZETOC (searched 25 Sept 2008)
bull SIGLE (searched 25 Sept 2008)
All of the databases were searched using the following text-word search strategy The strategy was adapted using either rsquorsquo or rsquo$rsquo as atruncation symbol
1 motor-cycl or motorcycl or motor-bik or motorbik or motor-scooter or motorscooter or scooter or moped or motocycl
or motocicl or moto
2 train or educat or instruct or program or course or qualif or learn or improv
3 1 and 2
H I S T O R Y
Protocol first published Issue 2 2005
Review first published Issue 10 2010
46Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4950
C O N T R I B U T I O N S O F A U T H O R S
Katina Kardamanidis was the first author of the protocol conducted the literature search extracted full studies assessed the studies
according to inclusion criteria assessed quality of the studies compared findings with the second author discussed disagreements with
the third author wrote the drafts and the final review
Rebecca Ivers was the second author of the protocol assessed the studies according to inclusion criteria assessed quality of the studies
compared findings with the first author edited the final draft and provided overall guidance for the process
Alexandra Martiniuk assessed quality of the studies compared findings with the first author discussed disagreements with the third
author and helped write the first draft and the final review
Mark Stevenson contributed to the protocol reviewed methods of studies and contributed to the final draft
Katrina Thistlethwaite conducted a second literature search extracted full studies and contributed to the final draft
D E C L A R A T I O N S O F I N T E R E S T
None known
S O U R C E S O F S U P P O R T
Internal sources
bull No sources of support supplied
External sources
bull National Health and Medical Research Council of Australia Australia
Salary support for Martiniuk Ivers and Stevenson
bull University of Sydney Sesqui Research Grant Australia
Direct research costs and salary support for Kardamanidis
D I F F E R E N C E S B E T W E E N P R O T O C O L A N D R E V I E W
There are no differences between the protocol and the review
I N D E X T E R M S
Medical Subject Headings (MeSH)
lowastMotorcycles Accident Prevention [lowastmethods] Accidents Traffic [lowastprevention amp control] Licensure Program Evaluation Wounds
and Injuries [mortality lowastprevention amp control]
47Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 5050
MeSH check words
Humans
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 1850
estimatedthat 62of thetotalstudysample received the interven-
tion The remaining studies did not report the percentage of thestudy population that received the intervention as was intended
or reported that this percentage was unknown (Raymond 1979
Kloeden 1994 Satten 1980 Jonah 1982 Mortimer 1984 Leung
1987 Cooper 1988 Mortimer 1988 Billheimer 1998 Perrino
2002 Savolainen 2007) Only Raymond 1979 reported the degree
to which their intervention was delivered as intended No studies
reported on possible lsquocontaminationrsquo of participants by other inter-
ventions than the one to which they were allocated Because Osga
1980 Satten 1980 Jonah 1982 Mortimer 1984 Cooper 1988
Mortimer 1988 Waller 1992 and Davis 1997 were retrospective
studies and participants were recruited from a list of graduates it
was known that they had completed and passed the course
Analysis
In some studies some participants who reported at baseline that
they were untrained decided to enroll in a rider training pro-
gramme after the baseline survey The non-randomised trial by
Adams 1985 experienced this move of subjects from the untrained
to the trained group but failed to conduct analyses by lsquointention-
to-treatrsquo This can bias the results in randomised and non-ran-
domised trials if participants who changed groups differ signifi-
cantly with regard to potentially confounding characteristics On
the other hand Perrino 2002 also experienced this move of sub-
jects from theuntrained to the trained groupmore than 100riders(5 of the total sample) originally stated that they were untrained
at baseline but then received training between the baseline and
follow-up surveys Yet since the Perrino 2002 study was a cohort
their analyses then proceeded by including those riders who were
lsquonewlyrsquo trained as exposed Riders who were lsquonewlyrsquo trained since
baseline but with a programme other than the one under study
were excluded from the analysis Nevertheless confounding fac-
tors were not accounted for in the analyses
McDavid 1989 matched exposed with unexposed subjects but
could not find suitable matches for the majority (60) of their
subjects They could therefore only conduct analysis on the result-
ing40 (139346) which is an inappropriate andinefficient anal-
ysis Some studies that reported on the issue of non-active licensedriders(Satten 1980 Jonah 1982 Mortimer 1984 Mortimer 1988
Kloeden 1994) excluded non-active riders from analysis intro-
ducing selection bias as it is likely that riders who continued to
ride differed with regards to potentially confounding character-
istics across intervention and control groups Kloeden 1994 for
example conducted several analyses to assess the effect of various
sub-sets of the control group Original analyses included all con-
trols and led to erroneous results since only a small proportion of
controls hadgone on to get a licence after their learnerrsquos permitand
were actually exposed to the possibility of experiencing a crash
Subsequent analyses compared only intervention and control sub-
jects who had a licence Waller 1992 reported crashes by person-
days of exposure but did not explain how they could be sure that
each study intervention and control participant was an active riderand actually exposed to the risk of experiencing a crash
Potential confounders that are unequally distributed by study arm
at baseline need to be dealt with in the analysis Satten 1980
Adams 1985 and Davis 1997 did not take into account any con-
founding factors either by design or in the analysis McDavid
1989 Waller 1992 and Billheimer 1998 accounted for potential
confounding factors at the design stage by matching the subjects
in the intervention and control groups Other studies (Raymond
1979 Mortimer 1984 Mortimer 1988 Perrino 2002 and also
Waller 1992) adjusted for one or two confounders by stratifica-
tionor standardisation (eg crashesmiles driven age group) Osga
1980 conducteda univariate analysis of crash rates in which no po-
tential confounders were accounted for but also calculated cumu-lative crash rates by mileage travelled Only Jonah 1982 Cooper
1988 and Savolainen 2007 conducted a multivariate analysis
Risk of bias in the case-control studies
A case-control design is suitable for the assessment of risk fac-
tors for outcomes such as motorcycle crashes injuries or deaths
but not appropriate for the evaluation of an intervention such
as motorcycle rider training It can potentially result in spurious
conclusions regarding the efficacy of the intervention because the
intervention has not been applied at random to an intervention
and control group Even with the control of possible confounders
using a case-control design to evaluate an intervention can lead tothese kinds of results since those who choose to undertake rider
training may differ on important prognostic factors that affect the
outcome compared to those who do not choose to participate in
rider training
Selection bias
A potential source of bias in case-control studies as with all other
study types where the intervention is not randomly allocated to
participants is that comparison groups are likely to differ on po-
tentially confounding characteristics This means that ideally all
cases need to be included in the study as well as a random sample
of controls taken from the population that gave rise to the cases If that is not possible cases and controls either need to be carefully
matched to make themcomparable or differences at baseline need
to be adjusted for in the analysis All four case-control studies were
population based Kraus 1975 included all fatalities and injured
cases leading to hospitalisation and appropriately recruited a ran-
dom sample of controls from the same population Hurt 1981 did
not manage to recruit all or a random sample of cases Only half
of the notifications of crashes to the police were received in time
by the researchers the research team for unreported reasons was
only able to respond to half of those again Data collection was
therefore completed for only about a fifth of all cases Neverthe-
less this was still a large sample of 900 cases Haworth 1997 using
16Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 1950
a similar study design was more likely to have included all cases
but this is not entirely clear from the study reported It is unclearfor example whether fatalities were included as they did not use
coronerrsquos reports for the recruitment of cases The MAIDS 2004
study was conducted in five European countries but it was unclear
how cases were recruited whether they were all cases during the
two years that the study lasted or a random sample of all cases
what the response rates were etc Controls were recruited from
randomly selected petrol stations but it was likewise not clear what
the eligibility criteria for inclusion in the study were how they
were recruited (eg randomly) and what the response rate was
Hurt 1981 and Haworth 1997 both matched controls to cases by
recruiting themat the sametime ofdayandsamedayof weekas the
crash occurred Hurt 1981 however recruited controls two years
later andit is likelythat there would have beendifferences betweenthe two comparison groups One important difference was that
during data collection for cases there was very little specialised
motorcycletrainingavailable in the studyregionwhereas two years
later when control data were collected there was some training
available
These four case-control studies also suffered from selection bias
due to potential differences between those that gave consent to
participate in the study and responded to interviews andquestion-
naires and those that did not None of the studies have assessed
such differences although Hurt 1981 did acknowledge that con-
trols giving consent to cooperate may have differed from those
motorcyclists who did not
Detection bias
Measures that can be taken to reduce detection bias in case-con-
trol studies are blinding of interviewers for case or control status
blinding of the participant for the study factor of interest validat-
ing measurement instruments and using these in an identical way
for cases and controls and cross checking self-reported informa-
tion with other more objective sources such as registers None of
the studies reported blinding of the intervention assessors to case
or control status but all used consistent measurement methods for
both case and control groups
All case-control studies used self-report as their main measure toassess whether cases and controls had received rider training and
if so the type of training The major potential source of bias in
case-control studies when self-report is used as a measure is recall
bias One way of reducing recall bias is by double-checking self-
reported data with more objective data Kraus 1975 also used po-
lice accident reports for information on rider characteristics but
such information was not available about rider training Haworth
1997 and MAIDS 2004 did not report on any measures to limit
recall bias Hurt 1981 reported further questioning of cases by
the interviewer when answers did not make sense or when par-
ticipants seemed to contradict themselves as part of the in-depth
unstructured interview technique This qualitative research tech-
nique can result in valuable information but as interviewers were
not blinded to case or control status of the participants it may have led to more thorough examination of cases compared to the
controls and thus differential detection bias
Intervention integrity
The level of detail about the rider training provided in the case-
control studies differed significantly but was minimal for each
Kraus 1975 simply asked whether lsquoanyrsquo motorcycle rider training
had lsquoeverrsquo been carried out Hurt 1981 asked participants whether
they hadbeen self taught informally taught or completed formally
taught rider training courses MAIDS 2004 asked whether partic-
ipants had had pre-license training additional (non-compulsory)
training other training or no training whereas Haworth 1997asked about the number of courses that a participant had com-
pleted whetherthey were basic intermediate or advanced courses
and how long ago the training had been done None of the case-
control studies reported collecting information about further de-
tails about any training programmes (eg type of training cur-
riculum duration of training)
Analysis
Kraus 1975 accounted for confounding factors through the exclu-
sion of females and stratification by age and frequency of motor-
cycle use Hurt 1981 matched cases and controls for time of day
and day of the week that the crash occurred and did not adjust forany confounding factors in the analysis Haworth 1997 matched
groups for time of day and day of the week that the crash occurred
and stratified by age blood alcohol level and licence status in the
analysis None of the studies used multivariate methods MAIDS
2004 did not provide any information about any participant char-
acteristics other than the study exposures of interest and did not
describe any methods to adjust for potential confounding factors
Effects of interventions
We have presented the results in a descriptive manner due to the
substantial heterogeneity in study types interventions and out-comes and the generally poor quality of the studies We have
grouped the interventions by mandatory and non-mandatory pre-
licence rider training and post-licence rider training and further
by study type and by outcome (crashes injurious and fatal crashes
and offences) We discuss results that are reported as statistically
significant (whether a P value was reported or not)
Pre-licence rider training - mandatory
Randomised trials
17Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2050
Crashes
Total group of participants
Anderson 1980 found a reduction in motorcycle crash rates in
the intervention group (MOST skill test with remedial training)
compared to the control group (who received the skill testwithout
a training component) The percent differenceranged from 143
six months after the programme (P lt 005) to 211 (P lt 0001)
after one year and 127 (P lt 001) after two years Hall 1988
found no statistically significant difference in crash rates between
intervention groups at any point in time
Subgroup of participants who obtained their licence only
Anderson 1980 found a reduction in motorcycle crash rates in the
intervention arm but this was statistically significant at one year
only (24 to 218 reduction P lt 005) Hall 1988 found no
statistically significant differences
Offences
Total group of participants
The percentage of offences in Anderson 1980 after six months
and one and two years was slightly higher in the intervention
arm compared to the control group but these differences were
not statistically significant Likewise the study by Hall 1988 did
not find a statistically significant difference in offences between
groups
Subgroup of participants who obtained their licence only
Hall 1988 found no significant difference in offence rates between
the intervention and control groups Anderson 1980 found thatthe percentage of offences increased in the intervention arms com-
pared to the control group This difference was statistically signif-
icant after two years (51 to 115 increase P lt 005)
Cohort studies
Crashes
Neitherof the cohort studies (Kloeden1994 and Billheimer1998)
found a statistically significant difference in crash rates between
the intervention and control groups
Offences
Billheimer 1998 (for the California Motorcyclist Safety Program
course) and Perrino 2002 did not find a statistically significant
difference in offence rates between the intervention and control
groups
Case-control studies
Crashes
MAIDS 2004 found that a slightly higher percentage of controls
(not involved in a crash) had had pre-license rider training than
cases (involved in a crash) This difference was not statistically significant (50 of controls vs 477 of cases OR 091 95 CI
076 to 110)
Pre-licence rider training - non-mandatory
Randomised trial
Injuries and deaths
Swaddiwudhipong 1998 found a significantly lower injury ratein the intervention villages (1051000 population 95 CI 92
to 120) compared to the control villages (1691000 population
95 CI 152 to 185) one year after the intervention This differ-
ence was not maintained two years after the intervention (127
95 CI 112 to 143 intervention 151 95 CI 135 to 167
control) There was no significant difference in death rates be-
tween groups
Cohort studies
Crashes
Cooper 1988 foundno statistically significant differencein crashes
between the intervention and control groups McDavid 1989
found a lower crash rate in the intervention group (14139) com-
pared to the control group (23139) but no test of statistical sig-
nificance was undertaken
Injuries and deaths
McDavid 1989 found that the intervention group had fewer in-
jurious crashes (814 crashes) compared to the control group (16
23 crashes) but no statistical significance test was reported
18Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2150
Offences
Cooper 1988 foundno statistically significantdifference in offencerates between the intervention and control groups McDavid 1989
foundno statistically significantdifference in offence ratesbetween
the intervention and control groups
Non-mandatory rider training
Non-randomised trials
Crashes
Neither Adams 1985 nor Leung 1987 found a statistically signif-icant effect of rider training on crash rates
Offences
Leung 1987 found a statistically significant lower offence rate in
the intervention group compared to the control group for both
males and females and for all age groups except for women 35
years or older (P values not reported)
Cohort studies
Crashes
Osga 1980 in the univariate analysis reported a lower percentage
of crashesin theintervention group (that continued to ride) (28
168595) than in the control group (34 112329) In the cu-
mulative crash rate analysis however Osga1980 found statistically
significant (at the P lt 001 level) higher crash rates in the interven-
tion group than in the control group at allmileage intervals (1000
to 29999 miles) travelled After riding more than 30000 miles
(when over 50 of the trained riders had been involved in a crash
compared to approximately 40 in untrained riders) this differ-
ence was no longer statistically significant Satten 1980 found a
lower percentage of crashes in the intervention group (13 969)
compared to the control group (25 2079) but the differencewas not assessed statistically Davis 1997 found a lower crash rate
in trained riders but it was unclear how this was calculated be-
cause the size of the control group was not reported Jonah 1982
and Billheimer 1998 (for the Experienced Rider Course) found no
statistically significant difference in crashes between the interven-
tion and control groups Although the difference between those
who completed the Experienced Rider Course and no course in
Billheimer 1998 neared significance at six months after the course
(P = 007) large P values were observed for one year after (P =
030) Raymond 1979 found a statistically significant higher crash
rate for the intervention group (1861000 miles ridden) com-
pared to the control group (1181000 miles ridden) in both males
and females (P lt 0001) especially during the first 5000 miles of
riding Waller 1992 found a statistically significant higher crashrate for the intervention group (27 crashes in three years or 070
crashes10000 person days) compared to the control group (5
crashes in three years or 013 crashes10000 person days) with a
crude relative risk of 538 (X 2=1513 plt005) Mortimer 1984
and Mortimer 1988 found no statistically significant difference in
crashes between the intervention and control groups
Savolainen 2007 found that the intervention group (those who
completed the Basic Rider Course) was 44 more likely to be
involved in crashes than the control group and those who com-
pleted the course multiple times were an additional 180 more
likely to be involved in a crash This finding was based on low
numbers and no P value or confidence level was given for either
result
Injuries and deaths
Satten 1980 also found a lower percentage of injurious crashes
(33 39) in the intervention group compared to the control
group(60 1220) although again no statistical significance test
was reported
Offences
Jonah 1982 did find a significantly lower self-reportedoffence rate
in the intervention group (15) compared to the control group(34 P lt 001) (no numbers given) Satten 1980 also found
a lower self-reported offence rate in the intervention group (1
71 versus 19100) but again this was not tested for statistical
significance Mortimer 1984 Mortimer1988 and Billheimer 1998
found no statistically significant differencein offence ratesbetween
the intervention and control groups
Case-control studies
CrashesMAIDS 2004 found that a slightly higher percentage of cases
(involved in a crash) had had additional non-mandatory rider
training than controls (not involved in a crash) This difference
was not statistically significant (17 of cases vs 12 of controls
OR 146 95 CI 064 to 337)
Rider training unspecified
Case-control studies
19Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2250
Injuries and deaths
Hurt 1981 found that the odds of injurious crashes were lower inthose who had received professional training compared to those
who had not Kraus 1975 found that in the younger age group
(less than 20 years) and the older age group (over 35 years) a larger
percentage of injured riders (12 and 21 respectively) reported
prior motorcycle rider training compared with the uninjured con-
trol group (6 and 5 respectively) (P lt 001) (no numbers
given) Haworth 1997 reported several different outcomes one
beingthat having completed at leastone trainingcoursecompared
to no training courses at all had no significant effect on the odds
of crashing (afteradjustment for age blood alcohol level or licence
status (learner permitprobationaryfull)) When comparing three
different levels of training courses (beginner intermediate and ad-
vanced) with no training no significant differences were found
in the odds of crashing However after adjustment for age and
blood alcohol level completing an advanced course as compared
to a beginners course was associated with a significant decrease in
the odds of severe motorcycle crashes resulting in serious injury or
death (OR 02 95 CI 01 to 08)
D I S C U S S I O N
The objective of this review was to quantify the effectiveness of
pre- andpost-licence motorcycle rider training on the reduction of
traffic crash involvement injuries deaths and offences committedby motorcycle riders Most included studies were cohort studies
There were a few randomised and non-randomised trials and case-
control studies
All studies compared either mandatory or non-mandatory formal
training with no formal training except for Haworth 1997rsquos case-
control study that compared levels of training (beginner interme-
diate and advanced) to each other
We found studies that evaluated the effectiveness of manda-
tory pre-licence training in reducing crash and offence rates
(Anderson 1980 Hall 1988 Kloeden 1994 Billheimer 1998
(the basic California Motorcyclist Safety Program component of
the study) Perrino 2002) non-mandatory training that when
passed resulted in a licence (Cooper 1988 McDavid 1989
Swaddiwudhipong 1998) non-mandatory training that could be
taken by motorcycle riders with or without a licence (Raymond
1979 Osga 1980 Satten 1980 Jonah 1982 Mortimer 1984
Adams 1985 Leung 1987 Mortimer 1988 Waller 1992 Davis
1997 Billheimer 1998 (the Experienced Rider Course compo-
nent of the study)) and unspecified rider training on crash injury
and death rates (Kraus 1975 Hurt 1981 Haworth 1997 MAIDS
2004)
The studies provided conflicting evidence regarding motorcycle
rider training which is likely due to the poor design quality and
reporting of the studies Most of the studies included in the review
were likely to be adversely affected by significant selection biasalong with substantial detection bias(measurement and recall bias)
which limits our ability to make conclusive comments
Mandatory pre-licence rider training
There were two RCTs three cohort studies and one case-control
study that examined pre-licence ridertraining The two RCTs eval-
uated the Motorcycle Skills Operating Test (MOST) (Anderson
1980 Hall 1988) Anderson 1980 found a statistically significant
reduction in crashes in the intervention group at six months one
and twoyearsfollow up However when only participantswho ob-
tainedtheirlicence were included in theanalysis thereduction wasstatistically significant only at one year follow up The other RCT
(Hall 1988) found no statistically significant differences in crash
rates at any point of time neither in the total population group
nor in the licensed subgroup Neither of the two cohort stud-
ies assessing crash rates found a statistically significant difference
in crash rates between intervention and control groups ( Kloeden
1994 Billheimer1998) andthe case-control study (MAIDS 2004)
also did not find a statistically significant difference in the odds of
experiencing a crash between trained and untrained riders With
regards to offences neither of the RCTs found a statistically sig-
nificant difference in offence rate between the intervention and
control groups in the analysis of the total group of participants
entering the trial However Anderson 1980 found a statistically significant increase in offence rates in the intervention group but
only after two years of follow up andonly in the analysis of partici-
pants who obtained their licence The two cohort studies assessing
offence rates found no statistically significant difference between
the groups (Billheimer 1998 Perrino 2002)
None of the five studies (RCTs or cohort) found a reduction in
crashes or offences following mandatory pre-licence rider training
except for Anderson 1980 who observed a statistically significant
reduction in crashes in the intervention group at one year in both
analyses It is not clear why both RCTs that evaluated the same
MOST and training intervention observed different results Both
studies enrolled large samples and observed a similar uptake of
licensure following the intervention It is possible that differencesmay have arisen due to different study samples (California ver-
sus New York) or that the intervention delivery differed between
studies
In both trials of the Motorcycle Skills Operating Test including
rider training only a third of motorcycle licence applicants ob-
tained their motorcycle licence It is possible that a licensing pro-
cess including a skills test may reduce crashes through decreased
exposure
Non-mandatory pre-licence training
20Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2350
Three studies assessed the effect of non-mandatory pre-licence
training The Cooper 1988 cohort study found no difference incrash or offence rates between the trained and untrained groups
We cannot however be confident about this result as the study
sufferedfrom selectionbiaswhichcouldhave biasedthe measure of
effect toward the null The RCT conducted by Swaddiwudhipong
1998 examined injury and death rates and found a statistically
significant lower injury rate in the intervention villages after one
year of follow up but no longer at two year follow up There
was no statistically significant difference in death rates between
intervention and control villages Our confidence in the results of
this study are hindered by the inadequacy of the randomisation
process that led to significant differences between the intervention
and control districts on two important variables The intervention
districts were more likely than the control districts to have a validlicence andto wear a helmetSincethese differences were nottaken
into account in later analyses they could explain the differences
observed in injury rates between districts Furthermore with only
six districts randomised the sample size was probably too small
to detect a difference between intervention and control groups
had there been one The McDavid 1989 cohort study found a
lower percentage of injuries in the intervention group although
a statistical test was not conducted McDavid 1989 drew their
intervention and control groups from different populations but
did adjust for confounders although they only analysed 40 of
the initial sample (139 trained and 139 untrained) Because of
these substantial weaknesses we are unable to draw conclusions
from this studyAn overall effect of pre-licence training on the reductionof crashes
injuries deaths and offences was not found although one should
be aware that the research of this issue to date has had significant
flaws However a mandatory skill test (such as the MOST) in
combination with and even without pre-licence training as part
of a licensing process seems to reduce the number of individuals
completing the licensing procedure and therefore indirectly re-
duces crash rates through a reduction in exposure to motorcycle
riding Making licensure a more difficult process may therefore be
seen as advantageous if it delays licensure and therefore reduces
exposure It is possible that the group who chose not to continue
riding a motorcycle would not have been the group to experience a
crash or commit an offence if they had continued to ride anyhowfor example because they are more safety conscious and this was
the reason for their giving up riding in the first place If this is the
case deterring individuals from completing their licence and rid-
ing may not necessarily decrease overall crashes deaths or offence
rates In addition other potentially adverse effects of complicat-
ing a licensing procedure need to be considered such as increases
in unlicensed riding In Thailand non-mandatory rider training
offered as a community intervention may have decreased injury
rates but this needs to be confirmed through further research
Non-mandatory training
The majority of the studies assessed non-mandatory training
(training independent of any licensing procedure) and the results
were varied The two non-randomised trials assessing crash rates found no statistically significant difference in crash rates between
the intervention and control groups (Adams 1985 Leung 1987)
Of the nine cohort studies Satten 1980 found a lower percentage
of crashesin the intervention group although it is unclear whether
this was a statistically significant result Davis 1997 also found a
lower percentage of crashesin theinterventiongroup although it is
unclear with which control group the intervention group wascom-
pared Raymond 1979 Osga 1980 Waller 1992 and Savolainen
2007 found a statistically significant higher crash rate in the in-
tervention group and the remaining cohort studies (Jonah 1982
Mortimer 1984 Mortimer 1988 Billheimer 1998) found no sta-
tistically significant difference in crash rates between intervention
and control groups No conclusions can be drawn from these stud-
ies with confidence The Satten 1980 and Davis 1997 studies did
not adjust for any confounding factors even though important
differences existed between the intervention and control groups
Although Raymond 1979 and Waller 1992 adjusted for a num-
ber of confounders at baseline (Waller 1992) and in the analy-
sis (Raymond 1979 Waller 1992) the intervention and control
groups were drawn from such different populations that it is pos-
sible that unmeasured confounders were responsible for the sta-
tistically significant increased risk of crashes observed in the inter-
vention group The observed difference could also be attributed totheobserved differential loss to followup (Raymond 1979) which
could bias the measure of effect towards or away from the null
The Osga 1980 study found a protective effect of rider training
in the univariate analysis but statistically significant higher crash
rates in the intervention group when accounting for riding expo-
sure Sex (a significantly higher proportion of females in the in-
tervention group) and other potentially confounding factors that
could explain thedifferencebetween the groups were howeverun-
accounted for Finally Savolainen 2007 did control for potential
confounding variables yet it is not clear whether variables were
controlled for simultaneously or one at a time One case-control
study (MAIDS 2004) found no statistically significant reduction
in the odds of experiencing a crash between trained and untrainedriders but the quality of this study was poor mainly because the
analysis did not account for any confounding factors
One cohort study (Satten 1980) assessed the effect of non-manda-
tory rider training on injurious crashes and found a lower percent-
age of injuries in the intervention group although a statistical test
was not conducted As mentioned above the Satten 1980 studyrsquos
validity is reduced due to its lack of adjustment for confounding
factors
The only non-randomised trial that assessed offence rates (Leung
1987) found a statistically significant lower offence rate in the in-
tervention group and of the five cohort studies that assessed of-
fence rates (Satten 1980 Jonah 1982 Mortimer 1984 Mortimer
21Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2450
1988) one study (Jonah 1982) also found a statistically significant
lower offence rate in the intervention group However we have lit-tle confidence in the results from Leung 1987 since potential con-
founding factors were not appropriately addressed Jonah 1982
did conduct a multivariate analysis but was hindered by an initial
low response rate The other four cohort studies did control for
several potential confounders and found no difference in offence
rates following non-mandatory post-licence rider training
The non-randomised trials and cohort studies evaluating non-
mandatory rider training were at high risk of substantial selection
bias due to the non-random allocation of participants as well
as low participation rates and attrition bias and could therefore
not provide conclusive evidence with regard to the effect of the
training It should be noted that despite the biases and other issues
with this group of studies most didnot detectmajor effects relatedto rider training in either direction This may suggest that any
potential impact of rider training will be modest
Studies examining unspecified modes of ridertraining
Three case-control studies assessed whether injured or deceased
motorcyclists were more or less likely to have had rider training
compared to control motorcyclists All had conflicting results
Hurt1981 foundthe odds of havingan injuriouscrash waslower in
those who had received rider training whereas Kraus 1975 found
that the odds of having an injurious crash were higher in youngerand older trained riders Haworth 1997 who compared advanced
versus beginner training found no effect on motorcycle injuries
or deaths although they did find that those who had completed
advanced training had reduced chances of crashing compared to
those who had done a beginnersrsquo training course Similar issues
with uncontrolled confounding and biases of selection as described
above arose with the three case-control studies Hurt 1981 suffers
from inadequate control of confounding via poor matching and
both the Kraus 1975 and the Haworth 1997 findings were only
controlled for limited factors by means of stratified analysis It is
likely that these limitations are the reason for the contradictory
directions of effect observed for the three case-control studies
Methodological weaknesses of the studies
In summary most studies investigating the effectiveness of mo-
torcycle rider training on reducing crashes injuries and death suf-
fered from serious methodological weaknesses
The majority of studies were non-randomised and were there-
fore limited mainly by selection bias Although the approaches of
matching stratification or regression analyses can minimise bias
(Fleiss 2004) they were poorly employed in the reviewed studies
The inadequate adjustment for confounders in particular adds
to our lack of confidence in the studiesrsquo results since it has been
demonstrated that individuals who choose to take motorcycle rider
training differ from those not taking rider training For instance
trained riders often differ from controls in age sex and experi-ence usually ride less and may have a more positive attitude to
safety as indicated by the reported more frequent wearing of seat
belts (Mortimer 1982)
The two large randomised trials reported that a large proportion
of their study group were never exposed to the risk of offences or
crashes because they never attained licensure This made it impos-
sible to assess whether a reduction in crashes in the intervention
group was a result of training or due to delayed or never attained
licensure
In some studies the sample sizes were too small to detect any sta-
tistically significant differences Follow-up time after training dif-
fered between studies and was an impediment to observing any
real effects of training if they existed As riding experience ac-cumulates over time any differences in outcomes between those
trained and untrained would diminish in studies with lengthy fol-
low-up times This dilution of any possible effect of rider training
may have also occurred in studies that mixed both novice and ex-
perienced riders in the study groups since it is hypothesised that
novice riders serve to benefit most from rider training This would
increase the difficulty in finding any effect of rider training par-
ticularly when sample sizes are small
Detection bias may also account for some of the differences ob-
served between trained and untrained riders Detection bias would
have occurred if the self-report hospital driver or police records
used to record the outcomes of crash injury death or offences
differed between the trained and untrained riders Research hasdemonstrated that not all crashes are reported to police for ex-
ample small crashes or off-road crashes (Kraus 1975) even serious
injury crashes have been found to be underreported (Anderson
1980) However this would only introduce detection bias if the
reporting of outcome differs for trained and untrained riders This
does appear possible in that responsible riders may be more likely
to enroll in a rider training course and also may be more likely
to report crashes to police than irresponsible riders meaning that
the number of crashes from riskier riders could be underrepre-
sented on records An analysis to investigate the effectiveness of
rider training may then show that rider training is not effective in
reducing crashes
Rather than using reported outcomes some studies used self-re-ported survey data although this technique suffers from non-
response biases In fact two studies reported differences in de-
mographic characteristics in response rates to questionnaires
(Anderson 1980 Hall 1988) Self-reported data also introduce
the risk of recall bias survival bias and social desirability response
bias Recall bias may have been equally distributed among trained
and untrained riders howeverit is hypothesised that trained riders
may have felt it was lsquoexpectedrsquo to report good driving histories
hence leading the rider training to appear effective
22Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2550
Limitations of this review
Limitations of this review were that we found very few recentstudies (the two most recent ones published in 2002 and 2007)
and that we found very few studies conducted in low-and middle
income countries even though our search strategy was set up to
identify such studies Anotherlimitation wasthat forthe reviewed
studies we could not assess all outcomes of rider training that are
potentially useful Some of the included studies assessed outcomes
such as use of safety clothing including helmets Such outcomes
are valuable since the use of motorcycle helmets has been demon-
strated to be effective in preventing head injuries (Liu 2004) De-
spite this the use of safety clothing was not included as an out-
come in this review because the studies measuring this outcome
were limited by poor data quality due to very low response rates
Another outcome of interest in assessing the effectiveness of rider
training is lsquotype of injuryrsquo As Haworth 2005 state in their review
of motorcycle licensing and training rider training could be con-
sidered effective if injuries become less severe because motorcycle
riders have learned how to avoid more serious injuries through
rider training The studies included in this review did not specify
type of injury and so this outcome could not be assessed Last
changes in knowledge and attitudes following rider training are
other potentially useful outcomes We have excluded such studies
(Osga 1979 Clayton 1990 Goldenbeld 2004) as these outcomes
are proxies rather than direct measures of the effectiveness of mo-
torcycle rider training We opted to report on the outcomes of
crashes injury death and offences alone since there is little ev-
idence about how changes in knowledge and attitudes relate to
these outcomes
A U T H O R S rsquo C O N C L U S I O N S
Implications for practice
Due to the poor quality of studies included in this review we were
unable to draw any conclusions about the effectiveness of rider
training on crash rates injurious or fatal or offence rates The
findings do suggest that mandatory pre-licence training may be
an impediment to completing a motorcycle licensing process thuspossibly indirectly reducing crash injury death and offence rates
through a reduction in exposure to riding a motorcycle On the
basis of the existing evidence it is not clear if and what type of
training reduces the risk of crashes injuries deaths or offences in
motorcyclists and the selection of the best rider training practice
can therefore not be recommended The suggestion that manda-
tory pre-licence training indirectly reduces exposure to riding the
motorcycle because trainees are impeded from completing the li-
censing process does not help those motorcyclists who persist in
obtaining their licence and riding the motorcycle Some type of
rider training is likely necessary to teach motorcyclists to ride a
motorcycle safely Good policy is based on sound evidence and
evaluation but such evidence for motorcycle rider training is seri-
ously lacking Policy makers are therefore encouraged to include arigorous evaluation component to any new or existing rider train-
ing programme Such an evaluation should ideally be randomized
and assess whether the training programme is effective and if so
which component(s) of the programme work best
Implications for research
Given theresults of this review which found littlereliable evidence
about the effectiveness of motorcycle rider training it is necessary
to build up an evidence base of the effectiveness of motorcycle
rider training in the reduction of crash injury death and offence
rates To this end well designed randomised controlled trials need
to be conducted that examine the impact of rider training oncrashes and offences A randomised controlled trial is the preferred
method as it would significantly reduce most biases described in
this review To minimise the impact of participants delaying or
never attaining their licence and therefore exposure as found in
the randomised trials in this review such a trial should include
riders who are already licensed and reliable measures of riding
exposure in a reasonably lengthy follow up period It is expected
that a randomised controlled trial of sufficient sample size would
observe equal balance on possible confounders by study arm how-
ever an assessment of potentially important confounders (such as
helmet wearing) by study group would enable control for these
variables in the analyses if necessary It is imperative that the effec-
tiveness of each component of the rider training (such as theorypractice knowledge skills and modification of behaviour) and the
synergy of such components are evaluated Outcomes should be
objective ideally those that are routinely collected rather than self-
reported outcomes Even though riding the motorcycle poses dif-
ferent challenges and incurs different risks lessons may be learned
from research into the effects of car driver training
Motorcycles are a common means of transport in low- and middle
income countries and usage is increasing As the burden of mo-
torcycle crash victims is mostly borne in such countries research
should be focused on the effectiveness of types of rider training on
the reduction of crashes injuries and fatalities in such countries
and be supported by higher income countries This review found
only one study conducted in a middle income country that sug-gested decreasing injury rates It would be beneficial to conduct
additional research to confirm this suggestion and to enhance our
knowledge of the role of rider training in low- and middle-income
settings
March 2006 saw the launch of the University of North Carolina
Highway Safety Research Centre partnership with the Motorcycle
Safety Foundation (MSF) and plans were revealed to conduct a
study of the potential benefits of motorcycle rider training The
study is funded by the MSF and the National Highway Traffic Sa-
fety Administration (NHTSA) Called rsquoThe Discovery Projectrsquo it
will compare data on crash rates from riders participating in a basic
23Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2650
motorcycle safety class with riders participating in a basic motor-
cycle safety class as well as a variety of continuing education classesThe collection of pilot data was due to start in March 2007 and
the data collection portion of the study is expected to take three
years (Highway Safety Research Center) It is vital that researchers
and policy makers collaborate to develop furtherhigh quality eval-
uations of motorcycle rider training programmes to help inform
development of new and effective training programmes
A C K N O W L E D G E M E N T S
Staff from The University of Sydney library for helping us locate
and retrieve full reports of many papers
R E F E R E N C E S
References to studies included in this review
Adams 1985 published data only
Adams A Collingwood V Job RFS Evaluation of the 1982 Australia
Post motorcycle rider training scheme NSW Australia Traffic
Accident Research Unit Traffic Authority NSW 1985 [ RN 85]
Anderson 1980 published data only
Anderson J Ford JL Peck RC Improved motorcyclist licensing and testing project Washington DC National Highway Traffic Safety
Administration (NHTSA) USA 19801ndash217 [
DOTndashHSndash805ndash598]
Billheimer 1998 published data only
Billheimer JW Evaluation of the California Motorcyclist Safety
Program Transportation Research Record 19981640100ndash9 [
TRB Report Number 980652]
Cooper 1988 published data only
Cooper PJ Rothe JP Motorcyclists who they are and why they do
what they do Transportation Research Record 1988116878ndash85
Davis 1997 published data only
David CF Evaluation of community traffic safety programs and
motorcycle operator training programs Connecticut Department of Transportation 199763
Hall 1988 published data only
Hall CJ Buchanan L Engle R Motorcycle rider education evaluationproject Washington DC USA National Highway Traffic Safety
Administration (NHTSA) US Department of Transportation
1988 [ DTNH 22ndash80ndashCndash05123]
Haworth 1997 published data only
Haworth N Smith R Brumen I Pronk N In Monash University
Accident Research Centre editor(s) Case-control study of motorcycle
crashes Victoria Australia Department of Transport and Regional
Development The Federal Office of Road Safety 19971ndash149 [
ISBN 0 642 25501 6 ]
Hurt 1981 published data only
Hurt H Ouellet J Thom D Motorcycle accident cause factors and
identification of countermeasures USA Department of
Transportation 1981 [ Technical Report DOT HS 806ndash862]
Jonah 1982 published data onlylowast Jonah BA Dawson N Bragg B Are formally trained motorcyclists
safer Accident Analysis amp Prevention 198214(4)247ndash55
Jonah BA Dawson NE Bragg BW Evaluation of the Canada
Safety Councilrsquos Motorcycle Training Program Proceedings of theInternational Motorcycle Safety Conference Washington DC
USA 198059ndash93
Kloeden 1994 published data only
Kloeden CN Moore VM McLean AJ Evaluation of the pre-license
training program for motorcyclists in South Australia Walkerville
South Australia Australia NHMRC Road Accident Research Unit
The University of Adelaide 19941ndash37 [ ISBN 0 7308 0607 3]
Kraus 1975 published data only
Kraus JF Riggins RS Franti CE Some epidemiological features of
motorcycle collision injuries American Journal of Epidemiology 1975102(1)74ndash98
Leung 1987 published data only
Leung KS Reding VA Evaluation of the Wisconsin motorcycle rider course Wisconsin Bureau of Policy Planning and Analysis
Wisconsin Department of Transportation 1987
MAIDS 2004 published and unpublished data
MAIDS In-depth investigations of accidents involving powered
two wheelers ACEM (Acemacemeu) 2004 Vol Final report
20179
McDavid 1989 published data only
McDavid JC Lohrmann BA Lohrmann G Does motorcycle
training reduce accidents Evidence from a longitudinal quasi-
experimental study Journal of Safety Research 198920(2)61ndash72
Mortimer 1984 published data only
Mortimer R Evaluation of the motorcycle rider course Accident
Analysis amp Prevention 198416(1)63ndash71
24Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2750
Mortimer 1988 published data only
Mortimer RG A further evaluation of the motorcycle rider courseJournal of Safety Research 198819(4)187ndash96
Osga 1980 published data only
Osga GA An investigation of the riding experiences of MSF rider
course participants Dissertation South Dakota USA The
University of South Dakota 19801ndash224
Perrino 2002 published data only
Perrino CS Ahmed A Callender A Rozier E Cantwell A Stewart
O The role of Marylandrsquos motorcycle rider course in promoting saver
behaviors and attitudes Baltimore USA National Transportation
Center Morgan State University 20021ndash52
Raymond 1979 published data only
Raymond S Tatum S An evaluation of the effectiveness of the RAC
ACU motorcycle training scheme - final report Road Safety ResearchUnit Dept of Civil Engineering University of Salford London
19791ndash131
Satten 1980 published data only
Satten RS Analysis and evaluation of the motorcycle rider courses
in thirteen Illinois counties International Motorcycle Safety
Conference Washington DC 1980145ndash93
Savolainen 2007 published and unpublished data
Savolainen Peter Mannering Fred Additional evidence on the
effectiveness of motorcycle training and motorcyclistsrsquo risk-taking
behaviour Transportation Research Board 2007 Vol March 20
19
Swaddiwudhipong 1998 published data only
Swaddiwudhipong W Boonmak C Nguntra P Mahasakpan P
Effect of motorcycle rider education on changes in risk behavioursand motorcycle-related injuries in rural Thailand Tropical Medicine
and International Health 19983(10)767ndash70
Waller 1992 published data only
Waller FL An analysis of the impact of the Texas Department of
Public Safety Motorcycle Operator Training Program upon
motocycle related traffic accidents Texas State University Public
Administration Program Applied Research Projects 199277
References to studies excluded from this review
Anderson 1980a published data only
Anderson LC Better biking a motorcycle rider improvement
program International Motorcycle Safety Conference Washington
DC USA Motorcycle Safety Foundation 1980231ndash53Aupetit 2008 published and unpublished data
Aupetit S Riff J Buttelli O Espie S Ergonomics research program
on motorcycle learning Contributions to the development of a
motorcycle simulator for training Transport Research Arena
Ljubljana 20089
Awane 1999 published data only
Awane T Integrating simulators in motorcycle safety education
Journal of International Association of Traffic and Safety Sciences 199923(1)26ndash35
Baldi 2005 published data only
Baldi S Baer JD Cook AL Identifying best practices states in
motorcycle rider education and licensing Journal of Safety Research200536(1)19ndash32
Batchler 1988 published data only
Batchler MW Evaluation of the pre-license motorcycle training course at the New South Wales traffic education centre Federal Office of
Road Safety (FORS) 19881ndash122 [ Report No CR 77 ]
Blanchard 2006 published data only
Blanchard HT Tabloski PA Motorcycle safety educating riders at
the teachable moment Mosby (CV) Company 200632330ndash2
Bonnett 2005 published data only
Bonnett J Mature age skills training for experienced motorcycle
riders Motorcycle ridersrsquo association of the Australian Capital
Territory 2005
Braver 2007 published data only
Braver ER et alPersuasion and licensure a randomized controlled
intervention trial to increase licensure rates among Maryland
motorcycle owners Traffic Injury Prevention 20078(1)39ndash46Clayton 1980 published data only
Clayton AB Sudlow DE An evaluation of the effectiveness of a one day training programme for learner motorcyclists Berkshire UK
Transport and Road Research Laboratory 19801ndash23 [ ISSN
0266ndash7045]lowast Clayton AB Sudlow DE An evaluation of the effectiveness of a one-day training programme for learner motorcyclists Berkshire UK
Road Safety Division Berkshire 19801ndash23 [ISSN 0266ndash7045]
Council 1978 published data only
Council FM Desper LP Dutt AK Driver education for motorcycle operation University of North Carolina Highway Safety Research
Center 19781ndash40
Crowther 2005 published data onlylowast Crowther G Engaging with motorcyclists UK police and theBikeSafe Road Safety Programme Published proceedings of the
Australasian Road Safety Research Policing Education Conference
Wellington New Zealand 2005 issue ISBN 0ndash473ndash10636ndash11ndash11
Daniello 2009 published data only
Daniello A Gabler HC Mehta YA Effectiveness of motorcycle
training and licensing Transportation Research Board 88th Annual
Meeting 200913
Dewanti 2007 published data only
Dewanti M Challenging efforts in promoting young motorcyclist
safety in Indonesia Road Safety on Four Continents 14th
International Conference Swedish National Road and Transport
Research Institute 2007
Du Plooy 1990 published data onlyDu Plooy DW Motorcycle training in South Africa The Human
Element International Motorcycle Safety Conference Orlando
Florida USA Motorcycle Safety Foundation 19904-1 to 4-9
Engel 1989 published data only
Engel U Krogsgaringrd Thomsen L [Faeligrdselsundervisning og
propaganda ndash en faeligrdselssikkerhedsfremmende foranstaltning]
Dansk Vejtidsskrift 198912295ndash7
Goldenbeld 2002 published data only
Goldenbeld C Houwing S de Craen S The development of
driving skills of young moped riders [De ontwikkeling van de
rijvaardigheid van jonge bromfietsers] SWOV Institute for Road
Safety Research (Stichting Wetenschappelijk Onderzoek
Verkeersveiligheid) Leidschendam the Netherlands 200263
25Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2850
Goldenbeld 2004 published data only
Goldenbeld C Twisk D de Craen S Short and long term effects of moped training a field experiment Transportation Research Part F 20047(1)1ndash16
Haworth 2000 published data only
Haworth N Smith R Kowadlo N Motorcycle training in Victoria
evaluation of rider training curriculum in Victoria Monash
University Accident Research Centre Reports 2000 Vol 165 issue
3131
Haworth 2005 published data only
Haworth N Mulvihill C Review of motorcycle licensing and
training Monash University Accident Research Centre 2005 [
Report no240]
Haworth 2006 published data only
Haworth N Mulvihill C Rowdon P Teaching old dogs new tricksTraining and older motorcyclists Journal of the Australasian College of Road Safety 200718(4)20ndash5lowast Haworth NL Mulvihill C Rowden PJ Teaching old dogs new
tricks Training and older motorcyclists Australasian Road Safety
Research Policing and Education Conference Gold Coast
Queensland 2006
Hill 1983 published data only
Hill D Motorcycle rider training testing and licensing Motor
Cycle Council of New South Wales Sydney 1983
Imatake 1980 published data only
Imatake Y On the activities in Japan of motorcycle safe riding
promotion committee (MSRPC) - especially the education of
unskilled riders of class 1 motorbikes International Motorcycle
Safety Conference Washington DC USA Motorcycle Safety
Foundation 1980255ndash68
Ingebritsen 1990 published and unpublished data
Ingebrigtsen S [Risikofaktorer ved ferdsel med moped og
motorsykkel] TOslashI-rapport 66 Oslo Transportoslashkonomisk
institutt 1990
Jain 2009 published and unpublished data
Jain A et alTwo wheeler accidents on Indian roads a study from
Mangalore India Elsevier 200916130ndash3
Jonah 1979 published data only
Jonah BA Dawson NE Validation of the motorcycle operator skill
test Accident Analysis amp Prevention 197911(3)163ndash71
Jonah 1981 published data onlyJonah BA Dawson NE Bragg WE Predicting accident
involvement with the Motorcycle Operator Skill Test Accident Analysis amp Prevention 198113(4)307ndash418
Kadar 1990 published data only
Kadar KJ Countermeasures motorcycle rider education delivery
system The Human Element International Motorcycle Safety
Conference Motorcycle Safety Foundation 19906-27 to 6-41
Kelsey 1986 published data only
Kelsey SL Liddicoat C Ratz M Licensing novice motorcyclists a
comparison of Californiarsquos standard test and the MOST II (Motorcycle
Operator Skill Test) administered at centralized testing offices Sacramento California USA California Department of Motor
Vehicles Research and Development Office 19861ndash74
Koch 1980 published data only
Koch H Advanced course for motorcycle riders InternationalMotorcycle Safety Conference Washington DC USA Motorcycle
Safety Foundation 1980269ndash81
Koch 1990 published data only
Koch H Brendicke R How to influence risk taking behaviour
within motorcycle rider training programs The Human Element
International Motorcycle Safety Conference Orlando Florida
USA Motorcycle Safety Foundation 1990Motorcycle Safety
Foundation 1990 p 9-1 to 9-25A
Labbett 2005 published data only
Labbett S Langham M Training can make the problem worse
Road Safety Congress 20051ndash12
Law 2005 published and unpublished data
Law TH Radin Umar RS Zulkaurnain S Kulanthayan S Impactof the effect of economic crisis and the targeted motorcycle safety
programme on motorcycle-related accidents injuries and fatalities
in Malaysia International Journal of Injury Control and Safety
Promotion 200512(1)9ndash21
Lowes 1990 published data only
Lowes BF In-traffic rider training The Human Element
International Motorcycle Safety Conference Orlando Florida
USA Motorcycle Safety Foundation 19907-46 to 7-57
Mahon 1976 published data only
Mahon J Teaching accident avoidance on the motorcycle Journal of Traffic Safety Education 1976231ndash29
Maxwell 1985 published data only
Maxwell GS Evaluation of a Program for Training MotorcycleRiders Evaluation and Program Planning 19858(4)339ndash47
Mayhew 1996 published data only
Mayhew DR Simpson SH Effectiveness and role of driver educationand training in a graduated licensing system Ottowa Canada The
Traffic Injury Research Foundation 19961ndash89 [ ISBN
0ndash920071ndash12ndash0]
McKnight 1987 published data only
McKnight AJ Milburn BG An evaluation of the modified
motorcycle operator skill test exercises NHTSA Report 1987 [
Report No DOT HS 807 212 ]
McPherson 1976 published data only
McPherson K McKnight AJ The development and evaluation of a
motorcycle skill test manual and knowledge test Washington DCUSA National Highway Safety Administration 1976
Mohan 2004 published data only
Mohan D Evidence-based interventions for road traffic injuries in
South Asia Journal of the College of Physicians and SurgeonsPakistan 200414(12)746ndash7
Morgan 1980 published data only
Morgan JK Advanced motorcycle training course content and
study evaluation International Motorcycle Safety Conference
Washington DC USA Motorcycle Safety Foundation 1980
283ndash312
Motoki 1990 published data only
Motoki M Yamazaki S A study on effective motorcycle rider
education The Human Element International Motorcycle Safety
26Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2950
Conference Orlando Florida USA Motorcycle Safety
Foundation 19909-26 to 9-57A
Nairn 1992 published data onlylowast Nairn RJ Motorcycle safety research literature review 1987 to
1991 Federal Office of Road Safety Australia 1993 Vol Report
CR 117
Newland 1980 published data only
Newland R Darby J Automotive practicespre motorcycle rider
program International Motorcycle Safety Conference Washington
DC USA Motorcycle Safety Foundation 1980127ndash9
Newland 1982 published data only
Newland R Motorcycle rider training and licensing requirements
Driver training steering a course for the future Melbourne
Victoria Transport Regulation Board 1982
Newland 1990 published data only
Newland R Motorcycle training and licensing in Victoria 1980-
1990 The Human Element International Motorcycle Safety
Conference Orlando Florida USA Motorcycle Safety
Foundation 1990
Ormston 2003 published data only
Ormston R Dudleston A Pearson S Stradling Evaluation of
Bikesafe Scotland Scottish Executive Social Research Edinburgh
2003
Osga 1979 published data only
Osga GS Ellingstad VS An evaluation of the effectiveness of the 1978
South Dakota motorcycle rider course program South Dakota USA
University of South Dakota 19791ndash89
Panou 2007 published data only
Panou M Bekiaris E A new concept on the integration of driving
simulators in driving training the train-all approach Road Safety
on Four Continents 14th International Conference Swedish
National Road and Transport Research Institute 2007
Peterson 1990 published data only
Peterson A Kadar K Status report 1980-1989 Minnesota rider
education statistics The Human Element International
Motorcycle Safety Conference Orlando Florida USA 199013-
113 to 13-129
Prem 1984 published data only
Prem H Good MC Motorcycle rider skills assessment ACT
Australia Office of Road Safety Department of Transport 1984
1ndash377 [ Report No CR 34 ]
Radin Umar 2006 published data only
Radin Umar RS Motorcycle safety programmes in Malaysia how
effective are they International Journal of Injury Control and Safety Promotion 200613(2)71ndash9
Rockwell 1990 published data only
Rockwell TH Kiger S Carnot M An evaluation of the Ohio
motorcyclist enrichment program Ohio Department of Highway
Safety 1990
Rothe 1987 published data only
Rothe JP A motorcycle riding course Motorcyclists Image and reality Canada Insurance Corporation of British Columbia 1987
59ndash83
Rowden 2007 published data only
Rowden P Watson B Haworth N What can riders tell us aboutmotorcycle training A view from the other side of the fence
Australasian Road Safety Research Policing and Education
Conference Melbourne Australia Melbourne 2007
RSC Victoria 1993 published data only
Road Safety Committee Inquiry into Motorcycle Safety in Victoria
Parliament of Victoria 1993
Russam 1979 published data only
Russam K Motorcycle training and licensing in Japan UK Transport
and Road Research Laboratory (TRRL) 1979 [ Report 916 ]
Saffron 1981 published data only
Saffron DG Driver and rider education training and licensing a
brief review Traffic Accident Research Unit Traffic Authority of
New South Wales NSW Australia Traffic Accident Research UnitTraffic Authority of New South Wales 1981
Schulz 1990 published data only
Schulz U Motorcycle accidents among learner riders Motorcycle
licensing Evaluation of the effectiveness of a Graduated Licensing Scheme Forshungshefte des Institut fur Zweiradsicherheit 1990
Sexton 2004 published data only (unpublished sought but not used)
Sexton B Baughan C Elliott M Maycock G The accident risk of
motorcyclists 2004TRL Report TRL 607(ISSN 09684107)58
Shephard 1986 published data only
Shepherd R Do Motorcycle Safety Foundation programs work
Driver 198619(9)16ndash8
Simpson 1990 published data onlySimpson HM Mayhew DR The promotion of motorcycle safety
training education and awareness Health Education Research
Theory and Practice 19905(2)257ndash64
Sokolov 1990 published data only
Sokolov O Safe motorcycle riding some aspects of motorcycle
riding in the USSR The Human Element International
Motorcycle Safety Conference Motorcycle Safety Foundation
19904ndash37 - 4-48
Steffens 1988 published data only
Steffens U Gawatz R Willmes G Effectiveness of mofa courses
[Wirksamkeit von Mofakursen] Unfall- und sicherheitsforschung
strasenverkehr 198867168
Sudlow 2003 published data only
Sudlow D Scoping study on motorcycle training Road Safety
Research Report London UK The British Institute of Traffic
Education Research Department for Transport 2003 Vol June
[ Report No 36]
Tasmania 1995 published data only
Mannering FL Grodsky LL Motorcycle rider training scheme
Traffic Education Research Centre Armidale Australia Driver
Training Australia Limited 1995 Vol April1ndash56
Tetard 1985 published data only
Tetard C Motorcycle ridersrsquo training to detect indices in an
intersection situation [Lrsquoentrainement a la prise drsquoinformation dans
lrsquoapprentissage de la conduite motocycliste] Le Travail Humain198548(2)147ndash59
27Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3050
Thackray 1980 published data only
Thackray RM Prescott JC Field test of the motorcycle ridercourse International Motorcycle Safety Conference Washington
DC USA Motorcycle Safety Foundation 1980195ndash7
Thompson 1994 published data only
Thompson M Evaluation of compulsory basic training for
motorcyclists Transport Research Laboratory Crowthorne
Berkshire UK 1994 [ Report 63]
Tobler 1990 published data only
Tobler U Individual driving instruction for motorcycle driving
learners the Swiss experiment of motorcycle driving school The
Human Element International Motorcycle Safety Conference
Orlando Florida USA Motorcycle Safety Foundation 1990
Transport 1989 published data only
Transport Communications Road Crash Statistics Australia Bureauof Transport and Communications Economics 198914ndash20
Tuan 2007 published data only
Tuan VA Shimizu T Towards development and evaluation of the
motorcycle drivers re-education program in Vietnam modeling of
motorcycle driverrsquos undesired behaviors 11th World Conference
on Transport Research 200725
Victor 1980 published data only
Victor DJ Traffic safety education for motorcyclists International
Motorcycle Safety Conference Washington DC USA Motorcycle
Safety Foundation 1980221ndash8
Waterer 1990 published data only
Waterer TH Compulsory training for motorcyclists from liberty
to compulsion The Human Element International Motorcycle
Safety Conference Orlando Florida USA Motorcycle Safety
Foundation 19904-49 to 4-57
Weller 1989 published data only
Weller RB Chandler EW Motorcycle safety and motorcycle
education past research and survey results Journal of Public Policy
amp Marketing 1989893ndash108
Wells 1981 published data only
Wells P Observations of some RACACU motorcycle training
schemes Transport and Road Research Laboratory Crowthorne
Berkshire UK 1981
Winn 1986 published data only
Winn G Two important trends in motorcycle safety regulations
rider education and conspicuity improvement American
Motorcyclist Association 1986147ndash52
Wisher 1988 published data only
Wisher G Reid J Motorcycle accidents in Norfolk a three-year study of motorcycle training and the Part 1 test Norfolk UK Road Safety
Division Highways Department Norfolk County Council 1988
Wood 1987 published data only
Wood HT Bowen R Evaluation of the revised learner permit scheme
July 1983 to December 1985 Victoria Australia Road Traffic
Authority 19871ndash27 [ ISBN 0 7306 0247 8]
Yamazaki 1990 published data only
Yamazaki S Nakanishi M Survey on effect of Honda rider
education The Human Element International Motorcycle Safety
Conference Orlando Florida USA 199012-167 to 12-187
References to ongoing studies
Burgess 2010 unpublished data only
Burgess C Broughton P Fylan F Stradling Interim evaluation of
the UKs National Ride program In L Dorn editor(s) Driver
behaviour and Training Vol IV Aldershot Ashgate Publishing
Limited 2010
Additional references
Anderson 1980
Anderson J Ford JL Peck RC Improved motorcyclist licensing and
testing project Volume I and II National Highway Traffic Safety
Administration US department of Transportation Washington
DC 1980
Bjornstig 1985
Bjornstig U Bylund P Lekander T Brorsson B Motorcycle fatalitiesin Sweden Acta Chirurgica Scandinavica 1985151(7)577ndash81
Clayton 1990
Clayton A Sudlow DE An evaluation of the effectiveness of a one-
day training program for learner motorcyclists The Human
Element International Motorcycle Safety Conference Orlando
Florida USA 199013-105 to 13-111
Cochrane Collaboration 2008
Higgins JP Green S Cochrane Handbook for Systematic Reviews of
Interventions 500 The Cochrane Collaboration 2008
Elliot 2003
Elliott MA Baughan CJT Broughton J Grayson G Knowles J et
alMotorcycle Safety a Scoping Study TRL [ Report No
0968ndash4107]Fleiss 2004
Fleiss JL Levin B Paik MC Statistical methods for rates and proportions 3rd Edition John Wiley amp Sons Inc 2004 [ ISBN
978ndash0ndash471ndash52629ndash2 ]
FORS 1999
FORS Federal Office of Road Safety Road risk for sober licensed
motorcyclists Vol Monograph 27 Federal Office of Road Safety
19991ndash2
Haworth 2005
Haworth N Mulvihill C Review of motorcycle licensing and training
Victoria Australia Monash University Accident Research Centre
20051ndash83 [ Report No 240]
Huwiler-Muumlntener 2002Huwiler-Muumlntener K Juumlni P Junker C Egger M Quality of
Reporting of Randomized Trials as a Measure of Methodologic
Quality JAMA 20022872801ndash4
Johnston 2008
Johnston P Brooks C Savage H Fatal and serious road crashes
involving motorcyclists Road Safety Research and Analysis Report
ATSB Monograph 20 Canberra Department of Infrastructure
Transport Regional Development and Local Government 0205
081ndash26
Juumlni 2001
Juumlni P Altman DG Egger M Systematic reviews in health care
Assessing the quality of controlled clinical trials BMJ 2001323
42ndash6
28Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3150
Kopits 2003
Kopits E Cropper M Traffic fatalities and economic growth TheWorld Bank Development Research Group Infrastructure and
Environment Policy research working paper 3035 2003
Leung 1983
Leung P Chau BTW Motorcycle accidents in Hong Kong pattern
of injury treatment causative factors and socio-economic
implications Injury 198315(1)23ndash9
Liu 2004
Liu B Ivers R Norton R Blows S Lo SK Helmets for preventing
injury in motorcycle riders Cochrane Database of Systematic Reviews 2004 Issue 2 [DOI 10100214651858CD004333pub3]
Mohan 2002
Mohan D Road safety in less-motorized environments future
concerns International Journal of Epidemiology 200231(3)527ndash32Mortimer 1982
Mortimer RG An evaluation of the motorcycle rider course
Proceedings of the 20th conference of the American Association for
Automotive Medicine Arlington Heights IL USA American
Association for Automotive Medicine 198277ndash96
Mullin 1997
Mullin BT Risk factors for motorcycle injury the role of age
gender experience training and alcohol PhD thesis Auckland
New Zealand University of Auckland 1997
Noordzij 2001
Noordzij PC Forke E Brendicke R Chinn BP Integration of needs
of moped and motorcycle riders into safety measures review and
statistical analysis in the framework of the European research
project PROMISING SWOV Institute for Road Safety Research
Leidschendam The Netherlands 2001 [ Report No Dndash2001ndash5]
Osga 1979
Osga G Ellingstad VS An evaluation of the effectiveness of the 1978 South Dakota Motorcycle Riders Course Program South Dakota
University of South Dakota Vermillion Human Factors Lab 1979
Oxford English Dictionary Online
httpoxforddictionariescom Oxford English Dictionary Online
Oxford University Press 2010
Peden 2004
Peden M Scurfield R Sleet D World Report on Road TrafficInjury Prevention Geneva World Health Organization Available
at httpwwwwhointviolence˙injury˙preventionpublications
road˙trafficworld˙reporten 2004Chapter 3
Radin Umar 1995
Radin Umar R Macakay G Hills B Preliminary analysis of
motorcycle accidents Short-term impacts of the running
headlights campaign and regulations in Malaysia J Traffic Med 199523(1)17ndash27
Tsai 1995
Tsai YJ Wang JD et alCase-control study of the effectiveness of
different types of helmets for the prevention of head injuries among
motorcycle riders in Taipei Taiwan American Journal of
Epidemiology 1995142(9)974ndash81
Vis 1995
Vis AA In-depth study of the hazards of motorcycling [De
onveiligheid van motorrijden nader bekeken] SWOV Institute for
Road Safety Research Leidschendam the Netherlands 1995 [
Report No Rndash95ndash69]
Weaver 1990
Weaver EA Motorcycle safety education in the United States
Motorcycle Safety Foundation Orlando Florida USA 1990
Winn 1987
Winn G Two important trends in motorcycle safety regulations
rider education and conspicuity improvement No more
information available Detroit MI US Society of Automotive
Engineers 1987147ndash52
Yuan 2000
Yuan W The effectiveness of the rsquoride-brightrsquo legislation for
motorcycles in Singapore Accident Analysis amp Prevention 200032
(4)559ndash63
Zambon 2006
Zambon F Hasselberg M Factors affecting the severity of injuries
among young motorcyclists a Swedish nationwide cohort study
Traffic Injury Prevention 20067(2)143ndash9lowast Indicates the major publication for the study
29Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3250
C H A R A C T E R I S T I C S O F S T U D I E S
Characteristics of included studies [ordered by study ID]
Adams 1985
Methods Non-randomised trial
Participants Initial sample size
Total of 162 Australia Post telegram delivery riders
Not clear how many in intervention and control groups
Analysis sample size
Intervention group 63Control group 35
Total 98 (61)
Interventions No formal training vs voluntary rider training course of 12 hours over 2 consecutive days (theory 4 hours practice
8 hours)
Intervention integrity
No information given
Outcomes Australia Post recorded crashes
Length of follow up 1 year
Notes Study conducted in NSW Australia
Anderson 1980
Methods Randomised trial
Participants Initial sample size
Total 41188 motorcycle licence applicants
Analysis sample size
Intervention group with skills training 13160
Intervention group without skills training 12634
Control group 15080
Total 40874 (99)
Interventions Standard licensing skill test vs the Motorcycle Operators Skill Test (MOST) with or without a training component
MOST is not to be confused with rider training because it is a licensing test The training component was for
participants who failed the skill test on the first attempt was of 3 hours duration and consisted of instructions about
the skills needed to pass the skill test The training consisted of 30 min theory and 2 frac12 hours practice on the
motorcycle
Intervention integrity
33 in MOST with skills training 31 in MOST without skills training and 43 in control group obtained their
motorcycle licence in the end
Outcomes Police recorded crashes and offences
Length of follow up 3 years
30Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3350
Anderson 1980 (Continued)
Notes Study conducted in California USA
Billheimer 1998
Methods Cohort study
Participants Initial sample size
Intervention group Unclear numbers of CMSP graduates and ERC graduates
Control group Over 16000 untrained motorcyclists
Analysis sample size
1139 matched CMSP-pairs and 1182 ERC-pairs
Interventions No formal training vs the mandatory (for those younger than 18 years) pre-licence 16-hours basic California
Motorcyclist Safety Program (CMSP) and the 8-hour Experienced Rider Course (ERC)
Intervention integrity
No information given
Outcomes Police recorded crashes and offences
Length of follow up 2 years
Notes Study conducted in California USA
Cooper 1988
Methods Cohort study
Participants Initial sample size
Intervention group 418 graduates and 57 who failed
Control group 402
Analysis sample size
Intervention group unclear
Control group unclear
Total 863 (984)
Interventions No formaltrainingvs non-mandatory pre-licence rider trainingThe content of thetrainingwas notfurther explainedIntervention integrity
Participants had all completed the rider training
Outcomes Police recorded crashes and offences
Length of follow up 7 years
Notes Study conducted in British Columbia Canada
31Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3450
Davis 1997
Methods Retrospective cohort study
Participants Initial sample size
Intervention group 9320 graduates
Control group unclear
Analysis sample size
Intervention group 9320 graduates
Control group unclear
Interventions No formal training vs the non-mandatory Connecticut Motorcycle Operator Training Program The content of
training was not further explained
Intervention integrity Participants had all completed the rider training
Outcomes Police recorded crashes severity of crashes
Length of follow up 1 year
Notes Study conducted in Connecticut USA
Hall 1988
Methods Randomised trial
Participants Initial sample sizeIntervention group A 6248
Intervention group B 6616
Intervention group C 6996
Control group 6604
Total 26464 motorcycle licence applicants
Analysis sample size
Intervention group A 6012 (96)
Intervention group B 6428 (97)
Intervention group C 6778 (97)
Control group 6502 (98)
Interventions Standard licensing skill test vs the Motorcycle Operator Skill Test (MOST II) without training (A) MOST II with
3 hours training (B) or MOST II with 20 hours training (C) The 3 hours training comprised instructions aboutthe skills needed to pass the skill test (30 min theory and 2 frac12 hours practice on the motorcycle) The 20 hours
training consisted of the Motorcycle Safety Foundationrsquos Motorcycle Rider Course (MRC) comprising of 8 hours
theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-road practice
Intervention integrity
25 and 26 obtained their motorcycle licence in the lsquoMOST 3-hour trainingrsquo and lsquoMOST 20-hour trainingrsquo
intervention arm respectively In the lsquoMOST testrsquo intervention arm 29 obtained their licence and in the control
group 35
Outcomes Police recorded crashes and offences
Length of follow up 3 years
32Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3550
Hall 1988 (Continued)
Notes Study conducted in New York USA
Haworth 1997
Methods Population based case-control study
Participants Cases Motorcycle riders or pillion riders involved in motorcycle crash (on road) Recruited from hospital and police
records Not clear whether all cases (especially fatalities) were included
Community controls Multiple matched sample of motorcycle rider or pillion who passed the crash site at the same
time of day and day of week that the crash occurred
Initial sample size
Cases 336
Controls 1195
Analysis sample size
Cases 234 (70)
Controls 561 (47)
Interventions Basic handling skills course beginner intermediate or advanced course measured through self-report not validated
Outcomes Injuries with an Injury Severity Score of 5 or greater or admission for at least four hours and fatalities from a
motorcycle crash (from hospital and police records)
Length of follow up two years or more prior to the interview
Notes Study conducted in Victoria Australia
Hurt 1981
Methods Population based case-control study
Participants Cases Motorcycle riders who had a motorcycle crash that was reported to police (traffic accident reports) Not all
cases included
Community controls Multiple matched sample of motorcycle riders who passed the crash site at the same time of
day and day of week as the crash occurred but 2 years later
Initial sample size
Cases 4500 notifications of crashes to police (only 50 received in time research team able to respond to 1126 for
unreported reasons data collection completed for 900)
Controls Number approached unreported 2310 consented and interviewed
Analysis sample size
Cases 808 (808900 data on rider training) - 18 of initial sample
Controls 670 (6702310 data on rider training) - lt 29 of initial sample
Interventions Training experience self taught friends-family school-club formal-professional others measured through self-
report validated
Outcomes Injuries and fatalities from a motorcycle crash (from hospital records and coronerrsquos office)
33Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3650
Hurt 1981 (Continued)
Notes Study conducted in Los Angeles USA
Jonah 1982
Methods Cohort study
Participants Initial sample size
Intervention group
2310 licensed MTP graduates who are active riders
Control group
2029 licensed motorcyclists who are active riders
Non-licensed and non-active riders excluded
Analysis sample size
Intervention group
811 interviews (35) 533 driver records (23)
Control group 1080 interviews (71) 905 driver records (45)
Interventions No training vs the non-mandatory Canada Safety Council Motorcycle Training Program (MTP) comprising of 4
hours theory and 16 hours practise (6 hrs on-road) over 2 weekends
Intervention integrity
Participants had all successfully completed the MTP
Outcomes Self-reported and police reported crashes and offencesLength of follow up 4 years
Notes Study conducted in Ontario Canada
Kloeden 1994
Methods Cohort study
Participants Initial sample size
Intervention group 2533 licence applicants
Control group Unclear
Sample of riders obtaining probationary licenceIntervention group 1625 (64)
Control group 5015 (unclear)
Non-active riders excluded from control group (~66)
Analysis sample size
Intervention group 631 (39)
Control group 318 (6)
Interventions No formal training vs the mandatory pre-licence Ridersafe course consisting of two levelsthe first is a training course
in the basics of motorcycle riding where a pass entitles the rider to obtain a learnerrsquos permit The second is a skill
test where a pass entitles the rider to obtain a probationary motorcycle licence The training course consists of two 4
hour training sessions on separate days involving instruction videos and actual riding on motorcycles The skill test
consists of a 4 hour preparation course of 74 minutes of lecturediscussion 38 minutes of video and 75 minutes of
34Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3750
Kloeden 1994 (Continued)
on-course riding which is the testing phase
Intervention integrity
Unclear what percentage in the Ridersafe group and the control group obtained their probationary licence halfway
and their full motorcycle licence in the end
Outcomes Police reported crashes
Length of follow up 1 year
Notes Study conducted in SA Australia
Kraus 1975
Methods Population based case-control study
Participants Cases All motorcycle riders with a confirmed medically treated motorcycle injury (including fatalities) from all
hospitals in study base and peripheral area hospitals and all deaths due to motorcycle crashes in the study base All
serious injuries (leading to hospitalisation) and fatalities likely to be included
Community controls A random sample of registered motorcycle owners in the same temporary and geographical
study base as the cases arose from
Initial sample size Cases 1273
Controls 738
Analysis sample size Cases 741 (58)
Controls 434 (59)
Interventions Motorcycle rider training (not specified) measured through self-report
Outcomes Injuries leading to hospitalisation fatalities
Length of follow up not specified
Notes Study conducted in Sacramento county USA
Leung 1987
Methods Non-randomised trial
Participants Initial sample sizeIntervention group unknown number of MRC graduates
Control group unknown number of licensed motorcyclists
Analysis sample size
Intervention group 2941
Control group 43094
Interventions No formal training vs non-mandatory Wisconsin motorcycle rider course a 22 hours off-street programme that
teaches the basic rules of safe motorcycling
Intervention integrity
50 in the MRC group and 100 in the control group obtained their motorcycle licence in the end
35Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3850
Leung 1987 (Continued)
Outcomes Police recorded crashes and offences
Length of follow up up to 3 years
Notes Study conducted in Wisconsin USA
MAIDS 2004
Methods Population based case-control study
Participants Cases Motorcycle riders who had a motorcycle crash Unclear whether all cases included
Community controls Motorcycle riders identified at petrol stations Unclear whether matched to cases
Initial sample size
Cases not reported
Controls not reported
Analysis sample size
Cases 921 (Italy 200 Netherlands 200 Spain 121 Germany 250 France 150)
Controls 923 (Italy 200 Netherlands 200 Spain123 Germany 250 France 150)
Interventions Pre-license rider training or additional training (non-compulsory)
Outcomes Crashes
Notes Study in five sampling areas located in five European countries (France Germany Netherlands Spain and Italy)
McDavid 1989
Methods Cohort study
Participants Sample size after first match
Intervention group 346 male training graduates
Control group 346 male motorcycle riders licensed in same year
Sample size after second match
Intervention group 139 (40)
Control group 139 (40)
Analysis sample size
Intervention group 139 (40)
Control group 139 (40)
Interventions No formal training vs the non-mandatory pre-licence British Columbia Safety Councilrsquos 37-hour motorcycle safety
programme
Intervention integrity
Participants had all completed and passed the training
Outcomes Police reported crashes injurious crashes
Length of follow up 1-6 years
36Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3950
McDavid 1989 (Continued)
Notes Study conducted in British Columbia Canada
Mortimer 1984
Methods Cohort study
Participants Initial sample size
Intervention group 608 MRC graduates
Control group 360 motorcycle riders
Response rates
Intervention group 360 (59)
Control group 324 (90)
Non-active riders excluded
Analysis sample size
Intervention group 213 (35)
Control group 303 (84)
Interventions No formal training vs the non-mandatory Motorcycle Safety Foundationrsquos Motorcycle Rider Course comprising of
8 hours theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-road practice
Intervention integrity
Participants had all completed and passed the MRC
Outcomes Self-reported crashes and offencesLength of follow up 1 year
Notes Study conducted in Illinois USA
Mortimer 1988
Methods Cohort study
Participants Initial sample size
Intervention group 2307 MRC graduates
Control group 556 active licensed motorcycle riders
Response ratesIntervention group 1298 (56)
Control group 500 (90)
Non-active riders excluded
Analysis sample size
Intervention group 913 (40)
Control group 500 (90)
Interventions No formal training vs the non-mandatory Motorcycle Safety Foundationrsquos Motorcycle Rider Course comprising of
8 hours theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-road practice
Intervention integrity
Participants had all completed and passed the MRC
37Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4050
Mortimer 1988 (Continued)
Outcomes Self reported crashes and offences
Length of follow up 1 year
Notes Study conducted in Illinois USA
Osga 1980
Methods Cohort study
Participants Initial sample sizeIntervention group 1163 MSF MRC participants invited 818 (74) responded
Control group 329 motorcycle riders responded to recruitment strategies
Non-active riders excluded from analysis
Analysis sample size
Intervention group 595 (73)
Control group 329 (100)
Interventions No formal training vs the non-mandatory Motorcycle Safety Foundationrsquos Motorcycle Rider Course (MRC) com-
prising of 8 hours theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-road
practice
Intervention integrity
Participants had all successfully completed the MRC
Outcomes Self-reported crashes police recorded crash reports
Length of follow up 6 months - 3 12 years
Notes Study conducted in South Dakota USA
Perrino 2002
Methods Cohort study
Participants Initial sample size
Intervention group 2144 (unclear) MRC graduates (motorcycle licence applicants)
Control group 2601 (unclear) regular licensing test graduates (motorcycle licence applicants)Analysis sample size
Intervention group 1070 (unclear)
Control group 867 (unclear)
Interventions No formal training vs the mandatory (for those younger than 18 years) pre-licence Maryland Motorcycle Rider
Course (MRC) consisting of 8 hours theory and 12 hours practice
Intervention integrity
No information given
Outcomes Self-reported and police recorded crashes and offences
Length of follow up 1 year
38Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4150
Perrino 2002 (Continued)
Notes Study conducted in Maryland USA
Raymond 1979
Methods Cohort study
Participants Initial sample size
Intervention group
258 volunteers mainly (but not all) novice riders
Control group
1104 provisional licence applicants
Analysis sample size
Intervention group 190 (74)
Control group 627 (57)
Interventions No formal training vs the non-mandatory RACACU (Royal Automobile ClubAuto-Cycle Union) training scheme
The training program consisted of either 24 one-hour sessions or 13 two-hour sessions Twelve hours of the training
focused on theory about basic operational procedures traffic laws and highway code The practical session focused
on basic operational skills riding on a training area and riding in traffic with an instructor A proficiency test was
administered at the end of the course
Intervention integrity
No information given
Outcomes Self-reported crashes (including minor non-injury non-damage and off-road crashes)
Length of follow up 1 year
Notes Study conducted in the UK
Satten 1980
Methods Cohort study
Participants Initial sample size
Intervention group interviewers kept interviewing until they had 100 riders selected from a population of 891
enrollees in the MSF MRC course (unknown how many were asked and how many refused to participate)Control group unknown number of registered motorcycle riders
Non-active riders excluded from analysis
Analysis sample size
Intervention group 71
Control group 100
Interventions No formal training vs the non-mandatory Motorcycle Safety Foundationrsquos (MSF) Motorcycle Rider Course (MRC)
comprising of 8 hours theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-
road practice
Intervention integrity
Participants had all successfully completed the MRC
39Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4250
Satten 1980 (Continued)
Outcomes Self-reported crashes injuries and offences
Length of follow up 3 years
Notes Study conducted in Illinois USA
Savolainen 2007
Methods Cohort study
Participants Initial sample size
Intervention group 4000 riders that had conducted the MSF MRC course (passed or failed)
Control group 4000 motorcycle owners or permit or endorsement holders from the Indiana Bureau of Motor
Vehicles database
Analysis sample size
Intervention group 739 (18)
Control group 588 (15)
Interventions Intervention integrity
No information given
Outcomes Probability of crash involvement
Length of follow up up to 5 years
Notes Study conducted in Indiana USA
Swaddiwudhipong 1998
Methods Randomised trial
Participants Initial sample size
Intervention group
all 16 villages of 3 sub-districts
(participants were practising motorcyclists licensed and unlicensed)
Control group
18 villages of 3 other sub-districtsAnalysis sample size
Intervention group 3 districts (100)
Control group 3 districts (100)
Interventions No rider education vs non-mandatory community based rider education consisting of theory on motorcycle injury
risk helmet use traffic laws vehicle regulations and traffic signs and a written and skill test leading to a licence
Intervention integrity
No information given
Outcomes Injuries fatalities
Length of follow up 2 years
40Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4350
Swaddiwudhipong 1998 (Continued)
Notes Study conducted in Tak Province rural Thailand
Waller 1992
Methods Cohort study
Participants Initial sample size
Intervention group 349 riders who succesfully completed the Department of Public Safety motorcycle operator
training in 1988
Control group 349 motorcyclists who never completed the DPS motorcycle operator training
Analysis sample size
Intervention group 349
Control group 349
Interventions No DPS motorcycle operator training vs non-mandatory DPS motorcycle operator training The DPS training is
the Motorcycle Safety Foundationrsquos Motorcycle Rider Course comprising eight hours of theory (lectures discussions
films and multiple choice tests) a textbook and 12 hours off-road practice
Intervention integrity
All 349 riders succesfully completed the rider training
Outcomes Police recorded crashes
Length of follow up 3 years
Notes Study conducted in
Characteristics of excluded studies [ordered by study ID]
Study Reason for exclusion
Anderson 1980a Study type review and process evaluation of a rider course
Aupetit 2008 Outcome proposal of a rider training curriculum that combines traditional training and a motorcycle riding
simulator
Awane 1999 Study type review on use of simulators in rider and driver training
Baldi 2005 Intervention not rider training but intervention integrity of rider training availabilty of rider education
implementation of graduated licensing system etc on a state-wide level
Batchler 1988 Study type description of a training course
Blanchard 2006 Study type not a study assessing the association between rider training and outcomes
Bonnett 2005 Outcome qualitative data regarding satisfaction with the training programme
41Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4450
(Continued)
Braver 2007 Outcome licensure rates
Clayton 1980 Outcome knowledge skills attitude
Council 1978 Outcome crash reduction in ANY vehicle not motorcycles exclusively
Crowther 2005 Study type qualitative study
Daniello 2009 Study type review of the effectiveness of motorcycle training and licensing
Dewanti 2007 Intervention emotional characteristics behaviour (ie not motocycle rider training)
Du Plooy 1990 Study type review of types of rider training courses in South Africa
Engel 1989 Study type uncontrolled before-after study
Goldenbeld 2002 Outcome skills
Goldenbeld 2004 Outcome knowledge and skills
Haworth 2000 Intervention balance between attitudinal andvehicle control skills acrosscurriculla of motorcycle rider training
courses
Haworth 2005 Study type review
Haworth 2006 Study type cross-sectional survey
Hill 1983 Study type review and description of training system
Imatake 1980 Study type course descriptionreview
Ingebritsen 1990 Intervention risk factors for motorcycle crashes
Jain 2009 Study type cross-sectional descriptive study of motorcycle crashes over a 5 year period
Jonah 1979 Study type cross-sectional survey Outcome scores on motorcycle operator skill test
Jonah 1981 Intervention evaluation of a motorcycle rider skill test not motorcycle rider training
Kadar 1990 Study type review of components of rider training courses
Kelsey 1986 Intervention evaluation of a motorcycle rider skill test not motorcycle rider training
Koch 1980 Study type reviewcourse description
Koch 1990 Study type description of course content
42Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4550
(Continued)
Labbett 2005 Study type essay about possible explanations for motorcycle deaths
Law 2005 Intervention targeted motorcycle safety programme not specifically rider training
Lowes 1990 Study type review and descriptions of rider training
Mahon 1976 Study type review of motorcycle rider training methods
Maxwell 1985 Outcomesrecommendations forprogramme operatorsand otherson howthe motorcycle trainingprogramme
works and how it can be improved
Mayhew 1996 Study type review
McKnight 1987 Study type evaluation of methodological and administrative aspects of a motorcycle operator skill test
McPherson 1976 Outcome knowledge and skills
Mohan 2004 Study type review of evidence of interventions to reduce road traffic injuries
Morgan 1980 Study type reviewcourse description
Motoki 1990 Study type proposal and development of rider training curriculum
Nairn 1992 Study type literature review
Newland 1980 Study type description of development of motorcycle safety course
Newland 1982 Study type descriptionreview of a rider training course
Newland 1990 Study type uncontrolled BeforeAfter study
Ormston 2003 Outcome changes in attitude and self reported behaviour
Osga 1979 Outcome knowledge attitudes and skills
Panou 2007 Intervention new simulation tool to teach driverrider
Peterson 1990 Study factor demographic details of course participants completing the course vs those not completing the
course
Prem 1984 Study type experimental
Study factor examines association between skills and performance in tests
Radin Umar 2006 Intervention motorcycle safety programme not specifically rider training
Rockwell 1990 Study type methodologically flawed study type
43Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4650
(Continued)
Rothe 1987 Study type qualitative study of rider training in general
Rowden 2007 Study type qualitative study
RSC Victoria 1993 Study type review of motorcycle rider training and education in Victoria
Russam 1979 Study type review of training programmes in Japan
Saffron 1981 Study type literature review
Schulz 1990 Study factor predictors of motorcycle crashes
Sexton 2004 Study type cross-sectional analytic analysis
Shephard 1986 Study factor reviewdescription of programme
Simpson 1990 Study type review of motorcycle rider trainingeducation
Sokolov 1990 Study factor review of motorcycle riding courses in USSR
Steffens 1988 Outcome self-reported behaviour experiences and researcher observed riding behaviour
Sudlow 2003 Study type review of motorcycle training in the UK
Tasmania 1995 Study type cross-sectional descriptiveprocess evaluation
Tetard 1985 Study type experimental
Outcome detection of indices
Thackray 1980 Outcome instructional effectiveness user acceptance administrative feasibility
Thompson 1994 Study type cross-sectional survey
Outcome attitudes and knowledge
Tobler 1990 Study type review of course curricula
Transport 1989 Intervention syllabus for rider training
Tuan 2007 Study type cross-sectional study (survey) which does not assess the effects of an intervention but examines
motocyclistsrsquo behaviour
Victor 1980 Study type review
Waterer 1990 Study type review of UK rider training issues
Weller 1989 Study type review of studies evaluating motorcycle safety and education measures
44Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4750
(Continued)
Wells 1981 Study type description of different methods of training
Winn 1986 Study type review of rsquohistoryrsquo of rider training
Wisher 1988 Study type cross-sectional survey
Wood 1987 Study type Interrupted Time Series but did not meet the criteria for an ITS as published by the Cochrane
Effective Practice and Organisation of Care Group
Yamazaki 1990 Study type cross-sectional survey
Outcome riding skills skills retention and attitude
Characteristics of ongoing studies [ordered by study ID]
Burgess 2010
Trial name or title UKs National Ride program
Methods Cohort study
Participants Motorcycle offenders
Interventions Police led diversion program
Outcomes Self reported attitudinal and behavioural measures
Starting date
Contact information
Notes Expected to be completed in early 2010
45Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4850
D A T A A N D A N A L Y S E S
This review has no analyses
A P P E N D I C E S
Appendix 1 Search strategy
We searched the following electronic databases
bull Cochrane Injuries Group Specialised Register (searched 22 Sept 2008)
bull CENTRAL (The Cochrane Library 2008 Issue 3)
bull TRANSPORT 1988 to 200706 (latest date available)
bull ATRI Australian Transport Index (to Sept 2008)
bull MEDLINE (Ovid SP) 1950 to Sept 2008
bull EMBASE (Ovid SP) 1980 to Sept 2008
bull CINAHL (EBSCO) 1982 to Sept 2008
bull WHOLIS (World Health Organization Library Information System) (to Sept 2008)
bull PsycINFO (Ovid SP) 1806 to Sept 2008
bull LILACS (BIREME) (Latin American and Caribbean Health Sciences) (1982 to Sept 2008)
bull ISI Web of Science Social Sciences Citation Index (SSCI) 1970 to Sept 2008
bull ERIC (searched 25 Sept 2008)
bull ZETOC (searched 25 Sept 2008)
bull SIGLE (searched 25 Sept 2008)
All of the databases were searched using the following text-word search strategy The strategy was adapted using either rsquorsquo or rsquo$rsquo as atruncation symbol
1 motor-cycl or motorcycl or motor-bik or motorbik or motor-scooter or motorscooter or scooter or moped or motocycl
or motocicl or moto
2 train or educat or instruct or program or course or qualif or learn or improv
3 1 and 2
H I S T O R Y
Protocol first published Issue 2 2005
Review first published Issue 10 2010
46Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4950
C O N T R I B U T I O N S O F A U T H O R S
Katina Kardamanidis was the first author of the protocol conducted the literature search extracted full studies assessed the studies
according to inclusion criteria assessed quality of the studies compared findings with the second author discussed disagreements with
the third author wrote the drafts and the final review
Rebecca Ivers was the second author of the protocol assessed the studies according to inclusion criteria assessed quality of the studies
compared findings with the first author edited the final draft and provided overall guidance for the process
Alexandra Martiniuk assessed quality of the studies compared findings with the first author discussed disagreements with the third
author and helped write the first draft and the final review
Mark Stevenson contributed to the protocol reviewed methods of studies and contributed to the final draft
Katrina Thistlethwaite conducted a second literature search extracted full studies and contributed to the final draft
D E C L A R A T I O N S O F I N T E R E S T
None known
S O U R C E S O F S U P P O R T
Internal sources
bull No sources of support supplied
External sources
bull National Health and Medical Research Council of Australia Australia
Salary support for Martiniuk Ivers and Stevenson
bull University of Sydney Sesqui Research Grant Australia
Direct research costs and salary support for Kardamanidis
D I F F E R E N C E S B E T W E E N P R O T O C O L A N D R E V I E W
There are no differences between the protocol and the review
I N D E X T E R M S
Medical Subject Headings (MeSH)
lowastMotorcycles Accident Prevention [lowastmethods] Accidents Traffic [lowastprevention amp control] Licensure Program Evaluation Wounds
and Injuries [mortality lowastprevention amp control]
47Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 5050
MeSH check words
Humans
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 1950
a similar study design was more likely to have included all cases
but this is not entirely clear from the study reported It is unclearfor example whether fatalities were included as they did not use
coronerrsquos reports for the recruitment of cases The MAIDS 2004
study was conducted in five European countries but it was unclear
how cases were recruited whether they were all cases during the
two years that the study lasted or a random sample of all cases
what the response rates were etc Controls were recruited from
randomly selected petrol stations but it was likewise not clear what
the eligibility criteria for inclusion in the study were how they
were recruited (eg randomly) and what the response rate was
Hurt 1981 and Haworth 1997 both matched controls to cases by
recruiting themat the sametime ofdayandsamedayof weekas the
crash occurred Hurt 1981 however recruited controls two years
later andit is likelythat there would have beendifferences betweenthe two comparison groups One important difference was that
during data collection for cases there was very little specialised
motorcycletrainingavailable in the studyregionwhereas two years
later when control data were collected there was some training
available
These four case-control studies also suffered from selection bias
due to potential differences between those that gave consent to
participate in the study and responded to interviews andquestion-
naires and those that did not None of the studies have assessed
such differences although Hurt 1981 did acknowledge that con-
trols giving consent to cooperate may have differed from those
motorcyclists who did not
Detection bias
Measures that can be taken to reduce detection bias in case-con-
trol studies are blinding of interviewers for case or control status
blinding of the participant for the study factor of interest validat-
ing measurement instruments and using these in an identical way
for cases and controls and cross checking self-reported informa-
tion with other more objective sources such as registers None of
the studies reported blinding of the intervention assessors to case
or control status but all used consistent measurement methods for
both case and control groups
All case-control studies used self-report as their main measure toassess whether cases and controls had received rider training and
if so the type of training The major potential source of bias in
case-control studies when self-report is used as a measure is recall
bias One way of reducing recall bias is by double-checking self-
reported data with more objective data Kraus 1975 also used po-
lice accident reports for information on rider characteristics but
such information was not available about rider training Haworth
1997 and MAIDS 2004 did not report on any measures to limit
recall bias Hurt 1981 reported further questioning of cases by
the interviewer when answers did not make sense or when par-
ticipants seemed to contradict themselves as part of the in-depth
unstructured interview technique This qualitative research tech-
nique can result in valuable information but as interviewers were
not blinded to case or control status of the participants it may have led to more thorough examination of cases compared to the
controls and thus differential detection bias
Intervention integrity
The level of detail about the rider training provided in the case-
control studies differed significantly but was minimal for each
Kraus 1975 simply asked whether lsquoanyrsquo motorcycle rider training
had lsquoeverrsquo been carried out Hurt 1981 asked participants whether
they hadbeen self taught informally taught or completed formally
taught rider training courses MAIDS 2004 asked whether partic-
ipants had had pre-license training additional (non-compulsory)
training other training or no training whereas Haworth 1997asked about the number of courses that a participant had com-
pleted whetherthey were basic intermediate or advanced courses
and how long ago the training had been done None of the case-
control studies reported collecting information about further de-
tails about any training programmes (eg type of training cur-
riculum duration of training)
Analysis
Kraus 1975 accounted for confounding factors through the exclu-
sion of females and stratification by age and frequency of motor-
cycle use Hurt 1981 matched cases and controls for time of day
and day of the week that the crash occurred and did not adjust forany confounding factors in the analysis Haworth 1997 matched
groups for time of day and day of the week that the crash occurred
and stratified by age blood alcohol level and licence status in the
analysis None of the studies used multivariate methods MAIDS
2004 did not provide any information about any participant char-
acteristics other than the study exposures of interest and did not
describe any methods to adjust for potential confounding factors
Effects of interventions
We have presented the results in a descriptive manner due to the
substantial heterogeneity in study types interventions and out-comes and the generally poor quality of the studies We have
grouped the interventions by mandatory and non-mandatory pre-
licence rider training and post-licence rider training and further
by study type and by outcome (crashes injurious and fatal crashes
and offences) We discuss results that are reported as statistically
significant (whether a P value was reported or not)
Pre-licence rider training - mandatory
Randomised trials
17Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2050
Crashes
Total group of participants
Anderson 1980 found a reduction in motorcycle crash rates in
the intervention group (MOST skill test with remedial training)
compared to the control group (who received the skill testwithout
a training component) The percent differenceranged from 143
six months after the programme (P lt 005) to 211 (P lt 0001)
after one year and 127 (P lt 001) after two years Hall 1988
found no statistically significant difference in crash rates between
intervention groups at any point in time
Subgroup of participants who obtained their licence only
Anderson 1980 found a reduction in motorcycle crash rates in the
intervention arm but this was statistically significant at one year
only (24 to 218 reduction P lt 005) Hall 1988 found no
statistically significant differences
Offences
Total group of participants
The percentage of offences in Anderson 1980 after six months
and one and two years was slightly higher in the intervention
arm compared to the control group but these differences were
not statistically significant Likewise the study by Hall 1988 did
not find a statistically significant difference in offences between
groups
Subgroup of participants who obtained their licence only
Hall 1988 found no significant difference in offence rates between
the intervention and control groups Anderson 1980 found thatthe percentage of offences increased in the intervention arms com-
pared to the control group This difference was statistically signif-
icant after two years (51 to 115 increase P lt 005)
Cohort studies
Crashes
Neitherof the cohort studies (Kloeden1994 and Billheimer1998)
found a statistically significant difference in crash rates between
the intervention and control groups
Offences
Billheimer 1998 (for the California Motorcyclist Safety Program
course) and Perrino 2002 did not find a statistically significant
difference in offence rates between the intervention and control
groups
Case-control studies
Crashes
MAIDS 2004 found that a slightly higher percentage of controls
(not involved in a crash) had had pre-license rider training than
cases (involved in a crash) This difference was not statistically significant (50 of controls vs 477 of cases OR 091 95 CI
076 to 110)
Pre-licence rider training - non-mandatory
Randomised trial
Injuries and deaths
Swaddiwudhipong 1998 found a significantly lower injury ratein the intervention villages (1051000 population 95 CI 92
to 120) compared to the control villages (1691000 population
95 CI 152 to 185) one year after the intervention This differ-
ence was not maintained two years after the intervention (127
95 CI 112 to 143 intervention 151 95 CI 135 to 167
control) There was no significant difference in death rates be-
tween groups
Cohort studies
Crashes
Cooper 1988 foundno statistically significant differencein crashes
between the intervention and control groups McDavid 1989
found a lower crash rate in the intervention group (14139) com-
pared to the control group (23139) but no test of statistical sig-
nificance was undertaken
Injuries and deaths
McDavid 1989 found that the intervention group had fewer in-
jurious crashes (814 crashes) compared to the control group (16
23 crashes) but no statistical significance test was reported
18Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2150
Offences
Cooper 1988 foundno statistically significantdifference in offencerates between the intervention and control groups McDavid 1989
foundno statistically significantdifference in offence ratesbetween
the intervention and control groups
Non-mandatory rider training
Non-randomised trials
Crashes
Neither Adams 1985 nor Leung 1987 found a statistically signif-icant effect of rider training on crash rates
Offences
Leung 1987 found a statistically significant lower offence rate in
the intervention group compared to the control group for both
males and females and for all age groups except for women 35
years or older (P values not reported)
Cohort studies
Crashes
Osga 1980 in the univariate analysis reported a lower percentage
of crashesin theintervention group (that continued to ride) (28
168595) than in the control group (34 112329) In the cu-
mulative crash rate analysis however Osga1980 found statistically
significant (at the P lt 001 level) higher crash rates in the interven-
tion group than in the control group at allmileage intervals (1000
to 29999 miles) travelled After riding more than 30000 miles
(when over 50 of the trained riders had been involved in a crash
compared to approximately 40 in untrained riders) this differ-
ence was no longer statistically significant Satten 1980 found a
lower percentage of crashes in the intervention group (13 969)
compared to the control group (25 2079) but the differencewas not assessed statistically Davis 1997 found a lower crash rate
in trained riders but it was unclear how this was calculated be-
cause the size of the control group was not reported Jonah 1982
and Billheimer 1998 (for the Experienced Rider Course) found no
statistically significant difference in crashes between the interven-
tion and control groups Although the difference between those
who completed the Experienced Rider Course and no course in
Billheimer 1998 neared significance at six months after the course
(P = 007) large P values were observed for one year after (P =
030) Raymond 1979 found a statistically significant higher crash
rate for the intervention group (1861000 miles ridden) com-
pared to the control group (1181000 miles ridden) in both males
and females (P lt 0001) especially during the first 5000 miles of
riding Waller 1992 found a statistically significant higher crashrate for the intervention group (27 crashes in three years or 070
crashes10000 person days) compared to the control group (5
crashes in three years or 013 crashes10000 person days) with a
crude relative risk of 538 (X 2=1513 plt005) Mortimer 1984
and Mortimer 1988 found no statistically significant difference in
crashes between the intervention and control groups
Savolainen 2007 found that the intervention group (those who
completed the Basic Rider Course) was 44 more likely to be
involved in crashes than the control group and those who com-
pleted the course multiple times were an additional 180 more
likely to be involved in a crash This finding was based on low
numbers and no P value or confidence level was given for either
result
Injuries and deaths
Satten 1980 also found a lower percentage of injurious crashes
(33 39) in the intervention group compared to the control
group(60 1220) although again no statistical significance test
was reported
Offences
Jonah 1982 did find a significantly lower self-reportedoffence rate
in the intervention group (15) compared to the control group(34 P lt 001) (no numbers given) Satten 1980 also found
a lower self-reported offence rate in the intervention group (1
71 versus 19100) but again this was not tested for statistical
significance Mortimer 1984 Mortimer1988 and Billheimer 1998
found no statistically significant differencein offence ratesbetween
the intervention and control groups
Case-control studies
CrashesMAIDS 2004 found that a slightly higher percentage of cases
(involved in a crash) had had additional non-mandatory rider
training than controls (not involved in a crash) This difference
was not statistically significant (17 of cases vs 12 of controls
OR 146 95 CI 064 to 337)
Rider training unspecified
Case-control studies
19Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2250
Injuries and deaths
Hurt 1981 found that the odds of injurious crashes were lower inthose who had received professional training compared to those
who had not Kraus 1975 found that in the younger age group
(less than 20 years) and the older age group (over 35 years) a larger
percentage of injured riders (12 and 21 respectively) reported
prior motorcycle rider training compared with the uninjured con-
trol group (6 and 5 respectively) (P lt 001) (no numbers
given) Haworth 1997 reported several different outcomes one
beingthat having completed at leastone trainingcoursecompared
to no training courses at all had no significant effect on the odds
of crashing (afteradjustment for age blood alcohol level or licence
status (learner permitprobationaryfull)) When comparing three
different levels of training courses (beginner intermediate and ad-
vanced) with no training no significant differences were found
in the odds of crashing However after adjustment for age and
blood alcohol level completing an advanced course as compared
to a beginners course was associated with a significant decrease in
the odds of severe motorcycle crashes resulting in serious injury or
death (OR 02 95 CI 01 to 08)
D I S C U S S I O N
The objective of this review was to quantify the effectiveness of
pre- andpost-licence motorcycle rider training on the reduction of
traffic crash involvement injuries deaths and offences committedby motorcycle riders Most included studies were cohort studies
There were a few randomised and non-randomised trials and case-
control studies
All studies compared either mandatory or non-mandatory formal
training with no formal training except for Haworth 1997rsquos case-
control study that compared levels of training (beginner interme-
diate and advanced) to each other
We found studies that evaluated the effectiveness of manda-
tory pre-licence training in reducing crash and offence rates
(Anderson 1980 Hall 1988 Kloeden 1994 Billheimer 1998
(the basic California Motorcyclist Safety Program component of
the study) Perrino 2002) non-mandatory training that when
passed resulted in a licence (Cooper 1988 McDavid 1989
Swaddiwudhipong 1998) non-mandatory training that could be
taken by motorcycle riders with or without a licence (Raymond
1979 Osga 1980 Satten 1980 Jonah 1982 Mortimer 1984
Adams 1985 Leung 1987 Mortimer 1988 Waller 1992 Davis
1997 Billheimer 1998 (the Experienced Rider Course compo-
nent of the study)) and unspecified rider training on crash injury
and death rates (Kraus 1975 Hurt 1981 Haworth 1997 MAIDS
2004)
The studies provided conflicting evidence regarding motorcycle
rider training which is likely due to the poor design quality and
reporting of the studies Most of the studies included in the review
were likely to be adversely affected by significant selection biasalong with substantial detection bias(measurement and recall bias)
which limits our ability to make conclusive comments
Mandatory pre-licence rider training
There were two RCTs three cohort studies and one case-control
study that examined pre-licence ridertraining The two RCTs eval-
uated the Motorcycle Skills Operating Test (MOST) (Anderson
1980 Hall 1988) Anderson 1980 found a statistically significant
reduction in crashes in the intervention group at six months one
and twoyearsfollow up However when only participantswho ob-
tainedtheirlicence were included in theanalysis thereduction wasstatistically significant only at one year follow up The other RCT
(Hall 1988) found no statistically significant differences in crash
rates at any point of time neither in the total population group
nor in the licensed subgroup Neither of the two cohort stud-
ies assessing crash rates found a statistically significant difference
in crash rates between intervention and control groups ( Kloeden
1994 Billheimer1998) andthe case-control study (MAIDS 2004)
also did not find a statistically significant difference in the odds of
experiencing a crash between trained and untrained riders With
regards to offences neither of the RCTs found a statistically sig-
nificant difference in offence rate between the intervention and
control groups in the analysis of the total group of participants
entering the trial However Anderson 1980 found a statistically significant increase in offence rates in the intervention group but
only after two years of follow up andonly in the analysis of partici-
pants who obtained their licence The two cohort studies assessing
offence rates found no statistically significant difference between
the groups (Billheimer 1998 Perrino 2002)
None of the five studies (RCTs or cohort) found a reduction in
crashes or offences following mandatory pre-licence rider training
except for Anderson 1980 who observed a statistically significant
reduction in crashes in the intervention group at one year in both
analyses It is not clear why both RCTs that evaluated the same
MOST and training intervention observed different results Both
studies enrolled large samples and observed a similar uptake of
licensure following the intervention It is possible that differencesmay have arisen due to different study samples (California ver-
sus New York) or that the intervention delivery differed between
studies
In both trials of the Motorcycle Skills Operating Test including
rider training only a third of motorcycle licence applicants ob-
tained their motorcycle licence It is possible that a licensing pro-
cess including a skills test may reduce crashes through decreased
exposure
Non-mandatory pre-licence training
20Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2350
Three studies assessed the effect of non-mandatory pre-licence
training The Cooper 1988 cohort study found no difference incrash or offence rates between the trained and untrained groups
We cannot however be confident about this result as the study
sufferedfrom selectionbiaswhichcouldhave biasedthe measure of
effect toward the null The RCT conducted by Swaddiwudhipong
1998 examined injury and death rates and found a statistically
significant lower injury rate in the intervention villages after one
year of follow up but no longer at two year follow up There
was no statistically significant difference in death rates between
intervention and control villages Our confidence in the results of
this study are hindered by the inadequacy of the randomisation
process that led to significant differences between the intervention
and control districts on two important variables The intervention
districts were more likely than the control districts to have a validlicence andto wear a helmetSincethese differences were nottaken
into account in later analyses they could explain the differences
observed in injury rates between districts Furthermore with only
six districts randomised the sample size was probably too small
to detect a difference between intervention and control groups
had there been one The McDavid 1989 cohort study found a
lower percentage of injuries in the intervention group although
a statistical test was not conducted McDavid 1989 drew their
intervention and control groups from different populations but
did adjust for confounders although they only analysed 40 of
the initial sample (139 trained and 139 untrained) Because of
these substantial weaknesses we are unable to draw conclusions
from this studyAn overall effect of pre-licence training on the reductionof crashes
injuries deaths and offences was not found although one should
be aware that the research of this issue to date has had significant
flaws However a mandatory skill test (such as the MOST) in
combination with and even without pre-licence training as part
of a licensing process seems to reduce the number of individuals
completing the licensing procedure and therefore indirectly re-
duces crash rates through a reduction in exposure to motorcycle
riding Making licensure a more difficult process may therefore be
seen as advantageous if it delays licensure and therefore reduces
exposure It is possible that the group who chose not to continue
riding a motorcycle would not have been the group to experience a
crash or commit an offence if they had continued to ride anyhowfor example because they are more safety conscious and this was
the reason for their giving up riding in the first place If this is the
case deterring individuals from completing their licence and rid-
ing may not necessarily decrease overall crashes deaths or offence
rates In addition other potentially adverse effects of complicat-
ing a licensing procedure need to be considered such as increases
in unlicensed riding In Thailand non-mandatory rider training
offered as a community intervention may have decreased injury
rates but this needs to be confirmed through further research
Non-mandatory training
The majority of the studies assessed non-mandatory training
(training independent of any licensing procedure) and the results
were varied The two non-randomised trials assessing crash rates found no statistically significant difference in crash rates between
the intervention and control groups (Adams 1985 Leung 1987)
Of the nine cohort studies Satten 1980 found a lower percentage
of crashesin the intervention group although it is unclear whether
this was a statistically significant result Davis 1997 also found a
lower percentage of crashesin theinterventiongroup although it is
unclear with which control group the intervention group wascom-
pared Raymond 1979 Osga 1980 Waller 1992 and Savolainen
2007 found a statistically significant higher crash rate in the in-
tervention group and the remaining cohort studies (Jonah 1982
Mortimer 1984 Mortimer 1988 Billheimer 1998) found no sta-
tistically significant difference in crash rates between intervention
and control groups No conclusions can be drawn from these stud-
ies with confidence The Satten 1980 and Davis 1997 studies did
not adjust for any confounding factors even though important
differences existed between the intervention and control groups
Although Raymond 1979 and Waller 1992 adjusted for a num-
ber of confounders at baseline (Waller 1992) and in the analy-
sis (Raymond 1979 Waller 1992) the intervention and control
groups were drawn from such different populations that it is pos-
sible that unmeasured confounders were responsible for the sta-
tistically significant increased risk of crashes observed in the inter-
vention group The observed difference could also be attributed totheobserved differential loss to followup (Raymond 1979) which
could bias the measure of effect towards or away from the null
The Osga 1980 study found a protective effect of rider training
in the univariate analysis but statistically significant higher crash
rates in the intervention group when accounting for riding expo-
sure Sex (a significantly higher proportion of females in the in-
tervention group) and other potentially confounding factors that
could explain thedifferencebetween the groups were howeverun-
accounted for Finally Savolainen 2007 did control for potential
confounding variables yet it is not clear whether variables were
controlled for simultaneously or one at a time One case-control
study (MAIDS 2004) found no statistically significant reduction
in the odds of experiencing a crash between trained and untrainedriders but the quality of this study was poor mainly because the
analysis did not account for any confounding factors
One cohort study (Satten 1980) assessed the effect of non-manda-
tory rider training on injurious crashes and found a lower percent-
age of injuries in the intervention group although a statistical test
was not conducted As mentioned above the Satten 1980 studyrsquos
validity is reduced due to its lack of adjustment for confounding
factors
The only non-randomised trial that assessed offence rates (Leung
1987) found a statistically significant lower offence rate in the in-
tervention group and of the five cohort studies that assessed of-
fence rates (Satten 1980 Jonah 1982 Mortimer 1984 Mortimer
21Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2450
1988) one study (Jonah 1982) also found a statistically significant
lower offence rate in the intervention group However we have lit-tle confidence in the results from Leung 1987 since potential con-
founding factors were not appropriately addressed Jonah 1982
did conduct a multivariate analysis but was hindered by an initial
low response rate The other four cohort studies did control for
several potential confounders and found no difference in offence
rates following non-mandatory post-licence rider training
The non-randomised trials and cohort studies evaluating non-
mandatory rider training were at high risk of substantial selection
bias due to the non-random allocation of participants as well
as low participation rates and attrition bias and could therefore
not provide conclusive evidence with regard to the effect of the
training It should be noted that despite the biases and other issues
with this group of studies most didnot detectmajor effects relatedto rider training in either direction This may suggest that any
potential impact of rider training will be modest
Studies examining unspecified modes of ridertraining
Three case-control studies assessed whether injured or deceased
motorcyclists were more or less likely to have had rider training
compared to control motorcyclists All had conflicting results
Hurt1981 foundthe odds of havingan injuriouscrash waslower in
those who had received rider training whereas Kraus 1975 found
that the odds of having an injurious crash were higher in youngerand older trained riders Haworth 1997 who compared advanced
versus beginner training found no effect on motorcycle injuries
or deaths although they did find that those who had completed
advanced training had reduced chances of crashing compared to
those who had done a beginnersrsquo training course Similar issues
with uncontrolled confounding and biases of selection as described
above arose with the three case-control studies Hurt 1981 suffers
from inadequate control of confounding via poor matching and
both the Kraus 1975 and the Haworth 1997 findings were only
controlled for limited factors by means of stratified analysis It is
likely that these limitations are the reason for the contradictory
directions of effect observed for the three case-control studies
Methodological weaknesses of the studies
In summary most studies investigating the effectiveness of mo-
torcycle rider training on reducing crashes injuries and death suf-
fered from serious methodological weaknesses
The majority of studies were non-randomised and were there-
fore limited mainly by selection bias Although the approaches of
matching stratification or regression analyses can minimise bias
(Fleiss 2004) they were poorly employed in the reviewed studies
The inadequate adjustment for confounders in particular adds
to our lack of confidence in the studiesrsquo results since it has been
demonstrated that individuals who choose to take motorcycle rider
training differ from those not taking rider training For instance
trained riders often differ from controls in age sex and experi-ence usually ride less and may have a more positive attitude to
safety as indicated by the reported more frequent wearing of seat
belts (Mortimer 1982)
The two large randomised trials reported that a large proportion
of their study group were never exposed to the risk of offences or
crashes because they never attained licensure This made it impos-
sible to assess whether a reduction in crashes in the intervention
group was a result of training or due to delayed or never attained
licensure
In some studies the sample sizes were too small to detect any sta-
tistically significant differences Follow-up time after training dif-
fered between studies and was an impediment to observing any
real effects of training if they existed As riding experience ac-cumulates over time any differences in outcomes between those
trained and untrained would diminish in studies with lengthy fol-
low-up times This dilution of any possible effect of rider training
may have also occurred in studies that mixed both novice and ex-
perienced riders in the study groups since it is hypothesised that
novice riders serve to benefit most from rider training This would
increase the difficulty in finding any effect of rider training par-
ticularly when sample sizes are small
Detection bias may also account for some of the differences ob-
served between trained and untrained riders Detection bias would
have occurred if the self-report hospital driver or police records
used to record the outcomes of crash injury death or offences
differed between the trained and untrained riders Research hasdemonstrated that not all crashes are reported to police for ex-
ample small crashes or off-road crashes (Kraus 1975) even serious
injury crashes have been found to be underreported (Anderson
1980) However this would only introduce detection bias if the
reporting of outcome differs for trained and untrained riders This
does appear possible in that responsible riders may be more likely
to enroll in a rider training course and also may be more likely
to report crashes to police than irresponsible riders meaning that
the number of crashes from riskier riders could be underrepre-
sented on records An analysis to investigate the effectiveness of
rider training may then show that rider training is not effective in
reducing crashes
Rather than using reported outcomes some studies used self-re-ported survey data although this technique suffers from non-
response biases In fact two studies reported differences in de-
mographic characteristics in response rates to questionnaires
(Anderson 1980 Hall 1988) Self-reported data also introduce
the risk of recall bias survival bias and social desirability response
bias Recall bias may have been equally distributed among trained
and untrained riders howeverit is hypothesised that trained riders
may have felt it was lsquoexpectedrsquo to report good driving histories
hence leading the rider training to appear effective
22Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2550
Limitations of this review
Limitations of this review were that we found very few recentstudies (the two most recent ones published in 2002 and 2007)
and that we found very few studies conducted in low-and middle
income countries even though our search strategy was set up to
identify such studies Anotherlimitation wasthat forthe reviewed
studies we could not assess all outcomes of rider training that are
potentially useful Some of the included studies assessed outcomes
such as use of safety clothing including helmets Such outcomes
are valuable since the use of motorcycle helmets has been demon-
strated to be effective in preventing head injuries (Liu 2004) De-
spite this the use of safety clothing was not included as an out-
come in this review because the studies measuring this outcome
were limited by poor data quality due to very low response rates
Another outcome of interest in assessing the effectiveness of rider
training is lsquotype of injuryrsquo As Haworth 2005 state in their review
of motorcycle licensing and training rider training could be con-
sidered effective if injuries become less severe because motorcycle
riders have learned how to avoid more serious injuries through
rider training The studies included in this review did not specify
type of injury and so this outcome could not be assessed Last
changes in knowledge and attitudes following rider training are
other potentially useful outcomes We have excluded such studies
(Osga 1979 Clayton 1990 Goldenbeld 2004) as these outcomes
are proxies rather than direct measures of the effectiveness of mo-
torcycle rider training We opted to report on the outcomes of
crashes injury death and offences alone since there is little ev-
idence about how changes in knowledge and attitudes relate to
these outcomes
A U T H O R S rsquo C O N C L U S I O N S
Implications for practice
Due to the poor quality of studies included in this review we were
unable to draw any conclusions about the effectiveness of rider
training on crash rates injurious or fatal or offence rates The
findings do suggest that mandatory pre-licence training may be
an impediment to completing a motorcycle licensing process thuspossibly indirectly reducing crash injury death and offence rates
through a reduction in exposure to riding a motorcycle On the
basis of the existing evidence it is not clear if and what type of
training reduces the risk of crashes injuries deaths or offences in
motorcyclists and the selection of the best rider training practice
can therefore not be recommended The suggestion that manda-
tory pre-licence training indirectly reduces exposure to riding the
motorcycle because trainees are impeded from completing the li-
censing process does not help those motorcyclists who persist in
obtaining their licence and riding the motorcycle Some type of
rider training is likely necessary to teach motorcyclists to ride a
motorcycle safely Good policy is based on sound evidence and
evaluation but such evidence for motorcycle rider training is seri-
ously lacking Policy makers are therefore encouraged to include arigorous evaluation component to any new or existing rider train-
ing programme Such an evaluation should ideally be randomized
and assess whether the training programme is effective and if so
which component(s) of the programme work best
Implications for research
Given theresults of this review which found littlereliable evidence
about the effectiveness of motorcycle rider training it is necessary
to build up an evidence base of the effectiveness of motorcycle
rider training in the reduction of crash injury death and offence
rates To this end well designed randomised controlled trials need
to be conducted that examine the impact of rider training oncrashes and offences A randomised controlled trial is the preferred
method as it would significantly reduce most biases described in
this review To minimise the impact of participants delaying or
never attaining their licence and therefore exposure as found in
the randomised trials in this review such a trial should include
riders who are already licensed and reliable measures of riding
exposure in a reasonably lengthy follow up period It is expected
that a randomised controlled trial of sufficient sample size would
observe equal balance on possible confounders by study arm how-
ever an assessment of potentially important confounders (such as
helmet wearing) by study group would enable control for these
variables in the analyses if necessary It is imperative that the effec-
tiveness of each component of the rider training (such as theorypractice knowledge skills and modification of behaviour) and the
synergy of such components are evaluated Outcomes should be
objective ideally those that are routinely collected rather than self-
reported outcomes Even though riding the motorcycle poses dif-
ferent challenges and incurs different risks lessons may be learned
from research into the effects of car driver training
Motorcycles are a common means of transport in low- and middle
income countries and usage is increasing As the burden of mo-
torcycle crash victims is mostly borne in such countries research
should be focused on the effectiveness of types of rider training on
the reduction of crashes injuries and fatalities in such countries
and be supported by higher income countries This review found
only one study conducted in a middle income country that sug-gested decreasing injury rates It would be beneficial to conduct
additional research to confirm this suggestion and to enhance our
knowledge of the role of rider training in low- and middle-income
settings
March 2006 saw the launch of the University of North Carolina
Highway Safety Research Centre partnership with the Motorcycle
Safety Foundation (MSF) and plans were revealed to conduct a
study of the potential benefits of motorcycle rider training The
study is funded by the MSF and the National Highway Traffic Sa-
fety Administration (NHTSA) Called rsquoThe Discovery Projectrsquo it
will compare data on crash rates from riders participating in a basic
23Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2650
motorcycle safety class with riders participating in a basic motor-
cycle safety class as well as a variety of continuing education classesThe collection of pilot data was due to start in March 2007 and
the data collection portion of the study is expected to take three
years (Highway Safety Research Center) It is vital that researchers
and policy makers collaborate to develop furtherhigh quality eval-
uations of motorcycle rider training programmes to help inform
development of new and effective training programmes
A C K N O W L E D G E M E N T S
Staff from The University of Sydney library for helping us locate
and retrieve full reports of many papers
R E F E R E N C E S
References to studies included in this review
Adams 1985 published data only
Adams A Collingwood V Job RFS Evaluation of the 1982 Australia
Post motorcycle rider training scheme NSW Australia Traffic
Accident Research Unit Traffic Authority NSW 1985 [ RN 85]
Anderson 1980 published data only
Anderson J Ford JL Peck RC Improved motorcyclist licensing and testing project Washington DC National Highway Traffic Safety
Administration (NHTSA) USA 19801ndash217 [
DOTndashHSndash805ndash598]
Billheimer 1998 published data only
Billheimer JW Evaluation of the California Motorcyclist Safety
Program Transportation Research Record 19981640100ndash9 [
TRB Report Number 980652]
Cooper 1988 published data only
Cooper PJ Rothe JP Motorcyclists who they are and why they do
what they do Transportation Research Record 1988116878ndash85
Davis 1997 published data only
David CF Evaluation of community traffic safety programs and
motorcycle operator training programs Connecticut Department of Transportation 199763
Hall 1988 published data only
Hall CJ Buchanan L Engle R Motorcycle rider education evaluationproject Washington DC USA National Highway Traffic Safety
Administration (NHTSA) US Department of Transportation
1988 [ DTNH 22ndash80ndashCndash05123]
Haworth 1997 published data only
Haworth N Smith R Brumen I Pronk N In Monash University
Accident Research Centre editor(s) Case-control study of motorcycle
crashes Victoria Australia Department of Transport and Regional
Development The Federal Office of Road Safety 19971ndash149 [
ISBN 0 642 25501 6 ]
Hurt 1981 published data only
Hurt H Ouellet J Thom D Motorcycle accident cause factors and
identification of countermeasures USA Department of
Transportation 1981 [ Technical Report DOT HS 806ndash862]
Jonah 1982 published data onlylowast Jonah BA Dawson N Bragg B Are formally trained motorcyclists
safer Accident Analysis amp Prevention 198214(4)247ndash55
Jonah BA Dawson NE Bragg BW Evaluation of the Canada
Safety Councilrsquos Motorcycle Training Program Proceedings of theInternational Motorcycle Safety Conference Washington DC
USA 198059ndash93
Kloeden 1994 published data only
Kloeden CN Moore VM McLean AJ Evaluation of the pre-license
training program for motorcyclists in South Australia Walkerville
South Australia Australia NHMRC Road Accident Research Unit
The University of Adelaide 19941ndash37 [ ISBN 0 7308 0607 3]
Kraus 1975 published data only
Kraus JF Riggins RS Franti CE Some epidemiological features of
motorcycle collision injuries American Journal of Epidemiology 1975102(1)74ndash98
Leung 1987 published data only
Leung KS Reding VA Evaluation of the Wisconsin motorcycle rider course Wisconsin Bureau of Policy Planning and Analysis
Wisconsin Department of Transportation 1987
MAIDS 2004 published and unpublished data
MAIDS In-depth investigations of accidents involving powered
two wheelers ACEM (Acemacemeu) 2004 Vol Final report
20179
McDavid 1989 published data only
McDavid JC Lohrmann BA Lohrmann G Does motorcycle
training reduce accidents Evidence from a longitudinal quasi-
experimental study Journal of Safety Research 198920(2)61ndash72
Mortimer 1984 published data only
Mortimer R Evaluation of the motorcycle rider course Accident
Analysis amp Prevention 198416(1)63ndash71
24Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2750
Mortimer 1988 published data only
Mortimer RG A further evaluation of the motorcycle rider courseJournal of Safety Research 198819(4)187ndash96
Osga 1980 published data only
Osga GA An investigation of the riding experiences of MSF rider
course participants Dissertation South Dakota USA The
University of South Dakota 19801ndash224
Perrino 2002 published data only
Perrino CS Ahmed A Callender A Rozier E Cantwell A Stewart
O The role of Marylandrsquos motorcycle rider course in promoting saver
behaviors and attitudes Baltimore USA National Transportation
Center Morgan State University 20021ndash52
Raymond 1979 published data only
Raymond S Tatum S An evaluation of the effectiveness of the RAC
ACU motorcycle training scheme - final report Road Safety ResearchUnit Dept of Civil Engineering University of Salford London
19791ndash131
Satten 1980 published data only
Satten RS Analysis and evaluation of the motorcycle rider courses
in thirteen Illinois counties International Motorcycle Safety
Conference Washington DC 1980145ndash93
Savolainen 2007 published and unpublished data
Savolainen Peter Mannering Fred Additional evidence on the
effectiveness of motorcycle training and motorcyclistsrsquo risk-taking
behaviour Transportation Research Board 2007 Vol March 20
19
Swaddiwudhipong 1998 published data only
Swaddiwudhipong W Boonmak C Nguntra P Mahasakpan P
Effect of motorcycle rider education on changes in risk behavioursand motorcycle-related injuries in rural Thailand Tropical Medicine
and International Health 19983(10)767ndash70
Waller 1992 published data only
Waller FL An analysis of the impact of the Texas Department of
Public Safety Motorcycle Operator Training Program upon
motocycle related traffic accidents Texas State University Public
Administration Program Applied Research Projects 199277
References to studies excluded from this review
Anderson 1980a published data only
Anderson LC Better biking a motorcycle rider improvement
program International Motorcycle Safety Conference Washington
DC USA Motorcycle Safety Foundation 1980231ndash53Aupetit 2008 published and unpublished data
Aupetit S Riff J Buttelli O Espie S Ergonomics research program
on motorcycle learning Contributions to the development of a
motorcycle simulator for training Transport Research Arena
Ljubljana 20089
Awane 1999 published data only
Awane T Integrating simulators in motorcycle safety education
Journal of International Association of Traffic and Safety Sciences 199923(1)26ndash35
Baldi 2005 published data only
Baldi S Baer JD Cook AL Identifying best practices states in
motorcycle rider education and licensing Journal of Safety Research200536(1)19ndash32
Batchler 1988 published data only
Batchler MW Evaluation of the pre-license motorcycle training course at the New South Wales traffic education centre Federal Office of
Road Safety (FORS) 19881ndash122 [ Report No CR 77 ]
Blanchard 2006 published data only
Blanchard HT Tabloski PA Motorcycle safety educating riders at
the teachable moment Mosby (CV) Company 200632330ndash2
Bonnett 2005 published data only
Bonnett J Mature age skills training for experienced motorcycle
riders Motorcycle ridersrsquo association of the Australian Capital
Territory 2005
Braver 2007 published data only
Braver ER et alPersuasion and licensure a randomized controlled
intervention trial to increase licensure rates among Maryland
motorcycle owners Traffic Injury Prevention 20078(1)39ndash46Clayton 1980 published data only
Clayton AB Sudlow DE An evaluation of the effectiveness of a one day training programme for learner motorcyclists Berkshire UK
Transport and Road Research Laboratory 19801ndash23 [ ISSN
0266ndash7045]lowast Clayton AB Sudlow DE An evaluation of the effectiveness of a one-day training programme for learner motorcyclists Berkshire UK
Road Safety Division Berkshire 19801ndash23 [ISSN 0266ndash7045]
Council 1978 published data only
Council FM Desper LP Dutt AK Driver education for motorcycle operation University of North Carolina Highway Safety Research
Center 19781ndash40
Crowther 2005 published data onlylowast Crowther G Engaging with motorcyclists UK police and theBikeSafe Road Safety Programme Published proceedings of the
Australasian Road Safety Research Policing Education Conference
Wellington New Zealand 2005 issue ISBN 0ndash473ndash10636ndash11ndash11
Daniello 2009 published data only
Daniello A Gabler HC Mehta YA Effectiveness of motorcycle
training and licensing Transportation Research Board 88th Annual
Meeting 200913
Dewanti 2007 published data only
Dewanti M Challenging efforts in promoting young motorcyclist
safety in Indonesia Road Safety on Four Continents 14th
International Conference Swedish National Road and Transport
Research Institute 2007
Du Plooy 1990 published data onlyDu Plooy DW Motorcycle training in South Africa The Human
Element International Motorcycle Safety Conference Orlando
Florida USA Motorcycle Safety Foundation 19904-1 to 4-9
Engel 1989 published data only
Engel U Krogsgaringrd Thomsen L [Faeligrdselsundervisning og
propaganda ndash en faeligrdselssikkerhedsfremmende foranstaltning]
Dansk Vejtidsskrift 198912295ndash7
Goldenbeld 2002 published data only
Goldenbeld C Houwing S de Craen S The development of
driving skills of young moped riders [De ontwikkeling van de
rijvaardigheid van jonge bromfietsers] SWOV Institute for Road
Safety Research (Stichting Wetenschappelijk Onderzoek
Verkeersveiligheid) Leidschendam the Netherlands 200263
25Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2850
Goldenbeld 2004 published data only
Goldenbeld C Twisk D de Craen S Short and long term effects of moped training a field experiment Transportation Research Part F 20047(1)1ndash16
Haworth 2000 published data only
Haworth N Smith R Kowadlo N Motorcycle training in Victoria
evaluation of rider training curriculum in Victoria Monash
University Accident Research Centre Reports 2000 Vol 165 issue
3131
Haworth 2005 published data only
Haworth N Mulvihill C Review of motorcycle licensing and
training Monash University Accident Research Centre 2005 [
Report no240]
Haworth 2006 published data only
Haworth N Mulvihill C Rowdon P Teaching old dogs new tricksTraining and older motorcyclists Journal of the Australasian College of Road Safety 200718(4)20ndash5lowast Haworth NL Mulvihill C Rowden PJ Teaching old dogs new
tricks Training and older motorcyclists Australasian Road Safety
Research Policing and Education Conference Gold Coast
Queensland 2006
Hill 1983 published data only
Hill D Motorcycle rider training testing and licensing Motor
Cycle Council of New South Wales Sydney 1983
Imatake 1980 published data only
Imatake Y On the activities in Japan of motorcycle safe riding
promotion committee (MSRPC) - especially the education of
unskilled riders of class 1 motorbikes International Motorcycle
Safety Conference Washington DC USA Motorcycle Safety
Foundation 1980255ndash68
Ingebritsen 1990 published and unpublished data
Ingebrigtsen S [Risikofaktorer ved ferdsel med moped og
motorsykkel] TOslashI-rapport 66 Oslo Transportoslashkonomisk
institutt 1990
Jain 2009 published and unpublished data
Jain A et alTwo wheeler accidents on Indian roads a study from
Mangalore India Elsevier 200916130ndash3
Jonah 1979 published data only
Jonah BA Dawson NE Validation of the motorcycle operator skill
test Accident Analysis amp Prevention 197911(3)163ndash71
Jonah 1981 published data onlyJonah BA Dawson NE Bragg WE Predicting accident
involvement with the Motorcycle Operator Skill Test Accident Analysis amp Prevention 198113(4)307ndash418
Kadar 1990 published data only
Kadar KJ Countermeasures motorcycle rider education delivery
system The Human Element International Motorcycle Safety
Conference Motorcycle Safety Foundation 19906-27 to 6-41
Kelsey 1986 published data only
Kelsey SL Liddicoat C Ratz M Licensing novice motorcyclists a
comparison of Californiarsquos standard test and the MOST II (Motorcycle
Operator Skill Test) administered at centralized testing offices Sacramento California USA California Department of Motor
Vehicles Research and Development Office 19861ndash74
Koch 1980 published data only
Koch H Advanced course for motorcycle riders InternationalMotorcycle Safety Conference Washington DC USA Motorcycle
Safety Foundation 1980269ndash81
Koch 1990 published data only
Koch H Brendicke R How to influence risk taking behaviour
within motorcycle rider training programs The Human Element
International Motorcycle Safety Conference Orlando Florida
USA Motorcycle Safety Foundation 1990Motorcycle Safety
Foundation 1990 p 9-1 to 9-25A
Labbett 2005 published data only
Labbett S Langham M Training can make the problem worse
Road Safety Congress 20051ndash12
Law 2005 published and unpublished data
Law TH Radin Umar RS Zulkaurnain S Kulanthayan S Impactof the effect of economic crisis and the targeted motorcycle safety
programme on motorcycle-related accidents injuries and fatalities
in Malaysia International Journal of Injury Control and Safety
Promotion 200512(1)9ndash21
Lowes 1990 published data only
Lowes BF In-traffic rider training The Human Element
International Motorcycle Safety Conference Orlando Florida
USA Motorcycle Safety Foundation 19907-46 to 7-57
Mahon 1976 published data only
Mahon J Teaching accident avoidance on the motorcycle Journal of Traffic Safety Education 1976231ndash29
Maxwell 1985 published data only
Maxwell GS Evaluation of a Program for Training MotorcycleRiders Evaluation and Program Planning 19858(4)339ndash47
Mayhew 1996 published data only
Mayhew DR Simpson SH Effectiveness and role of driver educationand training in a graduated licensing system Ottowa Canada The
Traffic Injury Research Foundation 19961ndash89 [ ISBN
0ndash920071ndash12ndash0]
McKnight 1987 published data only
McKnight AJ Milburn BG An evaluation of the modified
motorcycle operator skill test exercises NHTSA Report 1987 [
Report No DOT HS 807 212 ]
McPherson 1976 published data only
McPherson K McKnight AJ The development and evaluation of a
motorcycle skill test manual and knowledge test Washington DCUSA National Highway Safety Administration 1976
Mohan 2004 published data only
Mohan D Evidence-based interventions for road traffic injuries in
South Asia Journal of the College of Physicians and SurgeonsPakistan 200414(12)746ndash7
Morgan 1980 published data only
Morgan JK Advanced motorcycle training course content and
study evaluation International Motorcycle Safety Conference
Washington DC USA Motorcycle Safety Foundation 1980
283ndash312
Motoki 1990 published data only
Motoki M Yamazaki S A study on effective motorcycle rider
education The Human Element International Motorcycle Safety
26Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2950
Conference Orlando Florida USA Motorcycle Safety
Foundation 19909-26 to 9-57A
Nairn 1992 published data onlylowast Nairn RJ Motorcycle safety research literature review 1987 to
1991 Federal Office of Road Safety Australia 1993 Vol Report
CR 117
Newland 1980 published data only
Newland R Darby J Automotive practicespre motorcycle rider
program International Motorcycle Safety Conference Washington
DC USA Motorcycle Safety Foundation 1980127ndash9
Newland 1982 published data only
Newland R Motorcycle rider training and licensing requirements
Driver training steering a course for the future Melbourne
Victoria Transport Regulation Board 1982
Newland 1990 published data only
Newland R Motorcycle training and licensing in Victoria 1980-
1990 The Human Element International Motorcycle Safety
Conference Orlando Florida USA Motorcycle Safety
Foundation 1990
Ormston 2003 published data only
Ormston R Dudleston A Pearson S Stradling Evaluation of
Bikesafe Scotland Scottish Executive Social Research Edinburgh
2003
Osga 1979 published data only
Osga GS Ellingstad VS An evaluation of the effectiveness of the 1978
South Dakota motorcycle rider course program South Dakota USA
University of South Dakota 19791ndash89
Panou 2007 published data only
Panou M Bekiaris E A new concept on the integration of driving
simulators in driving training the train-all approach Road Safety
on Four Continents 14th International Conference Swedish
National Road and Transport Research Institute 2007
Peterson 1990 published data only
Peterson A Kadar K Status report 1980-1989 Minnesota rider
education statistics The Human Element International
Motorcycle Safety Conference Orlando Florida USA 199013-
113 to 13-129
Prem 1984 published data only
Prem H Good MC Motorcycle rider skills assessment ACT
Australia Office of Road Safety Department of Transport 1984
1ndash377 [ Report No CR 34 ]
Radin Umar 2006 published data only
Radin Umar RS Motorcycle safety programmes in Malaysia how
effective are they International Journal of Injury Control and Safety Promotion 200613(2)71ndash9
Rockwell 1990 published data only
Rockwell TH Kiger S Carnot M An evaluation of the Ohio
motorcyclist enrichment program Ohio Department of Highway
Safety 1990
Rothe 1987 published data only
Rothe JP A motorcycle riding course Motorcyclists Image and reality Canada Insurance Corporation of British Columbia 1987
59ndash83
Rowden 2007 published data only
Rowden P Watson B Haworth N What can riders tell us aboutmotorcycle training A view from the other side of the fence
Australasian Road Safety Research Policing and Education
Conference Melbourne Australia Melbourne 2007
RSC Victoria 1993 published data only
Road Safety Committee Inquiry into Motorcycle Safety in Victoria
Parliament of Victoria 1993
Russam 1979 published data only
Russam K Motorcycle training and licensing in Japan UK Transport
and Road Research Laboratory (TRRL) 1979 [ Report 916 ]
Saffron 1981 published data only
Saffron DG Driver and rider education training and licensing a
brief review Traffic Accident Research Unit Traffic Authority of
New South Wales NSW Australia Traffic Accident Research UnitTraffic Authority of New South Wales 1981
Schulz 1990 published data only
Schulz U Motorcycle accidents among learner riders Motorcycle
licensing Evaluation of the effectiveness of a Graduated Licensing Scheme Forshungshefte des Institut fur Zweiradsicherheit 1990
Sexton 2004 published data only (unpublished sought but not used)
Sexton B Baughan C Elliott M Maycock G The accident risk of
motorcyclists 2004TRL Report TRL 607(ISSN 09684107)58
Shephard 1986 published data only
Shepherd R Do Motorcycle Safety Foundation programs work
Driver 198619(9)16ndash8
Simpson 1990 published data onlySimpson HM Mayhew DR The promotion of motorcycle safety
training education and awareness Health Education Research
Theory and Practice 19905(2)257ndash64
Sokolov 1990 published data only
Sokolov O Safe motorcycle riding some aspects of motorcycle
riding in the USSR The Human Element International
Motorcycle Safety Conference Motorcycle Safety Foundation
19904ndash37 - 4-48
Steffens 1988 published data only
Steffens U Gawatz R Willmes G Effectiveness of mofa courses
[Wirksamkeit von Mofakursen] Unfall- und sicherheitsforschung
strasenverkehr 198867168
Sudlow 2003 published data only
Sudlow D Scoping study on motorcycle training Road Safety
Research Report London UK The British Institute of Traffic
Education Research Department for Transport 2003 Vol June
[ Report No 36]
Tasmania 1995 published data only
Mannering FL Grodsky LL Motorcycle rider training scheme
Traffic Education Research Centre Armidale Australia Driver
Training Australia Limited 1995 Vol April1ndash56
Tetard 1985 published data only
Tetard C Motorcycle ridersrsquo training to detect indices in an
intersection situation [Lrsquoentrainement a la prise drsquoinformation dans
lrsquoapprentissage de la conduite motocycliste] Le Travail Humain198548(2)147ndash59
27Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3050
Thackray 1980 published data only
Thackray RM Prescott JC Field test of the motorcycle ridercourse International Motorcycle Safety Conference Washington
DC USA Motorcycle Safety Foundation 1980195ndash7
Thompson 1994 published data only
Thompson M Evaluation of compulsory basic training for
motorcyclists Transport Research Laboratory Crowthorne
Berkshire UK 1994 [ Report 63]
Tobler 1990 published data only
Tobler U Individual driving instruction for motorcycle driving
learners the Swiss experiment of motorcycle driving school The
Human Element International Motorcycle Safety Conference
Orlando Florida USA Motorcycle Safety Foundation 1990
Transport 1989 published data only
Transport Communications Road Crash Statistics Australia Bureauof Transport and Communications Economics 198914ndash20
Tuan 2007 published data only
Tuan VA Shimizu T Towards development and evaluation of the
motorcycle drivers re-education program in Vietnam modeling of
motorcycle driverrsquos undesired behaviors 11th World Conference
on Transport Research 200725
Victor 1980 published data only
Victor DJ Traffic safety education for motorcyclists International
Motorcycle Safety Conference Washington DC USA Motorcycle
Safety Foundation 1980221ndash8
Waterer 1990 published data only
Waterer TH Compulsory training for motorcyclists from liberty
to compulsion The Human Element International Motorcycle
Safety Conference Orlando Florida USA Motorcycle Safety
Foundation 19904-49 to 4-57
Weller 1989 published data only
Weller RB Chandler EW Motorcycle safety and motorcycle
education past research and survey results Journal of Public Policy
amp Marketing 1989893ndash108
Wells 1981 published data only
Wells P Observations of some RACACU motorcycle training
schemes Transport and Road Research Laboratory Crowthorne
Berkshire UK 1981
Winn 1986 published data only
Winn G Two important trends in motorcycle safety regulations
rider education and conspicuity improvement American
Motorcyclist Association 1986147ndash52
Wisher 1988 published data only
Wisher G Reid J Motorcycle accidents in Norfolk a three-year study of motorcycle training and the Part 1 test Norfolk UK Road Safety
Division Highways Department Norfolk County Council 1988
Wood 1987 published data only
Wood HT Bowen R Evaluation of the revised learner permit scheme
July 1983 to December 1985 Victoria Australia Road Traffic
Authority 19871ndash27 [ ISBN 0 7306 0247 8]
Yamazaki 1990 published data only
Yamazaki S Nakanishi M Survey on effect of Honda rider
education The Human Element International Motorcycle Safety
Conference Orlando Florida USA 199012-167 to 12-187
References to ongoing studies
Burgess 2010 unpublished data only
Burgess C Broughton P Fylan F Stradling Interim evaluation of
the UKs National Ride program In L Dorn editor(s) Driver
behaviour and Training Vol IV Aldershot Ashgate Publishing
Limited 2010
Additional references
Anderson 1980
Anderson J Ford JL Peck RC Improved motorcyclist licensing and
testing project Volume I and II National Highway Traffic Safety
Administration US department of Transportation Washington
DC 1980
Bjornstig 1985
Bjornstig U Bylund P Lekander T Brorsson B Motorcycle fatalitiesin Sweden Acta Chirurgica Scandinavica 1985151(7)577ndash81
Clayton 1990
Clayton A Sudlow DE An evaluation of the effectiveness of a one-
day training program for learner motorcyclists The Human
Element International Motorcycle Safety Conference Orlando
Florida USA 199013-105 to 13-111
Cochrane Collaboration 2008
Higgins JP Green S Cochrane Handbook for Systematic Reviews of
Interventions 500 The Cochrane Collaboration 2008
Elliot 2003
Elliott MA Baughan CJT Broughton J Grayson G Knowles J et
alMotorcycle Safety a Scoping Study TRL [ Report No
0968ndash4107]Fleiss 2004
Fleiss JL Levin B Paik MC Statistical methods for rates and proportions 3rd Edition John Wiley amp Sons Inc 2004 [ ISBN
978ndash0ndash471ndash52629ndash2 ]
FORS 1999
FORS Federal Office of Road Safety Road risk for sober licensed
motorcyclists Vol Monograph 27 Federal Office of Road Safety
19991ndash2
Haworth 2005
Haworth N Mulvihill C Review of motorcycle licensing and training
Victoria Australia Monash University Accident Research Centre
20051ndash83 [ Report No 240]
Huwiler-Muumlntener 2002Huwiler-Muumlntener K Juumlni P Junker C Egger M Quality of
Reporting of Randomized Trials as a Measure of Methodologic
Quality JAMA 20022872801ndash4
Johnston 2008
Johnston P Brooks C Savage H Fatal and serious road crashes
involving motorcyclists Road Safety Research and Analysis Report
ATSB Monograph 20 Canberra Department of Infrastructure
Transport Regional Development and Local Government 0205
081ndash26
Juumlni 2001
Juumlni P Altman DG Egger M Systematic reviews in health care
Assessing the quality of controlled clinical trials BMJ 2001323
42ndash6
28Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3150
Kopits 2003
Kopits E Cropper M Traffic fatalities and economic growth TheWorld Bank Development Research Group Infrastructure and
Environment Policy research working paper 3035 2003
Leung 1983
Leung P Chau BTW Motorcycle accidents in Hong Kong pattern
of injury treatment causative factors and socio-economic
implications Injury 198315(1)23ndash9
Liu 2004
Liu B Ivers R Norton R Blows S Lo SK Helmets for preventing
injury in motorcycle riders Cochrane Database of Systematic Reviews 2004 Issue 2 [DOI 10100214651858CD004333pub3]
Mohan 2002
Mohan D Road safety in less-motorized environments future
concerns International Journal of Epidemiology 200231(3)527ndash32Mortimer 1982
Mortimer RG An evaluation of the motorcycle rider course
Proceedings of the 20th conference of the American Association for
Automotive Medicine Arlington Heights IL USA American
Association for Automotive Medicine 198277ndash96
Mullin 1997
Mullin BT Risk factors for motorcycle injury the role of age
gender experience training and alcohol PhD thesis Auckland
New Zealand University of Auckland 1997
Noordzij 2001
Noordzij PC Forke E Brendicke R Chinn BP Integration of needs
of moped and motorcycle riders into safety measures review and
statistical analysis in the framework of the European research
project PROMISING SWOV Institute for Road Safety Research
Leidschendam The Netherlands 2001 [ Report No Dndash2001ndash5]
Osga 1979
Osga G Ellingstad VS An evaluation of the effectiveness of the 1978 South Dakota Motorcycle Riders Course Program South Dakota
University of South Dakota Vermillion Human Factors Lab 1979
Oxford English Dictionary Online
httpoxforddictionariescom Oxford English Dictionary Online
Oxford University Press 2010
Peden 2004
Peden M Scurfield R Sleet D World Report on Road TrafficInjury Prevention Geneva World Health Organization Available
at httpwwwwhointviolence˙injury˙preventionpublications
road˙trafficworld˙reporten 2004Chapter 3
Radin Umar 1995
Radin Umar R Macakay G Hills B Preliminary analysis of
motorcycle accidents Short-term impacts of the running
headlights campaign and regulations in Malaysia J Traffic Med 199523(1)17ndash27
Tsai 1995
Tsai YJ Wang JD et alCase-control study of the effectiveness of
different types of helmets for the prevention of head injuries among
motorcycle riders in Taipei Taiwan American Journal of
Epidemiology 1995142(9)974ndash81
Vis 1995
Vis AA In-depth study of the hazards of motorcycling [De
onveiligheid van motorrijden nader bekeken] SWOV Institute for
Road Safety Research Leidschendam the Netherlands 1995 [
Report No Rndash95ndash69]
Weaver 1990
Weaver EA Motorcycle safety education in the United States
Motorcycle Safety Foundation Orlando Florida USA 1990
Winn 1987
Winn G Two important trends in motorcycle safety regulations
rider education and conspicuity improvement No more
information available Detroit MI US Society of Automotive
Engineers 1987147ndash52
Yuan 2000
Yuan W The effectiveness of the rsquoride-brightrsquo legislation for
motorcycles in Singapore Accident Analysis amp Prevention 200032
(4)559ndash63
Zambon 2006
Zambon F Hasselberg M Factors affecting the severity of injuries
among young motorcyclists a Swedish nationwide cohort study
Traffic Injury Prevention 20067(2)143ndash9lowast Indicates the major publication for the study
29Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3250
C H A R A C T E R I S T I C S O F S T U D I E S
Characteristics of included studies [ordered by study ID]
Adams 1985
Methods Non-randomised trial
Participants Initial sample size
Total of 162 Australia Post telegram delivery riders
Not clear how many in intervention and control groups
Analysis sample size
Intervention group 63Control group 35
Total 98 (61)
Interventions No formal training vs voluntary rider training course of 12 hours over 2 consecutive days (theory 4 hours practice
8 hours)
Intervention integrity
No information given
Outcomes Australia Post recorded crashes
Length of follow up 1 year
Notes Study conducted in NSW Australia
Anderson 1980
Methods Randomised trial
Participants Initial sample size
Total 41188 motorcycle licence applicants
Analysis sample size
Intervention group with skills training 13160
Intervention group without skills training 12634
Control group 15080
Total 40874 (99)
Interventions Standard licensing skill test vs the Motorcycle Operators Skill Test (MOST) with or without a training component
MOST is not to be confused with rider training because it is a licensing test The training component was for
participants who failed the skill test on the first attempt was of 3 hours duration and consisted of instructions about
the skills needed to pass the skill test The training consisted of 30 min theory and 2 frac12 hours practice on the
motorcycle
Intervention integrity
33 in MOST with skills training 31 in MOST without skills training and 43 in control group obtained their
motorcycle licence in the end
Outcomes Police recorded crashes and offences
Length of follow up 3 years
30Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3350
Anderson 1980 (Continued)
Notes Study conducted in California USA
Billheimer 1998
Methods Cohort study
Participants Initial sample size
Intervention group Unclear numbers of CMSP graduates and ERC graduates
Control group Over 16000 untrained motorcyclists
Analysis sample size
1139 matched CMSP-pairs and 1182 ERC-pairs
Interventions No formal training vs the mandatory (for those younger than 18 years) pre-licence 16-hours basic California
Motorcyclist Safety Program (CMSP) and the 8-hour Experienced Rider Course (ERC)
Intervention integrity
No information given
Outcomes Police recorded crashes and offences
Length of follow up 2 years
Notes Study conducted in California USA
Cooper 1988
Methods Cohort study
Participants Initial sample size
Intervention group 418 graduates and 57 who failed
Control group 402
Analysis sample size
Intervention group unclear
Control group unclear
Total 863 (984)
Interventions No formaltrainingvs non-mandatory pre-licence rider trainingThe content of thetrainingwas notfurther explainedIntervention integrity
Participants had all completed the rider training
Outcomes Police recorded crashes and offences
Length of follow up 7 years
Notes Study conducted in British Columbia Canada
31Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3450
Davis 1997
Methods Retrospective cohort study
Participants Initial sample size
Intervention group 9320 graduates
Control group unclear
Analysis sample size
Intervention group 9320 graduates
Control group unclear
Interventions No formal training vs the non-mandatory Connecticut Motorcycle Operator Training Program The content of
training was not further explained
Intervention integrity Participants had all completed the rider training
Outcomes Police recorded crashes severity of crashes
Length of follow up 1 year
Notes Study conducted in Connecticut USA
Hall 1988
Methods Randomised trial
Participants Initial sample sizeIntervention group A 6248
Intervention group B 6616
Intervention group C 6996
Control group 6604
Total 26464 motorcycle licence applicants
Analysis sample size
Intervention group A 6012 (96)
Intervention group B 6428 (97)
Intervention group C 6778 (97)
Control group 6502 (98)
Interventions Standard licensing skill test vs the Motorcycle Operator Skill Test (MOST II) without training (A) MOST II with
3 hours training (B) or MOST II with 20 hours training (C) The 3 hours training comprised instructions aboutthe skills needed to pass the skill test (30 min theory and 2 frac12 hours practice on the motorcycle) The 20 hours
training consisted of the Motorcycle Safety Foundationrsquos Motorcycle Rider Course (MRC) comprising of 8 hours
theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-road practice
Intervention integrity
25 and 26 obtained their motorcycle licence in the lsquoMOST 3-hour trainingrsquo and lsquoMOST 20-hour trainingrsquo
intervention arm respectively In the lsquoMOST testrsquo intervention arm 29 obtained their licence and in the control
group 35
Outcomes Police recorded crashes and offences
Length of follow up 3 years
32Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3550
Hall 1988 (Continued)
Notes Study conducted in New York USA
Haworth 1997
Methods Population based case-control study
Participants Cases Motorcycle riders or pillion riders involved in motorcycle crash (on road) Recruited from hospital and police
records Not clear whether all cases (especially fatalities) were included
Community controls Multiple matched sample of motorcycle rider or pillion who passed the crash site at the same
time of day and day of week that the crash occurred
Initial sample size
Cases 336
Controls 1195
Analysis sample size
Cases 234 (70)
Controls 561 (47)
Interventions Basic handling skills course beginner intermediate or advanced course measured through self-report not validated
Outcomes Injuries with an Injury Severity Score of 5 or greater or admission for at least four hours and fatalities from a
motorcycle crash (from hospital and police records)
Length of follow up two years or more prior to the interview
Notes Study conducted in Victoria Australia
Hurt 1981
Methods Population based case-control study
Participants Cases Motorcycle riders who had a motorcycle crash that was reported to police (traffic accident reports) Not all
cases included
Community controls Multiple matched sample of motorcycle riders who passed the crash site at the same time of
day and day of week as the crash occurred but 2 years later
Initial sample size
Cases 4500 notifications of crashes to police (only 50 received in time research team able to respond to 1126 for
unreported reasons data collection completed for 900)
Controls Number approached unreported 2310 consented and interviewed
Analysis sample size
Cases 808 (808900 data on rider training) - 18 of initial sample
Controls 670 (6702310 data on rider training) - lt 29 of initial sample
Interventions Training experience self taught friends-family school-club formal-professional others measured through self-
report validated
Outcomes Injuries and fatalities from a motorcycle crash (from hospital records and coronerrsquos office)
33Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3650
Hurt 1981 (Continued)
Notes Study conducted in Los Angeles USA
Jonah 1982
Methods Cohort study
Participants Initial sample size
Intervention group
2310 licensed MTP graduates who are active riders
Control group
2029 licensed motorcyclists who are active riders
Non-licensed and non-active riders excluded
Analysis sample size
Intervention group
811 interviews (35) 533 driver records (23)
Control group 1080 interviews (71) 905 driver records (45)
Interventions No training vs the non-mandatory Canada Safety Council Motorcycle Training Program (MTP) comprising of 4
hours theory and 16 hours practise (6 hrs on-road) over 2 weekends
Intervention integrity
Participants had all successfully completed the MTP
Outcomes Self-reported and police reported crashes and offencesLength of follow up 4 years
Notes Study conducted in Ontario Canada
Kloeden 1994
Methods Cohort study
Participants Initial sample size
Intervention group 2533 licence applicants
Control group Unclear
Sample of riders obtaining probationary licenceIntervention group 1625 (64)
Control group 5015 (unclear)
Non-active riders excluded from control group (~66)
Analysis sample size
Intervention group 631 (39)
Control group 318 (6)
Interventions No formal training vs the mandatory pre-licence Ridersafe course consisting of two levelsthe first is a training course
in the basics of motorcycle riding where a pass entitles the rider to obtain a learnerrsquos permit The second is a skill
test where a pass entitles the rider to obtain a probationary motorcycle licence The training course consists of two 4
hour training sessions on separate days involving instruction videos and actual riding on motorcycles The skill test
consists of a 4 hour preparation course of 74 minutes of lecturediscussion 38 minutes of video and 75 minutes of
34Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3750
Kloeden 1994 (Continued)
on-course riding which is the testing phase
Intervention integrity
Unclear what percentage in the Ridersafe group and the control group obtained their probationary licence halfway
and their full motorcycle licence in the end
Outcomes Police reported crashes
Length of follow up 1 year
Notes Study conducted in SA Australia
Kraus 1975
Methods Population based case-control study
Participants Cases All motorcycle riders with a confirmed medically treated motorcycle injury (including fatalities) from all
hospitals in study base and peripheral area hospitals and all deaths due to motorcycle crashes in the study base All
serious injuries (leading to hospitalisation) and fatalities likely to be included
Community controls A random sample of registered motorcycle owners in the same temporary and geographical
study base as the cases arose from
Initial sample size Cases 1273
Controls 738
Analysis sample size Cases 741 (58)
Controls 434 (59)
Interventions Motorcycle rider training (not specified) measured through self-report
Outcomes Injuries leading to hospitalisation fatalities
Length of follow up not specified
Notes Study conducted in Sacramento county USA
Leung 1987
Methods Non-randomised trial
Participants Initial sample sizeIntervention group unknown number of MRC graduates
Control group unknown number of licensed motorcyclists
Analysis sample size
Intervention group 2941
Control group 43094
Interventions No formal training vs non-mandatory Wisconsin motorcycle rider course a 22 hours off-street programme that
teaches the basic rules of safe motorcycling
Intervention integrity
50 in the MRC group and 100 in the control group obtained their motorcycle licence in the end
35Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3850
Leung 1987 (Continued)
Outcomes Police recorded crashes and offences
Length of follow up up to 3 years
Notes Study conducted in Wisconsin USA
MAIDS 2004
Methods Population based case-control study
Participants Cases Motorcycle riders who had a motorcycle crash Unclear whether all cases included
Community controls Motorcycle riders identified at petrol stations Unclear whether matched to cases
Initial sample size
Cases not reported
Controls not reported
Analysis sample size
Cases 921 (Italy 200 Netherlands 200 Spain 121 Germany 250 France 150)
Controls 923 (Italy 200 Netherlands 200 Spain123 Germany 250 France 150)
Interventions Pre-license rider training or additional training (non-compulsory)
Outcomes Crashes
Notes Study in five sampling areas located in five European countries (France Germany Netherlands Spain and Italy)
McDavid 1989
Methods Cohort study
Participants Sample size after first match
Intervention group 346 male training graduates
Control group 346 male motorcycle riders licensed in same year
Sample size after second match
Intervention group 139 (40)
Control group 139 (40)
Analysis sample size
Intervention group 139 (40)
Control group 139 (40)
Interventions No formal training vs the non-mandatory pre-licence British Columbia Safety Councilrsquos 37-hour motorcycle safety
programme
Intervention integrity
Participants had all completed and passed the training
Outcomes Police reported crashes injurious crashes
Length of follow up 1-6 years
36Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3950
McDavid 1989 (Continued)
Notes Study conducted in British Columbia Canada
Mortimer 1984
Methods Cohort study
Participants Initial sample size
Intervention group 608 MRC graduates
Control group 360 motorcycle riders
Response rates
Intervention group 360 (59)
Control group 324 (90)
Non-active riders excluded
Analysis sample size
Intervention group 213 (35)
Control group 303 (84)
Interventions No formal training vs the non-mandatory Motorcycle Safety Foundationrsquos Motorcycle Rider Course comprising of
8 hours theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-road practice
Intervention integrity
Participants had all completed and passed the MRC
Outcomes Self-reported crashes and offencesLength of follow up 1 year
Notes Study conducted in Illinois USA
Mortimer 1988
Methods Cohort study
Participants Initial sample size
Intervention group 2307 MRC graduates
Control group 556 active licensed motorcycle riders
Response ratesIntervention group 1298 (56)
Control group 500 (90)
Non-active riders excluded
Analysis sample size
Intervention group 913 (40)
Control group 500 (90)
Interventions No formal training vs the non-mandatory Motorcycle Safety Foundationrsquos Motorcycle Rider Course comprising of
8 hours theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-road practice
Intervention integrity
Participants had all completed and passed the MRC
37Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4050
Mortimer 1988 (Continued)
Outcomes Self reported crashes and offences
Length of follow up 1 year
Notes Study conducted in Illinois USA
Osga 1980
Methods Cohort study
Participants Initial sample sizeIntervention group 1163 MSF MRC participants invited 818 (74) responded
Control group 329 motorcycle riders responded to recruitment strategies
Non-active riders excluded from analysis
Analysis sample size
Intervention group 595 (73)
Control group 329 (100)
Interventions No formal training vs the non-mandatory Motorcycle Safety Foundationrsquos Motorcycle Rider Course (MRC) com-
prising of 8 hours theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-road
practice
Intervention integrity
Participants had all successfully completed the MRC
Outcomes Self-reported crashes police recorded crash reports
Length of follow up 6 months - 3 12 years
Notes Study conducted in South Dakota USA
Perrino 2002
Methods Cohort study
Participants Initial sample size
Intervention group 2144 (unclear) MRC graduates (motorcycle licence applicants)
Control group 2601 (unclear) regular licensing test graduates (motorcycle licence applicants)Analysis sample size
Intervention group 1070 (unclear)
Control group 867 (unclear)
Interventions No formal training vs the mandatory (for those younger than 18 years) pre-licence Maryland Motorcycle Rider
Course (MRC) consisting of 8 hours theory and 12 hours practice
Intervention integrity
No information given
Outcomes Self-reported and police recorded crashes and offences
Length of follow up 1 year
38Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4150
Perrino 2002 (Continued)
Notes Study conducted in Maryland USA
Raymond 1979
Methods Cohort study
Participants Initial sample size
Intervention group
258 volunteers mainly (but not all) novice riders
Control group
1104 provisional licence applicants
Analysis sample size
Intervention group 190 (74)
Control group 627 (57)
Interventions No formal training vs the non-mandatory RACACU (Royal Automobile ClubAuto-Cycle Union) training scheme
The training program consisted of either 24 one-hour sessions or 13 two-hour sessions Twelve hours of the training
focused on theory about basic operational procedures traffic laws and highway code The practical session focused
on basic operational skills riding on a training area and riding in traffic with an instructor A proficiency test was
administered at the end of the course
Intervention integrity
No information given
Outcomes Self-reported crashes (including minor non-injury non-damage and off-road crashes)
Length of follow up 1 year
Notes Study conducted in the UK
Satten 1980
Methods Cohort study
Participants Initial sample size
Intervention group interviewers kept interviewing until they had 100 riders selected from a population of 891
enrollees in the MSF MRC course (unknown how many were asked and how many refused to participate)Control group unknown number of registered motorcycle riders
Non-active riders excluded from analysis
Analysis sample size
Intervention group 71
Control group 100
Interventions No formal training vs the non-mandatory Motorcycle Safety Foundationrsquos (MSF) Motorcycle Rider Course (MRC)
comprising of 8 hours theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-
road practice
Intervention integrity
Participants had all successfully completed the MRC
39Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4250
Satten 1980 (Continued)
Outcomes Self-reported crashes injuries and offences
Length of follow up 3 years
Notes Study conducted in Illinois USA
Savolainen 2007
Methods Cohort study
Participants Initial sample size
Intervention group 4000 riders that had conducted the MSF MRC course (passed or failed)
Control group 4000 motorcycle owners or permit or endorsement holders from the Indiana Bureau of Motor
Vehicles database
Analysis sample size
Intervention group 739 (18)
Control group 588 (15)
Interventions Intervention integrity
No information given
Outcomes Probability of crash involvement
Length of follow up up to 5 years
Notes Study conducted in Indiana USA
Swaddiwudhipong 1998
Methods Randomised trial
Participants Initial sample size
Intervention group
all 16 villages of 3 sub-districts
(participants were practising motorcyclists licensed and unlicensed)
Control group
18 villages of 3 other sub-districtsAnalysis sample size
Intervention group 3 districts (100)
Control group 3 districts (100)
Interventions No rider education vs non-mandatory community based rider education consisting of theory on motorcycle injury
risk helmet use traffic laws vehicle regulations and traffic signs and a written and skill test leading to a licence
Intervention integrity
No information given
Outcomes Injuries fatalities
Length of follow up 2 years
40Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4350
Swaddiwudhipong 1998 (Continued)
Notes Study conducted in Tak Province rural Thailand
Waller 1992
Methods Cohort study
Participants Initial sample size
Intervention group 349 riders who succesfully completed the Department of Public Safety motorcycle operator
training in 1988
Control group 349 motorcyclists who never completed the DPS motorcycle operator training
Analysis sample size
Intervention group 349
Control group 349
Interventions No DPS motorcycle operator training vs non-mandatory DPS motorcycle operator training The DPS training is
the Motorcycle Safety Foundationrsquos Motorcycle Rider Course comprising eight hours of theory (lectures discussions
films and multiple choice tests) a textbook and 12 hours off-road practice
Intervention integrity
All 349 riders succesfully completed the rider training
Outcomes Police recorded crashes
Length of follow up 3 years
Notes Study conducted in
Characteristics of excluded studies [ordered by study ID]
Study Reason for exclusion
Anderson 1980a Study type review and process evaluation of a rider course
Aupetit 2008 Outcome proposal of a rider training curriculum that combines traditional training and a motorcycle riding
simulator
Awane 1999 Study type review on use of simulators in rider and driver training
Baldi 2005 Intervention not rider training but intervention integrity of rider training availabilty of rider education
implementation of graduated licensing system etc on a state-wide level
Batchler 1988 Study type description of a training course
Blanchard 2006 Study type not a study assessing the association between rider training and outcomes
Bonnett 2005 Outcome qualitative data regarding satisfaction with the training programme
41Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4450
(Continued)
Braver 2007 Outcome licensure rates
Clayton 1980 Outcome knowledge skills attitude
Council 1978 Outcome crash reduction in ANY vehicle not motorcycles exclusively
Crowther 2005 Study type qualitative study
Daniello 2009 Study type review of the effectiveness of motorcycle training and licensing
Dewanti 2007 Intervention emotional characteristics behaviour (ie not motocycle rider training)
Du Plooy 1990 Study type review of types of rider training courses in South Africa
Engel 1989 Study type uncontrolled before-after study
Goldenbeld 2002 Outcome skills
Goldenbeld 2004 Outcome knowledge and skills
Haworth 2000 Intervention balance between attitudinal andvehicle control skills acrosscurriculla of motorcycle rider training
courses
Haworth 2005 Study type review
Haworth 2006 Study type cross-sectional survey
Hill 1983 Study type review and description of training system
Imatake 1980 Study type course descriptionreview
Ingebritsen 1990 Intervention risk factors for motorcycle crashes
Jain 2009 Study type cross-sectional descriptive study of motorcycle crashes over a 5 year period
Jonah 1979 Study type cross-sectional survey Outcome scores on motorcycle operator skill test
Jonah 1981 Intervention evaluation of a motorcycle rider skill test not motorcycle rider training
Kadar 1990 Study type review of components of rider training courses
Kelsey 1986 Intervention evaluation of a motorcycle rider skill test not motorcycle rider training
Koch 1980 Study type reviewcourse description
Koch 1990 Study type description of course content
42Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4550
(Continued)
Labbett 2005 Study type essay about possible explanations for motorcycle deaths
Law 2005 Intervention targeted motorcycle safety programme not specifically rider training
Lowes 1990 Study type review and descriptions of rider training
Mahon 1976 Study type review of motorcycle rider training methods
Maxwell 1985 Outcomesrecommendations forprogramme operatorsand otherson howthe motorcycle trainingprogramme
works and how it can be improved
Mayhew 1996 Study type review
McKnight 1987 Study type evaluation of methodological and administrative aspects of a motorcycle operator skill test
McPherson 1976 Outcome knowledge and skills
Mohan 2004 Study type review of evidence of interventions to reduce road traffic injuries
Morgan 1980 Study type reviewcourse description
Motoki 1990 Study type proposal and development of rider training curriculum
Nairn 1992 Study type literature review
Newland 1980 Study type description of development of motorcycle safety course
Newland 1982 Study type descriptionreview of a rider training course
Newland 1990 Study type uncontrolled BeforeAfter study
Ormston 2003 Outcome changes in attitude and self reported behaviour
Osga 1979 Outcome knowledge attitudes and skills
Panou 2007 Intervention new simulation tool to teach driverrider
Peterson 1990 Study factor demographic details of course participants completing the course vs those not completing the
course
Prem 1984 Study type experimental
Study factor examines association between skills and performance in tests
Radin Umar 2006 Intervention motorcycle safety programme not specifically rider training
Rockwell 1990 Study type methodologically flawed study type
43Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4650
(Continued)
Rothe 1987 Study type qualitative study of rider training in general
Rowden 2007 Study type qualitative study
RSC Victoria 1993 Study type review of motorcycle rider training and education in Victoria
Russam 1979 Study type review of training programmes in Japan
Saffron 1981 Study type literature review
Schulz 1990 Study factor predictors of motorcycle crashes
Sexton 2004 Study type cross-sectional analytic analysis
Shephard 1986 Study factor reviewdescription of programme
Simpson 1990 Study type review of motorcycle rider trainingeducation
Sokolov 1990 Study factor review of motorcycle riding courses in USSR
Steffens 1988 Outcome self-reported behaviour experiences and researcher observed riding behaviour
Sudlow 2003 Study type review of motorcycle training in the UK
Tasmania 1995 Study type cross-sectional descriptiveprocess evaluation
Tetard 1985 Study type experimental
Outcome detection of indices
Thackray 1980 Outcome instructional effectiveness user acceptance administrative feasibility
Thompson 1994 Study type cross-sectional survey
Outcome attitudes and knowledge
Tobler 1990 Study type review of course curricula
Transport 1989 Intervention syllabus for rider training
Tuan 2007 Study type cross-sectional study (survey) which does not assess the effects of an intervention but examines
motocyclistsrsquo behaviour
Victor 1980 Study type review
Waterer 1990 Study type review of UK rider training issues
Weller 1989 Study type review of studies evaluating motorcycle safety and education measures
44Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4750
(Continued)
Wells 1981 Study type description of different methods of training
Winn 1986 Study type review of rsquohistoryrsquo of rider training
Wisher 1988 Study type cross-sectional survey
Wood 1987 Study type Interrupted Time Series but did not meet the criteria for an ITS as published by the Cochrane
Effective Practice and Organisation of Care Group
Yamazaki 1990 Study type cross-sectional survey
Outcome riding skills skills retention and attitude
Characteristics of ongoing studies [ordered by study ID]
Burgess 2010
Trial name or title UKs National Ride program
Methods Cohort study
Participants Motorcycle offenders
Interventions Police led diversion program
Outcomes Self reported attitudinal and behavioural measures
Starting date
Contact information
Notes Expected to be completed in early 2010
45Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4850
D A T A A N D A N A L Y S E S
This review has no analyses
A P P E N D I C E S
Appendix 1 Search strategy
We searched the following electronic databases
bull Cochrane Injuries Group Specialised Register (searched 22 Sept 2008)
bull CENTRAL (The Cochrane Library 2008 Issue 3)
bull TRANSPORT 1988 to 200706 (latest date available)
bull ATRI Australian Transport Index (to Sept 2008)
bull MEDLINE (Ovid SP) 1950 to Sept 2008
bull EMBASE (Ovid SP) 1980 to Sept 2008
bull CINAHL (EBSCO) 1982 to Sept 2008
bull WHOLIS (World Health Organization Library Information System) (to Sept 2008)
bull PsycINFO (Ovid SP) 1806 to Sept 2008
bull LILACS (BIREME) (Latin American and Caribbean Health Sciences) (1982 to Sept 2008)
bull ISI Web of Science Social Sciences Citation Index (SSCI) 1970 to Sept 2008
bull ERIC (searched 25 Sept 2008)
bull ZETOC (searched 25 Sept 2008)
bull SIGLE (searched 25 Sept 2008)
All of the databases were searched using the following text-word search strategy The strategy was adapted using either rsquorsquo or rsquo$rsquo as atruncation symbol
1 motor-cycl or motorcycl or motor-bik or motorbik or motor-scooter or motorscooter or scooter or moped or motocycl
or motocicl or moto
2 train or educat or instruct or program or course or qualif or learn or improv
3 1 and 2
H I S T O R Y
Protocol first published Issue 2 2005
Review first published Issue 10 2010
46Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4950
C O N T R I B U T I O N S O F A U T H O R S
Katina Kardamanidis was the first author of the protocol conducted the literature search extracted full studies assessed the studies
according to inclusion criteria assessed quality of the studies compared findings with the second author discussed disagreements with
the third author wrote the drafts and the final review
Rebecca Ivers was the second author of the protocol assessed the studies according to inclusion criteria assessed quality of the studies
compared findings with the first author edited the final draft and provided overall guidance for the process
Alexandra Martiniuk assessed quality of the studies compared findings with the first author discussed disagreements with the third
author and helped write the first draft and the final review
Mark Stevenson contributed to the protocol reviewed methods of studies and contributed to the final draft
Katrina Thistlethwaite conducted a second literature search extracted full studies and contributed to the final draft
D E C L A R A T I O N S O F I N T E R E S T
None known
S O U R C E S O F S U P P O R T
Internal sources
bull No sources of support supplied
External sources
bull National Health and Medical Research Council of Australia Australia
Salary support for Martiniuk Ivers and Stevenson
bull University of Sydney Sesqui Research Grant Australia
Direct research costs and salary support for Kardamanidis
D I F F E R E N C E S B E T W E E N P R O T O C O L A N D R E V I E W
There are no differences between the protocol and the review
I N D E X T E R M S
Medical Subject Headings (MeSH)
lowastMotorcycles Accident Prevention [lowastmethods] Accidents Traffic [lowastprevention amp control] Licensure Program Evaluation Wounds
and Injuries [mortality lowastprevention amp control]
47Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 5050
MeSH check words
Humans
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2050
Crashes
Total group of participants
Anderson 1980 found a reduction in motorcycle crash rates in
the intervention group (MOST skill test with remedial training)
compared to the control group (who received the skill testwithout
a training component) The percent differenceranged from 143
six months after the programme (P lt 005) to 211 (P lt 0001)
after one year and 127 (P lt 001) after two years Hall 1988
found no statistically significant difference in crash rates between
intervention groups at any point in time
Subgroup of participants who obtained their licence only
Anderson 1980 found a reduction in motorcycle crash rates in the
intervention arm but this was statistically significant at one year
only (24 to 218 reduction P lt 005) Hall 1988 found no
statistically significant differences
Offences
Total group of participants
The percentage of offences in Anderson 1980 after six months
and one and two years was slightly higher in the intervention
arm compared to the control group but these differences were
not statistically significant Likewise the study by Hall 1988 did
not find a statistically significant difference in offences between
groups
Subgroup of participants who obtained their licence only
Hall 1988 found no significant difference in offence rates between
the intervention and control groups Anderson 1980 found thatthe percentage of offences increased in the intervention arms com-
pared to the control group This difference was statistically signif-
icant after two years (51 to 115 increase P lt 005)
Cohort studies
Crashes
Neitherof the cohort studies (Kloeden1994 and Billheimer1998)
found a statistically significant difference in crash rates between
the intervention and control groups
Offences
Billheimer 1998 (for the California Motorcyclist Safety Program
course) and Perrino 2002 did not find a statistically significant
difference in offence rates between the intervention and control
groups
Case-control studies
Crashes
MAIDS 2004 found that a slightly higher percentage of controls
(not involved in a crash) had had pre-license rider training than
cases (involved in a crash) This difference was not statistically significant (50 of controls vs 477 of cases OR 091 95 CI
076 to 110)
Pre-licence rider training - non-mandatory
Randomised trial
Injuries and deaths
Swaddiwudhipong 1998 found a significantly lower injury ratein the intervention villages (1051000 population 95 CI 92
to 120) compared to the control villages (1691000 population
95 CI 152 to 185) one year after the intervention This differ-
ence was not maintained two years after the intervention (127
95 CI 112 to 143 intervention 151 95 CI 135 to 167
control) There was no significant difference in death rates be-
tween groups
Cohort studies
Crashes
Cooper 1988 foundno statistically significant differencein crashes
between the intervention and control groups McDavid 1989
found a lower crash rate in the intervention group (14139) com-
pared to the control group (23139) but no test of statistical sig-
nificance was undertaken
Injuries and deaths
McDavid 1989 found that the intervention group had fewer in-
jurious crashes (814 crashes) compared to the control group (16
23 crashes) but no statistical significance test was reported
18Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2150
Offences
Cooper 1988 foundno statistically significantdifference in offencerates between the intervention and control groups McDavid 1989
foundno statistically significantdifference in offence ratesbetween
the intervention and control groups
Non-mandatory rider training
Non-randomised trials
Crashes
Neither Adams 1985 nor Leung 1987 found a statistically signif-icant effect of rider training on crash rates
Offences
Leung 1987 found a statistically significant lower offence rate in
the intervention group compared to the control group for both
males and females and for all age groups except for women 35
years or older (P values not reported)
Cohort studies
Crashes
Osga 1980 in the univariate analysis reported a lower percentage
of crashesin theintervention group (that continued to ride) (28
168595) than in the control group (34 112329) In the cu-
mulative crash rate analysis however Osga1980 found statistically
significant (at the P lt 001 level) higher crash rates in the interven-
tion group than in the control group at allmileage intervals (1000
to 29999 miles) travelled After riding more than 30000 miles
(when over 50 of the trained riders had been involved in a crash
compared to approximately 40 in untrained riders) this differ-
ence was no longer statistically significant Satten 1980 found a
lower percentage of crashes in the intervention group (13 969)
compared to the control group (25 2079) but the differencewas not assessed statistically Davis 1997 found a lower crash rate
in trained riders but it was unclear how this was calculated be-
cause the size of the control group was not reported Jonah 1982
and Billheimer 1998 (for the Experienced Rider Course) found no
statistically significant difference in crashes between the interven-
tion and control groups Although the difference between those
who completed the Experienced Rider Course and no course in
Billheimer 1998 neared significance at six months after the course
(P = 007) large P values were observed for one year after (P =
030) Raymond 1979 found a statistically significant higher crash
rate for the intervention group (1861000 miles ridden) com-
pared to the control group (1181000 miles ridden) in both males
and females (P lt 0001) especially during the first 5000 miles of
riding Waller 1992 found a statistically significant higher crashrate for the intervention group (27 crashes in three years or 070
crashes10000 person days) compared to the control group (5
crashes in three years or 013 crashes10000 person days) with a
crude relative risk of 538 (X 2=1513 plt005) Mortimer 1984
and Mortimer 1988 found no statistically significant difference in
crashes between the intervention and control groups
Savolainen 2007 found that the intervention group (those who
completed the Basic Rider Course) was 44 more likely to be
involved in crashes than the control group and those who com-
pleted the course multiple times were an additional 180 more
likely to be involved in a crash This finding was based on low
numbers and no P value or confidence level was given for either
result
Injuries and deaths
Satten 1980 also found a lower percentage of injurious crashes
(33 39) in the intervention group compared to the control
group(60 1220) although again no statistical significance test
was reported
Offences
Jonah 1982 did find a significantly lower self-reportedoffence rate
in the intervention group (15) compared to the control group(34 P lt 001) (no numbers given) Satten 1980 also found
a lower self-reported offence rate in the intervention group (1
71 versus 19100) but again this was not tested for statistical
significance Mortimer 1984 Mortimer1988 and Billheimer 1998
found no statistically significant differencein offence ratesbetween
the intervention and control groups
Case-control studies
CrashesMAIDS 2004 found that a slightly higher percentage of cases
(involved in a crash) had had additional non-mandatory rider
training than controls (not involved in a crash) This difference
was not statistically significant (17 of cases vs 12 of controls
OR 146 95 CI 064 to 337)
Rider training unspecified
Case-control studies
19Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2250
Injuries and deaths
Hurt 1981 found that the odds of injurious crashes were lower inthose who had received professional training compared to those
who had not Kraus 1975 found that in the younger age group
(less than 20 years) and the older age group (over 35 years) a larger
percentage of injured riders (12 and 21 respectively) reported
prior motorcycle rider training compared with the uninjured con-
trol group (6 and 5 respectively) (P lt 001) (no numbers
given) Haworth 1997 reported several different outcomes one
beingthat having completed at leastone trainingcoursecompared
to no training courses at all had no significant effect on the odds
of crashing (afteradjustment for age blood alcohol level or licence
status (learner permitprobationaryfull)) When comparing three
different levels of training courses (beginner intermediate and ad-
vanced) with no training no significant differences were found
in the odds of crashing However after adjustment for age and
blood alcohol level completing an advanced course as compared
to a beginners course was associated with a significant decrease in
the odds of severe motorcycle crashes resulting in serious injury or
death (OR 02 95 CI 01 to 08)
D I S C U S S I O N
The objective of this review was to quantify the effectiveness of
pre- andpost-licence motorcycle rider training on the reduction of
traffic crash involvement injuries deaths and offences committedby motorcycle riders Most included studies were cohort studies
There were a few randomised and non-randomised trials and case-
control studies
All studies compared either mandatory or non-mandatory formal
training with no formal training except for Haworth 1997rsquos case-
control study that compared levels of training (beginner interme-
diate and advanced) to each other
We found studies that evaluated the effectiveness of manda-
tory pre-licence training in reducing crash and offence rates
(Anderson 1980 Hall 1988 Kloeden 1994 Billheimer 1998
(the basic California Motorcyclist Safety Program component of
the study) Perrino 2002) non-mandatory training that when
passed resulted in a licence (Cooper 1988 McDavid 1989
Swaddiwudhipong 1998) non-mandatory training that could be
taken by motorcycle riders with or without a licence (Raymond
1979 Osga 1980 Satten 1980 Jonah 1982 Mortimer 1984
Adams 1985 Leung 1987 Mortimer 1988 Waller 1992 Davis
1997 Billheimer 1998 (the Experienced Rider Course compo-
nent of the study)) and unspecified rider training on crash injury
and death rates (Kraus 1975 Hurt 1981 Haworth 1997 MAIDS
2004)
The studies provided conflicting evidence regarding motorcycle
rider training which is likely due to the poor design quality and
reporting of the studies Most of the studies included in the review
were likely to be adversely affected by significant selection biasalong with substantial detection bias(measurement and recall bias)
which limits our ability to make conclusive comments
Mandatory pre-licence rider training
There were two RCTs three cohort studies and one case-control
study that examined pre-licence ridertraining The two RCTs eval-
uated the Motorcycle Skills Operating Test (MOST) (Anderson
1980 Hall 1988) Anderson 1980 found a statistically significant
reduction in crashes in the intervention group at six months one
and twoyearsfollow up However when only participantswho ob-
tainedtheirlicence were included in theanalysis thereduction wasstatistically significant only at one year follow up The other RCT
(Hall 1988) found no statistically significant differences in crash
rates at any point of time neither in the total population group
nor in the licensed subgroup Neither of the two cohort stud-
ies assessing crash rates found a statistically significant difference
in crash rates between intervention and control groups ( Kloeden
1994 Billheimer1998) andthe case-control study (MAIDS 2004)
also did not find a statistically significant difference in the odds of
experiencing a crash between trained and untrained riders With
regards to offences neither of the RCTs found a statistically sig-
nificant difference in offence rate between the intervention and
control groups in the analysis of the total group of participants
entering the trial However Anderson 1980 found a statistically significant increase in offence rates in the intervention group but
only after two years of follow up andonly in the analysis of partici-
pants who obtained their licence The two cohort studies assessing
offence rates found no statistically significant difference between
the groups (Billheimer 1998 Perrino 2002)
None of the five studies (RCTs or cohort) found a reduction in
crashes or offences following mandatory pre-licence rider training
except for Anderson 1980 who observed a statistically significant
reduction in crashes in the intervention group at one year in both
analyses It is not clear why both RCTs that evaluated the same
MOST and training intervention observed different results Both
studies enrolled large samples and observed a similar uptake of
licensure following the intervention It is possible that differencesmay have arisen due to different study samples (California ver-
sus New York) or that the intervention delivery differed between
studies
In both trials of the Motorcycle Skills Operating Test including
rider training only a third of motorcycle licence applicants ob-
tained their motorcycle licence It is possible that a licensing pro-
cess including a skills test may reduce crashes through decreased
exposure
Non-mandatory pre-licence training
20Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2350
Three studies assessed the effect of non-mandatory pre-licence
training The Cooper 1988 cohort study found no difference incrash or offence rates between the trained and untrained groups
We cannot however be confident about this result as the study
sufferedfrom selectionbiaswhichcouldhave biasedthe measure of
effect toward the null The RCT conducted by Swaddiwudhipong
1998 examined injury and death rates and found a statistically
significant lower injury rate in the intervention villages after one
year of follow up but no longer at two year follow up There
was no statistically significant difference in death rates between
intervention and control villages Our confidence in the results of
this study are hindered by the inadequacy of the randomisation
process that led to significant differences between the intervention
and control districts on two important variables The intervention
districts were more likely than the control districts to have a validlicence andto wear a helmetSincethese differences were nottaken
into account in later analyses they could explain the differences
observed in injury rates between districts Furthermore with only
six districts randomised the sample size was probably too small
to detect a difference between intervention and control groups
had there been one The McDavid 1989 cohort study found a
lower percentage of injuries in the intervention group although
a statistical test was not conducted McDavid 1989 drew their
intervention and control groups from different populations but
did adjust for confounders although they only analysed 40 of
the initial sample (139 trained and 139 untrained) Because of
these substantial weaknesses we are unable to draw conclusions
from this studyAn overall effect of pre-licence training on the reductionof crashes
injuries deaths and offences was not found although one should
be aware that the research of this issue to date has had significant
flaws However a mandatory skill test (such as the MOST) in
combination with and even without pre-licence training as part
of a licensing process seems to reduce the number of individuals
completing the licensing procedure and therefore indirectly re-
duces crash rates through a reduction in exposure to motorcycle
riding Making licensure a more difficult process may therefore be
seen as advantageous if it delays licensure and therefore reduces
exposure It is possible that the group who chose not to continue
riding a motorcycle would not have been the group to experience a
crash or commit an offence if they had continued to ride anyhowfor example because they are more safety conscious and this was
the reason for their giving up riding in the first place If this is the
case deterring individuals from completing their licence and rid-
ing may not necessarily decrease overall crashes deaths or offence
rates In addition other potentially adverse effects of complicat-
ing a licensing procedure need to be considered such as increases
in unlicensed riding In Thailand non-mandatory rider training
offered as a community intervention may have decreased injury
rates but this needs to be confirmed through further research
Non-mandatory training
The majority of the studies assessed non-mandatory training
(training independent of any licensing procedure) and the results
were varied The two non-randomised trials assessing crash rates found no statistically significant difference in crash rates between
the intervention and control groups (Adams 1985 Leung 1987)
Of the nine cohort studies Satten 1980 found a lower percentage
of crashesin the intervention group although it is unclear whether
this was a statistically significant result Davis 1997 also found a
lower percentage of crashesin theinterventiongroup although it is
unclear with which control group the intervention group wascom-
pared Raymond 1979 Osga 1980 Waller 1992 and Savolainen
2007 found a statistically significant higher crash rate in the in-
tervention group and the remaining cohort studies (Jonah 1982
Mortimer 1984 Mortimer 1988 Billheimer 1998) found no sta-
tistically significant difference in crash rates between intervention
and control groups No conclusions can be drawn from these stud-
ies with confidence The Satten 1980 and Davis 1997 studies did
not adjust for any confounding factors even though important
differences existed between the intervention and control groups
Although Raymond 1979 and Waller 1992 adjusted for a num-
ber of confounders at baseline (Waller 1992) and in the analy-
sis (Raymond 1979 Waller 1992) the intervention and control
groups were drawn from such different populations that it is pos-
sible that unmeasured confounders were responsible for the sta-
tistically significant increased risk of crashes observed in the inter-
vention group The observed difference could also be attributed totheobserved differential loss to followup (Raymond 1979) which
could bias the measure of effect towards or away from the null
The Osga 1980 study found a protective effect of rider training
in the univariate analysis but statistically significant higher crash
rates in the intervention group when accounting for riding expo-
sure Sex (a significantly higher proportion of females in the in-
tervention group) and other potentially confounding factors that
could explain thedifferencebetween the groups were howeverun-
accounted for Finally Savolainen 2007 did control for potential
confounding variables yet it is not clear whether variables were
controlled for simultaneously or one at a time One case-control
study (MAIDS 2004) found no statistically significant reduction
in the odds of experiencing a crash between trained and untrainedriders but the quality of this study was poor mainly because the
analysis did not account for any confounding factors
One cohort study (Satten 1980) assessed the effect of non-manda-
tory rider training on injurious crashes and found a lower percent-
age of injuries in the intervention group although a statistical test
was not conducted As mentioned above the Satten 1980 studyrsquos
validity is reduced due to its lack of adjustment for confounding
factors
The only non-randomised trial that assessed offence rates (Leung
1987) found a statistically significant lower offence rate in the in-
tervention group and of the five cohort studies that assessed of-
fence rates (Satten 1980 Jonah 1982 Mortimer 1984 Mortimer
21Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2450
1988) one study (Jonah 1982) also found a statistically significant
lower offence rate in the intervention group However we have lit-tle confidence in the results from Leung 1987 since potential con-
founding factors were not appropriately addressed Jonah 1982
did conduct a multivariate analysis but was hindered by an initial
low response rate The other four cohort studies did control for
several potential confounders and found no difference in offence
rates following non-mandatory post-licence rider training
The non-randomised trials and cohort studies evaluating non-
mandatory rider training were at high risk of substantial selection
bias due to the non-random allocation of participants as well
as low participation rates and attrition bias and could therefore
not provide conclusive evidence with regard to the effect of the
training It should be noted that despite the biases and other issues
with this group of studies most didnot detectmajor effects relatedto rider training in either direction This may suggest that any
potential impact of rider training will be modest
Studies examining unspecified modes of ridertraining
Three case-control studies assessed whether injured or deceased
motorcyclists were more or less likely to have had rider training
compared to control motorcyclists All had conflicting results
Hurt1981 foundthe odds of havingan injuriouscrash waslower in
those who had received rider training whereas Kraus 1975 found
that the odds of having an injurious crash were higher in youngerand older trained riders Haworth 1997 who compared advanced
versus beginner training found no effect on motorcycle injuries
or deaths although they did find that those who had completed
advanced training had reduced chances of crashing compared to
those who had done a beginnersrsquo training course Similar issues
with uncontrolled confounding and biases of selection as described
above arose with the three case-control studies Hurt 1981 suffers
from inadequate control of confounding via poor matching and
both the Kraus 1975 and the Haworth 1997 findings were only
controlled for limited factors by means of stratified analysis It is
likely that these limitations are the reason for the contradictory
directions of effect observed for the three case-control studies
Methodological weaknesses of the studies
In summary most studies investigating the effectiveness of mo-
torcycle rider training on reducing crashes injuries and death suf-
fered from serious methodological weaknesses
The majority of studies were non-randomised and were there-
fore limited mainly by selection bias Although the approaches of
matching stratification or regression analyses can minimise bias
(Fleiss 2004) they were poorly employed in the reviewed studies
The inadequate adjustment for confounders in particular adds
to our lack of confidence in the studiesrsquo results since it has been
demonstrated that individuals who choose to take motorcycle rider
training differ from those not taking rider training For instance
trained riders often differ from controls in age sex and experi-ence usually ride less and may have a more positive attitude to
safety as indicated by the reported more frequent wearing of seat
belts (Mortimer 1982)
The two large randomised trials reported that a large proportion
of their study group were never exposed to the risk of offences or
crashes because they never attained licensure This made it impos-
sible to assess whether a reduction in crashes in the intervention
group was a result of training or due to delayed or never attained
licensure
In some studies the sample sizes were too small to detect any sta-
tistically significant differences Follow-up time after training dif-
fered between studies and was an impediment to observing any
real effects of training if they existed As riding experience ac-cumulates over time any differences in outcomes between those
trained and untrained would diminish in studies with lengthy fol-
low-up times This dilution of any possible effect of rider training
may have also occurred in studies that mixed both novice and ex-
perienced riders in the study groups since it is hypothesised that
novice riders serve to benefit most from rider training This would
increase the difficulty in finding any effect of rider training par-
ticularly when sample sizes are small
Detection bias may also account for some of the differences ob-
served between trained and untrained riders Detection bias would
have occurred if the self-report hospital driver or police records
used to record the outcomes of crash injury death or offences
differed between the trained and untrained riders Research hasdemonstrated that not all crashes are reported to police for ex-
ample small crashes or off-road crashes (Kraus 1975) even serious
injury crashes have been found to be underreported (Anderson
1980) However this would only introduce detection bias if the
reporting of outcome differs for trained and untrained riders This
does appear possible in that responsible riders may be more likely
to enroll in a rider training course and also may be more likely
to report crashes to police than irresponsible riders meaning that
the number of crashes from riskier riders could be underrepre-
sented on records An analysis to investigate the effectiveness of
rider training may then show that rider training is not effective in
reducing crashes
Rather than using reported outcomes some studies used self-re-ported survey data although this technique suffers from non-
response biases In fact two studies reported differences in de-
mographic characteristics in response rates to questionnaires
(Anderson 1980 Hall 1988) Self-reported data also introduce
the risk of recall bias survival bias and social desirability response
bias Recall bias may have been equally distributed among trained
and untrained riders howeverit is hypothesised that trained riders
may have felt it was lsquoexpectedrsquo to report good driving histories
hence leading the rider training to appear effective
22Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2550
Limitations of this review
Limitations of this review were that we found very few recentstudies (the two most recent ones published in 2002 and 2007)
and that we found very few studies conducted in low-and middle
income countries even though our search strategy was set up to
identify such studies Anotherlimitation wasthat forthe reviewed
studies we could not assess all outcomes of rider training that are
potentially useful Some of the included studies assessed outcomes
such as use of safety clothing including helmets Such outcomes
are valuable since the use of motorcycle helmets has been demon-
strated to be effective in preventing head injuries (Liu 2004) De-
spite this the use of safety clothing was not included as an out-
come in this review because the studies measuring this outcome
were limited by poor data quality due to very low response rates
Another outcome of interest in assessing the effectiveness of rider
training is lsquotype of injuryrsquo As Haworth 2005 state in their review
of motorcycle licensing and training rider training could be con-
sidered effective if injuries become less severe because motorcycle
riders have learned how to avoid more serious injuries through
rider training The studies included in this review did not specify
type of injury and so this outcome could not be assessed Last
changes in knowledge and attitudes following rider training are
other potentially useful outcomes We have excluded such studies
(Osga 1979 Clayton 1990 Goldenbeld 2004) as these outcomes
are proxies rather than direct measures of the effectiveness of mo-
torcycle rider training We opted to report on the outcomes of
crashes injury death and offences alone since there is little ev-
idence about how changes in knowledge and attitudes relate to
these outcomes
A U T H O R S rsquo C O N C L U S I O N S
Implications for practice
Due to the poor quality of studies included in this review we were
unable to draw any conclusions about the effectiveness of rider
training on crash rates injurious or fatal or offence rates The
findings do suggest that mandatory pre-licence training may be
an impediment to completing a motorcycle licensing process thuspossibly indirectly reducing crash injury death and offence rates
through a reduction in exposure to riding a motorcycle On the
basis of the existing evidence it is not clear if and what type of
training reduces the risk of crashes injuries deaths or offences in
motorcyclists and the selection of the best rider training practice
can therefore not be recommended The suggestion that manda-
tory pre-licence training indirectly reduces exposure to riding the
motorcycle because trainees are impeded from completing the li-
censing process does not help those motorcyclists who persist in
obtaining their licence and riding the motorcycle Some type of
rider training is likely necessary to teach motorcyclists to ride a
motorcycle safely Good policy is based on sound evidence and
evaluation but such evidence for motorcycle rider training is seri-
ously lacking Policy makers are therefore encouraged to include arigorous evaluation component to any new or existing rider train-
ing programme Such an evaluation should ideally be randomized
and assess whether the training programme is effective and if so
which component(s) of the programme work best
Implications for research
Given theresults of this review which found littlereliable evidence
about the effectiveness of motorcycle rider training it is necessary
to build up an evidence base of the effectiveness of motorcycle
rider training in the reduction of crash injury death and offence
rates To this end well designed randomised controlled trials need
to be conducted that examine the impact of rider training oncrashes and offences A randomised controlled trial is the preferred
method as it would significantly reduce most biases described in
this review To minimise the impact of participants delaying or
never attaining their licence and therefore exposure as found in
the randomised trials in this review such a trial should include
riders who are already licensed and reliable measures of riding
exposure in a reasonably lengthy follow up period It is expected
that a randomised controlled trial of sufficient sample size would
observe equal balance on possible confounders by study arm how-
ever an assessment of potentially important confounders (such as
helmet wearing) by study group would enable control for these
variables in the analyses if necessary It is imperative that the effec-
tiveness of each component of the rider training (such as theorypractice knowledge skills and modification of behaviour) and the
synergy of such components are evaluated Outcomes should be
objective ideally those that are routinely collected rather than self-
reported outcomes Even though riding the motorcycle poses dif-
ferent challenges and incurs different risks lessons may be learned
from research into the effects of car driver training
Motorcycles are a common means of transport in low- and middle
income countries and usage is increasing As the burden of mo-
torcycle crash victims is mostly borne in such countries research
should be focused on the effectiveness of types of rider training on
the reduction of crashes injuries and fatalities in such countries
and be supported by higher income countries This review found
only one study conducted in a middle income country that sug-gested decreasing injury rates It would be beneficial to conduct
additional research to confirm this suggestion and to enhance our
knowledge of the role of rider training in low- and middle-income
settings
March 2006 saw the launch of the University of North Carolina
Highway Safety Research Centre partnership with the Motorcycle
Safety Foundation (MSF) and plans were revealed to conduct a
study of the potential benefits of motorcycle rider training The
study is funded by the MSF and the National Highway Traffic Sa-
fety Administration (NHTSA) Called rsquoThe Discovery Projectrsquo it
will compare data on crash rates from riders participating in a basic
23Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2650
motorcycle safety class with riders participating in a basic motor-
cycle safety class as well as a variety of continuing education classesThe collection of pilot data was due to start in March 2007 and
the data collection portion of the study is expected to take three
years (Highway Safety Research Center) It is vital that researchers
and policy makers collaborate to develop furtherhigh quality eval-
uations of motorcycle rider training programmes to help inform
development of new and effective training programmes
A C K N O W L E D G E M E N T S
Staff from The University of Sydney library for helping us locate
and retrieve full reports of many papers
R E F E R E N C E S
References to studies included in this review
Adams 1985 published data only
Adams A Collingwood V Job RFS Evaluation of the 1982 Australia
Post motorcycle rider training scheme NSW Australia Traffic
Accident Research Unit Traffic Authority NSW 1985 [ RN 85]
Anderson 1980 published data only
Anderson J Ford JL Peck RC Improved motorcyclist licensing and testing project Washington DC National Highway Traffic Safety
Administration (NHTSA) USA 19801ndash217 [
DOTndashHSndash805ndash598]
Billheimer 1998 published data only
Billheimer JW Evaluation of the California Motorcyclist Safety
Program Transportation Research Record 19981640100ndash9 [
TRB Report Number 980652]
Cooper 1988 published data only
Cooper PJ Rothe JP Motorcyclists who they are and why they do
what they do Transportation Research Record 1988116878ndash85
Davis 1997 published data only
David CF Evaluation of community traffic safety programs and
motorcycle operator training programs Connecticut Department of Transportation 199763
Hall 1988 published data only
Hall CJ Buchanan L Engle R Motorcycle rider education evaluationproject Washington DC USA National Highway Traffic Safety
Administration (NHTSA) US Department of Transportation
1988 [ DTNH 22ndash80ndashCndash05123]
Haworth 1997 published data only
Haworth N Smith R Brumen I Pronk N In Monash University
Accident Research Centre editor(s) Case-control study of motorcycle
crashes Victoria Australia Department of Transport and Regional
Development The Federal Office of Road Safety 19971ndash149 [
ISBN 0 642 25501 6 ]
Hurt 1981 published data only
Hurt H Ouellet J Thom D Motorcycle accident cause factors and
identification of countermeasures USA Department of
Transportation 1981 [ Technical Report DOT HS 806ndash862]
Jonah 1982 published data onlylowast Jonah BA Dawson N Bragg B Are formally trained motorcyclists
safer Accident Analysis amp Prevention 198214(4)247ndash55
Jonah BA Dawson NE Bragg BW Evaluation of the Canada
Safety Councilrsquos Motorcycle Training Program Proceedings of theInternational Motorcycle Safety Conference Washington DC
USA 198059ndash93
Kloeden 1994 published data only
Kloeden CN Moore VM McLean AJ Evaluation of the pre-license
training program for motorcyclists in South Australia Walkerville
South Australia Australia NHMRC Road Accident Research Unit
The University of Adelaide 19941ndash37 [ ISBN 0 7308 0607 3]
Kraus 1975 published data only
Kraus JF Riggins RS Franti CE Some epidemiological features of
motorcycle collision injuries American Journal of Epidemiology 1975102(1)74ndash98
Leung 1987 published data only
Leung KS Reding VA Evaluation of the Wisconsin motorcycle rider course Wisconsin Bureau of Policy Planning and Analysis
Wisconsin Department of Transportation 1987
MAIDS 2004 published and unpublished data
MAIDS In-depth investigations of accidents involving powered
two wheelers ACEM (Acemacemeu) 2004 Vol Final report
20179
McDavid 1989 published data only
McDavid JC Lohrmann BA Lohrmann G Does motorcycle
training reduce accidents Evidence from a longitudinal quasi-
experimental study Journal of Safety Research 198920(2)61ndash72
Mortimer 1984 published data only
Mortimer R Evaluation of the motorcycle rider course Accident
Analysis amp Prevention 198416(1)63ndash71
24Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2750
Mortimer 1988 published data only
Mortimer RG A further evaluation of the motorcycle rider courseJournal of Safety Research 198819(4)187ndash96
Osga 1980 published data only
Osga GA An investigation of the riding experiences of MSF rider
course participants Dissertation South Dakota USA The
University of South Dakota 19801ndash224
Perrino 2002 published data only
Perrino CS Ahmed A Callender A Rozier E Cantwell A Stewart
O The role of Marylandrsquos motorcycle rider course in promoting saver
behaviors and attitudes Baltimore USA National Transportation
Center Morgan State University 20021ndash52
Raymond 1979 published data only
Raymond S Tatum S An evaluation of the effectiveness of the RAC
ACU motorcycle training scheme - final report Road Safety ResearchUnit Dept of Civil Engineering University of Salford London
19791ndash131
Satten 1980 published data only
Satten RS Analysis and evaluation of the motorcycle rider courses
in thirteen Illinois counties International Motorcycle Safety
Conference Washington DC 1980145ndash93
Savolainen 2007 published and unpublished data
Savolainen Peter Mannering Fred Additional evidence on the
effectiveness of motorcycle training and motorcyclistsrsquo risk-taking
behaviour Transportation Research Board 2007 Vol March 20
19
Swaddiwudhipong 1998 published data only
Swaddiwudhipong W Boonmak C Nguntra P Mahasakpan P
Effect of motorcycle rider education on changes in risk behavioursand motorcycle-related injuries in rural Thailand Tropical Medicine
and International Health 19983(10)767ndash70
Waller 1992 published data only
Waller FL An analysis of the impact of the Texas Department of
Public Safety Motorcycle Operator Training Program upon
motocycle related traffic accidents Texas State University Public
Administration Program Applied Research Projects 199277
References to studies excluded from this review
Anderson 1980a published data only
Anderson LC Better biking a motorcycle rider improvement
program International Motorcycle Safety Conference Washington
DC USA Motorcycle Safety Foundation 1980231ndash53Aupetit 2008 published and unpublished data
Aupetit S Riff J Buttelli O Espie S Ergonomics research program
on motorcycle learning Contributions to the development of a
motorcycle simulator for training Transport Research Arena
Ljubljana 20089
Awane 1999 published data only
Awane T Integrating simulators in motorcycle safety education
Journal of International Association of Traffic and Safety Sciences 199923(1)26ndash35
Baldi 2005 published data only
Baldi S Baer JD Cook AL Identifying best practices states in
motorcycle rider education and licensing Journal of Safety Research200536(1)19ndash32
Batchler 1988 published data only
Batchler MW Evaluation of the pre-license motorcycle training course at the New South Wales traffic education centre Federal Office of
Road Safety (FORS) 19881ndash122 [ Report No CR 77 ]
Blanchard 2006 published data only
Blanchard HT Tabloski PA Motorcycle safety educating riders at
the teachable moment Mosby (CV) Company 200632330ndash2
Bonnett 2005 published data only
Bonnett J Mature age skills training for experienced motorcycle
riders Motorcycle ridersrsquo association of the Australian Capital
Territory 2005
Braver 2007 published data only
Braver ER et alPersuasion and licensure a randomized controlled
intervention trial to increase licensure rates among Maryland
motorcycle owners Traffic Injury Prevention 20078(1)39ndash46Clayton 1980 published data only
Clayton AB Sudlow DE An evaluation of the effectiveness of a one day training programme for learner motorcyclists Berkshire UK
Transport and Road Research Laboratory 19801ndash23 [ ISSN
0266ndash7045]lowast Clayton AB Sudlow DE An evaluation of the effectiveness of a one-day training programme for learner motorcyclists Berkshire UK
Road Safety Division Berkshire 19801ndash23 [ISSN 0266ndash7045]
Council 1978 published data only
Council FM Desper LP Dutt AK Driver education for motorcycle operation University of North Carolina Highway Safety Research
Center 19781ndash40
Crowther 2005 published data onlylowast Crowther G Engaging with motorcyclists UK police and theBikeSafe Road Safety Programme Published proceedings of the
Australasian Road Safety Research Policing Education Conference
Wellington New Zealand 2005 issue ISBN 0ndash473ndash10636ndash11ndash11
Daniello 2009 published data only
Daniello A Gabler HC Mehta YA Effectiveness of motorcycle
training and licensing Transportation Research Board 88th Annual
Meeting 200913
Dewanti 2007 published data only
Dewanti M Challenging efforts in promoting young motorcyclist
safety in Indonesia Road Safety on Four Continents 14th
International Conference Swedish National Road and Transport
Research Institute 2007
Du Plooy 1990 published data onlyDu Plooy DW Motorcycle training in South Africa The Human
Element International Motorcycle Safety Conference Orlando
Florida USA Motorcycle Safety Foundation 19904-1 to 4-9
Engel 1989 published data only
Engel U Krogsgaringrd Thomsen L [Faeligrdselsundervisning og
propaganda ndash en faeligrdselssikkerhedsfremmende foranstaltning]
Dansk Vejtidsskrift 198912295ndash7
Goldenbeld 2002 published data only
Goldenbeld C Houwing S de Craen S The development of
driving skills of young moped riders [De ontwikkeling van de
rijvaardigheid van jonge bromfietsers] SWOV Institute for Road
Safety Research (Stichting Wetenschappelijk Onderzoek
Verkeersveiligheid) Leidschendam the Netherlands 200263
25Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2850
Goldenbeld 2004 published data only
Goldenbeld C Twisk D de Craen S Short and long term effects of moped training a field experiment Transportation Research Part F 20047(1)1ndash16
Haworth 2000 published data only
Haworth N Smith R Kowadlo N Motorcycle training in Victoria
evaluation of rider training curriculum in Victoria Monash
University Accident Research Centre Reports 2000 Vol 165 issue
3131
Haworth 2005 published data only
Haworth N Mulvihill C Review of motorcycle licensing and
training Monash University Accident Research Centre 2005 [
Report no240]
Haworth 2006 published data only
Haworth N Mulvihill C Rowdon P Teaching old dogs new tricksTraining and older motorcyclists Journal of the Australasian College of Road Safety 200718(4)20ndash5lowast Haworth NL Mulvihill C Rowden PJ Teaching old dogs new
tricks Training and older motorcyclists Australasian Road Safety
Research Policing and Education Conference Gold Coast
Queensland 2006
Hill 1983 published data only
Hill D Motorcycle rider training testing and licensing Motor
Cycle Council of New South Wales Sydney 1983
Imatake 1980 published data only
Imatake Y On the activities in Japan of motorcycle safe riding
promotion committee (MSRPC) - especially the education of
unskilled riders of class 1 motorbikes International Motorcycle
Safety Conference Washington DC USA Motorcycle Safety
Foundation 1980255ndash68
Ingebritsen 1990 published and unpublished data
Ingebrigtsen S [Risikofaktorer ved ferdsel med moped og
motorsykkel] TOslashI-rapport 66 Oslo Transportoslashkonomisk
institutt 1990
Jain 2009 published and unpublished data
Jain A et alTwo wheeler accidents on Indian roads a study from
Mangalore India Elsevier 200916130ndash3
Jonah 1979 published data only
Jonah BA Dawson NE Validation of the motorcycle operator skill
test Accident Analysis amp Prevention 197911(3)163ndash71
Jonah 1981 published data onlyJonah BA Dawson NE Bragg WE Predicting accident
involvement with the Motorcycle Operator Skill Test Accident Analysis amp Prevention 198113(4)307ndash418
Kadar 1990 published data only
Kadar KJ Countermeasures motorcycle rider education delivery
system The Human Element International Motorcycle Safety
Conference Motorcycle Safety Foundation 19906-27 to 6-41
Kelsey 1986 published data only
Kelsey SL Liddicoat C Ratz M Licensing novice motorcyclists a
comparison of Californiarsquos standard test and the MOST II (Motorcycle
Operator Skill Test) administered at centralized testing offices Sacramento California USA California Department of Motor
Vehicles Research and Development Office 19861ndash74
Koch 1980 published data only
Koch H Advanced course for motorcycle riders InternationalMotorcycle Safety Conference Washington DC USA Motorcycle
Safety Foundation 1980269ndash81
Koch 1990 published data only
Koch H Brendicke R How to influence risk taking behaviour
within motorcycle rider training programs The Human Element
International Motorcycle Safety Conference Orlando Florida
USA Motorcycle Safety Foundation 1990Motorcycle Safety
Foundation 1990 p 9-1 to 9-25A
Labbett 2005 published data only
Labbett S Langham M Training can make the problem worse
Road Safety Congress 20051ndash12
Law 2005 published and unpublished data
Law TH Radin Umar RS Zulkaurnain S Kulanthayan S Impactof the effect of economic crisis and the targeted motorcycle safety
programme on motorcycle-related accidents injuries and fatalities
in Malaysia International Journal of Injury Control and Safety
Promotion 200512(1)9ndash21
Lowes 1990 published data only
Lowes BF In-traffic rider training The Human Element
International Motorcycle Safety Conference Orlando Florida
USA Motorcycle Safety Foundation 19907-46 to 7-57
Mahon 1976 published data only
Mahon J Teaching accident avoidance on the motorcycle Journal of Traffic Safety Education 1976231ndash29
Maxwell 1985 published data only
Maxwell GS Evaluation of a Program for Training MotorcycleRiders Evaluation and Program Planning 19858(4)339ndash47
Mayhew 1996 published data only
Mayhew DR Simpson SH Effectiveness and role of driver educationand training in a graduated licensing system Ottowa Canada The
Traffic Injury Research Foundation 19961ndash89 [ ISBN
0ndash920071ndash12ndash0]
McKnight 1987 published data only
McKnight AJ Milburn BG An evaluation of the modified
motorcycle operator skill test exercises NHTSA Report 1987 [
Report No DOT HS 807 212 ]
McPherson 1976 published data only
McPherson K McKnight AJ The development and evaluation of a
motorcycle skill test manual and knowledge test Washington DCUSA National Highway Safety Administration 1976
Mohan 2004 published data only
Mohan D Evidence-based interventions for road traffic injuries in
South Asia Journal of the College of Physicians and SurgeonsPakistan 200414(12)746ndash7
Morgan 1980 published data only
Morgan JK Advanced motorcycle training course content and
study evaluation International Motorcycle Safety Conference
Washington DC USA Motorcycle Safety Foundation 1980
283ndash312
Motoki 1990 published data only
Motoki M Yamazaki S A study on effective motorcycle rider
education The Human Element International Motorcycle Safety
26Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2950
Conference Orlando Florida USA Motorcycle Safety
Foundation 19909-26 to 9-57A
Nairn 1992 published data onlylowast Nairn RJ Motorcycle safety research literature review 1987 to
1991 Federal Office of Road Safety Australia 1993 Vol Report
CR 117
Newland 1980 published data only
Newland R Darby J Automotive practicespre motorcycle rider
program International Motorcycle Safety Conference Washington
DC USA Motorcycle Safety Foundation 1980127ndash9
Newland 1982 published data only
Newland R Motorcycle rider training and licensing requirements
Driver training steering a course for the future Melbourne
Victoria Transport Regulation Board 1982
Newland 1990 published data only
Newland R Motorcycle training and licensing in Victoria 1980-
1990 The Human Element International Motorcycle Safety
Conference Orlando Florida USA Motorcycle Safety
Foundation 1990
Ormston 2003 published data only
Ormston R Dudleston A Pearson S Stradling Evaluation of
Bikesafe Scotland Scottish Executive Social Research Edinburgh
2003
Osga 1979 published data only
Osga GS Ellingstad VS An evaluation of the effectiveness of the 1978
South Dakota motorcycle rider course program South Dakota USA
University of South Dakota 19791ndash89
Panou 2007 published data only
Panou M Bekiaris E A new concept on the integration of driving
simulators in driving training the train-all approach Road Safety
on Four Continents 14th International Conference Swedish
National Road and Transport Research Institute 2007
Peterson 1990 published data only
Peterson A Kadar K Status report 1980-1989 Minnesota rider
education statistics The Human Element International
Motorcycle Safety Conference Orlando Florida USA 199013-
113 to 13-129
Prem 1984 published data only
Prem H Good MC Motorcycle rider skills assessment ACT
Australia Office of Road Safety Department of Transport 1984
1ndash377 [ Report No CR 34 ]
Radin Umar 2006 published data only
Radin Umar RS Motorcycle safety programmes in Malaysia how
effective are they International Journal of Injury Control and Safety Promotion 200613(2)71ndash9
Rockwell 1990 published data only
Rockwell TH Kiger S Carnot M An evaluation of the Ohio
motorcyclist enrichment program Ohio Department of Highway
Safety 1990
Rothe 1987 published data only
Rothe JP A motorcycle riding course Motorcyclists Image and reality Canada Insurance Corporation of British Columbia 1987
59ndash83
Rowden 2007 published data only
Rowden P Watson B Haworth N What can riders tell us aboutmotorcycle training A view from the other side of the fence
Australasian Road Safety Research Policing and Education
Conference Melbourne Australia Melbourne 2007
RSC Victoria 1993 published data only
Road Safety Committee Inquiry into Motorcycle Safety in Victoria
Parliament of Victoria 1993
Russam 1979 published data only
Russam K Motorcycle training and licensing in Japan UK Transport
and Road Research Laboratory (TRRL) 1979 [ Report 916 ]
Saffron 1981 published data only
Saffron DG Driver and rider education training and licensing a
brief review Traffic Accident Research Unit Traffic Authority of
New South Wales NSW Australia Traffic Accident Research UnitTraffic Authority of New South Wales 1981
Schulz 1990 published data only
Schulz U Motorcycle accidents among learner riders Motorcycle
licensing Evaluation of the effectiveness of a Graduated Licensing Scheme Forshungshefte des Institut fur Zweiradsicherheit 1990
Sexton 2004 published data only (unpublished sought but not used)
Sexton B Baughan C Elliott M Maycock G The accident risk of
motorcyclists 2004TRL Report TRL 607(ISSN 09684107)58
Shephard 1986 published data only
Shepherd R Do Motorcycle Safety Foundation programs work
Driver 198619(9)16ndash8
Simpson 1990 published data onlySimpson HM Mayhew DR The promotion of motorcycle safety
training education and awareness Health Education Research
Theory and Practice 19905(2)257ndash64
Sokolov 1990 published data only
Sokolov O Safe motorcycle riding some aspects of motorcycle
riding in the USSR The Human Element International
Motorcycle Safety Conference Motorcycle Safety Foundation
19904ndash37 - 4-48
Steffens 1988 published data only
Steffens U Gawatz R Willmes G Effectiveness of mofa courses
[Wirksamkeit von Mofakursen] Unfall- und sicherheitsforschung
strasenverkehr 198867168
Sudlow 2003 published data only
Sudlow D Scoping study on motorcycle training Road Safety
Research Report London UK The British Institute of Traffic
Education Research Department for Transport 2003 Vol June
[ Report No 36]
Tasmania 1995 published data only
Mannering FL Grodsky LL Motorcycle rider training scheme
Traffic Education Research Centre Armidale Australia Driver
Training Australia Limited 1995 Vol April1ndash56
Tetard 1985 published data only
Tetard C Motorcycle ridersrsquo training to detect indices in an
intersection situation [Lrsquoentrainement a la prise drsquoinformation dans
lrsquoapprentissage de la conduite motocycliste] Le Travail Humain198548(2)147ndash59
27Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3050
Thackray 1980 published data only
Thackray RM Prescott JC Field test of the motorcycle ridercourse International Motorcycle Safety Conference Washington
DC USA Motorcycle Safety Foundation 1980195ndash7
Thompson 1994 published data only
Thompson M Evaluation of compulsory basic training for
motorcyclists Transport Research Laboratory Crowthorne
Berkshire UK 1994 [ Report 63]
Tobler 1990 published data only
Tobler U Individual driving instruction for motorcycle driving
learners the Swiss experiment of motorcycle driving school The
Human Element International Motorcycle Safety Conference
Orlando Florida USA Motorcycle Safety Foundation 1990
Transport 1989 published data only
Transport Communications Road Crash Statistics Australia Bureauof Transport and Communications Economics 198914ndash20
Tuan 2007 published data only
Tuan VA Shimizu T Towards development and evaluation of the
motorcycle drivers re-education program in Vietnam modeling of
motorcycle driverrsquos undesired behaviors 11th World Conference
on Transport Research 200725
Victor 1980 published data only
Victor DJ Traffic safety education for motorcyclists International
Motorcycle Safety Conference Washington DC USA Motorcycle
Safety Foundation 1980221ndash8
Waterer 1990 published data only
Waterer TH Compulsory training for motorcyclists from liberty
to compulsion The Human Element International Motorcycle
Safety Conference Orlando Florida USA Motorcycle Safety
Foundation 19904-49 to 4-57
Weller 1989 published data only
Weller RB Chandler EW Motorcycle safety and motorcycle
education past research and survey results Journal of Public Policy
amp Marketing 1989893ndash108
Wells 1981 published data only
Wells P Observations of some RACACU motorcycle training
schemes Transport and Road Research Laboratory Crowthorne
Berkshire UK 1981
Winn 1986 published data only
Winn G Two important trends in motorcycle safety regulations
rider education and conspicuity improvement American
Motorcyclist Association 1986147ndash52
Wisher 1988 published data only
Wisher G Reid J Motorcycle accidents in Norfolk a three-year study of motorcycle training and the Part 1 test Norfolk UK Road Safety
Division Highways Department Norfolk County Council 1988
Wood 1987 published data only
Wood HT Bowen R Evaluation of the revised learner permit scheme
July 1983 to December 1985 Victoria Australia Road Traffic
Authority 19871ndash27 [ ISBN 0 7306 0247 8]
Yamazaki 1990 published data only
Yamazaki S Nakanishi M Survey on effect of Honda rider
education The Human Element International Motorcycle Safety
Conference Orlando Florida USA 199012-167 to 12-187
References to ongoing studies
Burgess 2010 unpublished data only
Burgess C Broughton P Fylan F Stradling Interim evaluation of
the UKs National Ride program In L Dorn editor(s) Driver
behaviour and Training Vol IV Aldershot Ashgate Publishing
Limited 2010
Additional references
Anderson 1980
Anderson J Ford JL Peck RC Improved motorcyclist licensing and
testing project Volume I and II National Highway Traffic Safety
Administration US department of Transportation Washington
DC 1980
Bjornstig 1985
Bjornstig U Bylund P Lekander T Brorsson B Motorcycle fatalitiesin Sweden Acta Chirurgica Scandinavica 1985151(7)577ndash81
Clayton 1990
Clayton A Sudlow DE An evaluation of the effectiveness of a one-
day training program for learner motorcyclists The Human
Element International Motorcycle Safety Conference Orlando
Florida USA 199013-105 to 13-111
Cochrane Collaboration 2008
Higgins JP Green S Cochrane Handbook for Systematic Reviews of
Interventions 500 The Cochrane Collaboration 2008
Elliot 2003
Elliott MA Baughan CJT Broughton J Grayson G Knowles J et
alMotorcycle Safety a Scoping Study TRL [ Report No
0968ndash4107]Fleiss 2004
Fleiss JL Levin B Paik MC Statistical methods for rates and proportions 3rd Edition John Wiley amp Sons Inc 2004 [ ISBN
978ndash0ndash471ndash52629ndash2 ]
FORS 1999
FORS Federal Office of Road Safety Road risk for sober licensed
motorcyclists Vol Monograph 27 Federal Office of Road Safety
19991ndash2
Haworth 2005
Haworth N Mulvihill C Review of motorcycle licensing and training
Victoria Australia Monash University Accident Research Centre
20051ndash83 [ Report No 240]
Huwiler-Muumlntener 2002Huwiler-Muumlntener K Juumlni P Junker C Egger M Quality of
Reporting of Randomized Trials as a Measure of Methodologic
Quality JAMA 20022872801ndash4
Johnston 2008
Johnston P Brooks C Savage H Fatal and serious road crashes
involving motorcyclists Road Safety Research and Analysis Report
ATSB Monograph 20 Canberra Department of Infrastructure
Transport Regional Development and Local Government 0205
081ndash26
Juumlni 2001
Juumlni P Altman DG Egger M Systematic reviews in health care
Assessing the quality of controlled clinical trials BMJ 2001323
42ndash6
28Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3150
Kopits 2003
Kopits E Cropper M Traffic fatalities and economic growth TheWorld Bank Development Research Group Infrastructure and
Environment Policy research working paper 3035 2003
Leung 1983
Leung P Chau BTW Motorcycle accidents in Hong Kong pattern
of injury treatment causative factors and socio-economic
implications Injury 198315(1)23ndash9
Liu 2004
Liu B Ivers R Norton R Blows S Lo SK Helmets for preventing
injury in motorcycle riders Cochrane Database of Systematic Reviews 2004 Issue 2 [DOI 10100214651858CD004333pub3]
Mohan 2002
Mohan D Road safety in less-motorized environments future
concerns International Journal of Epidemiology 200231(3)527ndash32Mortimer 1982
Mortimer RG An evaluation of the motorcycle rider course
Proceedings of the 20th conference of the American Association for
Automotive Medicine Arlington Heights IL USA American
Association for Automotive Medicine 198277ndash96
Mullin 1997
Mullin BT Risk factors for motorcycle injury the role of age
gender experience training and alcohol PhD thesis Auckland
New Zealand University of Auckland 1997
Noordzij 2001
Noordzij PC Forke E Brendicke R Chinn BP Integration of needs
of moped and motorcycle riders into safety measures review and
statistical analysis in the framework of the European research
project PROMISING SWOV Institute for Road Safety Research
Leidschendam The Netherlands 2001 [ Report No Dndash2001ndash5]
Osga 1979
Osga G Ellingstad VS An evaluation of the effectiveness of the 1978 South Dakota Motorcycle Riders Course Program South Dakota
University of South Dakota Vermillion Human Factors Lab 1979
Oxford English Dictionary Online
httpoxforddictionariescom Oxford English Dictionary Online
Oxford University Press 2010
Peden 2004
Peden M Scurfield R Sleet D World Report on Road TrafficInjury Prevention Geneva World Health Organization Available
at httpwwwwhointviolence˙injury˙preventionpublications
road˙trafficworld˙reporten 2004Chapter 3
Radin Umar 1995
Radin Umar R Macakay G Hills B Preliminary analysis of
motorcycle accidents Short-term impacts of the running
headlights campaign and regulations in Malaysia J Traffic Med 199523(1)17ndash27
Tsai 1995
Tsai YJ Wang JD et alCase-control study of the effectiveness of
different types of helmets for the prevention of head injuries among
motorcycle riders in Taipei Taiwan American Journal of
Epidemiology 1995142(9)974ndash81
Vis 1995
Vis AA In-depth study of the hazards of motorcycling [De
onveiligheid van motorrijden nader bekeken] SWOV Institute for
Road Safety Research Leidschendam the Netherlands 1995 [
Report No Rndash95ndash69]
Weaver 1990
Weaver EA Motorcycle safety education in the United States
Motorcycle Safety Foundation Orlando Florida USA 1990
Winn 1987
Winn G Two important trends in motorcycle safety regulations
rider education and conspicuity improvement No more
information available Detroit MI US Society of Automotive
Engineers 1987147ndash52
Yuan 2000
Yuan W The effectiveness of the rsquoride-brightrsquo legislation for
motorcycles in Singapore Accident Analysis amp Prevention 200032
(4)559ndash63
Zambon 2006
Zambon F Hasselberg M Factors affecting the severity of injuries
among young motorcyclists a Swedish nationwide cohort study
Traffic Injury Prevention 20067(2)143ndash9lowast Indicates the major publication for the study
29Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3250
C H A R A C T E R I S T I C S O F S T U D I E S
Characteristics of included studies [ordered by study ID]
Adams 1985
Methods Non-randomised trial
Participants Initial sample size
Total of 162 Australia Post telegram delivery riders
Not clear how many in intervention and control groups
Analysis sample size
Intervention group 63Control group 35
Total 98 (61)
Interventions No formal training vs voluntary rider training course of 12 hours over 2 consecutive days (theory 4 hours practice
8 hours)
Intervention integrity
No information given
Outcomes Australia Post recorded crashes
Length of follow up 1 year
Notes Study conducted in NSW Australia
Anderson 1980
Methods Randomised trial
Participants Initial sample size
Total 41188 motorcycle licence applicants
Analysis sample size
Intervention group with skills training 13160
Intervention group without skills training 12634
Control group 15080
Total 40874 (99)
Interventions Standard licensing skill test vs the Motorcycle Operators Skill Test (MOST) with or without a training component
MOST is not to be confused with rider training because it is a licensing test The training component was for
participants who failed the skill test on the first attempt was of 3 hours duration and consisted of instructions about
the skills needed to pass the skill test The training consisted of 30 min theory and 2 frac12 hours practice on the
motorcycle
Intervention integrity
33 in MOST with skills training 31 in MOST without skills training and 43 in control group obtained their
motorcycle licence in the end
Outcomes Police recorded crashes and offences
Length of follow up 3 years
30Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3350
Anderson 1980 (Continued)
Notes Study conducted in California USA
Billheimer 1998
Methods Cohort study
Participants Initial sample size
Intervention group Unclear numbers of CMSP graduates and ERC graduates
Control group Over 16000 untrained motorcyclists
Analysis sample size
1139 matched CMSP-pairs and 1182 ERC-pairs
Interventions No formal training vs the mandatory (for those younger than 18 years) pre-licence 16-hours basic California
Motorcyclist Safety Program (CMSP) and the 8-hour Experienced Rider Course (ERC)
Intervention integrity
No information given
Outcomes Police recorded crashes and offences
Length of follow up 2 years
Notes Study conducted in California USA
Cooper 1988
Methods Cohort study
Participants Initial sample size
Intervention group 418 graduates and 57 who failed
Control group 402
Analysis sample size
Intervention group unclear
Control group unclear
Total 863 (984)
Interventions No formaltrainingvs non-mandatory pre-licence rider trainingThe content of thetrainingwas notfurther explainedIntervention integrity
Participants had all completed the rider training
Outcomes Police recorded crashes and offences
Length of follow up 7 years
Notes Study conducted in British Columbia Canada
31Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3450
Davis 1997
Methods Retrospective cohort study
Participants Initial sample size
Intervention group 9320 graduates
Control group unclear
Analysis sample size
Intervention group 9320 graduates
Control group unclear
Interventions No formal training vs the non-mandatory Connecticut Motorcycle Operator Training Program The content of
training was not further explained
Intervention integrity Participants had all completed the rider training
Outcomes Police recorded crashes severity of crashes
Length of follow up 1 year
Notes Study conducted in Connecticut USA
Hall 1988
Methods Randomised trial
Participants Initial sample sizeIntervention group A 6248
Intervention group B 6616
Intervention group C 6996
Control group 6604
Total 26464 motorcycle licence applicants
Analysis sample size
Intervention group A 6012 (96)
Intervention group B 6428 (97)
Intervention group C 6778 (97)
Control group 6502 (98)
Interventions Standard licensing skill test vs the Motorcycle Operator Skill Test (MOST II) without training (A) MOST II with
3 hours training (B) or MOST II with 20 hours training (C) The 3 hours training comprised instructions aboutthe skills needed to pass the skill test (30 min theory and 2 frac12 hours practice on the motorcycle) The 20 hours
training consisted of the Motorcycle Safety Foundationrsquos Motorcycle Rider Course (MRC) comprising of 8 hours
theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-road practice
Intervention integrity
25 and 26 obtained their motorcycle licence in the lsquoMOST 3-hour trainingrsquo and lsquoMOST 20-hour trainingrsquo
intervention arm respectively In the lsquoMOST testrsquo intervention arm 29 obtained their licence and in the control
group 35
Outcomes Police recorded crashes and offences
Length of follow up 3 years
32Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3550
Hall 1988 (Continued)
Notes Study conducted in New York USA
Haworth 1997
Methods Population based case-control study
Participants Cases Motorcycle riders or pillion riders involved in motorcycle crash (on road) Recruited from hospital and police
records Not clear whether all cases (especially fatalities) were included
Community controls Multiple matched sample of motorcycle rider or pillion who passed the crash site at the same
time of day and day of week that the crash occurred
Initial sample size
Cases 336
Controls 1195
Analysis sample size
Cases 234 (70)
Controls 561 (47)
Interventions Basic handling skills course beginner intermediate or advanced course measured through self-report not validated
Outcomes Injuries with an Injury Severity Score of 5 or greater or admission for at least four hours and fatalities from a
motorcycle crash (from hospital and police records)
Length of follow up two years or more prior to the interview
Notes Study conducted in Victoria Australia
Hurt 1981
Methods Population based case-control study
Participants Cases Motorcycle riders who had a motorcycle crash that was reported to police (traffic accident reports) Not all
cases included
Community controls Multiple matched sample of motorcycle riders who passed the crash site at the same time of
day and day of week as the crash occurred but 2 years later
Initial sample size
Cases 4500 notifications of crashes to police (only 50 received in time research team able to respond to 1126 for
unreported reasons data collection completed for 900)
Controls Number approached unreported 2310 consented and interviewed
Analysis sample size
Cases 808 (808900 data on rider training) - 18 of initial sample
Controls 670 (6702310 data on rider training) - lt 29 of initial sample
Interventions Training experience self taught friends-family school-club formal-professional others measured through self-
report validated
Outcomes Injuries and fatalities from a motorcycle crash (from hospital records and coronerrsquos office)
33Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3650
Hurt 1981 (Continued)
Notes Study conducted in Los Angeles USA
Jonah 1982
Methods Cohort study
Participants Initial sample size
Intervention group
2310 licensed MTP graduates who are active riders
Control group
2029 licensed motorcyclists who are active riders
Non-licensed and non-active riders excluded
Analysis sample size
Intervention group
811 interviews (35) 533 driver records (23)
Control group 1080 interviews (71) 905 driver records (45)
Interventions No training vs the non-mandatory Canada Safety Council Motorcycle Training Program (MTP) comprising of 4
hours theory and 16 hours practise (6 hrs on-road) over 2 weekends
Intervention integrity
Participants had all successfully completed the MTP
Outcomes Self-reported and police reported crashes and offencesLength of follow up 4 years
Notes Study conducted in Ontario Canada
Kloeden 1994
Methods Cohort study
Participants Initial sample size
Intervention group 2533 licence applicants
Control group Unclear
Sample of riders obtaining probationary licenceIntervention group 1625 (64)
Control group 5015 (unclear)
Non-active riders excluded from control group (~66)
Analysis sample size
Intervention group 631 (39)
Control group 318 (6)
Interventions No formal training vs the mandatory pre-licence Ridersafe course consisting of two levelsthe first is a training course
in the basics of motorcycle riding where a pass entitles the rider to obtain a learnerrsquos permit The second is a skill
test where a pass entitles the rider to obtain a probationary motorcycle licence The training course consists of two 4
hour training sessions on separate days involving instruction videos and actual riding on motorcycles The skill test
consists of a 4 hour preparation course of 74 minutes of lecturediscussion 38 minutes of video and 75 minutes of
34Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3750
Kloeden 1994 (Continued)
on-course riding which is the testing phase
Intervention integrity
Unclear what percentage in the Ridersafe group and the control group obtained their probationary licence halfway
and their full motorcycle licence in the end
Outcomes Police reported crashes
Length of follow up 1 year
Notes Study conducted in SA Australia
Kraus 1975
Methods Population based case-control study
Participants Cases All motorcycle riders with a confirmed medically treated motorcycle injury (including fatalities) from all
hospitals in study base and peripheral area hospitals and all deaths due to motorcycle crashes in the study base All
serious injuries (leading to hospitalisation) and fatalities likely to be included
Community controls A random sample of registered motorcycle owners in the same temporary and geographical
study base as the cases arose from
Initial sample size Cases 1273
Controls 738
Analysis sample size Cases 741 (58)
Controls 434 (59)
Interventions Motorcycle rider training (not specified) measured through self-report
Outcomes Injuries leading to hospitalisation fatalities
Length of follow up not specified
Notes Study conducted in Sacramento county USA
Leung 1987
Methods Non-randomised trial
Participants Initial sample sizeIntervention group unknown number of MRC graduates
Control group unknown number of licensed motorcyclists
Analysis sample size
Intervention group 2941
Control group 43094
Interventions No formal training vs non-mandatory Wisconsin motorcycle rider course a 22 hours off-street programme that
teaches the basic rules of safe motorcycling
Intervention integrity
50 in the MRC group and 100 in the control group obtained their motorcycle licence in the end
35Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3850
Leung 1987 (Continued)
Outcomes Police recorded crashes and offences
Length of follow up up to 3 years
Notes Study conducted in Wisconsin USA
MAIDS 2004
Methods Population based case-control study
Participants Cases Motorcycle riders who had a motorcycle crash Unclear whether all cases included
Community controls Motorcycle riders identified at petrol stations Unclear whether matched to cases
Initial sample size
Cases not reported
Controls not reported
Analysis sample size
Cases 921 (Italy 200 Netherlands 200 Spain 121 Germany 250 France 150)
Controls 923 (Italy 200 Netherlands 200 Spain123 Germany 250 France 150)
Interventions Pre-license rider training or additional training (non-compulsory)
Outcomes Crashes
Notes Study in five sampling areas located in five European countries (France Germany Netherlands Spain and Italy)
McDavid 1989
Methods Cohort study
Participants Sample size after first match
Intervention group 346 male training graduates
Control group 346 male motorcycle riders licensed in same year
Sample size after second match
Intervention group 139 (40)
Control group 139 (40)
Analysis sample size
Intervention group 139 (40)
Control group 139 (40)
Interventions No formal training vs the non-mandatory pre-licence British Columbia Safety Councilrsquos 37-hour motorcycle safety
programme
Intervention integrity
Participants had all completed and passed the training
Outcomes Police reported crashes injurious crashes
Length of follow up 1-6 years
36Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3950
McDavid 1989 (Continued)
Notes Study conducted in British Columbia Canada
Mortimer 1984
Methods Cohort study
Participants Initial sample size
Intervention group 608 MRC graduates
Control group 360 motorcycle riders
Response rates
Intervention group 360 (59)
Control group 324 (90)
Non-active riders excluded
Analysis sample size
Intervention group 213 (35)
Control group 303 (84)
Interventions No formal training vs the non-mandatory Motorcycle Safety Foundationrsquos Motorcycle Rider Course comprising of
8 hours theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-road practice
Intervention integrity
Participants had all completed and passed the MRC
Outcomes Self-reported crashes and offencesLength of follow up 1 year
Notes Study conducted in Illinois USA
Mortimer 1988
Methods Cohort study
Participants Initial sample size
Intervention group 2307 MRC graduates
Control group 556 active licensed motorcycle riders
Response ratesIntervention group 1298 (56)
Control group 500 (90)
Non-active riders excluded
Analysis sample size
Intervention group 913 (40)
Control group 500 (90)
Interventions No formal training vs the non-mandatory Motorcycle Safety Foundationrsquos Motorcycle Rider Course comprising of
8 hours theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-road practice
Intervention integrity
Participants had all completed and passed the MRC
37Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4050
Mortimer 1988 (Continued)
Outcomes Self reported crashes and offences
Length of follow up 1 year
Notes Study conducted in Illinois USA
Osga 1980
Methods Cohort study
Participants Initial sample sizeIntervention group 1163 MSF MRC participants invited 818 (74) responded
Control group 329 motorcycle riders responded to recruitment strategies
Non-active riders excluded from analysis
Analysis sample size
Intervention group 595 (73)
Control group 329 (100)
Interventions No formal training vs the non-mandatory Motorcycle Safety Foundationrsquos Motorcycle Rider Course (MRC) com-
prising of 8 hours theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-road
practice
Intervention integrity
Participants had all successfully completed the MRC
Outcomes Self-reported crashes police recorded crash reports
Length of follow up 6 months - 3 12 years
Notes Study conducted in South Dakota USA
Perrino 2002
Methods Cohort study
Participants Initial sample size
Intervention group 2144 (unclear) MRC graduates (motorcycle licence applicants)
Control group 2601 (unclear) regular licensing test graduates (motorcycle licence applicants)Analysis sample size
Intervention group 1070 (unclear)
Control group 867 (unclear)
Interventions No formal training vs the mandatory (for those younger than 18 years) pre-licence Maryland Motorcycle Rider
Course (MRC) consisting of 8 hours theory and 12 hours practice
Intervention integrity
No information given
Outcomes Self-reported and police recorded crashes and offences
Length of follow up 1 year
38Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4150
Perrino 2002 (Continued)
Notes Study conducted in Maryland USA
Raymond 1979
Methods Cohort study
Participants Initial sample size
Intervention group
258 volunteers mainly (but not all) novice riders
Control group
1104 provisional licence applicants
Analysis sample size
Intervention group 190 (74)
Control group 627 (57)
Interventions No formal training vs the non-mandatory RACACU (Royal Automobile ClubAuto-Cycle Union) training scheme
The training program consisted of either 24 one-hour sessions or 13 two-hour sessions Twelve hours of the training
focused on theory about basic operational procedures traffic laws and highway code The practical session focused
on basic operational skills riding on a training area and riding in traffic with an instructor A proficiency test was
administered at the end of the course
Intervention integrity
No information given
Outcomes Self-reported crashes (including minor non-injury non-damage and off-road crashes)
Length of follow up 1 year
Notes Study conducted in the UK
Satten 1980
Methods Cohort study
Participants Initial sample size
Intervention group interviewers kept interviewing until they had 100 riders selected from a population of 891
enrollees in the MSF MRC course (unknown how many were asked and how many refused to participate)Control group unknown number of registered motorcycle riders
Non-active riders excluded from analysis
Analysis sample size
Intervention group 71
Control group 100
Interventions No formal training vs the non-mandatory Motorcycle Safety Foundationrsquos (MSF) Motorcycle Rider Course (MRC)
comprising of 8 hours theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-
road practice
Intervention integrity
Participants had all successfully completed the MRC
39Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4250
Satten 1980 (Continued)
Outcomes Self-reported crashes injuries and offences
Length of follow up 3 years
Notes Study conducted in Illinois USA
Savolainen 2007
Methods Cohort study
Participants Initial sample size
Intervention group 4000 riders that had conducted the MSF MRC course (passed or failed)
Control group 4000 motorcycle owners or permit or endorsement holders from the Indiana Bureau of Motor
Vehicles database
Analysis sample size
Intervention group 739 (18)
Control group 588 (15)
Interventions Intervention integrity
No information given
Outcomes Probability of crash involvement
Length of follow up up to 5 years
Notes Study conducted in Indiana USA
Swaddiwudhipong 1998
Methods Randomised trial
Participants Initial sample size
Intervention group
all 16 villages of 3 sub-districts
(participants were practising motorcyclists licensed and unlicensed)
Control group
18 villages of 3 other sub-districtsAnalysis sample size
Intervention group 3 districts (100)
Control group 3 districts (100)
Interventions No rider education vs non-mandatory community based rider education consisting of theory on motorcycle injury
risk helmet use traffic laws vehicle regulations and traffic signs and a written and skill test leading to a licence
Intervention integrity
No information given
Outcomes Injuries fatalities
Length of follow up 2 years
40Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4350
Swaddiwudhipong 1998 (Continued)
Notes Study conducted in Tak Province rural Thailand
Waller 1992
Methods Cohort study
Participants Initial sample size
Intervention group 349 riders who succesfully completed the Department of Public Safety motorcycle operator
training in 1988
Control group 349 motorcyclists who never completed the DPS motorcycle operator training
Analysis sample size
Intervention group 349
Control group 349
Interventions No DPS motorcycle operator training vs non-mandatory DPS motorcycle operator training The DPS training is
the Motorcycle Safety Foundationrsquos Motorcycle Rider Course comprising eight hours of theory (lectures discussions
films and multiple choice tests) a textbook and 12 hours off-road practice
Intervention integrity
All 349 riders succesfully completed the rider training
Outcomes Police recorded crashes
Length of follow up 3 years
Notes Study conducted in
Characteristics of excluded studies [ordered by study ID]
Study Reason for exclusion
Anderson 1980a Study type review and process evaluation of a rider course
Aupetit 2008 Outcome proposal of a rider training curriculum that combines traditional training and a motorcycle riding
simulator
Awane 1999 Study type review on use of simulators in rider and driver training
Baldi 2005 Intervention not rider training but intervention integrity of rider training availabilty of rider education
implementation of graduated licensing system etc on a state-wide level
Batchler 1988 Study type description of a training course
Blanchard 2006 Study type not a study assessing the association between rider training and outcomes
Bonnett 2005 Outcome qualitative data regarding satisfaction with the training programme
41Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4450
(Continued)
Braver 2007 Outcome licensure rates
Clayton 1980 Outcome knowledge skills attitude
Council 1978 Outcome crash reduction in ANY vehicle not motorcycles exclusively
Crowther 2005 Study type qualitative study
Daniello 2009 Study type review of the effectiveness of motorcycle training and licensing
Dewanti 2007 Intervention emotional characteristics behaviour (ie not motocycle rider training)
Du Plooy 1990 Study type review of types of rider training courses in South Africa
Engel 1989 Study type uncontrolled before-after study
Goldenbeld 2002 Outcome skills
Goldenbeld 2004 Outcome knowledge and skills
Haworth 2000 Intervention balance between attitudinal andvehicle control skills acrosscurriculla of motorcycle rider training
courses
Haworth 2005 Study type review
Haworth 2006 Study type cross-sectional survey
Hill 1983 Study type review and description of training system
Imatake 1980 Study type course descriptionreview
Ingebritsen 1990 Intervention risk factors for motorcycle crashes
Jain 2009 Study type cross-sectional descriptive study of motorcycle crashes over a 5 year period
Jonah 1979 Study type cross-sectional survey Outcome scores on motorcycle operator skill test
Jonah 1981 Intervention evaluation of a motorcycle rider skill test not motorcycle rider training
Kadar 1990 Study type review of components of rider training courses
Kelsey 1986 Intervention evaluation of a motorcycle rider skill test not motorcycle rider training
Koch 1980 Study type reviewcourse description
Koch 1990 Study type description of course content
42Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4550
(Continued)
Labbett 2005 Study type essay about possible explanations for motorcycle deaths
Law 2005 Intervention targeted motorcycle safety programme not specifically rider training
Lowes 1990 Study type review and descriptions of rider training
Mahon 1976 Study type review of motorcycle rider training methods
Maxwell 1985 Outcomesrecommendations forprogramme operatorsand otherson howthe motorcycle trainingprogramme
works and how it can be improved
Mayhew 1996 Study type review
McKnight 1987 Study type evaluation of methodological and administrative aspects of a motorcycle operator skill test
McPherson 1976 Outcome knowledge and skills
Mohan 2004 Study type review of evidence of interventions to reduce road traffic injuries
Morgan 1980 Study type reviewcourse description
Motoki 1990 Study type proposal and development of rider training curriculum
Nairn 1992 Study type literature review
Newland 1980 Study type description of development of motorcycle safety course
Newland 1982 Study type descriptionreview of a rider training course
Newland 1990 Study type uncontrolled BeforeAfter study
Ormston 2003 Outcome changes in attitude and self reported behaviour
Osga 1979 Outcome knowledge attitudes and skills
Panou 2007 Intervention new simulation tool to teach driverrider
Peterson 1990 Study factor demographic details of course participants completing the course vs those not completing the
course
Prem 1984 Study type experimental
Study factor examines association between skills and performance in tests
Radin Umar 2006 Intervention motorcycle safety programme not specifically rider training
Rockwell 1990 Study type methodologically flawed study type
43Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4650
(Continued)
Rothe 1987 Study type qualitative study of rider training in general
Rowden 2007 Study type qualitative study
RSC Victoria 1993 Study type review of motorcycle rider training and education in Victoria
Russam 1979 Study type review of training programmes in Japan
Saffron 1981 Study type literature review
Schulz 1990 Study factor predictors of motorcycle crashes
Sexton 2004 Study type cross-sectional analytic analysis
Shephard 1986 Study factor reviewdescription of programme
Simpson 1990 Study type review of motorcycle rider trainingeducation
Sokolov 1990 Study factor review of motorcycle riding courses in USSR
Steffens 1988 Outcome self-reported behaviour experiences and researcher observed riding behaviour
Sudlow 2003 Study type review of motorcycle training in the UK
Tasmania 1995 Study type cross-sectional descriptiveprocess evaluation
Tetard 1985 Study type experimental
Outcome detection of indices
Thackray 1980 Outcome instructional effectiveness user acceptance administrative feasibility
Thompson 1994 Study type cross-sectional survey
Outcome attitudes and knowledge
Tobler 1990 Study type review of course curricula
Transport 1989 Intervention syllabus for rider training
Tuan 2007 Study type cross-sectional study (survey) which does not assess the effects of an intervention but examines
motocyclistsrsquo behaviour
Victor 1980 Study type review
Waterer 1990 Study type review of UK rider training issues
Weller 1989 Study type review of studies evaluating motorcycle safety and education measures
44Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4750
(Continued)
Wells 1981 Study type description of different methods of training
Winn 1986 Study type review of rsquohistoryrsquo of rider training
Wisher 1988 Study type cross-sectional survey
Wood 1987 Study type Interrupted Time Series but did not meet the criteria for an ITS as published by the Cochrane
Effective Practice and Organisation of Care Group
Yamazaki 1990 Study type cross-sectional survey
Outcome riding skills skills retention and attitude
Characteristics of ongoing studies [ordered by study ID]
Burgess 2010
Trial name or title UKs National Ride program
Methods Cohort study
Participants Motorcycle offenders
Interventions Police led diversion program
Outcomes Self reported attitudinal and behavioural measures
Starting date
Contact information
Notes Expected to be completed in early 2010
45Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4850
D A T A A N D A N A L Y S E S
This review has no analyses
A P P E N D I C E S
Appendix 1 Search strategy
We searched the following electronic databases
bull Cochrane Injuries Group Specialised Register (searched 22 Sept 2008)
bull CENTRAL (The Cochrane Library 2008 Issue 3)
bull TRANSPORT 1988 to 200706 (latest date available)
bull ATRI Australian Transport Index (to Sept 2008)
bull MEDLINE (Ovid SP) 1950 to Sept 2008
bull EMBASE (Ovid SP) 1980 to Sept 2008
bull CINAHL (EBSCO) 1982 to Sept 2008
bull WHOLIS (World Health Organization Library Information System) (to Sept 2008)
bull PsycINFO (Ovid SP) 1806 to Sept 2008
bull LILACS (BIREME) (Latin American and Caribbean Health Sciences) (1982 to Sept 2008)
bull ISI Web of Science Social Sciences Citation Index (SSCI) 1970 to Sept 2008
bull ERIC (searched 25 Sept 2008)
bull ZETOC (searched 25 Sept 2008)
bull SIGLE (searched 25 Sept 2008)
All of the databases were searched using the following text-word search strategy The strategy was adapted using either rsquorsquo or rsquo$rsquo as atruncation symbol
1 motor-cycl or motorcycl or motor-bik or motorbik or motor-scooter or motorscooter or scooter or moped or motocycl
or motocicl or moto
2 train or educat or instruct or program or course or qualif or learn or improv
3 1 and 2
H I S T O R Y
Protocol first published Issue 2 2005
Review first published Issue 10 2010
46Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4950
C O N T R I B U T I O N S O F A U T H O R S
Katina Kardamanidis was the first author of the protocol conducted the literature search extracted full studies assessed the studies
according to inclusion criteria assessed quality of the studies compared findings with the second author discussed disagreements with
the third author wrote the drafts and the final review
Rebecca Ivers was the second author of the protocol assessed the studies according to inclusion criteria assessed quality of the studies
compared findings with the first author edited the final draft and provided overall guidance for the process
Alexandra Martiniuk assessed quality of the studies compared findings with the first author discussed disagreements with the third
author and helped write the first draft and the final review
Mark Stevenson contributed to the protocol reviewed methods of studies and contributed to the final draft
Katrina Thistlethwaite conducted a second literature search extracted full studies and contributed to the final draft
D E C L A R A T I O N S O F I N T E R E S T
None known
S O U R C E S O F S U P P O R T
Internal sources
bull No sources of support supplied
External sources
bull National Health and Medical Research Council of Australia Australia
Salary support for Martiniuk Ivers and Stevenson
bull University of Sydney Sesqui Research Grant Australia
Direct research costs and salary support for Kardamanidis
D I F F E R E N C E S B E T W E E N P R O T O C O L A N D R E V I E W
There are no differences between the protocol and the review
I N D E X T E R M S
Medical Subject Headings (MeSH)
lowastMotorcycles Accident Prevention [lowastmethods] Accidents Traffic [lowastprevention amp control] Licensure Program Evaluation Wounds
and Injuries [mortality lowastprevention amp control]
47Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 5050
MeSH check words
Humans
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2150
Offences
Cooper 1988 foundno statistically significantdifference in offencerates between the intervention and control groups McDavid 1989
foundno statistically significantdifference in offence ratesbetween
the intervention and control groups
Non-mandatory rider training
Non-randomised trials
Crashes
Neither Adams 1985 nor Leung 1987 found a statistically signif-icant effect of rider training on crash rates
Offences
Leung 1987 found a statistically significant lower offence rate in
the intervention group compared to the control group for both
males and females and for all age groups except for women 35
years or older (P values not reported)
Cohort studies
Crashes
Osga 1980 in the univariate analysis reported a lower percentage
of crashesin theintervention group (that continued to ride) (28
168595) than in the control group (34 112329) In the cu-
mulative crash rate analysis however Osga1980 found statistically
significant (at the P lt 001 level) higher crash rates in the interven-
tion group than in the control group at allmileage intervals (1000
to 29999 miles) travelled After riding more than 30000 miles
(when over 50 of the trained riders had been involved in a crash
compared to approximately 40 in untrained riders) this differ-
ence was no longer statistically significant Satten 1980 found a
lower percentage of crashes in the intervention group (13 969)
compared to the control group (25 2079) but the differencewas not assessed statistically Davis 1997 found a lower crash rate
in trained riders but it was unclear how this was calculated be-
cause the size of the control group was not reported Jonah 1982
and Billheimer 1998 (for the Experienced Rider Course) found no
statistically significant difference in crashes between the interven-
tion and control groups Although the difference between those
who completed the Experienced Rider Course and no course in
Billheimer 1998 neared significance at six months after the course
(P = 007) large P values were observed for one year after (P =
030) Raymond 1979 found a statistically significant higher crash
rate for the intervention group (1861000 miles ridden) com-
pared to the control group (1181000 miles ridden) in both males
and females (P lt 0001) especially during the first 5000 miles of
riding Waller 1992 found a statistically significant higher crashrate for the intervention group (27 crashes in three years or 070
crashes10000 person days) compared to the control group (5
crashes in three years or 013 crashes10000 person days) with a
crude relative risk of 538 (X 2=1513 plt005) Mortimer 1984
and Mortimer 1988 found no statistically significant difference in
crashes between the intervention and control groups
Savolainen 2007 found that the intervention group (those who
completed the Basic Rider Course) was 44 more likely to be
involved in crashes than the control group and those who com-
pleted the course multiple times were an additional 180 more
likely to be involved in a crash This finding was based on low
numbers and no P value or confidence level was given for either
result
Injuries and deaths
Satten 1980 also found a lower percentage of injurious crashes
(33 39) in the intervention group compared to the control
group(60 1220) although again no statistical significance test
was reported
Offences
Jonah 1982 did find a significantly lower self-reportedoffence rate
in the intervention group (15) compared to the control group(34 P lt 001) (no numbers given) Satten 1980 also found
a lower self-reported offence rate in the intervention group (1
71 versus 19100) but again this was not tested for statistical
significance Mortimer 1984 Mortimer1988 and Billheimer 1998
found no statistically significant differencein offence ratesbetween
the intervention and control groups
Case-control studies
CrashesMAIDS 2004 found that a slightly higher percentage of cases
(involved in a crash) had had additional non-mandatory rider
training than controls (not involved in a crash) This difference
was not statistically significant (17 of cases vs 12 of controls
OR 146 95 CI 064 to 337)
Rider training unspecified
Case-control studies
19Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2250
Injuries and deaths
Hurt 1981 found that the odds of injurious crashes were lower inthose who had received professional training compared to those
who had not Kraus 1975 found that in the younger age group
(less than 20 years) and the older age group (over 35 years) a larger
percentage of injured riders (12 and 21 respectively) reported
prior motorcycle rider training compared with the uninjured con-
trol group (6 and 5 respectively) (P lt 001) (no numbers
given) Haworth 1997 reported several different outcomes one
beingthat having completed at leastone trainingcoursecompared
to no training courses at all had no significant effect on the odds
of crashing (afteradjustment for age blood alcohol level or licence
status (learner permitprobationaryfull)) When comparing three
different levels of training courses (beginner intermediate and ad-
vanced) with no training no significant differences were found
in the odds of crashing However after adjustment for age and
blood alcohol level completing an advanced course as compared
to a beginners course was associated with a significant decrease in
the odds of severe motorcycle crashes resulting in serious injury or
death (OR 02 95 CI 01 to 08)
D I S C U S S I O N
The objective of this review was to quantify the effectiveness of
pre- andpost-licence motorcycle rider training on the reduction of
traffic crash involvement injuries deaths and offences committedby motorcycle riders Most included studies were cohort studies
There were a few randomised and non-randomised trials and case-
control studies
All studies compared either mandatory or non-mandatory formal
training with no formal training except for Haworth 1997rsquos case-
control study that compared levels of training (beginner interme-
diate and advanced) to each other
We found studies that evaluated the effectiveness of manda-
tory pre-licence training in reducing crash and offence rates
(Anderson 1980 Hall 1988 Kloeden 1994 Billheimer 1998
(the basic California Motorcyclist Safety Program component of
the study) Perrino 2002) non-mandatory training that when
passed resulted in a licence (Cooper 1988 McDavid 1989
Swaddiwudhipong 1998) non-mandatory training that could be
taken by motorcycle riders with or without a licence (Raymond
1979 Osga 1980 Satten 1980 Jonah 1982 Mortimer 1984
Adams 1985 Leung 1987 Mortimer 1988 Waller 1992 Davis
1997 Billheimer 1998 (the Experienced Rider Course compo-
nent of the study)) and unspecified rider training on crash injury
and death rates (Kraus 1975 Hurt 1981 Haworth 1997 MAIDS
2004)
The studies provided conflicting evidence regarding motorcycle
rider training which is likely due to the poor design quality and
reporting of the studies Most of the studies included in the review
were likely to be adversely affected by significant selection biasalong with substantial detection bias(measurement and recall bias)
which limits our ability to make conclusive comments
Mandatory pre-licence rider training
There were two RCTs three cohort studies and one case-control
study that examined pre-licence ridertraining The two RCTs eval-
uated the Motorcycle Skills Operating Test (MOST) (Anderson
1980 Hall 1988) Anderson 1980 found a statistically significant
reduction in crashes in the intervention group at six months one
and twoyearsfollow up However when only participantswho ob-
tainedtheirlicence were included in theanalysis thereduction wasstatistically significant only at one year follow up The other RCT
(Hall 1988) found no statistically significant differences in crash
rates at any point of time neither in the total population group
nor in the licensed subgroup Neither of the two cohort stud-
ies assessing crash rates found a statistically significant difference
in crash rates between intervention and control groups ( Kloeden
1994 Billheimer1998) andthe case-control study (MAIDS 2004)
also did not find a statistically significant difference in the odds of
experiencing a crash between trained and untrained riders With
regards to offences neither of the RCTs found a statistically sig-
nificant difference in offence rate between the intervention and
control groups in the analysis of the total group of participants
entering the trial However Anderson 1980 found a statistically significant increase in offence rates in the intervention group but
only after two years of follow up andonly in the analysis of partici-
pants who obtained their licence The two cohort studies assessing
offence rates found no statistically significant difference between
the groups (Billheimer 1998 Perrino 2002)
None of the five studies (RCTs or cohort) found a reduction in
crashes or offences following mandatory pre-licence rider training
except for Anderson 1980 who observed a statistically significant
reduction in crashes in the intervention group at one year in both
analyses It is not clear why both RCTs that evaluated the same
MOST and training intervention observed different results Both
studies enrolled large samples and observed a similar uptake of
licensure following the intervention It is possible that differencesmay have arisen due to different study samples (California ver-
sus New York) or that the intervention delivery differed between
studies
In both trials of the Motorcycle Skills Operating Test including
rider training only a third of motorcycle licence applicants ob-
tained their motorcycle licence It is possible that a licensing pro-
cess including a skills test may reduce crashes through decreased
exposure
Non-mandatory pre-licence training
20Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2350
Three studies assessed the effect of non-mandatory pre-licence
training The Cooper 1988 cohort study found no difference incrash or offence rates between the trained and untrained groups
We cannot however be confident about this result as the study
sufferedfrom selectionbiaswhichcouldhave biasedthe measure of
effect toward the null The RCT conducted by Swaddiwudhipong
1998 examined injury and death rates and found a statistically
significant lower injury rate in the intervention villages after one
year of follow up but no longer at two year follow up There
was no statistically significant difference in death rates between
intervention and control villages Our confidence in the results of
this study are hindered by the inadequacy of the randomisation
process that led to significant differences between the intervention
and control districts on two important variables The intervention
districts were more likely than the control districts to have a validlicence andto wear a helmetSincethese differences were nottaken
into account in later analyses they could explain the differences
observed in injury rates between districts Furthermore with only
six districts randomised the sample size was probably too small
to detect a difference between intervention and control groups
had there been one The McDavid 1989 cohort study found a
lower percentage of injuries in the intervention group although
a statistical test was not conducted McDavid 1989 drew their
intervention and control groups from different populations but
did adjust for confounders although they only analysed 40 of
the initial sample (139 trained and 139 untrained) Because of
these substantial weaknesses we are unable to draw conclusions
from this studyAn overall effect of pre-licence training on the reductionof crashes
injuries deaths and offences was not found although one should
be aware that the research of this issue to date has had significant
flaws However a mandatory skill test (such as the MOST) in
combination with and even without pre-licence training as part
of a licensing process seems to reduce the number of individuals
completing the licensing procedure and therefore indirectly re-
duces crash rates through a reduction in exposure to motorcycle
riding Making licensure a more difficult process may therefore be
seen as advantageous if it delays licensure and therefore reduces
exposure It is possible that the group who chose not to continue
riding a motorcycle would not have been the group to experience a
crash or commit an offence if they had continued to ride anyhowfor example because they are more safety conscious and this was
the reason for their giving up riding in the first place If this is the
case deterring individuals from completing their licence and rid-
ing may not necessarily decrease overall crashes deaths or offence
rates In addition other potentially adverse effects of complicat-
ing a licensing procedure need to be considered such as increases
in unlicensed riding In Thailand non-mandatory rider training
offered as a community intervention may have decreased injury
rates but this needs to be confirmed through further research
Non-mandatory training
The majority of the studies assessed non-mandatory training
(training independent of any licensing procedure) and the results
were varied The two non-randomised trials assessing crash rates found no statistically significant difference in crash rates between
the intervention and control groups (Adams 1985 Leung 1987)
Of the nine cohort studies Satten 1980 found a lower percentage
of crashesin the intervention group although it is unclear whether
this was a statistically significant result Davis 1997 also found a
lower percentage of crashesin theinterventiongroup although it is
unclear with which control group the intervention group wascom-
pared Raymond 1979 Osga 1980 Waller 1992 and Savolainen
2007 found a statistically significant higher crash rate in the in-
tervention group and the remaining cohort studies (Jonah 1982
Mortimer 1984 Mortimer 1988 Billheimer 1998) found no sta-
tistically significant difference in crash rates between intervention
and control groups No conclusions can be drawn from these stud-
ies with confidence The Satten 1980 and Davis 1997 studies did
not adjust for any confounding factors even though important
differences existed between the intervention and control groups
Although Raymond 1979 and Waller 1992 adjusted for a num-
ber of confounders at baseline (Waller 1992) and in the analy-
sis (Raymond 1979 Waller 1992) the intervention and control
groups were drawn from such different populations that it is pos-
sible that unmeasured confounders were responsible for the sta-
tistically significant increased risk of crashes observed in the inter-
vention group The observed difference could also be attributed totheobserved differential loss to followup (Raymond 1979) which
could bias the measure of effect towards or away from the null
The Osga 1980 study found a protective effect of rider training
in the univariate analysis but statistically significant higher crash
rates in the intervention group when accounting for riding expo-
sure Sex (a significantly higher proportion of females in the in-
tervention group) and other potentially confounding factors that
could explain thedifferencebetween the groups were howeverun-
accounted for Finally Savolainen 2007 did control for potential
confounding variables yet it is not clear whether variables were
controlled for simultaneously or one at a time One case-control
study (MAIDS 2004) found no statistically significant reduction
in the odds of experiencing a crash between trained and untrainedriders but the quality of this study was poor mainly because the
analysis did not account for any confounding factors
One cohort study (Satten 1980) assessed the effect of non-manda-
tory rider training on injurious crashes and found a lower percent-
age of injuries in the intervention group although a statistical test
was not conducted As mentioned above the Satten 1980 studyrsquos
validity is reduced due to its lack of adjustment for confounding
factors
The only non-randomised trial that assessed offence rates (Leung
1987) found a statistically significant lower offence rate in the in-
tervention group and of the five cohort studies that assessed of-
fence rates (Satten 1980 Jonah 1982 Mortimer 1984 Mortimer
21Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2450
1988) one study (Jonah 1982) also found a statistically significant
lower offence rate in the intervention group However we have lit-tle confidence in the results from Leung 1987 since potential con-
founding factors were not appropriately addressed Jonah 1982
did conduct a multivariate analysis but was hindered by an initial
low response rate The other four cohort studies did control for
several potential confounders and found no difference in offence
rates following non-mandatory post-licence rider training
The non-randomised trials and cohort studies evaluating non-
mandatory rider training were at high risk of substantial selection
bias due to the non-random allocation of participants as well
as low participation rates and attrition bias and could therefore
not provide conclusive evidence with regard to the effect of the
training It should be noted that despite the biases and other issues
with this group of studies most didnot detectmajor effects relatedto rider training in either direction This may suggest that any
potential impact of rider training will be modest
Studies examining unspecified modes of ridertraining
Three case-control studies assessed whether injured or deceased
motorcyclists were more or less likely to have had rider training
compared to control motorcyclists All had conflicting results
Hurt1981 foundthe odds of havingan injuriouscrash waslower in
those who had received rider training whereas Kraus 1975 found
that the odds of having an injurious crash were higher in youngerand older trained riders Haworth 1997 who compared advanced
versus beginner training found no effect on motorcycle injuries
or deaths although they did find that those who had completed
advanced training had reduced chances of crashing compared to
those who had done a beginnersrsquo training course Similar issues
with uncontrolled confounding and biases of selection as described
above arose with the three case-control studies Hurt 1981 suffers
from inadequate control of confounding via poor matching and
both the Kraus 1975 and the Haworth 1997 findings were only
controlled for limited factors by means of stratified analysis It is
likely that these limitations are the reason for the contradictory
directions of effect observed for the three case-control studies
Methodological weaknesses of the studies
In summary most studies investigating the effectiveness of mo-
torcycle rider training on reducing crashes injuries and death suf-
fered from serious methodological weaknesses
The majority of studies were non-randomised and were there-
fore limited mainly by selection bias Although the approaches of
matching stratification or regression analyses can minimise bias
(Fleiss 2004) they were poorly employed in the reviewed studies
The inadequate adjustment for confounders in particular adds
to our lack of confidence in the studiesrsquo results since it has been
demonstrated that individuals who choose to take motorcycle rider
training differ from those not taking rider training For instance
trained riders often differ from controls in age sex and experi-ence usually ride less and may have a more positive attitude to
safety as indicated by the reported more frequent wearing of seat
belts (Mortimer 1982)
The two large randomised trials reported that a large proportion
of their study group were never exposed to the risk of offences or
crashes because they never attained licensure This made it impos-
sible to assess whether a reduction in crashes in the intervention
group was a result of training or due to delayed or never attained
licensure
In some studies the sample sizes were too small to detect any sta-
tistically significant differences Follow-up time after training dif-
fered between studies and was an impediment to observing any
real effects of training if they existed As riding experience ac-cumulates over time any differences in outcomes between those
trained and untrained would diminish in studies with lengthy fol-
low-up times This dilution of any possible effect of rider training
may have also occurred in studies that mixed both novice and ex-
perienced riders in the study groups since it is hypothesised that
novice riders serve to benefit most from rider training This would
increase the difficulty in finding any effect of rider training par-
ticularly when sample sizes are small
Detection bias may also account for some of the differences ob-
served between trained and untrained riders Detection bias would
have occurred if the self-report hospital driver or police records
used to record the outcomes of crash injury death or offences
differed between the trained and untrained riders Research hasdemonstrated that not all crashes are reported to police for ex-
ample small crashes or off-road crashes (Kraus 1975) even serious
injury crashes have been found to be underreported (Anderson
1980) However this would only introduce detection bias if the
reporting of outcome differs for trained and untrained riders This
does appear possible in that responsible riders may be more likely
to enroll in a rider training course and also may be more likely
to report crashes to police than irresponsible riders meaning that
the number of crashes from riskier riders could be underrepre-
sented on records An analysis to investigate the effectiveness of
rider training may then show that rider training is not effective in
reducing crashes
Rather than using reported outcomes some studies used self-re-ported survey data although this technique suffers from non-
response biases In fact two studies reported differences in de-
mographic characteristics in response rates to questionnaires
(Anderson 1980 Hall 1988) Self-reported data also introduce
the risk of recall bias survival bias and social desirability response
bias Recall bias may have been equally distributed among trained
and untrained riders howeverit is hypothesised that trained riders
may have felt it was lsquoexpectedrsquo to report good driving histories
hence leading the rider training to appear effective
22Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2550
Limitations of this review
Limitations of this review were that we found very few recentstudies (the two most recent ones published in 2002 and 2007)
and that we found very few studies conducted in low-and middle
income countries even though our search strategy was set up to
identify such studies Anotherlimitation wasthat forthe reviewed
studies we could not assess all outcomes of rider training that are
potentially useful Some of the included studies assessed outcomes
such as use of safety clothing including helmets Such outcomes
are valuable since the use of motorcycle helmets has been demon-
strated to be effective in preventing head injuries (Liu 2004) De-
spite this the use of safety clothing was not included as an out-
come in this review because the studies measuring this outcome
were limited by poor data quality due to very low response rates
Another outcome of interest in assessing the effectiveness of rider
training is lsquotype of injuryrsquo As Haworth 2005 state in their review
of motorcycle licensing and training rider training could be con-
sidered effective if injuries become less severe because motorcycle
riders have learned how to avoid more serious injuries through
rider training The studies included in this review did not specify
type of injury and so this outcome could not be assessed Last
changes in knowledge and attitudes following rider training are
other potentially useful outcomes We have excluded such studies
(Osga 1979 Clayton 1990 Goldenbeld 2004) as these outcomes
are proxies rather than direct measures of the effectiveness of mo-
torcycle rider training We opted to report on the outcomes of
crashes injury death and offences alone since there is little ev-
idence about how changes in knowledge and attitudes relate to
these outcomes
A U T H O R S rsquo C O N C L U S I O N S
Implications for practice
Due to the poor quality of studies included in this review we were
unable to draw any conclusions about the effectiveness of rider
training on crash rates injurious or fatal or offence rates The
findings do suggest that mandatory pre-licence training may be
an impediment to completing a motorcycle licensing process thuspossibly indirectly reducing crash injury death and offence rates
through a reduction in exposure to riding a motorcycle On the
basis of the existing evidence it is not clear if and what type of
training reduces the risk of crashes injuries deaths or offences in
motorcyclists and the selection of the best rider training practice
can therefore not be recommended The suggestion that manda-
tory pre-licence training indirectly reduces exposure to riding the
motorcycle because trainees are impeded from completing the li-
censing process does not help those motorcyclists who persist in
obtaining their licence and riding the motorcycle Some type of
rider training is likely necessary to teach motorcyclists to ride a
motorcycle safely Good policy is based on sound evidence and
evaluation but such evidence for motorcycle rider training is seri-
ously lacking Policy makers are therefore encouraged to include arigorous evaluation component to any new or existing rider train-
ing programme Such an evaluation should ideally be randomized
and assess whether the training programme is effective and if so
which component(s) of the programme work best
Implications for research
Given theresults of this review which found littlereliable evidence
about the effectiveness of motorcycle rider training it is necessary
to build up an evidence base of the effectiveness of motorcycle
rider training in the reduction of crash injury death and offence
rates To this end well designed randomised controlled trials need
to be conducted that examine the impact of rider training oncrashes and offences A randomised controlled trial is the preferred
method as it would significantly reduce most biases described in
this review To minimise the impact of participants delaying or
never attaining their licence and therefore exposure as found in
the randomised trials in this review such a trial should include
riders who are already licensed and reliable measures of riding
exposure in a reasonably lengthy follow up period It is expected
that a randomised controlled trial of sufficient sample size would
observe equal balance on possible confounders by study arm how-
ever an assessment of potentially important confounders (such as
helmet wearing) by study group would enable control for these
variables in the analyses if necessary It is imperative that the effec-
tiveness of each component of the rider training (such as theorypractice knowledge skills and modification of behaviour) and the
synergy of such components are evaluated Outcomes should be
objective ideally those that are routinely collected rather than self-
reported outcomes Even though riding the motorcycle poses dif-
ferent challenges and incurs different risks lessons may be learned
from research into the effects of car driver training
Motorcycles are a common means of transport in low- and middle
income countries and usage is increasing As the burden of mo-
torcycle crash victims is mostly borne in such countries research
should be focused on the effectiveness of types of rider training on
the reduction of crashes injuries and fatalities in such countries
and be supported by higher income countries This review found
only one study conducted in a middle income country that sug-gested decreasing injury rates It would be beneficial to conduct
additional research to confirm this suggestion and to enhance our
knowledge of the role of rider training in low- and middle-income
settings
March 2006 saw the launch of the University of North Carolina
Highway Safety Research Centre partnership with the Motorcycle
Safety Foundation (MSF) and plans were revealed to conduct a
study of the potential benefits of motorcycle rider training The
study is funded by the MSF and the National Highway Traffic Sa-
fety Administration (NHTSA) Called rsquoThe Discovery Projectrsquo it
will compare data on crash rates from riders participating in a basic
23Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2650
motorcycle safety class with riders participating in a basic motor-
cycle safety class as well as a variety of continuing education classesThe collection of pilot data was due to start in March 2007 and
the data collection portion of the study is expected to take three
years (Highway Safety Research Center) It is vital that researchers
and policy makers collaborate to develop furtherhigh quality eval-
uations of motorcycle rider training programmes to help inform
development of new and effective training programmes
A C K N O W L E D G E M E N T S
Staff from The University of Sydney library for helping us locate
and retrieve full reports of many papers
R E F E R E N C E S
References to studies included in this review
Adams 1985 published data only
Adams A Collingwood V Job RFS Evaluation of the 1982 Australia
Post motorcycle rider training scheme NSW Australia Traffic
Accident Research Unit Traffic Authority NSW 1985 [ RN 85]
Anderson 1980 published data only
Anderson J Ford JL Peck RC Improved motorcyclist licensing and testing project Washington DC National Highway Traffic Safety
Administration (NHTSA) USA 19801ndash217 [
DOTndashHSndash805ndash598]
Billheimer 1998 published data only
Billheimer JW Evaluation of the California Motorcyclist Safety
Program Transportation Research Record 19981640100ndash9 [
TRB Report Number 980652]
Cooper 1988 published data only
Cooper PJ Rothe JP Motorcyclists who they are and why they do
what they do Transportation Research Record 1988116878ndash85
Davis 1997 published data only
David CF Evaluation of community traffic safety programs and
motorcycle operator training programs Connecticut Department of Transportation 199763
Hall 1988 published data only
Hall CJ Buchanan L Engle R Motorcycle rider education evaluationproject Washington DC USA National Highway Traffic Safety
Administration (NHTSA) US Department of Transportation
1988 [ DTNH 22ndash80ndashCndash05123]
Haworth 1997 published data only
Haworth N Smith R Brumen I Pronk N In Monash University
Accident Research Centre editor(s) Case-control study of motorcycle
crashes Victoria Australia Department of Transport and Regional
Development The Federal Office of Road Safety 19971ndash149 [
ISBN 0 642 25501 6 ]
Hurt 1981 published data only
Hurt H Ouellet J Thom D Motorcycle accident cause factors and
identification of countermeasures USA Department of
Transportation 1981 [ Technical Report DOT HS 806ndash862]
Jonah 1982 published data onlylowast Jonah BA Dawson N Bragg B Are formally trained motorcyclists
safer Accident Analysis amp Prevention 198214(4)247ndash55
Jonah BA Dawson NE Bragg BW Evaluation of the Canada
Safety Councilrsquos Motorcycle Training Program Proceedings of theInternational Motorcycle Safety Conference Washington DC
USA 198059ndash93
Kloeden 1994 published data only
Kloeden CN Moore VM McLean AJ Evaluation of the pre-license
training program for motorcyclists in South Australia Walkerville
South Australia Australia NHMRC Road Accident Research Unit
The University of Adelaide 19941ndash37 [ ISBN 0 7308 0607 3]
Kraus 1975 published data only
Kraus JF Riggins RS Franti CE Some epidemiological features of
motorcycle collision injuries American Journal of Epidemiology 1975102(1)74ndash98
Leung 1987 published data only
Leung KS Reding VA Evaluation of the Wisconsin motorcycle rider course Wisconsin Bureau of Policy Planning and Analysis
Wisconsin Department of Transportation 1987
MAIDS 2004 published and unpublished data
MAIDS In-depth investigations of accidents involving powered
two wheelers ACEM (Acemacemeu) 2004 Vol Final report
20179
McDavid 1989 published data only
McDavid JC Lohrmann BA Lohrmann G Does motorcycle
training reduce accidents Evidence from a longitudinal quasi-
experimental study Journal of Safety Research 198920(2)61ndash72
Mortimer 1984 published data only
Mortimer R Evaluation of the motorcycle rider course Accident
Analysis amp Prevention 198416(1)63ndash71
24Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2750
Mortimer 1988 published data only
Mortimer RG A further evaluation of the motorcycle rider courseJournal of Safety Research 198819(4)187ndash96
Osga 1980 published data only
Osga GA An investigation of the riding experiences of MSF rider
course participants Dissertation South Dakota USA The
University of South Dakota 19801ndash224
Perrino 2002 published data only
Perrino CS Ahmed A Callender A Rozier E Cantwell A Stewart
O The role of Marylandrsquos motorcycle rider course in promoting saver
behaviors and attitudes Baltimore USA National Transportation
Center Morgan State University 20021ndash52
Raymond 1979 published data only
Raymond S Tatum S An evaluation of the effectiveness of the RAC
ACU motorcycle training scheme - final report Road Safety ResearchUnit Dept of Civil Engineering University of Salford London
19791ndash131
Satten 1980 published data only
Satten RS Analysis and evaluation of the motorcycle rider courses
in thirteen Illinois counties International Motorcycle Safety
Conference Washington DC 1980145ndash93
Savolainen 2007 published and unpublished data
Savolainen Peter Mannering Fred Additional evidence on the
effectiveness of motorcycle training and motorcyclistsrsquo risk-taking
behaviour Transportation Research Board 2007 Vol March 20
19
Swaddiwudhipong 1998 published data only
Swaddiwudhipong W Boonmak C Nguntra P Mahasakpan P
Effect of motorcycle rider education on changes in risk behavioursand motorcycle-related injuries in rural Thailand Tropical Medicine
and International Health 19983(10)767ndash70
Waller 1992 published data only
Waller FL An analysis of the impact of the Texas Department of
Public Safety Motorcycle Operator Training Program upon
motocycle related traffic accidents Texas State University Public
Administration Program Applied Research Projects 199277
References to studies excluded from this review
Anderson 1980a published data only
Anderson LC Better biking a motorcycle rider improvement
program International Motorcycle Safety Conference Washington
DC USA Motorcycle Safety Foundation 1980231ndash53Aupetit 2008 published and unpublished data
Aupetit S Riff J Buttelli O Espie S Ergonomics research program
on motorcycle learning Contributions to the development of a
motorcycle simulator for training Transport Research Arena
Ljubljana 20089
Awane 1999 published data only
Awane T Integrating simulators in motorcycle safety education
Journal of International Association of Traffic and Safety Sciences 199923(1)26ndash35
Baldi 2005 published data only
Baldi S Baer JD Cook AL Identifying best practices states in
motorcycle rider education and licensing Journal of Safety Research200536(1)19ndash32
Batchler 1988 published data only
Batchler MW Evaluation of the pre-license motorcycle training course at the New South Wales traffic education centre Federal Office of
Road Safety (FORS) 19881ndash122 [ Report No CR 77 ]
Blanchard 2006 published data only
Blanchard HT Tabloski PA Motorcycle safety educating riders at
the teachable moment Mosby (CV) Company 200632330ndash2
Bonnett 2005 published data only
Bonnett J Mature age skills training for experienced motorcycle
riders Motorcycle ridersrsquo association of the Australian Capital
Territory 2005
Braver 2007 published data only
Braver ER et alPersuasion and licensure a randomized controlled
intervention trial to increase licensure rates among Maryland
motorcycle owners Traffic Injury Prevention 20078(1)39ndash46Clayton 1980 published data only
Clayton AB Sudlow DE An evaluation of the effectiveness of a one day training programme for learner motorcyclists Berkshire UK
Transport and Road Research Laboratory 19801ndash23 [ ISSN
0266ndash7045]lowast Clayton AB Sudlow DE An evaluation of the effectiveness of a one-day training programme for learner motorcyclists Berkshire UK
Road Safety Division Berkshire 19801ndash23 [ISSN 0266ndash7045]
Council 1978 published data only
Council FM Desper LP Dutt AK Driver education for motorcycle operation University of North Carolina Highway Safety Research
Center 19781ndash40
Crowther 2005 published data onlylowast Crowther G Engaging with motorcyclists UK police and theBikeSafe Road Safety Programme Published proceedings of the
Australasian Road Safety Research Policing Education Conference
Wellington New Zealand 2005 issue ISBN 0ndash473ndash10636ndash11ndash11
Daniello 2009 published data only
Daniello A Gabler HC Mehta YA Effectiveness of motorcycle
training and licensing Transportation Research Board 88th Annual
Meeting 200913
Dewanti 2007 published data only
Dewanti M Challenging efforts in promoting young motorcyclist
safety in Indonesia Road Safety on Four Continents 14th
International Conference Swedish National Road and Transport
Research Institute 2007
Du Plooy 1990 published data onlyDu Plooy DW Motorcycle training in South Africa The Human
Element International Motorcycle Safety Conference Orlando
Florida USA Motorcycle Safety Foundation 19904-1 to 4-9
Engel 1989 published data only
Engel U Krogsgaringrd Thomsen L [Faeligrdselsundervisning og
propaganda ndash en faeligrdselssikkerhedsfremmende foranstaltning]
Dansk Vejtidsskrift 198912295ndash7
Goldenbeld 2002 published data only
Goldenbeld C Houwing S de Craen S The development of
driving skills of young moped riders [De ontwikkeling van de
rijvaardigheid van jonge bromfietsers] SWOV Institute for Road
Safety Research (Stichting Wetenschappelijk Onderzoek
Verkeersveiligheid) Leidschendam the Netherlands 200263
25Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2850
Goldenbeld 2004 published data only
Goldenbeld C Twisk D de Craen S Short and long term effects of moped training a field experiment Transportation Research Part F 20047(1)1ndash16
Haworth 2000 published data only
Haworth N Smith R Kowadlo N Motorcycle training in Victoria
evaluation of rider training curriculum in Victoria Monash
University Accident Research Centre Reports 2000 Vol 165 issue
3131
Haworth 2005 published data only
Haworth N Mulvihill C Review of motorcycle licensing and
training Monash University Accident Research Centre 2005 [
Report no240]
Haworth 2006 published data only
Haworth N Mulvihill C Rowdon P Teaching old dogs new tricksTraining and older motorcyclists Journal of the Australasian College of Road Safety 200718(4)20ndash5lowast Haworth NL Mulvihill C Rowden PJ Teaching old dogs new
tricks Training and older motorcyclists Australasian Road Safety
Research Policing and Education Conference Gold Coast
Queensland 2006
Hill 1983 published data only
Hill D Motorcycle rider training testing and licensing Motor
Cycle Council of New South Wales Sydney 1983
Imatake 1980 published data only
Imatake Y On the activities in Japan of motorcycle safe riding
promotion committee (MSRPC) - especially the education of
unskilled riders of class 1 motorbikes International Motorcycle
Safety Conference Washington DC USA Motorcycle Safety
Foundation 1980255ndash68
Ingebritsen 1990 published and unpublished data
Ingebrigtsen S [Risikofaktorer ved ferdsel med moped og
motorsykkel] TOslashI-rapport 66 Oslo Transportoslashkonomisk
institutt 1990
Jain 2009 published and unpublished data
Jain A et alTwo wheeler accidents on Indian roads a study from
Mangalore India Elsevier 200916130ndash3
Jonah 1979 published data only
Jonah BA Dawson NE Validation of the motorcycle operator skill
test Accident Analysis amp Prevention 197911(3)163ndash71
Jonah 1981 published data onlyJonah BA Dawson NE Bragg WE Predicting accident
involvement with the Motorcycle Operator Skill Test Accident Analysis amp Prevention 198113(4)307ndash418
Kadar 1990 published data only
Kadar KJ Countermeasures motorcycle rider education delivery
system The Human Element International Motorcycle Safety
Conference Motorcycle Safety Foundation 19906-27 to 6-41
Kelsey 1986 published data only
Kelsey SL Liddicoat C Ratz M Licensing novice motorcyclists a
comparison of Californiarsquos standard test and the MOST II (Motorcycle
Operator Skill Test) administered at centralized testing offices Sacramento California USA California Department of Motor
Vehicles Research and Development Office 19861ndash74
Koch 1980 published data only
Koch H Advanced course for motorcycle riders InternationalMotorcycle Safety Conference Washington DC USA Motorcycle
Safety Foundation 1980269ndash81
Koch 1990 published data only
Koch H Brendicke R How to influence risk taking behaviour
within motorcycle rider training programs The Human Element
International Motorcycle Safety Conference Orlando Florida
USA Motorcycle Safety Foundation 1990Motorcycle Safety
Foundation 1990 p 9-1 to 9-25A
Labbett 2005 published data only
Labbett S Langham M Training can make the problem worse
Road Safety Congress 20051ndash12
Law 2005 published and unpublished data
Law TH Radin Umar RS Zulkaurnain S Kulanthayan S Impactof the effect of economic crisis and the targeted motorcycle safety
programme on motorcycle-related accidents injuries and fatalities
in Malaysia International Journal of Injury Control and Safety
Promotion 200512(1)9ndash21
Lowes 1990 published data only
Lowes BF In-traffic rider training The Human Element
International Motorcycle Safety Conference Orlando Florida
USA Motorcycle Safety Foundation 19907-46 to 7-57
Mahon 1976 published data only
Mahon J Teaching accident avoidance on the motorcycle Journal of Traffic Safety Education 1976231ndash29
Maxwell 1985 published data only
Maxwell GS Evaluation of a Program for Training MotorcycleRiders Evaluation and Program Planning 19858(4)339ndash47
Mayhew 1996 published data only
Mayhew DR Simpson SH Effectiveness and role of driver educationand training in a graduated licensing system Ottowa Canada The
Traffic Injury Research Foundation 19961ndash89 [ ISBN
0ndash920071ndash12ndash0]
McKnight 1987 published data only
McKnight AJ Milburn BG An evaluation of the modified
motorcycle operator skill test exercises NHTSA Report 1987 [
Report No DOT HS 807 212 ]
McPherson 1976 published data only
McPherson K McKnight AJ The development and evaluation of a
motorcycle skill test manual and knowledge test Washington DCUSA National Highway Safety Administration 1976
Mohan 2004 published data only
Mohan D Evidence-based interventions for road traffic injuries in
South Asia Journal of the College of Physicians and SurgeonsPakistan 200414(12)746ndash7
Morgan 1980 published data only
Morgan JK Advanced motorcycle training course content and
study evaluation International Motorcycle Safety Conference
Washington DC USA Motorcycle Safety Foundation 1980
283ndash312
Motoki 1990 published data only
Motoki M Yamazaki S A study on effective motorcycle rider
education The Human Element International Motorcycle Safety
26Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2950
Conference Orlando Florida USA Motorcycle Safety
Foundation 19909-26 to 9-57A
Nairn 1992 published data onlylowast Nairn RJ Motorcycle safety research literature review 1987 to
1991 Federal Office of Road Safety Australia 1993 Vol Report
CR 117
Newland 1980 published data only
Newland R Darby J Automotive practicespre motorcycle rider
program International Motorcycle Safety Conference Washington
DC USA Motorcycle Safety Foundation 1980127ndash9
Newland 1982 published data only
Newland R Motorcycle rider training and licensing requirements
Driver training steering a course for the future Melbourne
Victoria Transport Regulation Board 1982
Newland 1990 published data only
Newland R Motorcycle training and licensing in Victoria 1980-
1990 The Human Element International Motorcycle Safety
Conference Orlando Florida USA Motorcycle Safety
Foundation 1990
Ormston 2003 published data only
Ormston R Dudleston A Pearson S Stradling Evaluation of
Bikesafe Scotland Scottish Executive Social Research Edinburgh
2003
Osga 1979 published data only
Osga GS Ellingstad VS An evaluation of the effectiveness of the 1978
South Dakota motorcycle rider course program South Dakota USA
University of South Dakota 19791ndash89
Panou 2007 published data only
Panou M Bekiaris E A new concept on the integration of driving
simulators in driving training the train-all approach Road Safety
on Four Continents 14th International Conference Swedish
National Road and Transport Research Institute 2007
Peterson 1990 published data only
Peterson A Kadar K Status report 1980-1989 Minnesota rider
education statistics The Human Element International
Motorcycle Safety Conference Orlando Florida USA 199013-
113 to 13-129
Prem 1984 published data only
Prem H Good MC Motorcycle rider skills assessment ACT
Australia Office of Road Safety Department of Transport 1984
1ndash377 [ Report No CR 34 ]
Radin Umar 2006 published data only
Radin Umar RS Motorcycle safety programmes in Malaysia how
effective are they International Journal of Injury Control and Safety Promotion 200613(2)71ndash9
Rockwell 1990 published data only
Rockwell TH Kiger S Carnot M An evaluation of the Ohio
motorcyclist enrichment program Ohio Department of Highway
Safety 1990
Rothe 1987 published data only
Rothe JP A motorcycle riding course Motorcyclists Image and reality Canada Insurance Corporation of British Columbia 1987
59ndash83
Rowden 2007 published data only
Rowden P Watson B Haworth N What can riders tell us aboutmotorcycle training A view from the other side of the fence
Australasian Road Safety Research Policing and Education
Conference Melbourne Australia Melbourne 2007
RSC Victoria 1993 published data only
Road Safety Committee Inquiry into Motorcycle Safety in Victoria
Parliament of Victoria 1993
Russam 1979 published data only
Russam K Motorcycle training and licensing in Japan UK Transport
and Road Research Laboratory (TRRL) 1979 [ Report 916 ]
Saffron 1981 published data only
Saffron DG Driver and rider education training and licensing a
brief review Traffic Accident Research Unit Traffic Authority of
New South Wales NSW Australia Traffic Accident Research UnitTraffic Authority of New South Wales 1981
Schulz 1990 published data only
Schulz U Motorcycle accidents among learner riders Motorcycle
licensing Evaluation of the effectiveness of a Graduated Licensing Scheme Forshungshefte des Institut fur Zweiradsicherheit 1990
Sexton 2004 published data only (unpublished sought but not used)
Sexton B Baughan C Elliott M Maycock G The accident risk of
motorcyclists 2004TRL Report TRL 607(ISSN 09684107)58
Shephard 1986 published data only
Shepherd R Do Motorcycle Safety Foundation programs work
Driver 198619(9)16ndash8
Simpson 1990 published data onlySimpson HM Mayhew DR The promotion of motorcycle safety
training education and awareness Health Education Research
Theory and Practice 19905(2)257ndash64
Sokolov 1990 published data only
Sokolov O Safe motorcycle riding some aspects of motorcycle
riding in the USSR The Human Element International
Motorcycle Safety Conference Motorcycle Safety Foundation
19904ndash37 - 4-48
Steffens 1988 published data only
Steffens U Gawatz R Willmes G Effectiveness of mofa courses
[Wirksamkeit von Mofakursen] Unfall- und sicherheitsforschung
strasenverkehr 198867168
Sudlow 2003 published data only
Sudlow D Scoping study on motorcycle training Road Safety
Research Report London UK The British Institute of Traffic
Education Research Department for Transport 2003 Vol June
[ Report No 36]
Tasmania 1995 published data only
Mannering FL Grodsky LL Motorcycle rider training scheme
Traffic Education Research Centre Armidale Australia Driver
Training Australia Limited 1995 Vol April1ndash56
Tetard 1985 published data only
Tetard C Motorcycle ridersrsquo training to detect indices in an
intersection situation [Lrsquoentrainement a la prise drsquoinformation dans
lrsquoapprentissage de la conduite motocycliste] Le Travail Humain198548(2)147ndash59
27Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3050
Thackray 1980 published data only
Thackray RM Prescott JC Field test of the motorcycle ridercourse International Motorcycle Safety Conference Washington
DC USA Motorcycle Safety Foundation 1980195ndash7
Thompson 1994 published data only
Thompson M Evaluation of compulsory basic training for
motorcyclists Transport Research Laboratory Crowthorne
Berkshire UK 1994 [ Report 63]
Tobler 1990 published data only
Tobler U Individual driving instruction for motorcycle driving
learners the Swiss experiment of motorcycle driving school The
Human Element International Motorcycle Safety Conference
Orlando Florida USA Motorcycle Safety Foundation 1990
Transport 1989 published data only
Transport Communications Road Crash Statistics Australia Bureauof Transport and Communications Economics 198914ndash20
Tuan 2007 published data only
Tuan VA Shimizu T Towards development and evaluation of the
motorcycle drivers re-education program in Vietnam modeling of
motorcycle driverrsquos undesired behaviors 11th World Conference
on Transport Research 200725
Victor 1980 published data only
Victor DJ Traffic safety education for motorcyclists International
Motorcycle Safety Conference Washington DC USA Motorcycle
Safety Foundation 1980221ndash8
Waterer 1990 published data only
Waterer TH Compulsory training for motorcyclists from liberty
to compulsion The Human Element International Motorcycle
Safety Conference Orlando Florida USA Motorcycle Safety
Foundation 19904-49 to 4-57
Weller 1989 published data only
Weller RB Chandler EW Motorcycle safety and motorcycle
education past research and survey results Journal of Public Policy
amp Marketing 1989893ndash108
Wells 1981 published data only
Wells P Observations of some RACACU motorcycle training
schemes Transport and Road Research Laboratory Crowthorne
Berkshire UK 1981
Winn 1986 published data only
Winn G Two important trends in motorcycle safety regulations
rider education and conspicuity improvement American
Motorcyclist Association 1986147ndash52
Wisher 1988 published data only
Wisher G Reid J Motorcycle accidents in Norfolk a three-year study of motorcycle training and the Part 1 test Norfolk UK Road Safety
Division Highways Department Norfolk County Council 1988
Wood 1987 published data only
Wood HT Bowen R Evaluation of the revised learner permit scheme
July 1983 to December 1985 Victoria Australia Road Traffic
Authority 19871ndash27 [ ISBN 0 7306 0247 8]
Yamazaki 1990 published data only
Yamazaki S Nakanishi M Survey on effect of Honda rider
education The Human Element International Motorcycle Safety
Conference Orlando Florida USA 199012-167 to 12-187
References to ongoing studies
Burgess 2010 unpublished data only
Burgess C Broughton P Fylan F Stradling Interim evaluation of
the UKs National Ride program In L Dorn editor(s) Driver
behaviour and Training Vol IV Aldershot Ashgate Publishing
Limited 2010
Additional references
Anderson 1980
Anderson J Ford JL Peck RC Improved motorcyclist licensing and
testing project Volume I and II National Highway Traffic Safety
Administration US department of Transportation Washington
DC 1980
Bjornstig 1985
Bjornstig U Bylund P Lekander T Brorsson B Motorcycle fatalitiesin Sweden Acta Chirurgica Scandinavica 1985151(7)577ndash81
Clayton 1990
Clayton A Sudlow DE An evaluation of the effectiveness of a one-
day training program for learner motorcyclists The Human
Element International Motorcycle Safety Conference Orlando
Florida USA 199013-105 to 13-111
Cochrane Collaboration 2008
Higgins JP Green S Cochrane Handbook for Systematic Reviews of
Interventions 500 The Cochrane Collaboration 2008
Elliot 2003
Elliott MA Baughan CJT Broughton J Grayson G Knowles J et
alMotorcycle Safety a Scoping Study TRL [ Report No
0968ndash4107]Fleiss 2004
Fleiss JL Levin B Paik MC Statistical methods for rates and proportions 3rd Edition John Wiley amp Sons Inc 2004 [ ISBN
978ndash0ndash471ndash52629ndash2 ]
FORS 1999
FORS Federal Office of Road Safety Road risk for sober licensed
motorcyclists Vol Monograph 27 Federal Office of Road Safety
19991ndash2
Haworth 2005
Haworth N Mulvihill C Review of motorcycle licensing and training
Victoria Australia Monash University Accident Research Centre
20051ndash83 [ Report No 240]
Huwiler-Muumlntener 2002Huwiler-Muumlntener K Juumlni P Junker C Egger M Quality of
Reporting of Randomized Trials as a Measure of Methodologic
Quality JAMA 20022872801ndash4
Johnston 2008
Johnston P Brooks C Savage H Fatal and serious road crashes
involving motorcyclists Road Safety Research and Analysis Report
ATSB Monograph 20 Canberra Department of Infrastructure
Transport Regional Development and Local Government 0205
081ndash26
Juumlni 2001
Juumlni P Altman DG Egger M Systematic reviews in health care
Assessing the quality of controlled clinical trials BMJ 2001323
42ndash6
28Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3150
Kopits 2003
Kopits E Cropper M Traffic fatalities and economic growth TheWorld Bank Development Research Group Infrastructure and
Environment Policy research working paper 3035 2003
Leung 1983
Leung P Chau BTW Motorcycle accidents in Hong Kong pattern
of injury treatment causative factors and socio-economic
implications Injury 198315(1)23ndash9
Liu 2004
Liu B Ivers R Norton R Blows S Lo SK Helmets for preventing
injury in motorcycle riders Cochrane Database of Systematic Reviews 2004 Issue 2 [DOI 10100214651858CD004333pub3]
Mohan 2002
Mohan D Road safety in less-motorized environments future
concerns International Journal of Epidemiology 200231(3)527ndash32Mortimer 1982
Mortimer RG An evaluation of the motorcycle rider course
Proceedings of the 20th conference of the American Association for
Automotive Medicine Arlington Heights IL USA American
Association for Automotive Medicine 198277ndash96
Mullin 1997
Mullin BT Risk factors for motorcycle injury the role of age
gender experience training and alcohol PhD thesis Auckland
New Zealand University of Auckland 1997
Noordzij 2001
Noordzij PC Forke E Brendicke R Chinn BP Integration of needs
of moped and motorcycle riders into safety measures review and
statistical analysis in the framework of the European research
project PROMISING SWOV Institute for Road Safety Research
Leidschendam The Netherlands 2001 [ Report No Dndash2001ndash5]
Osga 1979
Osga G Ellingstad VS An evaluation of the effectiveness of the 1978 South Dakota Motorcycle Riders Course Program South Dakota
University of South Dakota Vermillion Human Factors Lab 1979
Oxford English Dictionary Online
httpoxforddictionariescom Oxford English Dictionary Online
Oxford University Press 2010
Peden 2004
Peden M Scurfield R Sleet D World Report on Road TrafficInjury Prevention Geneva World Health Organization Available
at httpwwwwhointviolence˙injury˙preventionpublications
road˙trafficworld˙reporten 2004Chapter 3
Radin Umar 1995
Radin Umar R Macakay G Hills B Preliminary analysis of
motorcycle accidents Short-term impacts of the running
headlights campaign and regulations in Malaysia J Traffic Med 199523(1)17ndash27
Tsai 1995
Tsai YJ Wang JD et alCase-control study of the effectiveness of
different types of helmets for the prevention of head injuries among
motorcycle riders in Taipei Taiwan American Journal of
Epidemiology 1995142(9)974ndash81
Vis 1995
Vis AA In-depth study of the hazards of motorcycling [De
onveiligheid van motorrijden nader bekeken] SWOV Institute for
Road Safety Research Leidschendam the Netherlands 1995 [
Report No Rndash95ndash69]
Weaver 1990
Weaver EA Motorcycle safety education in the United States
Motorcycle Safety Foundation Orlando Florida USA 1990
Winn 1987
Winn G Two important trends in motorcycle safety regulations
rider education and conspicuity improvement No more
information available Detroit MI US Society of Automotive
Engineers 1987147ndash52
Yuan 2000
Yuan W The effectiveness of the rsquoride-brightrsquo legislation for
motorcycles in Singapore Accident Analysis amp Prevention 200032
(4)559ndash63
Zambon 2006
Zambon F Hasselberg M Factors affecting the severity of injuries
among young motorcyclists a Swedish nationwide cohort study
Traffic Injury Prevention 20067(2)143ndash9lowast Indicates the major publication for the study
29Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3250
C H A R A C T E R I S T I C S O F S T U D I E S
Characteristics of included studies [ordered by study ID]
Adams 1985
Methods Non-randomised trial
Participants Initial sample size
Total of 162 Australia Post telegram delivery riders
Not clear how many in intervention and control groups
Analysis sample size
Intervention group 63Control group 35
Total 98 (61)
Interventions No formal training vs voluntary rider training course of 12 hours over 2 consecutive days (theory 4 hours practice
8 hours)
Intervention integrity
No information given
Outcomes Australia Post recorded crashes
Length of follow up 1 year
Notes Study conducted in NSW Australia
Anderson 1980
Methods Randomised trial
Participants Initial sample size
Total 41188 motorcycle licence applicants
Analysis sample size
Intervention group with skills training 13160
Intervention group without skills training 12634
Control group 15080
Total 40874 (99)
Interventions Standard licensing skill test vs the Motorcycle Operators Skill Test (MOST) with or without a training component
MOST is not to be confused with rider training because it is a licensing test The training component was for
participants who failed the skill test on the first attempt was of 3 hours duration and consisted of instructions about
the skills needed to pass the skill test The training consisted of 30 min theory and 2 frac12 hours practice on the
motorcycle
Intervention integrity
33 in MOST with skills training 31 in MOST without skills training and 43 in control group obtained their
motorcycle licence in the end
Outcomes Police recorded crashes and offences
Length of follow up 3 years
30Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3350
Anderson 1980 (Continued)
Notes Study conducted in California USA
Billheimer 1998
Methods Cohort study
Participants Initial sample size
Intervention group Unclear numbers of CMSP graduates and ERC graduates
Control group Over 16000 untrained motorcyclists
Analysis sample size
1139 matched CMSP-pairs and 1182 ERC-pairs
Interventions No formal training vs the mandatory (for those younger than 18 years) pre-licence 16-hours basic California
Motorcyclist Safety Program (CMSP) and the 8-hour Experienced Rider Course (ERC)
Intervention integrity
No information given
Outcomes Police recorded crashes and offences
Length of follow up 2 years
Notes Study conducted in California USA
Cooper 1988
Methods Cohort study
Participants Initial sample size
Intervention group 418 graduates and 57 who failed
Control group 402
Analysis sample size
Intervention group unclear
Control group unclear
Total 863 (984)
Interventions No formaltrainingvs non-mandatory pre-licence rider trainingThe content of thetrainingwas notfurther explainedIntervention integrity
Participants had all completed the rider training
Outcomes Police recorded crashes and offences
Length of follow up 7 years
Notes Study conducted in British Columbia Canada
31Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3450
Davis 1997
Methods Retrospective cohort study
Participants Initial sample size
Intervention group 9320 graduates
Control group unclear
Analysis sample size
Intervention group 9320 graduates
Control group unclear
Interventions No formal training vs the non-mandatory Connecticut Motorcycle Operator Training Program The content of
training was not further explained
Intervention integrity Participants had all completed the rider training
Outcomes Police recorded crashes severity of crashes
Length of follow up 1 year
Notes Study conducted in Connecticut USA
Hall 1988
Methods Randomised trial
Participants Initial sample sizeIntervention group A 6248
Intervention group B 6616
Intervention group C 6996
Control group 6604
Total 26464 motorcycle licence applicants
Analysis sample size
Intervention group A 6012 (96)
Intervention group B 6428 (97)
Intervention group C 6778 (97)
Control group 6502 (98)
Interventions Standard licensing skill test vs the Motorcycle Operator Skill Test (MOST II) without training (A) MOST II with
3 hours training (B) or MOST II with 20 hours training (C) The 3 hours training comprised instructions aboutthe skills needed to pass the skill test (30 min theory and 2 frac12 hours practice on the motorcycle) The 20 hours
training consisted of the Motorcycle Safety Foundationrsquos Motorcycle Rider Course (MRC) comprising of 8 hours
theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-road practice
Intervention integrity
25 and 26 obtained their motorcycle licence in the lsquoMOST 3-hour trainingrsquo and lsquoMOST 20-hour trainingrsquo
intervention arm respectively In the lsquoMOST testrsquo intervention arm 29 obtained their licence and in the control
group 35
Outcomes Police recorded crashes and offences
Length of follow up 3 years
32Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3550
Hall 1988 (Continued)
Notes Study conducted in New York USA
Haworth 1997
Methods Population based case-control study
Participants Cases Motorcycle riders or pillion riders involved in motorcycle crash (on road) Recruited from hospital and police
records Not clear whether all cases (especially fatalities) were included
Community controls Multiple matched sample of motorcycle rider or pillion who passed the crash site at the same
time of day and day of week that the crash occurred
Initial sample size
Cases 336
Controls 1195
Analysis sample size
Cases 234 (70)
Controls 561 (47)
Interventions Basic handling skills course beginner intermediate or advanced course measured through self-report not validated
Outcomes Injuries with an Injury Severity Score of 5 or greater or admission for at least four hours and fatalities from a
motorcycle crash (from hospital and police records)
Length of follow up two years or more prior to the interview
Notes Study conducted in Victoria Australia
Hurt 1981
Methods Population based case-control study
Participants Cases Motorcycle riders who had a motorcycle crash that was reported to police (traffic accident reports) Not all
cases included
Community controls Multiple matched sample of motorcycle riders who passed the crash site at the same time of
day and day of week as the crash occurred but 2 years later
Initial sample size
Cases 4500 notifications of crashes to police (only 50 received in time research team able to respond to 1126 for
unreported reasons data collection completed for 900)
Controls Number approached unreported 2310 consented and interviewed
Analysis sample size
Cases 808 (808900 data on rider training) - 18 of initial sample
Controls 670 (6702310 data on rider training) - lt 29 of initial sample
Interventions Training experience self taught friends-family school-club formal-professional others measured through self-
report validated
Outcomes Injuries and fatalities from a motorcycle crash (from hospital records and coronerrsquos office)
33Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3650
Hurt 1981 (Continued)
Notes Study conducted in Los Angeles USA
Jonah 1982
Methods Cohort study
Participants Initial sample size
Intervention group
2310 licensed MTP graduates who are active riders
Control group
2029 licensed motorcyclists who are active riders
Non-licensed and non-active riders excluded
Analysis sample size
Intervention group
811 interviews (35) 533 driver records (23)
Control group 1080 interviews (71) 905 driver records (45)
Interventions No training vs the non-mandatory Canada Safety Council Motorcycle Training Program (MTP) comprising of 4
hours theory and 16 hours practise (6 hrs on-road) over 2 weekends
Intervention integrity
Participants had all successfully completed the MTP
Outcomes Self-reported and police reported crashes and offencesLength of follow up 4 years
Notes Study conducted in Ontario Canada
Kloeden 1994
Methods Cohort study
Participants Initial sample size
Intervention group 2533 licence applicants
Control group Unclear
Sample of riders obtaining probationary licenceIntervention group 1625 (64)
Control group 5015 (unclear)
Non-active riders excluded from control group (~66)
Analysis sample size
Intervention group 631 (39)
Control group 318 (6)
Interventions No formal training vs the mandatory pre-licence Ridersafe course consisting of two levelsthe first is a training course
in the basics of motorcycle riding where a pass entitles the rider to obtain a learnerrsquos permit The second is a skill
test where a pass entitles the rider to obtain a probationary motorcycle licence The training course consists of two 4
hour training sessions on separate days involving instruction videos and actual riding on motorcycles The skill test
consists of a 4 hour preparation course of 74 minutes of lecturediscussion 38 minutes of video and 75 minutes of
34Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3750
Kloeden 1994 (Continued)
on-course riding which is the testing phase
Intervention integrity
Unclear what percentage in the Ridersafe group and the control group obtained their probationary licence halfway
and their full motorcycle licence in the end
Outcomes Police reported crashes
Length of follow up 1 year
Notes Study conducted in SA Australia
Kraus 1975
Methods Population based case-control study
Participants Cases All motorcycle riders with a confirmed medically treated motorcycle injury (including fatalities) from all
hospitals in study base and peripheral area hospitals and all deaths due to motorcycle crashes in the study base All
serious injuries (leading to hospitalisation) and fatalities likely to be included
Community controls A random sample of registered motorcycle owners in the same temporary and geographical
study base as the cases arose from
Initial sample size Cases 1273
Controls 738
Analysis sample size Cases 741 (58)
Controls 434 (59)
Interventions Motorcycle rider training (not specified) measured through self-report
Outcomes Injuries leading to hospitalisation fatalities
Length of follow up not specified
Notes Study conducted in Sacramento county USA
Leung 1987
Methods Non-randomised trial
Participants Initial sample sizeIntervention group unknown number of MRC graduates
Control group unknown number of licensed motorcyclists
Analysis sample size
Intervention group 2941
Control group 43094
Interventions No formal training vs non-mandatory Wisconsin motorcycle rider course a 22 hours off-street programme that
teaches the basic rules of safe motorcycling
Intervention integrity
50 in the MRC group and 100 in the control group obtained their motorcycle licence in the end
35Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3850
Leung 1987 (Continued)
Outcomes Police recorded crashes and offences
Length of follow up up to 3 years
Notes Study conducted in Wisconsin USA
MAIDS 2004
Methods Population based case-control study
Participants Cases Motorcycle riders who had a motorcycle crash Unclear whether all cases included
Community controls Motorcycle riders identified at petrol stations Unclear whether matched to cases
Initial sample size
Cases not reported
Controls not reported
Analysis sample size
Cases 921 (Italy 200 Netherlands 200 Spain 121 Germany 250 France 150)
Controls 923 (Italy 200 Netherlands 200 Spain123 Germany 250 France 150)
Interventions Pre-license rider training or additional training (non-compulsory)
Outcomes Crashes
Notes Study in five sampling areas located in five European countries (France Germany Netherlands Spain and Italy)
McDavid 1989
Methods Cohort study
Participants Sample size after first match
Intervention group 346 male training graduates
Control group 346 male motorcycle riders licensed in same year
Sample size after second match
Intervention group 139 (40)
Control group 139 (40)
Analysis sample size
Intervention group 139 (40)
Control group 139 (40)
Interventions No formal training vs the non-mandatory pre-licence British Columbia Safety Councilrsquos 37-hour motorcycle safety
programme
Intervention integrity
Participants had all completed and passed the training
Outcomes Police reported crashes injurious crashes
Length of follow up 1-6 years
36Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3950
McDavid 1989 (Continued)
Notes Study conducted in British Columbia Canada
Mortimer 1984
Methods Cohort study
Participants Initial sample size
Intervention group 608 MRC graduates
Control group 360 motorcycle riders
Response rates
Intervention group 360 (59)
Control group 324 (90)
Non-active riders excluded
Analysis sample size
Intervention group 213 (35)
Control group 303 (84)
Interventions No formal training vs the non-mandatory Motorcycle Safety Foundationrsquos Motorcycle Rider Course comprising of
8 hours theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-road practice
Intervention integrity
Participants had all completed and passed the MRC
Outcomes Self-reported crashes and offencesLength of follow up 1 year
Notes Study conducted in Illinois USA
Mortimer 1988
Methods Cohort study
Participants Initial sample size
Intervention group 2307 MRC graduates
Control group 556 active licensed motorcycle riders
Response ratesIntervention group 1298 (56)
Control group 500 (90)
Non-active riders excluded
Analysis sample size
Intervention group 913 (40)
Control group 500 (90)
Interventions No formal training vs the non-mandatory Motorcycle Safety Foundationrsquos Motorcycle Rider Course comprising of
8 hours theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-road practice
Intervention integrity
Participants had all completed and passed the MRC
37Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4050
Mortimer 1988 (Continued)
Outcomes Self reported crashes and offences
Length of follow up 1 year
Notes Study conducted in Illinois USA
Osga 1980
Methods Cohort study
Participants Initial sample sizeIntervention group 1163 MSF MRC participants invited 818 (74) responded
Control group 329 motorcycle riders responded to recruitment strategies
Non-active riders excluded from analysis
Analysis sample size
Intervention group 595 (73)
Control group 329 (100)
Interventions No formal training vs the non-mandatory Motorcycle Safety Foundationrsquos Motorcycle Rider Course (MRC) com-
prising of 8 hours theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-road
practice
Intervention integrity
Participants had all successfully completed the MRC
Outcomes Self-reported crashes police recorded crash reports
Length of follow up 6 months - 3 12 years
Notes Study conducted in South Dakota USA
Perrino 2002
Methods Cohort study
Participants Initial sample size
Intervention group 2144 (unclear) MRC graduates (motorcycle licence applicants)
Control group 2601 (unclear) regular licensing test graduates (motorcycle licence applicants)Analysis sample size
Intervention group 1070 (unclear)
Control group 867 (unclear)
Interventions No formal training vs the mandatory (for those younger than 18 years) pre-licence Maryland Motorcycle Rider
Course (MRC) consisting of 8 hours theory and 12 hours practice
Intervention integrity
No information given
Outcomes Self-reported and police recorded crashes and offences
Length of follow up 1 year
38Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4150
Perrino 2002 (Continued)
Notes Study conducted in Maryland USA
Raymond 1979
Methods Cohort study
Participants Initial sample size
Intervention group
258 volunteers mainly (but not all) novice riders
Control group
1104 provisional licence applicants
Analysis sample size
Intervention group 190 (74)
Control group 627 (57)
Interventions No formal training vs the non-mandatory RACACU (Royal Automobile ClubAuto-Cycle Union) training scheme
The training program consisted of either 24 one-hour sessions or 13 two-hour sessions Twelve hours of the training
focused on theory about basic operational procedures traffic laws and highway code The practical session focused
on basic operational skills riding on a training area and riding in traffic with an instructor A proficiency test was
administered at the end of the course
Intervention integrity
No information given
Outcomes Self-reported crashes (including minor non-injury non-damage and off-road crashes)
Length of follow up 1 year
Notes Study conducted in the UK
Satten 1980
Methods Cohort study
Participants Initial sample size
Intervention group interviewers kept interviewing until they had 100 riders selected from a population of 891
enrollees in the MSF MRC course (unknown how many were asked and how many refused to participate)Control group unknown number of registered motorcycle riders
Non-active riders excluded from analysis
Analysis sample size
Intervention group 71
Control group 100
Interventions No formal training vs the non-mandatory Motorcycle Safety Foundationrsquos (MSF) Motorcycle Rider Course (MRC)
comprising of 8 hours theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-
road practice
Intervention integrity
Participants had all successfully completed the MRC
39Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4250
Satten 1980 (Continued)
Outcomes Self-reported crashes injuries and offences
Length of follow up 3 years
Notes Study conducted in Illinois USA
Savolainen 2007
Methods Cohort study
Participants Initial sample size
Intervention group 4000 riders that had conducted the MSF MRC course (passed or failed)
Control group 4000 motorcycle owners or permit or endorsement holders from the Indiana Bureau of Motor
Vehicles database
Analysis sample size
Intervention group 739 (18)
Control group 588 (15)
Interventions Intervention integrity
No information given
Outcomes Probability of crash involvement
Length of follow up up to 5 years
Notes Study conducted in Indiana USA
Swaddiwudhipong 1998
Methods Randomised trial
Participants Initial sample size
Intervention group
all 16 villages of 3 sub-districts
(participants were practising motorcyclists licensed and unlicensed)
Control group
18 villages of 3 other sub-districtsAnalysis sample size
Intervention group 3 districts (100)
Control group 3 districts (100)
Interventions No rider education vs non-mandatory community based rider education consisting of theory on motorcycle injury
risk helmet use traffic laws vehicle regulations and traffic signs and a written and skill test leading to a licence
Intervention integrity
No information given
Outcomes Injuries fatalities
Length of follow up 2 years
40Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4350
Swaddiwudhipong 1998 (Continued)
Notes Study conducted in Tak Province rural Thailand
Waller 1992
Methods Cohort study
Participants Initial sample size
Intervention group 349 riders who succesfully completed the Department of Public Safety motorcycle operator
training in 1988
Control group 349 motorcyclists who never completed the DPS motorcycle operator training
Analysis sample size
Intervention group 349
Control group 349
Interventions No DPS motorcycle operator training vs non-mandatory DPS motorcycle operator training The DPS training is
the Motorcycle Safety Foundationrsquos Motorcycle Rider Course comprising eight hours of theory (lectures discussions
films and multiple choice tests) a textbook and 12 hours off-road practice
Intervention integrity
All 349 riders succesfully completed the rider training
Outcomes Police recorded crashes
Length of follow up 3 years
Notes Study conducted in
Characteristics of excluded studies [ordered by study ID]
Study Reason for exclusion
Anderson 1980a Study type review and process evaluation of a rider course
Aupetit 2008 Outcome proposal of a rider training curriculum that combines traditional training and a motorcycle riding
simulator
Awane 1999 Study type review on use of simulators in rider and driver training
Baldi 2005 Intervention not rider training but intervention integrity of rider training availabilty of rider education
implementation of graduated licensing system etc on a state-wide level
Batchler 1988 Study type description of a training course
Blanchard 2006 Study type not a study assessing the association between rider training and outcomes
Bonnett 2005 Outcome qualitative data regarding satisfaction with the training programme
41Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4450
(Continued)
Braver 2007 Outcome licensure rates
Clayton 1980 Outcome knowledge skills attitude
Council 1978 Outcome crash reduction in ANY vehicle not motorcycles exclusively
Crowther 2005 Study type qualitative study
Daniello 2009 Study type review of the effectiveness of motorcycle training and licensing
Dewanti 2007 Intervention emotional characteristics behaviour (ie not motocycle rider training)
Du Plooy 1990 Study type review of types of rider training courses in South Africa
Engel 1989 Study type uncontrolled before-after study
Goldenbeld 2002 Outcome skills
Goldenbeld 2004 Outcome knowledge and skills
Haworth 2000 Intervention balance between attitudinal andvehicle control skills acrosscurriculla of motorcycle rider training
courses
Haworth 2005 Study type review
Haworth 2006 Study type cross-sectional survey
Hill 1983 Study type review and description of training system
Imatake 1980 Study type course descriptionreview
Ingebritsen 1990 Intervention risk factors for motorcycle crashes
Jain 2009 Study type cross-sectional descriptive study of motorcycle crashes over a 5 year period
Jonah 1979 Study type cross-sectional survey Outcome scores on motorcycle operator skill test
Jonah 1981 Intervention evaluation of a motorcycle rider skill test not motorcycle rider training
Kadar 1990 Study type review of components of rider training courses
Kelsey 1986 Intervention evaluation of a motorcycle rider skill test not motorcycle rider training
Koch 1980 Study type reviewcourse description
Koch 1990 Study type description of course content
42Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4550
(Continued)
Labbett 2005 Study type essay about possible explanations for motorcycle deaths
Law 2005 Intervention targeted motorcycle safety programme not specifically rider training
Lowes 1990 Study type review and descriptions of rider training
Mahon 1976 Study type review of motorcycle rider training methods
Maxwell 1985 Outcomesrecommendations forprogramme operatorsand otherson howthe motorcycle trainingprogramme
works and how it can be improved
Mayhew 1996 Study type review
McKnight 1987 Study type evaluation of methodological and administrative aspects of a motorcycle operator skill test
McPherson 1976 Outcome knowledge and skills
Mohan 2004 Study type review of evidence of interventions to reduce road traffic injuries
Morgan 1980 Study type reviewcourse description
Motoki 1990 Study type proposal and development of rider training curriculum
Nairn 1992 Study type literature review
Newland 1980 Study type description of development of motorcycle safety course
Newland 1982 Study type descriptionreview of a rider training course
Newland 1990 Study type uncontrolled BeforeAfter study
Ormston 2003 Outcome changes in attitude and self reported behaviour
Osga 1979 Outcome knowledge attitudes and skills
Panou 2007 Intervention new simulation tool to teach driverrider
Peterson 1990 Study factor demographic details of course participants completing the course vs those not completing the
course
Prem 1984 Study type experimental
Study factor examines association between skills and performance in tests
Radin Umar 2006 Intervention motorcycle safety programme not specifically rider training
Rockwell 1990 Study type methodologically flawed study type
43Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4650
(Continued)
Rothe 1987 Study type qualitative study of rider training in general
Rowden 2007 Study type qualitative study
RSC Victoria 1993 Study type review of motorcycle rider training and education in Victoria
Russam 1979 Study type review of training programmes in Japan
Saffron 1981 Study type literature review
Schulz 1990 Study factor predictors of motorcycle crashes
Sexton 2004 Study type cross-sectional analytic analysis
Shephard 1986 Study factor reviewdescription of programme
Simpson 1990 Study type review of motorcycle rider trainingeducation
Sokolov 1990 Study factor review of motorcycle riding courses in USSR
Steffens 1988 Outcome self-reported behaviour experiences and researcher observed riding behaviour
Sudlow 2003 Study type review of motorcycle training in the UK
Tasmania 1995 Study type cross-sectional descriptiveprocess evaluation
Tetard 1985 Study type experimental
Outcome detection of indices
Thackray 1980 Outcome instructional effectiveness user acceptance administrative feasibility
Thompson 1994 Study type cross-sectional survey
Outcome attitudes and knowledge
Tobler 1990 Study type review of course curricula
Transport 1989 Intervention syllabus for rider training
Tuan 2007 Study type cross-sectional study (survey) which does not assess the effects of an intervention but examines
motocyclistsrsquo behaviour
Victor 1980 Study type review
Waterer 1990 Study type review of UK rider training issues
Weller 1989 Study type review of studies evaluating motorcycle safety and education measures
44Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4750
(Continued)
Wells 1981 Study type description of different methods of training
Winn 1986 Study type review of rsquohistoryrsquo of rider training
Wisher 1988 Study type cross-sectional survey
Wood 1987 Study type Interrupted Time Series but did not meet the criteria for an ITS as published by the Cochrane
Effective Practice and Organisation of Care Group
Yamazaki 1990 Study type cross-sectional survey
Outcome riding skills skills retention and attitude
Characteristics of ongoing studies [ordered by study ID]
Burgess 2010
Trial name or title UKs National Ride program
Methods Cohort study
Participants Motorcycle offenders
Interventions Police led diversion program
Outcomes Self reported attitudinal and behavioural measures
Starting date
Contact information
Notes Expected to be completed in early 2010
45Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4850
D A T A A N D A N A L Y S E S
This review has no analyses
A P P E N D I C E S
Appendix 1 Search strategy
We searched the following electronic databases
bull Cochrane Injuries Group Specialised Register (searched 22 Sept 2008)
bull CENTRAL (The Cochrane Library 2008 Issue 3)
bull TRANSPORT 1988 to 200706 (latest date available)
bull ATRI Australian Transport Index (to Sept 2008)
bull MEDLINE (Ovid SP) 1950 to Sept 2008
bull EMBASE (Ovid SP) 1980 to Sept 2008
bull CINAHL (EBSCO) 1982 to Sept 2008
bull WHOLIS (World Health Organization Library Information System) (to Sept 2008)
bull PsycINFO (Ovid SP) 1806 to Sept 2008
bull LILACS (BIREME) (Latin American and Caribbean Health Sciences) (1982 to Sept 2008)
bull ISI Web of Science Social Sciences Citation Index (SSCI) 1970 to Sept 2008
bull ERIC (searched 25 Sept 2008)
bull ZETOC (searched 25 Sept 2008)
bull SIGLE (searched 25 Sept 2008)
All of the databases were searched using the following text-word search strategy The strategy was adapted using either rsquorsquo or rsquo$rsquo as atruncation symbol
1 motor-cycl or motorcycl or motor-bik or motorbik or motor-scooter or motorscooter or scooter or moped or motocycl
or motocicl or moto
2 train or educat or instruct or program or course or qualif or learn or improv
3 1 and 2
H I S T O R Y
Protocol first published Issue 2 2005
Review first published Issue 10 2010
46Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4950
C O N T R I B U T I O N S O F A U T H O R S
Katina Kardamanidis was the first author of the protocol conducted the literature search extracted full studies assessed the studies
according to inclusion criteria assessed quality of the studies compared findings with the second author discussed disagreements with
the third author wrote the drafts and the final review
Rebecca Ivers was the second author of the protocol assessed the studies according to inclusion criteria assessed quality of the studies
compared findings with the first author edited the final draft and provided overall guidance for the process
Alexandra Martiniuk assessed quality of the studies compared findings with the first author discussed disagreements with the third
author and helped write the first draft and the final review
Mark Stevenson contributed to the protocol reviewed methods of studies and contributed to the final draft
Katrina Thistlethwaite conducted a second literature search extracted full studies and contributed to the final draft
D E C L A R A T I O N S O F I N T E R E S T
None known
S O U R C E S O F S U P P O R T
Internal sources
bull No sources of support supplied
External sources
bull National Health and Medical Research Council of Australia Australia
Salary support for Martiniuk Ivers and Stevenson
bull University of Sydney Sesqui Research Grant Australia
Direct research costs and salary support for Kardamanidis
D I F F E R E N C E S B E T W E E N P R O T O C O L A N D R E V I E W
There are no differences between the protocol and the review
I N D E X T E R M S
Medical Subject Headings (MeSH)
lowastMotorcycles Accident Prevention [lowastmethods] Accidents Traffic [lowastprevention amp control] Licensure Program Evaluation Wounds
and Injuries [mortality lowastprevention amp control]
47Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 5050
MeSH check words
Humans
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2250
Injuries and deaths
Hurt 1981 found that the odds of injurious crashes were lower inthose who had received professional training compared to those
who had not Kraus 1975 found that in the younger age group
(less than 20 years) and the older age group (over 35 years) a larger
percentage of injured riders (12 and 21 respectively) reported
prior motorcycle rider training compared with the uninjured con-
trol group (6 and 5 respectively) (P lt 001) (no numbers
given) Haworth 1997 reported several different outcomes one
beingthat having completed at leastone trainingcoursecompared
to no training courses at all had no significant effect on the odds
of crashing (afteradjustment for age blood alcohol level or licence
status (learner permitprobationaryfull)) When comparing three
different levels of training courses (beginner intermediate and ad-
vanced) with no training no significant differences were found
in the odds of crashing However after adjustment for age and
blood alcohol level completing an advanced course as compared
to a beginners course was associated with a significant decrease in
the odds of severe motorcycle crashes resulting in serious injury or
death (OR 02 95 CI 01 to 08)
D I S C U S S I O N
The objective of this review was to quantify the effectiveness of
pre- andpost-licence motorcycle rider training on the reduction of
traffic crash involvement injuries deaths and offences committedby motorcycle riders Most included studies were cohort studies
There were a few randomised and non-randomised trials and case-
control studies
All studies compared either mandatory or non-mandatory formal
training with no formal training except for Haworth 1997rsquos case-
control study that compared levels of training (beginner interme-
diate and advanced) to each other
We found studies that evaluated the effectiveness of manda-
tory pre-licence training in reducing crash and offence rates
(Anderson 1980 Hall 1988 Kloeden 1994 Billheimer 1998
(the basic California Motorcyclist Safety Program component of
the study) Perrino 2002) non-mandatory training that when
passed resulted in a licence (Cooper 1988 McDavid 1989
Swaddiwudhipong 1998) non-mandatory training that could be
taken by motorcycle riders with or without a licence (Raymond
1979 Osga 1980 Satten 1980 Jonah 1982 Mortimer 1984
Adams 1985 Leung 1987 Mortimer 1988 Waller 1992 Davis
1997 Billheimer 1998 (the Experienced Rider Course compo-
nent of the study)) and unspecified rider training on crash injury
and death rates (Kraus 1975 Hurt 1981 Haworth 1997 MAIDS
2004)
The studies provided conflicting evidence regarding motorcycle
rider training which is likely due to the poor design quality and
reporting of the studies Most of the studies included in the review
were likely to be adversely affected by significant selection biasalong with substantial detection bias(measurement and recall bias)
which limits our ability to make conclusive comments
Mandatory pre-licence rider training
There were two RCTs three cohort studies and one case-control
study that examined pre-licence ridertraining The two RCTs eval-
uated the Motorcycle Skills Operating Test (MOST) (Anderson
1980 Hall 1988) Anderson 1980 found a statistically significant
reduction in crashes in the intervention group at six months one
and twoyearsfollow up However when only participantswho ob-
tainedtheirlicence were included in theanalysis thereduction wasstatistically significant only at one year follow up The other RCT
(Hall 1988) found no statistically significant differences in crash
rates at any point of time neither in the total population group
nor in the licensed subgroup Neither of the two cohort stud-
ies assessing crash rates found a statistically significant difference
in crash rates between intervention and control groups ( Kloeden
1994 Billheimer1998) andthe case-control study (MAIDS 2004)
also did not find a statistically significant difference in the odds of
experiencing a crash between trained and untrained riders With
regards to offences neither of the RCTs found a statistically sig-
nificant difference in offence rate between the intervention and
control groups in the analysis of the total group of participants
entering the trial However Anderson 1980 found a statistically significant increase in offence rates in the intervention group but
only after two years of follow up andonly in the analysis of partici-
pants who obtained their licence The two cohort studies assessing
offence rates found no statistically significant difference between
the groups (Billheimer 1998 Perrino 2002)
None of the five studies (RCTs or cohort) found a reduction in
crashes or offences following mandatory pre-licence rider training
except for Anderson 1980 who observed a statistically significant
reduction in crashes in the intervention group at one year in both
analyses It is not clear why both RCTs that evaluated the same
MOST and training intervention observed different results Both
studies enrolled large samples and observed a similar uptake of
licensure following the intervention It is possible that differencesmay have arisen due to different study samples (California ver-
sus New York) or that the intervention delivery differed between
studies
In both trials of the Motorcycle Skills Operating Test including
rider training only a third of motorcycle licence applicants ob-
tained their motorcycle licence It is possible that a licensing pro-
cess including a skills test may reduce crashes through decreased
exposure
Non-mandatory pre-licence training
20Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2350
Three studies assessed the effect of non-mandatory pre-licence
training The Cooper 1988 cohort study found no difference incrash or offence rates between the trained and untrained groups
We cannot however be confident about this result as the study
sufferedfrom selectionbiaswhichcouldhave biasedthe measure of
effect toward the null The RCT conducted by Swaddiwudhipong
1998 examined injury and death rates and found a statistically
significant lower injury rate in the intervention villages after one
year of follow up but no longer at two year follow up There
was no statistically significant difference in death rates between
intervention and control villages Our confidence in the results of
this study are hindered by the inadequacy of the randomisation
process that led to significant differences between the intervention
and control districts on two important variables The intervention
districts were more likely than the control districts to have a validlicence andto wear a helmetSincethese differences were nottaken
into account in later analyses they could explain the differences
observed in injury rates between districts Furthermore with only
six districts randomised the sample size was probably too small
to detect a difference between intervention and control groups
had there been one The McDavid 1989 cohort study found a
lower percentage of injuries in the intervention group although
a statistical test was not conducted McDavid 1989 drew their
intervention and control groups from different populations but
did adjust for confounders although they only analysed 40 of
the initial sample (139 trained and 139 untrained) Because of
these substantial weaknesses we are unable to draw conclusions
from this studyAn overall effect of pre-licence training on the reductionof crashes
injuries deaths and offences was not found although one should
be aware that the research of this issue to date has had significant
flaws However a mandatory skill test (such as the MOST) in
combination with and even without pre-licence training as part
of a licensing process seems to reduce the number of individuals
completing the licensing procedure and therefore indirectly re-
duces crash rates through a reduction in exposure to motorcycle
riding Making licensure a more difficult process may therefore be
seen as advantageous if it delays licensure and therefore reduces
exposure It is possible that the group who chose not to continue
riding a motorcycle would not have been the group to experience a
crash or commit an offence if they had continued to ride anyhowfor example because they are more safety conscious and this was
the reason for their giving up riding in the first place If this is the
case deterring individuals from completing their licence and rid-
ing may not necessarily decrease overall crashes deaths or offence
rates In addition other potentially adverse effects of complicat-
ing a licensing procedure need to be considered such as increases
in unlicensed riding In Thailand non-mandatory rider training
offered as a community intervention may have decreased injury
rates but this needs to be confirmed through further research
Non-mandatory training
The majority of the studies assessed non-mandatory training
(training independent of any licensing procedure) and the results
were varied The two non-randomised trials assessing crash rates found no statistically significant difference in crash rates between
the intervention and control groups (Adams 1985 Leung 1987)
Of the nine cohort studies Satten 1980 found a lower percentage
of crashesin the intervention group although it is unclear whether
this was a statistically significant result Davis 1997 also found a
lower percentage of crashesin theinterventiongroup although it is
unclear with which control group the intervention group wascom-
pared Raymond 1979 Osga 1980 Waller 1992 and Savolainen
2007 found a statistically significant higher crash rate in the in-
tervention group and the remaining cohort studies (Jonah 1982
Mortimer 1984 Mortimer 1988 Billheimer 1998) found no sta-
tistically significant difference in crash rates between intervention
and control groups No conclusions can be drawn from these stud-
ies with confidence The Satten 1980 and Davis 1997 studies did
not adjust for any confounding factors even though important
differences existed between the intervention and control groups
Although Raymond 1979 and Waller 1992 adjusted for a num-
ber of confounders at baseline (Waller 1992) and in the analy-
sis (Raymond 1979 Waller 1992) the intervention and control
groups were drawn from such different populations that it is pos-
sible that unmeasured confounders were responsible for the sta-
tistically significant increased risk of crashes observed in the inter-
vention group The observed difference could also be attributed totheobserved differential loss to followup (Raymond 1979) which
could bias the measure of effect towards or away from the null
The Osga 1980 study found a protective effect of rider training
in the univariate analysis but statistically significant higher crash
rates in the intervention group when accounting for riding expo-
sure Sex (a significantly higher proportion of females in the in-
tervention group) and other potentially confounding factors that
could explain thedifferencebetween the groups were howeverun-
accounted for Finally Savolainen 2007 did control for potential
confounding variables yet it is not clear whether variables were
controlled for simultaneously or one at a time One case-control
study (MAIDS 2004) found no statistically significant reduction
in the odds of experiencing a crash between trained and untrainedriders but the quality of this study was poor mainly because the
analysis did not account for any confounding factors
One cohort study (Satten 1980) assessed the effect of non-manda-
tory rider training on injurious crashes and found a lower percent-
age of injuries in the intervention group although a statistical test
was not conducted As mentioned above the Satten 1980 studyrsquos
validity is reduced due to its lack of adjustment for confounding
factors
The only non-randomised trial that assessed offence rates (Leung
1987) found a statistically significant lower offence rate in the in-
tervention group and of the five cohort studies that assessed of-
fence rates (Satten 1980 Jonah 1982 Mortimer 1984 Mortimer
21Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2450
1988) one study (Jonah 1982) also found a statistically significant
lower offence rate in the intervention group However we have lit-tle confidence in the results from Leung 1987 since potential con-
founding factors were not appropriately addressed Jonah 1982
did conduct a multivariate analysis but was hindered by an initial
low response rate The other four cohort studies did control for
several potential confounders and found no difference in offence
rates following non-mandatory post-licence rider training
The non-randomised trials and cohort studies evaluating non-
mandatory rider training were at high risk of substantial selection
bias due to the non-random allocation of participants as well
as low participation rates and attrition bias and could therefore
not provide conclusive evidence with regard to the effect of the
training It should be noted that despite the biases and other issues
with this group of studies most didnot detectmajor effects relatedto rider training in either direction This may suggest that any
potential impact of rider training will be modest
Studies examining unspecified modes of ridertraining
Three case-control studies assessed whether injured or deceased
motorcyclists were more or less likely to have had rider training
compared to control motorcyclists All had conflicting results
Hurt1981 foundthe odds of havingan injuriouscrash waslower in
those who had received rider training whereas Kraus 1975 found
that the odds of having an injurious crash were higher in youngerand older trained riders Haworth 1997 who compared advanced
versus beginner training found no effect on motorcycle injuries
or deaths although they did find that those who had completed
advanced training had reduced chances of crashing compared to
those who had done a beginnersrsquo training course Similar issues
with uncontrolled confounding and biases of selection as described
above arose with the three case-control studies Hurt 1981 suffers
from inadequate control of confounding via poor matching and
both the Kraus 1975 and the Haworth 1997 findings were only
controlled for limited factors by means of stratified analysis It is
likely that these limitations are the reason for the contradictory
directions of effect observed for the three case-control studies
Methodological weaknesses of the studies
In summary most studies investigating the effectiveness of mo-
torcycle rider training on reducing crashes injuries and death suf-
fered from serious methodological weaknesses
The majority of studies were non-randomised and were there-
fore limited mainly by selection bias Although the approaches of
matching stratification or regression analyses can minimise bias
(Fleiss 2004) they were poorly employed in the reviewed studies
The inadequate adjustment for confounders in particular adds
to our lack of confidence in the studiesrsquo results since it has been
demonstrated that individuals who choose to take motorcycle rider
training differ from those not taking rider training For instance
trained riders often differ from controls in age sex and experi-ence usually ride less and may have a more positive attitude to
safety as indicated by the reported more frequent wearing of seat
belts (Mortimer 1982)
The two large randomised trials reported that a large proportion
of their study group were never exposed to the risk of offences or
crashes because they never attained licensure This made it impos-
sible to assess whether a reduction in crashes in the intervention
group was a result of training or due to delayed or never attained
licensure
In some studies the sample sizes were too small to detect any sta-
tistically significant differences Follow-up time after training dif-
fered between studies and was an impediment to observing any
real effects of training if they existed As riding experience ac-cumulates over time any differences in outcomes between those
trained and untrained would diminish in studies with lengthy fol-
low-up times This dilution of any possible effect of rider training
may have also occurred in studies that mixed both novice and ex-
perienced riders in the study groups since it is hypothesised that
novice riders serve to benefit most from rider training This would
increase the difficulty in finding any effect of rider training par-
ticularly when sample sizes are small
Detection bias may also account for some of the differences ob-
served between trained and untrained riders Detection bias would
have occurred if the self-report hospital driver or police records
used to record the outcomes of crash injury death or offences
differed between the trained and untrained riders Research hasdemonstrated that not all crashes are reported to police for ex-
ample small crashes or off-road crashes (Kraus 1975) even serious
injury crashes have been found to be underreported (Anderson
1980) However this would only introduce detection bias if the
reporting of outcome differs for trained and untrained riders This
does appear possible in that responsible riders may be more likely
to enroll in a rider training course and also may be more likely
to report crashes to police than irresponsible riders meaning that
the number of crashes from riskier riders could be underrepre-
sented on records An analysis to investigate the effectiveness of
rider training may then show that rider training is not effective in
reducing crashes
Rather than using reported outcomes some studies used self-re-ported survey data although this technique suffers from non-
response biases In fact two studies reported differences in de-
mographic characteristics in response rates to questionnaires
(Anderson 1980 Hall 1988) Self-reported data also introduce
the risk of recall bias survival bias and social desirability response
bias Recall bias may have been equally distributed among trained
and untrained riders howeverit is hypothesised that trained riders
may have felt it was lsquoexpectedrsquo to report good driving histories
hence leading the rider training to appear effective
22Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2550
Limitations of this review
Limitations of this review were that we found very few recentstudies (the two most recent ones published in 2002 and 2007)
and that we found very few studies conducted in low-and middle
income countries even though our search strategy was set up to
identify such studies Anotherlimitation wasthat forthe reviewed
studies we could not assess all outcomes of rider training that are
potentially useful Some of the included studies assessed outcomes
such as use of safety clothing including helmets Such outcomes
are valuable since the use of motorcycle helmets has been demon-
strated to be effective in preventing head injuries (Liu 2004) De-
spite this the use of safety clothing was not included as an out-
come in this review because the studies measuring this outcome
were limited by poor data quality due to very low response rates
Another outcome of interest in assessing the effectiveness of rider
training is lsquotype of injuryrsquo As Haworth 2005 state in their review
of motorcycle licensing and training rider training could be con-
sidered effective if injuries become less severe because motorcycle
riders have learned how to avoid more serious injuries through
rider training The studies included in this review did not specify
type of injury and so this outcome could not be assessed Last
changes in knowledge and attitudes following rider training are
other potentially useful outcomes We have excluded such studies
(Osga 1979 Clayton 1990 Goldenbeld 2004) as these outcomes
are proxies rather than direct measures of the effectiveness of mo-
torcycle rider training We opted to report on the outcomes of
crashes injury death and offences alone since there is little ev-
idence about how changes in knowledge and attitudes relate to
these outcomes
A U T H O R S rsquo C O N C L U S I O N S
Implications for practice
Due to the poor quality of studies included in this review we were
unable to draw any conclusions about the effectiveness of rider
training on crash rates injurious or fatal or offence rates The
findings do suggest that mandatory pre-licence training may be
an impediment to completing a motorcycle licensing process thuspossibly indirectly reducing crash injury death and offence rates
through a reduction in exposure to riding a motorcycle On the
basis of the existing evidence it is not clear if and what type of
training reduces the risk of crashes injuries deaths or offences in
motorcyclists and the selection of the best rider training practice
can therefore not be recommended The suggestion that manda-
tory pre-licence training indirectly reduces exposure to riding the
motorcycle because trainees are impeded from completing the li-
censing process does not help those motorcyclists who persist in
obtaining their licence and riding the motorcycle Some type of
rider training is likely necessary to teach motorcyclists to ride a
motorcycle safely Good policy is based on sound evidence and
evaluation but such evidence for motorcycle rider training is seri-
ously lacking Policy makers are therefore encouraged to include arigorous evaluation component to any new or existing rider train-
ing programme Such an evaluation should ideally be randomized
and assess whether the training programme is effective and if so
which component(s) of the programme work best
Implications for research
Given theresults of this review which found littlereliable evidence
about the effectiveness of motorcycle rider training it is necessary
to build up an evidence base of the effectiveness of motorcycle
rider training in the reduction of crash injury death and offence
rates To this end well designed randomised controlled trials need
to be conducted that examine the impact of rider training oncrashes and offences A randomised controlled trial is the preferred
method as it would significantly reduce most biases described in
this review To minimise the impact of participants delaying or
never attaining their licence and therefore exposure as found in
the randomised trials in this review such a trial should include
riders who are already licensed and reliable measures of riding
exposure in a reasonably lengthy follow up period It is expected
that a randomised controlled trial of sufficient sample size would
observe equal balance on possible confounders by study arm how-
ever an assessment of potentially important confounders (such as
helmet wearing) by study group would enable control for these
variables in the analyses if necessary It is imperative that the effec-
tiveness of each component of the rider training (such as theorypractice knowledge skills and modification of behaviour) and the
synergy of such components are evaluated Outcomes should be
objective ideally those that are routinely collected rather than self-
reported outcomes Even though riding the motorcycle poses dif-
ferent challenges and incurs different risks lessons may be learned
from research into the effects of car driver training
Motorcycles are a common means of transport in low- and middle
income countries and usage is increasing As the burden of mo-
torcycle crash victims is mostly borne in such countries research
should be focused on the effectiveness of types of rider training on
the reduction of crashes injuries and fatalities in such countries
and be supported by higher income countries This review found
only one study conducted in a middle income country that sug-gested decreasing injury rates It would be beneficial to conduct
additional research to confirm this suggestion and to enhance our
knowledge of the role of rider training in low- and middle-income
settings
March 2006 saw the launch of the University of North Carolina
Highway Safety Research Centre partnership with the Motorcycle
Safety Foundation (MSF) and plans were revealed to conduct a
study of the potential benefits of motorcycle rider training The
study is funded by the MSF and the National Highway Traffic Sa-
fety Administration (NHTSA) Called rsquoThe Discovery Projectrsquo it
will compare data on crash rates from riders participating in a basic
23Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2650
motorcycle safety class with riders participating in a basic motor-
cycle safety class as well as a variety of continuing education classesThe collection of pilot data was due to start in March 2007 and
the data collection portion of the study is expected to take three
years (Highway Safety Research Center) It is vital that researchers
and policy makers collaborate to develop furtherhigh quality eval-
uations of motorcycle rider training programmes to help inform
development of new and effective training programmes
A C K N O W L E D G E M E N T S
Staff from The University of Sydney library for helping us locate
and retrieve full reports of many papers
R E F E R E N C E S
References to studies included in this review
Adams 1985 published data only
Adams A Collingwood V Job RFS Evaluation of the 1982 Australia
Post motorcycle rider training scheme NSW Australia Traffic
Accident Research Unit Traffic Authority NSW 1985 [ RN 85]
Anderson 1980 published data only
Anderson J Ford JL Peck RC Improved motorcyclist licensing and testing project Washington DC National Highway Traffic Safety
Administration (NHTSA) USA 19801ndash217 [
DOTndashHSndash805ndash598]
Billheimer 1998 published data only
Billheimer JW Evaluation of the California Motorcyclist Safety
Program Transportation Research Record 19981640100ndash9 [
TRB Report Number 980652]
Cooper 1988 published data only
Cooper PJ Rothe JP Motorcyclists who they are and why they do
what they do Transportation Research Record 1988116878ndash85
Davis 1997 published data only
David CF Evaluation of community traffic safety programs and
motorcycle operator training programs Connecticut Department of Transportation 199763
Hall 1988 published data only
Hall CJ Buchanan L Engle R Motorcycle rider education evaluationproject Washington DC USA National Highway Traffic Safety
Administration (NHTSA) US Department of Transportation
1988 [ DTNH 22ndash80ndashCndash05123]
Haworth 1997 published data only
Haworth N Smith R Brumen I Pronk N In Monash University
Accident Research Centre editor(s) Case-control study of motorcycle
crashes Victoria Australia Department of Transport and Regional
Development The Federal Office of Road Safety 19971ndash149 [
ISBN 0 642 25501 6 ]
Hurt 1981 published data only
Hurt H Ouellet J Thom D Motorcycle accident cause factors and
identification of countermeasures USA Department of
Transportation 1981 [ Technical Report DOT HS 806ndash862]
Jonah 1982 published data onlylowast Jonah BA Dawson N Bragg B Are formally trained motorcyclists
safer Accident Analysis amp Prevention 198214(4)247ndash55
Jonah BA Dawson NE Bragg BW Evaluation of the Canada
Safety Councilrsquos Motorcycle Training Program Proceedings of theInternational Motorcycle Safety Conference Washington DC
USA 198059ndash93
Kloeden 1994 published data only
Kloeden CN Moore VM McLean AJ Evaluation of the pre-license
training program for motorcyclists in South Australia Walkerville
South Australia Australia NHMRC Road Accident Research Unit
The University of Adelaide 19941ndash37 [ ISBN 0 7308 0607 3]
Kraus 1975 published data only
Kraus JF Riggins RS Franti CE Some epidemiological features of
motorcycle collision injuries American Journal of Epidemiology 1975102(1)74ndash98
Leung 1987 published data only
Leung KS Reding VA Evaluation of the Wisconsin motorcycle rider course Wisconsin Bureau of Policy Planning and Analysis
Wisconsin Department of Transportation 1987
MAIDS 2004 published and unpublished data
MAIDS In-depth investigations of accidents involving powered
two wheelers ACEM (Acemacemeu) 2004 Vol Final report
20179
McDavid 1989 published data only
McDavid JC Lohrmann BA Lohrmann G Does motorcycle
training reduce accidents Evidence from a longitudinal quasi-
experimental study Journal of Safety Research 198920(2)61ndash72
Mortimer 1984 published data only
Mortimer R Evaluation of the motorcycle rider course Accident
Analysis amp Prevention 198416(1)63ndash71
24Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2750
Mortimer 1988 published data only
Mortimer RG A further evaluation of the motorcycle rider courseJournal of Safety Research 198819(4)187ndash96
Osga 1980 published data only
Osga GA An investigation of the riding experiences of MSF rider
course participants Dissertation South Dakota USA The
University of South Dakota 19801ndash224
Perrino 2002 published data only
Perrino CS Ahmed A Callender A Rozier E Cantwell A Stewart
O The role of Marylandrsquos motorcycle rider course in promoting saver
behaviors and attitudes Baltimore USA National Transportation
Center Morgan State University 20021ndash52
Raymond 1979 published data only
Raymond S Tatum S An evaluation of the effectiveness of the RAC
ACU motorcycle training scheme - final report Road Safety ResearchUnit Dept of Civil Engineering University of Salford London
19791ndash131
Satten 1980 published data only
Satten RS Analysis and evaluation of the motorcycle rider courses
in thirteen Illinois counties International Motorcycle Safety
Conference Washington DC 1980145ndash93
Savolainen 2007 published and unpublished data
Savolainen Peter Mannering Fred Additional evidence on the
effectiveness of motorcycle training and motorcyclistsrsquo risk-taking
behaviour Transportation Research Board 2007 Vol March 20
19
Swaddiwudhipong 1998 published data only
Swaddiwudhipong W Boonmak C Nguntra P Mahasakpan P
Effect of motorcycle rider education on changes in risk behavioursand motorcycle-related injuries in rural Thailand Tropical Medicine
and International Health 19983(10)767ndash70
Waller 1992 published data only
Waller FL An analysis of the impact of the Texas Department of
Public Safety Motorcycle Operator Training Program upon
motocycle related traffic accidents Texas State University Public
Administration Program Applied Research Projects 199277
References to studies excluded from this review
Anderson 1980a published data only
Anderson LC Better biking a motorcycle rider improvement
program International Motorcycle Safety Conference Washington
DC USA Motorcycle Safety Foundation 1980231ndash53Aupetit 2008 published and unpublished data
Aupetit S Riff J Buttelli O Espie S Ergonomics research program
on motorcycle learning Contributions to the development of a
motorcycle simulator for training Transport Research Arena
Ljubljana 20089
Awane 1999 published data only
Awane T Integrating simulators in motorcycle safety education
Journal of International Association of Traffic and Safety Sciences 199923(1)26ndash35
Baldi 2005 published data only
Baldi S Baer JD Cook AL Identifying best practices states in
motorcycle rider education and licensing Journal of Safety Research200536(1)19ndash32
Batchler 1988 published data only
Batchler MW Evaluation of the pre-license motorcycle training course at the New South Wales traffic education centre Federal Office of
Road Safety (FORS) 19881ndash122 [ Report No CR 77 ]
Blanchard 2006 published data only
Blanchard HT Tabloski PA Motorcycle safety educating riders at
the teachable moment Mosby (CV) Company 200632330ndash2
Bonnett 2005 published data only
Bonnett J Mature age skills training for experienced motorcycle
riders Motorcycle ridersrsquo association of the Australian Capital
Territory 2005
Braver 2007 published data only
Braver ER et alPersuasion and licensure a randomized controlled
intervention trial to increase licensure rates among Maryland
motorcycle owners Traffic Injury Prevention 20078(1)39ndash46Clayton 1980 published data only
Clayton AB Sudlow DE An evaluation of the effectiveness of a one day training programme for learner motorcyclists Berkshire UK
Transport and Road Research Laboratory 19801ndash23 [ ISSN
0266ndash7045]lowast Clayton AB Sudlow DE An evaluation of the effectiveness of a one-day training programme for learner motorcyclists Berkshire UK
Road Safety Division Berkshire 19801ndash23 [ISSN 0266ndash7045]
Council 1978 published data only
Council FM Desper LP Dutt AK Driver education for motorcycle operation University of North Carolina Highway Safety Research
Center 19781ndash40
Crowther 2005 published data onlylowast Crowther G Engaging with motorcyclists UK police and theBikeSafe Road Safety Programme Published proceedings of the
Australasian Road Safety Research Policing Education Conference
Wellington New Zealand 2005 issue ISBN 0ndash473ndash10636ndash11ndash11
Daniello 2009 published data only
Daniello A Gabler HC Mehta YA Effectiveness of motorcycle
training and licensing Transportation Research Board 88th Annual
Meeting 200913
Dewanti 2007 published data only
Dewanti M Challenging efforts in promoting young motorcyclist
safety in Indonesia Road Safety on Four Continents 14th
International Conference Swedish National Road and Transport
Research Institute 2007
Du Plooy 1990 published data onlyDu Plooy DW Motorcycle training in South Africa The Human
Element International Motorcycle Safety Conference Orlando
Florida USA Motorcycle Safety Foundation 19904-1 to 4-9
Engel 1989 published data only
Engel U Krogsgaringrd Thomsen L [Faeligrdselsundervisning og
propaganda ndash en faeligrdselssikkerhedsfremmende foranstaltning]
Dansk Vejtidsskrift 198912295ndash7
Goldenbeld 2002 published data only
Goldenbeld C Houwing S de Craen S The development of
driving skills of young moped riders [De ontwikkeling van de
rijvaardigheid van jonge bromfietsers] SWOV Institute for Road
Safety Research (Stichting Wetenschappelijk Onderzoek
Verkeersveiligheid) Leidschendam the Netherlands 200263
25Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2850
Goldenbeld 2004 published data only
Goldenbeld C Twisk D de Craen S Short and long term effects of moped training a field experiment Transportation Research Part F 20047(1)1ndash16
Haworth 2000 published data only
Haworth N Smith R Kowadlo N Motorcycle training in Victoria
evaluation of rider training curriculum in Victoria Monash
University Accident Research Centre Reports 2000 Vol 165 issue
3131
Haworth 2005 published data only
Haworth N Mulvihill C Review of motorcycle licensing and
training Monash University Accident Research Centre 2005 [
Report no240]
Haworth 2006 published data only
Haworth N Mulvihill C Rowdon P Teaching old dogs new tricksTraining and older motorcyclists Journal of the Australasian College of Road Safety 200718(4)20ndash5lowast Haworth NL Mulvihill C Rowden PJ Teaching old dogs new
tricks Training and older motorcyclists Australasian Road Safety
Research Policing and Education Conference Gold Coast
Queensland 2006
Hill 1983 published data only
Hill D Motorcycle rider training testing and licensing Motor
Cycle Council of New South Wales Sydney 1983
Imatake 1980 published data only
Imatake Y On the activities in Japan of motorcycle safe riding
promotion committee (MSRPC) - especially the education of
unskilled riders of class 1 motorbikes International Motorcycle
Safety Conference Washington DC USA Motorcycle Safety
Foundation 1980255ndash68
Ingebritsen 1990 published and unpublished data
Ingebrigtsen S [Risikofaktorer ved ferdsel med moped og
motorsykkel] TOslashI-rapport 66 Oslo Transportoslashkonomisk
institutt 1990
Jain 2009 published and unpublished data
Jain A et alTwo wheeler accidents on Indian roads a study from
Mangalore India Elsevier 200916130ndash3
Jonah 1979 published data only
Jonah BA Dawson NE Validation of the motorcycle operator skill
test Accident Analysis amp Prevention 197911(3)163ndash71
Jonah 1981 published data onlyJonah BA Dawson NE Bragg WE Predicting accident
involvement with the Motorcycle Operator Skill Test Accident Analysis amp Prevention 198113(4)307ndash418
Kadar 1990 published data only
Kadar KJ Countermeasures motorcycle rider education delivery
system The Human Element International Motorcycle Safety
Conference Motorcycle Safety Foundation 19906-27 to 6-41
Kelsey 1986 published data only
Kelsey SL Liddicoat C Ratz M Licensing novice motorcyclists a
comparison of Californiarsquos standard test and the MOST II (Motorcycle
Operator Skill Test) administered at centralized testing offices Sacramento California USA California Department of Motor
Vehicles Research and Development Office 19861ndash74
Koch 1980 published data only
Koch H Advanced course for motorcycle riders InternationalMotorcycle Safety Conference Washington DC USA Motorcycle
Safety Foundation 1980269ndash81
Koch 1990 published data only
Koch H Brendicke R How to influence risk taking behaviour
within motorcycle rider training programs The Human Element
International Motorcycle Safety Conference Orlando Florida
USA Motorcycle Safety Foundation 1990Motorcycle Safety
Foundation 1990 p 9-1 to 9-25A
Labbett 2005 published data only
Labbett S Langham M Training can make the problem worse
Road Safety Congress 20051ndash12
Law 2005 published and unpublished data
Law TH Radin Umar RS Zulkaurnain S Kulanthayan S Impactof the effect of economic crisis and the targeted motorcycle safety
programme on motorcycle-related accidents injuries and fatalities
in Malaysia International Journal of Injury Control and Safety
Promotion 200512(1)9ndash21
Lowes 1990 published data only
Lowes BF In-traffic rider training The Human Element
International Motorcycle Safety Conference Orlando Florida
USA Motorcycle Safety Foundation 19907-46 to 7-57
Mahon 1976 published data only
Mahon J Teaching accident avoidance on the motorcycle Journal of Traffic Safety Education 1976231ndash29
Maxwell 1985 published data only
Maxwell GS Evaluation of a Program for Training MotorcycleRiders Evaluation and Program Planning 19858(4)339ndash47
Mayhew 1996 published data only
Mayhew DR Simpson SH Effectiveness and role of driver educationand training in a graduated licensing system Ottowa Canada The
Traffic Injury Research Foundation 19961ndash89 [ ISBN
0ndash920071ndash12ndash0]
McKnight 1987 published data only
McKnight AJ Milburn BG An evaluation of the modified
motorcycle operator skill test exercises NHTSA Report 1987 [
Report No DOT HS 807 212 ]
McPherson 1976 published data only
McPherson K McKnight AJ The development and evaluation of a
motorcycle skill test manual and knowledge test Washington DCUSA National Highway Safety Administration 1976
Mohan 2004 published data only
Mohan D Evidence-based interventions for road traffic injuries in
South Asia Journal of the College of Physicians and SurgeonsPakistan 200414(12)746ndash7
Morgan 1980 published data only
Morgan JK Advanced motorcycle training course content and
study evaluation International Motorcycle Safety Conference
Washington DC USA Motorcycle Safety Foundation 1980
283ndash312
Motoki 1990 published data only
Motoki M Yamazaki S A study on effective motorcycle rider
education The Human Element International Motorcycle Safety
26Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2950
Conference Orlando Florida USA Motorcycle Safety
Foundation 19909-26 to 9-57A
Nairn 1992 published data onlylowast Nairn RJ Motorcycle safety research literature review 1987 to
1991 Federal Office of Road Safety Australia 1993 Vol Report
CR 117
Newland 1980 published data only
Newland R Darby J Automotive practicespre motorcycle rider
program International Motorcycle Safety Conference Washington
DC USA Motorcycle Safety Foundation 1980127ndash9
Newland 1982 published data only
Newland R Motorcycle rider training and licensing requirements
Driver training steering a course for the future Melbourne
Victoria Transport Regulation Board 1982
Newland 1990 published data only
Newland R Motorcycle training and licensing in Victoria 1980-
1990 The Human Element International Motorcycle Safety
Conference Orlando Florida USA Motorcycle Safety
Foundation 1990
Ormston 2003 published data only
Ormston R Dudleston A Pearson S Stradling Evaluation of
Bikesafe Scotland Scottish Executive Social Research Edinburgh
2003
Osga 1979 published data only
Osga GS Ellingstad VS An evaluation of the effectiveness of the 1978
South Dakota motorcycle rider course program South Dakota USA
University of South Dakota 19791ndash89
Panou 2007 published data only
Panou M Bekiaris E A new concept on the integration of driving
simulators in driving training the train-all approach Road Safety
on Four Continents 14th International Conference Swedish
National Road and Transport Research Institute 2007
Peterson 1990 published data only
Peterson A Kadar K Status report 1980-1989 Minnesota rider
education statistics The Human Element International
Motorcycle Safety Conference Orlando Florida USA 199013-
113 to 13-129
Prem 1984 published data only
Prem H Good MC Motorcycle rider skills assessment ACT
Australia Office of Road Safety Department of Transport 1984
1ndash377 [ Report No CR 34 ]
Radin Umar 2006 published data only
Radin Umar RS Motorcycle safety programmes in Malaysia how
effective are they International Journal of Injury Control and Safety Promotion 200613(2)71ndash9
Rockwell 1990 published data only
Rockwell TH Kiger S Carnot M An evaluation of the Ohio
motorcyclist enrichment program Ohio Department of Highway
Safety 1990
Rothe 1987 published data only
Rothe JP A motorcycle riding course Motorcyclists Image and reality Canada Insurance Corporation of British Columbia 1987
59ndash83
Rowden 2007 published data only
Rowden P Watson B Haworth N What can riders tell us aboutmotorcycle training A view from the other side of the fence
Australasian Road Safety Research Policing and Education
Conference Melbourne Australia Melbourne 2007
RSC Victoria 1993 published data only
Road Safety Committee Inquiry into Motorcycle Safety in Victoria
Parliament of Victoria 1993
Russam 1979 published data only
Russam K Motorcycle training and licensing in Japan UK Transport
and Road Research Laboratory (TRRL) 1979 [ Report 916 ]
Saffron 1981 published data only
Saffron DG Driver and rider education training and licensing a
brief review Traffic Accident Research Unit Traffic Authority of
New South Wales NSW Australia Traffic Accident Research UnitTraffic Authority of New South Wales 1981
Schulz 1990 published data only
Schulz U Motorcycle accidents among learner riders Motorcycle
licensing Evaluation of the effectiveness of a Graduated Licensing Scheme Forshungshefte des Institut fur Zweiradsicherheit 1990
Sexton 2004 published data only (unpublished sought but not used)
Sexton B Baughan C Elliott M Maycock G The accident risk of
motorcyclists 2004TRL Report TRL 607(ISSN 09684107)58
Shephard 1986 published data only
Shepherd R Do Motorcycle Safety Foundation programs work
Driver 198619(9)16ndash8
Simpson 1990 published data onlySimpson HM Mayhew DR The promotion of motorcycle safety
training education and awareness Health Education Research
Theory and Practice 19905(2)257ndash64
Sokolov 1990 published data only
Sokolov O Safe motorcycle riding some aspects of motorcycle
riding in the USSR The Human Element International
Motorcycle Safety Conference Motorcycle Safety Foundation
19904ndash37 - 4-48
Steffens 1988 published data only
Steffens U Gawatz R Willmes G Effectiveness of mofa courses
[Wirksamkeit von Mofakursen] Unfall- und sicherheitsforschung
strasenverkehr 198867168
Sudlow 2003 published data only
Sudlow D Scoping study on motorcycle training Road Safety
Research Report London UK The British Institute of Traffic
Education Research Department for Transport 2003 Vol June
[ Report No 36]
Tasmania 1995 published data only
Mannering FL Grodsky LL Motorcycle rider training scheme
Traffic Education Research Centre Armidale Australia Driver
Training Australia Limited 1995 Vol April1ndash56
Tetard 1985 published data only
Tetard C Motorcycle ridersrsquo training to detect indices in an
intersection situation [Lrsquoentrainement a la prise drsquoinformation dans
lrsquoapprentissage de la conduite motocycliste] Le Travail Humain198548(2)147ndash59
27Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3050
Thackray 1980 published data only
Thackray RM Prescott JC Field test of the motorcycle ridercourse International Motorcycle Safety Conference Washington
DC USA Motorcycle Safety Foundation 1980195ndash7
Thompson 1994 published data only
Thompson M Evaluation of compulsory basic training for
motorcyclists Transport Research Laboratory Crowthorne
Berkshire UK 1994 [ Report 63]
Tobler 1990 published data only
Tobler U Individual driving instruction for motorcycle driving
learners the Swiss experiment of motorcycle driving school The
Human Element International Motorcycle Safety Conference
Orlando Florida USA Motorcycle Safety Foundation 1990
Transport 1989 published data only
Transport Communications Road Crash Statistics Australia Bureauof Transport and Communications Economics 198914ndash20
Tuan 2007 published data only
Tuan VA Shimizu T Towards development and evaluation of the
motorcycle drivers re-education program in Vietnam modeling of
motorcycle driverrsquos undesired behaviors 11th World Conference
on Transport Research 200725
Victor 1980 published data only
Victor DJ Traffic safety education for motorcyclists International
Motorcycle Safety Conference Washington DC USA Motorcycle
Safety Foundation 1980221ndash8
Waterer 1990 published data only
Waterer TH Compulsory training for motorcyclists from liberty
to compulsion The Human Element International Motorcycle
Safety Conference Orlando Florida USA Motorcycle Safety
Foundation 19904-49 to 4-57
Weller 1989 published data only
Weller RB Chandler EW Motorcycle safety and motorcycle
education past research and survey results Journal of Public Policy
amp Marketing 1989893ndash108
Wells 1981 published data only
Wells P Observations of some RACACU motorcycle training
schemes Transport and Road Research Laboratory Crowthorne
Berkshire UK 1981
Winn 1986 published data only
Winn G Two important trends in motorcycle safety regulations
rider education and conspicuity improvement American
Motorcyclist Association 1986147ndash52
Wisher 1988 published data only
Wisher G Reid J Motorcycle accidents in Norfolk a three-year study of motorcycle training and the Part 1 test Norfolk UK Road Safety
Division Highways Department Norfolk County Council 1988
Wood 1987 published data only
Wood HT Bowen R Evaluation of the revised learner permit scheme
July 1983 to December 1985 Victoria Australia Road Traffic
Authority 19871ndash27 [ ISBN 0 7306 0247 8]
Yamazaki 1990 published data only
Yamazaki S Nakanishi M Survey on effect of Honda rider
education The Human Element International Motorcycle Safety
Conference Orlando Florida USA 199012-167 to 12-187
References to ongoing studies
Burgess 2010 unpublished data only
Burgess C Broughton P Fylan F Stradling Interim evaluation of
the UKs National Ride program In L Dorn editor(s) Driver
behaviour and Training Vol IV Aldershot Ashgate Publishing
Limited 2010
Additional references
Anderson 1980
Anderson J Ford JL Peck RC Improved motorcyclist licensing and
testing project Volume I and II National Highway Traffic Safety
Administration US department of Transportation Washington
DC 1980
Bjornstig 1985
Bjornstig U Bylund P Lekander T Brorsson B Motorcycle fatalitiesin Sweden Acta Chirurgica Scandinavica 1985151(7)577ndash81
Clayton 1990
Clayton A Sudlow DE An evaluation of the effectiveness of a one-
day training program for learner motorcyclists The Human
Element International Motorcycle Safety Conference Orlando
Florida USA 199013-105 to 13-111
Cochrane Collaboration 2008
Higgins JP Green S Cochrane Handbook for Systematic Reviews of
Interventions 500 The Cochrane Collaboration 2008
Elliot 2003
Elliott MA Baughan CJT Broughton J Grayson G Knowles J et
alMotorcycle Safety a Scoping Study TRL [ Report No
0968ndash4107]Fleiss 2004
Fleiss JL Levin B Paik MC Statistical methods for rates and proportions 3rd Edition John Wiley amp Sons Inc 2004 [ ISBN
978ndash0ndash471ndash52629ndash2 ]
FORS 1999
FORS Federal Office of Road Safety Road risk for sober licensed
motorcyclists Vol Monograph 27 Federal Office of Road Safety
19991ndash2
Haworth 2005
Haworth N Mulvihill C Review of motorcycle licensing and training
Victoria Australia Monash University Accident Research Centre
20051ndash83 [ Report No 240]
Huwiler-Muumlntener 2002Huwiler-Muumlntener K Juumlni P Junker C Egger M Quality of
Reporting of Randomized Trials as a Measure of Methodologic
Quality JAMA 20022872801ndash4
Johnston 2008
Johnston P Brooks C Savage H Fatal and serious road crashes
involving motorcyclists Road Safety Research and Analysis Report
ATSB Monograph 20 Canberra Department of Infrastructure
Transport Regional Development and Local Government 0205
081ndash26
Juumlni 2001
Juumlni P Altman DG Egger M Systematic reviews in health care
Assessing the quality of controlled clinical trials BMJ 2001323
42ndash6
28Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3150
Kopits 2003
Kopits E Cropper M Traffic fatalities and economic growth TheWorld Bank Development Research Group Infrastructure and
Environment Policy research working paper 3035 2003
Leung 1983
Leung P Chau BTW Motorcycle accidents in Hong Kong pattern
of injury treatment causative factors and socio-economic
implications Injury 198315(1)23ndash9
Liu 2004
Liu B Ivers R Norton R Blows S Lo SK Helmets for preventing
injury in motorcycle riders Cochrane Database of Systematic Reviews 2004 Issue 2 [DOI 10100214651858CD004333pub3]
Mohan 2002
Mohan D Road safety in less-motorized environments future
concerns International Journal of Epidemiology 200231(3)527ndash32Mortimer 1982
Mortimer RG An evaluation of the motorcycle rider course
Proceedings of the 20th conference of the American Association for
Automotive Medicine Arlington Heights IL USA American
Association for Automotive Medicine 198277ndash96
Mullin 1997
Mullin BT Risk factors for motorcycle injury the role of age
gender experience training and alcohol PhD thesis Auckland
New Zealand University of Auckland 1997
Noordzij 2001
Noordzij PC Forke E Brendicke R Chinn BP Integration of needs
of moped and motorcycle riders into safety measures review and
statistical analysis in the framework of the European research
project PROMISING SWOV Institute for Road Safety Research
Leidschendam The Netherlands 2001 [ Report No Dndash2001ndash5]
Osga 1979
Osga G Ellingstad VS An evaluation of the effectiveness of the 1978 South Dakota Motorcycle Riders Course Program South Dakota
University of South Dakota Vermillion Human Factors Lab 1979
Oxford English Dictionary Online
httpoxforddictionariescom Oxford English Dictionary Online
Oxford University Press 2010
Peden 2004
Peden M Scurfield R Sleet D World Report on Road TrafficInjury Prevention Geneva World Health Organization Available
at httpwwwwhointviolence˙injury˙preventionpublications
road˙trafficworld˙reporten 2004Chapter 3
Radin Umar 1995
Radin Umar R Macakay G Hills B Preliminary analysis of
motorcycle accidents Short-term impacts of the running
headlights campaign and regulations in Malaysia J Traffic Med 199523(1)17ndash27
Tsai 1995
Tsai YJ Wang JD et alCase-control study of the effectiveness of
different types of helmets for the prevention of head injuries among
motorcycle riders in Taipei Taiwan American Journal of
Epidemiology 1995142(9)974ndash81
Vis 1995
Vis AA In-depth study of the hazards of motorcycling [De
onveiligheid van motorrijden nader bekeken] SWOV Institute for
Road Safety Research Leidschendam the Netherlands 1995 [
Report No Rndash95ndash69]
Weaver 1990
Weaver EA Motorcycle safety education in the United States
Motorcycle Safety Foundation Orlando Florida USA 1990
Winn 1987
Winn G Two important trends in motorcycle safety regulations
rider education and conspicuity improvement No more
information available Detroit MI US Society of Automotive
Engineers 1987147ndash52
Yuan 2000
Yuan W The effectiveness of the rsquoride-brightrsquo legislation for
motorcycles in Singapore Accident Analysis amp Prevention 200032
(4)559ndash63
Zambon 2006
Zambon F Hasselberg M Factors affecting the severity of injuries
among young motorcyclists a Swedish nationwide cohort study
Traffic Injury Prevention 20067(2)143ndash9lowast Indicates the major publication for the study
29Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3250
C H A R A C T E R I S T I C S O F S T U D I E S
Characteristics of included studies [ordered by study ID]
Adams 1985
Methods Non-randomised trial
Participants Initial sample size
Total of 162 Australia Post telegram delivery riders
Not clear how many in intervention and control groups
Analysis sample size
Intervention group 63Control group 35
Total 98 (61)
Interventions No formal training vs voluntary rider training course of 12 hours over 2 consecutive days (theory 4 hours practice
8 hours)
Intervention integrity
No information given
Outcomes Australia Post recorded crashes
Length of follow up 1 year
Notes Study conducted in NSW Australia
Anderson 1980
Methods Randomised trial
Participants Initial sample size
Total 41188 motorcycle licence applicants
Analysis sample size
Intervention group with skills training 13160
Intervention group without skills training 12634
Control group 15080
Total 40874 (99)
Interventions Standard licensing skill test vs the Motorcycle Operators Skill Test (MOST) with or without a training component
MOST is not to be confused with rider training because it is a licensing test The training component was for
participants who failed the skill test on the first attempt was of 3 hours duration and consisted of instructions about
the skills needed to pass the skill test The training consisted of 30 min theory and 2 frac12 hours practice on the
motorcycle
Intervention integrity
33 in MOST with skills training 31 in MOST without skills training and 43 in control group obtained their
motorcycle licence in the end
Outcomes Police recorded crashes and offences
Length of follow up 3 years
30Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3350
Anderson 1980 (Continued)
Notes Study conducted in California USA
Billheimer 1998
Methods Cohort study
Participants Initial sample size
Intervention group Unclear numbers of CMSP graduates and ERC graduates
Control group Over 16000 untrained motorcyclists
Analysis sample size
1139 matched CMSP-pairs and 1182 ERC-pairs
Interventions No formal training vs the mandatory (for those younger than 18 years) pre-licence 16-hours basic California
Motorcyclist Safety Program (CMSP) and the 8-hour Experienced Rider Course (ERC)
Intervention integrity
No information given
Outcomes Police recorded crashes and offences
Length of follow up 2 years
Notes Study conducted in California USA
Cooper 1988
Methods Cohort study
Participants Initial sample size
Intervention group 418 graduates and 57 who failed
Control group 402
Analysis sample size
Intervention group unclear
Control group unclear
Total 863 (984)
Interventions No formaltrainingvs non-mandatory pre-licence rider trainingThe content of thetrainingwas notfurther explainedIntervention integrity
Participants had all completed the rider training
Outcomes Police recorded crashes and offences
Length of follow up 7 years
Notes Study conducted in British Columbia Canada
31Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3450
Davis 1997
Methods Retrospective cohort study
Participants Initial sample size
Intervention group 9320 graduates
Control group unclear
Analysis sample size
Intervention group 9320 graduates
Control group unclear
Interventions No formal training vs the non-mandatory Connecticut Motorcycle Operator Training Program The content of
training was not further explained
Intervention integrity Participants had all completed the rider training
Outcomes Police recorded crashes severity of crashes
Length of follow up 1 year
Notes Study conducted in Connecticut USA
Hall 1988
Methods Randomised trial
Participants Initial sample sizeIntervention group A 6248
Intervention group B 6616
Intervention group C 6996
Control group 6604
Total 26464 motorcycle licence applicants
Analysis sample size
Intervention group A 6012 (96)
Intervention group B 6428 (97)
Intervention group C 6778 (97)
Control group 6502 (98)
Interventions Standard licensing skill test vs the Motorcycle Operator Skill Test (MOST II) without training (A) MOST II with
3 hours training (B) or MOST II with 20 hours training (C) The 3 hours training comprised instructions aboutthe skills needed to pass the skill test (30 min theory and 2 frac12 hours practice on the motorcycle) The 20 hours
training consisted of the Motorcycle Safety Foundationrsquos Motorcycle Rider Course (MRC) comprising of 8 hours
theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-road practice
Intervention integrity
25 and 26 obtained their motorcycle licence in the lsquoMOST 3-hour trainingrsquo and lsquoMOST 20-hour trainingrsquo
intervention arm respectively In the lsquoMOST testrsquo intervention arm 29 obtained their licence and in the control
group 35
Outcomes Police recorded crashes and offences
Length of follow up 3 years
32Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3550
Hall 1988 (Continued)
Notes Study conducted in New York USA
Haworth 1997
Methods Population based case-control study
Participants Cases Motorcycle riders or pillion riders involved in motorcycle crash (on road) Recruited from hospital and police
records Not clear whether all cases (especially fatalities) were included
Community controls Multiple matched sample of motorcycle rider or pillion who passed the crash site at the same
time of day and day of week that the crash occurred
Initial sample size
Cases 336
Controls 1195
Analysis sample size
Cases 234 (70)
Controls 561 (47)
Interventions Basic handling skills course beginner intermediate or advanced course measured through self-report not validated
Outcomes Injuries with an Injury Severity Score of 5 or greater or admission for at least four hours and fatalities from a
motorcycle crash (from hospital and police records)
Length of follow up two years or more prior to the interview
Notes Study conducted in Victoria Australia
Hurt 1981
Methods Population based case-control study
Participants Cases Motorcycle riders who had a motorcycle crash that was reported to police (traffic accident reports) Not all
cases included
Community controls Multiple matched sample of motorcycle riders who passed the crash site at the same time of
day and day of week as the crash occurred but 2 years later
Initial sample size
Cases 4500 notifications of crashes to police (only 50 received in time research team able to respond to 1126 for
unreported reasons data collection completed for 900)
Controls Number approached unreported 2310 consented and interviewed
Analysis sample size
Cases 808 (808900 data on rider training) - 18 of initial sample
Controls 670 (6702310 data on rider training) - lt 29 of initial sample
Interventions Training experience self taught friends-family school-club formal-professional others measured through self-
report validated
Outcomes Injuries and fatalities from a motorcycle crash (from hospital records and coronerrsquos office)
33Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3650
Hurt 1981 (Continued)
Notes Study conducted in Los Angeles USA
Jonah 1982
Methods Cohort study
Participants Initial sample size
Intervention group
2310 licensed MTP graduates who are active riders
Control group
2029 licensed motorcyclists who are active riders
Non-licensed and non-active riders excluded
Analysis sample size
Intervention group
811 interviews (35) 533 driver records (23)
Control group 1080 interviews (71) 905 driver records (45)
Interventions No training vs the non-mandatory Canada Safety Council Motorcycle Training Program (MTP) comprising of 4
hours theory and 16 hours practise (6 hrs on-road) over 2 weekends
Intervention integrity
Participants had all successfully completed the MTP
Outcomes Self-reported and police reported crashes and offencesLength of follow up 4 years
Notes Study conducted in Ontario Canada
Kloeden 1994
Methods Cohort study
Participants Initial sample size
Intervention group 2533 licence applicants
Control group Unclear
Sample of riders obtaining probationary licenceIntervention group 1625 (64)
Control group 5015 (unclear)
Non-active riders excluded from control group (~66)
Analysis sample size
Intervention group 631 (39)
Control group 318 (6)
Interventions No formal training vs the mandatory pre-licence Ridersafe course consisting of two levelsthe first is a training course
in the basics of motorcycle riding where a pass entitles the rider to obtain a learnerrsquos permit The second is a skill
test where a pass entitles the rider to obtain a probationary motorcycle licence The training course consists of two 4
hour training sessions on separate days involving instruction videos and actual riding on motorcycles The skill test
consists of a 4 hour preparation course of 74 minutes of lecturediscussion 38 minutes of video and 75 minutes of
34Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3750
Kloeden 1994 (Continued)
on-course riding which is the testing phase
Intervention integrity
Unclear what percentage in the Ridersafe group and the control group obtained their probationary licence halfway
and their full motorcycle licence in the end
Outcomes Police reported crashes
Length of follow up 1 year
Notes Study conducted in SA Australia
Kraus 1975
Methods Population based case-control study
Participants Cases All motorcycle riders with a confirmed medically treated motorcycle injury (including fatalities) from all
hospitals in study base and peripheral area hospitals and all deaths due to motorcycle crashes in the study base All
serious injuries (leading to hospitalisation) and fatalities likely to be included
Community controls A random sample of registered motorcycle owners in the same temporary and geographical
study base as the cases arose from
Initial sample size Cases 1273
Controls 738
Analysis sample size Cases 741 (58)
Controls 434 (59)
Interventions Motorcycle rider training (not specified) measured through self-report
Outcomes Injuries leading to hospitalisation fatalities
Length of follow up not specified
Notes Study conducted in Sacramento county USA
Leung 1987
Methods Non-randomised trial
Participants Initial sample sizeIntervention group unknown number of MRC graduates
Control group unknown number of licensed motorcyclists
Analysis sample size
Intervention group 2941
Control group 43094
Interventions No formal training vs non-mandatory Wisconsin motorcycle rider course a 22 hours off-street programme that
teaches the basic rules of safe motorcycling
Intervention integrity
50 in the MRC group and 100 in the control group obtained their motorcycle licence in the end
35Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3850
Leung 1987 (Continued)
Outcomes Police recorded crashes and offences
Length of follow up up to 3 years
Notes Study conducted in Wisconsin USA
MAIDS 2004
Methods Population based case-control study
Participants Cases Motorcycle riders who had a motorcycle crash Unclear whether all cases included
Community controls Motorcycle riders identified at petrol stations Unclear whether matched to cases
Initial sample size
Cases not reported
Controls not reported
Analysis sample size
Cases 921 (Italy 200 Netherlands 200 Spain 121 Germany 250 France 150)
Controls 923 (Italy 200 Netherlands 200 Spain123 Germany 250 France 150)
Interventions Pre-license rider training or additional training (non-compulsory)
Outcomes Crashes
Notes Study in five sampling areas located in five European countries (France Germany Netherlands Spain and Italy)
McDavid 1989
Methods Cohort study
Participants Sample size after first match
Intervention group 346 male training graduates
Control group 346 male motorcycle riders licensed in same year
Sample size after second match
Intervention group 139 (40)
Control group 139 (40)
Analysis sample size
Intervention group 139 (40)
Control group 139 (40)
Interventions No formal training vs the non-mandatory pre-licence British Columbia Safety Councilrsquos 37-hour motorcycle safety
programme
Intervention integrity
Participants had all completed and passed the training
Outcomes Police reported crashes injurious crashes
Length of follow up 1-6 years
36Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3950
McDavid 1989 (Continued)
Notes Study conducted in British Columbia Canada
Mortimer 1984
Methods Cohort study
Participants Initial sample size
Intervention group 608 MRC graduates
Control group 360 motorcycle riders
Response rates
Intervention group 360 (59)
Control group 324 (90)
Non-active riders excluded
Analysis sample size
Intervention group 213 (35)
Control group 303 (84)
Interventions No formal training vs the non-mandatory Motorcycle Safety Foundationrsquos Motorcycle Rider Course comprising of
8 hours theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-road practice
Intervention integrity
Participants had all completed and passed the MRC
Outcomes Self-reported crashes and offencesLength of follow up 1 year
Notes Study conducted in Illinois USA
Mortimer 1988
Methods Cohort study
Participants Initial sample size
Intervention group 2307 MRC graduates
Control group 556 active licensed motorcycle riders
Response ratesIntervention group 1298 (56)
Control group 500 (90)
Non-active riders excluded
Analysis sample size
Intervention group 913 (40)
Control group 500 (90)
Interventions No formal training vs the non-mandatory Motorcycle Safety Foundationrsquos Motorcycle Rider Course comprising of
8 hours theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-road practice
Intervention integrity
Participants had all completed and passed the MRC
37Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4050
Mortimer 1988 (Continued)
Outcomes Self reported crashes and offences
Length of follow up 1 year
Notes Study conducted in Illinois USA
Osga 1980
Methods Cohort study
Participants Initial sample sizeIntervention group 1163 MSF MRC participants invited 818 (74) responded
Control group 329 motorcycle riders responded to recruitment strategies
Non-active riders excluded from analysis
Analysis sample size
Intervention group 595 (73)
Control group 329 (100)
Interventions No formal training vs the non-mandatory Motorcycle Safety Foundationrsquos Motorcycle Rider Course (MRC) com-
prising of 8 hours theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-road
practice
Intervention integrity
Participants had all successfully completed the MRC
Outcomes Self-reported crashes police recorded crash reports
Length of follow up 6 months - 3 12 years
Notes Study conducted in South Dakota USA
Perrino 2002
Methods Cohort study
Participants Initial sample size
Intervention group 2144 (unclear) MRC graduates (motorcycle licence applicants)
Control group 2601 (unclear) regular licensing test graduates (motorcycle licence applicants)Analysis sample size
Intervention group 1070 (unclear)
Control group 867 (unclear)
Interventions No formal training vs the mandatory (for those younger than 18 years) pre-licence Maryland Motorcycle Rider
Course (MRC) consisting of 8 hours theory and 12 hours practice
Intervention integrity
No information given
Outcomes Self-reported and police recorded crashes and offences
Length of follow up 1 year
38Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4150
Perrino 2002 (Continued)
Notes Study conducted in Maryland USA
Raymond 1979
Methods Cohort study
Participants Initial sample size
Intervention group
258 volunteers mainly (but not all) novice riders
Control group
1104 provisional licence applicants
Analysis sample size
Intervention group 190 (74)
Control group 627 (57)
Interventions No formal training vs the non-mandatory RACACU (Royal Automobile ClubAuto-Cycle Union) training scheme
The training program consisted of either 24 one-hour sessions or 13 two-hour sessions Twelve hours of the training
focused on theory about basic operational procedures traffic laws and highway code The practical session focused
on basic operational skills riding on a training area and riding in traffic with an instructor A proficiency test was
administered at the end of the course
Intervention integrity
No information given
Outcomes Self-reported crashes (including minor non-injury non-damage and off-road crashes)
Length of follow up 1 year
Notes Study conducted in the UK
Satten 1980
Methods Cohort study
Participants Initial sample size
Intervention group interviewers kept interviewing until they had 100 riders selected from a population of 891
enrollees in the MSF MRC course (unknown how many were asked and how many refused to participate)Control group unknown number of registered motorcycle riders
Non-active riders excluded from analysis
Analysis sample size
Intervention group 71
Control group 100
Interventions No formal training vs the non-mandatory Motorcycle Safety Foundationrsquos (MSF) Motorcycle Rider Course (MRC)
comprising of 8 hours theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-
road practice
Intervention integrity
Participants had all successfully completed the MRC
39Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4250
Satten 1980 (Continued)
Outcomes Self-reported crashes injuries and offences
Length of follow up 3 years
Notes Study conducted in Illinois USA
Savolainen 2007
Methods Cohort study
Participants Initial sample size
Intervention group 4000 riders that had conducted the MSF MRC course (passed or failed)
Control group 4000 motorcycle owners or permit or endorsement holders from the Indiana Bureau of Motor
Vehicles database
Analysis sample size
Intervention group 739 (18)
Control group 588 (15)
Interventions Intervention integrity
No information given
Outcomes Probability of crash involvement
Length of follow up up to 5 years
Notes Study conducted in Indiana USA
Swaddiwudhipong 1998
Methods Randomised trial
Participants Initial sample size
Intervention group
all 16 villages of 3 sub-districts
(participants were practising motorcyclists licensed and unlicensed)
Control group
18 villages of 3 other sub-districtsAnalysis sample size
Intervention group 3 districts (100)
Control group 3 districts (100)
Interventions No rider education vs non-mandatory community based rider education consisting of theory on motorcycle injury
risk helmet use traffic laws vehicle regulations and traffic signs and a written and skill test leading to a licence
Intervention integrity
No information given
Outcomes Injuries fatalities
Length of follow up 2 years
40Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4350
Swaddiwudhipong 1998 (Continued)
Notes Study conducted in Tak Province rural Thailand
Waller 1992
Methods Cohort study
Participants Initial sample size
Intervention group 349 riders who succesfully completed the Department of Public Safety motorcycle operator
training in 1988
Control group 349 motorcyclists who never completed the DPS motorcycle operator training
Analysis sample size
Intervention group 349
Control group 349
Interventions No DPS motorcycle operator training vs non-mandatory DPS motorcycle operator training The DPS training is
the Motorcycle Safety Foundationrsquos Motorcycle Rider Course comprising eight hours of theory (lectures discussions
films and multiple choice tests) a textbook and 12 hours off-road practice
Intervention integrity
All 349 riders succesfully completed the rider training
Outcomes Police recorded crashes
Length of follow up 3 years
Notes Study conducted in
Characteristics of excluded studies [ordered by study ID]
Study Reason for exclusion
Anderson 1980a Study type review and process evaluation of a rider course
Aupetit 2008 Outcome proposal of a rider training curriculum that combines traditional training and a motorcycle riding
simulator
Awane 1999 Study type review on use of simulators in rider and driver training
Baldi 2005 Intervention not rider training but intervention integrity of rider training availabilty of rider education
implementation of graduated licensing system etc on a state-wide level
Batchler 1988 Study type description of a training course
Blanchard 2006 Study type not a study assessing the association between rider training and outcomes
Bonnett 2005 Outcome qualitative data regarding satisfaction with the training programme
41Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4450
(Continued)
Braver 2007 Outcome licensure rates
Clayton 1980 Outcome knowledge skills attitude
Council 1978 Outcome crash reduction in ANY vehicle not motorcycles exclusively
Crowther 2005 Study type qualitative study
Daniello 2009 Study type review of the effectiveness of motorcycle training and licensing
Dewanti 2007 Intervention emotional characteristics behaviour (ie not motocycle rider training)
Du Plooy 1990 Study type review of types of rider training courses in South Africa
Engel 1989 Study type uncontrolled before-after study
Goldenbeld 2002 Outcome skills
Goldenbeld 2004 Outcome knowledge and skills
Haworth 2000 Intervention balance between attitudinal andvehicle control skills acrosscurriculla of motorcycle rider training
courses
Haworth 2005 Study type review
Haworth 2006 Study type cross-sectional survey
Hill 1983 Study type review and description of training system
Imatake 1980 Study type course descriptionreview
Ingebritsen 1990 Intervention risk factors for motorcycle crashes
Jain 2009 Study type cross-sectional descriptive study of motorcycle crashes over a 5 year period
Jonah 1979 Study type cross-sectional survey Outcome scores on motorcycle operator skill test
Jonah 1981 Intervention evaluation of a motorcycle rider skill test not motorcycle rider training
Kadar 1990 Study type review of components of rider training courses
Kelsey 1986 Intervention evaluation of a motorcycle rider skill test not motorcycle rider training
Koch 1980 Study type reviewcourse description
Koch 1990 Study type description of course content
42Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4550
(Continued)
Labbett 2005 Study type essay about possible explanations for motorcycle deaths
Law 2005 Intervention targeted motorcycle safety programme not specifically rider training
Lowes 1990 Study type review and descriptions of rider training
Mahon 1976 Study type review of motorcycle rider training methods
Maxwell 1985 Outcomesrecommendations forprogramme operatorsand otherson howthe motorcycle trainingprogramme
works and how it can be improved
Mayhew 1996 Study type review
McKnight 1987 Study type evaluation of methodological and administrative aspects of a motorcycle operator skill test
McPherson 1976 Outcome knowledge and skills
Mohan 2004 Study type review of evidence of interventions to reduce road traffic injuries
Morgan 1980 Study type reviewcourse description
Motoki 1990 Study type proposal and development of rider training curriculum
Nairn 1992 Study type literature review
Newland 1980 Study type description of development of motorcycle safety course
Newland 1982 Study type descriptionreview of a rider training course
Newland 1990 Study type uncontrolled BeforeAfter study
Ormston 2003 Outcome changes in attitude and self reported behaviour
Osga 1979 Outcome knowledge attitudes and skills
Panou 2007 Intervention new simulation tool to teach driverrider
Peterson 1990 Study factor demographic details of course participants completing the course vs those not completing the
course
Prem 1984 Study type experimental
Study factor examines association between skills and performance in tests
Radin Umar 2006 Intervention motorcycle safety programme not specifically rider training
Rockwell 1990 Study type methodologically flawed study type
43Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4650
(Continued)
Rothe 1987 Study type qualitative study of rider training in general
Rowden 2007 Study type qualitative study
RSC Victoria 1993 Study type review of motorcycle rider training and education in Victoria
Russam 1979 Study type review of training programmes in Japan
Saffron 1981 Study type literature review
Schulz 1990 Study factor predictors of motorcycle crashes
Sexton 2004 Study type cross-sectional analytic analysis
Shephard 1986 Study factor reviewdescription of programme
Simpson 1990 Study type review of motorcycle rider trainingeducation
Sokolov 1990 Study factor review of motorcycle riding courses in USSR
Steffens 1988 Outcome self-reported behaviour experiences and researcher observed riding behaviour
Sudlow 2003 Study type review of motorcycle training in the UK
Tasmania 1995 Study type cross-sectional descriptiveprocess evaluation
Tetard 1985 Study type experimental
Outcome detection of indices
Thackray 1980 Outcome instructional effectiveness user acceptance administrative feasibility
Thompson 1994 Study type cross-sectional survey
Outcome attitudes and knowledge
Tobler 1990 Study type review of course curricula
Transport 1989 Intervention syllabus for rider training
Tuan 2007 Study type cross-sectional study (survey) which does not assess the effects of an intervention but examines
motocyclistsrsquo behaviour
Victor 1980 Study type review
Waterer 1990 Study type review of UK rider training issues
Weller 1989 Study type review of studies evaluating motorcycle safety and education measures
44Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4750
(Continued)
Wells 1981 Study type description of different methods of training
Winn 1986 Study type review of rsquohistoryrsquo of rider training
Wisher 1988 Study type cross-sectional survey
Wood 1987 Study type Interrupted Time Series but did not meet the criteria for an ITS as published by the Cochrane
Effective Practice and Organisation of Care Group
Yamazaki 1990 Study type cross-sectional survey
Outcome riding skills skills retention and attitude
Characteristics of ongoing studies [ordered by study ID]
Burgess 2010
Trial name or title UKs National Ride program
Methods Cohort study
Participants Motorcycle offenders
Interventions Police led diversion program
Outcomes Self reported attitudinal and behavioural measures
Starting date
Contact information
Notes Expected to be completed in early 2010
45Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4850
D A T A A N D A N A L Y S E S
This review has no analyses
A P P E N D I C E S
Appendix 1 Search strategy
We searched the following electronic databases
bull Cochrane Injuries Group Specialised Register (searched 22 Sept 2008)
bull CENTRAL (The Cochrane Library 2008 Issue 3)
bull TRANSPORT 1988 to 200706 (latest date available)
bull ATRI Australian Transport Index (to Sept 2008)
bull MEDLINE (Ovid SP) 1950 to Sept 2008
bull EMBASE (Ovid SP) 1980 to Sept 2008
bull CINAHL (EBSCO) 1982 to Sept 2008
bull WHOLIS (World Health Organization Library Information System) (to Sept 2008)
bull PsycINFO (Ovid SP) 1806 to Sept 2008
bull LILACS (BIREME) (Latin American and Caribbean Health Sciences) (1982 to Sept 2008)
bull ISI Web of Science Social Sciences Citation Index (SSCI) 1970 to Sept 2008
bull ERIC (searched 25 Sept 2008)
bull ZETOC (searched 25 Sept 2008)
bull SIGLE (searched 25 Sept 2008)
All of the databases were searched using the following text-word search strategy The strategy was adapted using either rsquorsquo or rsquo$rsquo as atruncation symbol
1 motor-cycl or motorcycl or motor-bik or motorbik or motor-scooter or motorscooter or scooter or moped or motocycl
or motocicl or moto
2 train or educat or instruct or program or course or qualif or learn or improv
3 1 and 2
H I S T O R Y
Protocol first published Issue 2 2005
Review first published Issue 10 2010
46Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4950
C O N T R I B U T I O N S O F A U T H O R S
Katina Kardamanidis was the first author of the protocol conducted the literature search extracted full studies assessed the studies
according to inclusion criteria assessed quality of the studies compared findings with the second author discussed disagreements with
the third author wrote the drafts and the final review
Rebecca Ivers was the second author of the protocol assessed the studies according to inclusion criteria assessed quality of the studies
compared findings with the first author edited the final draft and provided overall guidance for the process
Alexandra Martiniuk assessed quality of the studies compared findings with the first author discussed disagreements with the third
author and helped write the first draft and the final review
Mark Stevenson contributed to the protocol reviewed methods of studies and contributed to the final draft
Katrina Thistlethwaite conducted a second literature search extracted full studies and contributed to the final draft
D E C L A R A T I O N S O F I N T E R E S T
None known
S O U R C E S O F S U P P O R T
Internal sources
bull No sources of support supplied
External sources
bull National Health and Medical Research Council of Australia Australia
Salary support for Martiniuk Ivers and Stevenson
bull University of Sydney Sesqui Research Grant Australia
Direct research costs and salary support for Kardamanidis
D I F F E R E N C E S B E T W E E N P R O T O C O L A N D R E V I E W
There are no differences between the protocol and the review
I N D E X T E R M S
Medical Subject Headings (MeSH)
lowastMotorcycles Accident Prevention [lowastmethods] Accidents Traffic [lowastprevention amp control] Licensure Program Evaluation Wounds
and Injuries [mortality lowastprevention amp control]
47Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 5050
MeSH check words
Humans
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2350
Three studies assessed the effect of non-mandatory pre-licence
training The Cooper 1988 cohort study found no difference incrash or offence rates between the trained and untrained groups
We cannot however be confident about this result as the study
sufferedfrom selectionbiaswhichcouldhave biasedthe measure of
effect toward the null The RCT conducted by Swaddiwudhipong
1998 examined injury and death rates and found a statistically
significant lower injury rate in the intervention villages after one
year of follow up but no longer at two year follow up There
was no statistically significant difference in death rates between
intervention and control villages Our confidence in the results of
this study are hindered by the inadequacy of the randomisation
process that led to significant differences between the intervention
and control districts on two important variables The intervention
districts were more likely than the control districts to have a validlicence andto wear a helmetSincethese differences were nottaken
into account in later analyses they could explain the differences
observed in injury rates between districts Furthermore with only
six districts randomised the sample size was probably too small
to detect a difference between intervention and control groups
had there been one The McDavid 1989 cohort study found a
lower percentage of injuries in the intervention group although
a statistical test was not conducted McDavid 1989 drew their
intervention and control groups from different populations but
did adjust for confounders although they only analysed 40 of
the initial sample (139 trained and 139 untrained) Because of
these substantial weaknesses we are unable to draw conclusions
from this studyAn overall effect of pre-licence training on the reductionof crashes
injuries deaths and offences was not found although one should
be aware that the research of this issue to date has had significant
flaws However a mandatory skill test (such as the MOST) in
combination with and even without pre-licence training as part
of a licensing process seems to reduce the number of individuals
completing the licensing procedure and therefore indirectly re-
duces crash rates through a reduction in exposure to motorcycle
riding Making licensure a more difficult process may therefore be
seen as advantageous if it delays licensure and therefore reduces
exposure It is possible that the group who chose not to continue
riding a motorcycle would not have been the group to experience a
crash or commit an offence if they had continued to ride anyhowfor example because they are more safety conscious and this was
the reason for their giving up riding in the first place If this is the
case deterring individuals from completing their licence and rid-
ing may not necessarily decrease overall crashes deaths or offence
rates In addition other potentially adverse effects of complicat-
ing a licensing procedure need to be considered such as increases
in unlicensed riding In Thailand non-mandatory rider training
offered as a community intervention may have decreased injury
rates but this needs to be confirmed through further research
Non-mandatory training
The majority of the studies assessed non-mandatory training
(training independent of any licensing procedure) and the results
were varied The two non-randomised trials assessing crash rates found no statistically significant difference in crash rates between
the intervention and control groups (Adams 1985 Leung 1987)
Of the nine cohort studies Satten 1980 found a lower percentage
of crashesin the intervention group although it is unclear whether
this was a statistically significant result Davis 1997 also found a
lower percentage of crashesin theinterventiongroup although it is
unclear with which control group the intervention group wascom-
pared Raymond 1979 Osga 1980 Waller 1992 and Savolainen
2007 found a statistically significant higher crash rate in the in-
tervention group and the remaining cohort studies (Jonah 1982
Mortimer 1984 Mortimer 1988 Billheimer 1998) found no sta-
tistically significant difference in crash rates between intervention
and control groups No conclusions can be drawn from these stud-
ies with confidence The Satten 1980 and Davis 1997 studies did
not adjust for any confounding factors even though important
differences existed between the intervention and control groups
Although Raymond 1979 and Waller 1992 adjusted for a num-
ber of confounders at baseline (Waller 1992) and in the analy-
sis (Raymond 1979 Waller 1992) the intervention and control
groups were drawn from such different populations that it is pos-
sible that unmeasured confounders were responsible for the sta-
tistically significant increased risk of crashes observed in the inter-
vention group The observed difference could also be attributed totheobserved differential loss to followup (Raymond 1979) which
could bias the measure of effect towards or away from the null
The Osga 1980 study found a protective effect of rider training
in the univariate analysis but statistically significant higher crash
rates in the intervention group when accounting for riding expo-
sure Sex (a significantly higher proportion of females in the in-
tervention group) and other potentially confounding factors that
could explain thedifferencebetween the groups were howeverun-
accounted for Finally Savolainen 2007 did control for potential
confounding variables yet it is not clear whether variables were
controlled for simultaneously or one at a time One case-control
study (MAIDS 2004) found no statistically significant reduction
in the odds of experiencing a crash between trained and untrainedriders but the quality of this study was poor mainly because the
analysis did not account for any confounding factors
One cohort study (Satten 1980) assessed the effect of non-manda-
tory rider training on injurious crashes and found a lower percent-
age of injuries in the intervention group although a statistical test
was not conducted As mentioned above the Satten 1980 studyrsquos
validity is reduced due to its lack of adjustment for confounding
factors
The only non-randomised trial that assessed offence rates (Leung
1987) found a statistically significant lower offence rate in the in-
tervention group and of the five cohort studies that assessed of-
fence rates (Satten 1980 Jonah 1982 Mortimer 1984 Mortimer
21Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2450
1988) one study (Jonah 1982) also found a statistically significant
lower offence rate in the intervention group However we have lit-tle confidence in the results from Leung 1987 since potential con-
founding factors were not appropriately addressed Jonah 1982
did conduct a multivariate analysis but was hindered by an initial
low response rate The other four cohort studies did control for
several potential confounders and found no difference in offence
rates following non-mandatory post-licence rider training
The non-randomised trials and cohort studies evaluating non-
mandatory rider training were at high risk of substantial selection
bias due to the non-random allocation of participants as well
as low participation rates and attrition bias and could therefore
not provide conclusive evidence with regard to the effect of the
training It should be noted that despite the biases and other issues
with this group of studies most didnot detectmajor effects relatedto rider training in either direction This may suggest that any
potential impact of rider training will be modest
Studies examining unspecified modes of ridertraining
Three case-control studies assessed whether injured or deceased
motorcyclists were more or less likely to have had rider training
compared to control motorcyclists All had conflicting results
Hurt1981 foundthe odds of havingan injuriouscrash waslower in
those who had received rider training whereas Kraus 1975 found
that the odds of having an injurious crash were higher in youngerand older trained riders Haworth 1997 who compared advanced
versus beginner training found no effect on motorcycle injuries
or deaths although they did find that those who had completed
advanced training had reduced chances of crashing compared to
those who had done a beginnersrsquo training course Similar issues
with uncontrolled confounding and biases of selection as described
above arose with the three case-control studies Hurt 1981 suffers
from inadequate control of confounding via poor matching and
both the Kraus 1975 and the Haworth 1997 findings were only
controlled for limited factors by means of stratified analysis It is
likely that these limitations are the reason for the contradictory
directions of effect observed for the three case-control studies
Methodological weaknesses of the studies
In summary most studies investigating the effectiveness of mo-
torcycle rider training on reducing crashes injuries and death suf-
fered from serious methodological weaknesses
The majority of studies were non-randomised and were there-
fore limited mainly by selection bias Although the approaches of
matching stratification or regression analyses can minimise bias
(Fleiss 2004) they were poorly employed in the reviewed studies
The inadequate adjustment for confounders in particular adds
to our lack of confidence in the studiesrsquo results since it has been
demonstrated that individuals who choose to take motorcycle rider
training differ from those not taking rider training For instance
trained riders often differ from controls in age sex and experi-ence usually ride less and may have a more positive attitude to
safety as indicated by the reported more frequent wearing of seat
belts (Mortimer 1982)
The two large randomised trials reported that a large proportion
of their study group were never exposed to the risk of offences or
crashes because they never attained licensure This made it impos-
sible to assess whether a reduction in crashes in the intervention
group was a result of training or due to delayed or never attained
licensure
In some studies the sample sizes were too small to detect any sta-
tistically significant differences Follow-up time after training dif-
fered between studies and was an impediment to observing any
real effects of training if they existed As riding experience ac-cumulates over time any differences in outcomes between those
trained and untrained would diminish in studies with lengthy fol-
low-up times This dilution of any possible effect of rider training
may have also occurred in studies that mixed both novice and ex-
perienced riders in the study groups since it is hypothesised that
novice riders serve to benefit most from rider training This would
increase the difficulty in finding any effect of rider training par-
ticularly when sample sizes are small
Detection bias may also account for some of the differences ob-
served between trained and untrained riders Detection bias would
have occurred if the self-report hospital driver or police records
used to record the outcomes of crash injury death or offences
differed between the trained and untrained riders Research hasdemonstrated that not all crashes are reported to police for ex-
ample small crashes or off-road crashes (Kraus 1975) even serious
injury crashes have been found to be underreported (Anderson
1980) However this would only introduce detection bias if the
reporting of outcome differs for trained and untrained riders This
does appear possible in that responsible riders may be more likely
to enroll in a rider training course and also may be more likely
to report crashes to police than irresponsible riders meaning that
the number of crashes from riskier riders could be underrepre-
sented on records An analysis to investigate the effectiveness of
rider training may then show that rider training is not effective in
reducing crashes
Rather than using reported outcomes some studies used self-re-ported survey data although this technique suffers from non-
response biases In fact two studies reported differences in de-
mographic characteristics in response rates to questionnaires
(Anderson 1980 Hall 1988) Self-reported data also introduce
the risk of recall bias survival bias and social desirability response
bias Recall bias may have been equally distributed among trained
and untrained riders howeverit is hypothesised that trained riders
may have felt it was lsquoexpectedrsquo to report good driving histories
hence leading the rider training to appear effective
22Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2550
Limitations of this review
Limitations of this review were that we found very few recentstudies (the two most recent ones published in 2002 and 2007)
and that we found very few studies conducted in low-and middle
income countries even though our search strategy was set up to
identify such studies Anotherlimitation wasthat forthe reviewed
studies we could not assess all outcomes of rider training that are
potentially useful Some of the included studies assessed outcomes
such as use of safety clothing including helmets Such outcomes
are valuable since the use of motorcycle helmets has been demon-
strated to be effective in preventing head injuries (Liu 2004) De-
spite this the use of safety clothing was not included as an out-
come in this review because the studies measuring this outcome
were limited by poor data quality due to very low response rates
Another outcome of interest in assessing the effectiveness of rider
training is lsquotype of injuryrsquo As Haworth 2005 state in their review
of motorcycle licensing and training rider training could be con-
sidered effective if injuries become less severe because motorcycle
riders have learned how to avoid more serious injuries through
rider training The studies included in this review did not specify
type of injury and so this outcome could not be assessed Last
changes in knowledge and attitudes following rider training are
other potentially useful outcomes We have excluded such studies
(Osga 1979 Clayton 1990 Goldenbeld 2004) as these outcomes
are proxies rather than direct measures of the effectiveness of mo-
torcycle rider training We opted to report on the outcomes of
crashes injury death and offences alone since there is little ev-
idence about how changes in knowledge and attitudes relate to
these outcomes
A U T H O R S rsquo C O N C L U S I O N S
Implications for practice
Due to the poor quality of studies included in this review we were
unable to draw any conclusions about the effectiveness of rider
training on crash rates injurious or fatal or offence rates The
findings do suggest that mandatory pre-licence training may be
an impediment to completing a motorcycle licensing process thuspossibly indirectly reducing crash injury death and offence rates
through a reduction in exposure to riding a motorcycle On the
basis of the existing evidence it is not clear if and what type of
training reduces the risk of crashes injuries deaths or offences in
motorcyclists and the selection of the best rider training practice
can therefore not be recommended The suggestion that manda-
tory pre-licence training indirectly reduces exposure to riding the
motorcycle because trainees are impeded from completing the li-
censing process does not help those motorcyclists who persist in
obtaining their licence and riding the motorcycle Some type of
rider training is likely necessary to teach motorcyclists to ride a
motorcycle safely Good policy is based on sound evidence and
evaluation but such evidence for motorcycle rider training is seri-
ously lacking Policy makers are therefore encouraged to include arigorous evaluation component to any new or existing rider train-
ing programme Such an evaluation should ideally be randomized
and assess whether the training programme is effective and if so
which component(s) of the programme work best
Implications for research
Given theresults of this review which found littlereliable evidence
about the effectiveness of motorcycle rider training it is necessary
to build up an evidence base of the effectiveness of motorcycle
rider training in the reduction of crash injury death and offence
rates To this end well designed randomised controlled trials need
to be conducted that examine the impact of rider training oncrashes and offences A randomised controlled trial is the preferred
method as it would significantly reduce most biases described in
this review To minimise the impact of participants delaying or
never attaining their licence and therefore exposure as found in
the randomised trials in this review such a trial should include
riders who are already licensed and reliable measures of riding
exposure in a reasonably lengthy follow up period It is expected
that a randomised controlled trial of sufficient sample size would
observe equal balance on possible confounders by study arm how-
ever an assessment of potentially important confounders (such as
helmet wearing) by study group would enable control for these
variables in the analyses if necessary It is imperative that the effec-
tiveness of each component of the rider training (such as theorypractice knowledge skills and modification of behaviour) and the
synergy of such components are evaluated Outcomes should be
objective ideally those that are routinely collected rather than self-
reported outcomes Even though riding the motorcycle poses dif-
ferent challenges and incurs different risks lessons may be learned
from research into the effects of car driver training
Motorcycles are a common means of transport in low- and middle
income countries and usage is increasing As the burden of mo-
torcycle crash victims is mostly borne in such countries research
should be focused on the effectiveness of types of rider training on
the reduction of crashes injuries and fatalities in such countries
and be supported by higher income countries This review found
only one study conducted in a middle income country that sug-gested decreasing injury rates It would be beneficial to conduct
additional research to confirm this suggestion and to enhance our
knowledge of the role of rider training in low- and middle-income
settings
March 2006 saw the launch of the University of North Carolina
Highway Safety Research Centre partnership with the Motorcycle
Safety Foundation (MSF) and plans were revealed to conduct a
study of the potential benefits of motorcycle rider training The
study is funded by the MSF and the National Highway Traffic Sa-
fety Administration (NHTSA) Called rsquoThe Discovery Projectrsquo it
will compare data on crash rates from riders participating in a basic
23Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2650
motorcycle safety class with riders participating in a basic motor-
cycle safety class as well as a variety of continuing education classesThe collection of pilot data was due to start in March 2007 and
the data collection portion of the study is expected to take three
years (Highway Safety Research Center) It is vital that researchers
and policy makers collaborate to develop furtherhigh quality eval-
uations of motorcycle rider training programmes to help inform
development of new and effective training programmes
A C K N O W L E D G E M E N T S
Staff from The University of Sydney library for helping us locate
and retrieve full reports of many papers
R E F E R E N C E S
References to studies included in this review
Adams 1985 published data only
Adams A Collingwood V Job RFS Evaluation of the 1982 Australia
Post motorcycle rider training scheme NSW Australia Traffic
Accident Research Unit Traffic Authority NSW 1985 [ RN 85]
Anderson 1980 published data only
Anderson J Ford JL Peck RC Improved motorcyclist licensing and testing project Washington DC National Highway Traffic Safety
Administration (NHTSA) USA 19801ndash217 [
DOTndashHSndash805ndash598]
Billheimer 1998 published data only
Billheimer JW Evaluation of the California Motorcyclist Safety
Program Transportation Research Record 19981640100ndash9 [
TRB Report Number 980652]
Cooper 1988 published data only
Cooper PJ Rothe JP Motorcyclists who they are and why they do
what they do Transportation Research Record 1988116878ndash85
Davis 1997 published data only
David CF Evaluation of community traffic safety programs and
motorcycle operator training programs Connecticut Department of Transportation 199763
Hall 1988 published data only
Hall CJ Buchanan L Engle R Motorcycle rider education evaluationproject Washington DC USA National Highway Traffic Safety
Administration (NHTSA) US Department of Transportation
1988 [ DTNH 22ndash80ndashCndash05123]
Haworth 1997 published data only
Haworth N Smith R Brumen I Pronk N In Monash University
Accident Research Centre editor(s) Case-control study of motorcycle
crashes Victoria Australia Department of Transport and Regional
Development The Federal Office of Road Safety 19971ndash149 [
ISBN 0 642 25501 6 ]
Hurt 1981 published data only
Hurt H Ouellet J Thom D Motorcycle accident cause factors and
identification of countermeasures USA Department of
Transportation 1981 [ Technical Report DOT HS 806ndash862]
Jonah 1982 published data onlylowast Jonah BA Dawson N Bragg B Are formally trained motorcyclists
safer Accident Analysis amp Prevention 198214(4)247ndash55
Jonah BA Dawson NE Bragg BW Evaluation of the Canada
Safety Councilrsquos Motorcycle Training Program Proceedings of theInternational Motorcycle Safety Conference Washington DC
USA 198059ndash93
Kloeden 1994 published data only
Kloeden CN Moore VM McLean AJ Evaluation of the pre-license
training program for motorcyclists in South Australia Walkerville
South Australia Australia NHMRC Road Accident Research Unit
The University of Adelaide 19941ndash37 [ ISBN 0 7308 0607 3]
Kraus 1975 published data only
Kraus JF Riggins RS Franti CE Some epidemiological features of
motorcycle collision injuries American Journal of Epidemiology 1975102(1)74ndash98
Leung 1987 published data only
Leung KS Reding VA Evaluation of the Wisconsin motorcycle rider course Wisconsin Bureau of Policy Planning and Analysis
Wisconsin Department of Transportation 1987
MAIDS 2004 published and unpublished data
MAIDS In-depth investigations of accidents involving powered
two wheelers ACEM (Acemacemeu) 2004 Vol Final report
20179
McDavid 1989 published data only
McDavid JC Lohrmann BA Lohrmann G Does motorcycle
training reduce accidents Evidence from a longitudinal quasi-
experimental study Journal of Safety Research 198920(2)61ndash72
Mortimer 1984 published data only
Mortimer R Evaluation of the motorcycle rider course Accident
Analysis amp Prevention 198416(1)63ndash71
24Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2750
Mortimer 1988 published data only
Mortimer RG A further evaluation of the motorcycle rider courseJournal of Safety Research 198819(4)187ndash96
Osga 1980 published data only
Osga GA An investigation of the riding experiences of MSF rider
course participants Dissertation South Dakota USA The
University of South Dakota 19801ndash224
Perrino 2002 published data only
Perrino CS Ahmed A Callender A Rozier E Cantwell A Stewart
O The role of Marylandrsquos motorcycle rider course in promoting saver
behaviors and attitudes Baltimore USA National Transportation
Center Morgan State University 20021ndash52
Raymond 1979 published data only
Raymond S Tatum S An evaluation of the effectiveness of the RAC
ACU motorcycle training scheme - final report Road Safety ResearchUnit Dept of Civil Engineering University of Salford London
19791ndash131
Satten 1980 published data only
Satten RS Analysis and evaluation of the motorcycle rider courses
in thirteen Illinois counties International Motorcycle Safety
Conference Washington DC 1980145ndash93
Savolainen 2007 published and unpublished data
Savolainen Peter Mannering Fred Additional evidence on the
effectiveness of motorcycle training and motorcyclistsrsquo risk-taking
behaviour Transportation Research Board 2007 Vol March 20
19
Swaddiwudhipong 1998 published data only
Swaddiwudhipong W Boonmak C Nguntra P Mahasakpan P
Effect of motorcycle rider education on changes in risk behavioursand motorcycle-related injuries in rural Thailand Tropical Medicine
and International Health 19983(10)767ndash70
Waller 1992 published data only
Waller FL An analysis of the impact of the Texas Department of
Public Safety Motorcycle Operator Training Program upon
motocycle related traffic accidents Texas State University Public
Administration Program Applied Research Projects 199277
References to studies excluded from this review
Anderson 1980a published data only
Anderson LC Better biking a motorcycle rider improvement
program International Motorcycle Safety Conference Washington
DC USA Motorcycle Safety Foundation 1980231ndash53Aupetit 2008 published and unpublished data
Aupetit S Riff J Buttelli O Espie S Ergonomics research program
on motorcycle learning Contributions to the development of a
motorcycle simulator for training Transport Research Arena
Ljubljana 20089
Awane 1999 published data only
Awane T Integrating simulators in motorcycle safety education
Journal of International Association of Traffic and Safety Sciences 199923(1)26ndash35
Baldi 2005 published data only
Baldi S Baer JD Cook AL Identifying best practices states in
motorcycle rider education and licensing Journal of Safety Research200536(1)19ndash32
Batchler 1988 published data only
Batchler MW Evaluation of the pre-license motorcycle training course at the New South Wales traffic education centre Federal Office of
Road Safety (FORS) 19881ndash122 [ Report No CR 77 ]
Blanchard 2006 published data only
Blanchard HT Tabloski PA Motorcycle safety educating riders at
the teachable moment Mosby (CV) Company 200632330ndash2
Bonnett 2005 published data only
Bonnett J Mature age skills training for experienced motorcycle
riders Motorcycle ridersrsquo association of the Australian Capital
Territory 2005
Braver 2007 published data only
Braver ER et alPersuasion and licensure a randomized controlled
intervention trial to increase licensure rates among Maryland
motorcycle owners Traffic Injury Prevention 20078(1)39ndash46Clayton 1980 published data only
Clayton AB Sudlow DE An evaluation of the effectiveness of a one day training programme for learner motorcyclists Berkshire UK
Transport and Road Research Laboratory 19801ndash23 [ ISSN
0266ndash7045]lowast Clayton AB Sudlow DE An evaluation of the effectiveness of a one-day training programme for learner motorcyclists Berkshire UK
Road Safety Division Berkshire 19801ndash23 [ISSN 0266ndash7045]
Council 1978 published data only
Council FM Desper LP Dutt AK Driver education for motorcycle operation University of North Carolina Highway Safety Research
Center 19781ndash40
Crowther 2005 published data onlylowast Crowther G Engaging with motorcyclists UK police and theBikeSafe Road Safety Programme Published proceedings of the
Australasian Road Safety Research Policing Education Conference
Wellington New Zealand 2005 issue ISBN 0ndash473ndash10636ndash11ndash11
Daniello 2009 published data only
Daniello A Gabler HC Mehta YA Effectiveness of motorcycle
training and licensing Transportation Research Board 88th Annual
Meeting 200913
Dewanti 2007 published data only
Dewanti M Challenging efforts in promoting young motorcyclist
safety in Indonesia Road Safety on Four Continents 14th
International Conference Swedish National Road and Transport
Research Institute 2007
Du Plooy 1990 published data onlyDu Plooy DW Motorcycle training in South Africa The Human
Element International Motorcycle Safety Conference Orlando
Florida USA Motorcycle Safety Foundation 19904-1 to 4-9
Engel 1989 published data only
Engel U Krogsgaringrd Thomsen L [Faeligrdselsundervisning og
propaganda ndash en faeligrdselssikkerhedsfremmende foranstaltning]
Dansk Vejtidsskrift 198912295ndash7
Goldenbeld 2002 published data only
Goldenbeld C Houwing S de Craen S The development of
driving skills of young moped riders [De ontwikkeling van de
rijvaardigheid van jonge bromfietsers] SWOV Institute for Road
Safety Research (Stichting Wetenschappelijk Onderzoek
Verkeersveiligheid) Leidschendam the Netherlands 200263
25Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2850
Goldenbeld 2004 published data only
Goldenbeld C Twisk D de Craen S Short and long term effects of moped training a field experiment Transportation Research Part F 20047(1)1ndash16
Haworth 2000 published data only
Haworth N Smith R Kowadlo N Motorcycle training in Victoria
evaluation of rider training curriculum in Victoria Monash
University Accident Research Centre Reports 2000 Vol 165 issue
3131
Haworth 2005 published data only
Haworth N Mulvihill C Review of motorcycle licensing and
training Monash University Accident Research Centre 2005 [
Report no240]
Haworth 2006 published data only
Haworth N Mulvihill C Rowdon P Teaching old dogs new tricksTraining and older motorcyclists Journal of the Australasian College of Road Safety 200718(4)20ndash5lowast Haworth NL Mulvihill C Rowden PJ Teaching old dogs new
tricks Training and older motorcyclists Australasian Road Safety
Research Policing and Education Conference Gold Coast
Queensland 2006
Hill 1983 published data only
Hill D Motorcycle rider training testing and licensing Motor
Cycle Council of New South Wales Sydney 1983
Imatake 1980 published data only
Imatake Y On the activities in Japan of motorcycle safe riding
promotion committee (MSRPC) - especially the education of
unskilled riders of class 1 motorbikes International Motorcycle
Safety Conference Washington DC USA Motorcycle Safety
Foundation 1980255ndash68
Ingebritsen 1990 published and unpublished data
Ingebrigtsen S [Risikofaktorer ved ferdsel med moped og
motorsykkel] TOslashI-rapport 66 Oslo Transportoslashkonomisk
institutt 1990
Jain 2009 published and unpublished data
Jain A et alTwo wheeler accidents on Indian roads a study from
Mangalore India Elsevier 200916130ndash3
Jonah 1979 published data only
Jonah BA Dawson NE Validation of the motorcycle operator skill
test Accident Analysis amp Prevention 197911(3)163ndash71
Jonah 1981 published data onlyJonah BA Dawson NE Bragg WE Predicting accident
involvement with the Motorcycle Operator Skill Test Accident Analysis amp Prevention 198113(4)307ndash418
Kadar 1990 published data only
Kadar KJ Countermeasures motorcycle rider education delivery
system The Human Element International Motorcycle Safety
Conference Motorcycle Safety Foundation 19906-27 to 6-41
Kelsey 1986 published data only
Kelsey SL Liddicoat C Ratz M Licensing novice motorcyclists a
comparison of Californiarsquos standard test and the MOST II (Motorcycle
Operator Skill Test) administered at centralized testing offices Sacramento California USA California Department of Motor
Vehicles Research and Development Office 19861ndash74
Koch 1980 published data only
Koch H Advanced course for motorcycle riders InternationalMotorcycle Safety Conference Washington DC USA Motorcycle
Safety Foundation 1980269ndash81
Koch 1990 published data only
Koch H Brendicke R How to influence risk taking behaviour
within motorcycle rider training programs The Human Element
International Motorcycle Safety Conference Orlando Florida
USA Motorcycle Safety Foundation 1990Motorcycle Safety
Foundation 1990 p 9-1 to 9-25A
Labbett 2005 published data only
Labbett S Langham M Training can make the problem worse
Road Safety Congress 20051ndash12
Law 2005 published and unpublished data
Law TH Radin Umar RS Zulkaurnain S Kulanthayan S Impactof the effect of economic crisis and the targeted motorcycle safety
programme on motorcycle-related accidents injuries and fatalities
in Malaysia International Journal of Injury Control and Safety
Promotion 200512(1)9ndash21
Lowes 1990 published data only
Lowes BF In-traffic rider training The Human Element
International Motorcycle Safety Conference Orlando Florida
USA Motorcycle Safety Foundation 19907-46 to 7-57
Mahon 1976 published data only
Mahon J Teaching accident avoidance on the motorcycle Journal of Traffic Safety Education 1976231ndash29
Maxwell 1985 published data only
Maxwell GS Evaluation of a Program for Training MotorcycleRiders Evaluation and Program Planning 19858(4)339ndash47
Mayhew 1996 published data only
Mayhew DR Simpson SH Effectiveness and role of driver educationand training in a graduated licensing system Ottowa Canada The
Traffic Injury Research Foundation 19961ndash89 [ ISBN
0ndash920071ndash12ndash0]
McKnight 1987 published data only
McKnight AJ Milburn BG An evaluation of the modified
motorcycle operator skill test exercises NHTSA Report 1987 [
Report No DOT HS 807 212 ]
McPherson 1976 published data only
McPherson K McKnight AJ The development and evaluation of a
motorcycle skill test manual and knowledge test Washington DCUSA National Highway Safety Administration 1976
Mohan 2004 published data only
Mohan D Evidence-based interventions for road traffic injuries in
South Asia Journal of the College of Physicians and SurgeonsPakistan 200414(12)746ndash7
Morgan 1980 published data only
Morgan JK Advanced motorcycle training course content and
study evaluation International Motorcycle Safety Conference
Washington DC USA Motorcycle Safety Foundation 1980
283ndash312
Motoki 1990 published data only
Motoki M Yamazaki S A study on effective motorcycle rider
education The Human Element International Motorcycle Safety
26Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2950
Conference Orlando Florida USA Motorcycle Safety
Foundation 19909-26 to 9-57A
Nairn 1992 published data onlylowast Nairn RJ Motorcycle safety research literature review 1987 to
1991 Federal Office of Road Safety Australia 1993 Vol Report
CR 117
Newland 1980 published data only
Newland R Darby J Automotive practicespre motorcycle rider
program International Motorcycle Safety Conference Washington
DC USA Motorcycle Safety Foundation 1980127ndash9
Newland 1982 published data only
Newland R Motorcycle rider training and licensing requirements
Driver training steering a course for the future Melbourne
Victoria Transport Regulation Board 1982
Newland 1990 published data only
Newland R Motorcycle training and licensing in Victoria 1980-
1990 The Human Element International Motorcycle Safety
Conference Orlando Florida USA Motorcycle Safety
Foundation 1990
Ormston 2003 published data only
Ormston R Dudleston A Pearson S Stradling Evaluation of
Bikesafe Scotland Scottish Executive Social Research Edinburgh
2003
Osga 1979 published data only
Osga GS Ellingstad VS An evaluation of the effectiveness of the 1978
South Dakota motorcycle rider course program South Dakota USA
University of South Dakota 19791ndash89
Panou 2007 published data only
Panou M Bekiaris E A new concept on the integration of driving
simulators in driving training the train-all approach Road Safety
on Four Continents 14th International Conference Swedish
National Road and Transport Research Institute 2007
Peterson 1990 published data only
Peterson A Kadar K Status report 1980-1989 Minnesota rider
education statistics The Human Element International
Motorcycle Safety Conference Orlando Florida USA 199013-
113 to 13-129
Prem 1984 published data only
Prem H Good MC Motorcycle rider skills assessment ACT
Australia Office of Road Safety Department of Transport 1984
1ndash377 [ Report No CR 34 ]
Radin Umar 2006 published data only
Radin Umar RS Motorcycle safety programmes in Malaysia how
effective are they International Journal of Injury Control and Safety Promotion 200613(2)71ndash9
Rockwell 1990 published data only
Rockwell TH Kiger S Carnot M An evaluation of the Ohio
motorcyclist enrichment program Ohio Department of Highway
Safety 1990
Rothe 1987 published data only
Rothe JP A motorcycle riding course Motorcyclists Image and reality Canada Insurance Corporation of British Columbia 1987
59ndash83
Rowden 2007 published data only
Rowden P Watson B Haworth N What can riders tell us aboutmotorcycle training A view from the other side of the fence
Australasian Road Safety Research Policing and Education
Conference Melbourne Australia Melbourne 2007
RSC Victoria 1993 published data only
Road Safety Committee Inquiry into Motorcycle Safety in Victoria
Parliament of Victoria 1993
Russam 1979 published data only
Russam K Motorcycle training and licensing in Japan UK Transport
and Road Research Laboratory (TRRL) 1979 [ Report 916 ]
Saffron 1981 published data only
Saffron DG Driver and rider education training and licensing a
brief review Traffic Accident Research Unit Traffic Authority of
New South Wales NSW Australia Traffic Accident Research UnitTraffic Authority of New South Wales 1981
Schulz 1990 published data only
Schulz U Motorcycle accidents among learner riders Motorcycle
licensing Evaluation of the effectiveness of a Graduated Licensing Scheme Forshungshefte des Institut fur Zweiradsicherheit 1990
Sexton 2004 published data only (unpublished sought but not used)
Sexton B Baughan C Elliott M Maycock G The accident risk of
motorcyclists 2004TRL Report TRL 607(ISSN 09684107)58
Shephard 1986 published data only
Shepherd R Do Motorcycle Safety Foundation programs work
Driver 198619(9)16ndash8
Simpson 1990 published data onlySimpson HM Mayhew DR The promotion of motorcycle safety
training education and awareness Health Education Research
Theory and Practice 19905(2)257ndash64
Sokolov 1990 published data only
Sokolov O Safe motorcycle riding some aspects of motorcycle
riding in the USSR The Human Element International
Motorcycle Safety Conference Motorcycle Safety Foundation
19904ndash37 - 4-48
Steffens 1988 published data only
Steffens U Gawatz R Willmes G Effectiveness of mofa courses
[Wirksamkeit von Mofakursen] Unfall- und sicherheitsforschung
strasenverkehr 198867168
Sudlow 2003 published data only
Sudlow D Scoping study on motorcycle training Road Safety
Research Report London UK The British Institute of Traffic
Education Research Department for Transport 2003 Vol June
[ Report No 36]
Tasmania 1995 published data only
Mannering FL Grodsky LL Motorcycle rider training scheme
Traffic Education Research Centre Armidale Australia Driver
Training Australia Limited 1995 Vol April1ndash56
Tetard 1985 published data only
Tetard C Motorcycle ridersrsquo training to detect indices in an
intersection situation [Lrsquoentrainement a la prise drsquoinformation dans
lrsquoapprentissage de la conduite motocycliste] Le Travail Humain198548(2)147ndash59
27Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3050
Thackray 1980 published data only
Thackray RM Prescott JC Field test of the motorcycle ridercourse International Motorcycle Safety Conference Washington
DC USA Motorcycle Safety Foundation 1980195ndash7
Thompson 1994 published data only
Thompson M Evaluation of compulsory basic training for
motorcyclists Transport Research Laboratory Crowthorne
Berkshire UK 1994 [ Report 63]
Tobler 1990 published data only
Tobler U Individual driving instruction for motorcycle driving
learners the Swiss experiment of motorcycle driving school The
Human Element International Motorcycle Safety Conference
Orlando Florida USA Motorcycle Safety Foundation 1990
Transport 1989 published data only
Transport Communications Road Crash Statistics Australia Bureauof Transport and Communications Economics 198914ndash20
Tuan 2007 published data only
Tuan VA Shimizu T Towards development and evaluation of the
motorcycle drivers re-education program in Vietnam modeling of
motorcycle driverrsquos undesired behaviors 11th World Conference
on Transport Research 200725
Victor 1980 published data only
Victor DJ Traffic safety education for motorcyclists International
Motorcycle Safety Conference Washington DC USA Motorcycle
Safety Foundation 1980221ndash8
Waterer 1990 published data only
Waterer TH Compulsory training for motorcyclists from liberty
to compulsion The Human Element International Motorcycle
Safety Conference Orlando Florida USA Motorcycle Safety
Foundation 19904-49 to 4-57
Weller 1989 published data only
Weller RB Chandler EW Motorcycle safety and motorcycle
education past research and survey results Journal of Public Policy
amp Marketing 1989893ndash108
Wells 1981 published data only
Wells P Observations of some RACACU motorcycle training
schemes Transport and Road Research Laboratory Crowthorne
Berkshire UK 1981
Winn 1986 published data only
Winn G Two important trends in motorcycle safety regulations
rider education and conspicuity improvement American
Motorcyclist Association 1986147ndash52
Wisher 1988 published data only
Wisher G Reid J Motorcycle accidents in Norfolk a three-year study of motorcycle training and the Part 1 test Norfolk UK Road Safety
Division Highways Department Norfolk County Council 1988
Wood 1987 published data only
Wood HT Bowen R Evaluation of the revised learner permit scheme
July 1983 to December 1985 Victoria Australia Road Traffic
Authority 19871ndash27 [ ISBN 0 7306 0247 8]
Yamazaki 1990 published data only
Yamazaki S Nakanishi M Survey on effect of Honda rider
education The Human Element International Motorcycle Safety
Conference Orlando Florida USA 199012-167 to 12-187
References to ongoing studies
Burgess 2010 unpublished data only
Burgess C Broughton P Fylan F Stradling Interim evaluation of
the UKs National Ride program In L Dorn editor(s) Driver
behaviour and Training Vol IV Aldershot Ashgate Publishing
Limited 2010
Additional references
Anderson 1980
Anderson J Ford JL Peck RC Improved motorcyclist licensing and
testing project Volume I and II National Highway Traffic Safety
Administration US department of Transportation Washington
DC 1980
Bjornstig 1985
Bjornstig U Bylund P Lekander T Brorsson B Motorcycle fatalitiesin Sweden Acta Chirurgica Scandinavica 1985151(7)577ndash81
Clayton 1990
Clayton A Sudlow DE An evaluation of the effectiveness of a one-
day training program for learner motorcyclists The Human
Element International Motorcycle Safety Conference Orlando
Florida USA 199013-105 to 13-111
Cochrane Collaboration 2008
Higgins JP Green S Cochrane Handbook for Systematic Reviews of
Interventions 500 The Cochrane Collaboration 2008
Elliot 2003
Elliott MA Baughan CJT Broughton J Grayson G Knowles J et
alMotorcycle Safety a Scoping Study TRL [ Report No
0968ndash4107]Fleiss 2004
Fleiss JL Levin B Paik MC Statistical methods for rates and proportions 3rd Edition John Wiley amp Sons Inc 2004 [ ISBN
978ndash0ndash471ndash52629ndash2 ]
FORS 1999
FORS Federal Office of Road Safety Road risk for sober licensed
motorcyclists Vol Monograph 27 Federal Office of Road Safety
19991ndash2
Haworth 2005
Haworth N Mulvihill C Review of motorcycle licensing and training
Victoria Australia Monash University Accident Research Centre
20051ndash83 [ Report No 240]
Huwiler-Muumlntener 2002Huwiler-Muumlntener K Juumlni P Junker C Egger M Quality of
Reporting of Randomized Trials as a Measure of Methodologic
Quality JAMA 20022872801ndash4
Johnston 2008
Johnston P Brooks C Savage H Fatal and serious road crashes
involving motorcyclists Road Safety Research and Analysis Report
ATSB Monograph 20 Canberra Department of Infrastructure
Transport Regional Development and Local Government 0205
081ndash26
Juumlni 2001
Juumlni P Altman DG Egger M Systematic reviews in health care
Assessing the quality of controlled clinical trials BMJ 2001323
42ndash6
28Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3150
Kopits 2003
Kopits E Cropper M Traffic fatalities and economic growth TheWorld Bank Development Research Group Infrastructure and
Environment Policy research working paper 3035 2003
Leung 1983
Leung P Chau BTW Motorcycle accidents in Hong Kong pattern
of injury treatment causative factors and socio-economic
implications Injury 198315(1)23ndash9
Liu 2004
Liu B Ivers R Norton R Blows S Lo SK Helmets for preventing
injury in motorcycle riders Cochrane Database of Systematic Reviews 2004 Issue 2 [DOI 10100214651858CD004333pub3]
Mohan 2002
Mohan D Road safety in less-motorized environments future
concerns International Journal of Epidemiology 200231(3)527ndash32Mortimer 1982
Mortimer RG An evaluation of the motorcycle rider course
Proceedings of the 20th conference of the American Association for
Automotive Medicine Arlington Heights IL USA American
Association for Automotive Medicine 198277ndash96
Mullin 1997
Mullin BT Risk factors for motorcycle injury the role of age
gender experience training and alcohol PhD thesis Auckland
New Zealand University of Auckland 1997
Noordzij 2001
Noordzij PC Forke E Brendicke R Chinn BP Integration of needs
of moped and motorcycle riders into safety measures review and
statistical analysis in the framework of the European research
project PROMISING SWOV Institute for Road Safety Research
Leidschendam The Netherlands 2001 [ Report No Dndash2001ndash5]
Osga 1979
Osga G Ellingstad VS An evaluation of the effectiveness of the 1978 South Dakota Motorcycle Riders Course Program South Dakota
University of South Dakota Vermillion Human Factors Lab 1979
Oxford English Dictionary Online
httpoxforddictionariescom Oxford English Dictionary Online
Oxford University Press 2010
Peden 2004
Peden M Scurfield R Sleet D World Report on Road TrafficInjury Prevention Geneva World Health Organization Available
at httpwwwwhointviolence˙injury˙preventionpublications
road˙trafficworld˙reporten 2004Chapter 3
Radin Umar 1995
Radin Umar R Macakay G Hills B Preliminary analysis of
motorcycle accidents Short-term impacts of the running
headlights campaign and regulations in Malaysia J Traffic Med 199523(1)17ndash27
Tsai 1995
Tsai YJ Wang JD et alCase-control study of the effectiveness of
different types of helmets for the prevention of head injuries among
motorcycle riders in Taipei Taiwan American Journal of
Epidemiology 1995142(9)974ndash81
Vis 1995
Vis AA In-depth study of the hazards of motorcycling [De
onveiligheid van motorrijden nader bekeken] SWOV Institute for
Road Safety Research Leidschendam the Netherlands 1995 [
Report No Rndash95ndash69]
Weaver 1990
Weaver EA Motorcycle safety education in the United States
Motorcycle Safety Foundation Orlando Florida USA 1990
Winn 1987
Winn G Two important trends in motorcycle safety regulations
rider education and conspicuity improvement No more
information available Detroit MI US Society of Automotive
Engineers 1987147ndash52
Yuan 2000
Yuan W The effectiveness of the rsquoride-brightrsquo legislation for
motorcycles in Singapore Accident Analysis amp Prevention 200032
(4)559ndash63
Zambon 2006
Zambon F Hasselberg M Factors affecting the severity of injuries
among young motorcyclists a Swedish nationwide cohort study
Traffic Injury Prevention 20067(2)143ndash9lowast Indicates the major publication for the study
29Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3250
C H A R A C T E R I S T I C S O F S T U D I E S
Characteristics of included studies [ordered by study ID]
Adams 1985
Methods Non-randomised trial
Participants Initial sample size
Total of 162 Australia Post telegram delivery riders
Not clear how many in intervention and control groups
Analysis sample size
Intervention group 63Control group 35
Total 98 (61)
Interventions No formal training vs voluntary rider training course of 12 hours over 2 consecutive days (theory 4 hours practice
8 hours)
Intervention integrity
No information given
Outcomes Australia Post recorded crashes
Length of follow up 1 year
Notes Study conducted in NSW Australia
Anderson 1980
Methods Randomised trial
Participants Initial sample size
Total 41188 motorcycle licence applicants
Analysis sample size
Intervention group with skills training 13160
Intervention group without skills training 12634
Control group 15080
Total 40874 (99)
Interventions Standard licensing skill test vs the Motorcycle Operators Skill Test (MOST) with or without a training component
MOST is not to be confused with rider training because it is a licensing test The training component was for
participants who failed the skill test on the first attempt was of 3 hours duration and consisted of instructions about
the skills needed to pass the skill test The training consisted of 30 min theory and 2 frac12 hours practice on the
motorcycle
Intervention integrity
33 in MOST with skills training 31 in MOST without skills training and 43 in control group obtained their
motorcycle licence in the end
Outcomes Police recorded crashes and offences
Length of follow up 3 years
30Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3350
Anderson 1980 (Continued)
Notes Study conducted in California USA
Billheimer 1998
Methods Cohort study
Participants Initial sample size
Intervention group Unclear numbers of CMSP graduates and ERC graduates
Control group Over 16000 untrained motorcyclists
Analysis sample size
1139 matched CMSP-pairs and 1182 ERC-pairs
Interventions No formal training vs the mandatory (for those younger than 18 years) pre-licence 16-hours basic California
Motorcyclist Safety Program (CMSP) and the 8-hour Experienced Rider Course (ERC)
Intervention integrity
No information given
Outcomes Police recorded crashes and offences
Length of follow up 2 years
Notes Study conducted in California USA
Cooper 1988
Methods Cohort study
Participants Initial sample size
Intervention group 418 graduates and 57 who failed
Control group 402
Analysis sample size
Intervention group unclear
Control group unclear
Total 863 (984)
Interventions No formaltrainingvs non-mandatory pre-licence rider trainingThe content of thetrainingwas notfurther explainedIntervention integrity
Participants had all completed the rider training
Outcomes Police recorded crashes and offences
Length of follow up 7 years
Notes Study conducted in British Columbia Canada
31Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3450
Davis 1997
Methods Retrospective cohort study
Participants Initial sample size
Intervention group 9320 graduates
Control group unclear
Analysis sample size
Intervention group 9320 graduates
Control group unclear
Interventions No formal training vs the non-mandatory Connecticut Motorcycle Operator Training Program The content of
training was not further explained
Intervention integrity Participants had all completed the rider training
Outcomes Police recorded crashes severity of crashes
Length of follow up 1 year
Notes Study conducted in Connecticut USA
Hall 1988
Methods Randomised trial
Participants Initial sample sizeIntervention group A 6248
Intervention group B 6616
Intervention group C 6996
Control group 6604
Total 26464 motorcycle licence applicants
Analysis sample size
Intervention group A 6012 (96)
Intervention group B 6428 (97)
Intervention group C 6778 (97)
Control group 6502 (98)
Interventions Standard licensing skill test vs the Motorcycle Operator Skill Test (MOST II) without training (A) MOST II with
3 hours training (B) or MOST II with 20 hours training (C) The 3 hours training comprised instructions aboutthe skills needed to pass the skill test (30 min theory and 2 frac12 hours practice on the motorcycle) The 20 hours
training consisted of the Motorcycle Safety Foundationrsquos Motorcycle Rider Course (MRC) comprising of 8 hours
theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-road practice
Intervention integrity
25 and 26 obtained their motorcycle licence in the lsquoMOST 3-hour trainingrsquo and lsquoMOST 20-hour trainingrsquo
intervention arm respectively In the lsquoMOST testrsquo intervention arm 29 obtained their licence and in the control
group 35
Outcomes Police recorded crashes and offences
Length of follow up 3 years
32Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3550
Hall 1988 (Continued)
Notes Study conducted in New York USA
Haworth 1997
Methods Population based case-control study
Participants Cases Motorcycle riders or pillion riders involved in motorcycle crash (on road) Recruited from hospital and police
records Not clear whether all cases (especially fatalities) were included
Community controls Multiple matched sample of motorcycle rider or pillion who passed the crash site at the same
time of day and day of week that the crash occurred
Initial sample size
Cases 336
Controls 1195
Analysis sample size
Cases 234 (70)
Controls 561 (47)
Interventions Basic handling skills course beginner intermediate or advanced course measured through self-report not validated
Outcomes Injuries with an Injury Severity Score of 5 or greater or admission for at least four hours and fatalities from a
motorcycle crash (from hospital and police records)
Length of follow up two years or more prior to the interview
Notes Study conducted in Victoria Australia
Hurt 1981
Methods Population based case-control study
Participants Cases Motorcycle riders who had a motorcycle crash that was reported to police (traffic accident reports) Not all
cases included
Community controls Multiple matched sample of motorcycle riders who passed the crash site at the same time of
day and day of week as the crash occurred but 2 years later
Initial sample size
Cases 4500 notifications of crashes to police (only 50 received in time research team able to respond to 1126 for
unreported reasons data collection completed for 900)
Controls Number approached unreported 2310 consented and interviewed
Analysis sample size
Cases 808 (808900 data on rider training) - 18 of initial sample
Controls 670 (6702310 data on rider training) - lt 29 of initial sample
Interventions Training experience self taught friends-family school-club formal-professional others measured through self-
report validated
Outcomes Injuries and fatalities from a motorcycle crash (from hospital records and coronerrsquos office)
33Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3650
Hurt 1981 (Continued)
Notes Study conducted in Los Angeles USA
Jonah 1982
Methods Cohort study
Participants Initial sample size
Intervention group
2310 licensed MTP graduates who are active riders
Control group
2029 licensed motorcyclists who are active riders
Non-licensed and non-active riders excluded
Analysis sample size
Intervention group
811 interviews (35) 533 driver records (23)
Control group 1080 interviews (71) 905 driver records (45)
Interventions No training vs the non-mandatory Canada Safety Council Motorcycle Training Program (MTP) comprising of 4
hours theory and 16 hours practise (6 hrs on-road) over 2 weekends
Intervention integrity
Participants had all successfully completed the MTP
Outcomes Self-reported and police reported crashes and offencesLength of follow up 4 years
Notes Study conducted in Ontario Canada
Kloeden 1994
Methods Cohort study
Participants Initial sample size
Intervention group 2533 licence applicants
Control group Unclear
Sample of riders obtaining probationary licenceIntervention group 1625 (64)
Control group 5015 (unclear)
Non-active riders excluded from control group (~66)
Analysis sample size
Intervention group 631 (39)
Control group 318 (6)
Interventions No formal training vs the mandatory pre-licence Ridersafe course consisting of two levelsthe first is a training course
in the basics of motorcycle riding where a pass entitles the rider to obtain a learnerrsquos permit The second is a skill
test where a pass entitles the rider to obtain a probationary motorcycle licence The training course consists of two 4
hour training sessions on separate days involving instruction videos and actual riding on motorcycles The skill test
consists of a 4 hour preparation course of 74 minutes of lecturediscussion 38 minutes of video and 75 minutes of
34Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3750
Kloeden 1994 (Continued)
on-course riding which is the testing phase
Intervention integrity
Unclear what percentage in the Ridersafe group and the control group obtained their probationary licence halfway
and their full motorcycle licence in the end
Outcomes Police reported crashes
Length of follow up 1 year
Notes Study conducted in SA Australia
Kraus 1975
Methods Population based case-control study
Participants Cases All motorcycle riders with a confirmed medically treated motorcycle injury (including fatalities) from all
hospitals in study base and peripheral area hospitals and all deaths due to motorcycle crashes in the study base All
serious injuries (leading to hospitalisation) and fatalities likely to be included
Community controls A random sample of registered motorcycle owners in the same temporary and geographical
study base as the cases arose from
Initial sample size Cases 1273
Controls 738
Analysis sample size Cases 741 (58)
Controls 434 (59)
Interventions Motorcycle rider training (not specified) measured through self-report
Outcomes Injuries leading to hospitalisation fatalities
Length of follow up not specified
Notes Study conducted in Sacramento county USA
Leung 1987
Methods Non-randomised trial
Participants Initial sample sizeIntervention group unknown number of MRC graduates
Control group unknown number of licensed motorcyclists
Analysis sample size
Intervention group 2941
Control group 43094
Interventions No formal training vs non-mandatory Wisconsin motorcycle rider course a 22 hours off-street programme that
teaches the basic rules of safe motorcycling
Intervention integrity
50 in the MRC group and 100 in the control group obtained their motorcycle licence in the end
35Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3850
Leung 1987 (Continued)
Outcomes Police recorded crashes and offences
Length of follow up up to 3 years
Notes Study conducted in Wisconsin USA
MAIDS 2004
Methods Population based case-control study
Participants Cases Motorcycle riders who had a motorcycle crash Unclear whether all cases included
Community controls Motorcycle riders identified at petrol stations Unclear whether matched to cases
Initial sample size
Cases not reported
Controls not reported
Analysis sample size
Cases 921 (Italy 200 Netherlands 200 Spain 121 Germany 250 France 150)
Controls 923 (Italy 200 Netherlands 200 Spain123 Germany 250 France 150)
Interventions Pre-license rider training or additional training (non-compulsory)
Outcomes Crashes
Notes Study in five sampling areas located in five European countries (France Germany Netherlands Spain and Italy)
McDavid 1989
Methods Cohort study
Participants Sample size after first match
Intervention group 346 male training graduates
Control group 346 male motorcycle riders licensed in same year
Sample size after second match
Intervention group 139 (40)
Control group 139 (40)
Analysis sample size
Intervention group 139 (40)
Control group 139 (40)
Interventions No formal training vs the non-mandatory pre-licence British Columbia Safety Councilrsquos 37-hour motorcycle safety
programme
Intervention integrity
Participants had all completed and passed the training
Outcomes Police reported crashes injurious crashes
Length of follow up 1-6 years
36Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3950
McDavid 1989 (Continued)
Notes Study conducted in British Columbia Canada
Mortimer 1984
Methods Cohort study
Participants Initial sample size
Intervention group 608 MRC graduates
Control group 360 motorcycle riders
Response rates
Intervention group 360 (59)
Control group 324 (90)
Non-active riders excluded
Analysis sample size
Intervention group 213 (35)
Control group 303 (84)
Interventions No formal training vs the non-mandatory Motorcycle Safety Foundationrsquos Motorcycle Rider Course comprising of
8 hours theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-road practice
Intervention integrity
Participants had all completed and passed the MRC
Outcomes Self-reported crashes and offencesLength of follow up 1 year
Notes Study conducted in Illinois USA
Mortimer 1988
Methods Cohort study
Participants Initial sample size
Intervention group 2307 MRC graduates
Control group 556 active licensed motorcycle riders
Response ratesIntervention group 1298 (56)
Control group 500 (90)
Non-active riders excluded
Analysis sample size
Intervention group 913 (40)
Control group 500 (90)
Interventions No formal training vs the non-mandatory Motorcycle Safety Foundationrsquos Motorcycle Rider Course comprising of
8 hours theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-road practice
Intervention integrity
Participants had all completed and passed the MRC
37Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4050
Mortimer 1988 (Continued)
Outcomes Self reported crashes and offences
Length of follow up 1 year
Notes Study conducted in Illinois USA
Osga 1980
Methods Cohort study
Participants Initial sample sizeIntervention group 1163 MSF MRC participants invited 818 (74) responded
Control group 329 motorcycle riders responded to recruitment strategies
Non-active riders excluded from analysis
Analysis sample size
Intervention group 595 (73)
Control group 329 (100)
Interventions No formal training vs the non-mandatory Motorcycle Safety Foundationrsquos Motorcycle Rider Course (MRC) com-
prising of 8 hours theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-road
practice
Intervention integrity
Participants had all successfully completed the MRC
Outcomes Self-reported crashes police recorded crash reports
Length of follow up 6 months - 3 12 years
Notes Study conducted in South Dakota USA
Perrino 2002
Methods Cohort study
Participants Initial sample size
Intervention group 2144 (unclear) MRC graduates (motorcycle licence applicants)
Control group 2601 (unclear) regular licensing test graduates (motorcycle licence applicants)Analysis sample size
Intervention group 1070 (unclear)
Control group 867 (unclear)
Interventions No formal training vs the mandatory (for those younger than 18 years) pre-licence Maryland Motorcycle Rider
Course (MRC) consisting of 8 hours theory and 12 hours practice
Intervention integrity
No information given
Outcomes Self-reported and police recorded crashes and offences
Length of follow up 1 year
38Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4150
Perrino 2002 (Continued)
Notes Study conducted in Maryland USA
Raymond 1979
Methods Cohort study
Participants Initial sample size
Intervention group
258 volunteers mainly (but not all) novice riders
Control group
1104 provisional licence applicants
Analysis sample size
Intervention group 190 (74)
Control group 627 (57)
Interventions No formal training vs the non-mandatory RACACU (Royal Automobile ClubAuto-Cycle Union) training scheme
The training program consisted of either 24 one-hour sessions or 13 two-hour sessions Twelve hours of the training
focused on theory about basic operational procedures traffic laws and highway code The practical session focused
on basic operational skills riding on a training area and riding in traffic with an instructor A proficiency test was
administered at the end of the course
Intervention integrity
No information given
Outcomes Self-reported crashes (including minor non-injury non-damage and off-road crashes)
Length of follow up 1 year
Notes Study conducted in the UK
Satten 1980
Methods Cohort study
Participants Initial sample size
Intervention group interviewers kept interviewing until they had 100 riders selected from a population of 891
enrollees in the MSF MRC course (unknown how many were asked and how many refused to participate)Control group unknown number of registered motorcycle riders
Non-active riders excluded from analysis
Analysis sample size
Intervention group 71
Control group 100
Interventions No formal training vs the non-mandatory Motorcycle Safety Foundationrsquos (MSF) Motorcycle Rider Course (MRC)
comprising of 8 hours theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-
road practice
Intervention integrity
Participants had all successfully completed the MRC
39Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4250
Satten 1980 (Continued)
Outcomes Self-reported crashes injuries and offences
Length of follow up 3 years
Notes Study conducted in Illinois USA
Savolainen 2007
Methods Cohort study
Participants Initial sample size
Intervention group 4000 riders that had conducted the MSF MRC course (passed or failed)
Control group 4000 motorcycle owners or permit or endorsement holders from the Indiana Bureau of Motor
Vehicles database
Analysis sample size
Intervention group 739 (18)
Control group 588 (15)
Interventions Intervention integrity
No information given
Outcomes Probability of crash involvement
Length of follow up up to 5 years
Notes Study conducted in Indiana USA
Swaddiwudhipong 1998
Methods Randomised trial
Participants Initial sample size
Intervention group
all 16 villages of 3 sub-districts
(participants were practising motorcyclists licensed and unlicensed)
Control group
18 villages of 3 other sub-districtsAnalysis sample size
Intervention group 3 districts (100)
Control group 3 districts (100)
Interventions No rider education vs non-mandatory community based rider education consisting of theory on motorcycle injury
risk helmet use traffic laws vehicle regulations and traffic signs and a written and skill test leading to a licence
Intervention integrity
No information given
Outcomes Injuries fatalities
Length of follow up 2 years
40Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4350
Swaddiwudhipong 1998 (Continued)
Notes Study conducted in Tak Province rural Thailand
Waller 1992
Methods Cohort study
Participants Initial sample size
Intervention group 349 riders who succesfully completed the Department of Public Safety motorcycle operator
training in 1988
Control group 349 motorcyclists who never completed the DPS motorcycle operator training
Analysis sample size
Intervention group 349
Control group 349
Interventions No DPS motorcycle operator training vs non-mandatory DPS motorcycle operator training The DPS training is
the Motorcycle Safety Foundationrsquos Motorcycle Rider Course comprising eight hours of theory (lectures discussions
films and multiple choice tests) a textbook and 12 hours off-road practice
Intervention integrity
All 349 riders succesfully completed the rider training
Outcomes Police recorded crashes
Length of follow up 3 years
Notes Study conducted in
Characteristics of excluded studies [ordered by study ID]
Study Reason for exclusion
Anderson 1980a Study type review and process evaluation of a rider course
Aupetit 2008 Outcome proposal of a rider training curriculum that combines traditional training and a motorcycle riding
simulator
Awane 1999 Study type review on use of simulators in rider and driver training
Baldi 2005 Intervention not rider training but intervention integrity of rider training availabilty of rider education
implementation of graduated licensing system etc on a state-wide level
Batchler 1988 Study type description of a training course
Blanchard 2006 Study type not a study assessing the association between rider training and outcomes
Bonnett 2005 Outcome qualitative data regarding satisfaction with the training programme
41Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4450
(Continued)
Braver 2007 Outcome licensure rates
Clayton 1980 Outcome knowledge skills attitude
Council 1978 Outcome crash reduction in ANY vehicle not motorcycles exclusively
Crowther 2005 Study type qualitative study
Daniello 2009 Study type review of the effectiveness of motorcycle training and licensing
Dewanti 2007 Intervention emotional characteristics behaviour (ie not motocycle rider training)
Du Plooy 1990 Study type review of types of rider training courses in South Africa
Engel 1989 Study type uncontrolled before-after study
Goldenbeld 2002 Outcome skills
Goldenbeld 2004 Outcome knowledge and skills
Haworth 2000 Intervention balance between attitudinal andvehicle control skills acrosscurriculla of motorcycle rider training
courses
Haworth 2005 Study type review
Haworth 2006 Study type cross-sectional survey
Hill 1983 Study type review and description of training system
Imatake 1980 Study type course descriptionreview
Ingebritsen 1990 Intervention risk factors for motorcycle crashes
Jain 2009 Study type cross-sectional descriptive study of motorcycle crashes over a 5 year period
Jonah 1979 Study type cross-sectional survey Outcome scores on motorcycle operator skill test
Jonah 1981 Intervention evaluation of a motorcycle rider skill test not motorcycle rider training
Kadar 1990 Study type review of components of rider training courses
Kelsey 1986 Intervention evaluation of a motorcycle rider skill test not motorcycle rider training
Koch 1980 Study type reviewcourse description
Koch 1990 Study type description of course content
42Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4550
(Continued)
Labbett 2005 Study type essay about possible explanations for motorcycle deaths
Law 2005 Intervention targeted motorcycle safety programme not specifically rider training
Lowes 1990 Study type review and descriptions of rider training
Mahon 1976 Study type review of motorcycle rider training methods
Maxwell 1985 Outcomesrecommendations forprogramme operatorsand otherson howthe motorcycle trainingprogramme
works and how it can be improved
Mayhew 1996 Study type review
McKnight 1987 Study type evaluation of methodological and administrative aspects of a motorcycle operator skill test
McPherson 1976 Outcome knowledge and skills
Mohan 2004 Study type review of evidence of interventions to reduce road traffic injuries
Morgan 1980 Study type reviewcourse description
Motoki 1990 Study type proposal and development of rider training curriculum
Nairn 1992 Study type literature review
Newland 1980 Study type description of development of motorcycle safety course
Newland 1982 Study type descriptionreview of a rider training course
Newland 1990 Study type uncontrolled BeforeAfter study
Ormston 2003 Outcome changes in attitude and self reported behaviour
Osga 1979 Outcome knowledge attitudes and skills
Panou 2007 Intervention new simulation tool to teach driverrider
Peterson 1990 Study factor demographic details of course participants completing the course vs those not completing the
course
Prem 1984 Study type experimental
Study factor examines association between skills and performance in tests
Radin Umar 2006 Intervention motorcycle safety programme not specifically rider training
Rockwell 1990 Study type methodologically flawed study type
43Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4650
(Continued)
Rothe 1987 Study type qualitative study of rider training in general
Rowden 2007 Study type qualitative study
RSC Victoria 1993 Study type review of motorcycle rider training and education in Victoria
Russam 1979 Study type review of training programmes in Japan
Saffron 1981 Study type literature review
Schulz 1990 Study factor predictors of motorcycle crashes
Sexton 2004 Study type cross-sectional analytic analysis
Shephard 1986 Study factor reviewdescription of programme
Simpson 1990 Study type review of motorcycle rider trainingeducation
Sokolov 1990 Study factor review of motorcycle riding courses in USSR
Steffens 1988 Outcome self-reported behaviour experiences and researcher observed riding behaviour
Sudlow 2003 Study type review of motorcycle training in the UK
Tasmania 1995 Study type cross-sectional descriptiveprocess evaluation
Tetard 1985 Study type experimental
Outcome detection of indices
Thackray 1980 Outcome instructional effectiveness user acceptance administrative feasibility
Thompson 1994 Study type cross-sectional survey
Outcome attitudes and knowledge
Tobler 1990 Study type review of course curricula
Transport 1989 Intervention syllabus for rider training
Tuan 2007 Study type cross-sectional study (survey) which does not assess the effects of an intervention but examines
motocyclistsrsquo behaviour
Victor 1980 Study type review
Waterer 1990 Study type review of UK rider training issues
Weller 1989 Study type review of studies evaluating motorcycle safety and education measures
44Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4750
(Continued)
Wells 1981 Study type description of different methods of training
Winn 1986 Study type review of rsquohistoryrsquo of rider training
Wisher 1988 Study type cross-sectional survey
Wood 1987 Study type Interrupted Time Series but did not meet the criteria for an ITS as published by the Cochrane
Effective Practice and Organisation of Care Group
Yamazaki 1990 Study type cross-sectional survey
Outcome riding skills skills retention and attitude
Characteristics of ongoing studies [ordered by study ID]
Burgess 2010
Trial name or title UKs National Ride program
Methods Cohort study
Participants Motorcycle offenders
Interventions Police led diversion program
Outcomes Self reported attitudinal and behavioural measures
Starting date
Contact information
Notes Expected to be completed in early 2010
45Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4850
D A T A A N D A N A L Y S E S
This review has no analyses
A P P E N D I C E S
Appendix 1 Search strategy
We searched the following electronic databases
bull Cochrane Injuries Group Specialised Register (searched 22 Sept 2008)
bull CENTRAL (The Cochrane Library 2008 Issue 3)
bull TRANSPORT 1988 to 200706 (latest date available)
bull ATRI Australian Transport Index (to Sept 2008)
bull MEDLINE (Ovid SP) 1950 to Sept 2008
bull EMBASE (Ovid SP) 1980 to Sept 2008
bull CINAHL (EBSCO) 1982 to Sept 2008
bull WHOLIS (World Health Organization Library Information System) (to Sept 2008)
bull PsycINFO (Ovid SP) 1806 to Sept 2008
bull LILACS (BIREME) (Latin American and Caribbean Health Sciences) (1982 to Sept 2008)
bull ISI Web of Science Social Sciences Citation Index (SSCI) 1970 to Sept 2008
bull ERIC (searched 25 Sept 2008)
bull ZETOC (searched 25 Sept 2008)
bull SIGLE (searched 25 Sept 2008)
All of the databases were searched using the following text-word search strategy The strategy was adapted using either rsquorsquo or rsquo$rsquo as atruncation symbol
1 motor-cycl or motorcycl or motor-bik or motorbik or motor-scooter or motorscooter or scooter or moped or motocycl
or motocicl or moto
2 train or educat or instruct or program or course or qualif or learn or improv
3 1 and 2
H I S T O R Y
Protocol first published Issue 2 2005
Review first published Issue 10 2010
46Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4950
C O N T R I B U T I O N S O F A U T H O R S
Katina Kardamanidis was the first author of the protocol conducted the literature search extracted full studies assessed the studies
according to inclusion criteria assessed quality of the studies compared findings with the second author discussed disagreements with
the third author wrote the drafts and the final review
Rebecca Ivers was the second author of the protocol assessed the studies according to inclusion criteria assessed quality of the studies
compared findings with the first author edited the final draft and provided overall guidance for the process
Alexandra Martiniuk assessed quality of the studies compared findings with the first author discussed disagreements with the third
author and helped write the first draft and the final review
Mark Stevenson contributed to the protocol reviewed methods of studies and contributed to the final draft
Katrina Thistlethwaite conducted a second literature search extracted full studies and contributed to the final draft
D E C L A R A T I O N S O F I N T E R E S T
None known
S O U R C E S O F S U P P O R T
Internal sources
bull No sources of support supplied
External sources
bull National Health and Medical Research Council of Australia Australia
Salary support for Martiniuk Ivers and Stevenson
bull University of Sydney Sesqui Research Grant Australia
Direct research costs and salary support for Kardamanidis
D I F F E R E N C E S B E T W E E N P R O T O C O L A N D R E V I E W
There are no differences between the protocol and the review
I N D E X T E R M S
Medical Subject Headings (MeSH)
lowastMotorcycles Accident Prevention [lowastmethods] Accidents Traffic [lowastprevention amp control] Licensure Program Evaluation Wounds
and Injuries [mortality lowastprevention amp control]
47Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 5050
MeSH check words
Humans
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2450
1988) one study (Jonah 1982) also found a statistically significant
lower offence rate in the intervention group However we have lit-tle confidence in the results from Leung 1987 since potential con-
founding factors were not appropriately addressed Jonah 1982
did conduct a multivariate analysis but was hindered by an initial
low response rate The other four cohort studies did control for
several potential confounders and found no difference in offence
rates following non-mandatory post-licence rider training
The non-randomised trials and cohort studies evaluating non-
mandatory rider training were at high risk of substantial selection
bias due to the non-random allocation of participants as well
as low participation rates and attrition bias and could therefore
not provide conclusive evidence with regard to the effect of the
training It should be noted that despite the biases and other issues
with this group of studies most didnot detectmajor effects relatedto rider training in either direction This may suggest that any
potential impact of rider training will be modest
Studies examining unspecified modes of ridertraining
Three case-control studies assessed whether injured or deceased
motorcyclists were more or less likely to have had rider training
compared to control motorcyclists All had conflicting results
Hurt1981 foundthe odds of havingan injuriouscrash waslower in
those who had received rider training whereas Kraus 1975 found
that the odds of having an injurious crash were higher in youngerand older trained riders Haworth 1997 who compared advanced
versus beginner training found no effect on motorcycle injuries
or deaths although they did find that those who had completed
advanced training had reduced chances of crashing compared to
those who had done a beginnersrsquo training course Similar issues
with uncontrolled confounding and biases of selection as described
above arose with the three case-control studies Hurt 1981 suffers
from inadequate control of confounding via poor matching and
both the Kraus 1975 and the Haworth 1997 findings were only
controlled for limited factors by means of stratified analysis It is
likely that these limitations are the reason for the contradictory
directions of effect observed for the three case-control studies
Methodological weaknesses of the studies
In summary most studies investigating the effectiveness of mo-
torcycle rider training on reducing crashes injuries and death suf-
fered from serious methodological weaknesses
The majority of studies were non-randomised and were there-
fore limited mainly by selection bias Although the approaches of
matching stratification or regression analyses can minimise bias
(Fleiss 2004) they were poorly employed in the reviewed studies
The inadequate adjustment for confounders in particular adds
to our lack of confidence in the studiesrsquo results since it has been
demonstrated that individuals who choose to take motorcycle rider
training differ from those not taking rider training For instance
trained riders often differ from controls in age sex and experi-ence usually ride less and may have a more positive attitude to
safety as indicated by the reported more frequent wearing of seat
belts (Mortimer 1982)
The two large randomised trials reported that a large proportion
of their study group were never exposed to the risk of offences or
crashes because they never attained licensure This made it impos-
sible to assess whether a reduction in crashes in the intervention
group was a result of training or due to delayed or never attained
licensure
In some studies the sample sizes were too small to detect any sta-
tistically significant differences Follow-up time after training dif-
fered between studies and was an impediment to observing any
real effects of training if they existed As riding experience ac-cumulates over time any differences in outcomes between those
trained and untrained would diminish in studies with lengthy fol-
low-up times This dilution of any possible effect of rider training
may have also occurred in studies that mixed both novice and ex-
perienced riders in the study groups since it is hypothesised that
novice riders serve to benefit most from rider training This would
increase the difficulty in finding any effect of rider training par-
ticularly when sample sizes are small
Detection bias may also account for some of the differences ob-
served between trained and untrained riders Detection bias would
have occurred if the self-report hospital driver or police records
used to record the outcomes of crash injury death or offences
differed between the trained and untrained riders Research hasdemonstrated that not all crashes are reported to police for ex-
ample small crashes or off-road crashes (Kraus 1975) even serious
injury crashes have been found to be underreported (Anderson
1980) However this would only introduce detection bias if the
reporting of outcome differs for trained and untrained riders This
does appear possible in that responsible riders may be more likely
to enroll in a rider training course and also may be more likely
to report crashes to police than irresponsible riders meaning that
the number of crashes from riskier riders could be underrepre-
sented on records An analysis to investigate the effectiveness of
rider training may then show that rider training is not effective in
reducing crashes
Rather than using reported outcomes some studies used self-re-ported survey data although this technique suffers from non-
response biases In fact two studies reported differences in de-
mographic characteristics in response rates to questionnaires
(Anderson 1980 Hall 1988) Self-reported data also introduce
the risk of recall bias survival bias and social desirability response
bias Recall bias may have been equally distributed among trained
and untrained riders howeverit is hypothesised that trained riders
may have felt it was lsquoexpectedrsquo to report good driving histories
hence leading the rider training to appear effective
22Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2550
Limitations of this review
Limitations of this review were that we found very few recentstudies (the two most recent ones published in 2002 and 2007)
and that we found very few studies conducted in low-and middle
income countries even though our search strategy was set up to
identify such studies Anotherlimitation wasthat forthe reviewed
studies we could not assess all outcomes of rider training that are
potentially useful Some of the included studies assessed outcomes
such as use of safety clothing including helmets Such outcomes
are valuable since the use of motorcycle helmets has been demon-
strated to be effective in preventing head injuries (Liu 2004) De-
spite this the use of safety clothing was not included as an out-
come in this review because the studies measuring this outcome
were limited by poor data quality due to very low response rates
Another outcome of interest in assessing the effectiveness of rider
training is lsquotype of injuryrsquo As Haworth 2005 state in their review
of motorcycle licensing and training rider training could be con-
sidered effective if injuries become less severe because motorcycle
riders have learned how to avoid more serious injuries through
rider training The studies included in this review did not specify
type of injury and so this outcome could not be assessed Last
changes in knowledge and attitudes following rider training are
other potentially useful outcomes We have excluded such studies
(Osga 1979 Clayton 1990 Goldenbeld 2004) as these outcomes
are proxies rather than direct measures of the effectiveness of mo-
torcycle rider training We opted to report on the outcomes of
crashes injury death and offences alone since there is little ev-
idence about how changes in knowledge and attitudes relate to
these outcomes
A U T H O R S rsquo C O N C L U S I O N S
Implications for practice
Due to the poor quality of studies included in this review we were
unable to draw any conclusions about the effectiveness of rider
training on crash rates injurious or fatal or offence rates The
findings do suggest that mandatory pre-licence training may be
an impediment to completing a motorcycle licensing process thuspossibly indirectly reducing crash injury death and offence rates
through a reduction in exposure to riding a motorcycle On the
basis of the existing evidence it is not clear if and what type of
training reduces the risk of crashes injuries deaths or offences in
motorcyclists and the selection of the best rider training practice
can therefore not be recommended The suggestion that manda-
tory pre-licence training indirectly reduces exposure to riding the
motorcycle because trainees are impeded from completing the li-
censing process does not help those motorcyclists who persist in
obtaining their licence and riding the motorcycle Some type of
rider training is likely necessary to teach motorcyclists to ride a
motorcycle safely Good policy is based on sound evidence and
evaluation but such evidence for motorcycle rider training is seri-
ously lacking Policy makers are therefore encouraged to include arigorous evaluation component to any new or existing rider train-
ing programme Such an evaluation should ideally be randomized
and assess whether the training programme is effective and if so
which component(s) of the programme work best
Implications for research
Given theresults of this review which found littlereliable evidence
about the effectiveness of motorcycle rider training it is necessary
to build up an evidence base of the effectiveness of motorcycle
rider training in the reduction of crash injury death and offence
rates To this end well designed randomised controlled trials need
to be conducted that examine the impact of rider training oncrashes and offences A randomised controlled trial is the preferred
method as it would significantly reduce most biases described in
this review To minimise the impact of participants delaying or
never attaining their licence and therefore exposure as found in
the randomised trials in this review such a trial should include
riders who are already licensed and reliable measures of riding
exposure in a reasonably lengthy follow up period It is expected
that a randomised controlled trial of sufficient sample size would
observe equal balance on possible confounders by study arm how-
ever an assessment of potentially important confounders (such as
helmet wearing) by study group would enable control for these
variables in the analyses if necessary It is imperative that the effec-
tiveness of each component of the rider training (such as theorypractice knowledge skills and modification of behaviour) and the
synergy of such components are evaluated Outcomes should be
objective ideally those that are routinely collected rather than self-
reported outcomes Even though riding the motorcycle poses dif-
ferent challenges and incurs different risks lessons may be learned
from research into the effects of car driver training
Motorcycles are a common means of transport in low- and middle
income countries and usage is increasing As the burden of mo-
torcycle crash victims is mostly borne in such countries research
should be focused on the effectiveness of types of rider training on
the reduction of crashes injuries and fatalities in such countries
and be supported by higher income countries This review found
only one study conducted in a middle income country that sug-gested decreasing injury rates It would be beneficial to conduct
additional research to confirm this suggestion and to enhance our
knowledge of the role of rider training in low- and middle-income
settings
March 2006 saw the launch of the University of North Carolina
Highway Safety Research Centre partnership with the Motorcycle
Safety Foundation (MSF) and plans were revealed to conduct a
study of the potential benefits of motorcycle rider training The
study is funded by the MSF and the National Highway Traffic Sa-
fety Administration (NHTSA) Called rsquoThe Discovery Projectrsquo it
will compare data on crash rates from riders participating in a basic
23Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2650
motorcycle safety class with riders participating in a basic motor-
cycle safety class as well as a variety of continuing education classesThe collection of pilot data was due to start in March 2007 and
the data collection portion of the study is expected to take three
years (Highway Safety Research Center) It is vital that researchers
and policy makers collaborate to develop furtherhigh quality eval-
uations of motorcycle rider training programmes to help inform
development of new and effective training programmes
A C K N O W L E D G E M E N T S
Staff from The University of Sydney library for helping us locate
and retrieve full reports of many papers
R E F E R E N C E S
References to studies included in this review
Adams 1985 published data only
Adams A Collingwood V Job RFS Evaluation of the 1982 Australia
Post motorcycle rider training scheme NSW Australia Traffic
Accident Research Unit Traffic Authority NSW 1985 [ RN 85]
Anderson 1980 published data only
Anderson J Ford JL Peck RC Improved motorcyclist licensing and testing project Washington DC National Highway Traffic Safety
Administration (NHTSA) USA 19801ndash217 [
DOTndashHSndash805ndash598]
Billheimer 1998 published data only
Billheimer JW Evaluation of the California Motorcyclist Safety
Program Transportation Research Record 19981640100ndash9 [
TRB Report Number 980652]
Cooper 1988 published data only
Cooper PJ Rothe JP Motorcyclists who they are and why they do
what they do Transportation Research Record 1988116878ndash85
Davis 1997 published data only
David CF Evaluation of community traffic safety programs and
motorcycle operator training programs Connecticut Department of Transportation 199763
Hall 1988 published data only
Hall CJ Buchanan L Engle R Motorcycle rider education evaluationproject Washington DC USA National Highway Traffic Safety
Administration (NHTSA) US Department of Transportation
1988 [ DTNH 22ndash80ndashCndash05123]
Haworth 1997 published data only
Haworth N Smith R Brumen I Pronk N In Monash University
Accident Research Centre editor(s) Case-control study of motorcycle
crashes Victoria Australia Department of Transport and Regional
Development The Federal Office of Road Safety 19971ndash149 [
ISBN 0 642 25501 6 ]
Hurt 1981 published data only
Hurt H Ouellet J Thom D Motorcycle accident cause factors and
identification of countermeasures USA Department of
Transportation 1981 [ Technical Report DOT HS 806ndash862]
Jonah 1982 published data onlylowast Jonah BA Dawson N Bragg B Are formally trained motorcyclists
safer Accident Analysis amp Prevention 198214(4)247ndash55
Jonah BA Dawson NE Bragg BW Evaluation of the Canada
Safety Councilrsquos Motorcycle Training Program Proceedings of theInternational Motorcycle Safety Conference Washington DC
USA 198059ndash93
Kloeden 1994 published data only
Kloeden CN Moore VM McLean AJ Evaluation of the pre-license
training program for motorcyclists in South Australia Walkerville
South Australia Australia NHMRC Road Accident Research Unit
The University of Adelaide 19941ndash37 [ ISBN 0 7308 0607 3]
Kraus 1975 published data only
Kraus JF Riggins RS Franti CE Some epidemiological features of
motorcycle collision injuries American Journal of Epidemiology 1975102(1)74ndash98
Leung 1987 published data only
Leung KS Reding VA Evaluation of the Wisconsin motorcycle rider course Wisconsin Bureau of Policy Planning and Analysis
Wisconsin Department of Transportation 1987
MAIDS 2004 published and unpublished data
MAIDS In-depth investigations of accidents involving powered
two wheelers ACEM (Acemacemeu) 2004 Vol Final report
20179
McDavid 1989 published data only
McDavid JC Lohrmann BA Lohrmann G Does motorcycle
training reduce accidents Evidence from a longitudinal quasi-
experimental study Journal of Safety Research 198920(2)61ndash72
Mortimer 1984 published data only
Mortimer R Evaluation of the motorcycle rider course Accident
Analysis amp Prevention 198416(1)63ndash71
24Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2750
Mortimer 1988 published data only
Mortimer RG A further evaluation of the motorcycle rider courseJournal of Safety Research 198819(4)187ndash96
Osga 1980 published data only
Osga GA An investigation of the riding experiences of MSF rider
course participants Dissertation South Dakota USA The
University of South Dakota 19801ndash224
Perrino 2002 published data only
Perrino CS Ahmed A Callender A Rozier E Cantwell A Stewart
O The role of Marylandrsquos motorcycle rider course in promoting saver
behaviors and attitudes Baltimore USA National Transportation
Center Morgan State University 20021ndash52
Raymond 1979 published data only
Raymond S Tatum S An evaluation of the effectiveness of the RAC
ACU motorcycle training scheme - final report Road Safety ResearchUnit Dept of Civil Engineering University of Salford London
19791ndash131
Satten 1980 published data only
Satten RS Analysis and evaluation of the motorcycle rider courses
in thirteen Illinois counties International Motorcycle Safety
Conference Washington DC 1980145ndash93
Savolainen 2007 published and unpublished data
Savolainen Peter Mannering Fred Additional evidence on the
effectiveness of motorcycle training and motorcyclistsrsquo risk-taking
behaviour Transportation Research Board 2007 Vol March 20
19
Swaddiwudhipong 1998 published data only
Swaddiwudhipong W Boonmak C Nguntra P Mahasakpan P
Effect of motorcycle rider education on changes in risk behavioursand motorcycle-related injuries in rural Thailand Tropical Medicine
and International Health 19983(10)767ndash70
Waller 1992 published data only
Waller FL An analysis of the impact of the Texas Department of
Public Safety Motorcycle Operator Training Program upon
motocycle related traffic accidents Texas State University Public
Administration Program Applied Research Projects 199277
References to studies excluded from this review
Anderson 1980a published data only
Anderson LC Better biking a motorcycle rider improvement
program International Motorcycle Safety Conference Washington
DC USA Motorcycle Safety Foundation 1980231ndash53Aupetit 2008 published and unpublished data
Aupetit S Riff J Buttelli O Espie S Ergonomics research program
on motorcycle learning Contributions to the development of a
motorcycle simulator for training Transport Research Arena
Ljubljana 20089
Awane 1999 published data only
Awane T Integrating simulators in motorcycle safety education
Journal of International Association of Traffic and Safety Sciences 199923(1)26ndash35
Baldi 2005 published data only
Baldi S Baer JD Cook AL Identifying best practices states in
motorcycle rider education and licensing Journal of Safety Research200536(1)19ndash32
Batchler 1988 published data only
Batchler MW Evaluation of the pre-license motorcycle training course at the New South Wales traffic education centre Federal Office of
Road Safety (FORS) 19881ndash122 [ Report No CR 77 ]
Blanchard 2006 published data only
Blanchard HT Tabloski PA Motorcycle safety educating riders at
the teachable moment Mosby (CV) Company 200632330ndash2
Bonnett 2005 published data only
Bonnett J Mature age skills training for experienced motorcycle
riders Motorcycle ridersrsquo association of the Australian Capital
Territory 2005
Braver 2007 published data only
Braver ER et alPersuasion and licensure a randomized controlled
intervention trial to increase licensure rates among Maryland
motorcycle owners Traffic Injury Prevention 20078(1)39ndash46Clayton 1980 published data only
Clayton AB Sudlow DE An evaluation of the effectiveness of a one day training programme for learner motorcyclists Berkshire UK
Transport and Road Research Laboratory 19801ndash23 [ ISSN
0266ndash7045]lowast Clayton AB Sudlow DE An evaluation of the effectiveness of a one-day training programme for learner motorcyclists Berkshire UK
Road Safety Division Berkshire 19801ndash23 [ISSN 0266ndash7045]
Council 1978 published data only
Council FM Desper LP Dutt AK Driver education for motorcycle operation University of North Carolina Highway Safety Research
Center 19781ndash40
Crowther 2005 published data onlylowast Crowther G Engaging with motorcyclists UK police and theBikeSafe Road Safety Programme Published proceedings of the
Australasian Road Safety Research Policing Education Conference
Wellington New Zealand 2005 issue ISBN 0ndash473ndash10636ndash11ndash11
Daniello 2009 published data only
Daniello A Gabler HC Mehta YA Effectiveness of motorcycle
training and licensing Transportation Research Board 88th Annual
Meeting 200913
Dewanti 2007 published data only
Dewanti M Challenging efforts in promoting young motorcyclist
safety in Indonesia Road Safety on Four Continents 14th
International Conference Swedish National Road and Transport
Research Institute 2007
Du Plooy 1990 published data onlyDu Plooy DW Motorcycle training in South Africa The Human
Element International Motorcycle Safety Conference Orlando
Florida USA Motorcycle Safety Foundation 19904-1 to 4-9
Engel 1989 published data only
Engel U Krogsgaringrd Thomsen L [Faeligrdselsundervisning og
propaganda ndash en faeligrdselssikkerhedsfremmende foranstaltning]
Dansk Vejtidsskrift 198912295ndash7
Goldenbeld 2002 published data only
Goldenbeld C Houwing S de Craen S The development of
driving skills of young moped riders [De ontwikkeling van de
rijvaardigheid van jonge bromfietsers] SWOV Institute for Road
Safety Research (Stichting Wetenschappelijk Onderzoek
Verkeersveiligheid) Leidschendam the Netherlands 200263
25Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2850
Goldenbeld 2004 published data only
Goldenbeld C Twisk D de Craen S Short and long term effects of moped training a field experiment Transportation Research Part F 20047(1)1ndash16
Haworth 2000 published data only
Haworth N Smith R Kowadlo N Motorcycle training in Victoria
evaluation of rider training curriculum in Victoria Monash
University Accident Research Centre Reports 2000 Vol 165 issue
3131
Haworth 2005 published data only
Haworth N Mulvihill C Review of motorcycle licensing and
training Monash University Accident Research Centre 2005 [
Report no240]
Haworth 2006 published data only
Haworth N Mulvihill C Rowdon P Teaching old dogs new tricksTraining and older motorcyclists Journal of the Australasian College of Road Safety 200718(4)20ndash5lowast Haworth NL Mulvihill C Rowden PJ Teaching old dogs new
tricks Training and older motorcyclists Australasian Road Safety
Research Policing and Education Conference Gold Coast
Queensland 2006
Hill 1983 published data only
Hill D Motorcycle rider training testing and licensing Motor
Cycle Council of New South Wales Sydney 1983
Imatake 1980 published data only
Imatake Y On the activities in Japan of motorcycle safe riding
promotion committee (MSRPC) - especially the education of
unskilled riders of class 1 motorbikes International Motorcycle
Safety Conference Washington DC USA Motorcycle Safety
Foundation 1980255ndash68
Ingebritsen 1990 published and unpublished data
Ingebrigtsen S [Risikofaktorer ved ferdsel med moped og
motorsykkel] TOslashI-rapport 66 Oslo Transportoslashkonomisk
institutt 1990
Jain 2009 published and unpublished data
Jain A et alTwo wheeler accidents on Indian roads a study from
Mangalore India Elsevier 200916130ndash3
Jonah 1979 published data only
Jonah BA Dawson NE Validation of the motorcycle operator skill
test Accident Analysis amp Prevention 197911(3)163ndash71
Jonah 1981 published data onlyJonah BA Dawson NE Bragg WE Predicting accident
involvement with the Motorcycle Operator Skill Test Accident Analysis amp Prevention 198113(4)307ndash418
Kadar 1990 published data only
Kadar KJ Countermeasures motorcycle rider education delivery
system The Human Element International Motorcycle Safety
Conference Motorcycle Safety Foundation 19906-27 to 6-41
Kelsey 1986 published data only
Kelsey SL Liddicoat C Ratz M Licensing novice motorcyclists a
comparison of Californiarsquos standard test and the MOST II (Motorcycle
Operator Skill Test) administered at centralized testing offices Sacramento California USA California Department of Motor
Vehicles Research and Development Office 19861ndash74
Koch 1980 published data only
Koch H Advanced course for motorcycle riders InternationalMotorcycle Safety Conference Washington DC USA Motorcycle
Safety Foundation 1980269ndash81
Koch 1990 published data only
Koch H Brendicke R How to influence risk taking behaviour
within motorcycle rider training programs The Human Element
International Motorcycle Safety Conference Orlando Florida
USA Motorcycle Safety Foundation 1990Motorcycle Safety
Foundation 1990 p 9-1 to 9-25A
Labbett 2005 published data only
Labbett S Langham M Training can make the problem worse
Road Safety Congress 20051ndash12
Law 2005 published and unpublished data
Law TH Radin Umar RS Zulkaurnain S Kulanthayan S Impactof the effect of economic crisis and the targeted motorcycle safety
programme on motorcycle-related accidents injuries and fatalities
in Malaysia International Journal of Injury Control and Safety
Promotion 200512(1)9ndash21
Lowes 1990 published data only
Lowes BF In-traffic rider training The Human Element
International Motorcycle Safety Conference Orlando Florida
USA Motorcycle Safety Foundation 19907-46 to 7-57
Mahon 1976 published data only
Mahon J Teaching accident avoidance on the motorcycle Journal of Traffic Safety Education 1976231ndash29
Maxwell 1985 published data only
Maxwell GS Evaluation of a Program for Training MotorcycleRiders Evaluation and Program Planning 19858(4)339ndash47
Mayhew 1996 published data only
Mayhew DR Simpson SH Effectiveness and role of driver educationand training in a graduated licensing system Ottowa Canada The
Traffic Injury Research Foundation 19961ndash89 [ ISBN
0ndash920071ndash12ndash0]
McKnight 1987 published data only
McKnight AJ Milburn BG An evaluation of the modified
motorcycle operator skill test exercises NHTSA Report 1987 [
Report No DOT HS 807 212 ]
McPherson 1976 published data only
McPherson K McKnight AJ The development and evaluation of a
motorcycle skill test manual and knowledge test Washington DCUSA National Highway Safety Administration 1976
Mohan 2004 published data only
Mohan D Evidence-based interventions for road traffic injuries in
South Asia Journal of the College of Physicians and SurgeonsPakistan 200414(12)746ndash7
Morgan 1980 published data only
Morgan JK Advanced motorcycle training course content and
study evaluation International Motorcycle Safety Conference
Washington DC USA Motorcycle Safety Foundation 1980
283ndash312
Motoki 1990 published data only
Motoki M Yamazaki S A study on effective motorcycle rider
education The Human Element International Motorcycle Safety
26Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2950
Conference Orlando Florida USA Motorcycle Safety
Foundation 19909-26 to 9-57A
Nairn 1992 published data onlylowast Nairn RJ Motorcycle safety research literature review 1987 to
1991 Federal Office of Road Safety Australia 1993 Vol Report
CR 117
Newland 1980 published data only
Newland R Darby J Automotive practicespre motorcycle rider
program International Motorcycle Safety Conference Washington
DC USA Motorcycle Safety Foundation 1980127ndash9
Newland 1982 published data only
Newland R Motorcycle rider training and licensing requirements
Driver training steering a course for the future Melbourne
Victoria Transport Regulation Board 1982
Newland 1990 published data only
Newland R Motorcycle training and licensing in Victoria 1980-
1990 The Human Element International Motorcycle Safety
Conference Orlando Florida USA Motorcycle Safety
Foundation 1990
Ormston 2003 published data only
Ormston R Dudleston A Pearson S Stradling Evaluation of
Bikesafe Scotland Scottish Executive Social Research Edinburgh
2003
Osga 1979 published data only
Osga GS Ellingstad VS An evaluation of the effectiveness of the 1978
South Dakota motorcycle rider course program South Dakota USA
University of South Dakota 19791ndash89
Panou 2007 published data only
Panou M Bekiaris E A new concept on the integration of driving
simulators in driving training the train-all approach Road Safety
on Four Continents 14th International Conference Swedish
National Road and Transport Research Institute 2007
Peterson 1990 published data only
Peterson A Kadar K Status report 1980-1989 Minnesota rider
education statistics The Human Element International
Motorcycle Safety Conference Orlando Florida USA 199013-
113 to 13-129
Prem 1984 published data only
Prem H Good MC Motorcycle rider skills assessment ACT
Australia Office of Road Safety Department of Transport 1984
1ndash377 [ Report No CR 34 ]
Radin Umar 2006 published data only
Radin Umar RS Motorcycle safety programmes in Malaysia how
effective are they International Journal of Injury Control and Safety Promotion 200613(2)71ndash9
Rockwell 1990 published data only
Rockwell TH Kiger S Carnot M An evaluation of the Ohio
motorcyclist enrichment program Ohio Department of Highway
Safety 1990
Rothe 1987 published data only
Rothe JP A motorcycle riding course Motorcyclists Image and reality Canada Insurance Corporation of British Columbia 1987
59ndash83
Rowden 2007 published data only
Rowden P Watson B Haworth N What can riders tell us aboutmotorcycle training A view from the other side of the fence
Australasian Road Safety Research Policing and Education
Conference Melbourne Australia Melbourne 2007
RSC Victoria 1993 published data only
Road Safety Committee Inquiry into Motorcycle Safety in Victoria
Parliament of Victoria 1993
Russam 1979 published data only
Russam K Motorcycle training and licensing in Japan UK Transport
and Road Research Laboratory (TRRL) 1979 [ Report 916 ]
Saffron 1981 published data only
Saffron DG Driver and rider education training and licensing a
brief review Traffic Accident Research Unit Traffic Authority of
New South Wales NSW Australia Traffic Accident Research UnitTraffic Authority of New South Wales 1981
Schulz 1990 published data only
Schulz U Motorcycle accidents among learner riders Motorcycle
licensing Evaluation of the effectiveness of a Graduated Licensing Scheme Forshungshefte des Institut fur Zweiradsicherheit 1990
Sexton 2004 published data only (unpublished sought but not used)
Sexton B Baughan C Elliott M Maycock G The accident risk of
motorcyclists 2004TRL Report TRL 607(ISSN 09684107)58
Shephard 1986 published data only
Shepherd R Do Motorcycle Safety Foundation programs work
Driver 198619(9)16ndash8
Simpson 1990 published data onlySimpson HM Mayhew DR The promotion of motorcycle safety
training education and awareness Health Education Research
Theory and Practice 19905(2)257ndash64
Sokolov 1990 published data only
Sokolov O Safe motorcycle riding some aspects of motorcycle
riding in the USSR The Human Element International
Motorcycle Safety Conference Motorcycle Safety Foundation
19904ndash37 - 4-48
Steffens 1988 published data only
Steffens U Gawatz R Willmes G Effectiveness of mofa courses
[Wirksamkeit von Mofakursen] Unfall- und sicherheitsforschung
strasenverkehr 198867168
Sudlow 2003 published data only
Sudlow D Scoping study on motorcycle training Road Safety
Research Report London UK The British Institute of Traffic
Education Research Department for Transport 2003 Vol June
[ Report No 36]
Tasmania 1995 published data only
Mannering FL Grodsky LL Motorcycle rider training scheme
Traffic Education Research Centre Armidale Australia Driver
Training Australia Limited 1995 Vol April1ndash56
Tetard 1985 published data only
Tetard C Motorcycle ridersrsquo training to detect indices in an
intersection situation [Lrsquoentrainement a la prise drsquoinformation dans
lrsquoapprentissage de la conduite motocycliste] Le Travail Humain198548(2)147ndash59
27Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3050
Thackray 1980 published data only
Thackray RM Prescott JC Field test of the motorcycle ridercourse International Motorcycle Safety Conference Washington
DC USA Motorcycle Safety Foundation 1980195ndash7
Thompson 1994 published data only
Thompson M Evaluation of compulsory basic training for
motorcyclists Transport Research Laboratory Crowthorne
Berkshire UK 1994 [ Report 63]
Tobler 1990 published data only
Tobler U Individual driving instruction for motorcycle driving
learners the Swiss experiment of motorcycle driving school The
Human Element International Motorcycle Safety Conference
Orlando Florida USA Motorcycle Safety Foundation 1990
Transport 1989 published data only
Transport Communications Road Crash Statistics Australia Bureauof Transport and Communications Economics 198914ndash20
Tuan 2007 published data only
Tuan VA Shimizu T Towards development and evaluation of the
motorcycle drivers re-education program in Vietnam modeling of
motorcycle driverrsquos undesired behaviors 11th World Conference
on Transport Research 200725
Victor 1980 published data only
Victor DJ Traffic safety education for motorcyclists International
Motorcycle Safety Conference Washington DC USA Motorcycle
Safety Foundation 1980221ndash8
Waterer 1990 published data only
Waterer TH Compulsory training for motorcyclists from liberty
to compulsion The Human Element International Motorcycle
Safety Conference Orlando Florida USA Motorcycle Safety
Foundation 19904-49 to 4-57
Weller 1989 published data only
Weller RB Chandler EW Motorcycle safety and motorcycle
education past research and survey results Journal of Public Policy
amp Marketing 1989893ndash108
Wells 1981 published data only
Wells P Observations of some RACACU motorcycle training
schemes Transport and Road Research Laboratory Crowthorne
Berkshire UK 1981
Winn 1986 published data only
Winn G Two important trends in motorcycle safety regulations
rider education and conspicuity improvement American
Motorcyclist Association 1986147ndash52
Wisher 1988 published data only
Wisher G Reid J Motorcycle accidents in Norfolk a three-year study of motorcycle training and the Part 1 test Norfolk UK Road Safety
Division Highways Department Norfolk County Council 1988
Wood 1987 published data only
Wood HT Bowen R Evaluation of the revised learner permit scheme
July 1983 to December 1985 Victoria Australia Road Traffic
Authority 19871ndash27 [ ISBN 0 7306 0247 8]
Yamazaki 1990 published data only
Yamazaki S Nakanishi M Survey on effect of Honda rider
education The Human Element International Motorcycle Safety
Conference Orlando Florida USA 199012-167 to 12-187
References to ongoing studies
Burgess 2010 unpublished data only
Burgess C Broughton P Fylan F Stradling Interim evaluation of
the UKs National Ride program In L Dorn editor(s) Driver
behaviour and Training Vol IV Aldershot Ashgate Publishing
Limited 2010
Additional references
Anderson 1980
Anderson J Ford JL Peck RC Improved motorcyclist licensing and
testing project Volume I and II National Highway Traffic Safety
Administration US department of Transportation Washington
DC 1980
Bjornstig 1985
Bjornstig U Bylund P Lekander T Brorsson B Motorcycle fatalitiesin Sweden Acta Chirurgica Scandinavica 1985151(7)577ndash81
Clayton 1990
Clayton A Sudlow DE An evaluation of the effectiveness of a one-
day training program for learner motorcyclists The Human
Element International Motorcycle Safety Conference Orlando
Florida USA 199013-105 to 13-111
Cochrane Collaboration 2008
Higgins JP Green S Cochrane Handbook for Systematic Reviews of
Interventions 500 The Cochrane Collaboration 2008
Elliot 2003
Elliott MA Baughan CJT Broughton J Grayson G Knowles J et
alMotorcycle Safety a Scoping Study TRL [ Report No
0968ndash4107]Fleiss 2004
Fleiss JL Levin B Paik MC Statistical methods for rates and proportions 3rd Edition John Wiley amp Sons Inc 2004 [ ISBN
978ndash0ndash471ndash52629ndash2 ]
FORS 1999
FORS Federal Office of Road Safety Road risk for sober licensed
motorcyclists Vol Monograph 27 Federal Office of Road Safety
19991ndash2
Haworth 2005
Haworth N Mulvihill C Review of motorcycle licensing and training
Victoria Australia Monash University Accident Research Centre
20051ndash83 [ Report No 240]
Huwiler-Muumlntener 2002Huwiler-Muumlntener K Juumlni P Junker C Egger M Quality of
Reporting of Randomized Trials as a Measure of Methodologic
Quality JAMA 20022872801ndash4
Johnston 2008
Johnston P Brooks C Savage H Fatal and serious road crashes
involving motorcyclists Road Safety Research and Analysis Report
ATSB Monograph 20 Canberra Department of Infrastructure
Transport Regional Development and Local Government 0205
081ndash26
Juumlni 2001
Juumlni P Altman DG Egger M Systematic reviews in health care
Assessing the quality of controlled clinical trials BMJ 2001323
42ndash6
28Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3150
Kopits 2003
Kopits E Cropper M Traffic fatalities and economic growth TheWorld Bank Development Research Group Infrastructure and
Environment Policy research working paper 3035 2003
Leung 1983
Leung P Chau BTW Motorcycle accidents in Hong Kong pattern
of injury treatment causative factors and socio-economic
implications Injury 198315(1)23ndash9
Liu 2004
Liu B Ivers R Norton R Blows S Lo SK Helmets for preventing
injury in motorcycle riders Cochrane Database of Systematic Reviews 2004 Issue 2 [DOI 10100214651858CD004333pub3]
Mohan 2002
Mohan D Road safety in less-motorized environments future
concerns International Journal of Epidemiology 200231(3)527ndash32Mortimer 1982
Mortimer RG An evaluation of the motorcycle rider course
Proceedings of the 20th conference of the American Association for
Automotive Medicine Arlington Heights IL USA American
Association for Automotive Medicine 198277ndash96
Mullin 1997
Mullin BT Risk factors for motorcycle injury the role of age
gender experience training and alcohol PhD thesis Auckland
New Zealand University of Auckland 1997
Noordzij 2001
Noordzij PC Forke E Brendicke R Chinn BP Integration of needs
of moped and motorcycle riders into safety measures review and
statistical analysis in the framework of the European research
project PROMISING SWOV Institute for Road Safety Research
Leidschendam The Netherlands 2001 [ Report No Dndash2001ndash5]
Osga 1979
Osga G Ellingstad VS An evaluation of the effectiveness of the 1978 South Dakota Motorcycle Riders Course Program South Dakota
University of South Dakota Vermillion Human Factors Lab 1979
Oxford English Dictionary Online
httpoxforddictionariescom Oxford English Dictionary Online
Oxford University Press 2010
Peden 2004
Peden M Scurfield R Sleet D World Report on Road TrafficInjury Prevention Geneva World Health Organization Available
at httpwwwwhointviolence˙injury˙preventionpublications
road˙trafficworld˙reporten 2004Chapter 3
Radin Umar 1995
Radin Umar R Macakay G Hills B Preliminary analysis of
motorcycle accidents Short-term impacts of the running
headlights campaign and regulations in Malaysia J Traffic Med 199523(1)17ndash27
Tsai 1995
Tsai YJ Wang JD et alCase-control study of the effectiveness of
different types of helmets for the prevention of head injuries among
motorcycle riders in Taipei Taiwan American Journal of
Epidemiology 1995142(9)974ndash81
Vis 1995
Vis AA In-depth study of the hazards of motorcycling [De
onveiligheid van motorrijden nader bekeken] SWOV Institute for
Road Safety Research Leidschendam the Netherlands 1995 [
Report No Rndash95ndash69]
Weaver 1990
Weaver EA Motorcycle safety education in the United States
Motorcycle Safety Foundation Orlando Florida USA 1990
Winn 1987
Winn G Two important trends in motorcycle safety regulations
rider education and conspicuity improvement No more
information available Detroit MI US Society of Automotive
Engineers 1987147ndash52
Yuan 2000
Yuan W The effectiveness of the rsquoride-brightrsquo legislation for
motorcycles in Singapore Accident Analysis amp Prevention 200032
(4)559ndash63
Zambon 2006
Zambon F Hasselberg M Factors affecting the severity of injuries
among young motorcyclists a Swedish nationwide cohort study
Traffic Injury Prevention 20067(2)143ndash9lowast Indicates the major publication for the study
29Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3250
C H A R A C T E R I S T I C S O F S T U D I E S
Characteristics of included studies [ordered by study ID]
Adams 1985
Methods Non-randomised trial
Participants Initial sample size
Total of 162 Australia Post telegram delivery riders
Not clear how many in intervention and control groups
Analysis sample size
Intervention group 63Control group 35
Total 98 (61)
Interventions No formal training vs voluntary rider training course of 12 hours over 2 consecutive days (theory 4 hours practice
8 hours)
Intervention integrity
No information given
Outcomes Australia Post recorded crashes
Length of follow up 1 year
Notes Study conducted in NSW Australia
Anderson 1980
Methods Randomised trial
Participants Initial sample size
Total 41188 motorcycle licence applicants
Analysis sample size
Intervention group with skills training 13160
Intervention group without skills training 12634
Control group 15080
Total 40874 (99)
Interventions Standard licensing skill test vs the Motorcycle Operators Skill Test (MOST) with or without a training component
MOST is not to be confused with rider training because it is a licensing test The training component was for
participants who failed the skill test on the first attempt was of 3 hours duration and consisted of instructions about
the skills needed to pass the skill test The training consisted of 30 min theory and 2 frac12 hours practice on the
motorcycle
Intervention integrity
33 in MOST with skills training 31 in MOST without skills training and 43 in control group obtained their
motorcycle licence in the end
Outcomes Police recorded crashes and offences
Length of follow up 3 years
30Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3350
Anderson 1980 (Continued)
Notes Study conducted in California USA
Billheimer 1998
Methods Cohort study
Participants Initial sample size
Intervention group Unclear numbers of CMSP graduates and ERC graduates
Control group Over 16000 untrained motorcyclists
Analysis sample size
1139 matched CMSP-pairs and 1182 ERC-pairs
Interventions No formal training vs the mandatory (for those younger than 18 years) pre-licence 16-hours basic California
Motorcyclist Safety Program (CMSP) and the 8-hour Experienced Rider Course (ERC)
Intervention integrity
No information given
Outcomes Police recorded crashes and offences
Length of follow up 2 years
Notes Study conducted in California USA
Cooper 1988
Methods Cohort study
Participants Initial sample size
Intervention group 418 graduates and 57 who failed
Control group 402
Analysis sample size
Intervention group unclear
Control group unclear
Total 863 (984)
Interventions No formaltrainingvs non-mandatory pre-licence rider trainingThe content of thetrainingwas notfurther explainedIntervention integrity
Participants had all completed the rider training
Outcomes Police recorded crashes and offences
Length of follow up 7 years
Notes Study conducted in British Columbia Canada
31Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3450
Davis 1997
Methods Retrospective cohort study
Participants Initial sample size
Intervention group 9320 graduates
Control group unclear
Analysis sample size
Intervention group 9320 graduates
Control group unclear
Interventions No formal training vs the non-mandatory Connecticut Motorcycle Operator Training Program The content of
training was not further explained
Intervention integrity Participants had all completed the rider training
Outcomes Police recorded crashes severity of crashes
Length of follow up 1 year
Notes Study conducted in Connecticut USA
Hall 1988
Methods Randomised trial
Participants Initial sample sizeIntervention group A 6248
Intervention group B 6616
Intervention group C 6996
Control group 6604
Total 26464 motorcycle licence applicants
Analysis sample size
Intervention group A 6012 (96)
Intervention group B 6428 (97)
Intervention group C 6778 (97)
Control group 6502 (98)
Interventions Standard licensing skill test vs the Motorcycle Operator Skill Test (MOST II) without training (A) MOST II with
3 hours training (B) or MOST II with 20 hours training (C) The 3 hours training comprised instructions aboutthe skills needed to pass the skill test (30 min theory and 2 frac12 hours practice on the motorcycle) The 20 hours
training consisted of the Motorcycle Safety Foundationrsquos Motorcycle Rider Course (MRC) comprising of 8 hours
theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-road practice
Intervention integrity
25 and 26 obtained their motorcycle licence in the lsquoMOST 3-hour trainingrsquo and lsquoMOST 20-hour trainingrsquo
intervention arm respectively In the lsquoMOST testrsquo intervention arm 29 obtained their licence and in the control
group 35
Outcomes Police recorded crashes and offences
Length of follow up 3 years
32Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3550
Hall 1988 (Continued)
Notes Study conducted in New York USA
Haworth 1997
Methods Population based case-control study
Participants Cases Motorcycle riders or pillion riders involved in motorcycle crash (on road) Recruited from hospital and police
records Not clear whether all cases (especially fatalities) were included
Community controls Multiple matched sample of motorcycle rider or pillion who passed the crash site at the same
time of day and day of week that the crash occurred
Initial sample size
Cases 336
Controls 1195
Analysis sample size
Cases 234 (70)
Controls 561 (47)
Interventions Basic handling skills course beginner intermediate or advanced course measured through self-report not validated
Outcomes Injuries with an Injury Severity Score of 5 or greater or admission for at least four hours and fatalities from a
motorcycle crash (from hospital and police records)
Length of follow up two years or more prior to the interview
Notes Study conducted in Victoria Australia
Hurt 1981
Methods Population based case-control study
Participants Cases Motorcycle riders who had a motorcycle crash that was reported to police (traffic accident reports) Not all
cases included
Community controls Multiple matched sample of motorcycle riders who passed the crash site at the same time of
day and day of week as the crash occurred but 2 years later
Initial sample size
Cases 4500 notifications of crashes to police (only 50 received in time research team able to respond to 1126 for
unreported reasons data collection completed for 900)
Controls Number approached unreported 2310 consented and interviewed
Analysis sample size
Cases 808 (808900 data on rider training) - 18 of initial sample
Controls 670 (6702310 data on rider training) - lt 29 of initial sample
Interventions Training experience self taught friends-family school-club formal-professional others measured through self-
report validated
Outcomes Injuries and fatalities from a motorcycle crash (from hospital records and coronerrsquos office)
33Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3650
Hurt 1981 (Continued)
Notes Study conducted in Los Angeles USA
Jonah 1982
Methods Cohort study
Participants Initial sample size
Intervention group
2310 licensed MTP graduates who are active riders
Control group
2029 licensed motorcyclists who are active riders
Non-licensed and non-active riders excluded
Analysis sample size
Intervention group
811 interviews (35) 533 driver records (23)
Control group 1080 interviews (71) 905 driver records (45)
Interventions No training vs the non-mandatory Canada Safety Council Motorcycle Training Program (MTP) comprising of 4
hours theory and 16 hours practise (6 hrs on-road) over 2 weekends
Intervention integrity
Participants had all successfully completed the MTP
Outcomes Self-reported and police reported crashes and offencesLength of follow up 4 years
Notes Study conducted in Ontario Canada
Kloeden 1994
Methods Cohort study
Participants Initial sample size
Intervention group 2533 licence applicants
Control group Unclear
Sample of riders obtaining probationary licenceIntervention group 1625 (64)
Control group 5015 (unclear)
Non-active riders excluded from control group (~66)
Analysis sample size
Intervention group 631 (39)
Control group 318 (6)
Interventions No formal training vs the mandatory pre-licence Ridersafe course consisting of two levelsthe first is a training course
in the basics of motorcycle riding where a pass entitles the rider to obtain a learnerrsquos permit The second is a skill
test where a pass entitles the rider to obtain a probationary motorcycle licence The training course consists of two 4
hour training sessions on separate days involving instruction videos and actual riding on motorcycles The skill test
consists of a 4 hour preparation course of 74 minutes of lecturediscussion 38 minutes of video and 75 minutes of
34Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3750
Kloeden 1994 (Continued)
on-course riding which is the testing phase
Intervention integrity
Unclear what percentage in the Ridersafe group and the control group obtained their probationary licence halfway
and their full motorcycle licence in the end
Outcomes Police reported crashes
Length of follow up 1 year
Notes Study conducted in SA Australia
Kraus 1975
Methods Population based case-control study
Participants Cases All motorcycle riders with a confirmed medically treated motorcycle injury (including fatalities) from all
hospitals in study base and peripheral area hospitals and all deaths due to motorcycle crashes in the study base All
serious injuries (leading to hospitalisation) and fatalities likely to be included
Community controls A random sample of registered motorcycle owners in the same temporary and geographical
study base as the cases arose from
Initial sample size Cases 1273
Controls 738
Analysis sample size Cases 741 (58)
Controls 434 (59)
Interventions Motorcycle rider training (not specified) measured through self-report
Outcomes Injuries leading to hospitalisation fatalities
Length of follow up not specified
Notes Study conducted in Sacramento county USA
Leung 1987
Methods Non-randomised trial
Participants Initial sample sizeIntervention group unknown number of MRC graduates
Control group unknown number of licensed motorcyclists
Analysis sample size
Intervention group 2941
Control group 43094
Interventions No formal training vs non-mandatory Wisconsin motorcycle rider course a 22 hours off-street programme that
teaches the basic rules of safe motorcycling
Intervention integrity
50 in the MRC group and 100 in the control group obtained their motorcycle licence in the end
35Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3850
Leung 1987 (Continued)
Outcomes Police recorded crashes and offences
Length of follow up up to 3 years
Notes Study conducted in Wisconsin USA
MAIDS 2004
Methods Population based case-control study
Participants Cases Motorcycle riders who had a motorcycle crash Unclear whether all cases included
Community controls Motorcycle riders identified at petrol stations Unclear whether matched to cases
Initial sample size
Cases not reported
Controls not reported
Analysis sample size
Cases 921 (Italy 200 Netherlands 200 Spain 121 Germany 250 France 150)
Controls 923 (Italy 200 Netherlands 200 Spain123 Germany 250 France 150)
Interventions Pre-license rider training or additional training (non-compulsory)
Outcomes Crashes
Notes Study in five sampling areas located in five European countries (France Germany Netherlands Spain and Italy)
McDavid 1989
Methods Cohort study
Participants Sample size after first match
Intervention group 346 male training graduates
Control group 346 male motorcycle riders licensed in same year
Sample size after second match
Intervention group 139 (40)
Control group 139 (40)
Analysis sample size
Intervention group 139 (40)
Control group 139 (40)
Interventions No formal training vs the non-mandatory pre-licence British Columbia Safety Councilrsquos 37-hour motorcycle safety
programme
Intervention integrity
Participants had all completed and passed the training
Outcomes Police reported crashes injurious crashes
Length of follow up 1-6 years
36Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3950
McDavid 1989 (Continued)
Notes Study conducted in British Columbia Canada
Mortimer 1984
Methods Cohort study
Participants Initial sample size
Intervention group 608 MRC graduates
Control group 360 motorcycle riders
Response rates
Intervention group 360 (59)
Control group 324 (90)
Non-active riders excluded
Analysis sample size
Intervention group 213 (35)
Control group 303 (84)
Interventions No formal training vs the non-mandatory Motorcycle Safety Foundationrsquos Motorcycle Rider Course comprising of
8 hours theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-road practice
Intervention integrity
Participants had all completed and passed the MRC
Outcomes Self-reported crashes and offencesLength of follow up 1 year
Notes Study conducted in Illinois USA
Mortimer 1988
Methods Cohort study
Participants Initial sample size
Intervention group 2307 MRC graduates
Control group 556 active licensed motorcycle riders
Response ratesIntervention group 1298 (56)
Control group 500 (90)
Non-active riders excluded
Analysis sample size
Intervention group 913 (40)
Control group 500 (90)
Interventions No formal training vs the non-mandatory Motorcycle Safety Foundationrsquos Motorcycle Rider Course comprising of
8 hours theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-road practice
Intervention integrity
Participants had all completed and passed the MRC
37Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4050
Mortimer 1988 (Continued)
Outcomes Self reported crashes and offences
Length of follow up 1 year
Notes Study conducted in Illinois USA
Osga 1980
Methods Cohort study
Participants Initial sample sizeIntervention group 1163 MSF MRC participants invited 818 (74) responded
Control group 329 motorcycle riders responded to recruitment strategies
Non-active riders excluded from analysis
Analysis sample size
Intervention group 595 (73)
Control group 329 (100)
Interventions No formal training vs the non-mandatory Motorcycle Safety Foundationrsquos Motorcycle Rider Course (MRC) com-
prising of 8 hours theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-road
practice
Intervention integrity
Participants had all successfully completed the MRC
Outcomes Self-reported crashes police recorded crash reports
Length of follow up 6 months - 3 12 years
Notes Study conducted in South Dakota USA
Perrino 2002
Methods Cohort study
Participants Initial sample size
Intervention group 2144 (unclear) MRC graduates (motorcycle licence applicants)
Control group 2601 (unclear) regular licensing test graduates (motorcycle licence applicants)Analysis sample size
Intervention group 1070 (unclear)
Control group 867 (unclear)
Interventions No formal training vs the mandatory (for those younger than 18 years) pre-licence Maryland Motorcycle Rider
Course (MRC) consisting of 8 hours theory and 12 hours practice
Intervention integrity
No information given
Outcomes Self-reported and police recorded crashes and offences
Length of follow up 1 year
38Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4150
Perrino 2002 (Continued)
Notes Study conducted in Maryland USA
Raymond 1979
Methods Cohort study
Participants Initial sample size
Intervention group
258 volunteers mainly (but not all) novice riders
Control group
1104 provisional licence applicants
Analysis sample size
Intervention group 190 (74)
Control group 627 (57)
Interventions No formal training vs the non-mandatory RACACU (Royal Automobile ClubAuto-Cycle Union) training scheme
The training program consisted of either 24 one-hour sessions or 13 two-hour sessions Twelve hours of the training
focused on theory about basic operational procedures traffic laws and highway code The practical session focused
on basic operational skills riding on a training area and riding in traffic with an instructor A proficiency test was
administered at the end of the course
Intervention integrity
No information given
Outcomes Self-reported crashes (including minor non-injury non-damage and off-road crashes)
Length of follow up 1 year
Notes Study conducted in the UK
Satten 1980
Methods Cohort study
Participants Initial sample size
Intervention group interviewers kept interviewing until they had 100 riders selected from a population of 891
enrollees in the MSF MRC course (unknown how many were asked and how many refused to participate)Control group unknown number of registered motorcycle riders
Non-active riders excluded from analysis
Analysis sample size
Intervention group 71
Control group 100
Interventions No formal training vs the non-mandatory Motorcycle Safety Foundationrsquos (MSF) Motorcycle Rider Course (MRC)
comprising of 8 hours theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-
road practice
Intervention integrity
Participants had all successfully completed the MRC
39Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4250
Satten 1980 (Continued)
Outcomes Self-reported crashes injuries and offences
Length of follow up 3 years
Notes Study conducted in Illinois USA
Savolainen 2007
Methods Cohort study
Participants Initial sample size
Intervention group 4000 riders that had conducted the MSF MRC course (passed or failed)
Control group 4000 motorcycle owners or permit or endorsement holders from the Indiana Bureau of Motor
Vehicles database
Analysis sample size
Intervention group 739 (18)
Control group 588 (15)
Interventions Intervention integrity
No information given
Outcomes Probability of crash involvement
Length of follow up up to 5 years
Notes Study conducted in Indiana USA
Swaddiwudhipong 1998
Methods Randomised trial
Participants Initial sample size
Intervention group
all 16 villages of 3 sub-districts
(participants were practising motorcyclists licensed and unlicensed)
Control group
18 villages of 3 other sub-districtsAnalysis sample size
Intervention group 3 districts (100)
Control group 3 districts (100)
Interventions No rider education vs non-mandatory community based rider education consisting of theory on motorcycle injury
risk helmet use traffic laws vehicle regulations and traffic signs and a written and skill test leading to a licence
Intervention integrity
No information given
Outcomes Injuries fatalities
Length of follow up 2 years
40Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4350
Swaddiwudhipong 1998 (Continued)
Notes Study conducted in Tak Province rural Thailand
Waller 1992
Methods Cohort study
Participants Initial sample size
Intervention group 349 riders who succesfully completed the Department of Public Safety motorcycle operator
training in 1988
Control group 349 motorcyclists who never completed the DPS motorcycle operator training
Analysis sample size
Intervention group 349
Control group 349
Interventions No DPS motorcycle operator training vs non-mandatory DPS motorcycle operator training The DPS training is
the Motorcycle Safety Foundationrsquos Motorcycle Rider Course comprising eight hours of theory (lectures discussions
films and multiple choice tests) a textbook and 12 hours off-road practice
Intervention integrity
All 349 riders succesfully completed the rider training
Outcomes Police recorded crashes
Length of follow up 3 years
Notes Study conducted in
Characteristics of excluded studies [ordered by study ID]
Study Reason for exclusion
Anderson 1980a Study type review and process evaluation of a rider course
Aupetit 2008 Outcome proposal of a rider training curriculum that combines traditional training and a motorcycle riding
simulator
Awane 1999 Study type review on use of simulators in rider and driver training
Baldi 2005 Intervention not rider training but intervention integrity of rider training availabilty of rider education
implementation of graduated licensing system etc on a state-wide level
Batchler 1988 Study type description of a training course
Blanchard 2006 Study type not a study assessing the association between rider training and outcomes
Bonnett 2005 Outcome qualitative data regarding satisfaction with the training programme
41Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4450
(Continued)
Braver 2007 Outcome licensure rates
Clayton 1980 Outcome knowledge skills attitude
Council 1978 Outcome crash reduction in ANY vehicle not motorcycles exclusively
Crowther 2005 Study type qualitative study
Daniello 2009 Study type review of the effectiveness of motorcycle training and licensing
Dewanti 2007 Intervention emotional characteristics behaviour (ie not motocycle rider training)
Du Plooy 1990 Study type review of types of rider training courses in South Africa
Engel 1989 Study type uncontrolled before-after study
Goldenbeld 2002 Outcome skills
Goldenbeld 2004 Outcome knowledge and skills
Haworth 2000 Intervention balance between attitudinal andvehicle control skills acrosscurriculla of motorcycle rider training
courses
Haworth 2005 Study type review
Haworth 2006 Study type cross-sectional survey
Hill 1983 Study type review and description of training system
Imatake 1980 Study type course descriptionreview
Ingebritsen 1990 Intervention risk factors for motorcycle crashes
Jain 2009 Study type cross-sectional descriptive study of motorcycle crashes over a 5 year period
Jonah 1979 Study type cross-sectional survey Outcome scores on motorcycle operator skill test
Jonah 1981 Intervention evaluation of a motorcycle rider skill test not motorcycle rider training
Kadar 1990 Study type review of components of rider training courses
Kelsey 1986 Intervention evaluation of a motorcycle rider skill test not motorcycle rider training
Koch 1980 Study type reviewcourse description
Koch 1990 Study type description of course content
42Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4550
(Continued)
Labbett 2005 Study type essay about possible explanations for motorcycle deaths
Law 2005 Intervention targeted motorcycle safety programme not specifically rider training
Lowes 1990 Study type review and descriptions of rider training
Mahon 1976 Study type review of motorcycle rider training methods
Maxwell 1985 Outcomesrecommendations forprogramme operatorsand otherson howthe motorcycle trainingprogramme
works and how it can be improved
Mayhew 1996 Study type review
McKnight 1987 Study type evaluation of methodological and administrative aspects of a motorcycle operator skill test
McPherson 1976 Outcome knowledge and skills
Mohan 2004 Study type review of evidence of interventions to reduce road traffic injuries
Morgan 1980 Study type reviewcourse description
Motoki 1990 Study type proposal and development of rider training curriculum
Nairn 1992 Study type literature review
Newland 1980 Study type description of development of motorcycle safety course
Newland 1982 Study type descriptionreview of a rider training course
Newland 1990 Study type uncontrolled BeforeAfter study
Ormston 2003 Outcome changes in attitude and self reported behaviour
Osga 1979 Outcome knowledge attitudes and skills
Panou 2007 Intervention new simulation tool to teach driverrider
Peterson 1990 Study factor demographic details of course participants completing the course vs those not completing the
course
Prem 1984 Study type experimental
Study factor examines association between skills and performance in tests
Radin Umar 2006 Intervention motorcycle safety programme not specifically rider training
Rockwell 1990 Study type methodologically flawed study type
43Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4650
(Continued)
Rothe 1987 Study type qualitative study of rider training in general
Rowden 2007 Study type qualitative study
RSC Victoria 1993 Study type review of motorcycle rider training and education in Victoria
Russam 1979 Study type review of training programmes in Japan
Saffron 1981 Study type literature review
Schulz 1990 Study factor predictors of motorcycle crashes
Sexton 2004 Study type cross-sectional analytic analysis
Shephard 1986 Study factor reviewdescription of programme
Simpson 1990 Study type review of motorcycle rider trainingeducation
Sokolov 1990 Study factor review of motorcycle riding courses in USSR
Steffens 1988 Outcome self-reported behaviour experiences and researcher observed riding behaviour
Sudlow 2003 Study type review of motorcycle training in the UK
Tasmania 1995 Study type cross-sectional descriptiveprocess evaluation
Tetard 1985 Study type experimental
Outcome detection of indices
Thackray 1980 Outcome instructional effectiveness user acceptance administrative feasibility
Thompson 1994 Study type cross-sectional survey
Outcome attitudes and knowledge
Tobler 1990 Study type review of course curricula
Transport 1989 Intervention syllabus for rider training
Tuan 2007 Study type cross-sectional study (survey) which does not assess the effects of an intervention but examines
motocyclistsrsquo behaviour
Victor 1980 Study type review
Waterer 1990 Study type review of UK rider training issues
Weller 1989 Study type review of studies evaluating motorcycle safety and education measures
44Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4750
(Continued)
Wells 1981 Study type description of different methods of training
Winn 1986 Study type review of rsquohistoryrsquo of rider training
Wisher 1988 Study type cross-sectional survey
Wood 1987 Study type Interrupted Time Series but did not meet the criteria for an ITS as published by the Cochrane
Effective Practice and Organisation of Care Group
Yamazaki 1990 Study type cross-sectional survey
Outcome riding skills skills retention and attitude
Characteristics of ongoing studies [ordered by study ID]
Burgess 2010
Trial name or title UKs National Ride program
Methods Cohort study
Participants Motorcycle offenders
Interventions Police led diversion program
Outcomes Self reported attitudinal and behavioural measures
Starting date
Contact information
Notes Expected to be completed in early 2010
45Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4850
D A T A A N D A N A L Y S E S
This review has no analyses
A P P E N D I C E S
Appendix 1 Search strategy
We searched the following electronic databases
bull Cochrane Injuries Group Specialised Register (searched 22 Sept 2008)
bull CENTRAL (The Cochrane Library 2008 Issue 3)
bull TRANSPORT 1988 to 200706 (latest date available)
bull ATRI Australian Transport Index (to Sept 2008)
bull MEDLINE (Ovid SP) 1950 to Sept 2008
bull EMBASE (Ovid SP) 1980 to Sept 2008
bull CINAHL (EBSCO) 1982 to Sept 2008
bull WHOLIS (World Health Organization Library Information System) (to Sept 2008)
bull PsycINFO (Ovid SP) 1806 to Sept 2008
bull LILACS (BIREME) (Latin American and Caribbean Health Sciences) (1982 to Sept 2008)
bull ISI Web of Science Social Sciences Citation Index (SSCI) 1970 to Sept 2008
bull ERIC (searched 25 Sept 2008)
bull ZETOC (searched 25 Sept 2008)
bull SIGLE (searched 25 Sept 2008)
All of the databases were searched using the following text-word search strategy The strategy was adapted using either rsquorsquo or rsquo$rsquo as atruncation symbol
1 motor-cycl or motorcycl or motor-bik or motorbik or motor-scooter or motorscooter or scooter or moped or motocycl
or motocicl or moto
2 train or educat or instruct or program or course or qualif or learn or improv
3 1 and 2
H I S T O R Y
Protocol first published Issue 2 2005
Review first published Issue 10 2010
46Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4950
C O N T R I B U T I O N S O F A U T H O R S
Katina Kardamanidis was the first author of the protocol conducted the literature search extracted full studies assessed the studies
according to inclusion criteria assessed quality of the studies compared findings with the second author discussed disagreements with
the third author wrote the drafts and the final review
Rebecca Ivers was the second author of the protocol assessed the studies according to inclusion criteria assessed quality of the studies
compared findings with the first author edited the final draft and provided overall guidance for the process
Alexandra Martiniuk assessed quality of the studies compared findings with the first author discussed disagreements with the third
author and helped write the first draft and the final review
Mark Stevenson contributed to the protocol reviewed methods of studies and contributed to the final draft
Katrina Thistlethwaite conducted a second literature search extracted full studies and contributed to the final draft
D E C L A R A T I O N S O F I N T E R E S T
None known
S O U R C E S O F S U P P O R T
Internal sources
bull No sources of support supplied
External sources
bull National Health and Medical Research Council of Australia Australia
Salary support for Martiniuk Ivers and Stevenson
bull University of Sydney Sesqui Research Grant Australia
Direct research costs and salary support for Kardamanidis
D I F F E R E N C E S B E T W E E N P R O T O C O L A N D R E V I E W
There are no differences between the protocol and the review
I N D E X T E R M S
Medical Subject Headings (MeSH)
lowastMotorcycles Accident Prevention [lowastmethods] Accidents Traffic [lowastprevention amp control] Licensure Program Evaluation Wounds
and Injuries [mortality lowastprevention amp control]
47Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 5050
MeSH check words
Humans
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2550
Limitations of this review
Limitations of this review were that we found very few recentstudies (the two most recent ones published in 2002 and 2007)
and that we found very few studies conducted in low-and middle
income countries even though our search strategy was set up to
identify such studies Anotherlimitation wasthat forthe reviewed
studies we could not assess all outcomes of rider training that are
potentially useful Some of the included studies assessed outcomes
such as use of safety clothing including helmets Such outcomes
are valuable since the use of motorcycle helmets has been demon-
strated to be effective in preventing head injuries (Liu 2004) De-
spite this the use of safety clothing was not included as an out-
come in this review because the studies measuring this outcome
were limited by poor data quality due to very low response rates
Another outcome of interest in assessing the effectiveness of rider
training is lsquotype of injuryrsquo As Haworth 2005 state in their review
of motorcycle licensing and training rider training could be con-
sidered effective if injuries become less severe because motorcycle
riders have learned how to avoid more serious injuries through
rider training The studies included in this review did not specify
type of injury and so this outcome could not be assessed Last
changes in knowledge and attitudes following rider training are
other potentially useful outcomes We have excluded such studies
(Osga 1979 Clayton 1990 Goldenbeld 2004) as these outcomes
are proxies rather than direct measures of the effectiveness of mo-
torcycle rider training We opted to report on the outcomes of
crashes injury death and offences alone since there is little ev-
idence about how changes in knowledge and attitudes relate to
these outcomes
A U T H O R S rsquo C O N C L U S I O N S
Implications for practice
Due to the poor quality of studies included in this review we were
unable to draw any conclusions about the effectiveness of rider
training on crash rates injurious or fatal or offence rates The
findings do suggest that mandatory pre-licence training may be
an impediment to completing a motorcycle licensing process thuspossibly indirectly reducing crash injury death and offence rates
through a reduction in exposure to riding a motorcycle On the
basis of the existing evidence it is not clear if and what type of
training reduces the risk of crashes injuries deaths or offences in
motorcyclists and the selection of the best rider training practice
can therefore not be recommended The suggestion that manda-
tory pre-licence training indirectly reduces exposure to riding the
motorcycle because trainees are impeded from completing the li-
censing process does not help those motorcyclists who persist in
obtaining their licence and riding the motorcycle Some type of
rider training is likely necessary to teach motorcyclists to ride a
motorcycle safely Good policy is based on sound evidence and
evaluation but such evidence for motorcycle rider training is seri-
ously lacking Policy makers are therefore encouraged to include arigorous evaluation component to any new or existing rider train-
ing programme Such an evaluation should ideally be randomized
and assess whether the training programme is effective and if so
which component(s) of the programme work best
Implications for research
Given theresults of this review which found littlereliable evidence
about the effectiveness of motorcycle rider training it is necessary
to build up an evidence base of the effectiveness of motorcycle
rider training in the reduction of crash injury death and offence
rates To this end well designed randomised controlled trials need
to be conducted that examine the impact of rider training oncrashes and offences A randomised controlled trial is the preferred
method as it would significantly reduce most biases described in
this review To minimise the impact of participants delaying or
never attaining their licence and therefore exposure as found in
the randomised trials in this review such a trial should include
riders who are already licensed and reliable measures of riding
exposure in a reasonably lengthy follow up period It is expected
that a randomised controlled trial of sufficient sample size would
observe equal balance on possible confounders by study arm how-
ever an assessment of potentially important confounders (such as
helmet wearing) by study group would enable control for these
variables in the analyses if necessary It is imperative that the effec-
tiveness of each component of the rider training (such as theorypractice knowledge skills and modification of behaviour) and the
synergy of such components are evaluated Outcomes should be
objective ideally those that are routinely collected rather than self-
reported outcomes Even though riding the motorcycle poses dif-
ferent challenges and incurs different risks lessons may be learned
from research into the effects of car driver training
Motorcycles are a common means of transport in low- and middle
income countries and usage is increasing As the burden of mo-
torcycle crash victims is mostly borne in such countries research
should be focused on the effectiveness of types of rider training on
the reduction of crashes injuries and fatalities in such countries
and be supported by higher income countries This review found
only one study conducted in a middle income country that sug-gested decreasing injury rates It would be beneficial to conduct
additional research to confirm this suggestion and to enhance our
knowledge of the role of rider training in low- and middle-income
settings
March 2006 saw the launch of the University of North Carolina
Highway Safety Research Centre partnership with the Motorcycle
Safety Foundation (MSF) and plans were revealed to conduct a
study of the potential benefits of motorcycle rider training The
study is funded by the MSF and the National Highway Traffic Sa-
fety Administration (NHTSA) Called rsquoThe Discovery Projectrsquo it
will compare data on crash rates from riders participating in a basic
23Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2650
motorcycle safety class with riders participating in a basic motor-
cycle safety class as well as a variety of continuing education classesThe collection of pilot data was due to start in March 2007 and
the data collection portion of the study is expected to take three
years (Highway Safety Research Center) It is vital that researchers
and policy makers collaborate to develop furtherhigh quality eval-
uations of motorcycle rider training programmes to help inform
development of new and effective training programmes
A C K N O W L E D G E M E N T S
Staff from The University of Sydney library for helping us locate
and retrieve full reports of many papers
R E F E R E N C E S
References to studies included in this review
Adams 1985 published data only
Adams A Collingwood V Job RFS Evaluation of the 1982 Australia
Post motorcycle rider training scheme NSW Australia Traffic
Accident Research Unit Traffic Authority NSW 1985 [ RN 85]
Anderson 1980 published data only
Anderson J Ford JL Peck RC Improved motorcyclist licensing and testing project Washington DC National Highway Traffic Safety
Administration (NHTSA) USA 19801ndash217 [
DOTndashHSndash805ndash598]
Billheimer 1998 published data only
Billheimer JW Evaluation of the California Motorcyclist Safety
Program Transportation Research Record 19981640100ndash9 [
TRB Report Number 980652]
Cooper 1988 published data only
Cooper PJ Rothe JP Motorcyclists who they are and why they do
what they do Transportation Research Record 1988116878ndash85
Davis 1997 published data only
David CF Evaluation of community traffic safety programs and
motorcycle operator training programs Connecticut Department of Transportation 199763
Hall 1988 published data only
Hall CJ Buchanan L Engle R Motorcycle rider education evaluationproject Washington DC USA National Highway Traffic Safety
Administration (NHTSA) US Department of Transportation
1988 [ DTNH 22ndash80ndashCndash05123]
Haworth 1997 published data only
Haworth N Smith R Brumen I Pronk N In Monash University
Accident Research Centre editor(s) Case-control study of motorcycle
crashes Victoria Australia Department of Transport and Regional
Development The Federal Office of Road Safety 19971ndash149 [
ISBN 0 642 25501 6 ]
Hurt 1981 published data only
Hurt H Ouellet J Thom D Motorcycle accident cause factors and
identification of countermeasures USA Department of
Transportation 1981 [ Technical Report DOT HS 806ndash862]
Jonah 1982 published data onlylowast Jonah BA Dawson N Bragg B Are formally trained motorcyclists
safer Accident Analysis amp Prevention 198214(4)247ndash55
Jonah BA Dawson NE Bragg BW Evaluation of the Canada
Safety Councilrsquos Motorcycle Training Program Proceedings of theInternational Motorcycle Safety Conference Washington DC
USA 198059ndash93
Kloeden 1994 published data only
Kloeden CN Moore VM McLean AJ Evaluation of the pre-license
training program for motorcyclists in South Australia Walkerville
South Australia Australia NHMRC Road Accident Research Unit
The University of Adelaide 19941ndash37 [ ISBN 0 7308 0607 3]
Kraus 1975 published data only
Kraus JF Riggins RS Franti CE Some epidemiological features of
motorcycle collision injuries American Journal of Epidemiology 1975102(1)74ndash98
Leung 1987 published data only
Leung KS Reding VA Evaluation of the Wisconsin motorcycle rider course Wisconsin Bureau of Policy Planning and Analysis
Wisconsin Department of Transportation 1987
MAIDS 2004 published and unpublished data
MAIDS In-depth investigations of accidents involving powered
two wheelers ACEM (Acemacemeu) 2004 Vol Final report
20179
McDavid 1989 published data only
McDavid JC Lohrmann BA Lohrmann G Does motorcycle
training reduce accidents Evidence from a longitudinal quasi-
experimental study Journal of Safety Research 198920(2)61ndash72
Mortimer 1984 published data only
Mortimer R Evaluation of the motorcycle rider course Accident
Analysis amp Prevention 198416(1)63ndash71
24Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2750
Mortimer 1988 published data only
Mortimer RG A further evaluation of the motorcycle rider courseJournal of Safety Research 198819(4)187ndash96
Osga 1980 published data only
Osga GA An investigation of the riding experiences of MSF rider
course participants Dissertation South Dakota USA The
University of South Dakota 19801ndash224
Perrino 2002 published data only
Perrino CS Ahmed A Callender A Rozier E Cantwell A Stewart
O The role of Marylandrsquos motorcycle rider course in promoting saver
behaviors and attitudes Baltimore USA National Transportation
Center Morgan State University 20021ndash52
Raymond 1979 published data only
Raymond S Tatum S An evaluation of the effectiveness of the RAC
ACU motorcycle training scheme - final report Road Safety ResearchUnit Dept of Civil Engineering University of Salford London
19791ndash131
Satten 1980 published data only
Satten RS Analysis and evaluation of the motorcycle rider courses
in thirteen Illinois counties International Motorcycle Safety
Conference Washington DC 1980145ndash93
Savolainen 2007 published and unpublished data
Savolainen Peter Mannering Fred Additional evidence on the
effectiveness of motorcycle training and motorcyclistsrsquo risk-taking
behaviour Transportation Research Board 2007 Vol March 20
19
Swaddiwudhipong 1998 published data only
Swaddiwudhipong W Boonmak C Nguntra P Mahasakpan P
Effect of motorcycle rider education on changes in risk behavioursand motorcycle-related injuries in rural Thailand Tropical Medicine
and International Health 19983(10)767ndash70
Waller 1992 published data only
Waller FL An analysis of the impact of the Texas Department of
Public Safety Motorcycle Operator Training Program upon
motocycle related traffic accidents Texas State University Public
Administration Program Applied Research Projects 199277
References to studies excluded from this review
Anderson 1980a published data only
Anderson LC Better biking a motorcycle rider improvement
program International Motorcycle Safety Conference Washington
DC USA Motorcycle Safety Foundation 1980231ndash53Aupetit 2008 published and unpublished data
Aupetit S Riff J Buttelli O Espie S Ergonomics research program
on motorcycle learning Contributions to the development of a
motorcycle simulator for training Transport Research Arena
Ljubljana 20089
Awane 1999 published data only
Awane T Integrating simulators in motorcycle safety education
Journal of International Association of Traffic and Safety Sciences 199923(1)26ndash35
Baldi 2005 published data only
Baldi S Baer JD Cook AL Identifying best practices states in
motorcycle rider education and licensing Journal of Safety Research200536(1)19ndash32
Batchler 1988 published data only
Batchler MW Evaluation of the pre-license motorcycle training course at the New South Wales traffic education centre Federal Office of
Road Safety (FORS) 19881ndash122 [ Report No CR 77 ]
Blanchard 2006 published data only
Blanchard HT Tabloski PA Motorcycle safety educating riders at
the teachable moment Mosby (CV) Company 200632330ndash2
Bonnett 2005 published data only
Bonnett J Mature age skills training for experienced motorcycle
riders Motorcycle ridersrsquo association of the Australian Capital
Territory 2005
Braver 2007 published data only
Braver ER et alPersuasion and licensure a randomized controlled
intervention trial to increase licensure rates among Maryland
motorcycle owners Traffic Injury Prevention 20078(1)39ndash46Clayton 1980 published data only
Clayton AB Sudlow DE An evaluation of the effectiveness of a one day training programme for learner motorcyclists Berkshire UK
Transport and Road Research Laboratory 19801ndash23 [ ISSN
0266ndash7045]lowast Clayton AB Sudlow DE An evaluation of the effectiveness of a one-day training programme for learner motorcyclists Berkshire UK
Road Safety Division Berkshire 19801ndash23 [ISSN 0266ndash7045]
Council 1978 published data only
Council FM Desper LP Dutt AK Driver education for motorcycle operation University of North Carolina Highway Safety Research
Center 19781ndash40
Crowther 2005 published data onlylowast Crowther G Engaging with motorcyclists UK police and theBikeSafe Road Safety Programme Published proceedings of the
Australasian Road Safety Research Policing Education Conference
Wellington New Zealand 2005 issue ISBN 0ndash473ndash10636ndash11ndash11
Daniello 2009 published data only
Daniello A Gabler HC Mehta YA Effectiveness of motorcycle
training and licensing Transportation Research Board 88th Annual
Meeting 200913
Dewanti 2007 published data only
Dewanti M Challenging efforts in promoting young motorcyclist
safety in Indonesia Road Safety on Four Continents 14th
International Conference Swedish National Road and Transport
Research Institute 2007
Du Plooy 1990 published data onlyDu Plooy DW Motorcycle training in South Africa The Human
Element International Motorcycle Safety Conference Orlando
Florida USA Motorcycle Safety Foundation 19904-1 to 4-9
Engel 1989 published data only
Engel U Krogsgaringrd Thomsen L [Faeligrdselsundervisning og
propaganda ndash en faeligrdselssikkerhedsfremmende foranstaltning]
Dansk Vejtidsskrift 198912295ndash7
Goldenbeld 2002 published data only
Goldenbeld C Houwing S de Craen S The development of
driving skills of young moped riders [De ontwikkeling van de
rijvaardigheid van jonge bromfietsers] SWOV Institute for Road
Safety Research (Stichting Wetenschappelijk Onderzoek
Verkeersveiligheid) Leidschendam the Netherlands 200263
25Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2850
Goldenbeld 2004 published data only
Goldenbeld C Twisk D de Craen S Short and long term effects of moped training a field experiment Transportation Research Part F 20047(1)1ndash16
Haworth 2000 published data only
Haworth N Smith R Kowadlo N Motorcycle training in Victoria
evaluation of rider training curriculum in Victoria Monash
University Accident Research Centre Reports 2000 Vol 165 issue
3131
Haworth 2005 published data only
Haworth N Mulvihill C Review of motorcycle licensing and
training Monash University Accident Research Centre 2005 [
Report no240]
Haworth 2006 published data only
Haworth N Mulvihill C Rowdon P Teaching old dogs new tricksTraining and older motorcyclists Journal of the Australasian College of Road Safety 200718(4)20ndash5lowast Haworth NL Mulvihill C Rowden PJ Teaching old dogs new
tricks Training and older motorcyclists Australasian Road Safety
Research Policing and Education Conference Gold Coast
Queensland 2006
Hill 1983 published data only
Hill D Motorcycle rider training testing and licensing Motor
Cycle Council of New South Wales Sydney 1983
Imatake 1980 published data only
Imatake Y On the activities in Japan of motorcycle safe riding
promotion committee (MSRPC) - especially the education of
unskilled riders of class 1 motorbikes International Motorcycle
Safety Conference Washington DC USA Motorcycle Safety
Foundation 1980255ndash68
Ingebritsen 1990 published and unpublished data
Ingebrigtsen S [Risikofaktorer ved ferdsel med moped og
motorsykkel] TOslashI-rapport 66 Oslo Transportoslashkonomisk
institutt 1990
Jain 2009 published and unpublished data
Jain A et alTwo wheeler accidents on Indian roads a study from
Mangalore India Elsevier 200916130ndash3
Jonah 1979 published data only
Jonah BA Dawson NE Validation of the motorcycle operator skill
test Accident Analysis amp Prevention 197911(3)163ndash71
Jonah 1981 published data onlyJonah BA Dawson NE Bragg WE Predicting accident
involvement with the Motorcycle Operator Skill Test Accident Analysis amp Prevention 198113(4)307ndash418
Kadar 1990 published data only
Kadar KJ Countermeasures motorcycle rider education delivery
system The Human Element International Motorcycle Safety
Conference Motorcycle Safety Foundation 19906-27 to 6-41
Kelsey 1986 published data only
Kelsey SL Liddicoat C Ratz M Licensing novice motorcyclists a
comparison of Californiarsquos standard test and the MOST II (Motorcycle
Operator Skill Test) administered at centralized testing offices Sacramento California USA California Department of Motor
Vehicles Research and Development Office 19861ndash74
Koch 1980 published data only
Koch H Advanced course for motorcycle riders InternationalMotorcycle Safety Conference Washington DC USA Motorcycle
Safety Foundation 1980269ndash81
Koch 1990 published data only
Koch H Brendicke R How to influence risk taking behaviour
within motorcycle rider training programs The Human Element
International Motorcycle Safety Conference Orlando Florida
USA Motorcycle Safety Foundation 1990Motorcycle Safety
Foundation 1990 p 9-1 to 9-25A
Labbett 2005 published data only
Labbett S Langham M Training can make the problem worse
Road Safety Congress 20051ndash12
Law 2005 published and unpublished data
Law TH Radin Umar RS Zulkaurnain S Kulanthayan S Impactof the effect of economic crisis and the targeted motorcycle safety
programme on motorcycle-related accidents injuries and fatalities
in Malaysia International Journal of Injury Control and Safety
Promotion 200512(1)9ndash21
Lowes 1990 published data only
Lowes BF In-traffic rider training The Human Element
International Motorcycle Safety Conference Orlando Florida
USA Motorcycle Safety Foundation 19907-46 to 7-57
Mahon 1976 published data only
Mahon J Teaching accident avoidance on the motorcycle Journal of Traffic Safety Education 1976231ndash29
Maxwell 1985 published data only
Maxwell GS Evaluation of a Program for Training MotorcycleRiders Evaluation and Program Planning 19858(4)339ndash47
Mayhew 1996 published data only
Mayhew DR Simpson SH Effectiveness and role of driver educationand training in a graduated licensing system Ottowa Canada The
Traffic Injury Research Foundation 19961ndash89 [ ISBN
0ndash920071ndash12ndash0]
McKnight 1987 published data only
McKnight AJ Milburn BG An evaluation of the modified
motorcycle operator skill test exercises NHTSA Report 1987 [
Report No DOT HS 807 212 ]
McPherson 1976 published data only
McPherson K McKnight AJ The development and evaluation of a
motorcycle skill test manual and knowledge test Washington DCUSA National Highway Safety Administration 1976
Mohan 2004 published data only
Mohan D Evidence-based interventions for road traffic injuries in
South Asia Journal of the College of Physicians and SurgeonsPakistan 200414(12)746ndash7
Morgan 1980 published data only
Morgan JK Advanced motorcycle training course content and
study evaluation International Motorcycle Safety Conference
Washington DC USA Motorcycle Safety Foundation 1980
283ndash312
Motoki 1990 published data only
Motoki M Yamazaki S A study on effective motorcycle rider
education The Human Element International Motorcycle Safety
26Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2950
Conference Orlando Florida USA Motorcycle Safety
Foundation 19909-26 to 9-57A
Nairn 1992 published data onlylowast Nairn RJ Motorcycle safety research literature review 1987 to
1991 Federal Office of Road Safety Australia 1993 Vol Report
CR 117
Newland 1980 published data only
Newland R Darby J Automotive practicespre motorcycle rider
program International Motorcycle Safety Conference Washington
DC USA Motorcycle Safety Foundation 1980127ndash9
Newland 1982 published data only
Newland R Motorcycle rider training and licensing requirements
Driver training steering a course for the future Melbourne
Victoria Transport Regulation Board 1982
Newland 1990 published data only
Newland R Motorcycle training and licensing in Victoria 1980-
1990 The Human Element International Motorcycle Safety
Conference Orlando Florida USA Motorcycle Safety
Foundation 1990
Ormston 2003 published data only
Ormston R Dudleston A Pearson S Stradling Evaluation of
Bikesafe Scotland Scottish Executive Social Research Edinburgh
2003
Osga 1979 published data only
Osga GS Ellingstad VS An evaluation of the effectiveness of the 1978
South Dakota motorcycle rider course program South Dakota USA
University of South Dakota 19791ndash89
Panou 2007 published data only
Panou M Bekiaris E A new concept on the integration of driving
simulators in driving training the train-all approach Road Safety
on Four Continents 14th International Conference Swedish
National Road and Transport Research Institute 2007
Peterson 1990 published data only
Peterson A Kadar K Status report 1980-1989 Minnesota rider
education statistics The Human Element International
Motorcycle Safety Conference Orlando Florida USA 199013-
113 to 13-129
Prem 1984 published data only
Prem H Good MC Motorcycle rider skills assessment ACT
Australia Office of Road Safety Department of Transport 1984
1ndash377 [ Report No CR 34 ]
Radin Umar 2006 published data only
Radin Umar RS Motorcycle safety programmes in Malaysia how
effective are they International Journal of Injury Control and Safety Promotion 200613(2)71ndash9
Rockwell 1990 published data only
Rockwell TH Kiger S Carnot M An evaluation of the Ohio
motorcyclist enrichment program Ohio Department of Highway
Safety 1990
Rothe 1987 published data only
Rothe JP A motorcycle riding course Motorcyclists Image and reality Canada Insurance Corporation of British Columbia 1987
59ndash83
Rowden 2007 published data only
Rowden P Watson B Haworth N What can riders tell us aboutmotorcycle training A view from the other side of the fence
Australasian Road Safety Research Policing and Education
Conference Melbourne Australia Melbourne 2007
RSC Victoria 1993 published data only
Road Safety Committee Inquiry into Motorcycle Safety in Victoria
Parliament of Victoria 1993
Russam 1979 published data only
Russam K Motorcycle training and licensing in Japan UK Transport
and Road Research Laboratory (TRRL) 1979 [ Report 916 ]
Saffron 1981 published data only
Saffron DG Driver and rider education training and licensing a
brief review Traffic Accident Research Unit Traffic Authority of
New South Wales NSW Australia Traffic Accident Research UnitTraffic Authority of New South Wales 1981
Schulz 1990 published data only
Schulz U Motorcycle accidents among learner riders Motorcycle
licensing Evaluation of the effectiveness of a Graduated Licensing Scheme Forshungshefte des Institut fur Zweiradsicherheit 1990
Sexton 2004 published data only (unpublished sought but not used)
Sexton B Baughan C Elliott M Maycock G The accident risk of
motorcyclists 2004TRL Report TRL 607(ISSN 09684107)58
Shephard 1986 published data only
Shepherd R Do Motorcycle Safety Foundation programs work
Driver 198619(9)16ndash8
Simpson 1990 published data onlySimpson HM Mayhew DR The promotion of motorcycle safety
training education and awareness Health Education Research
Theory and Practice 19905(2)257ndash64
Sokolov 1990 published data only
Sokolov O Safe motorcycle riding some aspects of motorcycle
riding in the USSR The Human Element International
Motorcycle Safety Conference Motorcycle Safety Foundation
19904ndash37 - 4-48
Steffens 1988 published data only
Steffens U Gawatz R Willmes G Effectiveness of mofa courses
[Wirksamkeit von Mofakursen] Unfall- und sicherheitsforschung
strasenverkehr 198867168
Sudlow 2003 published data only
Sudlow D Scoping study on motorcycle training Road Safety
Research Report London UK The British Institute of Traffic
Education Research Department for Transport 2003 Vol June
[ Report No 36]
Tasmania 1995 published data only
Mannering FL Grodsky LL Motorcycle rider training scheme
Traffic Education Research Centre Armidale Australia Driver
Training Australia Limited 1995 Vol April1ndash56
Tetard 1985 published data only
Tetard C Motorcycle ridersrsquo training to detect indices in an
intersection situation [Lrsquoentrainement a la prise drsquoinformation dans
lrsquoapprentissage de la conduite motocycliste] Le Travail Humain198548(2)147ndash59
27Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3050
Thackray 1980 published data only
Thackray RM Prescott JC Field test of the motorcycle ridercourse International Motorcycle Safety Conference Washington
DC USA Motorcycle Safety Foundation 1980195ndash7
Thompson 1994 published data only
Thompson M Evaluation of compulsory basic training for
motorcyclists Transport Research Laboratory Crowthorne
Berkshire UK 1994 [ Report 63]
Tobler 1990 published data only
Tobler U Individual driving instruction for motorcycle driving
learners the Swiss experiment of motorcycle driving school The
Human Element International Motorcycle Safety Conference
Orlando Florida USA Motorcycle Safety Foundation 1990
Transport 1989 published data only
Transport Communications Road Crash Statistics Australia Bureauof Transport and Communications Economics 198914ndash20
Tuan 2007 published data only
Tuan VA Shimizu T Towards development and evaluation of the
motorcycle drivers re-education program in Vietnam modeling of
motorcycle driverrsquos undesired behaviors 11th World Conference
on Transport Research 200725
Victor 1980 published data only
Victor DJ Traffic safety education for motorcyclists International
Motorcycle Safety Conference Washington DC USA Motorcycle
Safety Foundation 1980221ndash8
Waterer 1990 published data only
Waterer TH Compulsory training for motorcyclists from liberty
to compulsion The Human Element International Motorcycle
Safety Conference Orlando Florida USA Motorcycle Safety
Foundation 19904-49 to 4-57
Weller 1989 published data only
Weller RB Chandler EW Motorcycle safety and motorcycle
education past research and survey results Journal of Public Policy
amp Marketing 1989893ndash108
Wells 1981 published data only
Wells P Observations of some RACACU motorcycle training
schemes Transport and Road Research Laboratory Crowthorne
Berkshire UK 1981
Winn 1986 published data only
Winn G Two important trends in motorcycle safety regulations
rider education and conspicuity improvement American
Motorcyclist Association 1986147ndash52
Wisher 1988 published data only
Wisher G Reid J Motorcycle accidents in Norfolk a three-year study of motorcycle training and the Part 1 test Norfolk UK Road Safety
Division Highways Department Norfolk County Council 1988
Wood 1987 published data only
Wood HT Bowen R Evaluation of the revised learner permit scheme
July 1983 to December 1985 Victoria Australia Road Traffic
Authority 19871ndash27 [ ISBN 0 7306 0247 8]
Yamazaki 1990 published data only
Yamazaki S Nakanishi M Survey on effect of Honda rider
education The Human Element International Motorcycle Safety
Conference Orlando Florida USA 199012-167 to 12-187
References to ongoing studies
Burgess 2010 unpublished data only
Burgess C Broughton P Fylan F Stradling Interim evaluation of
the UKs National Ride program In L Dorn editor(s) Driver
behaviour and Training Vol IV Aldershot Ashgate Publishing
Limited 2010
Additional references
Anderson 1980
Anderson J Ford JL Peck RC Improved motorcyclist licensing and
testing project Volume I and II National Highway Traffic Safety
Administration US department of Transportation Washington
DC 1980
Bjornstig 1985
Bjornstig U Bylund P Lekander T Brorsson B Motorcycle fatalitiesin Sweden Acta Chirurgica Scandinavica 1985151(7)577ndash81
Clayton 1990
Clayton A Sudlow DE An evaluation of the effectiveness of a one-
day training program for learner motorcyclists The Human
Element International Motorcycle Safety Conference Orlando
Florida USA 199013-105 to 13-111
Cochrane Collaboration 2008
Higgins JP Green S Cochrane Handbook for Systematic Reviews of
Interventions 500 The Cochrane Collaboration 2008
Elliot 2003
Elliott MA Baughan CJT Broughton J Grayson G Knowles J et
alMotorcycle Safety a Scoping Study TRL [ Report No
0968ndash4107]Fleiss 2004
Fleiss JL Levin B Paik MC Statistical methods for rates and proportions 3rd Edition John Wiley amp Sons Inc 2004 [ ISBN
978ndash0ndash471ndash52629ndash2 ]
FORS 1999
FORS Federal Office of Road Safety Road risk for sober licensed
motorcyclists Vol Monograph 27 Federal Office of Road Safety
19991ndash2
Haworth 2005
Haworth N Mulvihill C Review of motorcycle licensing and training
Victoria Australia Monash University Accident Research Centre
20051ndash83 [ Report No 240]
Huwiler-Muumlntener 2002Huwiler-Muumlntener K Juumlni P Junker C Egger M Quality of
Reporting of Randomized Trials as a Measure of Methodologic
Quality JAMA 20022872801ndash4
Johnston 2008
Johnston P Brooks C Savage H Fatal and serious road crashes
involving motorcyclists Road Safety Research and Analysis Report
ATSB Monograph 20 Canberra Department of Infrastructure
Transport Regional Development and Local Government 0205
081ndash26
Juumlni 2001
Juumlni P Altman DG Egger M Systematic reviews in health care
Assessing the quality of controlled clinical trials BMJ 2001323
42ndash6
28Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3150
Kopits 2003
Kopits E Cropper M Traffic fatalities and economic growth TheWorld Bank Development Research Group Infrastructure and
Environment Policy research working paper 3035 2003
Leung 1983
Leung P Chau BTW Motorcycle accidents in Hong Kong pattern
of injury treatment causative factors and socio-economic
implications Injury 198315(1)23ndash9
Liu 2004
Liu B Ivers R Norton R Blows S Lo SK Helmets for preventing
injury in motorcycle riders Cochrane Database of Systematic Reviews 2004 Issue 2 [DOI 10100214651858CD004333pub3]
Mohan 2002
Mohan D Road safety in less-motorized environments future
concerns International Journal of Epidemiology 200231(3)527ndash32Mortimer 1982
Mortimer RG An evaluation of the motorcycle rider course
Proceedings of the 20th conference of the American Association for
Automotive Medicine Arlington Heights IL USA American
Association for Automotive Medicine 198277ndash96
Mullin 1997
Mullin BT Risk factors for motorcycle injury the role of age
gender experience training and alcohol PhD thesis Auckland
New Zealand University of Auckland 1997
Noordzij 2001
Noordzij PC Forke E Brendicke R Chinn BP Integration of needs
of moped and motorcycle riders into safety measures review and
statistical analysis in the framework of the European research
project PROMISING SWOV Institute for Road Safety Research
Leidschendam The Netherlands 2001 [ Report No Dndash2001ndash5]
Osga 1979
Osga G Ellingstad VS An evaluation of the effectiveness of the 1978 South Dakota Motorcycle Riders Course Program South Dakota
University of South Dakota Vermillion Human Factors Lab 1979
Oxford English Dictionary Online
httpoxforddictionariescom Oxford English Dictionary Online
Oxford University Press 2010
Peden 2004
Peden M Scurfield R Sleet D World Report on Road TrafficInjury Prevention Geneva World Health Organization Available
at httpwwwwhointviolence˙injury˙preventionpublications
road˙trafficworld˙reporten 2004Chapter 3
Radin Umar 1995
Radin Umar R Macakay G Hills B Preliminary analysis of
motorcycle accidents Short-term impacts of the running
headlights campaign and regulations in Malaysia J Traffic Med 199523(1)17ndash27
Tsai 1995
Tsai YJ Wang JD et alCase-control study of the effectiveness of
different types of helmets for the prevention of head injuries among
motorcycle riders in Taipei Taiwan American Journal of
Epidemiology 1995142(9)974ndash81
Vis 1995
Vis AA In-depth study of the hazards of motorcycling [De
onveiligheid van motorrijden nader bekeken] SWOV Institute for
Road Safety Research Leidschendam the Netherlands 1995 [
Report No Rndash95ndash69]
Weaver 1990
Weaver EA Motorcycle safety education in the United States
Motorcycle Safety Foundation Orlando Florida USA 1990
Winn 1987
Winn G Two important trends in motorcycle safety regulations
rider education and conspicuity improvement No more
information available Detroit MI US Society of Automotive
Engineers 1987147ndash52
Yuan 2000
Yuan W The effectiveness of the rsquoride-brightrsquo legislation for
motorcycles in Singapore Accident Analysis amp Prevention 200032
(4)559ndash63
Zambon 2006
Zambon F Hasselberg M Factors affecting the severity of injuries
among young motorcyclists a Swedish nationwide cohort study
Traffic Injury Prevention 20067(2)143ndash9lowast Indicates the major publication for the study
29Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3250
C H A R A C T E R I S T I C S O F S T U D I E S
Characteristics of included studies [ordered by study ID]
Adams 1985
Methods Non-randomised trial
Participants Initial sample size
Total of 162 Australia Post telegram delivery riders
Not clear how many in intervention and control groups
Analysis sample size
Intervention group 63Control group 35
Total 98 (61)
Interventions No formal training vs voluntary rider training course of 12 hours over 2 consecutive days (theory 4 hours practice
8 hours)
Intervention integrity
No information given
Outcomes Australia Post recorded crashes
Length of follow up 1 year
Notes Study conducted in NSW Australia
Anderson 1980
Methods Randomised trial
Participants Initial sample size
Total 41188 motorcycle licence applicants
Analysis sample size
Intervention group with skills training 13160
Intervention group without skills training 12634
Control group 15080
Total 40874 (99)
Interventions Standard licensing skill test vs the Motorcycle Operators Skill Test (MOST) with or without a training component
MOST is not to be confused with rider training because it is a licensing test The training component was for
participants who failed the skill test on the first attempt was of 3 hours duration and consisted of instructions about
the skills needed to pass the skill test The training consisted of 30 min theory and 2 frac12 hours practice on the
motorcycle
Intervention integrity
33 in MOST with skills training 31 in MOST without skills training and 43 in control group obtained their
motorcycle licence in the end
Outcomes Police recorded crashes and offences
Length of follow up 3 years
30Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3350
Anderson 1980 (Continued)
Notes Study conducted in California USA
Billheimer 1998
Methods Cohort study
Participants Initial sample size
Intervention group Unclear numbers of CMSP graduates and ERC graduates
Control group Over 16000 untrained motorcyclists
Analysis sample size
1139 matched CMSP-pairs and 1182 ERC-pairs
Interventions No formal training vs the mandatory (for those younger than 18 years) pre-licence 16-hours basic California
Motorcyclist Safety Program (CMSP) and the 8-hour Experienced Rider Course (ERC)
Intervention integrity
No information given
Outcomes Police recorded crashes and offences
Length of follow up 2 years
Notes Study conducted in California USA
Cooper 1988
Methods Cohort study
Participants Initial sample size
Intervention group 418 graduates and 57 who failed
Control group 402
Analysis sample size
Intervention group unclear
Control group unclear
Total 863 (984)
Interventions No formaltrainingvs non-mandatory pre-licence rider trainingThe content of thetrainingwas notfurther explainedIntervention integrity
Participants had all completed the rider training
Outcomes Police recorded crashes and offences
Length of follow up 7 years
Notes Study conducted in British Columbia Canada
31Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3450
Davis 1997
Methods Retrospective cohort study
Participants Initial sample size
Intervention group 9320 graduates
Control group unclear
Analysis sample size
Intervention group 9320 graduates
Control group unclear
Interventions No formal training vs the non-mandatory Connecticut Motorcycle Operator Training Program The content of
training was not further explained
Intervention integrity Participants had all completed the rider training
Outcomes Police recorded crashes severity of crashes
Length of follow up 1 year
Notes Study conducted in Connecticut USA
Hall 1988
Methods Randomised trial
Participants Initial sample sizeIntervention group A 6248
Intervention group B 6616
Intervention group C 6996
Control group 6604
Total 26464 motorcycle licence applicants
Analysis sample size
Intervention group A 6012 (96)
Intervention group B 6428 (97)
Intervention group C 6778 (97)
Control group 6502 (98)
Interventions Standard licensing skill test vs the Motorcycle Operator Skill Test (MOST II) without training (A) MOST II with
3 hours training (B) or MOST II with 20 hours training (C) The 3 hours training comprised instructions aboutthe skills needed to pass the skill test (30 min theory and 2 frac12 hours practice on the motorcycle) The 20 hours
training consisted of the Motorcycle Safety Foundationrsquos Motorcycle Rider Course (MRC) comprising of 8 hours
theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-road practice
Intervention integrity
25 and 26 obtained their motorcycle licence in the lsquoMOST 3-hour trainingrsquo and lsquoMOST 20-hour trainingrsquo
intervention arm respectively In the lsquoMOST testrsquo intervention arm 29 obtained their licence and in the control
group 35
Outcomes Police recorded crashes and offences
Length of follow up 3 years
32Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3550
Hall 1988 (Continued)
Notes Study conducted in New York USA
Haworth 1997
Methods Population based case-control study
Participants Cases Motorcycle riders or pillion riders involved in motorcycle crash (on road) Recruited from hospital and police
records Not clear whether all cases (especially fatalities) were included
Community controls Multiple matched sample of motorcycle rider or pillion who passed the crash site at the same
time of day and day of week that the crash occurred
Initial sample size
Cases 336
Controls 1195
Analysis sample size
Cases 234 (70)
Controls 561 (47)
Interventions Basic handling skills course beginner intermediate or advanced course measured through self-report not validated
Outcomes Injuries with an Injury Severity Score of 5 or greater or admission for at least four hours and fatalities from a
motorcycle crash (from hospital and police records)
Length of follow up two years or more prior to the interview
Notes Study conducted in Victoria Australia
Hurt 1981
Methods Population based case-control study
Participants Cases Motorcycle riders who had a motorcycle crash that was reported to police (traffic accident reports) Not all
cases included
Community controls Multiple matched sample of motorcycle riders who passed the crash site at the same time of
day and day of week as the crash occurred but 2 years later
Initial sample size
Cases 4500 notifications of crashes to police (only 50 received in time research team able to respond to 1126 for
unreported reasons data collection completed for 900)
Controls Number approached unreported 2310 consented and interviewed
Analysis sample size
Cases 808 (808900 data on rider training) - 18 of initial sample
Controls 670 (6702310 data on rider training) - lt 29 of initial sample
Interventions Training experience self taught friends-family school-club formal-professional others measured through self-
report validated
Outcomes Injuries and fatalities from a motorcycle crash (from hospital records and coronerrsquos office)
33Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3650
Hurt 1981 (Continued)
Notes Study conducted in Los Angeles USA
Jonah 1982
Methods Cohort study
Participants Initial sample size
Intervention group
2310 licensed MTP graduates who are active riders
Control group
2029 licensed motorcyclists who are active riders
Non-licensed and non-active riders excluded
Analysis sample size
Intervention group
811 interviews (35) 533 driver records (23)
Control group 1080 interviews (71) 905 driver records (45)
Interventions No training vs the non-mandatory Canada Safety Council Motorcycle Training Program (MTP) comprising of 4
hours theory and 16 hours practise (6 hrs on-road) over 2 weekends
Intervention integrity
Participants had all successfully completed the MTP
Outcomes Self-reported and police reported crashes and offencesLength of follow up 4 years
Notes Study conducted in Ontario Canada
Kloeden 1994
Methods Cohort study
Participants Initial sample size
Intervention group 2533 licence applicants
Control group Unclear
Sample of riders obtaining probationary licenceIntervention group 1625 (64)
Control group 5015 (unclear)
Non-active riders excluded from control group (~66)
Analysis sample size
Intervention group 631 (39)
Control group 318 (6)
Interventions No formal training vs the mandatory pre-licence Ridersafe course consisting of two levelsthe first is a training course
in the basics of motorcycle riding where a pass entitles the rider to obtain a learnerrsquos permit The second is a skill
test where a pass entitles the rider to obtain a probationary motorcycle licence The training course consists of two 4
hour training sessions on separate days involving instruction videos and actual riding on motorcycles The skill test
consists of a 4 hour preparation course of 74 minutes of lecturediscussion 38 minutes of video and 75 minutes of
34Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3750
Kloeden 1994 (Continued)
on-course riding which is the testing phase
Intervention integrity
Unclear what percentage in the Ridersafe group and the control group obtained their probationary licence halfway
and their full motorcycle licence in the end
Outcomes Police reported crashes
Length of follow up 1 year
Notes Study conducted in SA Australia
Kraus 1975
Methods Population based case-control study
Participants Cases All motorcycle riders with a confirmed medically treated motorcycle injury (including fatalities) from all
hospitals in study base and peripheral area hospitals and all deaths due to motorcycle crashes in the study base All
serious injuries (leading to hospitalisation) and fatalities likely to be included
Community controls A random sample of registered motorcycle owners in the same temporary and geographical
study base as the cases arose from
Initial sample size Cases 1273
Controls 738
Analysis sample size Cases 741 (58)
Controls 434 (59)
Interventions Motorcycle rider training (not specified) measured through self-report
Outcomes Injuries leading to hospitalisation fatalities
Length of follow up not specified
Notes Study conducted in Sacramento county USA
Leung 1987
Methods Non-randomised trial
Participants Initial sample sizeIntervention group unknown number of MRC graduates
Control group unknown number of licensed motorcyclists
Analysis sample size
Intervention group 2941
Control group 43094
Interventions No formal training vs non-mandatory Wisconsin motorcycle rider course a 22 hours off-street programme that
teaches the basic rules of safe motorcycling
Intervention integrity
50 in the MRC group and 100 in the control group obtained their motorcycle licence in the end
35Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3850
Leung 1987 (Continued)
Outcomes Police recorded crashes and offences
Length of follow up up to 3 years
Notes Study conducted in Wisconsin USA
MAIDS 2004
Methods Population based case-control study
Participants Cases Motorcycle riders who had a motorcycle crash Unclear whether all cases included
Community controls Motorcycle riders identified at petrol stations Unclear whether matched to cases
Initial sample size
Cases not reported
Controls not reported
Analysis sample size
Cases 921 (Italy 200 Netherlands 200 Spain 121 Germany 250 France 150)
Controls 923 (Italy 200 Netherlands 200 Spain123 Germany 250 France 150)
Interventions Pre-license rider training or additional training (non-compulsory)
Outcomes Crashes
Notes Study in five sampling areas located in five European countries (France Germany Netherlands Spain and Italy)
McDavid 1989
Methods Cohort study
Participants Sample size after first match
Intervention group 346 male training graduates
Control group 346 male motorcycle riders licensed in same year
Sample size after second match
Intervention group 139 (40)
Control group 139 (40)
Analysis sample size
Intervention group 139 (40)
Control group 139 (40)
Interventions No formal training vs the non-mandatory pre-licence British Columbia Safety Councilrsquos 37-hour motorcycle safety
programme
Intervention integrity
Participants had all completed and passed the training
Outcomes Police reported crashes injurious crashes
Length of follow up 1-6 years
36Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3950
McDavid 1989 (Continued)
Notes Study conducted in British Columbia Canada
Mortimer 1984
Methods Cohort study
Participants Initial sample size
Intervention group 608 MRC graduates
Control group 360 motorcycle riders
Response rates
Intervention group 360 (59)
Control group 324 (90)
Non-active riders excluded
Analysis sample size
Intervention group 213 (35)
Control group 303 (84)
Interventions No formal training vs the non-mandatory Motorcycle Safety Foundationrsquos Motorcycle Rider Course comprising of
8 hours theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-road practice
Intervention integrity
Participants had all completed and passed the MRC
Outcomes Self-reported crashes and offencesLength of follow up 1 year
Notes Study conducted in Illinois USA
Mortimer 1988
Methods Cohort study
Participants Initial sample size
Intervention group 2307 MRC graduates
Control group 556 active licensed motorcycle riders
Response ratesIntervention group 1298 (56)
Control group 500 (90)
Non-active riders excluded
Analysis sample size
Intervention group 913 (40)
Control group 500 (90)
Interventions No formal training vs the non-mandatory Motorcycle Safety Foundationrsquos Motorcycle Rider Course comprising of
8 hours theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-road practice
Intervention integrity
Participants had all completed and passed the MRC
37Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4050
Mortimer 1988 (Continued)
Outcomes Self reported crashes and offences
Length of follow up 1 year
Notes Study conducted in Illinois USA
Osga 1980
Methods Cohort study
Participants Initial sample sizeIntervention group 1163 MSF MRC participants invited 818 (74) responded
Control group 329 motorcycle riders responded to recruitment strategies
Non-active riders excluded from analysis
Analysis sample size
Intervention group 595 (73)
Control group 329 (100)
Interventions No formal training vs the non-mandatory Motorcycle Safety Foundationrsquos Motorcycle Rider Course (MRC) com-
prising of 8 hours theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-road
practice
Intervention integrity
Participants had all successfully completed the MRC
Outcomes Self-reported crashes police recorded crash reports
Length of follow up 6 months - 3 12 years
Notes Study conducted in South Dakota USA
Perrino 2002
Methods Cohort study
Participants Initial sample size
Intervention group 2144 (unclear) MRC graduates (motorcycle licence applicants)
Control group 2601 (unclear) regular licensing test graduates (motorcycle licence applicants)Analysis sample size
Intervention group 1070 (unclear)
Control group 867 (unclear)
Interventions No formal training vs the mandatory (for those younger than 18 years) pre-licence Maryland Motorcycle Rider
Course (MRC) consisting of 8 hours theory and 12 hours practice
Intervention integrity
No information given
Outcomes Self-reported and police recorded crashes and offences
Length of follow up 1 year
38Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4150
Perrino 2002 (Continued)
Notes Study conducted in Maryland USA
Raymond 1979
Methods Cohort study
Participants Initial sample size
Intervention group
258 volunteers mainly (but not all) novice riders
Control group
1104 provisional licence applicants
Analysis sample size
Intervention group 190 (74)
Control group 627 (57)
Interventions No formal training vs the non-mandatory RACACU (Royal Automobile ClubAuto-Cycle Union) training scheme
The training program consisted of either 24 one-hour sessions or 13 two-hour sessions Twelve hours of the training
focused on theory about basic operational procedures traffic laws and highway code The practical session focused
on basic operational skills riding on a training area and riding in traffic with an instructor A proficiency test was
administered at the end of the course
Intervention integrity
No information given
Outcomes Self-reported crashes (including minor non-injury non-damage and off-road crashes)
Length of follow up 1 year
Notes Study conducted in the UK
Satten 1980
Methods Cohort study
Participants Initial sample size
Intervention group interviewers kept interviewing until they had 100 riders selected from a population of 891
enrollees in the MSF MRC course (unknown how many were asked and how many refused to participate)Control group unknown number of registered motorcycle riders
Non-active riders excluded from analysis
Analysis sample size
Intervention group 71
Control group 100
Interventions No formal training vs the non-mandatory Motorcycle Safety Foundationrsquos (MSF) Motorcycle Rider Course (MRC)
comprising of 8 hours theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-
road practice
Intervention integrity
Participants had all successfully completed the MRC
39Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4250
Satten 1980 (Continued)
Outcomes Self-reported crashes injuries and offences
Length of follow up 3 years
Notes Study conducted in Illinois USA
Savolainen 2007
Methods Cohort study
Participants Initial sample size
Intervention group 4000 riders that had conducted the MSF MRC course (passed or failed)
Control group 4000 motorcycle owners or permit or endorsement holders from the Indiana Bureau of Motor
Vehicles database
Analysis sample size
Intervention group 739 (18)
Control group 588 (15)
Interventions Intervention integrity
No information given
Outcomes Probability of crash involvement
Length of follow up up to 5 years
Notes Study conducted in Indiana USA
Swaddiwudhipong 1998
Methods Randomised trial
Participants Initial sample size
Intervention group
all 16 villages of 3 sub-districts
(participants were practising motorcyclists licensed and unlicensed)
Control group
18 villages of 3 other sub-districtsAnalysis sample size
Intervention group 3 districts (100)
Control group 3 districts (100)
Interventions No rider education vs non-mandatory community based rider education consisting of theory on motorcycle injury
risk helmet use traffic laws vehicle regulations and traffic signs and a written and skill test leading to a licence
Intervention integrity
No information given
Outcomes Injuries fatalities
Length of follow up 2 years
40Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4350
Swaddiwudhipong 1998 (Continued)
Notes Study conducted in Tak Province rural Thailand
Waller 1992
Methods Cohort study
Participants Initial sample size
Intervention group 349 riders who succesfully completed the Department of Public Safety motorcycle operator
training in 1988
Control group 349 motorcyclists who never completed the DPS motorcycle operator training
Analysis sample size
Intervention group 349
Control group 349
Interventions No DPS motorcycle operator training vs non-mandatory DPS motorcycle operator training The DPS training is
the Motorcycle Safety Foundationrsquos Motorcycle Rider Course comprising eight hours of theory (lectures discussions
films and multiple choice tests) a textbook and 12 hours off-road practice
Intervention integrity
All 349 riders succesfully completed the rider training
Outcomes Police recorded crashes
Length of follow up 3 years
Notes Study conducted in
Characteristics of excluded studies [ordered by study ID]
Study Reason for exclusion
Anderson 1980a Study type review and process evaluation of a rider course
Aupetit 2008 Outcome proposal of a rider training curriculum that combines traditional training and a motorcycle riding
simulator
Awane 1999 Study type review on use of simulators in rider and driver training
Baldi 2005 Intervention not rider training but intervention integrity of rider training availabilty of rider education
implementation of graduated licensing system etc on a state-wide level
Batchler 1988 Study type description of a training course
Blanchard 2006 Study type not a study assessing the association between rider training and outcomes
Bonnett 2005 Outcome qualitative data regarding satisfaction with the training programme
41Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4450
(Continued)
Braver 2007 Outcome licensure rates
Clayton 1980 Outcome knowledge skills attitude
Council 1978 Outcome crash reduction in ANY vehicle not motorcycles exclusively
Crowther 2005 Study type qualitative study
Daniello 2009 Study type review of the effectiveness of motorcycle training and licensing
Dewanti 2007 Intervention emotional characteristics behaviour (ie not motocycle rider training)
Du Plooy 1990 Study type review of types of rider training courses in South Africa
Engel 1989 Study type uncontrolled before-after study
Goldenbeld 2002 Outcome skills
Goldenbeld 2004 Outcome knowledge and skills
Haworth 2000 Intervention balance between attitudinal andvehicle control skills acrosscurriculla of motorcycle rider training
courses
Haworth 2005 Study type review
Haworth 2006 Study type cross-sectional survey
Hill 1983 Study type review and description of training system
Imatake 1980 Study type course descriptionreview
Ingebritsen 1990 Intervention risk factors for motorcycle crashes
Jain 2009 Study type cross-sectional descriptive study of motorcycle crashes over a 5 year period
Jonah 1979 Study type cross-sectional survey Outcome scores on motorcycle operator skill test
Jonah 1981 Intervention evaluation of a motorcycle rider skill test not motorcycle rider training
Kadar 1990 Study type review of components of rider training courses
Kelsey 1986 Intervention evaluation of a motorcycle rider skill test not motorcycle rider training
Koch 1980 Study type reviewcourse description
Koch 1990 Study type description of course content
42Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4550
(Continued)
Labbett 2005 Study type essay about possible explanations for motorcycle deaths
Law 2005 Intervention targeted motorcycle safety programme not specifically rider training
Lowes 1990 Study type review and descriptions of rider training
Mahon 1976 Study type review of motorcycle rider training methods
Maxwell 1985 Outcomesrecommendations forprogramme operatorsand otherson howthe motorcycle trainingprogramme
works and how it can be improved
Mayhew 1996 Study type review
McKnight 1987 Study type evaluation of methodological and administrative aspects of a motorcycle operator skill test
McPherson 1976 Outcome knowledge and skills
Mohan 2004 Study type review of evidence of interventions to reduce road traffic injuries
Morgan 1980 Study type reviewcourse description
Motoki 1990 Study type proposal and development of rider training curriculum
Nairn 1992 Study type literature review
Newland 1980 Study type description of development of motorcycle safety course
Newland 1982 Study type descriptionreview of a rider training course
Newland 1990 Study type uncontrolled BeforeAfter study
Ormston 2003 Outcome changes in attitude and self reported behaviour
Osga 1979 Outcome knowledge attitudes and skills
Panou 2007 Intervention new simulation tool to teach driverrider
Peterson 1990 Study factor demographic details of course participants completing the course vs those not completing the
course
Prem 1984 Study type experimental
Study factor examines association between skills and performance in tests
Radin Umar 2006 Intervention motorcycle safety programme not specifically rider training
Rockwell 1990 Study type methodologically flawed study type
43Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4650
(Continued)
Rothe 1987 Study type qualitative study of rider training in general
Rowden 2007 Study type qualitative study
RSC Victoria 1993 Study type review of motorcycle rider training and education in Victoria
Russam 1979 Study type review of training programmes in Japan
Saffron 1981 Study type literature review
Schulz 1990 Study factor predictors of motorcycle crashes
Sexton 2004 Study type cross-sectional analytic analysis
Shephard 1986 Study factor reviewdescription of programme
Simpson 1990 Study type review of motorcycle rider trainingeducation
Sokolov 1990 Study factor review of motorcycle riding courses in USSR
Steffens 1988 Outcome self-reported behaviour experiences and researcher observed riding behaviour
Sudlow 2003 Study type review of motorcycle training in the UK
Tasmania 1995 Study type cross-sectional descriptiveprocess evaluation
Tetard 1985 Study type experimental
Outcome detection of indices
Thackray 1980 Outcome instructional effectiveness user acceptance administrative feasibility
Thompson 1994 Study type cross-sectional survey
Outcome attitudes and knowledge
Tobler 1990 Study type review of course curricula
Transport 1989 Intervention syllabus for rider training
Tuan 2007 Study type cross-sectional study (survey) which does not assess the effects of an intervention but examines
motocyclistsrsquo behaviour
Victor 1980 Study type review
Waterer 1990 Study type review of UK rider training issues
Weller 1989 Study type review of studies evaluating motorcycle safety and education measures
44Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4750
(Continued)
Wells 1981 Study type description of different methods of training
Winn 1986 Study type review of rsquohistoryrsquo of rider training
Wisher 1988 Study type cross-sectional survey
Wood 1987 Study type Interrupted Time Series but did not meet the criteria for an ITS as published by the Cochrane
Effective Practice and Organisation of Care Group
Yamazaki 1990 Study type cross-sectional survey
Outcome riding skills skills retention and attitude
Characteristics of ongoing studies [ordered by study ID]
Burgess 2010
Trial name or title UKs National Ride program
Methods Cohort study
Participants Motorcycle offenders
Interventions Police led diversion program
Outcomes Self reported attitudinal and behavioural measures
Starting date
Contact information
Notes Expected to be completed in early 2010
45Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4850
D A T A A N D A N A L Y S E S
This review has no analyses
A P P E N D I C E S
Appendix 1 Search strategy
We searched the following electronic databases
bull Cochrane Injuries Group Specialised Register (searched 22 Sept 2008)
bull CENTRAL (The Cochrane Library 2008 Issue 3)
bull TRANSPORT 1988 to 200706 (latest date available)
bull ATRI Australian Transport Index (to Sept 2008)
bull MEDLINE (Ovid SP) 1950 to Sept 2008
bull EMBASE (Ovid SP) 1980 to Sept 2008
bull CINAHL (EBSCO) 1982 to Sept 2008
bull WHOLIS (World Health Organization Library Information System) (to Sept 2008)
bull PsycINFO (Ovid SP) 1806 to Sept 2008
bull LILACS (BIREME) (Latin American and Caribbean Health Sciences) (1982 to Sept 2008)
bull ISI Web of Science Social Sciences Citation Index (SSCI) 1970 to Sept 2008
bull ERIC (searched 25 Sept 2008)
bull ZETOC (searched 25 Sept 2008)
bull SIGLE (searched 25 Sept 2008)
All of the databases were searched using the following text-word search strategy The strategy was adapted using either rsquorsquo or rsquo$rsquo as atruncation symbol
1 motor-cycl or motorcycl or motor-bik or motorbik or motor-scooter or motorscooter or scooter or moped or motocycl
or motocicl or moto
2 train or educat or instruct or program or course or qualif or learn or improv
3 1 and 2
H I S T O R Y
Protocol first published Issue 2 2005
Review first published Issue 10 2010
46Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4950
C O N T R I B U T I O N S O F A U T H O R S
Katina Kardamanidis was the first author of the protocol conducted the literature search extracted full studies assessed the studies
according to inclusion criteria assessed quality of the studies compared findings with the second author discussed disagreements with
the third author wrote the drafts and the final review
Rebecca Ivers was the second author of the protocol assessed the studies according to inclusion criteria assessed quality of the studies
compared findings with the first author edited the final draft and provided overall guidance for the process
Alexandra Martiniuk assessed quality of the studies compared findings with the first author discussed disagreements with the third
author and helped write the first draft and the final review
Mark Stevenson contributed to the protocol reviewed methods of studies and contributed to the final draft
Katrina Thistlethwaite conducted a second literature search extracted full studies and contributed to the final draft
D E C L A R A T I O N S O F I N T E R E S T
None known
S O U R C E S O F S U P P O R T
Internal sources
bull No sources of support supplied
External sources
bull National Health and Medical Research Council of Australia Australia
Salary support for Martiniuk Ivers and Stevenson
bull University of Sydney Sesqui Research Grant Australia
Direct research costs and salary support for Kardamanidis
D I F F E R E N C E S B E T W E E N P R O T O C O L A N D R E V I E W
There are no differences between the protocol and the review
I N D E X T E R M S
Medical Subject Headings (MeSH)
lowastMotorcycles Accident Prevention [lowastmethods] Accidents Traffic [lowastprevention amp control] Licensure Program Evaluation Wounds
and Injuries [mortality lowastprevention amp control]
47Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 5050
MeSH check words
Humans
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2650
motorcycle safety class with riders participating in a basic motor-
cycle safety class as well as a variety of continuing education classesThe collection of pilot data was due to start in March 2007 and
the data collection portion of the study is expected to take three
years (Highway Safety Research Center) It is vital that researchers
and policy makers collaborate to develop furtherhigh quality eval-
uations of motorcycle rider training programmes to help inform
development of new and effective training programmes
A C K N O W L E D G E M E N T S
Staff from The University of Sydney library for helping us locate
and retrieve full reports of many papers
R E F E R E N C E S
References to studies included in this review
Adams 1985 published data only
Adams A Collingwood V Job RFS Evaluation of the 1982 Australia
Post motorcycle rider training scheme NSW Australia Traffic
Accident Research Unit Traffic Authority NSW 1985 [ RN 85]
Anderson 1980 published data only
Anderson J Ford JL Peck RC Improved motorcyclist licensing and testing project Washington DC National Highway Traffic Safety
Administration (NHTSA) USA 19801ndash217 [
DOTndashHSndash805ndash598]
Billheimer 1998 published data only
Billheimer JW Evaluation of the California Motorcyclist Safety
Program Transportation Research Record 19981640100ndash9 [
TRB Report Number 980652]
Cooper 1988 published data only
Cooper PJ Rothe JP Motorcyclists who they are and why they do
what they do Transportation Research Record 1988116878ndash85
Davis 1997 published data only
David CF Evaluation of community traffic safety programs and
motorcycle operator training programs Connecticut Department of Transportation 199763
Hall 1988 published data only
Hall CJ Buchanan L Engle R Motorcycle rider education evaluationproject Washington DC USA National Highway Traffic Safety
Administration (NHTSA) US Department of Transportation
1988 [ DTNH 22ndash80ndashCndash05123]
Haworth 1997 published data only
Haworth N Smith R Brumen I Pronk N In Monash University
Accident Research Centre editor(s) Case-control study of motorcycle
crashes Victoria Australia Department of Transport and Regional
Development The Federal Office of Road Safety 19971ndash149 [
ISBN 0 642 25501 6 ]
Hurt 1981 published data only
Hurt H Ouellet J Thom D Motorcycle accident cause factors and
identification of countermeasures USA Department of
Transportation 1981 [ Technical Report DOT HS 806ndash862]
Jonah 1982 published data onlylowast Jonah BA Dawson N Bragg B Are formally trained motorcyclists
safer Accident Analysis amp Prevention 198214(4)247ndash55
Jonah BA Dawson NE Bragg BW Evaluation of the Canada
Safety Councilrsquos Motorcycle Training Program Proceedings of theInternational Motorcycle Safety Conference Washington DC
USA 198059ndash93
Kloeden 1994 published data only
Kloeden CN Moore VM McLean AJ Evaluation of the pre-license
training program for motorcyclists in South Australia Walkerville
South Australia Australia NHMRC Road Accident Research Unit
The University of Adelaide 19941ndash37 [ ISBN 0 7308 0607 3]
Kraus 1975 published data only
Kraus JF Riggins RS Franti CE Some epidemiological features of
motorcycle collision injuries American Journal of Epidemiology 1975102(1)74ndash98
Leung 1987 published data only
Leung KS Reding VA Evaluation of the Wisconsin motorcycle rider course Wisconsin Bureau of Policy Planning and Analysis
Wisconsin Department of Transportation 1987
MAIDS 2004 published and unpublished data
MAIDS In-depth investigations of accidents involving powered
two wheelers ACEM (Acemacemeu) 2004 Vol Final report
20179
McDavid 1989 published data only
McDavid JC Lohrmann BA Lohrmann G Does motorcycle
training reduce accidents Evidence from a longitudinal quasi-
experimental study Journal of Safety Research 198920(2)61ndash72
Mortimer 1984 published data only
Mortimer R Evaluation of the motorcycle rider course Accident
Analysis amp Prevention 198416(1)63ndash71
24Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2750
Mortimer 1988 published data only
Mortimer RG A further evaluation of the motorcycle rider courseJournal of Safety Research 198819(4)187ndash96
Osga 1980 published data only
Osga GA An investigation of the riding experiences of MSF rider
course participants Dissertation South Dakota USA The
University of South Dakota 19801ndash224
Perrino 2002 published data only
Perrino CS Ahmed A Callender A Rozier E Cantwell A Stewart
O The role of Marylandrsquos motorcycle rider course in promoting saver
behaviors and attitudes Baltimore USA National Transportation
Center Morgan State University 20021ndash52
Raymond 1979 published data only
Raymond S Tatum S An evaluation of the effectiveness of the RAC
ACU motorcycle training scheme - final report Road Safety ResearchUnit Dept of Civil Engineering University of Salford London
19791ndash131
Satten 1980 published data only
Satten RS Analysis and evaluation of the motorcycle rider courses
in thirteen Illinois counties International Motorcycle Safety
Conference Washington DC 1980145ndash93
Savolainen 2007 published and unpublished data
Savolainen Peter Mannering Fred Additional evidence on the
effectiveness of motorcycle training and motorcyclistsrsquo risk-taking
behaviour Transportation Research Board 2007 Vol March 20
19
Swaddiwudhipong 1998 published data only
Swaddiwudhipong W Boonmak C Nguntra P Mahasakpan P
Effect of motorcycle rider education on changes in risk behavioursand motorcycle-related injuries in rural Thailand Tropical Medicine
and International Health 19983(10)767ndash70
Waller 1992 published data only
Waller FL An analysis of the impact of the Texas Department of
Public Safety Motorcycle Operator Training Program upon
motocycle related traffic accidents Texas State University Public
Administration Program Applied Research Projects 199277
References to studies excluded from this review
Anderson 1980a published data only
Anderson LC Better biking a motorcycle rider improvement
program International Motorcycle Safety Conference Washington
DC USA Motorcycle Safety Foundation 1980231ndash53Aupetit 2008 published and unpublished data
Aupetit S Riff J Buttelli O Espie S Ergonomics research program
on motorcycle learning Contributions to the development of a
motorcycle simulator for training Transport Research Arena
Ljubljana 20089
Awane 1999 published data only
Awane T Integrating simulators in motorcycle safety education
Journal of International Association of Traffic and Safety Sciences 199923(1)26ndash35
Baldi 2005 published data only
Baldi S Baer JD Cook AL Identifying best practices states in
motorcycle rider education and licensing Journal of Safety Research200536(1)19ndash32
Batchler 1988 published data only
Batchler MW Evaluation of the pre-license motorcycle training course at the New South Wales traffic education centre Federal Office of
Road Safety (FORS) 19881ndash122 [ Report No CR 77 ]
Blanchard 2006 published data only
Blanchard HT Tabloski PA Motorcycle safety educating riders at
the teachable moment Mosby (CV) Company 200632330ndash2
Bonnett 2005 published data only
Bonnett J Mature age skills training for experienced motorcycle
riders Motorcycle ridersrsquo association of the Australian Capital
Territory 2005
Braver 2007 published data only
Braver ER et alPersuasion and licensure a randomized controlled
intervention trial to increase licensure rates among Maryland
motorcycle owners Traffic Injury Prevention 20078(1)39ndash46Clayton 1980 published data only
Clayton AB Sudlow DE An evaluation of the effectiveness of a one day training programme for learner motorcyclists Berkshire UK
Transport and Road Research Laboratory 19801ndash23 [ ISSN
0266ndash7045]lowast Clayton AB Sudlow DE An evaluation of the effectiveness of a one-day training programme for learner motorcyclists Berkshire UK
Road Safety Division Berkshire 19801ndash23 [ISSN 0266ndash7045]
Council 1978 published data only
Council FM Desper LP Dutt AK Driver education for motorcycle operation University of North Carolina Highway Safety Research
Center 19781ndash40
Crowther 2005 published data onlylowast Crowther G Engaging with motorcyclists UK police and theBikeSafe Road Safety Programme Published proceedings of the
Australasian Road Safety Research Policing Education Conference
Wellington New Zealand 2005 issue ISBN 0ndash473ndash10636ndash11ndash11
Daniello 2009 published data only
Daniello A Gabler HC Mehta YA Effectiveness of motorcycle
training and licensing Transportation Research Board 88th Annual
Meeting 200913
Dewanti 2007 published data only
Dewanti M Challenging efforts in promoting young motorcyclist
safety in Indonesia Road Safety on Four Continents 14th
International Conference Swedish National Road and Transport
Research Institute 2007
Du Plooy 1990 published data onlyDu Plooy DW Motorcycle training in South Africa The Human
Element International Motorcycle Safety Conference Orlando
Florida USA Motorcycle Safety Foundation 19904-1 to 4-9
Engel 1989 published data only
Engel U Krogsgaringrd Thomsen L [Faeligrdselsundervisning og
propaganda ndash en faeligrdselssikkerhedsfremmende foranstaltning]
Dansk Vejtidsskrift 198912295ndash7
Goldenbeld 2002 published data only
Goldenbeld C Houwing S de Craen S The development of
driving skills of young moped riders [De ontwikkeling van de
rijvaardigheid van jonge bromfietsers] SWOV Institute for Road
Safety Research (Stichting Wetenschappelijk Onderzoek
Verkeersveiligheid) Leidschendam the Netherlands 200263
25Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2850
Goldenbeld 2004 published data only
Goldenbeld C Twisk D de Craen S Short and long term effects of moped training a field experiment Transportation Research Part F 20047(1)1ndash16
Haworth 2000 published data only
Haworth N Smith R Kowadlo N Motorcycle training in Victoria
evaluation of rider training curriculum in Victoria Monash
University Accident Research Centre Reports 2000 Vol 165 issue
3131
Haworth 2005 published data only
Haworth N Mulvihill C Review of motorcycle licensing and
training Monash University Accident Research Centre 2005 [
Report no240]
Haworth 2006 published data only
Haworth N Mulvihill C Rowdon P Teaching old dogs new tricksTraining and older motorcyclists Journal of the Australasian College of Road Safety 200718(4)20ndash5lowast Haworth NL Mulvihill C Rowden PJ Teaching old dogs new
tricks Training and older motorcyclists Australasian Road Safety
Research Policing and Education Conference Gold Coast
Queensland 2006
Hill 1983 published data only
Hill D Motorcycle rider training testing and licensing Motor
Cycle Council of New South Wales Sydney 1983
Imatake 1980 published data only
Imatake Y On the activities in Japan of motorcycle safe riding
promotion committee (MSRPC) - especially the education of
unskilled riders of class 1 motorbikes International Motorcycle
Safety Conference Washington DC USA Motorcycle Safety
Foundation 1980255ndash68
Ingebritsen 1990 published and unpublished data
Ingebrigtsen S [Risikofaktorer ved ferdsel med moped og
motorsykkel] TOslashI-rapport 66 Oslo Transportoslashkonomisk
institutt 1990
Jain 2009 published and unpublished data
Jain A et alTwo wheeler accidents on Indian roads a study from
Mangalore India Elsevier 200916130ndash3
Jonah 1979 published data only
Jonah BA Dawson NE Validation of the motorcycle operator skill
test Accident Analysis amp Prevention 197911(3)163ndash71
Jonah 1981 published data onlyJonah BA Dawson NE Bragg WE Predicting accident
involvement with the Motorcycle Operator Skill Test Accident Analysis amp Prevention 198113(4)307ndash418
Kadar 1990 published data only
Kadar KJ Countermeasures motorcycle rider education delivery
system The Human Element International Motorcycle Safety
Conference Motorcycle Safety Foundation 19906-27 to 6-41
Kelsey 1986 published data only
Kelsey SL Liddicoat C Ratz M Licensing novice motorcyclists a
comparison of Californiarsquos standard test and the MOST II (Motorcycle
Operator Skill Test) administered at centralized testing offices Sacramento California USA California Department of Motor
Vehicles Research and Development Office 19861ndash74
Koch 1980 published data only
Koch H Advanced course for motorcycle riders InternationalMotorcycle Safety Conference Washington DC USA Motorcycle
Safety Foundation 1980269ndash81
Koch 1990 published data only
Koch H Brendicke R How to influence risk taking behaviour
within motorcycle rider training programs The Human Element
International Motorcycle Safety Conference Orlando Florida
USA Motorcycle Safety Foundation 1990Motorcycle Safety
Foundation 1990 p 9-1 to 9-25A
Labbett 2005 published data only
Labbett S Langham M Training can make the problem worse
Road Safety Congress 20051ndash12
Law 2005 published and unpublished data
Law TH Radin Umar RS Zulkaurnain S Kulanthayan S Impactof the effect of economic crisis and the targeted motorcycle safety
programme on motorcycle-related accidents injuries and fatalities
in Malaysia International Journal of Injury Control and Safety
Promotion 200512(1)9ndash21
Lowes 1990 published data only
Lowes BF In-traffic rider training The Human Element
International Motorcycle Safety Conference Orlando Florida
USA Motorcycle Safety Foundation 19907-46 to 7-57
Mahon 1976 published data only
Mahon J Teaching accident avoidance on the motorcycle Journal of Traffic Safety Education 1976231ndash29
Maxwell 1985 published data only
Maxwell GS Evaluation of a Program for Training MotorcycleRiders Evaluation and Program Planning 19858(4)339ndash47
Mayhew 1996 published data only
Mayhew DR Simpson SH Effectiveness and role of driver educationand training in a graduated licensing system Ottowa Canada The
Traffic Injury Research Foundation 19961ndash89 [ ISBN
0ndash920071ndash12ndash0]
McKnight 1987 published data only
McKnight AJ Milburn BG An evaluation of the modified
motorcycle operator skill test exercises NHTSA Report 1987 [
Report No DOT HS 807 212 ]
McPherson 1976 published data only
McPherson K McKnight AJ The development and evaluation of a
motorcycle skill test manual and knowledge test Washington DCUSA National Highway Safety Administration 1976
Mohan 2004 published data only
Mohan D Evidence-based interventions for road traffic injuries in
South Asia Journal of the College of Physicians and SurgeonsPakistan 200414(12)746ndash7
Morgan 1980 published data only
Morgan JK Advanced motorcycle training course content and
study evaluation International Motorcycle Safety Conference
Washington DC USA Motorcycle Safety Foundation 1980
283ndash312
Motoki 1990 published data only
Motoki M Yamazaki S A study on effective motorcycle rider
education The Human Element International Motorcycle Safety
26Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2950
Conference Orlando Florida USA Motorcycle Safety
Foundation 19909-26 to 9-57A
Nairn 1992 published data onlylowast Nairn RJ Motorcycle safety research literature review 1987 to
1991 Federal Office of Road Safety Australia 1993 Vol Report
CR 117
Newland 1980 published data only
Newland R Darby J Automotive practicespre motorcycle rider
program International Motorcycle Safety Conference Washington
DC USA Motorcycle Safety Foundation 1980127ndash9
Newland 1982 published data only
Newland R Motorcycle rider training and licensing requirements
Driver training steering a course for the future Melbourne
Victoria Transport Regulation Board 1982
Newland 1990 published data only
Newland R Motorcycle training and licensing in Victoria 1980-
1990 The Human Element International Motorcycle Safety
Conference Orlando Florida USA Motorcycle Safety
Foundation 1990
Ormston 2003 published data only
Ormston R Dudleston A Pearson S Stradling Evaluation of
Bikesafe Scotland Scottish Executive Social Research Edinburgh
2003
Osga 1979 published data only
Osga GS Ellingstad VS An evaluation of the effectiveness of the 1978
South Dakota motorcycle rider course program South Dakota USA
University of South Dakota 19791ndash89
Panou 2007 published data only
Panou M Bekiaris E A new concept on the integration of driving
simulators in driving training the train-all approach Road Safety
on Four Continents 14th International Conference Swedish
National Road and Transport Research Institute 2007
Peterson 1990 published data only
Peterson A Kadar K Status report 1980-1989 Minnesota rider
education statistics The Human Element International
Motorcycle Safety Conference Orlando Florida USA 199013-
113 to 13-129
Prem 1984 published data only
Prem H Good MC Motorcycle rider skills assessment ACT
Australia Office of Road Safety Department of Transport 1984
1ndash377 [ Report No CR 34 ]
Radin Umar 2006 published data only
Radin Umar RS Motorcycle safety programmes in Malaysia how
effective are they International Journal of Injury Control and Safety Promotion 200613(2)71ndash9
Rockwell 1990 published data only
Rockwell TH Kiger S Carnot M An evaluation of the Ohio
motorcyclist enrichment program Ohio Department of Highway
Safety 1990
Rothe 1987 published data only
Rothe JP A motorcycle riding course Motorcyclists Image and reality Canada Insurance Corporation of British Columbia 1987
59ndash83
Rowden 2007 published data only
Rowden P Watson B Haworth N What can riders tell us aboutmotorcycle training A view from the other side of the fence
Australasian Road Safety Research Policing and Education
Conference Melbourne Australia Melbourne 2007
RSC Victoria 1993 published data only
Road Safety Committee Inquiry into Motorcycle Safety in Victoria
Parliament of Victoria 1993
Russam 1979 published data only
Russam K Motorcycle training and licensing in Japan UK Transport
and Road Research Laboratory (TRRL) 1979 [ Report 916 ]
Saffron 1981 published data only
Saffron DG Driver and rider education training and licensing a
brief review Traffic Accident Research Unit Traffic Authority of
New South Wales NSW Australia Traffic Accident Research UnitTraffic Authority of New South Wales 1981
Schulz 1990 published data only
Schulz U Motorcycle accidents among learner riders Motorcycle
licensing Evaluation of the effectiveness of a Graduated Licensing Scheme Forshungshefte des Institut fur Zweiradsicherheit 1990
Sexton 2004 published data only (unpublished sought but not used)
Sexton B Baughan C Elliott M Maycock G The accident risk of
motorcyclists 2004TRL Report TRL 607(ISSN 09684107)58
Shephard 1986 published data only
Shepherd R Do Motorcycle Safety Foundation programs work
Driver 198619(9)16ndash8
Simpson 1990 published data onlySimpson HM Mayhew DR The promotion of motorcycle safety
training education and awareness Health Education Research
Theory and Practice 19905(2)257ndash64
Sokolov 1990 published data only
Sokolov O Safe motorcycle riding some aspects of motorcycle
riding in the USSR The Human Element International
Motorcycle Safety Conference Motorcycle Safety Foundation
19904ndash37 - 4-48
Steffens 1988 published data only
Steffens U Gawatz R Willmes G Effectiveness of mofa courses
[Wirksamkeit von Mofakursen] Unfall- und sicherheitsforschung
strasenverkehr 198867168
Sudlow 2003 published data only
Sudlow D Scoping study on motorcycle training Road Safety
Research Report London UK The British Institute of Traffic
Education Research Department for Transport 2003 Vol June
[ Report No 36]
Tasmania 1995 published data only
Mannering FL Grodsky LL Motorcycle rider training scheme
Traffic Education Research Centre Armidale Australia Driver
Training Australia Limited 1995 Vol April1ndash56
Tetard 1985 published data only
Tetard C Motorcycle ridersrsquo training to detect indices in an
intersection situation [Lrsquoentrainement a la prise drsquoinformation dans
lrsquoapprentissage de la conduite motocycliste] Le Travail Humain198548(2)147ndash59
27Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3050
Thackray 1980 published data only
Thackray RM Prescott JC Field test of the motorcycle ridercourse International Motorcycle Safety Conference Washington
DC USA Motorcycle Safety Foundation 1980195ndash7
Thompson 1994 published data only
Thompson M Evaluation of compulsory basic training for
motorcyclists Transport Research Laboratory Crowthorne
Berkshire UK 1994 [ Report 63]
Tobler 1990 published data only
Tobler U Individual driving instruction for motorcycle driving
learners the Swiss experiment of motorcycle driving school The
Human Element International Motorcycle Safety Conference
Orlando Florida USA Motorcycle Safety Foundation 1990
Transport 1989 published data only
Transport Communications Road Crash Statistics Australia Bureauof Transport and Communications Economics 198914ndash20
Tuan 2007 published data only
Tuan VA Shimizu T Towards development and evaluation of the
motorcycle drivers re-education program in Vietnam modeling of
motorcycle driverrsquos undesired behaviors 11th World Conference
on Transport Research 200725
Victor 1980 published data only
Victor DJ Traffic safety education for motorcyclists International
Motorcycle Safety Conference Washington DC USA Motorcycle
Safety Foundation 1980221ndash8
Waterer 1990 published data only
Waterer TH Compulsory training for motorcyclists from liberty
to compulsion The Human Element International Motorcycle
Safety Conference Orlando Florida USA Motorcycle Safety
Foundation 19904-49 to 4-57
Weller 1989 published data only
Weller RB Chandler EW Motorcycle safety and motorcycle
education past research and survey results Journal of Public Policy
amp Marketing 1989893ndash108
Wells 1981 published data only
Wells P Observations of some RACACU motorcycle training
schemes Transport and Road Research Laboratory Crowthorne
Berkshire UK 1981
Winn 1986 published data only
Winn G Two important trends in motorcycle safety regulations
rider education and conspicuity improvement American
Motorcyclist Association 1986147ndash52
Wisher 1988 published data only
Wisher G Reid J Motorcycle accidents in Norfolk a three-year study of motorcycle training and the Part 1 test Norfolk UK Road Safety
Division Highways Department Norfolk County Council 1988
Wood 1987 published data only
Wood HT Bowen R Evaluation of the revised learner permit scheme
July 1983 to December 1985 Victoria Australia Road Traffic
Authority 19871ndash27 [ ISBN 0 7306 0247 8]
Yamazaki 1990 published data only
Yamazaki S Nakanishi M Survey on effect of Honda rider
education The Human Element International Motorcycle Safety
Conference Orlando Florida USA 199012-167 to 12-187
References to ongoing studies
Burgess 2010 unpublished data only
Burgess C Broughton P Fylan F Stradling Interim evaluation of
the UKs National Ride program In L Dorn editor(s) Driver
behaviour and Training Vol IV Aldershot Ashgate Publishing
Limited 2010
Additional references
Anderson 1980
Anderson J Ford JL Peck RC Improved motorcyclist licensing and
testing project Volume I and II National Highway Traffic Safety
Administration US department of Transportation Washington
DC 1980
Bjornstig 1985
Bjornstig U Bylund P Lekander T Brorsson B Motorcycle fatalitiesin Sweden Acta Chirurgica Scandinavica 1985151(7)577ndash81
Clayton 1990
Clayton A Sudlow DE An evaluation of the effectiveness of a one-
day training program for learner motorcyclists The Human
Element International Motorcycle Safety Conference Orlando
Florida USA 199013-105 to 13-111
Cochrane Collaboration 2008
Higgins JP Green S Cochrane Handbook for Systematic Reviews of
Interventions 500 The Cochrane Collaboration 2008
Elliot 2003
Elliott MA Baughan CJT Broughton J Grayson G Knowles J et
alMotorcycle Safety a Scoping Study TRL [ Report No
0968ndash4107]Fleiss 2004
Fleiss JL Levin B Paik MC Statistical methods for rates and proportions 3rd Edition John Wiley amp Sons Inc 2004 [ ISBN
978ndash0ndash471ndash52629ndash2 ]
FORS 1999
FORS Federal Office of Road Safety Road risk for sober licensed
motorcyclists Vol Monograph 27 Federal Office of Road Safety
19991ndash2
Haworth 2005
Haworth N Mulvihill C Review of motorcycle licensing and training
Victoria Australia Monash University Accident Research Centre
20051ndash83 [ Report No 240]
Huwiler-Muumlntener 2002Huwiler-Muumlntener K Juumlni P Junker C Egger M Quality of
Reporting of Randomized Trials as a Measure of Methodologic
Quality JAMA 20022872801ndash4
Johnston 2008
Johnston P Brooks C Savage H Fatal and serious road crashes
involving motorcyclists Road Safety Research and Analysis Report
ATSB Monograph 20 Canberra Department of Infrastructure
Transport Regional Development and Local Government 0205
081ndash26
Juumlni 2001
Juumlni P Altman DG Egger M Systematic reviews in health care
Assessing the quality of controlled clinical trials BMJ 2001323
42ndash6
28Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3150
Kopits 2003
Kopits E Cropper M Traffic fatalities and economic growth TheWorld Bank Development Research Group Infrastructure and
Environment Policy research working paper 3035 2003
Leung 1983
Leung P Chau BTW Motorcycle accidents in Hong Kong pattern
of injury treatment causative factors and socio-economic
implications Injury 198315(1)23ndash9
Liu 2004
Liu B Ivers R Norton R Blows S Lo SK Helmets for preventing
injury in motorcycle riders Cochrane Database of Systematic Reviews 2004 Issue 2 [DOI 10100214651858CD004333pub3]
Mohan 2002
Mohan D Road safety in less-motorized environments future
concerns International Journal of Epidemiology 200231(3)527ndash32Mortimer 1982
Mortimer RG An evaluation of the motorcycle rider course
Proceedings of the 20th conference of the American Association for
Automotive Medicine Arlington Heights IL USA American
Association for Automotive Medicine 198277ndash96
Mullin 1997
Mullin BT Risk factors for motorcycle injury the role of age
gender experience training and alcohol PhD thesis Auckland
New Zealand University of Auckland 1997
Noordzij 2001
Noordzij PC Forke E Brendicke R Chinn BP Integration of needs
of moped and motorcycle riders into safety measures review and
statistical analysis in the framework of the European research
project PROMISING SWOV Institute for Road Safety Research
Leidschendam The Netherlands 2001 [ Report No Dndash2001ndash5]
Osga 1979
Osga G Ellingstad VS An evaluation of the effectiveness of the 1978 South Dakota Motorcycle Riders Course Program South Dakota
University of South Dakota Vermillion Human Factors Lab 1979
Oxford English Dictionary Online
httpoxforddictionariescom Oxford English Dictionary Online
Oxford University Press 2010
Peden 2004
Peden M Scurfield R Sleet D World Report on Road TrafficInjury Prevention Geneva World Health Organization Available
at httpwwwwhointviolence˙injury˙preventionpublications
road˙trafficworld˙reporten 2004Chapter 3
Radin Umar 1995
Radin Umar R Macakay G Hills B Preliminary analysis of
motorcycle accidents Short-term impacts of the running
headlights campaign and regulations in Malaysia J Traffic Med 199523(1)17ndash27
Tsai 1995
Tsai YJ Wang JD et alCase-control study of the effectiveness of
different types of helmets for the prevention of head injuries among
motorcycle riders in Taipei Taiwan American Journal of
Epidemiology 1995142(9)974ndash81
Vis 1995
Vis AA In-depth study of the hazards of motorcycling [De
onveiligheid van motorrijden nader bekeken] SWOV Institute for
Road Safety Research Leidschendam the Netherlands 1995 [
Report No Rndash95ndash69]
Weaver 1990
Weaver EA Motorcycle safety education in the United States
Motorcycle Safety Foundation Orlando Florida USA 1990
Winn 1987
Winn G Two important trends in motorcycle safety regulations
rider education and conspicuity improvement No more
information available Detroit MI US Society of Automotive
Engineers 1987147ndash52
Yuan 2000
Yuan W The effectiveness of the rsquoride-brightrsquo legislation for
motorcycles in Singapore Accident Analysis amp Prevention 200032
(4)559ndash63
Zambon 2006
Zambon F Hasselberg M Factors affecting the severity of injuries
among young motorcyclists a Swedish nationwide cohort study
Traffic Injury Prevention 20067(2)143ndash9lowast Indicates the major publication for the study
29Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3250
C H A R A C T E R I S T I C S O F S T U D I E S
Characteristics of included studies [ordered by study ID]
Adams 1985
Methods Non-randomised trial
Participants Initial sample size
Total of 162 Australia Post telegram delivery riders
Not clear how many in intervention and control groups
Analysis sample size
Intervention group 63Control group 35
Total 98 (61)
Interventions No formal training vs voluntary rider training course of 12 hours over 2 consecutive days (theory 4 hours practice
8 hours)
Intervention integrity
No information given
Outcomes Australia Post recorded crashes
Length of follow up 1 year
Notes Study conducted in NSW Australia
Anderson 1980
Methods Randomised trial
Participants Initial sample size
Total 41188 motorcycle licence applicants
Analysis sample size
Intervention group with skills training 13160
Intervention group without skills training 12634
Control group 15080
Total 40874 (99)
Interventions Standard licensing skill test vs the Motorcycle Operators Skill Test (MOST) with or without a training component
MOST is not to be confused with rider training because it is a licensing test The training component was for
participants who failed the skill test on the first attempt was of 3 hours duration and consisted of instructions about
the skills needed to pass the skill test The training consisted of 30 min theory and 2 frac12 hours practice on the
motorcycle
Intervention integrity
33 in MOST with skills training 31 in MOST without skills training and 43 in control group obtained their
motorcycle licence in the end
Outcomes Police recorded crashes and offences
Length of follow up 3 years
30Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3350
Anderson 1980 (Continued)
Notes Study conducted in California USA
Billheimer 1998
Methods Cohort study
Participants Initial sample size
Intervention group Unclear numbers of CMSP graduates and ERC graduates
Control group Over 16000 untrained motorcyclists
Analysis sample size
1139 matched CMSP-pairs and 1182 ERC-pairs
Interventions No formal training vs the mandatory (for those younger than 18 years) pre-licence 16-hours basic California
Motorcyclist Safety Program (CMSP) and the 8-hour Experienced Rider Course (ERC)
Intervention integrity
No information given
Outcomes Police recorded crashes and offences
Length of follow up 2 years
Notes Study conducted in California USA
Cooper 1988
Methods Cohort study
Participants Initial sample size
Intervention group 418 graduates and 57 who failed
Control group 402
Analysis sample size
Intervention group unclear
Control group unclear
Total 863 (984)
Interventions No formaltrainingvs non-mandatory pre-licence rider trainingThe content of thetrainingwas notfurther explainedIntervention integrity
Participants had all completed the rider training
Outcomes Police recorded crashes and offences
Length of follow up 7 years
Notes Study conducted in British Columbia Canada
31Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3450
Davis 1997
Methods Retrospective cohort study
Participants Initial sample size
Intervention group 9320 graduates
Control group unclear
Analysis sample size
Intervention group 9320 graduates
Control group unclear
Interventions No formal training vs the non-mandatory Connecticut Motorcycle Operator Training Program The content of
training was not further explained
Intervention integrity Participants had all completed the rider training
Outcomes Police recorded crashes severity of crashes
Length of follow up 1 year
Notes Study conducted in Connecticut USA
Hall 1988
Methods Randomised trial
Participants Initial sample sizeIntervention group A 6248
Intervention group B 6616
Intervention group C 6996
Control group 6604
Total 26464 motorcycle licence applicants
Analysis sample size
Intervention group A 6012 (96)
Intervention group B 6428 (97)
Intervention group C 6778 (97)
Control group 6502 (98)
Interventions Standard licensing skill test vs the Motorcycle Operator Skill Test (MOST II) without training (A) MOST II with
3 hours training (B) or MOST II with 20 hours training (C) The 3 hours training comprised instructions aboutthe skills needed to pass the skill test (30 min theory and 2 frac12 hours practice on the motorcycle) The 20 hours
training consisted of the Motorcycle Safety Foundationrsquos Motorcycle Rider Course (MRC) comprising of 8 hours
theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-road practice
Intervention integrity
25 and 26 obtained their motorcycle licence in the lsquoMOST 3-hour trainingrsquo and lsquoMOST 20-hour trainingrsquo
intervention arm respectively In the lsquoMOST testrsquo intervention arm 29 obtained their licence and in the control
group 35
Outcomes Police recorded crashes and offences
Length of follow up 3 years
32Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3550
Hall 1988 (Continued)
Notes Study conducted in New York USA
Haworth 1997
Methods Population based case-control study
Participants Cases Motorcycle riders or pillion riders involved in motorcycle crash (on road) Recruited from hospital and police
records Not clear whether all cases (especially fatalities) were included
Community controls Multiple matched sample of motorcycle rider or pillion who passed the crash site at the same
time of day and day of week that the crash occurred
Initial sample size
Cases 336
Controls 1195
Analysis sample size
Cases 234 (70)
Controls 561 (47)
Interventions Basic handling skills course beginner intermediate or advanced course measured through self-report not validated
Outcomes Injuries with an Injury Severity Score of 5 or greater or admission for at least four hours and fatalities from a
motorcycle crash (from hospital and police records)
Length of follow up two years or more prior to the interview
Notes Study conducted in Victoria Australia
Hurt 1981
Methods Population based case-control study
Participants Cases Motorcycle riders who had a motorcycle crash that was reported to police (traffic accident reports) Not all
cases included
Community controls Multiple matched sample of motorcycle riders who passed the crash site at the same time of
day and day of week as the crash occurred but 2 years later
Initial sample size
Cases 4500 notifications of crashes to police (only 50 received in time research team able to respond to 1126 for
unreported reasons data collection completed for 900)
Controls Number approached unreported 2310 consented and interviewed
Analysis sample size
Cases 808 (808900 data on rider training) - 18 of initial sample
Controls 670 (6702310 data on rider training) - lt 29 of initial sample
Interventions Training experience self taught friends-family school-club formal-professional others measured through self-
report validated
Outcomes Injuries and fatalities from a motorcycle crash (from hospital records and coronerrsquos office)
33Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3650
Hurt 1981 (Continued)
Notes Study conducted in Los Angeles USA
Jonah 1982
Methods Cohort study
Participants Initial sample size
Intervention group
2310 licensed MTP graduates who are active riders
Control group
2029 licensed motorcyclists who are active riders
Non-licensed and non-active riders excluded
Analysis sample size
Intervention group
811 interviews (35) 533 driver records (23)
Control group 1080 interviews (71) 905 driver records (45)
Interventions No training vs the non-mandatory Canada Safety Council Motorcycle Training Program (MTP) comprising of 4
hours theory and 16 hours practise (6 hrs on-road) over 2 weekends
Intervention integrity
Participants had all successfully completed the MTP
Outcomes Self-reported and police reported crashes and offencesLength of follow up 4 years
Notes Study conducted in Ontario Canada
Kloeden 1994
Methods Cohort study
Participants Initial sample size
Intervention group 2533 licence applicants
Control group Unclear
Sample of riders obtaining probationary licenceIntervention group 1625 (64)
Control group 5015 (unclear)
Non-active riders excluded from control group (~66)
Analysis sample size
Intervention group 631 (39)
Control group 318 (6)
Interventions No formal training vs the mandatory pre-licence Ridersafe course consisting of two levelsthe first is a training course
in the basics of motorcycle riding where a pass entitles the rider to obtain a learnerrsquos permit The second is a skill
test where a pass entitles the rider to obtain a probationary motorcycle licence The training course consists of two 4
hour training sessions on separate days involving instruction videos and actual riding on motorcycles The skill test
consists of a 4 hour preparation course of 74 minutes of lecturediscussion 38 minutes of video and 75 minutes of
34Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3750
Kloeden 1994 (Continued)
on-course riding which is the testing phase
Intervention integrity
Unclear what percentage in the Ridersafe group and the control group obtained their probationary licence halfway
and their full motorcycle licence in the end
Outcomes Police reported crashes
Length of follow up 1 year
Notes Study conducted in SA Australia
Kraus 1975
Methods Population based case-control study
Participants Cases All motorcycle riders with a confirmed medically treated motorcycle injury (including fatalities) from all
hospitals in study base and peripheral area hospitals and all deaths due to motorcycle crashes in the study base All
serious injuries (leading to hospitalisation) and fatalities likely to be included
Community controls A random sample of registered motorcycle owners in the same temporary and geographical
study base as the cases arose from
Initial sample size Cases 1273
Controls 738
Analysis sample size Cases 741 (58)
Controls 434 (59)
Interventions Motorcycle rider training (not specified) measured through self-report
Outcomes Injuries leading to hospitalisation fatalities
Length of follow up not specified
Notes Study conducted in Sacramento county USA
Leung 1987
Methods Non-randomised trial
Participants Initial sample sizeIntervention group unknown number of MRC graduates
Control group unknown number of licensed motorcyclists
Analysis sample size
Intervention group 2941
Control group 43094
Interventions No formal training vs non-mandatory Wisconsin motorcycle rider course a 22 hours off-street programme that
teaches the basic rules of safe motorcycling
Intervention integrity
50 in the MRC group and 100 in the control group obtained their motorcycle licence in the end
35Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3850
Leung 1987 (Continued)
Outcomes Police recorded crashes and offences
Length of follow up up to 3 years
Notes Study conducted in Wisconsin USA
MAIDS 2004
Methods Population based case-control study
Participants Cases Motorcycle riders who had a motorcycle crash Unclear whether all cases included
Community controls Motorcycle riders identified at petrol stations Unclear whether matched to cases
Initial sample size
Cases not reported
Controls not reported
Analysis sample size
Cases 921 (Italy 200 Netherlands 200 Spain 121 Germany 250 France 150)
Controls 923 (Italy 200 Netherlands 200 Spain123 Germany 250 France 150)
Interventions Pre-license rider training or additional training (non-compulsory)
Outcomes Crashes
Notes Study in five sampling areas located in five European countries (France Germany Netherlands Spain and Italy)
McDavid 1989
Methods Cohort study
Participants Sample size after first match
Intervention group 346 male training graduates
Control group 346 male motorcycle riders licensed in same year
Sample size after second match
Intervention group 139 (40)
Control group 139 (40)
Analysis sample size
Intervention group 139 (40)
Control group 139 (40)
Interventions No formal training vs the non-mandatory pre-licence British Columbia Safety Councilrsquos 37-hour motorcycle safety
programme
Intervention integrity
Participants had all completed and passed the training
Outcomes Police reported crashes injurious crashes
Length of follow up 1-6 years
36Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3950
McDavid 1989 (Continued)
Notes Study conducted in British Columbia Canada
Mortimer 1984
Methods Cohort study
Participants Initial sample size
Intervention group 608 MRC graduates
Control group 360 motorcycle riders
Response rates
Intervention group 360 (59)
Control group 324 (90)
Non-active riders excluded
Analysis sample size
Intervention group 213 (35)
Control group 303 (84)
Interventions No formal training vs the non-mandatory Motorcycle Safety Foundationrsquos Motorcycle Rider Course comprising of
8 hours theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-road practice
Intervention integrity
Participants had all completed and passed the MRC
Outcomes Self-reported crashes and offencesLength of follow up 1 year
Notes Study conducted in Illinois USA
Mortimer 1988
Methods Cohort study
Participants Initial sample size
Intervention group 2307 MRC graduates
Control group 556 active licensed motorcycle riders
Response ratesIntervention group 1298 (56)
Control group 500 (90)
Non-active riders excluded
Analysis sample size
Intervention group 913 (40)
Control group 500 (90)
Interventions No formal training vs the non-mandatory Motorcycle Safety Foundationrsquos Motorcycle Rider Course comprising of
8 hours theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-road practice
Intervention integrity
Participants had all completed and passed the MRC
37Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4050
Mortimer 1988 (Continued)
Outcomes Self reported crashes and offences
Length of follow up 1 year
Notes Study conducted in Illinois USA
Osga 1980
Methods Cohort study
Participants Initial sample sizeIntervention group 1163 MSF MRC participants invited 818 (74) responded
Control group 329 motorcycle riders responded to recruitment strategies
Non-active riders excluded from analysis
Analysis sample size
Intervention group 595 (73)
Control group 329 (100)
Interventions No formal training vs the non-mandatory Motorcycle Safety Foundationrsquos Motorcycle Rider Course (MRC) com-
prising of 8 hours theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-road
practice
Intervention integrity
Participants had all successfully completed the MRC
Outcomes Self-reported crashes police recorded crash reports
Length of follow up 6 months - 3 12 years
Notes Study conducted in South Dakota USA
Perrino 2002
Methods Cohort study
Participants Initial sample size
Intervention group 2144 (unclear) MRC graduates (motorcycle licence applicants)
Control group 2601 (unclear) regular licensing test graduates (motorcycle licence applicants)Analysis sample size
Intervention group 1070 (unclear)
Control group 867 (unclear)
Interventions No formal training vs the mandatory (for those younger than 18 years) pre-licence Maryland Motorcycle Rider
Course (MRC) consisting of 8 hours theory and 12 hours practice
Intervention integrity
No information given
Outcomes Self-reported and police recorded crashes and offences
Length of follow up 1 year
38Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4150
Perrino 2002 (Continued)
Notes Study conducted in Maryland USA
Raymond 1979
Methods Cohort study
Participants Initial sample size
Intervention group
258 volunteers mainly (but not all) novice riders
Control group
1104 provisional licence applicants
Analysis sample size
Intervention group 190 (74)
Control group 627 (57)
Interventions No formal training vs the non-mandatory RACACU (Royal Automobile ClubAuto-Cycle Union) training scheme
The training program consisted of either 24 one-hour sessions or 13 two-hour sessions Twelve hours of the training
focused on theory about basic operational procedures traffic laws and highway code The practical session focused
on basic operational skills riding on a training area and riding in traffic with an instructor A proficiency test was
administered at the end of the course
Intervention integrity
No information given
Outcomes Self-reported crashes (including minor non-injury non-damage and off-road crashes)
Length of follow up 1 year
Notes Study conducted in the UK
Satten 1980
Methods Cohort study
Participants Initial sample size
Intervention group interviewers kept interviewing until they had 100 riders selected from a population of 891
enrollees in the MSF MRC course (unknown how many were asked and how many refused to participate)Control group unknown number of registered motorcycle riders
Non-active riders excluded from analysis
Analysis sample size
Intervention group 71
Control group 100
Interventions No formal training vs the non-mandatory Motorcycle Safety Foundationrsquos (MSF) Motorcycle Rider Course (MRC)
comprising of 8 hours theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-
road practice
Intervention integrity
Participants had all successfully completed the MRC
39Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4250
Satten 1980 (Continued)
Outcomes Self-reported crashes injuries and offences
Length of follow up 3 years
Notes Study conducted in Illinois USA
Savolainen 2007
Methods Cohort study
Participants Initial sample size
Intervention group 4000 riders that had conducted the MSF MRC course (passed or failed)
Control group 4000 motorcycle owners or permit or endorsement holders from the Indiana Bureau of Motor
Vehicles database
Analysis sample size
Intervention group 739 (18)
Control group 588 (15)
Interventions Intervention integrity
No information given
Outcomes Probability of crash involvement
Length of follow up up to 5 years
Notes Study conducted in Indiana USA
Swaddiwudhipong 1998
Methods Randomised trial
Participants Initial sample size
Intervention group
all 16 villages of 3 sub-districts
(participants were practising motorcyclists licensed and unlicensed)
Control group
18 villages of 3 other sub-districtsAnalysis sample size
Intervention group 3 districts (100)
Control group 3 districts (100)
Interventions No rider education vs non-mandatory community based rider education consisting of theory on motorcycle injury
risk helmet use traffic laws vehicle regulations and traffic signs and a written and skill test leading to a licence
Intervention integrity
No information given
Outcomes Injuries fatalities
Length of follow up 2 years
40Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4350
Swaddiwudhipong 1998 (Continued)
Notes Study conducted in Tak Province rural Thailand
Waller 1992
Methods Cohort study
Participants Initial sample size
Intervention group 349 riders who succesfully completed the Department of Public Safety motorcycle operator
training in 1988
Control group 349 motorcyclists who never completed the DPS motorcycle operator training
Analysis sample size
Intervention group 349
Control group 349
Interventions No DPS motorcycle operator training vs non-mandatory DPS motorcycle operator training The DPS training is
the Motorcycle Safety Foundationrsquos Motorcycle Rider Course comprising eight hours of theory (lectures discussions
films and multiple choice tests) a textbook and 12 hours off-road practice
Intervention integrity
All 349 riders succesfully completed the rider training
Outcomes Police recorded crashes
Length of follow up 3 years
Notes Study conducted in
Characteristics of excluded studies [ordered by study ID]
Study Reason for exclusion
Anderson 1980a Study type review and process evaluation of a rider course
Aupetit 2008 Outcome proposal of a rider training curriculum that combines traditional training and a motorcycle riding
simulator
Awane 1999 Study type review on use of simulators in rider and driver training
Baldi 2005 Intervention not rider training but intervention integrity of rider training availabilty of rider education
implementation of graduated licensing system etc on a state-wide level
Batchler 1988 Study type description of a training course
Blanchard 2006 Study type not a study assessing the association between rider training and outcomes
Bonnett 2005 Outcome qualitative data regarding satisfaction with the training programme
41Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4450
(Continued)
Braver 2007 Outcome licensure rates
Clayton 1980 Outcome knowledge skills attitude
Council 1978 Outcome crash reduction in ANY vehicle not motorcycles exclusively
Crowther 2005 Study type qualitative study
Daniello 2009 Study type review of the effectiveness of motorcycle training and licensing
Dewanti 2007 Intervention emotional characteristics behaviour (ie not motocycle rider training)
Du Plooy 1990 Study type review of types of rider training courses in South Africa
Engel 1989 Study type uncontrolled before-after study
Goldenbeld 2002 Outcome skills
Goldenbeld 2004 Outcome knowledge and skills
Haworth 2000 Intervention balance between attitudinal andvehicle control skills acrosscurriculla of motorcycle rider training
courses
Haworth 2005 Study type review
Haworth 2006 Study type cross-sectional survey
Hill 1983 Study type review and description of training system
Imatake 1980 Study type course descriptionreview
Ingebritsen 1990 Intervention risk factors for motorcycle crashes
Jain 2009 Study type cross-sectional descriptive study of motorcycle crashes over a 5 year period
Jonah 1979 Study type cross-sectional survey Outcome scores on motorcycle operator skill test
Jonah 1981 Intervention evaluation of a motorcycle rider skill test not motorcycle rider training
Kadar 1990 Study type review of components of rider training courses
Kelsey 1986 Intervention evaluation of a motorcycle rider skill test not motorcycle rider training
Koch 1980 Study type reviewcourse description
Koch 1990 Study type description of course content
42Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4550
(Continued)
Labbett 2005 Study type essay about possible explanations for motorcycle deaths
Law 2005 Intervention targeted motorcycle safety programme not specifically rider training
Lowes 1990 Study type review and descriptions of rider training
Mahon 1976 Study type review of motorcycle rider training methods
Maxwell 1985 Outcomesrecommendations forprogramme operatorsand otherson howthe motorcycle trainingprogramme
works and how it can be improved
Mayhew 1996 Study type review
McKnight 1987 Study type evaluation of methodological and administrative aspects of a motorcycle operator skill test
McPherson 1976 Outcome knowledge and skills
Mohan 2004 Study type review of evidence of interventions to reduce road traffic injuries
Morgan 1980 Study type reviewcourse description
Motoki 1990 Study type proposal and development of rider training curriculum
Nairn 1992 Study type literature review
Newland 1980 Study type description of development of motorcycle safety course
Newland 1982 Study type descriptionreview of a rider training course
Newland 1990 Study type uncontrolled BeforeAfter study
Ormston 2003 Outcome changes in attitude and self reported behaviour
Osga 1979 Outcome knowledge attitudes and skills
Panou 2007 Intervention new simulation tool to teach driverrider
Peterson 1990 Study factor demographic details of course participants completing the course vs those not completing the
course
Prem 1984 Study type experimental
Study factor examines association between skills and performance in tests
Radin Umar 2006 Intervention motorcycle safety programme not specifically rider training
Rockwell 1990 Study type methodologically flawed study type
43Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4650
(Continued)
Rothe 1987 Study type qualitative study of rider training in general
Rowden 2007 Study type qualitative study
RSC Victoria 1993 Study type review of motorcycle rider training and education in Victoria
Russam 1979 Study type review of training programmes in Japan
Saffron 1981 Study type literature review
Schulz 1990 Study factor predictors of motorcycle crashes
Sexton 2004 Study type cross-sectional analytic analysis
Shephard 1986 Study factor reviewdescription of programme
Simpson 1990 Study type review of motorcycle rider trainingeducation
Sokolov 1990 Study factor review of motorcycle riding courses in USSR
Steffens 1988 Outcome self-reported behaviour experiences and researcher observed riding behaviour
Sudlow 2003 Study type review of motorcycle training in the UK
Tasmania 1995 Study type cross-sectional descriptiveprocess evaluation
Tetard 1985 Study type experimental
Outcome detection of indices
Thackray 1980 Outcome instructional effectiveness user acceptance administrative feasibility
Thompson 1994 Study type cross-sectional survey
Outcome attitudes and knowledge
Tobler 1990 Study type review of course curricula
Transport 1989 Intervention syllabus for rider training
Tuan 2007 Study type cross-sectional study (survey) which does not assess the effects of an intervention but examines
motocyclistsrsquo behaviour
Victor 1980 Study type review
Waterer 1990 Study type review of UK rider training issues
Weller 1989 Study type review of studies evaluating motorcycle safety and education measures
44Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4750
(Continued)
Wells 1981 Study type description of different methods of training
Winn 1986 Study type review of rsquohistoryrsquo of rider training
Wisher 1988 Study type cross-sectional survey
Wood 1987 Study type Interrupted Time Series but did not meet the criteria for an ITS as published by the Cochrane
Effective Practice and Organisation of Care Group
Yamazaki 1990 Study type cross-sectional survey
Outcome riding skills skills retention and attitude
Characteristics of ongoing studies [ordered by study ID]
Burgess 2010
Trial name or title UKs National Ride program
Methods Cohort study
Participants Motorcycle offenders
Interventions Police led diversion program
Outcomes Self reported attitudinal and behavioural measures
Starting date
Contact information
Notes Expected to be completed in early 2010
45Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4850
D A T A A N D A N A L Y S E S
This review has no analyses
A P P E N D I C E S
Appendix 1 Search strategy
We searched the following electronic databases
bull Cochrane Injuries Group Specialised Register (searched 22 Sept 2008)
bull CENTRAL (The Cochrane Library 2008 Issue 3)
bull TRANSPORT 1988 to 200706 (latest date available)
bull ATRI Australian Transport Index (to Sept 2008)
bull MEDLINE (Ovid SP) 1950 to Sept 2008
bull EMBASE (Ovid SP) 1980 to Sept 2008
bull CINAHL (EBSCO) 1982 to Sept 2008
bull WHOLIS (World Health Organization Library Information System) (to Sept 2008)
bull PsycINFO (Ovid SP) 1806 to Sept 2008
bull LILACS (BIREME) (Latin American and Caribbean Health Sciences) (1982 to Sept 2008)
bull ISI Web of Science Social Sciences Citation Index (SSCI) 1970 to Sept 2008
bull ERIC (searched 25 Sept 2008)
bull ZETOC (searched 25 Sept 2008)
bull SIGLE (searched 25 Sept 2008)
All of the databases were searched using the following text-word search strategy The strategy was adapted using either rsquorsquo or rsquo$rsquo as atruncation symbol
1 motor-cycl or motorcycl or motor-bik or motorbik or motor-scooter or motorscooter or scooter or moped or motocycl
or motocicl or moto
2 train or educat or instruct or program or course or qualif or learn or improv
3 1 and 2
H I S T O R Y
Protocol first published Issue 2 2005
Review first published Issue 10 2010
46Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4950
C O N T R I B U T I O N S O F A U T H O R S
Katina Kardamanidis was the first author of the protocol conducted the literature search extracted full studies assessed the studies
according to inclusion criteria assessed quality of the studies compared findings with the second author discussed disagreements with
the third author wrote the drafts and the final review
Rebecca Ivers was the second author of the protocol assessed the studies according to inclusion criteria assessed quality of the studies
compared findings with the first author edited the final draft and provided overall guidance for the process
Alexandra Martiniuk assessed quality of the studies compared findings with the first author discussed disagreements with the third
author and helped write the first draft and the final review
Mark Stevenson contributed to the protocol reviewed methods of studies and contributed to the final draft
Katrina Thistlethwaite conducted a second literature search extracted full studies and contributed to the final draft
D E C L A R A T I O N S O F I N T E R E S T
None known
S O U R C E S O F S U P P O R T
Internal sources
bull No sources of support supplied
External sources
bull National Health and Medical Research Council of Australia Australia
Salary support for Martiniuk Ivers and Stevenson
bull University of Sydney Sesqui Research Grant Australia
Direct research costs and salary support for Kardamanidis
D I F F E R E N C E S B E T W E E N P R O T O C O L A N D R E V I E W
There are no differences between the protocol and the review
I N D E X T E R M S
Medical Subject Headings (MeSH)
lowastMotorcycles Accident Prevention [lowastmethods] Accidents Traffic [lowastprevention amp control] Licensure Program Evaluation Wounds
and Injuries [mortality lowastprevention amp control]
47Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 5050
MeSH check words
Humans
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2750
Mortimer 1988 published data only
Mortimer RG A further evaluation of the motorcycle rider courseJournal of Safety Research 198819(4)187ndash96
Osga 1980 published data only
Osga GA An investigation of the riding experiences of MSF rider
course participants Dissertation South Dakota USA The
University of South Dakota 19801ndash224
Perrino 2002 published data only
Perrino CS Ahmed A Callender A Rozier E Cantwell A Stewart
O The role of Marylandrsquos motorcycle rider course in promoting saver
behaviors and attitudes Baltimore USA National Transportation
Center Morgan State University 20021ndash52
Raymond 1979 published data only
Raymond S Tatum S An evaluation of the effectiveness of the RAC
ACU motorcycle training scheme - final report Road Safety ResearchUnit Dept of Civil Engineering University of Salford London
19791ndash131
Satten 1980 published data only
Satten RS Analysis and evaluation of the motorcycle rider courses
in thirteen Illinois counties International Motorcycle Safety
Conference Washington DC 1980145ndash93
Savolainen 2007 published and unpublished data
Savolainen Peter Mannering Fred Additional evidence on the
effectiveness of motorcycle training and motorcyclistsrsquo risk-taking
behaviour Transportation Research Board 2007 Vol March 20
19
Swaddiwudhipong 1998 published data only
Swaddiwudhipong W Boonmak C Nguntra P Mahasakpan P
Effect of motorcycle rider education on changes in risk behavioursand motorcycle-related injuries in rural Thailand Tropical Medicine
and International Health 19983(10)767ndash70
Waller 1992 published data only
Waller FL An analysis of the impact of the Texas Department of
Public Safety Motorcycle Operator Training Program upon
motocycle related traffic accidents Texas State University Public
Administration Program Applied Research Projects 199277
References to studies excluded from this review
Anderson 1980a published data only
Anderson LC Better biking a motorcycle rider improvement
program International Motorcycle Safety Conference Washington
DC USA Motorcycle Safety Foundation 1980231ndash53Aupetit 2008 published and unpublished data
Aupetit S Riff J Buttelli O Espie S Ergonomics research program
on motorcycle learning Contributions to the development of a
motorcycle simulator for training Transport Research Arena
Ljubljana 20089
Awane 1999 published data only
Awane T Integrating simulators in motorcycle safety education
Journal of International Association of Traffic and Safety Sciences 199923(1)26ndash35
Baldi 2005 published data only
Baldi S Baer JD Cook AL Identifying best practices states in
motorcycle rider education and licensing Journal of Safety Research200536(1)19ndash32
Batchler 1988 published data only
Batchler MW Evaluation of the pre-license motorcycle training course at the New South Wales traffic education centre Federal Office of
Road Safety (FORS) 19881ndash122 [ Report No CR 77 ]
Blanchard 2006 published data only
Blanchard HT Tabloski PA Motorcycle safety educating riders at
the teachable moment Mosby (CV) Company 200632330ndash2
Bonnett 2005 published data only
Bonnett J Mature age skills training for experienced motorcycle
riders Motorcycle ridersrsquo association of the Australian Capital
Territory 2005
Braver 2007 published data only
Braver ER et alPersuasion and licensure a randomized controlled
intervention trial to increase licensure rates among Maryland
motorcycle owners Traffic Injury Prevention 20078(1)39ndash46Clayton 1980 published data only
Clayton AB Sudlow DE An evaluation of the effectiveness of a one day training programme for learner motorcyclists Berkshire UK
Transport and Road Research Laboratory 19801ndash23 [ ISSN
0266ndash7045]lowast Clayton AB Sudlow DE An evaluation of the effectiveness of a one-day training programme for learner motorcyclists Berkshire UK
Road Safety Division Berkshire 19801ndash23 [ISSN 0266ndash7045]
Council 1978 published data only
Council FM Desper LP Dutt AK Driver education for motorcycle operation University of North Carolina Highway Safety Research
Center 19781ndash40
Crowther 2005 published data onlylowast Crowther G Engaging with motorcyclists UK police and theBikeSafe Road Safety Programme Published proceedings of the
Australasian Road Safety Research Policing Education Conference
Wellington New Zealand 2005 issue ISBN 0ndash473ndash10636ndash11ndash11
Daniello 2009 published data only
Daniello A Gabler HC Mehta YA Effectiveness of motorcycle
training and licensing Transportation Research Board 88th Annual
Meeting 200913
Dewanti 2007 published data only
Dewanti M Challenging efforts in promoting young motorcyclist
safety in Indonesia Road Safety on Four Continents 14th
International Conference Swedish National Road and Transport
Research Institute 2007
Du Plooy 1990 published data onlyDu Plooy DW Motorcycle training in South Africa The Human
Element International Motorcycle Safety Conference Orlando
Florida USA Motorcycle Safety Foundation 19904-1 to 4-9
Engel 1989 published data only
Engel U Krogsgaringrd Thomsen L [Faeligrdselsundervisning og
propaganda ndash en faeligrdselssikkerhedsfremmende foranstaltning]
Dansk Vejtidsskrift 198912295ndash7
Goldenbeld 2002 published data only
Goldenbeld C Houwing S de Craen S The development of
driving skills of young moped riders [De ontwikkeling van de
rijvaardigheid van jonge bromfietsers] SWOV Institute for Road
Safety Research (Stichting Wetenschappelijk Onderzoek
Verkeersveiligheid) Leidschendam the Netherlands 200263
25Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2850
Goldenbeld 2004 published data only
Goldenbeld C Twisk D de Craen S Short and long term effects of moped training a field experiment Transportation Research Part F 20047(1)1ndash16
Haworth 2000 published data only
Haworth N Smith R Kowadlo N Motorcycle training in Victoria
evaluation of rider training curriculum in Victoria Monash
University Accident Research Centre Reports 2000 Vol 165 issue
3131
Haworth 2005 published data only
Haworth N Mulvihill C Review of motorcycle licensing and
training Monash University Accident Research Centre 2005 [
Report no240]
Haworth 2006 published data only
Haworth N Mulvihill C Rowdon P Teaching old dogs new tricksTraining and older motorcyclists Journal of the Australasian College of Road Safety 200718(4)20ndash5lowast Haworth NL Mulvihill C Rowden PJ Teaching old dogs new
tricks Training and older motorcyclists Australasian Road Safety
Research Policing and Education Conference Gold Coast
Queensland 2006
Hill 1983 published data only
Hill D Motorcycle rider training testing and licensing Motor
Cycle Council of New South Wales Sydney 1983
Imatake 1980 published data only
Imatake Y On the activities in Japan of motorcycle safe riding
promotion committee (MSRPC) - especially the education of
unskilled riders of class 1 motorbikes International Motorcycle
Safety Conference Washington DC USA Motorcycle Safety
Foundation 1980255ndash68
Ingebritsen 1990 published and unpublished data
Ingebrigtsen S [Risikofaktorer ved ferdsel med moped og
motorsykkel] TOslashI-rapport 66 Oslo Transportoslashkonomisk
institutt 1990
Jain 2009 published and unpublished data
Jain A et alTwo wheeler accidents on Indian roads a study from
Mangalore India Elsevier 200916130ndash3
Jonah 1979 published data only
Jonah BA Dawson NE Validation of the motorcycle operator skill
test Accident Analysis amp Prevention 197911(3)163ndash71
Jonah 1981 published data onlyJonah BA Dawson NE Bragg WE Predicting accident
involvement with the Motorcycle Operator Skill Test Accident Analysis amp Prevention 198113(4)307ndash418
Kadar 1990 published data only
Kadar KJ Countermeasures motorcycle rider education delivery
system The Human Element International Motorcycle Safety
Conference Motorcycle Safety Foundation 19906-27 to 6-41
Kelsey 1986 published data only
Kelsey SL Liddicoat C Ratz M Licensing novice motorcyclists a
comparison of Californiarsquos standard test and the MOST II (Motorcycle
Operator Skill Test) administered at centralized testing offices Sacramento California USA California Department of Motor
Vehicles Research and Development Office 19861ndash74
Koch 1980 published data only
Koch H Advanced course for motorcycle riders InternationalMotorcycle Safety Conference Washington DC USA Motorcycle
Safety Foundation 1980269ndash81
Koch 1990 published data only
Koch H Brendicke R How to influence risk taking behaviour
within motorcycle rider training programs The Human Element
International Motorcycle Safety Conference Orlando Florida
USA Motorcycle Safety Foundation 1990Motorcycle Safety
Foundation 1990 p 9-1 to 9-25A
Labbett 2005 published data only
Labbett S Langham M Training can make the problem worse
Road Safety Congress 20051ndash12
Law 2005 published and unpublished data
Law TH Radin Umar RS Zulkaurnain S Kulanthayan S Impactof the effect of economic crisis and the targeted motorcycle safety
programme on motorcycle-related accidents injuries and fatalities
in Malaysia International Journal of Injury Control and Safety
Promotion 200512(1)9ndash21
Lowes 1990 published data only
Lowes BF In-traffic rider training The Human Element
International Motorcycle Safety Conference Orlando Florida
USA Motorcycle Safety Foundation 19907-46 to 7-57
Mahon 1976 published data only
Mahon J Teaching accident avoidance on the motorcycle Journal of Traffic Safety Education 1976231ndash29
Maxwell 1985 published data only
Maxwell GS Evaluation of a Program for Training MotorcycleRiders Evaluation and Program Planning 19858(4)339ndash47
Mayhew 1996 published data only
Mayhew DR Simpson SH Effectiveness and role of driver educationand training in a graduated licensing system Ottowa Canada The
Traffic Injury Research Foundation 19961ndash89 [ ISBN
0ndash920071ndash12ndash0]
McKnight 1987 published data only
McKnight AJ Milburn BG An evaluation of the modified
motorcycle operator skill test exercises NHTSA Report 1987 [
Report No DOT HS 807 212 ]
McPherson 1976 published data only
McPherson K McKnight AJ The development and evaluation of a
motorcycle skill test manual and knowledge test Washington DCUSA National Highway Safety Administration 1976
Mohan 2004 published data only
Mohan D Evidence-based interventions for road traffic injuries in
South Asia Journal of the College of Physicians and SurgeonsPakistan 200414(12)746ndash7
Morgan 1980 published data only
Morgan JK Advanced motorcycle training course content and
study evaluation International Motorcycle Safety Conference
Washington DC USA Motorcycle Safety Foundation 1980
283ndash312
Motoki 1990 published data only
Motoki M Yamazaki S A study on effective motorcycle rider
education The Human Element International Motorcycle Safety
26Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2950
Conference Orlando Florida USA Motorcycle Safety
Foundation 19909-26 to 9-57A
Nairn 1992 published data onlylowast Nairn RJ Motorcycle safety research literature review 1987 to
1991 Federal Office of Road Safety Australia 1993 Vol Report
CR 117
Newland 1980 published data only
Newland R Darby J Automotive practicespre motorcycle rider
program International Motorcycle Safety Conference Washington
DC USA Motorcycle Safety Foundation 1980127ndash9
Newland 1982 published data only
Newland R Motorcycle rider training and licensing requirements
Driver training steering a course for the future Melbourne
Victoria Transport Regulation Board 1982
Newland 1990 published data only
Newland R Motorcycle training and licensing in Victoria 1980-
1990 The Human Element International Motorcycle Safety
Conference Orlando Florida USA Motorcycle Safety
Foundation 1990
Ormston 2003 published data only
Ormston R Dudleston A Pearson S Stradling Evaluation of
Bikesafe Scotland Scottish Executive Social Research Edinburgh
2003
Osga 1979 published data only
Osga GS Ellingstad VS An evaluation of the effectiveness of the 1978
South Dakota motorcycle rider course program South Dakota USA
University of South Dakota 19791ndash89
Panou 2007 published data only
Panou M Bekiaris E A new concept on the integration of driving
simulators in driving training the train-all approach Road Safety
on Four Continents 14th International Conference Swedish
National Road and Transport Research Institute 2007
Peterson 1990 published data only
Peterson A Kadar K Status report 1980-1989 Minnesota rider
education statistics The Human Element International
Motorcycle Safety Conference Orlando Florida USA 199013-
113 to 13-129
Prem 1984 published data only
Prem H Good MC Motorcycle rider skills assessment ACT
Australia Office of Road Safety Department of Transport 1984
1ndash377 [ Report No CR 34 ]
Radin Umar 2006 published data only
Radin Umar RS Motorcycle safety programmes in Malaysia how
effective are they International Journal of Injury Control and Safety Promotion 200613(2)71ndash9
Rockwell 1990 published data only
Rockwell TH Kiger S Carnot M An evaluation of the Ohio
motorcyclist enrichment program Ohio Department of Highway
Safety 1990
Rothe 1987 published data only
Rothe JP A motorcycle riding course Motorcyclists Image and reality Canada Insurance Corporation of British Columbia 1987
59ndash83
Rowden 2007 published data only
Rowden P Watson B Haworth N What can riders tell us aboutmotorcycle training A view from the other side of the fence
Australasian Road Safety Research Policing and Education
Conference Melbourne Australia Melbourne 2007
RSC Victoria 1993 published data only
Road Safety Committee Inquiry into Motorcycle Safety in Victoria
Parliament of Victoria 1993
Russam 1979 published data only
Russam K Motorcycle training and licensing in Japan UK Transport
and Road Research Laboratory (TRRL) 1979 [ Report 916 ]
Saffron 1981 published data only
Saffron DG Driver and rider education training and licensing a
brief review Traffic Accident Research Unit Traffic Authority of
New South Wales NSW Australia Traffic Accident Research UnitTraffic Authority of New South Wales 1981
Schulz 1990 published data only
Schulz U Motorcycle accidents among learner riders Motorcycle
licensing Evaluation of the effectiveness of a Graduated Licensing Scheme Forshungshefte des Institut fur Zweiradsicherheit 1990
Sexton 2004 published data only (unpublished sought but not used)
Sexton B Baughan C Elliott M Maycock G The accident risk of
motorcyclists 2004TRL Report TRL 607(ISSN 09684107)58
Shephard 1986 published data only
Shepherd R Do Motorcycle Safety Foundation programs work
Driver 198619(9)16ndash8
Simpson 1990 published data onlySimpson HM Mayhew DR The promotion of motorcycle safety
training education and awareness Health Education Research
Theory and Practice 19905(2)257ndash64
Sokolov 1990 published data only
Sokolov O Safe motorcycle riding some aspects of motorcycle
riding in the USSR The Human Element International
Motorcycle Safety Conference Motorcycle Safety Foundation
19904ndash37 - 4-48
Steffens 1988 published data only
Steffens U Gawatz R Willmes G Effectiveness of mofa courses
[Wirksamkeit von Mofakursen] Unfall- und sicherheitsforschung
strasenverkehr 198867168
Sudlow 2003 published data only
Sudlow D Scoping study on motorcycle training Road Safety
Research Report London UK The British Institute of Traffic
Education Research Department for Transport 2003 Vol June
[ Report No 36]
Tasmania 1995 published data only
Mannering FL Grodsky LL Motorcycle rider training scheme
Traffic Education Research Centre Armidale Australia Driver
Training Australia Limited 1995 Vol April1ndash56
Tetard 1985 published data only
Tetard C Motorcycle ridersrsquo training to detect indices in an
intersection situation [Lrsquoentrainement a la prise drsquoinformation dans
lrsquoapprentissage de la conduite motocycliste] Le Travail Humain198548(2)147ndash59
27Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3050
Thackray 1980 published data only
Thackray RM Prescott JC Field test of the motorcycle ridercourse International Motorcycle Safety Conference Washington
DC USA Motorcycle Safety Foundation 1980195ndash7
Thompson 1994 published data only
Thompson M Evaluation of compulsory basic training for
motorcyclists Transport Research Laboratory Crowthorne
Berkshire UK 1994 [ Report 63]
Tobler 1990 published data only
Tobler U Individual driving instruction for motorcycle driving
learners the Swiss experiment of motorcycle driving school The
Human Element International Motorcycle Safety Conference
Orlando Florida USA Motorcycle Safety Foundation 1990
Transport 1989 published data only
Transport Communications Road Crash Statistics Australia Bureauof Transport and Communications Economics 198914ndash20
Tuan 2007 published data only
Tuan VA Shimizu T Towards development and evaluation of the
motorcycle drivers re-education program in Vietnam modeling of
motorcycle driverrsquos undesired behaviors 11th World Conference
on Transport Research 200725
Victor 1980 published data only
Victor DJ Traffic safety education for motorcyclists International
Motorcycle Safety Conference Washington DC USA Motorcycle
Safety Foundation 1980221ndash8
Waterer 1990 published data only
Waterer TH Compulsory training for motorcyclists from liberty
to compulsion The Human Element International Motorcycle
Safety Conference Orlando Florida USA Motorcycle Safety
Foundation 19904-49 to 4-57
Weller 1989 published data only
Weller RB Chandler EW Motorcycle safety and motorcycle
education past research and survey results Journal of Public Policy
amp Marketing 1989893ndash108
Wells 1981 published data only
Wells P Observations of some RACACU motorcycle training
schemes Transport and Road Research Laboratory Crowthorne
Berkshire UK 1981
Winn 1986 published data only
Winn G Two important trends in motorcycle safety regulations
rider education and conspicuity improvement American
Motorcyclist Association 1986147ndash52
Wisher 1988 published data only
Wisher G Reid J Motorcycle accidents in Norfolk a three-year study of motorcycle training and the Part 1 test Norfolk UK Road Safety
Division Highways Department Norfolk County Council 1988
Wood 1987 published data only
Wood HT Bowen R Evaluation of the revised learner permit scheme
July 1983 to December 1985 Victoria Australia Road Traffic
Authority 19871ndash27 [ ISBN 0 7306 0247 8]
Yamazaki 1990 published data only
Yamazaki S Nakanishi M Survey on effect of Honda rider
education The Human Element International Motorcycle Safety
Conference Orlando Florida USA 199012-167 to 12-187
References to ongoing studies
Burgess 2010 unpublished data only
Burgess C Broughton P Fylan F Stradling Interim evaluation of
the UKs National Ride program In L Dorn editor(s) Driver
behaviour and Training Vol IV Aldershot Ashgate Publishing
Limited 2010
Additional references
Anderson 1980
Anderson J Ford JL Peck RC Improved motorcyclist licensing and
testing project Volume I and II National Highway Traffic Safety
Administration US department of Transportation Washington
DC 1980
Bjornstig 1985
Bjornstig U Bylund P Lekander T Brorsson B Motorcycle fatalitiesin Sweden Acta Chirurgica Scandinavica 1985151(7)577ndash81
Clayton 1990
Clayton A Sudlow DE An evaluation of the effectiveness of a one-
day training program for learner motorcyclists The Human
Element International Motorcycle Safety Conference Orlando
Florida USA 199013-105 to 13-111
Cochrane Collaboration 2008
Higgins JP Green S Cochrane Handbook for Systematic Reviews of
Interventions 500 The Cochrane Collaboration 2008
Elliot 2003
Elliott MA Baughan CJT Broughton J Grayson G Knowles J et
alMotorcycle Safety a Scoping Study TRL [ Report No
0968ndash4107]Fleiss 2004
Fleiss JL Levin B Paik MC Statistical methods for rates and proportions 3rd Edition John Wiley amp Sons Inc 2004 [ ISBN
978ndash0ndash471ndash52629ndash2 ]
FORS 1999
FORS Federal Office of Road Safety Road risk for sober licensed
motorcyclists Vol Monograph 27 Federal Office of Road Safety
19991ndash2
Haworth 2005
Haworth N Mulvihill C Review of motorcycle licensing and training
Victoria Australia Monash University Accident Research Centre
20051ndash83 [ Report No 240]
Huwiler-Muumlntener 2002Huwiler-Muumlntener K Juumlni P Junker C Egger M Quality of
Reporting of Randomized Trials as a Measure of Methodologic
Quality JAMA 20022872801ndash4
Johnston 2008
Johnston P Brooks C Savage H Fatal and serious road crashes
involving motorcyclists Road Safety Research and Analysis Report
ATSB Monograph 20 Canberra Department of Infrastructure
Transport Regional Development and Local Government 0205
081ndash26
Juumlni 2001
Juumlni P Altman DG Egger M Systematic reviews in health care
Assessing the quality of controlled clinical trials BMJ 2001323
42ndash6
28Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3150
Kopits 2003
Kopits E Cropper M Traffic fatalities and economic growth TheWorld Bank Development Research Group Infrastructure and
Environment Policy research working paper 3035 2003
Leung 1983
Leung P Chau BTW Motorcycle accidents in Hong Kong pattern
of injury treatment causative factors and socio-economic
implications Injury 198315(1)23ndash9
Liu 2004
Liu B Ivers R Norton R Blows S Lo SK Helmets for preventing
injury in motorcycle riders Cochrane Database of Systematic Reviews 2004 Issue 2 [DOI 10100214651858CD004333pub3]
Mohan 2002
Mohan D Road safety in less-motorized environments future
concerns International Journal of Epidemiology 200231(3)527ndash32Mortimer 1982
Mortimer RG An evaluation of the motorcycle rider course
Proceedings of the 20th conference of the American Association for
Automotive Medicine Arlington Heights IL USA American
Association for Automotive Medicine 198277ndash96
Mullin 1997
Mullin BT Risk factors for motorcycle injury the role of age
gender experience training and alcohol PhD thesis Auckland
New Zealand University of Auckland 1997
Noordzij 2001
Noordzij PC Forke E Brendicke R Chinn BP Integration of needs
of moped and motorcycle riders into safety measures review and
statistical analysis in the framework of the European research
project PROMISING SWOV Institute for Road Safety Research
Leidschendam The Netherlands 2001 [ Report No Dndash2001ndash5]
Osga 1979
Osga G Ellingstad VS An evaluation of the effectiveness of the 1978 South Dakota Motorcycle Riders Course Program South Dakota
University of South Dakota Vermillion Human Factors Lab 1979
Oxford English Dictionary Online
httpoxforddictionariescom Oxford English Dictionary Online
Oxford University Press 2010
Peden 2004
Peden M Scurfield R Sleet D World Report on Road TrafficInjury Prevention Geneva World Health Organization Available
at httpwwwwhointviolence˙injury˙preventionpublications
road˙trafficworld˙reporten 2004Chapter 3
Radin Umar 1995
Radin Umar R Macakay G Hills B Preliminary analysis of
motorcycle accidents Short-term impacts of the running
headlights campaign and regulations in Malaysia J Traffic Med 199523(1)17ndash27
Tsai 1995
Tsai YJ Wang JD et alCase-control study of the effectiveness of
different types of helmets for the prevention of head injuries among
motorcycle riders in Taipei Taiwan American Journal of
Epidemiology 1995142(9)974ndash81
Vis 1995
Vis AA In-depth study of the hazards of motorcycling [De
onveiligheid van motorrijden nader bekeken] SWOV Institute for
Road Safety Research Leidschendam the Netherlands 1995 [
Report No Rndash95ndash69]
Weaver 1990
Weaver EA Motorcycle safety education in the United States
Motorcycle Safety Foundation Orlando Florida USA 1990
Winn 1987
Winn G Two important trends in motorcycle safety regulations
rider education and conspicuity improvement No more
information available Detroit MI US Society of Automotive
Engineers 1987147ndash52
Yuan 2000
Yuan W The effectiveness of the rsquoride-brightrsquo legislation for
motorcycles in Singapore Accident Analysis amp Prevention 200032
(4)559ndash63
Zambon 2006
Zambon F Hasselberg M Factors affecting the severity of injuries
among young motorcyclists a Swedish nationwide cohort study
Traffic Injury Prevention 20067(2)143ndash9lowast Indicates the major publication for the study
29Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3250
C H A R A C T E R I S T I C S O F S T U D I E S
Characteristics of included studies [ordered by study ID]
Adams 1985
Methods Non-randomised trial
Participants Initial sample size
Total of 162 Australia Post telegram delivery riders
Not clear how many in intervention and control groups
Analysis sample size
Intervention group 63Control group 35
Total 98 (61)
Interventions No formal training vs voluntary rider training course of 12 hours over 2 consecutive days (theory 4 hours practice
8 hours)
Intervention integrity
No information given
Outcomes Australia Post recorded crashes
Length of follow up 1 year
Notes Study conducted in NSW Australia
Anderson 1980
Methods Randomised trial
Participants Initial sample size
Total 41188 motorcycle licence applicants
Analysis sample size
Intervention group with skills training 13160
Intervention group without skills training 12634
Control group 15080
Total 40874 (99)
Interventions Standard licensing skill test vs the Motorcycle Operators Skill Test (MOST) with or without a training component
MOST is not to be confused with rider training because it is a licensing test The training component was for
participants who failed the skill test on the first attempt was of 3 hours duration and consisted of instructions about
the skills needed to pass the skill test The training consisted of 30 min theory and 2 frac12 hours practice on the
motorcycle
Intervention integrity
33 in MOST with skills training 31 in MOST without skills training and 43 in control group obtained their
motorcycle licence in the end
Outcomes Police recorded crashes and offences
Length of follow up 3 years
30Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3350
Anderson 1980 (Continued)
Notes Study conducted in California USA
Billheimer 1998
Methods Cohort study
Participants Initial sample size
Intervention group Unclear numbers of CMSP graduates and ERC graduates
Control group Over 16000 untrained motorcyclists
Analysis sample size
1139 matched CMSP-pairs and 1182 ERC-pairs
Interventions No formal training vs the mandatory (for those younger than 18 years) pre-licence 16-hours basic California
Motorcyclist Safety Program (CMSP) and the 8-hour Experienced Rider Course (ERC)
Intervention integrity
No information given
Outcomes Police recorded crashes and offences
Length of follow up 2 years
Notes Study conducted in California USA
Cooper 1988
Methods Cohort study
Participants Initial sample size
Intervention group 418 graduates and 57 who failed
Control group 402
Analysis sample size
Intervention group unclear
Control group unclear
Total 863 (984)
Interventions No formaltrainingvs non-mandatory pre-licence rider trainingThe content of thetrainingwas notfurther explainedIntervention integrity
Participants had all completed the rider training
Outcomes Police recorded crashes and offences
Length of follow up 7 years
Notes Study conducted in British Columbia Canada
31Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3450
Davis 1997
Methods Retrospective cohort study
Participants Initial sample size
Intervention group 9320 graduates
Control group unclear
Analysis sample size
Intervention group 9320 graduates
Control group unclear
Interventions No formal training vs the non-mandatory Connecticut Motorcycle Operator Training Program The content of
training was not further explained
Intervention integrity Participants had all completed the rider training
Outcomes Police recorded crashes severity of crashes
Length of follow up 1 year
Notes Study conducted in Connecticut USA
Hall 1988
Methods Randomised trial
Participants Initial sample sizeIntervention group A 6248
Intervention group B 6616
Intervention group C 6996
Control group 6604
Total 26464 motorcycle licence applicants
Analysis sample size
Intervention group A 6012 (96)
Intervention group B 6428 (97)
Intervention group C 6778 (97)
Control group 6502 (98)
Interventions Standard licensing skill test vs the Motorcycle Operator Skill Test (MOST II) without training (A) MOST II with
3 hours training (B) or MOST II with 20 hours training (C) The 3 hours training comprised instructions aboutthe skills needed to pass the skill test (30 min theory and 2 frac12 hours practice on the motorcycle) The 20 hours
training consisted of the Motorcycle Safety Foundationrsquos Motorcycle Rider Course (MRC) comprising of 8 hours
theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-road practice
Intervention integrity
25 and 26 obtained their motorcycle licence in the lsquoMOST 3-hour trainingrsquo and lsquoMOST 20-hour trainingrsquo
intervention arm respectively In the lsquoMOST testrsquo intervention arm 29 obtained their licence and in the control
group 35
Outcomes Police recorded crashes and offences
Length of follow up 3 years
32Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3550
Hall 1988 (Continued)
Notes Study conducted in New York USA
Haworth 1997
Methods Population based case-control study
Participants Cases Motorcycle riders or pillion riders involved in motorcycle crash (on road) Recruited from hospital and police
records Not clear whether all cases (especially fatalities) were included
Community controls Multiple matched sample of motorcycle rider or pillion who passed the crash site at the same
time of day and day of week that the crash occurred
Initial sample size
Cases 336
Controls 1195
Analysis sample size
Cases 234 (70)
Controls 561 (47)
Interventions Basic handling skills course beginner intermediate or advanced course measured through self-report not validated
Outcomes Injuries with an Injury Severity Score of 5 or greater or admission for at least four hours and fatalities from a
motorcycle crash (from hospital and police records)
Length of follow up two years or more prior to the interview
Notes Study conducted in Victoria Australia
Hurt 1981
Methods Population based case-control study
Participants Cases Motorcycle riders who had a motorcycle crash that was reported to police (traffic accident reports) Not all
cases included
Community controls Multiple matched sample of motorcycle riders who passed the crash site at the same time of
day and day of week as the crash occurred but 2 years later
Initial sample size
Cases 4500 notifications of crashes to police (only 50 received in time research team able to respond to 1126 for
unreported reasons data collection completed for 900)
Controls Number approached unreported 2310 consented and interviewed
Analysis sample size
Cases 808 (808900 data on rider training) - 18 of initial sample
Controls 670 (6702310 data on rider training) - lt 29 of initial sample
Interventions Training experience self taught friends-family school-club formal-professional others measured through self-
report validated
Outcomes Injuries and fatalities from a motorcycle crash (from hospital records and coronerrsquos office)
33Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3650
Hurt 1981 (Continued)
Notes Study conducted in Los Angeles USA
Jonah 1982
Methods Cohort study
Participants Initial sample size
Intervention group
2310 licensed MTP graduates who are active riders
Control group
2029 licensed motorcyclists who are active riders
Non-licensed and non-active riders excluded
Analysis sample size
Intervention group
811 interviews (35) 533 driver records (23)
Control group 1080 interviews (71) 905 driver records (45)
Interventions No training vs the non-mandatory Canada Safety Council Motorcycle Training Program (MTP) comprising of 4
hours theory and 16 hours practise (6 hrs on-road) over 2 weekends
Intervention integrity
Participants had all successfully completed the MTP
Outcomes Self-reported and police reported crashes and offencesLength of follow up 4 years
Notes Study conducted in Ontario Canada
Kloeden 1994
Methods Cohort study
Participants Initial sample size
Intervention group 2533 licence applicants
Control group Unclear
Sample of riders obtaining probationary licenceIntervention group 1625 (64)
Control group 5015 (unclear)
Non-active riders excluded from control group (~66)
Analysis sample size
Intervention group 631 (39)
Control group 318 (6)
Interventions No formal training vs the mandatory pre-licence Ridersafe course consisting of two levelsthe first is a training course
in the basics of motorcycle riding where a pass entitles the rider to obtain a learnerrsquos permit The second is a skill
test where a pass entitles the rider to obtain a probationary motorcycle licence The training course consists of two 4
hour training sessions on separate days involving instruction videos and actual riding on motorcycles The skill test
consists of a 4 hour preparation course of 74 minutes of lecturediscussion 38 minutes of video and 75 minutes of
34Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3750
Kloeden 1994 (Continued)
on-course riding which is the testing phase
Intervention integrity
Unclear what percentage in the Ridersafe group and the control group obtained their probationary licence halfway
and their full motorcycle licence in the end
Outcomes Police reported crashes
Length of follow up 1 year
Notes Study conducted in SA Australia
Kraus 1975
Methods Population based case-control study
Participants Cases All motorcycle riders with a confirmed medically treated motorcycle injury (including fatalities) from all
hospitals in study base and peripheral area hospitals and all deaths due to motorcycle crashes in the study base All
serious injuries (leading to hospitalisation) and fatalities likely to be included
Community controls A random sample of registered motorcycle owners in the same temporary and geographical
study base as the cases arose from
Initial sample size Cases 1273
Controls 738
Analysis sample size Cases 741 (58)
Controls 434 (59)
Interventions Motorcycle rider training (not specified) measured through self-report
Outcomes Injuries leading to hospitalisation fatalities
Length of follow up not specified
Notes Study conducted in Sacramento county USA
Leung 1987
Methods Non-randomised trial
Participants Initial sample sizeIntervention group unknown number of MRC graduates
Control group unknown number of licensed motorcyclists
Analysis sample size
Intervention group 2941
Control group 43094
Interventions No formal training vs non-mandatory Wisconsin motorcycle rider course a 22 hours off-street programme that
teaches the basic rules of safe motorcycling
Intervention integrity
50 in the MRC group and 100 in the control group obtained their motorcycle licence in the end
35Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3850
Leung 1987 (Continued)
Outcomes Police recorded crashes and offences
Length of follow up up to 3 years
Notes Study conducted in Wisconsin USA
MAIDS 2004
Methods Population based case-control study
Participants Cases Motorcycle riders who had a motorcycle crash Unclear whether all cases included
Community controls Motorcycle riders identified at petrol stations Unclear whether matched to cases
Initial sample size
Cases not reported
Controls not reported
Analysis sample size
Cases 921 (Italy 200 Netherlands 200 Spain 121 Germany 250 France 150)
Controls 923 (Italy 200 Netherlands 200 Spain123 Germany 250 France 150)
Interventions Pre-license rider training or additional training (non-compulsory)
Outcomes Crashes
Notes Study in five sampling areas located in five European countries (France Germany Netherlands Spain and Italy)
McDavid 1989
Methods Cohort study
Participants Sample size after first match
Intervention group 346 male training graduates
Control group 346 male motorcycle riders licensed in same year
Sample size after second match
Intervention group 139 (40)
Control group 139 (40)
Analysis sample size
Intervention group 139 (40)
Control group 139 (40)
Interventions No formal training vs the non-mandatory pre-licence British Columbia Safety Councilrsquos 37-hour motorcycle safety
programme
Intervention integrity
Participants had all completed and passed the training
Outcomes Police reported crashes injurious crashes
Length of follow up 1-6 years
36Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3950
McDavid 1989 (Continued)
Notes Study conducted in British Columbia Canada
Mortimer 1984
Methods Cohort study
Participants Initial sample size
Intervention group 608 MRC graduates
Control group 360 motorcycle riders
Response rates
Intervention group 360 (59)
Control group 324 (90)
Non-active riders excluded
Analysis sample size
Intervention group 213 (35)
Control group 303 (84)
Interventions No formal training vs the non-mandatory Motorcycle Safety Foundationrsquos Motorcycle Rider Course comprising of
8 hours theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-road practice
Intervention integrity
Participants had all completed and passed the MRC
Outcomes Self-reported crashes and offencesLength of follow up 1 year
Notes Study conducted in Illinois USA
Mortimer 1988
Methods Cohort study
Participants Initial sample size
Intervention group 2307 MRC graduates
Control group 556 active licensed motorcycle riders
Response ratesIntervention group 1298 (56)
Control group 500 (90)
Non-active riders excluded
Analysis sample size
Intervention group 913 (40)
Control group 500 (90)
Interventions No formal training vs the non-mandatory Motorcycle Safety Foundationrsquos Motorcycle Rider Course comprising of
8 hours theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-road practice
Intervention integrity
Participants had all completed and passed the MRC
37Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4050
Mortimer 1988 (Continued)
Outcomes Self reported crashes and offences
Length of follow up 1 year
Notes Study conducted in Illinois USA
Osga 1980
Methods Cohort study
Participants Initial sample sizeIntervention group 1163 MSF MRC participants invited 818 (74) responded
Control group 329 motorcycle riders responded to recruitment strategies
Non-active riders excluded from analysis
Analysis sample size
Intervention group 595 (73)
Control group 329 (100)
Interventions No formal training vs the non-mandatory Motorcycle Safety Foundationrsquos Motorcycle Rider Course (MRC) com-
prising of 8 hours theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-road
practice
Intervention integrity
Participants had all successfully completed the MRC
Outcomes Self-reported crashes police recorded crash reports
Length of follow up 6 months - 3 12 years
Notes Study conducted in South Dakota USA
Perrino 2002
Methods Cohort study
Participants Initial sample size
Intervention group 2144 (unclear) MRC graduates (motorcycle licence applicants)
Control group 2601 (unclear) regular licensing test graduates (motorcycle licence applicants)Analysis sample size
Intervention group 1070 (unclear)
Control group 867 (unclear)
Interventions No formal training vs the mandatory (for those younger than 18 years) pre-licence Maryland Motorcycle Rider
Course (MRC) consisting of 8 hours theory and 12 hours practice
Intervention integrity
No information given
Outcomes Self-reported and police recorded crashes and offences
Length of follow up 1 year
38Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4150
Perrino 2002 (Continued)
Notes Study conducted in Maryland USA
Raymond 1979
Methods Cohort study
Participants Initial sample size
Intervention group
258 volunteers mainly (but not all) novice riders
Control group
1104 provisional licence applicants
Analysis sample size
Intervention group 190 (74)
Control group 627 (57)
Interventions No formal training vs the non-mandatory RACACU (Royal Automobile ClubAuto-Cycle Union) training scheme
The training program consisted of either 24 one-hour sessions or 13 two-hour sessions Twelve hours of the training
focused on theory about basic operational procedures traffic laws and highway code The practical session focused
on basic operational skills riding on a training area and riding in traffic with an instructor A proficiency test was
administered at the end of the course
Intervention integrity
No information given
Outcomes Self-reported crashes (including minor non-injury non-damage and off-road crashes)
Length of follow up 1 year
Notes Study conducted in the UK
Satten 1980
Methods Cohort study
Participants Initial sample size
Intervention group interviewers kept interviewing until they had 100 riders selected from a population of 891
enrollees in the MSF MRC course (unknown how many were asked and how many refused to participate)Control group unknown number of registered motorcycle riders
Non-active riders excluded from analysis
Analysis sample size
Intervention group 71
Control group 100
Interventions No formal training vs the non-mandatory Motorcycle Safety Foundationrsquos (MSF) Motorcycle Rider Course (MRC)
comprising of 8 hours theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-
road practice
Intervention integrity
Participants had all successfully completed the MRC
39Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4250
Satten 1980 (Continued)
Outcomes Self-reported crashes injuries and offences
Length of follow up 3 years
Notes Study conducted in Illinois USA
Savolainen 2007
Methods Cohort study
Participants Initial sample size
Intervention group 4000 riders that had conducted the MSF MRC course (passed or failed)
Control group 4000 motorcycle owners or permit or endorsement holders from the Indiana Bureau of Motor
Vehicles database
Analysis sample size
Intervention group 739 (18)
Control group 588 (15)
Interventions Intervention integrity
No information given
Outcomes Probability of crash involvement
Length of follow up up to 5 years
Notes Study conducted in Indiana USA
Swaddiwudhipong 1998
Methods Randomised trial
Participants Initial sample size
Intervention group
all 16 villages of 3 sub-districts
(participants were practising motorcyclists licensed and unlicensed)
Control group
18 villages of 3 other sub-districtsAnalysis sample size
Intervention group 3 districts (100)
Control group 3 districts (100)
Interventions No rider education vs non-mandatory community based rider education consisting of theory on motorcycle injury
risk helmet use traffic laws vehicle regulations and traffic signs and a written and skill test leading to a licence
Intervention integrity
No information given
Outcomes Injuries fatalities
Length of follow up 2 years
40Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4350
Swaddiwudhipong 1998 (Continued)
Notes Study conducted in Tak Province rural Thailand
Waller 1992
Methods Cohort study
Participants Initial sample size
Intervention group 349 riders who succesfully completed the Department of Public Safety motorcycle operator
training in 1988
Control group 349 motorcyclists who never completed the DPS motorcycle operator training
Analysis sample size
Intervention group 349
Control group 349
Interventions No DPS motorcycle operator training vs non-mandatory DPS motorcycle operator training The DPS training is
the Motorcycle Safety Foundationrsquos Motorcycle Rider Course comprising eight hours of theory (lectures discussions
films and multiple choice tests) a textbook and 12 hours off-road practice
Intervention integrity
All 349 riders succesfully completed the rider training
Outcomes Police recorded crashes
Length of follow up 3 years
Notes Study conducted in
Characteristics of excluded studies [ordered by study ID]
Study Reason for exclusion
Anderson 1980a Study type review and process evaluation of a rider course
Aupetit 2008 Outcome proposal of a rider training curriculum that combines traditional training and a motorcycle riding
simulator
Awane 1999 Study type review on use of simulators in rider and driver training
Baldi 2005 Intervention not rider training but intervention integrity of rider training availabilty of rider education
implementation of graduated licensing system etc on a state-wide level
Batchler 1988 Study type description of a training course
Blanchard 2006 Study type not a study assessing the association between rider training and outcomes
Bonnett 2005 Outcome qualitative data regarding satisfaction with the training programme
41Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4450
(Continued)
Braver 2007 Outcome licensure rates
Clayton 1980 Outcome knowledge skills attitude
Council 1978 Outcome crash reduction in ANY vehicle not motorcycles exclusively
Crowther 2005 Study type qualitative study
Daniello 2009 Study type review of the effectiveness of motorcycle training and licensing
Dewanti 2007 Intervention emotional characteristics behaviour (ie not motocycle rider training)
Du Plooy 1990 Study type review of types of rider training courses in South Africa
Engel 1989 Study type uncontrolled before-after study
Goldenbeld 2002 Outcome skills
Goldenbeld 2004 Outcome knowledge and skills
Haworth 2000 Intervention balance between attitudinal andvehicle control skills acrosscurriculla of motorcycle rider training
courses
Haworth 2005 Study type review
Haworth 2006 Study type cross-sectional survey
Hill 1983 Study type review and description of training system
Imatake 1980 Study type course descriptionreview
Ingebritsen 1990 Intervention risk factors for motorcycle crashes
Jain 2009 Study type cross-sectional descriptive study of motorcycle crashes over a 5 year period
Jonah 1979 Study type cross-sectional survey Outcome scores on motorcycle operator skill test
Jonah 1981 Intervention evaluation of a motorcycle rider skill test not motorcycle rider training
Kadar 1990 Study type review of components of rider training courses
Kelsey 1986 Intervention evaluation of a motorcycle rider skill test not motorcycle rider training
Koch 1980 Study type reviewcourse description
Koch 1990 Study type description of course content
42Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4550
(Continued)
Labbett 2005 Study type essay about possible explanations for motorcycle deaths
Law 2005 Intervention targeted motorcycle safety programme not specifically rider training
Lowes 1990 Study type review and descriptions of rider training
Mahon 1976 Study type review of motorcycle rider training methods
Maxwell 1985 Outcomesrecommendations forprogramme operatorsand otherson howthe motorcycle trainingprogramme
works and how it can be improved
Mayhew 1996 Study type review
McKnight 1987 Study type evaluation of methodological and administrative aspects of a motorcycle operator skill test
McPherson 1976 Outcome knowledge and skills
Mohan 2004 Study type review of evidence of interventions to reduce road traffic injuries
Morgan 1980 Study type reviewcourse description
Motoki 1990 Study type proposal and development of rider training curriculum
Nairn 1992 Study type literature review
Newland 1980 Study type description of development of motorcycle safety course
Newland 1982 Study type descriptionreview of a rider training course
Newland 1990 Study type uncontrolled BeforeAfter study
Ormston 2003 Outcome changes in attitude and self reported behaviour
Osga 1979 Outcome knowledge attitudes and skills
Panou 2007 Intervention new simulation tool to teach driverrider
Peterson 1990 Study factor demographic details of course participants completing the course vs those not completing the
course
Prem 1984 Study type experimental
Study factor examines association between skills and performance in tests
Radin Umar 2006 Intervention motorcycle safety programme not specifically rider training
Rockwell 1990 Study type methodologically flawed study type
43Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4650
(Continued)
Rothe 1987 Study type qualitative study of rider training in general
Rowden 2007 Study type qualitative study
RSC Victoria 1993 Study type review of motorcycle rider training and education in Victoria
Russam 1979 Study type review of training programmes in Japan
Saffron 1981 Study type literature review
Schulz 1990 Study factor predictors of motorcycle crashes
Sexton 2004 Study type cross-sectional analytic analysis
Shephard 1986 Study factor reviewdescription of programme
Simpson 1990 Study type review of motorcycle rider trainingeducation
Sokolov 1990 Study factor review of motorcycle riding courses in USSR
Steffens 1988 Outcome self-reported behaviour experiences and researcher observed riding behaviour
Sudlow 2003 Study type review of motorcycle training in the UK
Tasmania 1995 Study type cross-sectional descriptiveprocess evaluation
Tetard 1985 Study type experimental
Outcome detection of indices
Thackray 1980 Outcome instructional effectiveness user acceptance administrative feasibility
Thompson 1994 Study type cross-sectional survey
Outcome attitudes and knowledge
Tobler 1990 Study type review of course curricula
Transport 1989 Intervention syllabus for rider training
Tuan 2007 Study type cross-sectional study (survey) which does not assess the effects of an intervention but examines
motocyclistsrsquo behaviour
Victor 1980 Study type review
Waterer 1990 Study type review of UK rider training issues
Weller 1989 Study type review of studies evaluating motorcycle safety and education measures
44Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4750
(Continued)
Wells 1981 Study type description of different methods of training
Winn 1986 Study type review of rsquohistoryrsquo of rider training
Wisher 1988 Study type cross-sectional survey
Wood 1987 Study type Interrupted Time Series but did not meet the criteria for an ITS as published by the Cochrane
Effective Practice and Organisation of Care Group
Yamazaki 1990 Study type cross-sectional survey
Outcome riding skills skills retention and attitude
Characteristics of ongoing studies [ordered by study ID]
Burgess 2010
Trial name or title UKs National Ride program
Methods Cohort study
Participants Motorcycle offenders
Interventions Police led diversion program
Outcomes Self reported attitudinal and behavioural measures
Starting date
Contact information
Notes Expected to be completed in early 2010
45Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4850
D A T A A N D A N A L Y S E S
This review has no analyses
A P P E N D I C E S
Appendix 1 Search strategy
We searched the following electronic databases
bull Cochrane Injuries Group Specialised Register (searched 22 Sept 2008)
bull CENTRAL (The Cochrane Library 2008 Issue 3)
bull TRANSPORT 1988 to 200706 (latest date available)
bull ATRI Australian Transport Index (to Sept 2008)
bull MEDLINE (Ovid SP) 1950 to Sept 2008
bull EMBASE (Ovid SP) 1980 to Sept 2008
bull CINAHL (EBSCO) 1982 to Sept 2008
bull WHOLIS (World Health Organization Library Information System) (to Sept 2008)
bull PsycINFO (Ovid SP) 1806 to Sept 2008
bull LILACS (BIREME) (Latin American and Caribbean Health Sciences) (1982 to Sept 2008)
bull ISI Web of Science Social Sciences Citation Index (SSCI) 1970 to Sept 2008
bull ERIC (searched 25 Sept 2008)
bull ZETOC (searched 25 Sept 2008)
bull SIGLE (searched 25 Sept 2008)
All of the databases were searched using the following text-word search strategy The strategy was adapted using either rsquorsquo or rsquo$rsquo as atruncation symbol
1 motor-cycl or motorcycl or motor-bik or motorbik or motor-scooter or motorscooter or scooter or moped or motocycl
or motocicl or moto
2 train or educat or instruct or program or course or qualif or learn or improv
3 1 and 2
H I S T O R Y
Protocol first published Issue 2 2005
Review first published Issue 10 2010
46Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4950
C O N T R I B U T I O N S O F A U T H O R S
Katina Kardamanidis was the first author of the protocol conducted the literature search extracted full studies assessed the studies
according to inclusion criteria assessed quality of the studies compared findings with the second author discussed disagreements with
the third author wrote the drafts and the final review
Rebecca Ivers was the second author of the protocol assessed the studies according to inclusion criteria assessed quality of the studies
compared findings with the first author edited the final draft and provided overall guidance for the process
Alexandra Martiniuk assessed quality of the studies compared findings with the first author discussed disagreements with the third
author and helped write the first draft and the final review
Mark Stevenson contributed to the protocol reviewed methods of studies and contributed to the final draft
Katrina Thistlethwaite conducted a second literature search extracted full studies and contributed to the final draft
D E C L A R A T I O N S O F I N T E R E S T
None known
S O U R C E S O F S U P P O R T
Internal sources
bull No sources of support supplied
External sources
bull National Health and Medical Research Council of Australia Australia
Salary support for Martiniuk Ivers and Stevenson
bull University of Sydney Sesqui Research Grant Australia
Direct research costs and salary support for Kardamanidis
D I F F E R E N C E S B E T W E E N P R O T O C O L A N D R E V I E W
There are no differences between the protocol and the review
I N D E X T E R M S
Medical Subject Headings (MeSH)
lowastMotorcycles Accident Prevention [lowastmethods] Accidents Traffic [lowastprevention amp control] Licensure Program Evaluation Wounds
and Injuries [mortality lowastprevention amp control]
47Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 5050
MeSH check words
Humans
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2850
Goldenbeld 2004 published data only
Goldenbeld C Twisk D de Craen S Short and long term effects of moped training a field experiment Transportation Research Part F 20047(1)1ndash16
Haworth 2000 published data only
Haworth N Smith R Kowadlo N Motorcycle training in Victoria
evaluation of rider training curriculum in Victoria Monash
University Accident Research Centre Reports 2000 Vol 165 issue
3131
Haworth 2005 published data only
Haworth N Mulvihill C Review of motorcycle licensing and
training Monash University Accident Research Centre 2005 [
Report no240]
Haworth 2006 published data only
Haworth N Mulvihill C Rowdon P Teaching old dogs new tricksTraining and older motorcyclists Journal of the Australasian College of Road Safety 200718(4)20ndash5lowast Haworth NL Mulvihill C Rowden PJ Teaching old dogs new
tricks Training and older motorcyclists Australasian Road Safety
Research Policing and Education Conference Gold Coast
Queensland 2006
Hill 1983 published data only
Hill D Motorcycle rider training testing and licensing Motor
Cycle Council of New South Wales Sydney 1983
Imatake 1980 published data only
Imatake Y On the activities in Japan of motorcycle safe riding
promotion committee (MSRPC) - especially the education of
unskilled riders of class 1 motorbikes International Motorcycle
Safety Conference Washington DC USA Motorcycle Safety
Foundation 1980255ndash68
Ingebritsen 1990 published and unpublished data
Ingebrigtsen S [Risikofaktorer ved ferdsel med moped og
motorsykkel] TOslashI-rapport 66 Oslo Transportoslashkonomisk
institutt 1990
Jain 2009 published and unpublished data
Jain A et alTwo wheeler accidents on Indian roads a study from
Mangalore India Elsevier 200916130ndash3
Jonah 1979 published data only
Jonah BA Dawson NE Validation of the motorcycle operator skill
test Accident Analysis amp Prevention 197911(3)163ndash71
Jonah 1981 published data onlyJonah BA Dawson NE Bragg WE Predicting accident
involvement with the Motorcycle Operator Skill Test Accident Analysis amp Prevention 198113(4)307ndash418
Kadar 1990 published data only
Kadar KJ Countermeasures motorcycle rider education delivery
system The Human Element International Motorcycle Safety
Conference Motorcycle Safety Foundation 19906-27 to 6-41
Kelsey 1986 published data only
Kelsey SL Liddicoat C Ratz M Licensing novice motorcyclists a
comparison of Californiarsquos standard test and the MOST II (Motorcycle
Operator Skill Test) administered at centralized testing offices Sacramento California USA California Department of Motor
Vehicles Research and Development Office 19861ndash74
Koch 1980 published data only
Koch H Advanced course for motorcycle riders InternationalMotorcycle Safety Conference Washington DC USA Motorcycle
Safety Foundation 1980269ndash81
Koch 1990 published data only
Koch H Brendicke R How to influence risk taking behaviour
within motorcycle rider training programs The Human Element
International Motorcycle Safety Conference Orlando Florida
USA Motorcycle Safety Foundation 1990Motorcycle Safety
Foundation 1990 p 9-1 to 9-25A
Labbett 2005 published data only
Labbett S Langham M Training can make the problem worse
Road Safety Congress 20051ndash12
Law 2005 published and unpublished data
Law TH Radin Umar RS Zulkaurnain S Kulanthayan S Impactof the effect of economic crisis and the targeted motorcycle safety
programme on motorcycle-related accidents injuries and fatalities
in Malaysia International Journal of Injury Control and Safety
Promotion 200512(1)9ndash21
Lowes 1990 published data only
Lowes BF In-traffic rider training The Human Element
International Motorcycle Safety Conference Orlando Florida
USA Motorcycle Safety Foundation 19907-46 to 7-57
Mahon 1976 published data only
Mahon J Teaching accident avoidance on the motorcycle Journal of Traffic Safety Education 1976231ndash29
Maxwell 1985 published data only
Maxwell GS Evaluation of a Program for Training MotorcycleRiders Evaluation and Program Planning 19858(4)339ndash47
Mayhew 1996 published data only
Mayhew DR Simpson SH Effectiveness and role of driver educationand training in a graduated licensing system Ottowa Canada The
Traffic Injury Research Foundation 19961ndash89 [ ISBN
0ndash920071ndash12ndash0]
McKnight 1987 published data only
McKnight AJ Milburn BG An evaluation of the modified
motorcycle operator skill test exercises NHTSA Report 1987 [
Report No DOT HS 807 212 ]
McPherson 1976 published data only
McPherson K McKnight AJ The development and evaluation of a
motorcycle skill test manual and knowledge test Washington DCUSA National Highway Safety Administration 1976
Mohan 2004 published data only
Mohan D Evidence-based interventions for road traffic injuries in
South Asia Journal of the College of Physicians and SurgeonsPakistan 200414(12)746ndash7
Morgan 1980 published data only
Morgan JK Advanced motorcycle training course content and
study evaluation International Motorcycle Safety Conference
Washington DC USA Motorcycle Safety Foundation 1980
283ndash312
Motoki 1990 published data only
Motoki M Yamazaki S A study on effective motorcycle rider
education The Human Element International Motorcycle Safety
26Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2950
Conference Orlando Florida USA Motorcycle Safety
Foundation 19909-26 to 9-57A
Nairn 1992 published data onlylowast Nairn RJ Motorcycle safety research literature review 1987 to
1991 Federal Office of Road Safety Australia 1993 Vol Report
CR 117
Newland 1980 published data only
Newland R Darby J Automotive practicespre motorcycle rider
program International Motorcycle Safety Conference Washington
DC USA Motorcycle Safety Foundation 1980127ndash9
Newland 1982 published data only
Newland R Motorcycle rider training and licensing requirements
Driver training steering a course for the future Melbourne
Victoria Transport Regulation Board 1982
Newland 1990 published data only
Newland R Motorcycle training and licensing in Victoria 1980-
1990 The Human Element International Motorcycle Safety
Conference Orlando Florida USA Motorcycle Safety
Foundation 1990
Ormston 2003 published data only
Ormston R Dudleston A Pearson S Stradling Evaluation of
Bikesafe Scotland Scottish Executive Social Research Edinburgh
2003
Osga 1979 published data only
Osga GS Ellingstad VS An evaluation of the effectiveness of the 1978
South Dakota motorcycle rider course program South Dakota USA
University of South Dakota 19791ndash89
Panou 2007 published data only
Panou M Bekiaris E A new concept on the integration of driving
simulators in driving training the train-all approach Road Safety
on Four Continents 14th International Conference Swedish
National Road and Transport Research Institute 2007
Peterson 1990 published data only
Peterson A Kadar K Status report 1980-1989 Minnesota rider
education statistics The Human Element International
Motorcycle Safety Conference Orlando Florida USA 199013-
113 to 13-129
Prem 1984 published data only
Prem H Good MC Motorcycle rider skills assessment ACT
Australia Office of Road Safety Department of Transport 1984
1ndash377 [ Report No CR 34 ]
Radin Umar 2006 published data only
Radin Umar RS Motorcycle safety programmes in Malaysia how
effective are they International Journal of Injury Control and Safety Promotion 200613(2)71ndash9
Rockwell 1990 published data only
Rockwell TH Kiger S Carnot M An evaluation of the Ohio
motorcyclist enrichment program Ohio Department of Highway
Safety 1990
Rothe 1987 published data only
Rothe JP A motorcycle riding course Motorcyclists Image and reality Canada Insurance Corporation of British Columbia 1987
59ndash83
Rowden 2007 published data only
Rowden P Watson B Haworth N What can riders tell us aboutmotorcycle training A view from the other side of the fence
Australasian Road Safety Research Policing and Education
Conference Melbourne Australia Melbourne 2007
RSC Victoria 1993 published data only
Road Safety Committee Inquiry into Motorcycle Safety in Victoria
Parliament of Victoria 1993
Russam 1979 published data only
Russam K Motorcycle training and licensing in Japan UK Transport
and Road Research Laboratory (TRRL) 1979 [ Report 916 ]
Saffron 1981 published data only
Saffron DG Driver and rider education training and licensing a
brief review Traffic Accident Research Unit Traffic Authority of
New South Wales NSW Australia Traffic Accident Research UnitTraffic Authority of New South Wales 1981
Schulz 1990 published data only
Schulz U Motorcycle accidents among learner riders Motorcycle
licensing Evaluation of the effectiveness of a Graduated Licensing Scheme Forshungshefte des Institut fur Zweiradsicherheit 1990
Sexton 2004 published data only (unpublished sought but not used)
Sexton B Baughan C Elliott M Maycock G The accident risk of
motorcyclists 2004TRL Report TRL 607(ISSN 09684107)58
Shephard 1986 published data only
Shepherd R Do Motorcycle Safety Foundation programs work
Driver 198619(9)16ndash8
Simpson 1990 published data onlySimpson HM Mayhew DR The promotion of motorcycle safety
training education and awareness Health Education Research
Theory and Practice 19905(2)257ndash64
Sokolov 1990 published data only
Sokolov O Safe motorcycle riding some aspects of motorcycle
riding in the USSR The Human Element International
Motorcycle Safety Conference Motorcycle Safety Foundation
19904ndash37 - 4-48
Steffens 1988 published data only
Steffens U Gawatz R Willmes G Effectiveness of mofa courses
[Wirksamkeit von Mofakursen] Unfall- und sicherheitsforschung
strasenverkehr 198867168
Sudlow 2003 published data only
Sudlow D Scoping study on motorcycle training Road Safety
Research Report London UK The British Institute of Traffic
Education Research Department for Transport 2003 Vol June
[ Report No 36]
Tasmania 1995 published data only
Mannering FL Grodsky LL Motorcycle rider training scheme
Traffic Education Research Centre Armidale Australia Driver
Training Australia Limited 1995 Vol April1ndash56
Tetard 1985 published data only
Tetard C Motorcycle ridersrsquo training to detect indices in an
intersection situation [Lrsquoentrainement a la prise drsquoinformation dans
lrsquoapprentissage de la conduite motocycliste] Le Travail Humain198548(2)147ndash59
27Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3050
Thackray 1980 published data only
Thackray RM Prescott JC Field test of the motorcycle ridercourse International Motorcycle Safety Conference Washington
DC USA Motorcycle Safety Foundation 1980195ndash7
Thompson 1994 published data only
Thompson M Evaluation of compulsory basic training for
motorcyclists Transport Research Laboratory Crowthorne
Berkshire UK 1994 [ Report 63]
Tobler 1990 published data only
Tobler U Individual driving instruction for motorcycle driving
learners the Swiss experiment of motorcycle driving school The
Human Element International Motorcycle Safety Conference
Orlando Florida USA Motorcycle Safety Foundation 1990
Transport 1989 published data only
Transport Communications Road Crash Statistics Australia Bureauof Transport and Communications Economics 198914ndash20
Tuan 2007 published data only
Tuan VA Shimizu T Towards development and evaluation of the
motorcycle drivers re-education program in Vietnam modeling of
motorcycle driverrsquos undesired behaviors 11th World Conference
on Transport Research 200725
Victor 1980 published data only
Victor DJ Traffic safety education for motorcyclists International
Motorcycle Safety Conference Washington DC USA Motorcycle
Safety Foundation 1980221ndash8
Waterer 1990 published data only
Waterer TH Compulsory training for motorcyclists from liberty
to compulsion The Human Element International Motorcycle
Safety Conference Orlando Florida USA Motorcycle Safety
Foundation 19904-49 to 4-57
Weller 1989 published data only
Weller RB Chandler EW Motorcycle safety and motorcycle
education past research and survey results Journal of Public Policy
amp Marketing 1989893ndash108
Wells 1981 published data only
Wells P Observations of some RACACU motorcycle training
schemes Transport and Road Research Laboratory Crowthorne
Berkshire UK 1981
Winn 1986 published data only
Winn G Two important trends in motorcycle safety regulations
rider education and conspicuity improvement American
Motorcyclist Association 1986147ndash52
Wisher 1988 published data only
Wisher G Reid J Motorcycle accidents in Norfolk a three-year study of motorcycle training and the Part 1 test Norfolk UK Road Safety
Division Highways Department Norfolk County Council 1988
Wood 1987 published data only
Wood HT Bowen R Evaluation of the revised learner permit scheme
July 1983 to December 1985 Victoria Australia Road Traffic
Authority 19871ndash27 [ ISBN 0 7306 0247 8]
Yamazaki 1990 published data only
Yamazaki S Nakanishi M Survey on effect of Honda rider
education The Human Element International Motorcycle Safety
Conference Orlando Florida USA 199012-167 to 12-187
References to ongoing studies
Burgess 2010 unpublished data only
Burgess C Broughton P Fylan F Stradling Interim evaluation of
the UKs National Ride program In L Dorn editor(s) Driver
behaviour and Training Vol IV Aldershot Ashgate Publishing
Limited 2010
Additional references
Anderson 1980
Anderson J Ford JL Peck RC Improved motorcyclist licensing and
testing project Volume I and II National Highway Traffic Safety
Administration US department of Transportation Washington
DC 1980
Bjornstig 1985
Bjornstig U Bylund P Lekander T Brorsson B Motorcycle fatalitiesin Sweden Acta Chirurgica Scandinavica 1985151(7)577ndash81
Clayton 1990
Clayton A Sudlow DE An evaluation of the effectiveness of a one-
day training program for learner motorcyclists The Human
Element International Motorcycle Safety Conference Orlando
Florida USA 199013-105 to 13-111
Cochrane Collaboration 2008
Higgins JP Green S Cochrane Handbook for Systematic Reviews of
Interventions 500 The Cochrane Collaboration 2008
Elliot 2003
Elliott MA Baughan CJT Broughton J Grayson G Knowles J et
alMotorcycle Safety a Scoping Study TRL [ Report No
0968ndash4107]Fleiss 2004
Fleiss JL Levin B Paik MC Statistical methods for rates and proportions 3rd Edition John Wiley amp Sons Inc 2004 [ ISBN
978ndash0ndash471ndash52629ndash2 ]
FORS 1999
FORS Federal Office of Road Safety Road risk for sober licensed
motorcyclists Vol Monograph 27 Federal Office of Road Safety
19991ndash2
Haworth 2005
Haworth N Mulvihill C Review of motorcycle licensing and training
Victoria Australia Monash University Accident Research Centre
20051ndash83 [ Report No 240]
Huwiler-Muumlntener 2002Huwiler-Muumlntener K Juumlni P Junker C Egger M Quality of
Reporting of Randomized Trials as a Measure of Methodologic
Quality JAMA 20022872801ndash4
Johnston 2008
Johnston P Brooks C Savage H Fatal and serious road crashes
involving motorcyclists Road Safety Research and Analysis Report
ATSB Monograph 20 Canberra Department of Infrastructure
Transport Regional Development and Local Government 0205
081ndash26
Juumlni 2001
Juumlni P Altman DG Egger M Systematic reviews in health care
Assessing the quality of controlled clinical trials BMJ 2001323
42ndash6
28Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3150
Kopits 2003
Kopits E Cropper M Traffic fatalities and economic growth TheWorld Bank Development Research Group Infrastructure and
Environment Policy research working paper 3035 2003
Leung 1983
Leung P Chau BTW Motorcycle accidents in Hong Kong pattern
of injury treatment causative factors and socio-economic
implications Injury 198315(1)23ndash9
Liu 2004
Liu B Ivers R Norton R Blows S Lo SK Helmets for preventing
injury in motorcycle riders Cochrane Database of Systematic Reviews 2004 Issue 2 [DOI 10100214651858CD004333pub3]
Mohan 2002
Mohan D Road safety in less-motorized environments future
concerns International Journal of Epidemiology 200231(3)527ndash32Mortimer 1982
Mortimer RG An evaluation of the motorcycle rider course
Proceedings of the 20th conference of the American Association for
Automotive Medicine Arlington Heights IL USA American
Association for Automotive Medicine 198277ndash96
Mullin 1997
Mullin BT Risk factors for motorcycle injury the role of age
gender experience training and alcohol PhD thesis Auckland
New Zealand University of Auckland 1997
Noordzij 2001
Noordzij PC Forke E Brendicke R Chinn BP Integration of needs
of moped and motorcycle riders into safety measures review and
statistical analysis in the framework of the European research
project PROMISING SWOV Institute for Road Safety Research
Leidschendam The Netherlands 2001 [ Report No Dndash2001ndash5]
Osga 1979
Osga G Ellingstad VS An evaluation of the effectiveness of the 1978 South Dakota Motorcycle Riders Course Program South Dakota
University of South Dakota Vermillion Human Factors Lab 1979
Oxford English Dictionary Online
httpoxforddictionariescom Oxford English Dictionary Online
Oxford University Press 2010
Peden 2004
Peden M Scurfield R Sleet D World Report on Road TrafficInjury Prevention Geneva World Health Organization Available
at httpwwwwhointviolence˙injury˙preventionpublications
road˙trafficworld˙reporten 2004Chapter 3
Radin Umar 1995
Radin Umar R Macakay G Hills B Preliminary analysis of
motorcycle accidents Short-term impacts of the running
headlights campaign and regulations in Malaysia J Traffic Med 199523(1)17ndash27
Tsai 1995
Tsai YJ Wang JD et alCase-control study of the effectiveness of
different types of helmets for the prevention of head injuries among
motorcycle riders in Taipei Taiwan American Journal of
Epidemiology 1995142(9)974ndash81
Vis 1995
Vis AA In-depth study of the hazards of motorcycling [De
onveiligheid van motorrijden nader bekeken] SWOV Institute for
Road Safety Research Leidschendam the Netherlands 1995 [
Report No Rndash95ndash69]
Weaver 1990
Weaver EA Motorcycle safety education in the United States
Motorcycle Safety Foundation Orlando Florida USA 1990
Winn 1987
Winn G Two important trends in motorcycle safety regulations
rider education and conspicuity improvement No more
information available Detroit MI US Society of Automotive
Engineers 1987147ndash52
Yuan 2000
Yuan W The effectiveness of the rsquoride-brightrsquo legislation for
motorcycles in Singapore Accident Analysis amp Prevention 200032
(4)559ndash63
Zambon 2006
Zambon F Hasselberg M Factors affecting the severity of injuries
among young motorcyclists a Swedish nationwide cohort study
Traffic Injury Prevention 20067(2)143ndash9lowast Indicates the major publication for the study
29Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3250
C H A R A C T E R I S T I C S O F S T U D I E S
Characteristics of included studies [ordered by study ID]
Adams 1985
Methods Non-randomised trial
Participants Initial sample size
Total of 162 Australia Post telegram delivery riders
Not clear how many in intervention and control groups
Analysis sample size
Intervention group 63Control group 35
Total 98 (61)
Interventions No formal training vs voluntary rider training course of 12 hours over 2 consecutive days (theory 4 hours practice
8 hours)
Intervention integrity
No information given
Outcomes Australia Post recorded crashes
Length of follow up 1 year
Notes Study conducted in NSW Australia
Anderson 1980
Methods Randomised trial
Participants Initial sample size
Total 41188 motorcycle licence applicants
Analysis sample size
Intervention group with skills training 13160
Intervention group without skills training 12634
Control group 15080
Total 40874 (99)
Interventions Standard licensing skill test vs the Motorcycle Operators Skill Test (MOST) with or without a training component
MOST is not to be confused with rider training because it is a licensing test The training component was for
participants who failed the skill test on the first attempt was of 3 hours duration and consisted of instructions about
the skills needed to pass the skill test The training consisted of 30 min theory and 2 frac12 hours practice on the
motorcycle
Intervention integrity
33 in MOST with skills training 31 in MOST without skills training and 43 in control group obtained their
motorcycle licence in the end
Outcomes Police recorded crashes and offences
Length of follow up 3 years
30Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3350
Anderson 1980 (Continued)
Notes Study conducted in California USA
Billheimer 1998
Methods Cohort study
Participants Initial sample size
Intervention group Unclear numbers of CMSP graduates and ERC graduates
Control group Over 16000 untrained motorcyclists
Analysis sample size
1139 matched CMSP-pairs and 1182 ERC-pairs
Interventions No formal training vs the mandatory (for those younger than 18 years) pre-licence 16-hours basic California
Motorcyclist Safety Program (CMSP) and the 8-hour Experienced Rider Course (ERC)
Intervention integrity
No information given
Outcomes Police recorded crashes and offences
Length of follow up 2 years
Notes Study conducted in California USA
Cooper 1988
Methods Cohort study
Participants Initial sample size
Intervention group 418 graduates and 57 who failed
Control group 402
Analysis sample size
Intervention group unclear
Control group unclear
Total 863 (984)
Interventions No formaltrainingvs non-mandatory pre-licence rider trainingThe content of thetrainingwas notfurther explainedIntervention integrity
Participants had all completed the rider training
Outcomes Police recorded crashes and offences
Length of follow up 7 years
Notes Study conducted in British Columbia Canada
31Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3450
Davis 1997
Methods Retrospective cohort study
Participants Initial sample size
Intervention group 9320 graduates
Control group unclear
Analysis sample size
Intervention group 9320 graduates
Control group unclear
Interventions No formal training vs the non-mandatory Connecticut Motorcycle Operator Training Program The content of
training was not further explained
Intervention integrity Participants had all completed the rider training
Outcomes Police recorded crashes severity of crashes
Length of follow up 1 year
Notes Study conducted in Connecticut USA
Hall 1988
Methods Randomised trial
Participants Initial sample sizeIntervention group A 6248
Intervention group B 6616
Intervention group C 6996
Control group 6604
Total 26464 motorcycle licence applicants
Analysis sample size
Intervention group A 6012 (96)
Intervention group B 6428 (97)
Intervention group C 6778 (97)
Control group 6502 (98)
Interventions Standard licensing skill test vs the Motorcycle Operator Skill Test (MOST II) without training (A) MOST II with
3 hours training (B) or MOST II with 20 hours training (C) The 3 hours training comprised instructions aboutthe skills needed to pass the skill test (30 min theory and 2 frac12 hours practice on the motorcycle) The 20 hours
training consisted of the Motorcycle Safety Foundationrsquos Motorcycle Rider Course (MRC) comprising of 8 hours
theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-road practice
Intervention integrity
25 and 26 obtained their motorcycle licence in the lsquoMOST 3-hour trainingrsquo and lsquoMOST 20-hour trainingrsquo
intervention arm respectively In the lsquoMOST testrsquo intervention arm 29 obtained their licence and in the control
group 35
Outcomes Police recorded crashes and offences
Length of follow up 3 years
32Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3550
Hall 1988 (Continued)
Notes Study conducted in New York USA
Haworth 1997
Methods Population based case-control study
Participants Cases Motorcycle riders or pillion riders involved in motorcycle crash (on road) Recruited from hospital and police
records Not clear whether all cases (especially fatalities) were included
Community controls Multiple matched sample of motorcycle rider or pillion who passed the crash site at the same
time of day and day of week that the crash occurred
Initial sample size
Cases 336
Controls 1195
Analysis sample size
Cases 234 (70)
Controls 561 (47)
Interventions Basic handling skills course beginner intermediate or advanced course measured through self-report not validated
Outcomes Injuries with an Injury Severity Score of 5 or greater or admission for at least four hours and fatalities from a
motorcycle crash (from hospital and police records)
Length of follow up two years or more prior to the interview
Notes Study conducted in Victoria Australia
Hurt 1981
Methods Population based case-control study
Participants Cases Motorcycle riders who had a motorcycle crash that was reported to police (traffic accident reports) Not all
cases included
Community controls Multiple matched sample of motorcycle riders who passed the crash site at the same time of
day and day of week as the crash occurred but 2 years later
Initial sample size
Cases 4500 notifications of crashes to police (only 50 received in time research team able to respond to 1126 for
unreported reasons data collection completed for 900)
Controls Number approached unreported 2310 consented and interviewed
Analysis sample size
Cases 808 (808900 data on rider training) - 18 of initial sample
Controls 670 (6702310 data on rider training) - lt 29 of initial sample
Interventions Training experience self taught friends-family school-club formal-professional others measured through self-
report validated
Outcomes Injuries and fatalities from a motorcycle crash (from hospital records and coronerrsquos office)
33Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3650
Hurt 1981 (Continued)
Notes Study conducted in Los Angeles USA
Jonah 1982
Methods Cohort study
Participants Initial sample size
Intervention group
2310 licensed MTP graduates who are active riders
Control group
2029 licensed motorcyclists who are active riders
Non-licensed and non-active riders excluded
Analysis sample size
Intervention group
811 interviews (35) 533 driver records (23)
Control group 1080 interviews (71) 905 driver records (45)
Interventions No training vs the non-mandatory Canada Safety Council Motorcycle Training Program (MTP) comprising of 4
hours theory and 16 hours practise (6 hrs on-road) over 2 weekends
Intervention integrity
Participants had all successfully completed the MTP
Outcomes Self-reported and police reported crashes and offencesLength of follow up 4 years
Notes Study conducted in Ontario Canada
Kloeden 1994
Methods Cohort study
Participants Initial sample size
Intervention group 2533 licence applicants
Control group Unclear
Sample of riders obtaining probationary licenceIntervention group 1625 (64)
Control group 5015 (unclear)
Non-active riders excluded from control group (~66)
Analysis sample size
Intervention group 631 (39)
Control group 318 (6)
Interventions No formal training vs the mandatory pre-licence Ridersafe course consisting of two levelsthe first is a training course
in the basics of motorcycle riding where a pass entitles the rider to obtain a learnerrsquos permit The second is a skill
test where a pass entitles the rider to obtain a probationary motorcycle licence The training course consists of two 4
hour training sessions on separate days involving instruction videos and actual riding on motorcycles The skill test
consists of a 4 hour preparation course of 74 minutes of lecturediscussion 38 minutes of video and 75 minutes of
34Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3750
Kloeden 1994 (Continued)
on-course riding which is the testing phase
Intervention integrity
Unclear what percentage in the Ridersafe group and the control group obtained their probationary licence halfway
and their full motorcycle licence in the end
Outcomes Police reported crashes
Length of follow up 1 year
Notes Study conducted in SA Australia
Kraus 1975
Methods Population based case-control study
Participants Cases All motorcycle riders with a confirmed medically treated motorcycle injury (including fatalities) from all
hospitals in study base and peripheral area hospitals and all deaths due to motorcycle crashes in the study base All
serious injuries (leading to hospitalisation) and fatalities likely to be included
Community controls A random sample of registered motorcycle owners in the same temporary and geographical
study base as the cases arose from
Initial sample size Cases 1273
Controls 738
Analysis sample size Cases 741 (58)
Controls 434 (59)
Interventions Motorcycle rider training (not specified) measured through self-report
Outcomes Injuries leading to hospitalisation fatalities
Length of follow up not specified
Notes Study conducted in Sacramento county USA
Leung 1987
Methods Non-randomised trial
Participants Initial sample sizeIntervention group unknown number of MRC graduates
Control group unknown number of licensed motorcyclists
Analysis sample size
Intervention group 2941
Control group 43094
Interventions No formal training vs non-mandatory Wisconsin motorcycle rider course a 22 hours off-street programme that
teaches the basic rules of safe motorcycling
Intervention integrity
50 in the MRC group and 100 in the control group obtained their motorcycle licence in the end
35Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3850
Leung 1987 (Continued)
Outcomes Police recorded crashes and offences
Length of follow up up to 3 years
Notes Study conducted in Wisconsin USA
MAIDS 2004
Methods Population based case-control study
Participants Cases Motorcycle riders who had a motorcycle crash Unclear whether all cases included
Community controls Motorcycle riders identified at petrol stations Unclear whether matched to cases
Initial sample size
Cases not reported
Controls not reported
Analysis sample size
Cases 921 (Italy 200 Netherlands 200 Spain 121 Germany 250 France 150)
Controls 923 (Italy 200 Netherlands 200 Spain123 Germany 250 France 150)
Interventions Pre-license rider training or additional training (non-compulsory)
Outcomes Crashes
Notes Study in five sampling areas located in five European countries (France Germany Netherlands Spain and Italy)
McDavid 1989
Methods Cohort study
Participants Sample size after first match
Intervention group 346 male training graduates
Control group 346 male motorcycle riders licensed in same year
Sample size after second match
Intervention group 139 (40)
Control group 139 (40)
Analysis sample size
Intervention group 139 (40)
Control group 139 (40)
Interventions No formal training vs the non-mandatory pre-licence British Columbia Safety Councilrsquos 37-hour motorcycle safety
programme
Intervention integrity
Participants had all completed and passed the training
Outcomes Police reported crashes injurious crashes
Length of follow up 1-6 years
36Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3950
McDavid 1989 (Continued)
Notes Study conducted in British Columbia Canada
Mortimer 1984
Methods Cohort study
Participants Initial sample size
Intervention group 608 MRC graduates
Control group 360 motorcycle riders
Response rates
Intervention group 360 (59)
Control group 324 (90)
Non-active riders excluded
Analysis sample size
Intervention group 213 (35)
Control group 303 (84)
Interventions No formal training vs the non-mandatory Motorcycle Safety Foundationrsquos Motorcycle Rider Course comprising of
8 hours theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-road practice
Intervention integrity
Participants had all completed and passed the MRC
Outcomes Self-reported crashes and offencesLength of follow up 1 year
Notes Study conducted in Illinois USA
Mortimer 1988
Methods Cohort study
Participants Initial sample size
Intervention group 2307 MRC graduates
Control group 556 active licensed motorcycle riders
Response ratesIntervention group 1298 (56)
Control group 500 (90)
Non-active riders excluded
Analysis sample size
Intervention group 913 (40)
Control group 500 (90)
Interventions No formal training vs the non-mandatory Motorcycle Safety Foundationrsquos Motorcycle Rider Course comprising of
8 hours theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-road practice
Intervention integrity
Participants had all completed and passed the MRC
37Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4050
Mortimer 1988 (Continued)
Outcomes Self reported crashes and offences
Length of follow up 1 year
Notes Study conducted in Illinois USA
Osga 1980
Methods Cohort study
Participants Initial sample sizeIntervention group 1163 MSF MRC participants invited 818 (74) responded
Control group 329 motorcycle riders responded to recruitment strategies
Non-active riders excluded from analysis
Analysis sample size
Intervention group 595 (73)
Control group 329 (100)
Interventions No formal training vs the non-mandatory Motorcycle Safety Foundationrsquos Motorcycle Rider Course (MRC) com-
prising of 8 hours theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-road
practice
Intervention integrity
Participants had all successfully completed the MRC
Outcomes Self-reported crashes police recorded crash reports
Length of follow up 6 months - 3 12 years
Notes Study conducted in South Dakota USA
Perrino 2002
Methods Cohort study
Participants Initial sample size
Intervention group 2144 (unclear) MRC graduates (motorcycle licence applicants)
Control group 2601 (unclear) regular licensing test graduates (motorcycle licence applicants)Analysis sample size
Intervention group 1070 (unclear)
Control group 867 (unclear)
Interventions No formal training vs the mandatory (for those younger than 18 years) pre-licence Maryland Motorcycle Rider
Course (MRC) consisting of 8 hours theory and 12 hours practice
Intervention integrity
No information given
Outcomes Self-reported and police recorded crashes and offences
Length of follow up 1 year
38Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4150
Perrino 2002 (Continued)
Notes Study conducted in Maryland USA
Raymond 1979
Methods Cohort study
Participants Initial sample size
Intervention group
258 volunteers mainly (but not all) novice riders
Control group
1104 provisional licence applicants
Analysis sample size
Intervention group 190 (74)
Control group 627 (57)
Interventions No formal training vs the non-mandatory RACACU (Royal Automobile ClubAuto-Cycle Union) training scheme
The training program consisted of either 24 one-hour sessions or 13 two-hour sessions Twelve hours of the training
focused on theory about basic operational procedures traffic laws and highway code The practical session focused
on basic operational skills riding on a training area and riding in traffic with an instructor A proficiency test was
administered at the end of the course
Intervention integrity
No information given
Outcomes Self-reported crashes (including minor non-injury non-damage and off-road crashes)
Length of follow up 1 year
Notes Study conducted in the UK
Satten 1980
Methods Cohort study
Participants Initial sample size
Intervention group interviewers kept interviewing until they had 100 riders selected from a population of 891
enrollees in the MSF MRC course (unknown how many were asked and how many refused to participate)Control group unknown number of registered motorcycle riders
Non-active riders excluded from analysis
Analysis sample size
Intervention group 71
Control group 100
Interventions No formal training vs the non-mandatory Motorcycle Safety Foundationrsquos (MSF) Motorcycle Rider Course (MRC)
comprising of 8 hours theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-
road practice
Intervention integrity
Participants had all successfully completed the MRC
39Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4250
Satten 1980 (Continued)
Outcomes Self-reported crashes injuries and offences
Length of follow up 3 years
Notes Study conducted in Illinois USA
Savolainen 2007
Methods Cohort study
Participants Initial sample size
Intervention group 4000 riders that had conducted the MSF MRC course (passed or failed)
Control group 4000 motorcycle owners or permit or endorsement holders from the Indiana Bureau of Motor
Vehicles database
Analysis sample size
Intervention group 739 (18)
Control group 588 (15)
Interventions Intervention integrity
No information given
Outcomes Probability of crash involvement
Length of follow up up to 5 years
Notes Study conducted in Indiana USA
Swaddiwudhipong 1998
Methods Randomised trial
Participants Initial sample size
Intervention group
all 16 villages of 3 sub-districts
(participants were practising motorcyclists licensed and unlicensed)
Control group
18 villages of 3 other sub-districtsAnalysis sample size
Intervention group 3 districts (100)
Control group 3 districts (100)
Interventions No rider education vs non-mandatory community based rider education consisting of theory on motorcycle injury
risk helmet use traffic laws vehicle regulations and traffic signs and a written and skill test leading to a licence
Intervention integrity
No information given
Outcomes Injuries fatalities
Length of follow up 2 years
40Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4350
Swaddiwudhipong 1998 (Continued)
Notes Study conducted in Tak Province rural Thailand
Waller 1992
Methods Cohort study
Participants Initial sample size
Intervention group 349 riders who succesfully completed the Department of Public Safety motorcycle operator
training in 1988
Control group 349 motorcyclists who never completed the DPS motorcycle operator training
Analysis sample size
Intervention group 349
Control group 349
Interventions No DPS motorcycle operator training vs non-mandatory DPS motorcycle operator training The DPS training is
the Motorcycle Safety Foundationrsquos Motorcycle Rider Course comprising eight hours of theory (lectures discussions
films and multiple choice tests) a textbook and 12 hours off-road practice
Intervention integrity
All 349 riders succesfully completed the rider training
Outcomes Police recorded crashes
Length of follow up 3 years
Notes Study conducted in
Characteristics of excluded studies [ordered by study ID]
Study Reason for exclusion
Anderson 1980a Study type review and process evaluation of a rider course
Aupetit 2008 Outcome proposal of a rider training curriculum that combines traditional training and a motorcycle riding
simulator
Awane 1999 Study type review on use of simulators in rider and driver training
Baldi 2005 Intervention not rider training but intervention integrity of rider training availabilty of rider education
implementation of graduated licensing system etc on a state-wide level
Batchler 1988 Study type description of a training course
Blanchard 2006 Study type not a study assessing the association between rider training and outcomes
Bonnett 2005 Outcome qualitative data regarding satisfaction with the training programme
41Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4450
(Continued)
Braver 2007 Outcome licensure rates
Clayton 1980 Outcome knowledge skills attitude
Council 1978 Outcome crash reduction in ANY vehicle not motorcycles exclusively
Crowther 2005 Study type qualitative study
Daniello 2009 Study type review of the effectiveness of motorcycle training and licensing
Dewanti 2007 Intervention emotional characteristics behaviour (ie not motocycle rider training)
Du Plooy 1990 Study type review of types of rider training courses in South Africa
Engel 1989 Study type uncontrolled before-after study
Goldenbeld 2002 Outcome skills
Goldenbeld 2004 Outcome knowledge and skills
Haworth 2000 Intervention balance between attitudinal andvehicle control skills acrosscurriculla of motorcycle rider training
courses
Haworth 2005 Study type review
Haworth 2006 Study type cross-sectional survey
Hill 1983 Study type review and description of training system
Imatake 1980 Study type course descriptionreview
Ingebritsen 1990 Intervention risk factors for motorcycle crashes
Jain 2009 Study type cross-sectional descriptive study of motorcycle crashes over a 5 year period
Jonah 1979 Study type cross-sectional survey Outcome scores on motorcycle operator skill test
Jonah 1981 Intervention evaluation of a motorcycle rider skill test not motorcycle rider training
Kadar 1990 Study type review of components of rider training courses
Kelsey 1986 Intervention evaluation of a motorcycle rider skill test not motorcycle rider training
Koch 1980 Study type reviewcourse description
Koch 1990 Study type description of course content
42Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4550
(Continued)
Labbett 2005 Study type essay about possible explanations for motorcycle deaths
Law 2005 Intervention targeted motorcycle safety programme not specifically rider training
Lowes 1990 Study type review and descriptions of rider training
Mahon 1976 Study type review of motorcycle rider training methods
Maxwell 1985 Outcomesrecommendations forprogramme operatorsand otherson howthe motorcycle trainingprogramme
works and how it can be improved
Mayhew 1996 Study type review
McKnight 1987 Study type evaluation of methodological and administrative aspects of a motorcycle operator skill test
McPherson 1976 Outcome knowledge and skills
Mohan 2004 Study type review of evidence of interventions to reduce road traffic injuries
Morgan 1980 Study type reviewcourse description
Motoki 1990 Study type proposal and development of rider training curriculum
Nairn 1992 Study type literature review
Newland 1980 Study type description of development of motorcycle safety course
Newland 1982 Study type descriptionreview of a rider training course
Newland 1990 Study type uncontrolled BeforeAfter study
Ormston 2003 Outcome changes in attitude and self reported behaviour
Osga 1979 Outcome knowledge attitudes and skills
Panou 2007 Intervention new simulation tool to teach driverrider
Peterson 1990 Study factor demographic details of course participants completing the course vs those not completing the
course
Prem 1984 Study type experimental
Study factor examines association between skills and performance in tests
Radin Umar 2006 Intervention motorcycle safety programme not specifically rider training
Rockwell 1990 Study type methodologically flawed study type
43Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4650
(Continued)
Rothe 1987 Study type qualitative study of rider training in general
Rowden 2007 Study type qualitative study
RSC Victoria 1993 Study type review of motorcycle rider training and education in Victoria
Russam 1979 Study type review of training programmes in Japan
Saffron 1981 Study type literature review
Schulz 1990 Study factor predictors of motorcycle crashes
Sexton 2004 Study type cross-sectional analytic analysis
Shephard 1986 Study factor reviewdescription of programme
Simpson 1990 Study type review of motorcycle rider trainingeducation
Sokolov 1990 Study factor review of motorcycle riding courses in USSR
Steffens 1988 Outcome self-reported behaviour experiences and researcher observed riding behaviour
Sudlow 2003 Study type review of motorcycle training in the UK
Tasmania 1995 Study type cross-sectional descriptiveprocess evaluation
Tetard 1985 Study type experimental
Outcome detection of indices
Thackray 1980 Outcome instructional effectiveness user acceptance administrative feasibility
Thompson 1994 Study type cross-sectional survey
Outcome attitudes and knowledge
Tobler 1990 Study type review of course curricula
Transport 1989 Intervention syllabus for rider training
Tuan 2007 Study type cross-sectional study (survey) which does not assess the effects of an intervention but examines
motocyclistsrsquo behaviour
Victor 1980 Study type review
Waterer 1990 Study type review of UK rider training issues
Weller 1989 Study type review of studies evaluating motorcycle safety and education measures
44Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4750
(Continued)
Wells 1981 Study type description of different methods of training
Winn 1986 Study type review of rsquohistoryrsquo of rider training
Wisher 1988 Study type cross-sectional survey
Wood 1987 Study type Interrupted Time Series but did not meet the criteria for an ITS as published by the Cochrane
Effective Practice and Organisation of Care Group
Yamazaki 1990 Study type cross-sectional survey
Outcome riding skills skills retention and attitude
Characteristics of ongoing studies [ordered by study ID]
Burgess 2010
Trial name or title UKs National Ride program
Methods Cohort study
Participants Motorcycle offenders
Interventions Police led diversion program
Outcomes Self reported attitudinal and behavioural measures
Starting date
Contact information
Notes Expected to be completed in early 2010
45Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4850
D A T A A N D A N A L Y S E S
This review has no analyses
A P P E N D I C E S
Appendix 1 Search strategy
We searched the following electronic databases
bull Cochrane Injuries Group Specialised Register (searched 22 Sept 2008)
bull CENTRAL (The Cochrane Library 2008 Issue 3)
bull TRANSPORT 1988 to 200706 (latest date available)
bull ATRI Australian Transport Index (to Sept 2008)
bull MEDLINE (Ovid SP) 1950 to Sept 2008
bull EMBASE (Ovid SP) 1980 to Sept 2008
bull CINAHL (EBSCO) 1982 to Sept 2008
bull WHOLIS (World Health Organization Library Information System) (to Sept 2008)
bull PsycINFO (Ovid SP) 1806 to Sept 2008
bull LILACS (BIREME) (Latin American and Caribbean Health Sciences) (1982 to Sept 2008)
bull ISI Web of Science Social Sciences Citation Index (SSCI) 1970 to Sept 2008
bull ERIC (searched 25 Sept 2008)
bull ZETOC (searched 25 Sept 2008)
bull SIGLE (searched 25 Sept 2008)
All of the databases were searched using the following text-word search strategy The strategy was adapted using either rsquorsquo or rsquo$rsquo as atruncation symbol
1 motor-cycl or motorcycl or motor-bik or motorbik or motor-scooter or motorscooter or scooter or moped or motocycl
or motocicl or moto
2 train or educat or instruct or program or course or qualif or learn or improv
3 1 and 2
H I S T O R Y
Protocol first published Issue 2 2005
Review first published Issue 10 2010
46Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4950
C O N T R I B U T I O N S O F A U T H O R S
Katina Kardamanidis was the first author of the protocol conducted the literature search extracted full studies assessed the studies
according to inclusion criteria assessed quality of the studies compared findings with the second author discussed disagreements with
the third author wrote the drafts and the final review
Rebecca Ivers was the second author of the protocol assessed the studies according to inclusion criteria assessed quality of the studies
compared findings with the first author edited the final draft and provided overall guidance for the process
Alexandra Martiniuk assessed quality of the studies compared findings with the first author discussed disagreements with the third
author and helped write the first draft and the final review
Mark Stevenson contributed to the protocol reviewed methods of studies and contributed to the final draft
Katrina Thistlethwaite conducted a second literature search extracted full studies and contributed to the final draft
D E C L A R A T I O N S O F I N T E R E S T
None known
S O U R C E S O F S U P P O R T
Internal sources
bull No sources of support supplied
External sources
bull National Health and Medical Research Council of Australia Australia
Salary support for Martiniuk Ivers and Stevenson
bull University of Sydney Sesqui Research Grant Australia
Direct research costs and salary support for Kardamanidis
D I F F E R E N C E S B E T W E E N P R O T O C O L A N D R E V I E W
There are no differences between the protocol and the review
I N D E X T E R M S
Medical Subject Headings (MeSH)
lowastMotorcycles Accident Prevention [lowastmethods] Accidents Traffic [lowastprevention amp control] Licensure Program Evaluation Wounds
and Injuries [mortality lowastprevention amp control]
47Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 5050
MeSH check words
Humans
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 2950
Conference Orlando Florida USA Motorcycle Safety
Foundation 19909-26 to 9-57A
Nairn 1992 published data onlylowast Nairn RJ Motorcycle safety research literature review 1987 to
1991 Federal Office of Road Safety Australia 1993 Vol Report
CR 117
Newland 1980 published data only
Newland R Darby J Automotive practicespre motorcycle rider
program International Motorcycle Safety Conference Washington
DC USA Motorcycle Safety Foundation 1980127ndash9
Newland 1982 published data only
Newland R Motorcycle rider training and licensing requirements
Driver training steering a course for the future Melbourne
Victoria Transport Regulation Board 1982
Newland 1990 published data only
Newland R Motorcycle training and licensing in Victoria 1980-
1990 The Human Element International Motorcycle Safety
Conference Orlando Florida USA Motorcycle Safety
Foundation 1990
Ormston 2003 published data only
Ormston R Dudleston A Pearson S Stradling Evaluation of
Bikesafe Scotland Scottish Executive Social Research Edinburgh
2003
Osga 1979 published data only
Osga GS Ellingstad VS An evaluation of the effectiveness of the 1978
South Dakota motorcycle rider course program South Dakota USA
University of South Dakota 19791ndash89
Panou 2007 published data only
Panou M Bekiaris E A new concept on the integration of driving
simulators in driving training the train-all approach Road Safety
on Four Continents 14th International Conference Swedish
National Road and Transport Research Institute 2007
Peterson 1990 published data only
Peterson A Kadar K Status report 1980-1989 Minnesota rider
education statistics The Human Element International
Motorcycle Safety Conference Orlando Florida USA 199013-
113 to 13-129
Prem 1984 published data only
Prem H Good MC Motorcycle rider skills assessment ACT
Australia Office of Road Safety Department of Transport 1984
1ndash377 [ Report No CR 34 ]
Radin Umar 2006 published data only
Radin Umar RS Motorcycle safety programmes in Malaysia how
effective are they International Journal of Injury Control and Safety Promotion 200613(2)71ndash9
Rockwell 1990 published data only
Rockwell TH Kiger S Carnot M An evaluation of the Ohio
motorcyclist enrichment program Ohio Department of Highway
Safety 1990
Rothe 1987 published data only
Rothe JP A motorcycle riding course Motorcyclists Image and reality Canada Insurance Corporation of British Columbia 1987
59ndash83
Rowden 2007 published data only
Rowden P Watson B Haworth N What can riders tell us aboutmotorcycle training A view from the other side of the fence
Australasian Road Safety Research Policing and Education
Conference Melbourne Australia Melbourne 2007
RSC Victoria 1993 published data only
Road Safety Committee Inquiry into Motorcycle Safety in Victoria
Parliament of Victoria 1993
Russam 1979 published data only
Russam K Motorcycle training and licensing in Japan UK Transport
and Road Research Laboratory (TRRL) 1979 [ Report 916 ]
Saffron 1981 published data only
Saffron DG Driver and rider education training and licensing a
brief review Traffic Accident Research Unit Traffic Authority of
New South Wales NSW Australia Traffic Accident Research UnitTraffic Authority of New South Wales 1981
Schulz 1990 published data only
Schulz U Motorcycle accidents among learner riders Motorcycle
licensing Evaluation of the effectiveness of a Graduated Licensing Scheme Forshungshefte des Institut fur Zweiradsicherheit 1990
Sexton 2004 published data only (unpublished sought but not used)
Sexton B Baughan C Elliott M Maycock G The accident risk of
motorcyclists 2004TRL Report TRL 607(ISSN 09684107)58
Shephard 1986 published data only
Shepherd R Do Motorcycle Safety Foundation programs work
Driver 198619(9)16ndash8
Simpson 1990 published data onlySimpson HM Mayhew DR The promotion of motorcycle safety
training education and awareness Health Education Research
Theory and Practice 19905(2)257ndash64
Sokolov 1990 published data only
Sokolov O Safe motorcycle riding some aspects of motorcycle
riding in the USSR The Human Element International
Motorcycle Safety Conference Motorcycle Safety Foundation
19904ndash37 - 4-48
Steffens 1988 published data only
Steffens U Gawatz R Willmes G Effectiveness of mofa courses
[Wirksamkeit von Mofakursen] Unfall- und sicherheitsforschung
strasenverkehr 198867168
Sudlow 2003 published data only
Sudlow D Scoping study on motorcycle training Road Safety
Research Report London UK The British Institute of Traffic
Education Research Department for Transport 2003 Vol June
[ Report No 36]
Tasmania 1995 published data only
Mannering FL Grodsky LL Motorcycle rider training scheme
Traffic Education Research Centre Armidale Australia Driver
Training Australia Limited 1995 Vol April1ndash56
Tetard 1985 published data only
Tetard C Motorcycle ridersrsquo training to detect indices in an
intersection situation [Lrsquoentrainement a la prise drsquoinformation dans
lrsquoapprentissage de la conduite motocycliste] Le Travail Humain198548(2)147ndash59
27Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3050
Thackray 1980 published data only
Thackray RM Prescott JC Field test of the motorcycle ridercourse International Motorcycle Safety Conference Washington
DC USA Motorcycle Safety Foundation 1980195ndash7
Thompson 1994 published data only
Thompson M Evaluation of compulsory basic training for
motorcyclists Transport Research Laboratory Crowthorne
Berkshire UK 1994 [ Report 63]
Tobler 1990 published data only
Tobler U Individual driving instruction for motorcycle driving
learners the Swiss experiment of motorcycle driving school The
Human Element International Motorcycle Safety Conference
Orlando Florida USA Motorcycle Safety Foundation 1990
Transport 1989 published data only
Transport Communications Road Crash Statistics Australia Bureauof Transport and Communications Economics 198914ndash20
Tuan 2007 published data only
Tuan VA Shimizu T Towards development and evaluation of the
motorcycle drivers re-education program in Vietnam modeling of
motorcycle driverrsquos undesired behaviors 11th World Conference
on Transport Research 200725
Victor 1980 published data only
Victor DJ Traffic safety education for motorcyclists International
Motorcycle Safety Conference Washington DC USA Motorcycle
Safety Foundation 1980221ndash8
Waterer 1990 published data only
Waterer TH Compulsory training for motorcyclists from liberty
to compulsion The Human Element International Motorcycle
Safety Conference Orlando Florida USA Motorcycle Safety
Foundation 19904-49 to 4-57
Weller 1989 published data only
Weller RB Chandler EW Motorcycle safety and motorcycle
education past research and survey results Journal of Public Policy
amp Marketing 1989893ndash108
Wells 1981 published data only
Wells P Observations of some RACACU motorcycle training
schemes Transport and Road Research Laboratory Crowthorne
Berkshire UK 1981
Winn 1986 published data only
Winn G Two important trends in motorcycle safety regulations
rider education and conspicuity improvement American
Motorcyclist Association 1986147ndash52
Wisher 1988 published data only
Wisher G Reid J Motorcycle accidents in Norfolk a three-year study of motorcycle training and the Part 1 test Norfolk UK Road Safety
Division Highways Department Norfolk County Council 1988
Wood 1987 published data only
Wood HT Bowen R Evaluation of the revised learner permit scheme
July 1983 to December 1985 Victoria Australia Road Traffic
Authority 19871ndash27 [ ISBN 0 7306 0247 8]
Yamazaki 1990 published data only
Yamazaki S Nakanishi M Survey on effect of Honda rider
education The Human Element International Motorcycle Safety
Conference Orlando Florida USA 199012-167 to 12-187
References to ongoing studies
Burgess 2010 unpublished data only
Burgess C Broughton P Fylan F Stradling Interim evaluation of
the UKs National Ride program In L Dorn editor(s) Driver
behaviour and Training Vol IV Aldershot Ashgate Publishing
Limited 2010
Additional references
Anderson 1980
Anderson J Ford JL Peck RC Improved motorcyclist licensing and
testing project Volume I and II National Highway Traffic Safety
Administration US department of Transportation Washington
DC 1980
Bjornstig 1985
Bjornstig U Bylund P Lekander T Brorsson B Motorcycle fatalitiesin Sweden Acta Chirurgica Scandinavica 1985151(7)577ndash81
Clayton 1990
Clayton A Sudlow DE An evaluation of the effectiveness of a one-
day training program for learner motorcyclists The Human
Element International Motorcycle Safety Conference Orlando
Florida USA 199013-105 to 13-111
Cochrane Collaboration 2008
Higgins JP Green S Cochrane Handbook for Systematic Reviews of
Interventions 500 The Cochrane Collaboration 2008
Elliot 2003
Elliott MA Baughan CJT Broughton J Grayson G Knowles J et
alMotorcycle Safety a Scoping Study TRL [ Report No
0968ndash4107]Fleiss 2004
Fleiss JL Levin B Paik MC Statistical methods for rates and proportions 3rd Edition John Wiley amp Sons Inc 2004 [ ISBN
978ndash0ndash471ndash52629ndash2 ]
FORS 1999
FORS Federal Office of Road Safety Road risk for sober licensed
motorcyclists Vol Monograph 27 Federal Office of Road Safety
19991ndash2
Haworth 2005
Haworth N Mulvihill C Review of motorcycle licensing and training
Victoria Australia Monash University Accident Research Centre
20051ndash83 [ Report No 240]
Huwiler-Muumlntener 2002Huwiler-Muumlntener K Juumlni P Junker C Egger M Quality of
Reporting of Randomized Trials as a Measure of Methodologic
Quality JAMA 20022872801ndash4
Johnston 2008
Johnston P Brooks C Savage H Fatal and serious road crashes
involving motorcyclists Road Safety Research and Analysis Report
ATSB Monograph 20 Canberra Department of Infrastructure
Transport Regional Development and Local Government 0205
081ndash26
Juumlni 2001
Juumlni P Altman DG Egger M Systematic reviews in health care
Assessing the quality of controlled clinical trials BMJ 2001323
42ndash6
28Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3150
Kopits 2003
Kopits E Cropper M Traffic fatalities and economic growth TheWorld Bank Development Research Group Infrastructure and
Environment Policy research working paper 3035 2003
Leung 1983
Leung P Chau BTW Motorcycle accidents in Hong Kong pattern
of injury treatment causative factors and socio-economic
implications Injury 198315(1)23ndash9
Liu 2004
Liu B Ivers R Norton R Blows S Lo SK Helmets for preventing
injury in motorcycle riders Cochrane Database of Systematic Reviews 2004 Issue 2 [DOI 10100214651858CD004333pub3]
Mohan 2002
Mohan D Road safety in less-motorized environments future
concerns International Journal of Epidemiology 200231(3)527ndash32Mortimer 1982
Mortimer RG An evaluation of the motorcycle rider course
Proceedings of the 20th conference of the American Association for
Automotive Medicine Arlington Heights IL USA American
Association for Automotive Medicine 198277ndash96
Mullin 1997
Mullin BT Risk factors for motorcycle injury the role of age
gender experience training and alcohol PhD thesis Auckland
New Zealand University of Auckland 1997
Noordzij 2001
Noordzij PC Forke E Brendicke R Chinn BP Integration of needs
of moped and motorcycle riders into safety measures review and
statistical analysis in the framework of the European research
project PROMISING SWOV Institute for Road Safety Research
Leidschendam The Netherlands 2001 [ Report No Dndash2001ndash5]
Osga 1979
Osga G Ellingstad VS An evaluation of the effectiveness of the 1978 South Dakota Motorcycle Riders Course Program South Dakota
University of South Dakota Vermillion Human Factors Lab 1979
Oxford English Dictionary Online
httpoxforddictionariescom Oxford English Dictionary Online
Oxford University Press 2010
Peden 2004
Peden M Scurfield R Sleet D World Report on Road TrafficInjury Prevention Geneva World Health Organization Available
at httpwwwwhointviolence˙injury˙preventionpublications
road˙trafficworld˙reporten 2004Chapter 3
Radin Umar 1995
Radin Umar R Macakay G Hills B Preliminary analysis of
motorcycle accidents Short-term impacts of the running
headlights campaign and regulations in Malaysia J Traffic Med 199523(1)17ndash27
Tsai 1995
Tsai YJ Wang JD et alCase-control study of the effectiveness of
different types of helmets for the prevention of head injuries among
motorcycle riders in Taipei Taiwan American Journal of
Epidemiology 1995142(9)974ndash81
Vis 1995
Vis AA In-depth study of the hazards of motorcycling [De
onveiligheid van motorrijden nader bekeken] SWOV Institute for
Road Safety Research Leidschendam the Netherlands 1995 [
Report No Rndash95ndash69]
Weaver 1990
Weaver EA Motorcycle safety education in the United States
Motorcycle Safety Foundation Orlando Florida USA 1990
Winn 1987
Winn G Two important trends in motorcycle safety regulations
rider education and conspicuity improvement No more
information available Detroit MI US Society of Automotive
Engineers 1987147ndash52
Yuan 2000
Yuan W The effectiveness of the rsquoride-brightrsquo legislation for
motorcycles in Singapore Accident Analysis amp Prevention 200032
(4)559ndash63
Zambon 2006
Zambon F Hasselberg M Factors affecting the severity of injuries
among young motorcyclists a Swedish nationwide cohort study
Traffic Injury Prevention 20067(2)143ndash9lowast Indicates the major publication for the study
29Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3250
C H A R A C T E R I S T I C S O F S T U D I E S
Characteristics of included studies [ordered by study ID]
Adams 1985
Methods Non-randomised trial
Participants Initial sample size
Total of 162 Australia Post telegram delivery riders
Not clear how many in intervention and control groups
Analysis sample size
Intervention group 63Control group 35
Total 98 (61)
Interventions No formal training vs voluntary rider training course of 12 hours over 2 consecutive days (theory 4 hours practice
8 hours)
Intervention integrity
No information given
Outcomes Australia Post recorded crashes
Length of follow up 1 year
Notes Study conducted in NSW Australia
Anderson 1980
Methods Randomised trial
Participants Initial sample size
Total 41188 motorcycle licence applicants
Analysis sample size
Intervention group with skills training 13160
Intervention group without skills training 12634
Control group 15080
Total 40874 (99)
Interventions Standard licensing skill test vs the Motorcycle Operators Skill Test (MOST) with or without a training component
MOST is not to be confused with rider training because it is a licensing test The training component was for
participants who failed the skill test on the first attempt was of 3 hours duration and consisted of instructions about
the skills needed to pass the skill test The training consisted of 30 min theory and 2 frac12 hours practice on the
motorcycle
Intervention integrity
33 in MOST with skills training 31 in MOST without skills training and 43 in control group obtained their
motorcycle licence in the end
Outcomes Police recorded crashes and offences
Length of follow up 3 years
30Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3350
Anderson 1980 (Continued)
Notes Study conducted in California USA
Billheimer 1998
Methods Cohort study
Participants Initial sample size
Intervention group Unclear numbers of CMSP graduates and ERC graduates
Control group Over 16000 untrained motorcyclists
Analysis sample size
1139 matched CMSP-pairs and 1182 ERC-pairs
Interventions No formal training vs the mandatory (for those younger than 18 years) pre-licence 16-hours basic California
Motorcyclist Safety Program (CMSP) and the 8-hour Experienced Rider Course (ERC)
Intervention integrity
No information given
Outcomes Police recorded crashes and offences
Length of follow up 2 years
Notes Study conducted in California USA
Cooper 1988
Methods Cohort study
Participants Initial sample size
Intervention group 418 graduates and 57 who failed
Control group 402
Analysis sample size
Intervention group unclear
Control group unclear
Total 863 (984)
Interventions No formaltrainingvs non-mandatory pre-licence rider trainingThe content of thetrainingwas notfurther explainedIntervention integrity
Participants had all completed the rider training
Outcomes Police recorded crashes and offences
Length of follow up 7 years
Notes Study conducted in British Columbia Canada
31Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3450
Davis 1997
Methods Retrospective cohort study
Participants Initial sample size
Intervention group 9320 graduates
Control group unclear
Analysis sample size
Intervention group 9320 graduates
Control group unclear
Interventions No formal training vs the non-mandatory Connecticut Motorcycle Operator Training Program The content of
training was not further explained
Intervention integrity Participants had all completed the rider training
Outcomes Police recorded crashes severity of crashes
Length of follow up 1 year
Notes Study conducted in Connecticut USA
Hall 1988
Methods Randomised trial
Participants Initial sample sizeIntervention group A 6248
Intervention group B 6616
Intervention group C 6996
Control group 6604
Total 26464 motorcycle licence applicants
Analysis sample size
Intervention group A 6012 (96)
Intervention group B 6428 (97)
Intervention group C 6778 (97)
Control group 6502 (98)
Interventions Standard licensing skill test vs the Motorcycle Operator Skill Test (MOST II) without training (A) MOST II with
3 hours training (B) or MOST II with 20 hours training (C) The 3 hours training comprised instructions aboutthe skills needed to pass the skill test (30 min theory and 2 frac12 hours practice on the motorcycle) The 20 hours
training consisted of the Motorcycle Safety Foundationrsquos Motorcycle Rider Course (MRC) comprising of 8 hours
theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-road practice
Intervention integrity
25 and 26 obtained their motorcycle licence in the lsquoMOST 3-hour trainingrsquo and lsquoMOST 20-hour trainingrsquo
intervention arm respectively In the lsquoMOST testrsquo intervention arm 29 obtained their licence and in the control
group 35
Outcomes Police recorded crashes and offences
Length of follow up 3 years
32Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3550
Hall 1988 (Continued)
Notes Study conducted in New York USA
Haworth 1997
Methods Population based case-control study
Participants Cases Motorcycle riders or pillion riders involved in motorcycle crash (on road) Recruited from hospital and police
records Not clear whether all cases (especially fatalities) were included
Community controls Multiple matched sample of motorcycle rider or pillion who passed the crash site at the same
time of day and day of week that the crash occurred
Initial sample size
Cases 336
Controls 1195
Analysis sample size
Cases 234 (70)
Controls 561 (47)
Interventions Basic handling skills course beginner intermediate or advanced course measured through self-report not validated
Outcomes Injuries with an Injury Severity Score of 5 or greater or admission for at least four hours and fatalities from a
motorcycle crash (from hospital and police records)
Length of follow up two years or more prior to the interview
Notes Study conducted in Victoria Australia
Hurt 1981
Methods Population based case-control study
Participants Cases Motorcycle riders who had a motorcycle crash that was reported to police (traffic accident reports) Not all
cases included
Community controls Multiple matched sample of motorcycle riders who passed the crash site at the same time of
day and day of week as the crash occurred but 2 years later
Initial sample size
Cases 4500 notifications of crashes to police (only 50 received in time research team able to respond to 1126 for
unreported reasons data collection completed for 900)
Controls Number approached unreported 2310 consented and interviewed
Analysis sample size
Cases 808 (808900 data on rider training) - 18 of initial sample
Controls 670 (6702310 data on rider training) - lt 29 of initial sample
Interventions Training experience self taught friends-family school-club formal-professional others measured through self-
report validated
Outcomes Injuries and fatalities from a motorcycle crash (from hospital records and coronerrsquos office)
33Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3650
Hurt 1981 (Continued)
Notes Study conducted in Los Angeles USA
Jonah 1982
Methods Cohort study
Participants Initial sample size
Intervention group
2310 licensed MTP graduates who are active riders
Control group
2029 licensed motorcyclists who are active riders
Non-licensed and non-active riders excluded
Analysis sample size
Intervention group
811 interviews (35) 533 driver records (23)
Control group 1080 interviews (71) 905 driver records (45)
Interventions No training vs the non-mandatory Canada Safety Council Motorcycle Training Program (MTP) comprising of 4
hours theory and 16 hours practise (6 hrs on-road) over 2 weekends
Intervention integrity
Participants had all successfully completed the MTP
Outcomes Self-reported and police reported crashes and offencesLength of follow up 4 years
Notes Study conducted in Ontario Canada
Kloeden 1994
Methods Cohort study
Participants Initial sample size
Intervention group 2533 licence applicants
Control group Unclear
Sample of riders obtaining probationary licenceIntervention group 1625 (64)
Control group 5015 (unclear)
Non-active riders excluded from control group (~66)
Analysis sample size
Intervention group 631 (39)
Control group 318 (6)
Interventions No formal training vs the mandatory pre-licence Ridersafe course consisting of two levelsthe first is a training course
in the basics of motorcycle riding where a pass entitles the rider to obtain a learnerrsquos permit The second is a skill
test where a pass entitles the rider to obtain a probationary motorcycle licence The training course consists of two 4
hour training sessions on separate days involving instruction videos and actual riding on motorcycles The skill test
consists of a 4 hour preparation course of 74 minutes of lecturediscussion 38 minutes of video and 75 minutes of
34Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3750
Kloeden 1994 (Continued)
on-course riding which is the testing phase
Intervention integrity
Unclear what percentage in the Ridersafe group and the control group obtained their probationary licence halfway
and their full motorcycle licence in the end
Outcomes Police reported crashes
Length of follow up 1 year
Notes Study conducted in SA Australia
Kraus 1975
Methods Population based case-control study
Participants Cases All motorcycle riders with a confirmed medically treated motorcycle injury (including fatalities) from all
hospitals in study base and peripheral area hospitals and all deaths due to motorcycle crashes in the study base All
serious injuries (leading to hospitalisation) and fatalities likely to be included
Community controls A random sample of registered motorcycle owners in the same temporary and geographical
study base as the cases arose from
Initial sample size Cases 1273
Controls 738
Analysis sample size Cases 741 (58)
Controls 434 (59)
Interventions Motorcycle rider training (not specified) measured through self-report
Outcomes Injuries leading to hospitalisation fatalities
Length of follow up not specified
Notes Study conducted in Sacramento county USA
Leung 1987
Methods Non-randomised trial
Participants Initial sample sizeIntervention group unknown number of MRC graduates
Control group unknown number of licensed motorcyclists
Analysis sample size
Intervention group 2941
Control group 43094
Interventions No formal training vs non-mandatory Wisconsin motorcycle rider course a 22 hours off-street programme that
teaches the basic rules of safe motorcycling
Intervention integrity
50 in the MRC group and 100 in the control group obtained their motorcycle licence in the end
35Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3850
Leung 1987 (Continued)
Outcomes Police recorded crashes and offences
Length of follow up up to 3 years
Notes Study conducted in Wisconsin USA
MAIDS 2004
Methods Population based case-control study
Participants Cases Motorcycle riders who had a motorcycle crash Unclear whether all cases included
Community controls Motorcycle riders identified at petrol stations Unclear whether matched to cases
Initial sample size
Cases not reported
Controls not reported
Analysis sample size
Cases 921 (Italy 200 Netherlands 200 Spain 121 Germany 250 France 150)
Controls 923 (Italy 200 Netherlands 200 Spain123 Germany 250 France 150)
Interventions Pre-license rider training or additional training (non-compulsory)
Outcomes Crashes
Notes Study in five sampling areas located in five European countries (France Germany Netherlands Spain and Italy)
McDavid 1989
Methods Cohort study
Participants Sample size after first match
Intervention group 346 male training graduates
Control group 346 male motorcycle riders licensed in same year
Sample size after second match
Intervention group 139 (40)
Control group 139 (40)
Analysis sample size
Intervention group 139 (40)
Control group 139 (40)
Interventions No formal training vs the non-mandatory pre-licence British Columbia Safety Councilrsquos 37-hour motorcycle safety
programme
Intervention integrity
Participants had all completed and passed the training
Outcomes Police reported crashes injurious crashes
Length of follow up 1-6 years
36Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3950
McDavid 1989 (Continued)
Notes Study conducted in British Columbia Canada
Mortimer 1984
Methods Cohort study
Participants Initial sample size
Intervention group 608 MRC graduates
Control group 360 motorcycle riders
Response rates
Intervention group 360 (59)
Control group 324 (90)
Non-active riders excluded
Analysis sample size
Intervention group 213 (35)
Control group 303 (84)
Interventions No formal training vs the non-mandatory Motorcycle Safety Foundationrsquos Motorcycle Rider Course comprising of
8 hours theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-road practice
Intervention integrity
Participants had all completed and passed the MRC
Outcomes Self-reported crashes and offencesLength of follow up 1 year
Notes Study conducted in Illinois USA
Mortimer 1988
Methods Cohort study
Participants Initial sample size
Intervention group 2307 MRC graduates
Control group 556 active licensed motorcycle riders
Response ratesIntervention group 1298 (56)
Control group 500 (90)
Non-active riders excluded
Analysis sample size
Intervention group 913 (40)
Control group 500 (90)
Interventions No formal training vs the non-mandatory Motorcycle Safety Foundationrsquos Motorcycle Rider Course comprising of
8 hours theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-road practice
Intervention integrity
Participants had all completed and passed the MRC
37Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4050
Mortimer 1988 (Continued)
Outcomes Self reported crashes and offences
Length of follow up 1 year
Notes Study conducted in Illinois USA
Osga 1980
Methods Cohort study
Participants Initial sample sizeIntervention group 1163 MSF MRC participants invited 818 (74) responded
Control group 329 motorcycle riders responded to recruitment strategies
Non-active riders excluded from analysis
Analysis sample size
Intervention group 595 (73)
Control group 329 (100)
Interventions No formal training vs the non-mandatory Motorcycle Safety Foundationrsquos Motorcycle Rider Course (MRC) com-
prising of 8 hours theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-road
practice
Intervention integrity
Participants had all successfully completed the MRC
Outcomes Self-reported crashes police recorded crash reports
Length of follow up 6 months - 3 12 years
Notes Study conducted in South Dakota USA
Perrino 2002
Methods Cohort study
Participants Initial sample size
Intervention group 2144 (unclear) MRC graduates (motorcycle licence applicants)
Control group 2601 (unclear) regular licensing test graduates (motorcycle licence applicants)Analysis sample size
Intervention group 1070 (unclear)
Control group 867 (unclear)
Interventions No formal training vs the mandatory (for those younger than 18 years) pre-licence Maryland Motorcycle Rider
Course (MRC) consisting of 8 hours theory and 12 hours practice
Intervention integrity
No information given
Outcomes Self-reported and police recorded crashes and offences
Length of follow up 1 year
38Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4150
Perrino 2002 (Continued)
Notes Study conducted in Maryland USA
Raymond 1979
Methods Cohort study
Participants Initial sample size
Intervention group
258 volunteers mainly (but not all) novice riders
Control group
1104 provisional licence applicants
Analysis sample size
Intervention group 190 (74)
Control group 627 (57)
Interventions No formal training vs the non-mandatory RACACU (Royal Automobile ClubAuto-Cycle Union) training scheme
The training program consisted of either 24 one-hour sessions or 13 two-hour sessions Twelve hours of the training
focused on theory about basic operational procedures traffic laws and highway code The practical session focused
on basic operational skills riding on a training area and riding in traffic with an instructor A proficiency test was
administered at the end of the course
Intervention integrity
No information given
Outcomes Self-reported crashes (including minor non-injury non-damage and off-road crashes)
Length of follow up 1 year
Notes Study conducted in the UK
Satten 1980
Methods Cohort study
Participants Initial sample size
Intervention group interviewers kept interviewing until they had 100 riders selected from a population of 891
enrollees in the MSF MRC course (unknown how many were asked and how many refused to participate)Control group unknown number of registered motorcycle riders
Non-active riders excluded from analysis
Analysis sample size
Intervention group 71
Control group 100
Interventions No formal training vs the non-mandatory Motorcycle Safety Foundationrsquos (MSF) Motorcycle Rider Course (MRC)
comprising of 8 hours theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-
road practice
Intervention integrity
Participants had all successfully completed the MRC
39Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4250
Satten 1980 (Continued)
Outcomes Self-reported crashes injuries and offences
Length of follow up 3 years
Notes Study conducted in Illinois USA
Savolainen 2007
Methods Cohort study
Participants Initial sample size
Intervention group 4000 riders that had conducted the MSF MRC course (passed or failed)
Control group 4000 motorcycle owners or permit or endorsement holders from the Indiana Bureau of Motor
Vehicles database
Analysis sample size
Intervention group 739 (18)
Control group 588 (15)
Interventions Intervention integrity
No information given
Outcomes Probability of crash involvement
Length of follow up up to 5 years
Notes Study conducted in Indiana USA
Swaddiwudhipong 1998
Methods Randomised trial
Participants Initial sample size
Intervention group
all 16 villages of 3 sub-districts
(participants were practising motorcyclists licensed and unlicensed)
Control group
18 villages of 3 other sub-districtsAnalysis sample size
Intervention group 3 districts (100)
Control group 3 districts (100)
Interventions No rider education vs non-mandatory community based rider education consisting of theory on motorcycle injury
risk helmet use traffic laws vehicle regulations and traffic signs and a written and skill test leading to a licence
Intervention integrity
No information given
Outcomes Injuries fatalities
Length of follow up 2 years
40Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4350
Swaddiwudhipong 1998 (Continued)
Notes Study conducted in Tak Province rural Thailand
Waller 1992
Methods Cohort study
Participants Initial sample size
Intervention group 349 riders who succesfully completed the Department of Public Safety motorcycle operator
training in 1988
Control group 349 motorcyclists who never completed the DPS motorcycle operator training
Analysis sample size
Intervention group 349
Control group 349
Interventions No DPS motorcycle operator training vs non-mandatory DPS motorcycle operator training The DPS training is
the Motorcycle Safety Foundationrsquos Motorcycle Rider Course comprising eight hours of theory (lectures discussions
films and multiple choice tests) a textbook and 12 hours off-road practice
Intervention integrity
All 349 riders succesfully completed the rider training
Outcomes Police recorded crashes
Length of follow up 3 years
Notes Study conducted in
Characteristics of excluded studies [ordered by study ID]
Study Reason for exclusion
Anderson 1980a Study type review and process evaluation of a rider course
Aupetit 2008 Outcome proposal of a rider training curriculum that combines traditional training and a motorcycle riding
simulator
Awane 1999 Study type review on use of simulators in rider and driver training
Baldi 2005 Intervention not rider training but intervention integrity of rider training availabilty of rider education
implementation of graduated licensing system etc on a state-wide level
Batchler 1988 Study type description of a training course
Blanchard 2006 Study type not a study assessing the association between rider training and outcomes
Bonnett 2005 Outcome qualitative data regarding satisfaction with the training programme
41Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4450
(Continued)
Braver 2007 Outcome licensure rates
Clayton 1980 Outcome knowledge skills attitude
Council 1978 Outcome crash reduction in ANY vehicle not motorcycles exclusively
Crowther 2005 Study type qualitative study
Daniello 2009 Study type review of the effectiveness of motorcycle training and licensing
Dewanti 2007 Intervention emotional characteristics behaviour (ie not motocycle rider training)
Du Plooy 1990 Study type review of types of rider training courses in South Africa
Engel 1989 Study type uncontrolled before-after study
Goldenbeld 2002 Outcome skills
Goldenbeld 2004 Outcome knowledge and skills
Haworth 2000 Intervention balance between attitudinal andvehicle control skills acrosscurriculla of motorcycle rider training
courses
Haworth 2005 Study type review
Haworth 2006 Study type cross-sectional survey
Hill 1983 Study type review and description of training system
Imatake 1980 Study type course descriptionreview
Ingebritsen 1990 Intervention risk factors for motorcycle crashes
Jain 2009 Study type cross-sectional descriptive study of motorcycle crashes over a 5 year period
Jonah 1979 Study type cross-sectional survey Outcome scores on motorcycle operator skill test
Jonah 1981 Intervention evaluation of a motorcycle rider skill test not motorcycle rider training
Kadar 1990 Study type review of components of rider training courses
Kelsey 1986 Intervention evaluation of a motorcycle rider skill test not motorcycle rider training
Koch 1980 Study type reviewcourse description
Koch 1990 Study type description of course content
42Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4550
(Continued)
Labbett 2005 Study type essay about possible explanations for motorcycle deaths
Law 2005 Intervention targeted motorcycle safety programme not specifically rider training
Lowes 1990 Study type review and descriptions of rider training
Mahon 1976 Study type review of motorcycle rider training methods
Maxwell 1985 Outcomesrecommendations forprogramme operatorsand otherson howthe motorcycle trainingprogramme
works and how it can be improved
Mayhew 1996 Study type review
McKnight 1987 Study type evaluation of methodological and administrative aspects of a motorcycle operator skill test
McPherson 1976 Outcome knowledge and skills
Mohan 2004 Study type review of evidence of interventions to reduce road traffic injuries
Morgan 1980 Study type reviewcourse description
Motoki 1990 Study type proposal and development of rider training curriculum
Nairn 1992 Study type literature review
Newland 1980 Study type description of development of motorcycle safety course
Newland 1982 Study type descriptionreview of a rider training course
Newland 1990 Study type uncontrolled BeforeAfter study
Ormston 2003 Outcome changes in attitude and self reported behaviour
Osga 1979 Outcome knowledge attitudes and skills
Panou 2007 Intervention new simulation tool to teach driverrider
Peterson 1990 Study factor demographic details of course participants completing the course vs those not completing the
course
Prem 1984 Study type experimental
Study factor examines association between skills and performance in tests
Radin Umar 2006 Intervention motorcycle safety programme not specifically rider training
Rockwell 1990 Study type methodologically flawed study type
43Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4650
(Continued)
Rothe 1987 Study type qualitative study of rider training in general
Rowden 2007 Study type qualitative study
RSC Victoria 1993 Study type review of motorcycle rider training and education in Victoria
Russam 1979 Study type review of training programmes in Japan
Saffron 1981 Study type literature review
Schulz 1990 Study factor predictors of motorcycle crashes
Sexton 2004 Study type cross-sectional analytic analysis
Shephard 1986 Study factor reviewdescription of programme
Simpson 1990 Study type review of motorcycle rider trainingeducation
Sokolov 1990 Study factor review of motorcycle riding courses in USSR
Steffens 1988 Outcome self-reported behaviour experiences and researcher observed riding behaviour
Sudlow 2003 Study type review of motorcycle training in the UK
Tasmania 1995 Study type cross-sectional descriptiveprocess evaluation
Tetard 1985 Study type experimental
Outcome detection of indices
Thackray 1980 Outcome instructional effectiveness user acceptance administrative feasibility
Thompson 1994 Study type cross-sectional survey
Outcome attitudes and knowledge
Tobler 1990 Study type review of course curricula
Transport 1989 Intervention syllabus for rider training
Tuan 2007 Study type cross-sectional study (survey) which does not assess the effects of an intervention but examines
motocyclistsrsquo behaviour
Victor 1980 Study type review
Waterer 1990 Study type review of UK rider training issues
Weller 1989 Study type review of studies evaluating motorcycle safety and education measures
44Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4750
(Continued)
Wells 1981 Study type description of different methods of training
Winn 1986 Study type review of rsquohistoryrsquo of rider training
Wisher 1988 Study type cross-sectional survey
Wood 1987 Study type Interrupted Time Series but did not meet the criteria for an ITS as published by the Cochrane
Effective Practice and Organisation of Care Group
Yamazaki 1990 Study type cross-sectional survey
Outcome riding skills skills retention and attitude
Characteristics of ongoing studies [ordered by study ID]
Burgess 2010
Trial name or title UKs National Ride program
Methods Cohort study
Participants Motorcycle offenders
Interventions Police led diversion program
Outcomes Self reported attitudinal and behavioural measures
Starting date
Contact information
Notes Expected to be completed in early 2010
45Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4850
D A T A A N D A N A L Y S E S
This review has no analyses
A P P E N D I C E S
Appendix 1 Search strategy
We searched the following electronic databases
bull Cochrane Injuries Group Specialised Register (searched 22 Sept 2008)
bull CENTRAL (The Cochrane Library 2008 Issue 3)
bull TRANSPORT 1988 to 200706 (latest date available)
bull ATRI Australian Transport Index (to Sept 2008)
bull MEDLINE (Ovid SP) 1950 to Sept 2008
bull EMBASE (Ovid SP) 1980 to Sept 2008
bull CINAHL (EBSCO) 1982 to Sept 2008
bull WHOLIS (World Health Organization Library Information System) (to Sept 2008)
bull PsycINFO (Ovid SP) 1806 to Sept 2008
bull LILACS (BIREME) (Latin American and Caribbean Health Sciences) (1982 to Sept 2008)
bull ISI Web of Science Social Sciences Citation Index (SSCI) 1970 to Sept 2008
bull ERIC (searched 25 Sept 2008)
bull ZETOC (searched 25 Sept 2008)
bull SIGLE (searched 25 Sept 2008)
All of the databases were searched using the following text-word search strategy The strategy was adapted using either rsquorsquo or rsquo$rsquo as atruncation symbol
1 motor-cycl or motorcycl or motor-bik or motorbik or motor-scooter or motorscooter or scooter or moped or motocycl
or motocicl or moto
2 train or educat or instruct or program or course or qualif or learn or improv
3 1 and 2
H I S T O R Y
Protocol first published Issue 2 2005
Review first published Issue 10 2010
46Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4950
C O N T R I B U T I O N S O F A U T H O R S
Katina Kardamanidis was the first author of the protocol conducted the literature search extracted full studies assessed the studies
according to inclusion criteria assessed quality of the studies compared findings with the second author discussed disagreements with
the third author wrote the drafts and the final review
Rebecca Ivers was the second author of the protocol assessed the studies according to inclusion criteria assessed quality of the studies
compared findings with the first author edited the final draft and provided overall guidance for the process
Alexandra Martiniuk assessed quality of the studies compared findings with the first author discussed disagreements with the third
author and helped write the first draft and the final review
Mark Stevenson contributed to the protocol reviewed methods of studies and contributed to the final draft
Katrina Thistlethwaite conducted a second literature search extracted full studies and contributed to the final draft
D E C L A R A T I O N S O F I N T E R E S T
None known
S O U R C E S O F S U P P O R T
Internal sources
bull No sources of support supplied
External sources
bull National Health and Medical Research Council of Australia Australia
Salary support for Martiniuk Ivers and Stevenson
bull University of Sydney Sesqui Research Grant Australia
Direct research costs and salary support for Kardamanidis
D I F F E R E N C E S B E T W E E N P R O T O C O L A N D R E V I E W
There are no differences between the protocol and the review
I N D E X T E R M S
Medical Subject Headings (MeSH)
lowastMotorcycles Accident Prevention [lowastmethods] Accidents Traffic [lowastprevention amp control] Licensure Program Evaluation Wounds
and Injuries [mortality lowastprevention amp control]
47Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 5050
MeSH check words
Humans
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3050
Thackray 1980 published data only
Thackray RM Prescott JC Field test of the motorcycle ridercourse International Motorcycle Safety Conference Washington
DC USA Motorcycle Safety Foundation 1980195ndash7
Thompson 1994 published data only
Thompson M Evaluation of compulsory basic training for
motorcyclists Transport Research Laboratory Crowthorne
Berkshire UK 1994 [ Report 63]
Tobler 1990 published data only
Tobler U Individual driving instruction for motorcycle driving
learners the Swiss experiment of motorcycle driving school The
Human Element International Motorcycle Safety Conference
Orlando Florida USA Motorcycle Safety Foundation 1990
Transport 1989 published data only
Transport Communications Road Crash Statistics Australia Bureauof Transport and Communications Economics 198914ndash20
Tuan 2007 published data only
Tuan VA Shimizu T Towards development and evaluation of the
motorcycle drivers re-education program in Vietnam modeling of
motorcycle driverrsquos undesired behaviors 11th World Conference
on Transport Research 200725
Victor 1980 published data only
Victor DJ Traffic safety education for motorcyclists International
Motorcycle Safety Conference Washington DC USA Motorcycle
Safety Foundation 1980221ndash8
Waterer 1990 published data only
Waterer TH Compulsory training for motorcyclists from liberty
to compulsion The Human Element International Motorcycle
Safety Conference Orlando Florida USA Motorcycle Safety
Foundation 19904-49 to 4-57
Weller 1989 published data only
Weller RB Chandler EW Motorcycle safety and motorcycle
education past research and survey results Journal of Public Policy
amp Marketing 1989893ndash108
Wells 1981 published data only
Wells P Observations of some RACACU motorcycle training
schemes Transport and Road Research Laboratory Crowthorne
Berkshire UK 1981
Winn 1986 published data only
Winn G Two important trends in motorcycle safety regulations
rider education and conspicuity improvement American
Motorcyclist Association 1986147ndash52
Wisher 1988 published data only
Wisher G Reid J Motorcycle accidents in Norfolk a three-year study of motorcycle training and the Part 1 test Norfolk UK Road Safety
Division Highways Department Norfolk County Council 1988
Wood 1987 published data only
Wood HT Bowen R Evaluation of the revised learner permit scheme
July 1983 to December 1985 Victoria Australia Road Traffic
Authority 19871ndash27 [ ISBN 0 7306 0247 8]
Yamazaki 1990 published data only
Yamazaki S Nakanishi M Survey on effect of Honda rider
education The Human Element International Motorcycle Safety
Conference Orlando Florida USA 199012-167 to 12-187
References to ongoing studies
Burgess 2010 unpublished data only
Burgess C Broughton P Fylan F Stradling Interim evaluation of
the UKs National Ride program In L Dorn editor(s) Driver
behaviour and Training Vol IV Aldershot Ashgate Publishing
Limited 2010
Additional references
Anderson 1980
Anderson J Ford JL Peck RC Improved motorcyclist licensing and
testing project Volume I and II National Highway Traffic Safety
Administration US department of Transportation Washington
DC 1980
Bjornstig 1985
Bjornstig U Bylund P Lekander T Brorsson B Motorcycle fatalitiesin Sweden Acta Chirurgica Scandinavica 1985151(7)577ndash81
Clayton 1990
Clayton A Sudlow DE An evaluation of the effectiveness of a one-
day training program for learner motorcyclists The Human
Element International Motorcycle Safety Conference Orlando
Florida USA 199013-105 to 13-111
Cochrane Collaboration 2008
Higgins JP Green S Cochrane Handbook for Systematic Reviews of
Interventions 500 The Cochrane Collaboration 2008
Elliot 2003
Elliott MA Baughan CJT Broughton J Grayson G Knowles J et
alMotorcycle Safety a Scoping Study TRL [ Report No
0968ndash4107]Fleiss 2004
Fleiss JL Levin B Paik MC Statistical methods for rates and proportions 3rd Edition John Wiley amp Sons Inc 2004 [ ISBN
978ndash0ndash471ndash52629ndash2 ]
FORS 1999
FORS Federal Office of Road Safety Road risk for sober licensed
motorcyclists Vol Monograph 27 Federal Office of Road Safety
19991ndash2
Haworth 2005
Haworth N Mulvihill C Review of motorcycle licensing and training
Victoria Australia Monash University Accident Research Centre
20051ndash83 [ Report No 240]
Huwiler-Muumlntener 2002Huwiler-Muumlntener K Juumlni P Junker C Egger M Quality of
Reporting of Randomized Trials as a Measure of Methodologic
Quality JAMA 20022872801ndash4
Johnston 2008
Johnston P Brooks C Savage H Fatal and serious road crashes
involving motorcyclists Road Safety Research and Analysis Report
ATSB Monograph 20 Canberra Department of Infrastructure
Transport Regional Development and Local Government 0205
081ndash26
Juumlni 2001
Juumlni P Altman DG Egger M Systematic reviews in health care
Assessing the quality of controlled clinical trials BMJ 2001323
42ndash6
28Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3150
Kopits 2003
Kopits E Cropper M Traffic fatalities and economic growth TheWorld Bank Development Research Group Infrastructure and
Environment Policy research working paper 3035 2003
Leung 1983
Leung P Chau BTW Motorcycle accidents in Hong Kong pattern
of injury treatment causative factors and socio-economic
implications Injury 198315(1)23ndash9
Liu 2004
Liu B Ivers R Norton R Blows S Lo SK Helmets for preventing
injury in motorcycle riders Cochrane Database of Systematic Reviews 2004 Issue 2 [DOI 10100214651858CD004333pub3]
Mohan 2002
Mohan D Road safety in less-motorized environments future
concerns International Journal of Epidemiology 200231(3)527ndash32Mortimer 1982
Mortimer RG An evaluation of the motorcycle rider course
Proceedings of the 20th conference of the American Association for
Automotive Medicine Arlington Heights IL USA American
Association for Automotive Medicine 198277ndash96
Mullin 1997
Mullin BT Risk factors for motorcycle injury the role of age
gender experience training and alcohol PhD thesis Auckland
New Zealand University of Auckland 1997
Noordzij 2001
Noordzij PC Forke E Brendicke R Chinn BP Integration of needs
of moped and motorcycle riders into safety measures review and
statistical analysis in the framework of the European research
project PROMISING SWOV Institute for Road Safety Research
Leidschendam The Netherlands 2001 [ Report No Dndash2001ndash5]
Osga 1979
Osga G Ellingstad VS An evaluation of the effectiveness of the 1978 South Dakota Motorcycle Riders Course Program South Dakota
University of South Dakota Vermillion Human Factors Lab 1979
Oxford English Dictionary Online
httpoxforddictionariescom Oxford English Dictionary Online
Oxford University Press 2010
Peden 2004
Peden M Scurfield R Sleet D World Report on Road TrafficInjury Prevention Geneva World Health Organization Available
at httpwwwwhointviolence˙injury˙preventionpublications
road˙trafficworld˙reporten 2004Chapter 3
Radin Umar 1995
Radin Umar R Macakay G Hills B Preliminary analysis of
motorcycle accidents Short-term impacts of the running
headlights campaign and regulations in Malaysia J Traffic Med 199523(1)17ndash27
Tsai 1995
Tsai YJ Wang JD et alCase-control study of the effectiveness of
different types of helmets for the prevention of head injuries among
motorcycle riders in Taipei Taiwan American Journal of
Epidemiology 1995142(9)974ndash81
Vis 1995
Vis AA In-depth study of the hazards of motorcycling [De
onveiligheid van motorrijden nader bekeken] SWOV Institute for
Road Safety Research Leidschendam the Netherlands 1995 [
Report No Rndash95ndash69]
Weaver 1990
Weaver EA Motorcycle safety education in the United States
Motorcycle Safety Foundation Orlando Florida USA 1990
Winn 1987
Winn G Two important trends in motorcycle safety regulations
rider education and conspicuity improvement No more
information available Detroit MI US Society of Automotive
Engineers 1987147ndash52
Yuan 2000
Yuan W The effectiveness of the rsquoride-brightrsquo legislation for
motorcycles in Singapore Accident Analysis amp Prevention 200032
(4)559ndash63
Zambon 2006
Zambon F Hasselberg M Factors affecting the severity of injuries
among young motorcyclists a Swedish nationwide cohort study
Traffic Injury Prevention 20067(2)143ndash9lowast Indicates the major publication for the study
29Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3250
C H A R A C T E R I S T I C S O F S T U D I E S
Characteristics of included studies [ordered by study ID]
Adams 1985
Methods Non-randomised trial
Participants Initial sample size
Total of 162 Australia Post telegram delivery riders
Not clear how many in intervention and control groups
Analysis sample size
Intervention group 63Control group 35
Total 98 (61)
Interventions No formal training vs voluntary rider training course of 12 hours over 2 consecutive days (theory 4 hours practice
8 hours)
Intervention integrity
No information given
Outcomes Australia Post recorded crashes
Length of follow up 1 year
Notes Study conducted in NSW Australia
Anderson 1980
Methods Randomised trial
Participants Initial sample size
Total 41188 motorcycle licence applicants
Analysis sample size
Intervention group with skills training 13160
Intervention group without skills training 12634
Control group 15080
Total 40874 (99)
Interventions Standard licensing skill test vs the Motorcycle Operators Skill Test (MOST) with or without a training component
MOST is not to be confused with rider training because it is a licensing test The training component was for
participants who failed the skill test on the first attempt was of 3 hours duration and consisted of instructions about
the skills needed to pass the skill test The training consisted of 30 min theory and 2 frac12 hours practice on the
motorcycle
Intervention integrity
33 in MOST with skills training 31 in MOST without skills training and 43 in control group obtained their
motorcycle licence in the end
Outcomes Police recorded crashes and offences
Length of follow up 3 years
30Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3350
Anderson 1980 (Continued)
Notes Study conducted in California USA
Billheimer 1998
Methods Cohort study
Participants Initial sample size
Intervention group Unclear numbers of CMSP graduates and ERC graduates
Control group Over 16000 untrained motorcyclists
Analysis sample size
1139 matched CMSP-pairs and 1182 ERC-pairs
Interventions No formal training vs the mandatory (for those younger than 18 years) pre-licence 16-hours basic California
Motorcyclist Safety Program (CMSP) and the 8-hour Experienced Rider Course (ERC)
Intervention integrity
No information given
Outcomes Police recorded crashes and offences
Length of follow up 2 years
Notes Study conducted in California USA
Cooper 1988
Methods Cohort study
Participants Initial sample size
Intervention group 418 graduates and 57 who failed
Control group 402
Analysis sample size
Intervention group unclear
Control group unclear
Total 863 (984)
Interventions No formaltrainingvs non-mandatory pre-licence rider trainingThe content of thetrainingwas notfurther explainedIntervention integrity
Participants had all completed the rider training
Outcomes Police recorded crashes and offences
Length of follow up 7 years
Notes Study conducted in British Columbia Canada
31Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3450
Davis 1997
Methods Retrospective cohort study
Participants Initial sample size
Intervention group 9320 graduates
Control group unclear
Analysis sample size
Intervention group 9320 graduates
Control group unclear
Interventions No formal training vs the non-mandatory Connecticut Motorcycle Operator Training Program The content of
training was not further explained
Intervention integrity Participants had all completed the rider training
Outcomes Police recorded crashes severity of crashes
Length of follow up 1 year
Notes Study conducted in Connecticut USA
Hall 1988
Methods Randomised trial
Participants Initial sample sizeIntervention group A 6248
Intervention group B 6616
Intervention group C 6996
Control group 6604
Total 26464 motorcycle licence applicants
Analysis sample size
Intervention group A 6012 (96)
Intervention group B 6428 (97)
Intervention group C 6778 (97)
Control group 6502 (98)
Interventions Standard licensing skill test vs the Motorcycle Operator Skill Test (MOST II) without training (A) MOST II with
3 hours training (B) or MOST II with 20 hours training (C) The 3 hours training comprised instructions aboutthe skills needed to pass the skill test (30 min theory and 2 frac12 hours practice on the motorcycle) The 20 hours
training consisted of the Motorcycle Safety Foundationrsquos Motorcycle Rider Course (MRC) comprising of 8 hours
theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-road practice
Intervention integrity
25 and 26 obtained their motorcycle licence in the lsquoMOST 3-hour trainingrsquo and lsquoMOST 20-hour trainingrsquo
intervention arm respectively In the lsquoMOST testrsquo intervention arm 29 obtained their licence and in the control
group 35
Outcomes Police recorded crashes and offences
Length of follow up 3 years
32Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3550
Hall 1988 (Continued)
Notes Study conducted in New York USA
Haworth 1997
Methods Population based case-control study
Participants Cases Motorcycle riders or pillion riders involved in motorcycle crash (on road) Recruited from hospital and police
records Not clear whether all cases (especially fatalities) were included
Community controls Multiple matched sample of motorcycle rider or pillion who passed the crash site at the same
time of day and day of week that the crash occurred
Initial sample size
Cases 336
Controls 1195
Analysis sample size
Cases 234 (70)
Controls 561 (47)
Interventions Basic handling skills course beginner intermediate or advanced course measured through self-report not validated
Outcomes Injuries with an Injury Severity Score of 5 or greater or admission for at least four hours and fatalities from a
motorcycle crash (from hospital and police records)
Length of follow up two years or more prior to the interview
Notes Study conducted in Victoria Australia
Hurt 1981
Methods Population based case-control study
Participants Cases Motorcycle riders who had a motorcycle crash that was reported to police (traffic accident reports) Not all
cases included
Community controls Multiple matched sample of motorcycle riders who passed the crash site at the same time of
day and day of week as the crash occurred but 2 years later
Initial sample size
Cases 4500 notifications of crashes to police (only 50 received in time research team able to respond to 1126 for
unreported reasons data collection completed for 900)
Controls Number approached unreported 2310 consented and interviewed
Analysis sample size
Cases 808 (808900 data on rider training) - 18 of initial sample
Controls 670 (6702310 data on rider training) - lt 29 of initial sample
Interventions Training experience self taught friends-family school-club formal-professional others measured through self-
report validated
Outcomes Injuries and fatalities from a motorcycle crash (from hospital records and coronerrsquos office)
33Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3650
Hurt 1981 (Continued)
Notes Study conducted in Los Angeles USA
Jonah 1982
Methods Cohort study
Participants Initial sample size
Intervention group
2310 licensed MTP graduates who are active riders
Control group
2029 licensed motorcyclists who are active riders
Non-licensed and non-active riders excluded
Analysis sample size
Intervention group
811 interviews (35) 533 driver records (23)
Control group 1080 interviews (71) 905 driver records (45)
Interventions No training vs the non-mandatory Canada Safety Council Motorcycle Training Program (MTP) comprising of 4
hours theory and 16 hours practise (6 hrs on-road) over 2 weekends
Intervention integrity
Participants had all successfully completed the MTP
Outcomes Self-reported and police reported crashes and offencesLength of follow up 4 years
Notes Study conducted in Ontario Canada
Kloeden 1994
Methods Cohort study
Participants Initial sample size
Intervention group 2533 licence applicants
Control group Unclear
Sample of riders obtaining probationary licenceIntervention group 1625 (64)
Control group 5015 (unclear)
Non-active riders excluded from control group (~66)
Analysis sample size
Intervention group 631 (39)
Control group 318 (6)
Interventions No formal training vs the mandatory pre-licence Ridersafe course consisting of two levelsthe first is a training course
in the basics of motorcycle riding where a pass entitles the rider to obtain a learnerrsquos permit The second is a skill
test where a pass entitles the rider to obtain a probationary motorcycle licence The training course consists of two 4
hour training sessions on separate days involving instruction videos and actual riding on motorcycles The skill test
consists of a 4 hour preparation course of 74 minutes of lecturediscussion 38 minutes of video and 75 minutes of
34Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3750
Kloeden 1994 (Continued)
on-course riding which is the testing phase
Intervention integrity
Unclear what percentage in the Ridersafe group and the control group obtained their probationary licence halfway
and their full motorcycle licence in the end
Outcomes Police reported crashes
Length of follow up 1 year
Notes Study conducted in SA Australia
Kraus 1975
Methods Population based case-control study
Participants Cases All motorcycle riders with a confirmed medically treated motorcycle injury (including fatalities) from all
hospitals in study base and peripheral area hospitals and all deaths due to motorcycle crashes in the study base All
serious injuries (leading to hospitalisation) and fatalities likely to be included
Community controls A random sample of registered motorcycle owners in the same temporary and geographical
study base as the cases arose from
Initial sample size Cases 1273
Controls 738
Analysis sample size Cases 741 (58)
Controls 434 (59)
Interventions Motorcycle rider training (not specified) measured through self-report
Outcomes Injuries leading to hospitalisation fatalities
Length of follow up not specified
Notes Study conducted in Sacramento county USA
Leung 1987
Methods Non-randomised trial
Participants Initial sample sizeIntervention group unknown number of MRC graduates
Control group unknown number of licensed motorcyclists
Analysis sample size
Intervention group 2941
Control group 43094
Interventions No formal training vs non-mandatory Wisconsin motorcycle rider course a 22 hours off-street programme that
teaches the basic rules of safe motorcycling
Intervention integrity
50 in the MRC group and 100 in the control group obtained their motorcycle licence in the end
35Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3850
Leung 1987 (Continued)
Outcomes Police recorded crashes and offences
Length of follow up up to 3 years
Notes Study conducted in Wisconsin USA
MAIDS 2004
Methods Population based case-control study
Participants Cases Motorcycle riders who had a motorcycle crash Unclear whether all cases included
Community controls Motorcycle riders identified at petrol stations Unclear whether matched to cases
Initial sample size
Cases not reported
Controls not reported
Analysis sample size
Cases 921 (Italy 200 Netherlands 200 Spain 121 Germany 250 France 150)
Controls 923 (Italy 200 Netherlands 200 Spain123 Germany 250 France 150)
Interventions Pre-license rider training or additional training (non-compulsory)
Outcomes Crashes
Notes Study in five sampling areas located in five European countries (France Germany Netherlands Spain and Italy)
McDavid 1989
Methods Cohort study
Participants Sample size after first match
Intervention group 346 male training graduates
Control group 346 male motorcycle riders licensed in same year
Sample size after second match
Intervention group 139 (40)
Control group 139 (40)
Analysis sample size
Intervention group 139 (40)
Control group 139 (40)
Interventions No formal training vs the non-mandatory pre-licence British Columbia Safety Councilrsquos 37-hour motorcycle safety
programme
Intervention integrity
Participants had all completed and passed the training
Outcomes Police reported crashes injurious crashes
Length of follow up 1-6 years
36Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3950
McDavid 1989 (Continued)
Notes Study conducted in British Columbia Canada
Mortimer 1984
Methods Cohort study
Participants Initial sample size
Intervention group 608 MRC graduates
Control group 360 motorcycle riders
Response rates
Intervention group 360 (59)
Control group 324 (90)
Non-active riders excluded
Analysis sample size
Intervention group 213 (35)
Control group 303 (84)
Interventions No formal training vs the non-mandatory Motorcycle Safety Foundationrsquos Motorcycle Rider Course comprising of
8 hours theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-road practice
Intervention integrity
Participants had all completed and passed the MRC
Outcomes Self-reported crashes and offencesLength of follow up 1 year
Notes Study conducted in Illinois USA
Mortimer 1988
Methods Cohort study
Participants Initial sample size
Intervention group 2307 MRC graduates
Control group 556 active licensed motorcycle riders
Response ratesIntervention group 1298 (56)
Control group 500 (90)
Non-active riders excluded
Analysis sample size
Intervention group 913 (40)
Control group 500 (90)
Interventions No formal training vs the non-mandatory Motorcycle Safety Foundationrsquos Motorcycle Rider Course comprising of
8 hours theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-road practice
Intervention integrity
Participants had all completed and passed the MRC
37Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4050
Mortimer 1988 (Continued)
Outcomes Self reported crashes and offences
Length of follow up 1 year
Notes Study conducted in Illinois USA
Osga 1980
Methods Cohort study
Participants Initial sample sizeIntervention group 1163 MSF MRC participants invited 818 (74) responded
Control group 329 motorcycle riders responded to recruitment strategies
Non-active riders excluded from analysis
Analysis sample size
Intervention group 595 (73)
Control group 329 (100)
Interventions No formal training vs the non-mandatory Motorcycle Safety Foundationrsquos Motorcycle Rider Course (MRC) com-
prising of 8 hours theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-road
practice
Intervention integrity
Participants had all successfully completed the MRC
Outcomes Self-reported crashes police recorded crash reports
Length of follow up 6 months - 3 12 years
Notes Study conducted in South Dakota USA
Perrino 2002
Methods Cohort study
Participants Initial sample size
Intervention group 2144 (unclear) MRC graduates (motorcycle licence applicants)
Control group 2601 (unclear) regular licensing test graduates (motorcycle licence applicants)Analysis sample size
Intervention group 1070 (unclear)
Control group 867 (unclear)
Interventions No formal training vs the mandatory (for those younger than 18 years) pre-licence Maryland Motorcycle Rider
Course (MRC) consisting of 8 hours theory and 12 hours practice
Intervention integrity
No information given
Outcomes Self-reported and police recorded crashes and offences
Length of follow up 1 year
38Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4150
Perrino 2002 (Continued)
Notes Study conducted in Maryland USA
Raymond 1979
Methods Cohort study
Participants Initial sample size
Intervention group
258 volunteers mainly (but not all) novice riders
Control group
1104 provisional licence applicants
Analysis sample size
Intervention group 190 (74)
Control group 627 (57)
Interventions No formal training vs the non-mandatory RACACU (Royal Automobile ClubAuto-Cycle Union) training scheme
The training program consisted of either 24 one-hour sessions or 13 two-hour sessions Twelve hours of the training
focused on theory about basic operational procedures traffic laws and highway code The practical session focused
on basic operational skills riding on a training area and riding in traffic with an instructor A proficiency test was
administered at the end of the course
Intervention integrity
No information given
Outcomes Self-reported crashes (including minor non-injury non-damage and off-road crashes)
Length of follow up 1 year
Notes Study conducted in the UK
Satten 1980
Methods Cohort study
Participants Initial sample size
Intervention group interviewers kept interviewing until they had 100 riders selected from a population of 891
enrollees in the MSF MRC course (unknown how many were asked and how many refused to participate)Control group unknown number of registered motorcycle riders
Non-active riders excluded from analysis
Analysis sample size
Intervention group 71
Control group 100
Interventions No formal training vs the non-mandatory Motorcycle Safety Foundationrsquos (MSF) Motorcycle Rider Course (MRC)
comprising of 8 hours theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-
road practice
Intervention integrity
Participants had all successfully completed the MRC
39Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4250
Satten 1980 (Continued)
Outcomes Self-reported crashes injuries and offences
Length of follow up 3 years
Notes Study conducted in Illinois USA
Savolainen 2007
Methods Cohort study
Participants Initial sample size
Intervention group 4000 riders that had conducted the MSF MRC course (passed or failed)
Control group 4000 motorcycle owners or permit or endorsement holders from the Indiana Bureau of Motor
Vehicles database
Analysis sample size
Intervention group 739 (18)
Control group 588 (15)
Interventions Intervention integrity
No information given
Outcomes Probability of crash involvement
Length of follow up up to 5 years
Notes Study conducted in Indiana USA
Swaddiwudhipong 1998
Methods Randomised trial
Participants Initial sample size
Intervention group
all 16 villages of 3 sub-districts
(participants were practising motorcyclists licensed and unlicensed)
Control group
18 villages of 3 other sub-districtsAnalysis sample size
Intervention group 3 districts (100)
Control group 3 districts (100)
Interventions No rider education vs non-mandatory community based rider education consisting of theory on motorcycle injury
risk helmet use traffic laws vehicle regulations and traffic signs and a written and skill test leading to a licence
Intervention integrity
No information given
Outcomes Injuries fatalities
Length of follow up 2 years
40Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4350
Swaddiwudhipong 1998 (Continued)
Notes Study conducted in Tak Province rural Thailand
Waller 1992
Methods Cohort study
Participants Initial sample size
Intervention group 349 riders who succesfully completed the Department of Public Safety motorcycle operator
training in 1988
Control group 349 motorcyclists who never completed the DPS motorcycle operator training
Analysis sample size
Intervention group 349
Control group 349
Interventions No DPS motorcycle operator training vs non-mandatory DPS motorcycle operator training The DPS training is
the Motorcycle Safety Foundationrsquos Motorcycle Rider Course comprising eight hours of theory (lectures discussions
films and multiple choice tests) a textbook and 12 hours off-road practice
Intervention integrity
All 349 riders succesfully completed the rider training
Outcomes Police recorded crashes
Length of follow up 3 years
Notes Study conducted in
Characteristics of excluded studies [ordered by study ID]
Study Reason for exclusion
Anderson 1980a Study type review and process evaluation of a rider course
Aupetit 2008 Outcome proposal of a rider training curriculum that combines traditional training and a motorcycle riding
simulator
Awane 1999 Study type review on use of simulators in rider and driver training
Baldi 2005 Intervention not rider training but intervention integrity of rider training availabilty of rider education
implementation of graduated licensing system etc on a state-wide level
Batchler 1988 Study type description of a training course
Blanchard 2006 Study type not a study assessing the association between rider training and outcomes
Bonnett 2005 Outcome qualitative data regarding satisfaction with the training programme
41Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4450
(Continued)
Braver 2007 Outcome licensure rates
Clayton 1980 Outcome knowledge skills attitude
Council 1978 Outcome crash reduction in ANY vehicle not motorcycles exclusively
Crowther 2005 Study type qualitative study
Daniello 2009 Study type review of the effectiveness of motorcycle training and licensing
Dewanti 2007 Intervention emotional characteristics behaviour (ie not motocycle rider training)
Du Plooy 1990 Study type review of types of rider training courses in South Africa
Engel 1989 Study type uncontrolled before-after study
Goldenbeld 2002 Outcome skills
Goldenbeld 2004 Outcome knowledge and skills
Haworth 2000 Intervention balance between attitudinal andvehicle control skills acrosscurriculla of motorcycle rider training
courses
Haworth 2005 Study type review
Haworth 2006 Study type cross-sectional survey
Hill 1983 Study type review and description of training system
Imatake 1980 Study type course descriptionreview
Ingebritsen 1990 Intervention risk factors for motorcycle crashes
Jain 2009 Study type cross-sectional descriptive study of motorcycle crashes over a 5 year period
Jonah 1979 Study type cross-sectional survey Outcome scores on motorcycle operator skill test
Jonah 1981 Intervention evaluation of a motorcycle rider skill test not motorcycle rider training
Kadar 1990 Study type review of components of rider training courses
Kelsey 1986 Intervention evaluation of a motorcycle rider skill test not motorcycle rider training
Koch 1980 Study type reviewcourse description
Koch 1990 Study type description of course content
42Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4550
(Continued)
Labbett 2005 Study type essay about possible explanations for motorcycle deaths
Law 2005 Intervention targeted motorcycle safety programme not specifically rider training
Lowes 1990 Study type review and descriptions of rider training
Mahon 1976 Study type review of motorcycle rider training methods
Maxwell 1985 Outcomesrecommendations forprogramme operatorsand otherson howthe motorcycle trainingprogramme
works and how it can be improved
Mayhew 1996 Study type review
McKnight 1987 Study type evaluation of methodological and administrative aspects of a motorcycle operator skill test
McPherson 1976 Outcome knowledge and skills
Mohan 2004 Study type review of evidence of interventions to reduce road traffic injuries
Morgan 1980 Study type reviewcourse description
Motoki 1990 Study type proposal and development of rider training curriculum
Nairn 1992 Study type literature review
Newland 1980 Study type description of development of motorcycle safety course
Newland 1982 Study type descriptionreview of a rider training course
Newland 1990 Study type uncontrolled BeforeAfter study
Ormston 2003 Outcome changes in attitude and self reported behaviour
Osga 1979 Outcome knowledge attitudes and skills
Panou 2007 Intervention new simulation tool to teach driverrider
Peterson 1990 Study factor demographic details of course participants completing the course vs those not completing the
course
Prem 1984 Study type experimental
Study factor examines association between skills and performance in tests
Radin Umar 2006 Intervention motorcycle safety programme not specifically rider training
Rockwell 1990 Study type methodologically flawed study type
43Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4650
(Continued)
Rothe 1987 Study type qualitative study of rider training in general
Rowden 2007 Study type qualitative study
RSC Victoria 1993 Study type review of motorcycle rider training and education in Victoria
Russam 1979 Study type review of training programmes in Japan
Saffron 1981 Study type literature review
Schulz 1990 Study factor predictors of motorcycle crashes
Sexton 2004 Study type cross-sectional analytic analysis
Shephard 1986 Study factor reviewdescription of programme
Simpson 1990 Study type review of motorcycle rider trainingeducation
Sokolov 1990 Study factor review of motorcycle riding courses in USSR
Steffens 1988 Outcome self-reported behaviour experiences and researcher observed riding behaviour
Sudlow 2003 Study type review of motorcycle training in the UK
Tasmania 1995 Study type cross-sectional descriptiveprocess evaluation
Tetard 1985 Study type experimental
Outcome detection of indices
Thackray 1980 Outcome instructional effectiveness user acceptance administrative feasibility
Thompson 1994 Study type cross-sectional survey
Outcome attitudes and knowledge
Tobler 1990 Study type review of course curricula
Transport 1989 Intervention syllabus for rider training
Tuan 2007 Study type cross-sectional study (survey) which does not assess the effects of an intervention but examines
motocyclistsrsquo behaviour
Victor 1980 Study type review
Waterer 1990 Study type review of UK rider training issues
Weller 1989 Study type review of studies evaluating motorcycle safety and education measures
44Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4750
(Continued)
Wells 1981 Study type description of different methods of training
Winn 1986 Study type review of rsquohistoryrsquo of rider training
Wisher 1988 Study type cross-sectional survey
Wood 1987 Study type Interrupted Time Series but did not meet the criteria for an ITS as published by the Cochrane
Effective Practice and Organisation of Care Group
Yamazaki 1990 Study type cross-sectional survey
Outcome riding skills skills retention and attitude
Characteristics of ongoing studies [ordered by study ID]
Burgess 2010
Trial name or title UKs National Ride program
Methods Cohort study
Participants Motorcycle offenders
Interventions Police led diversion program
Outcomes Self reported attitudinal and behavioural measures
Starting date
Contact information
Notes Expected to be completed in early 2010
45Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4850
D A T A A N D A N A L Y S E S
This review has no analyses
A P P E N D I C E S
Appendix 1 Search strategy
We searched the following electronic databases
bull Cochrane Injuries Group Specialised Register (searched 22 Sept 2008)
bull CENTRAL (The Cochrane Library 2008 Issue 3)
bull TRANSPORT 1988 to 200706 (latest date available)
bull ATRI Australian Transport Index (to Sept 2008)
bull MEDLINE (Ovid SP) 1950 to Sept 2008
bull EMBASE (Ovid SP) 1980 to Sept 2008
bull CINAHL (EBSCO) 1982 to Sept 2008
bull WHOLIS (World Health Organization Library Information System) (to Sept 2008)
bull PsycINFO (Ovid SP) 1806 to Sept 2008
bull LILACS (BIREME) (Latin American and Caribbean Health Sciences) (1982 to Sept 2008)
bull ISI Web of Science Social Sciences Citation Index (SSCI) 1970 to Sept 2008
bull ERIC (searched 25 Sept 2008)
bull ZETOC (searched 25 Sept 2008)
bull SIGLE (searched 25 Sept 2008)
All of the databases were searched using the following text-word search strategy The strategy was adapted using either rsquorsquo or rsquo$rsquo as atruncation symbol
1 motor-cycl or motorcycl or motor-bik or motorbik or motor-scooter or motorscooter or scooter or moped or motocycl
or motocicl or moto
2 train or educat or instruct or program or course or qualif or learn or improv
3 1 and 2
H I S T O R Y
Protocol first published Issue 2 2005
Review first published Issue 10 2010
46Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4950
C O N T R I B U T I O N S O F A U T H O R S
Katina Kardamanidis was the first author of the protocol conducted the literature search extracted full studies assessed the studies
according to inclusion criteria assessed quality of the studies compared findings with the second author discussed disagreements with
the third author wrote the drafts and the final review
Rebecca Ivers was the second author of the protocol assessed the studies according to inclusion criteria assessed quality of the studies
compared findings with the first author edited the final draft and provided overall guidance for the process
Alexandra Martiniuk assessed quality of the studies compared findings with the first author discussed disagreements with the third
author and helped write the first draft and the final review
Mark Stevenson contributed to the protocol reviewed methods of studies and contributed to the final draft
Katrina Thistlethwaite conducted a second literature search extracted full studies and contributed to the final draft
D E C L A R A T I O N S O F I N T E R E S T
None known
S O U R C E S O F S U P P O R T
Internal sources
bull No sources of support supplied
External sources
bull National Health and Medical Research Council of Australia Australia
Salary support for Martiniuk Ivers and Stevenson
bull University of Sydney Sesqui Research Grant Australia
Direct research costs and salary support for Kardamanidis
D I F F E R E N C E S B E T W E E N P R O T O C O L A N D R E V I E W
There are no differences between the protocol and the review
I N D E X T E R M S
Medical Subject Headings (MeSH)
lowastMotorcycles Accident Prevention [lowastmethods] Accidents Traffic [lowastprevention amp control] Licensure Program Evaluation Wounds
and Injuries [mortality lowastprevention amp control]
47Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 5050
MeSH check words
Humans
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3150
Kopits 2003
Kopits E Cropper M Traffic fatalities and economic growth TheWorld Bank Development Research Group Infrastructure and
Environment Policy research working paper 3035 2003
Leung 1983
Leung P Chau BTW Motorcycle accidents in Hong Kong pattern
of injury treatment causative factors and socio-economic
implications Injury 198315(1)23ndash9
Liu 2004
Liu B Ivers R Norton R Blows S Lo SK Helmets for preventing
injury in motorcycle riders Cochrane Database of Systematic Reviews 2004 Issue 2 [DOI 10100214651858CD004333pub3]
Mohan 2002
Mohan D Road safety in less-motorized environments future
concerns International Journal of Epidemiology 200231(3)527ndash32Mortimer 1982
Mortimer RG An evaluation of the motorcycle rider course
Proceedings of the 20th conference of the American Association for
Automotive Medicine Arlington Heights IL USA American
Association for Automotive Medicine 198277ndash96
Mullin 1997
Mullin BT Risk factors for motorcycle injury the role of age
gender experience training and alcohol PhD thesis Auckland
New Zealand University of Auckland 1997
Noordzij 2001
Noordzij PC Forke E Brendicke R Chinn BP Integration of needs
of moped and motorcycle riders into safety measures review and
statistical analysis in the framework of the European research
project PROMISING SWOV Institute for Road Safety Research
Leidschendam The Netherlands 2001 [ Report No Dndash2001ndash5]
Osga 1979
Osga G Ellingstad VS An evaluation of the effectiveness of the 1978 South Dakota Motorcycle Riders Course Program South Dakota
University of South Dakota Vermillion Human Factors Lab 1979
Oxford English Dictionary Online
httpoxforddictionariescom Oxford English Dictionary Online
Oxford University Press 2010
Peden 2004
Peden M Scurfield R Sleet D World Report on Road TrafficInjury Prevention Geneva World Health Organization Available
at httpwwwwhointviolence˙injury˙preventionpublications
road˙trafficworld˙reporten 2004Chapter 3
Radin Umar 1995
Radin Umar R Macakay G Hills B Preliminary analysis of
motorcycle accidents Short-term impacts of the running
headlights campaign and regulations in Malaysia J Traffic Med 199523(1)17ndash27
Tsai 1995
Tsai YJ Wang JD et alCase-control study of the effectiveness of
different types of helmets for the prevention of head injuries among
motorcycle riders in Taipei Taiwan American Journal of
Epidemiology 1995142(9)974ndash81
Vis 1995
Vis AA In-depth study of the hazards of motorcycling [De
onveiligheid van motorrijden nader bekeken] SWOV Institute for
Road Safety Research Leidschendam the Netherlands 1995 [
Report No Rndash95ndash69]
Weaver 1990
Weaver EA Motorcycle safety education in the United States
Motorcycle Safety Foundation Orlando Florida USA 1990
Winn 1987
Winn G Two important trends in motorcycle safety regulations
rider education and conspicuity improvement No more
information available Detroit MI US Society of Automotive
Engineers 1987147ndash52
Yuan 2000
Yuan W The effectiveness of the rsquoride-brightrsquo legislation for
motorcycles in Singapore Accident Analysis amp Prevention 200032
(4)559ndash63
Zambon 2006
Zambon F Hasselberg M Factors affecting the severity of injuries
among young motorcyclists a Swedish nationwide cohort study
Traffic Injury Prevention 20067(2)143ndash9lowast Indicates the major publication for the study
29Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3250
C H A R A C T E R I S T I C S O F S T U D I E S
Characteristics of included studies [ordered by study ID]
Adams 1985
Methods Non-randomised trial
Participants Initial sample size
Total of 162 Australia Post telegram delivery riders
Not clear how many in intervention and control groups
Analysis sample size
Intervention group 63Control group 35
Total 98 (61)
Interventions No formal training vs voluntary rider training course of 12 hours over 2 consecutive days (theory 4 hours practice
8 hours)
Intervention integrity
No information given
Outcomes Australia Post recorded crashes
Length of follow up 1 year
Notes Study conducted in NSW Australia
Anderson 1980
Methods Randomised trial
Participants Initial sample size
Total 41188 motorcycle licence applicants
Analysis sample size
Intervention group with skills training 13160
Intervention group without skills training 12634
Control group 15080
Total 40874 (99)
Interventions Standard licensing skill test vs the Motorcycle Operators Skill Test (MOST) with or without a training component
MOST is not to be confused with rider training because it is a licensing test The training component was for
participants who failed the skill test on the first attempt was of 3 hours duration and consisted of instructions about
the skills needed to pass the skill test The training consisted of 30 min theory and 2 frac12 hours practice on the
motorcycle
Intervention integrity
33 in MOST with skills training 31 in MOST without skills training and 43 in control group obtained their
motorcycle licence in the end
Outcomes Police recorded crashes and offences
Length of follow up 3 years
30Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3350
Anderson 1980 (Continued)
Notes Study conducted in California USA
Billheimer 1998
Methods Cohort study
Participants Initial sample size
Intervention group Unclear numbers of CMSP graduates and ERC graduates
Control group Over 16000 untrained motorcyclists
Analysis sample size
1139 matched CMSP-pairs and 1182 ERC-pairs
Interventions No formal training vs the mandatory (for those younger than 18 years) pre-licence 16-hours basic California
Motorcyclist Safety Program (CMSP) and the 8-hour Experienced Rider Course (ERC)
Intervention integrity
No information given
Outcomes Police recorded crashes and offences
Length of follow up 2 years
Notes Study conducted in California USA
Cooper 1988
Methods Cohort study
Participants Initial sample size
Intervention group 418 graduates and 57 who failed
Control group 402
Analysis sample size
Intervention group unclear
Control group unclear
Total 863 (984)
Interventions No formaltrainingvs non-mandatory pre-licence rider trainingThe content of thetrainingwas notfurther explainedIntervention integrity
Participants had all completed the rider training
Outcomes Police recorded crashes and offences
Length of follow up 7 years
Notes Study conducted in British Columbia Canada
31Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3450
Davis 1997
Methods Retrospective cohort study
Participants Initial sample size
Intervention group 9320 graduates
Control group unclear
Analysis sample size
Intervention group 9320 graduates
Control group unclear
Interventions No formal training vs the non-mandatory Connecticut Motorcycle Operator Training Program The content of
training was not further explained
Intervention integrity Participants had all completed the rider training
Outcomes Police recorded crashes severity of crashes
Length of follow up 1 year
Notes Study conducted in Connecticut USA
Hall 1988
Methods Randomised trial
Participants Initial sample sizeIntervention group A 6248
Intervention group B 6616
Intervention group C 6996
Control group 6604
Total 26464 motorcycle licence applicants
Analysis sample size
Intervention group A 6012 (96)
Intervention group B 6428 (97)
Intervention group C 6778 (97)
Control group 6502 (98)
Interventions Standard licensing skill test vs the Motorcycle Operator Skill Test (MOST II) without training (A) MOST II with
3 hours training (B) or MOST II with 20 hours training (C) The 3 hours training comprised instructions aboutthe skills needed to pass the skill test (30 min theory and 2 frac12 hours practice on the motorcycle) The 20 hours
training consisted of the Motorcycle Safety Foundationrsquos Motorcycle Rider Course (MRC) comprising of 8 hours
theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-road practice
Intervention integrity
25 and 26 obtained their motorcycle licence in the lsquoMOST 3-hour trainingrsquo and lsquoMOST 20-hour trainingrsquo
intervention arm respectively In the lsquoMOST testrsquo intervention arm 29 obtained their licence and in the control
group 35
Outcomes Police recorded crashes and offences
Length of follow up 3 years
32Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3550
Hall 1988 (Continued)
Notes Study conducted in New York USA
Haworth 1997
Methods Population based case-control study
Participants Cases Motorcycle riders or pillion riders involved in motorcycle crash (on road) Recruited from hospital and police
records Not clear whether all cases (especially fatalities) were included
Community controls Multiple matched sample of motorcycle rider or pillion who passed the crash site at the same
time of day and day of week that the crash occurred
Initial sample size
Cases 336
Controls 1195
Analysis sample size
Cases 234 (70)
Controls 561 (47)
Interventions Basic handling skills course beginner intermediate or advanced course measured through self-report not validated
Outcomes Injuries with an Injury Severity Score of 5 or greater or admission for at least four hours and fatalities from a
motorcycle crash (from hospital and police records)
Length of follow up two years or more prior to the interview
Notes Study conducted in Victoria Australia
Hurt 1981
Methods Population based case-control study
Participants Cases Motorcycle riders who had a motorcycle crash that was reported to police (traffic accident reports) Not all
cases included
Community controls Multiple matched sample of motorcycle riders who passed the crash site at the same time of
day and day of week as the crash occurred but 2 years later
Initial sample size
Cases 4500 notifications of crashes to police (only 50 received in time research team able to respond to 1126 for
unreported reasons data collection completed for 900)
Controls Number approached unreported 2310 consented and interviewed
Analysis sample size
Cases 808 (808900 data on rider training) - 18 of initial sample
Controls 670 (6702310 data on rider training) - lt 29 of initial sample
Interventions Training experience self taught friends-family school-club formal-professional others measured through self-
report validated
Outcomes Injuries and fatalities from a motorcycle crash (from hospital records and coronerrsquos office)
33Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3650
Hurt 1981 (Continued)
Notes Study conducted in Los Angeles USA
Jonah 1982
Methods Cohort study
Participants Initial sample size
Intervention group
2310 licensed MTP graduates who are active riders
Control group
2029 licensed motorcyclists who are active riders
Non-licensed and non-active riders excluded
Analysis sample size
Intervention group
811 interviews (35) 533 driver records (23)
Control group 1080 interviews (71) 905 driver records (45)
Interventions No training vs the non-mandatory Canada Safety Council Motorcycle Training Program (MTP) comprising of 4
hours theory and 16 hours practise (6 hrs on-road) over 2 weekends
Intervention integrity
Participants had all successfully completed the MTP
Outcomes Self-reported and police reported crashes and offencesLength of follow up 4 years
Notes Study conducted in Ontario Canada
Kloeden 1994
Methods Cohort study
Participants Initial sample size
Intervention group 2533 licence applicants
Control group Unclear
Sample of riders obtaining probationary licenceIntervention group 1625 (64)
Control group 5015 (unclear)
Non-active riders excluded from control group (~66)
Analysis sample size
Intervention group 631 (39)
Control group 318 (6)
Interventions No formal training vs the mandatory pre-licence Ridersafe course consisting of two levelsthe first is a training course
in the basics of motorcycle riding where a pass entitles the rider to obtain a learnerrsquos permit The second is a skill
test where a pass entitles the rider to obtain a probationary motorcycle licence The training course consists of two 4
hour training sessions on separate days involving instruction videos and actual riding on motorcycles The skill test
consists of a 4 hour preparation course of 74 minutes of lecturediscussion 38 minutes of video and 75 minutes of
34Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3750
Kloeden 1994 (Continued)
on-course riding which is the testing phase
Intervention integrity
Unclear what percentage in the Ridersafe group and the control group obtained their probationary licence halfway
and their full motorcycle licence in the end
Outcomes Police reported crashes
Length of follow up 1 year
Notes Study conducted in SA Australia
Kraus 1975
Methods Population based case-control study
Participants Cases All motorcycle riders with a confirmed medically treated motorcycle injury (including fatalities) from all
hospitals in study base and peripheral area hospitals and all deaths due to motorcycle crashes in the study base All
serious injuries (leading to hospitalisation) and fatalities likely to be included
Community controls A random sample of registered motorcycle owners in the same temporary and geographical
study base as the cases arose from
Initial sample size Cases 1273
Controls 738
Analysis sample size Cases 741 (58)
Controls 434 (59)
Interventions Motorcycle rider training (not specified) measured through self-report
Outcomes Injuries leading to hospitalisation fatalities
Length of follow up not specified
Notes Study conducted in Sacramento county USA
Leung 1987
Methods Non-randomised trial
Participants Initial sample sizeIntervention group unknown number of MRC graduates
Control group unknown number of licensed motorcyclists
Analysis sample size
Intervention group 2941
Control group 43094
Interventions No formal training vs non-mandatory Wisconsin motorcycle rider course a 22 hours off-street programme that
teaches the basic rules of safe motorcycling
Intervention integrity
50 in the MRC group and 100 in the control group obtained their motorcycle licence in the end
35Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3850
Leung 1987 (Continued)
Outcomes Police recorded crashes and offences
Length of follow up up to 3 years
Notes Study conducted in Wisconsin USA
MAIDS 2004
Methods Population based case-control study
Participants Cases Motorcycle riders who had a motorcycle crash Unclear whether all cases included
Community controls Motorcycle riders identified at petrol stations Unclear whether matched to cases
Initial sample size
Cases not reported
Controls not reported
Analysis sample size
Cases 921 (Italy 200 Netherlands 200 Spain 121 Germany 250 France 150)
Controls 923 (Italy 200 Netherlands 200 Spain123 Germany 250 France 150)
Interventions Pre-license rider training or additional training (non-compulsory)
Outcomes Crashes
Notes Study in five sampling areas located in five European countries (France Germany Netherlands Spain and Italy)
McDavid 1989
Methods Cohort study
Participants Sample size after first match
Intervention group 346 male training graduates
Control group 346 male motorcycle riders licensed in same year
Sample size after second match
Intervention group 139 (40)
Control group 139 (40)
Analysis sample size
Intervention group 139 (40)
Control group 139 (40)
Interventions No formal training vs the non-mandatory pre-licence British Columbia Safety Councilrsquos 37-hour motorcycle safety
programme
Intervention integrity
Participants had all completed and passed the training
Outcomes Police reported crashes injurious crashes
Length of follow up 1-6 years
36Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3950
McDavid 1989 (Continued)
Notes Study conducted in British Columbia Canada
Mortimer 1984
Methods Cohort study
Participants Initial sample size
Intervention group 608 MRC graduates
Control group 360 motorcycle riders
Response rates
Intervention group 360 (59)
Control group 324 (90)
Non-active riders excluded
Analysis sample size
Intervention group 213 (35)
Control group 303 (84)
Interventions No formal training vs the non-mandatory Motorcycle Safety Foundationrsquos Motorcycle Rider Course comprising of
8 hours theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-road practice
Intervention integrity
Participants had all completed and passed the MRC
Outcomes Self-reported crashes and offencesLength of follow up 1 year
Notes Study conducted in Illinois USA
Mortimer 1988
Methods Cohort study
Participants Initial sample size
Intervention group 2307 MRC graduates
Control group 556 active licensed motorcycle riders
Response ratesIntervention group 1298 (56)
Control group 500 (90)
Non-active riders excluded
Analysis sample size
Intervention group 913 (40)
Control group 500 (90)
Interventions No formal training vs the non-mandatory Motorcycle Safety Foundationrsquos Motorcycle Rider Course comprising of
8 hours theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-road practice
Intervention integrity
Participants had all completed and passed the MRC
37Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4050
Mortimer 1988 (Continued)
Outcomes Self reported crashes and offences
Length of follow up 1 year
Notes Study conducted in Illinois USA
Osga 1980
Methods Cohort study
Participants Initial sample sizeIntervention group 1163 MSF MRC participants invited 818 (74) responded
Control group 329 motorcycle riders responded to recruitment strategies
Non-active riders excluded from analysis
Analysis sample size
Intervention group 595 (73)
Control group 329 (100)
Interventions No formal training vs the non-mandatory Motorcycle Safety Foundationrsquos Motorcycle Rider Course (MRC) com-
prising of 8 hours theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-road
practice
Intervention integrity
Participants had all successfully completed the MRC
Outcomes Self-reported crashes police recorded crash reports
Length of follow up 6 months - 3 12 years
Notes Study conducted in South Dakota USA
Perrino 2002
Methods Cohort study
Participants Initial sample size
Intervention group 2144 (unclear) MRC graduates (motorcycle licence applicants)
Control group 2601 (unclear) regular licensing test graduates (motorcycle licence applicants)Analysis sample size
Intervention group 1070 (unclear)
Control group 867 (unclear)
Interventions No formal training vs the mandatory (for those younger than 18 years) pre-licence Maryland Motorcycle Rider
Course (MRC) consisting of 8 hours theory and 12 hours practice
Intervention integrity
No information given
Outcomes Self-reported and police recorded crashes and offences
Length of follow up 1 year
38Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4150
Perrino 2002 (Continued)
Notes Study conducted in Maryland USA
Raymond 1979
Methods Cohort study
Participants Initial sample size
Intervention group
258 volunteers mainly (but not all) novice riders
Control group
1104 provisional licence applicants
Analysis sample size
Intervention group 190 (74)
Control group 627 (57)
Interventions No formal training vs the non-mandatory RACACU (Royal Automobile ClubAuto-Cycle Union) training scheme
The training program consisted of either 24 one-hour sessions or 13 two-hour sessions Twelve hours of the training
focused on theory about basic operational procedures traffic laws and highway code The practical session focused
on basic operational skills riding on a training area and riding in traffic with an instructor A proficiency test was
administered at the end of the course
Intervention integrity
No information given
Outcomes Self-reported crashes (including minor non-injury non-damage and off-road crashes)
Length of follow up 1 year
Notes Study conducted in the UK
Satten 1980
Methods Cohort study
Participants Initial sample size
Intervention group interviewers kept interviewing until they had 100 riders selected from a population of 891
enrollees in the MSF MRC course (unknown how many were asked and how many refused to participate)Control group unknown number of registered motorcycle riders
Non-active riders excluded from analysis
Analysis sample size
Intervention group 71
Control group 100
Interventions No formal training vs the non-mandatory Motorcycle Safety Foundationrsquos (MSF) Motorcycle Rider Course (MRC)
comprising of 8 hours theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-
road practice
Intervention integrity
Participants had all successfully completed the MRC
39Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4250
Satten 1980 (Continued)
Outcomes Self-reported crashes injuries and offences
Length of follow up 3 years
Notes Study conducted in Illinois USA
Savolainen 2007
Methods Cohort study
Participants Initial sample size
Intervention group 4000 riders that had conducted the MSF MRC course (passed or failed)
Control group 4000 motorcycle owners or permit or endorsement holders from the Indiana Bureau of Motor
Vehicles database
Analysis sample size
Intervention group 739 (18)
Control group 588 (15)
Interventions Intervention integrity
No information given
Outcomes Probability of crash involvement
Length of follow up up to 5 years
Notes Study conducted in Indiana USA
Swaddiwudhipong 1998
Methods Randomised trial
Participants Initial sample size
Intervention group
all 16 villages of 3 sub-districts
(participants were practising motorcyclists licensed and unlicensed)
Control group
18 villages of 3 other sub-districtsAnalysis sample size
Intervention group 3 districts (100)
Control group 3 districts (100)
Interventions No rider education vs non-mandatory community based rider education consisting of theory on motorcycle injury
risk helmet use traffic laws vehicle regulations and traffic signs and a written and skill test leading to a licence
Intervention integrity
No information given
Outcomes Injuries fatalities
Length of follow up 2 years
40Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4350
Swaddiwudhipong 1998 (Continued)
Notes Study conducted in Tak Province rural Thailand
Waller 1992
Methods Cohort study
Participants Initial sample size
Intervention group 349 riders who succesfully completed the Department of Public Safety motorcycle operator
training in 1988
Control group 349 motorcyclists who never completed the DPS motorcycle operator training
Analysis sample size
Intervention group 349
Control group 349
Interventions No DPS motorcycle operator training vs non-mandatory DPS motorcycle operator training The DPS training is
the Motorcycle Safety Foundationrsquos Motorcycle Rider Course comprising eight hours of theory (lectures discussions
films and multiple choice tests) a textbook and 12 hours off-road practice
Intervention integrity
All 349 riders succesfully completed the rider training
Outcomes Police recorded crashes
Length of follow up 3 years
Notes Study conducted in
Characteristics of excluded studies [ordered by study ID]
Study Reason for exclusion
Anderson 1980a Study type review and process evaluation of a rider course
Aupetit 2008 Outcome proposal of a rider training curriculum that combines traditional training and a motorcycle riding
simulator
Awane 1999 Study type review on use of simulators in rider and driver training
Baldi 2005 Intervention not rider training but intervention integrity of rider training availabilty of rider education
implementation of graduated licensing system etc on a state-wide level
Batchler 1988 Study type description of a training course
Blanchard 2006 Study type not a study assessing the association between rider training and outcomes
Bonnett 2005 Outcome qualitative data regarding satisfaction with the training programme
41Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4450
(Continued)
Braver 2007 Outcome licensure rates
Clayton 1980 Outcome knowledge skills attitude
Council 1978 Outcome crash reduction in ANY vehicle not motorcycles exclusively
Crowther 2005 Study type qualitative study
Daniello 2009 Study type review of the effectiveness of motorcycle training and licensing
Dewanti 2007 Intervention emotional characteristics behaviour (ie not motocycle rider training)
Du Plooy 1990 Study type review of types of rider training courses in South Africa
Engel 1989 Study type uncontrolled before-after study
Goldenbeld 2002 Outcome skills
Goldenbeld 2004 Outcome knowledge and skills
Haworth 2000 Intervention balance between attitudinal andvehicle control skills acrosscurriculla of motorcycle rider training
courses
Haworth 2005 Study type review
Haworth 2006 Study type cross-sectional survey
Hill 1983 Study type review and description of training system
Imatake 1980 Study type course descriptionreview
Ingebritsen 1990 Intervention risk factors for motorcycle crashes
Jain 2009 Study type cross-sectional descriptive study of motorcycle crashes over a 5 year period
Jonah 1979 Study type cross-sectional survey Outcome scores on motorcycle operator skill test
Jonah 1981 Intervention evaluation of a motorcycle rider skill test not motorcycle rider training
Kadar 1990 Study type review of components of rider training courses
Kelsey 1986 Intervention evaluation of a motorcycle rider skill test not motorcycle rider training
Koch 1980 Study type reviewcourse description
Koch 1990 Study type description of course content
42Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4550
(Continued)
Labbett 2005 Study type essay about possible explanations for motorcycle deaths
Law 2005 Intervention targeted motorcycle safety programme not specifically rider training
Lowes 1990 Study type review and descriptions of rider training
Mahon 1976 Study type review of motorcycle rider training methods
Maxwell 1985 Outcomesrecommendations forprogramme operatorsand otherson howthe motorcycle trainingprogramme
works and how it can be improved
Mayhew 1996 Study type review
McKnight 1987 Study type evaluation of methodological and administrative aspects of a motorcycle operator skill test
McPherson 1976 Outcome knowledge and skills
Mohan 2004 Study type review of evidence of interventions to reduce road traffic injuries
Morgan 1980 Study type reviewcourse description
Motoki 1990 Study type proposal and development of rider training curriculum
Nairn 1992 Study type literature review
Newland 1980 Study type description of development of motorcycle safety course
Newland 1982 Study type descriptionreview of a rider training course
Newland 1990 Study type uncontrolled BeforeAfter study
Ormston 2003 Outcome changes in attitude and self reported behaviour
Osga 1979 Outcome knowledge attitudes and skills
Panou 2007 Intervention new simulation tool to teach driverrider
Peterson 1990 Study factor demographic details of course participants completing the course vs those not completing the
course
Prem 1984 Study type experimental
Study factor examines association between skills and performance in tests
Radin Umar 2006 Intervention motorcycle safety programme not specifically rider training
Rockwell 1990 Study type methodologically flawed study type
43Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4650
(Continued)
Rothe 1987 Study type qualitative study of rider training in general
Rowden 2007 Study type qualitative study
RSC Victoria 1993 Study type review of motorcycle rider training and education in Victoria
Russam 1979 Study type review of training programmes in Japan
Saffron 1981 Study type literature review
Schulz 1990 Study factor predictors of motorcycle crashes
Sexton 2004 Study type cross-sectional analytic analysis
Shephard 1986 Study factor reviewdescription of programme
Simpson 1990 Study type review of motorcycle rider trainingeducation
Sokolov 1990 Study factor review of motorcycle riding courses in USSR
Steffens 1988 Outcome self-reported behaviour experiences and researcher observed riding behaviour
Sudlow 2003 Study type review of motorcycle training in the UK
Tasmania 1995 Study type cross-sectional descriptiveprocess evaluation
Tetard 1985 Study type experimental
Outcome detection of indices
Thackray 1980 Outcome instructional effectiveness user acceptance administrative feasibility
Thompson 1994 Study type cross-sectional survey
Outcome attitudes and knowledge
Tobler 1990 Study type review of course curricula
Transport 1989 Intervention syllabus for rider training
Tuan 2007 Study type cross-sectional study (survey) which does not assess the effects of an intervention but examines
motocyclistsrsquo behaviour
Victor 1980 Study type review
Waterer 1990 Study type review of UK rider training issues
Weller 1989 Study type review of studies evaluating motorcycle safety and education measures
44Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4750
(Continued)
Wells 1981 Study type description of different methods of training
Winn 1986 Study type review of rsquohistoryrsquo of rider training
Wisher 1988 Study type cross-sectional survey
Wood 1987 Study type Interrupted Time Series but did not meet the criteria for an ITS as published by the Cochrane
Effective Practice and Organisation of Care Group
Yamazaki 1990 Study type cross-sectional survey
Outcome riding skills skills retention and attitude
Characteristics of ongoing studies [ordered by study ID]
Burgess 2010
Trial name or title UKs National Ride program
Methods Cohort study
Participants Motorcycle offenders
Interventions Police led diversion program
Outcomes Self reported attitudinal and behavioural measures
Starting date
Contact information
Notes Expected to be completed in early 2010
45Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4850
D A T A A N D A N A L Y S E S
This review has no analyses
A P P E N D I C E S
Appendix 1 Search strategy
We searched the following electronic databases
bull Cochrane Injuries Group Specialised Register (searched 22 Sept 2008)
bull CENTRAL (The Cochrane Library 2008 Issue 3)
bull TRANSPORT 1988 to 200706 (latest date available)
bull ATRI Australian Transport Index (to Sept 2008)
bull MEDLINE (Ovid SP) 1950 to Sept 2008
bull EMBASE (Ovid SP) 1980 to Sept 2008
bull CINAHL (EBSCO) 1982 to Sept 2008
bull WHOLIS (World Health Organization Library Information System) (to Sept 2008)
bull PsycINFO (Ovid SP) 1806 to Sept 2008
bull LILACS (BIREME) (Latin American and Caribbean Health Sciences) (1982 to Sept 2008)
bull ISI Web of Science Social Sciences Citation Index (SSCI) 1970 to Sept 2008
bull ERIC (searched 25 Sept 2008)
bull ZETOC (searched 25 Sept 2008)
bull SIGLE (searched 25 Sept 2008)
All of the databases were searched using the following text-word search strategy The strategy was adapted using either rsquorsquo or rsquo$rsquo as atruncation symbol
1 motor-cycl or motorcycl or motor-bik or motorbik or motor-scooter or motorscooter or scooter or moped or motocycl
or motocicl or moto
2 train or educat or instruct or program or course or qualif or learn or improv
3 1 and 2
H I S T O R Y
Protocol first published Issue 2 2005
Review first published Issue 10 2010
46Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4950
C O N T R I B U T I O N S O F A U T H O R S
Katina Kardamanidis was the first author of the protocol conducted the literature search extracted full studies assessed the studies
according to inclusion criteria assessed quality of the studies compared findings with the second author discussed disagreements with
the third author wrote the drafts and the final review
Rebecca Ivers was the second author of the protocol assessed the studies according to inclusion criteria assessed quality of the studies
compared findings with the first author edited the final draft and provided overall guidance for the process
Alexandra Martiniuk assessed quality of the studies compared findings with the first author discussed disagreements with the third
author and helped write the first draft and the final review
Mark Stevenson contributed to the protocol reviewed methods of studies and contributed to the final draft
Katrina Thistlethwaite conducted a second literature search extracted full studies and contributed to the final draft
D E C L A R A T I O N S O F I N T E R E S T
None known
S O U R C E S O F S U P P O R T
Internal sources
bull No sources of support supplied
External sources
bull National Health and Medical Research Council of Australia Australia
Salary support for Martiniuk Ivers and Stevenson
bull University of Sydney Sesqui Research Grant Australia
Direct research costs and salary support for Kardamanidis
D I F F E R E N C E S B E T W E E N P R O T O C O L A N D R E V I E W
There are no differences between the protocol and the review
I N D E X T E R M S
Medical Subject Headings (MeSH)
lowastMotorcycles Accident Prevention [lowastmethods] Accidents Traffic [lowastprevention amp control] Licensure Program Evaluation Wounds
and Injuries [mortality lowastprevention amp control]
47Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 5050
MeSH check words
Humans
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3250
C H A R A C T E R I S T I C S O F S T U D I E S
Characteristics of included studies [ordered by study ID]
Adams 1985
Methods Non-randomised trial
Participants Initial sample size
Total of 162 Australia Post telegram delivery riders
Not clear how many in intervention and control groups
Analysis sample size
Intervention group 63Control group 35
Total 98 (61)
Interventions No formal training vs voluntary rider training course of 12 hours over 2 consecutive days (theory 4 hours practice
8 hours)
Intervention integrity
No information given
Outcomes Australia Post recorded crashes
Length of follow up 1 year
Notes Study conducted in NSW Australia
Anderson 1980
Methods Randomised trial
Participants Initial sample size
Total 41188 motorcycle licence applicants
Analysis sample size
Intervention group with skills training 13160
Intervention group without skills training 12634
Control group 15080
Total 40874 (99)
Interventions Standard licensing skill test vs the Motorcycle Operators Skill Test (MOST) with or without a training component
MOST is not to be confused with rider training because it is a licensing test The training component was for
participants who failed the skill test on the first attempt was of 3 hours duration and consisted of instructions about
the skills needed to pass the skill test The training consisted of 30 min theory and 2 frac12 hours practice on the
motorcycle
Intervention integrity
33 in MOST with skills training 31 in MOST without skills training and 43 in control group obtained their
motorcycle licence in the end
Outcomes Police recorded crashes and offences
Length of follow up 3 years
30Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3350
Anderson 1980 (Continued)
Notes Study conducted in California USA
Billheimer 1998
Methods Cohort study
Participants Initial sample size
Intervention group Unclear numbers of CMSP graduates and ERC graduates
Control group Over 16000 untrained motorcyclists
Analysis sample size
1139 matched CMSP-pairs and 1182 ERC-pairs
Interventions No formal training vs the mandatory (for those younger than 18 years) pre-licence 16-hours basic California
Motorcyclist Safety Program (CMSP) and the 8-hour Experienced Rider Course (ERC)
Intervention integrity
No information given
Outcomes Police recorded crashes and offences
Length of follow up 2 years
Notes Study conducted in California USA
Cooper 1988
Methods Cohort study
Participants Initial sample size
Intervention group 418 graduates and 57 who failed
Control group 402
Analysis sample size
Intervention group unclear
Control group unclear
Total 863 (984)
Interventions No formaltrainingvs non-mandatory pre-licence rider trainingThe content of thetrainingwas notfurther explainedIntervention integrity
Participants had all completed the rider training
Outcomes Police recorded crashes and offences
Length of follow up 7 years
Notes Study conducted in British Columbia Canada
31Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3450
Davis 1997
Methods Retrospective cohort study
Participants Initial sample size
Intervention group 9320 graduates
Control group unclear
Analysis sample size
Intervention group 9320 graduates
Control group unclear
Interventions No formal training vs the non-mandatory Connecticut Motorcycle Operator Training Program The content of
training was not further explained
Intervention integrity Participants had all completed the rider training
Outcomes Police recorded crashes severity of crashes
Length of follow up 1 year
Notes Study conducted in Connecticut USA
Hall 1988
Methods Randomised trial
Participants Initial sample sizeIntervention group A 6248
Intervention group B 6616
Intervention group C 6996
Control group 6604
Total 26464 motorcycle licence applicants
Analysis sample size
Intervention group A 6012 (96)
Intervention group B 6428 (97)
Intervention group C 6778 (97)
Control group 6502 (98)
Interventions Standard licensing skill test vs the Motorcycle Operator Skill Test (MOST II) without training (A) MOST II with
3 hours training (B) or MOST II with 20 hours training (C) The 3 hours training comprised instructions aboutthe skills needed to pass the skill test (30 min theory and 2 frac12 hours practice on the motorcycle) The 20 hours
training consisted of the Motorcycle Safety Foundationrsquos Motorcycle Rider Course (MRC) comprising of 8 hours
theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-road practice
Intervention integrity
25 and 26 obtained their motorcycle licence in the lsquoMOST 3-hour trainingrsquo and lsquoMOST 20-hour trainingrsquo
intervention arm respectively In the lsquoMOST testrsquo intervention arm 29 obtained their licence and in the control
group 35
Outcomes Police recorded crashes and offences
Length of follow up 3 years
32Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3550
Hall 1988 (Continued)
Notes Study conducted in New York USA
Haworth 1997
Methods Population based case-control study
Participants Cases Motorcycle riders or pillion riders involved in motorcycle crash (on road) Recruited from hospital and police
records Not clear whether all cases (especially fatalities) were included
Community controls Multiple matched sample of motorcycle rider or pillion who passed the crash site at the same
time of day and day of week that the crash occurred
Initial sample size
Cases 336
Controls 1195
Analysis sample size
Cases 234 (70)
Controls 561 (47)
Interventions Basic handling skills course beginner intermediate or advanced course measured through self-report not validated
Outcomes Injuries with an Injury Severity Score of 5 or greater or admission for at least four hours and fatalities from a
motorcycle crash (from hospital and police records)
Length of follow up two years or more prior to the interview
Notes Study conducted in Victoria Australia
Hurt 1981
Methods Population based case-control study
Participants Cases Motorcycle riders who had a motorcycle crash that was reported to police (traffic accident reports) Not all
cases included
Community controls Multiple matched sample of motorcycle riders who passed the crash site at the same time of
day and day of week as the crash occurred but 2 years later
Initial sample size
Cases 4500 notifications of crashes to police (only 50 received in time research team able to respond to 1126 for
unreported reasons data collection completed for 900)
Controls Number approached unreported 2310 consented and interviewed
Analysis sample size
Cases 808 (808900 data on rider training) - 18 of initial sample
Controls 670 (6702310 data on rider training) - lt 29 of initial sample
Interventions Training experience self taught friends-family school-club formal-professional others measured through self-
report validated
Outcomes Injuries and fatalities from a motorcycle crash (from hospital records and coronerrsquos office)
33Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3650
Hurt 1981 (Continued)
Notes Study conducted in Los Angeles USA
Jonah 1982
Methods Cohort study
Participants Initial sample size
Intervention group
2310 licensed MTP graduates who are active riders
Control group
2029 licensed motorcyclists who are active riders
Non-licensed and non-active riders excluded
Analysis sample size
Intervention group
811 interviews (35) 533 driver records (23)
Control group 1080 interviews (71) 905 driver records (45)
Interventions No training vs the non-mandatory Canada Safety Council Motorcycle Training Program (MTP) comprising of 4
hours theory and 16 hours practise (6 hrs on-road) over 2 weekends
Intervention integrity
Participants had all successfully completed the MTP
Outcomes Self-reported and police reported crashes and offencesLength of follow up 4 years
Notes Study conducted in Ontario Canada
Kloeden 1994
Methods Cohort study
Participants Initial sample size
Intervention group 2533 licence applicants
Control group Unclear
Sample of riders obtaining probationary licenceIntervention group 1625 (64)
Control group 5015 (unclear)
Non-active riders excluded from control group (~66)
Analysis sample size
Intervention group 631 (39)
Control group 318 (6)
Interventions No formal training vs the mandatory pre-licence Ridersafe course consisting of two levelsthe first is a training course
in the basics of motorcycle riding where a pass entitles the rider to obtain a learnerrsquos permit The second is a skill
test where a pass entitles the rider to obtain a probationary motorcycle licence The training course consists of two 4
hour training sessions on separate days involving instruction videos and actual riding on motorcycles The skill test
consists of a 4 hour preparation course of 74 minutes of lecturediscussion 38 minutes of video and 75 minutes of
34Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3750
Kloeden 1994 (Continued)
on-course riding which is the testing phase
Intervention integrity
Unclear what percentage in the Ridersafe group and the control group obtained their probationary licence halfway
and their full motorcycle licence in the end
Outcomes Police reported crashes
Length of follow up 1 year
Notes Study conducted in SA Australia
Kraus 1975
Methods Population based case-control study
Participants Cases All motorcycle riders with a confirmed medically treated motorcycle injury (including fatalities) from all
hospitals in study base and peripheral area hospitals and all deaths due to motorcycle crashes in the study base All
serious injuries (leading to hospitalisation) and fatalities likely to be included
Community controls A random sample of registered motorcycle owners in the same temporary and geographical
study base as the cases arose from
Initial sample size Cases 1273
Controls 738
Analysis sample size Cases 741 (58)
Controls 434 (59)
Interventions Motorcycle rider training (not specified) measured through self-report
Outcomes Injuries leading to hospitalisation fatalities
Length of follow up not specified
Notes Study conducted in Sacramento county USA
Leung 1987
Methods Non-randomised trial
Participants Initial sample sizeIntervention group unknown number of MRC graduates
Control group unknown number of licensed motorcyclists
Analysis sample size
Intervention group 2941
Control group 43094
Interventions No formal training vs non-mandatory Wisconsin motorcycle rider course a 22 hours off-street programme that
teaches the basic rules of safe motorcycling
Intervention integrity
50 in the MRC group and 100 in the control group obtained their motorcycle licence in the end
35Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3850
Leung 1987 (Continued)
Outcomes Police recorded crashes and offences
Length of follow up up to 3 years
Notes Study conducted in Wisconsin USA
MAIDS 2004
Methods Population based case-control study
Participants Cases Motorcycle riders who had a motorcycle crash Unclear whether all cases included
Community controls Motorcycle riders identified at petrol stations Unclear whether matched to cases
Initial sample size
Cases not reported
Controls not reported
Analysis sample size
Cases 921 (Italy 200 Netherlands 200 Spain 121 Germany 250 France 150)
Controls 923 (Italy 200 Netherlands 200 Spain123 Germany 250 France 150)
Interventions Pre-license rider training or additional training (non-compulsory)
Outcomes Crashes
Notes Study in five sampling areas located in five European countries (France Germany Netherlands Spain and Italy)
McDavid 1989
Methods Cohort study
Participants Sample size after first match
Intervention group 346 male training graduates
Control group 346 male motorcycle riders licensed in same year
Sample size after second match
Intervention group 139 (40)
Control group 139 (40)
Analysis sample size
Intervention group 139 (40)
Control group 139 (40)
Interventions No formal training vs the non-mandatory pre-licence British Columbia Safety Councilrsquos 37-hour motorcycle safety
programme
Intervention integrity
Participants had all completed and passed the training
Outcomes Police reported crashes injurious crashes
Length of follow up 1-6 years
36Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3950
McDavid 1989 (Continued)
Notes Study conducted in British Columbia Canada
Mortimer 1984
Methods Cohort study
Participants Initial sample size
Intervention group 608 MRC graduates
Control group 360 motorcycle riders
Response rates
Intervention group 360 (59)
Control group 324 (90)
Non-active riders excluded
Analysis sample size
Intervention group 213 (35)
Control group 303 (84)
Interventions No formal training vs the non-mandatory Motorcycle Safety Foundationrsquos Motorcycle Rider Course comprising of
8 hours theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-road practice
Intervention integrity
Participants had all completed and passed the MRC
Outcomes Self-reported crashes and offencesLength of follow up 1 year
Notes Study conducted in Illinois USA
Mortimer 1988
Methods Cohort study
Participants Initial sample size
Intervention group 2307 MRC graduates
Control group 556 active licensed motorcycle riders
Response ratesIntervention group 1298 (56)
Control group 500 (90)
Non-active riders excluded
Analysis sample size
Intervention group 913 (40)
Control group 500 (90)
Interventions No formal training vs the non-mandatory Motorcycle Safety Foundationrsquos Motorcycle Rider Course comprising of
8 hours theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-road practice
Intervention integrity
Participants had all completed and passed the MRC
37Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4050
Mortimer 1988 (Continued)
Outcomes Self reported crashes and offences
Length of follow up 1 year
Notes Study conducted in Illinois USA
Osga 1980
Methods Cohort study
Participants Initial sample sizeIntervention group 1163 MSF MRC participants invited 818 (74) responded
Control group 329 motorcycle riders responded to recruitment strategies
Non-active riders excluded from analysis
Analysis sample size
Intervention group 595 (73)
Control group 329 (100)
Interventions No formal training vs the non-mandatory Motorcycle Safety Foundationrsquos Motorcycle Rider Course (MRC) com-
prising of 8 hours theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-road
practice
Intervention integrity
Participants had all successfully completed the MRC
Outcomes Self-reported crashes police recorded crash reports
Length of follow up 6 months - 3 12 years
Notes Study conducted in South Dakota USA
Perrino 2002
Methods Cohort study
Participants Initial sample size
Intervention group 2144 (unclear) MRC graduates (motorcycle licence applicants)
Control group 2601 (unclear) regular licensing test graduates (motorcycle licence applicants)Analysis sample size
Intervention group 1070 (unclear)
Control group 867 (unclear)
Interventions No formal training vs the mandatory (for those younger than 18 years) pre-licence Maryland Motorcycle Rider
Course (MRC) consisting of 8 hours theory and 12 hours practice
Intervention integrity
No information given
Outcomes Self-reported and police recorded crashes and offences
Length of follow up 1 year
38Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4150
Perrino 2002 (Continued)
Notes Study conducted in Maryland USA
Raymond 1979
Methods Cohort study
Participants Initial sample size
Intervention group
258 volunteers mainly (but not all) novice riders
Control group
1104 provisional licence applicants
Analysis sample size
Intervention group 190 (74)
Control group 627 (57)
Interventions No formal training vs the non-mandatory RACACU (Royal Automobile ClubAuto-Cycle Union) training scheme
The training program consisted of either 24 one-hour sessions or 13 two-hour sessions Twelve hours of the training
focused on theory about basic operational procedures traffic laws and highway code The practical session focused
on basic operational skills riding on a training area and riding in traffic with an instructor A proficiency test was
administered at the end of the course
Intervention integrity
No information given
Outcomes Self-reported crashes (including minor non-injury non-damage and off-road crashes)
Length of follow up 1 year
Notes Study conducted in the UK
Satten 1980
Methods Cohort study
Participants Initial sample size
Intervention group interviewers kept interviewing until they had 100 riders selected from a population of 891
enrollees in the MSF MRC course (unknown how many were asked and how many refused to participate)Control group unknown number of registered motorcycle riders
Non-active riders excluded from analysis
Analysis sample size
Intervention group 71
Control group 100
Interventions No formal training vs the non-mandatory Motorcycle Safety Foundationrsquos (MSF) Motorcycle Rider Course (MRC)
comprising of 8 hours theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-
road practice
Intervention integrity
Participants had all successfully completed the MRC
39Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4250
Satten 1980 (Continued)
Outcomes Self-reported crashes injuries and offences
Length of follow up 3 years
Notes Study conducted in Illinois USA
Savolainen 2007
Methods Cohort study
Participants Initial sample size
Intervention group 4000 riders that had conducted the MSF MRC course (passed or failed)
Control group 4000 motorcycle owners or permit or endorsement holders from the Indiana Bureau of Motor
Vehicles database
Analysis sample size
Intervention group 739 (18)
Control group 588 (15)
Interventions Intervention integrity
No information given
Outcomes Probability of crash involvement
Length of follow up up to 5 years
Notes Study conducted in Indiana USA
Swaddiwudhipong 1998
Methods Randomised trial
Participants Initial sample size
Intervention group
all 16 villages of 3 sub-districts
(participants were practising motorcyclists licensed and unlicensed)
Control group
18 villages of 3 other sub-districtsAnalysis sample size
Intervention group 3 districts (100)
Control group 3 districts (100)
Interventions No rider education vs non-mandatory community based rider education consisting of theory on motorcycle injury
risk helmet use traffic laws vehicle regulations and traffic signs and a written and skill test leading to a licence
Intervention integrity
No information given
Outcomes Injuries fatalities
Length of follow up 2 years
40Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4350
Swaddiwudhipong 1998 (Continued)
Notes Study conducted in Tak Province rural Thailand
Waller 1992
Methods Cohort study
Participants Initial sample size
Intervention group 349 riders who succesfully completed the Department of Public Safety motorcycle operator
training in 1988
Control group 349 motorcyclists who never completed the DPS motorcycle operator training
Analysis sample size
Intervention group 349
Control group 349
Interventions No DPS motorcycle operator training vs non-mandatory DPS motorcycle operator training The DPS training is
the Motorcycle Safety Foundationrsquos Motorcycle Rider Course comprising eight hours of theory (lectures discussions
films and multiple choice tests) a textbook and 12 hours off-road practice
Intervention integrity
All 349 riders succesfully completed the rider training
Outcomes Police recorded crashes
Length of follow up 3 years
Notes Study conducted in
Characteristics of excluded studies [ordered by study ID]
Study Reason for exclusion
Anderson 1980a Study type review and process evaluation of a rider course
Aupetit 2008 Outcome proposal of a rider training curriculum that combines traditional training and a motorcycle riding
simulator
Awane 1999 Study type review on use of simulators in rider and driver training
Baldi 2005 Intervention not rider training but intervention integrity of rider training availabilty of rider education
implementation of graduated licensing system etc on a state-wide level
Batchler 1988 Study type description of a training course
Blanchard 2006 Study type not a study assessing the association between rider training and outcomes
Bonnett 2005 Outcome qualitative data regarding satisfaction with the training programme
41Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4450
(Continued)
Braver 2007 Outcome licensure rates
Clayton 1980 Outcome knowledge skills attitude
Council 1978 Outcome crash reduction in ANY vehicle not motorcycles exclusively
Crowther 2005 Study type qualitative study
Daniello 2009 Study type review of the effectiveness of motorcycle training and licensing
Dewanti 2007 Intervention emotional characteristics behaviour (ie not motocycle rider training)
Du Plooy 1990 Study type review of types of rider training courses in South Africa
Engel 1989 Study type uncontrolled before-after study
Goldenbeld 2002 Outcome skills
Goldenbeld 2004 Outcome knowledge and skills
Haworth 2000 Intervention balance between attitudinal andvehicle control skills acrosscurriculla of motorcycle rider training
courses
Haworth 2005 Study type review
Haworth 2006 Study type cross-sectional survey
Hill 1983 Study type review and description of training system
Imatake 1980 Study type course descriptionreview
Ingebritsen 1990 Intervention risk factors for motorcycle crashes
Jain 2009 Study type cross-sectional descriptive study of motorcycle crashes over a 5 year period
Jonah 1979 Study type cross-sectional survey Outcome scores on motorcycle operator skill test
Jonah 1981 Intervention evaluation of a motorcycle rider skill test not motorcycle rider training
Kadar 1990 Study type review of components of rider training courses
Kelsey 1986 Intervention evaluation of a motorcycle rider skill test not motorcycle rider training
Koch 1980 Study type reviewcourse description
Koch 1990 Study type description of course content
42Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4550
(Continued)
Labbett 2005 Study type essay about possible explanations for motorcycle deaths
Law 2005 Intervention targeted motorcycle safety programme not specifically rider training
Lowes 1990 Study type review and descriptions of rider training
Mahon 1976 Study type review of motorcycle rider training methods
Maxwell 1985 Outcomesrecommendations forprogramme operatorsand otherson howthe motorcycle trainingprogramme
works and how it can be improved
Mayhew 1996 Study type review
McKnight 1987 Study type evaluation of methodological and administrative aspects of a motorcycle operator skill test
McPherson 1976 Outcome knowledge and skills
Mohan 2004 Study type review of evidence of interventions to reduce road traffic injuries
Morgan 1980 Study type reviewcourse description
Motoki 1990 Study type proposal and development of rider training curriculum
Nairn 1992 Study type literature review
Newland 1980 Study type description of development of motorcycle safety course
Newland 1982 Study type descriptionreview of a rider training course
Newland 1990 Study type uncontrolled BeforeAfter study
Ormston 2003 Outcome changes in attitude and self reported behaviour
Osga 1979 Outcome knowledge attitudes and skills
Panou 2007 Intervention new simulation tool to teach driverrider
Peterson 1990 Study factor demographic details of course participants completing the course vs those not completing the
course
Prem 1984 Study type experimental
Study factor examines association between skills and performance in tests
Radin Umar 2006 Intervention motorcycle safety programme not specifically rider training
Rockwell 1990 Study type methodologically flawed study type
43Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4650
(Continued)
Rothe 1987 Study type qualitative study of rider training in general
Rowden 2007 Study type qualitative study
RSC Victoria 1993 Study type review of motorcycle rider training and education in Victoria
Russam 1979 Study type review of training programmes in Japan
Saffron 1981 Study type literature review
Schulz 1990 Study factor predictors of motorcycle crashes
Sexton 2004 Study type cross-sectional analytic analysis
Shephard 1986 Study factor reviewdescription of programme
Simpson 1990 Study type review of motorcycle rider trainingeducation
Sokolov 1990 Study factor review of motorcycle riding courses in USSR
Steffens 1988 Outcome self-reported behaviour experiences and researcher observed riding behaviour
Sudlow 2003 Study type review of motorcycle training in the UK
Tasmania 1995 Study type cross-sectional descriptiveprocess evaluation
Tetard 1985 Study type experimental
Outcome detection of indices
Thackray 1980 Outcome instructional effectiveness user acceptance administrative feasibility
Thompson 1994 Study type cross-sectional survey
Outcome attitudes and knowledge
Tobler 1990 Study type review of course curricula
Transport 1989 Intervention syllabus for rider training
Tuan 2007 Study type cross-sectional study (survey) which does not assess the effects of an intervention but examines
motocyclistsrsquo behaviour
Victor 1980 Study type review
Waterer 1990 Study type review of UK rider training issues
Weller 1989 Study type review of studies evaluating motorcycle safety and education measures
44Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4750
(Continued)
Wells 1981 Study type description of different methods of training
Winn 1986 Study type review of rsquohistoryrsquo of rider training
Wisher 1988 Study type cross-sectional survey
Wood 1987 Study type Interrupted Time Series but did not meet the criteria for an ITS as published by the Cochrane
Effective Practice and Organisation of Care Group
Yamazaki 1990 Study type cross-sectional survey
Outcome riding skills skills retention and attitude
Characteristics of ongoing studies [ordered by study ID]
Burgess 2010
Trial name or title UKs National Ride program
Methods Cohort study
Participants Motorcycle offenders
Interventions Police led diversion program
Outcomes Self reported attitudinal and behavioural measures
Starting date
Contact information
Notes Expected to be completed in early 2010
45Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4850
D A T A A N D A N A L Y S E S
This review has no analyses
A P P E N D I C E S
Appendix 1 Search strategy
We searched the following electronic databases
bull Cochrane Injuries Group Specialised Register (searched 22 Sept 2008)
bull CENTRAL (The Cochrane Library 2008 Issue 3)
bull TRANSPORT 1988 to 200706 (latest date available)
bull ATRI Australian Transport Index (to Sept 2008)
bull MEDLINE (Ovid SP) 1950 to Sept 2008
bull EMBASE (Ovid SP) 1980 to Sept 2008
bull CINAHL (EBSCO) 1982 to Sept 2008
bull WHOLIS (World Health Organization Library Information System) (to Sept 2008)
bull PsycINFO (Ovid SP) 1806 to Sept 2008
bull LILACS (BIREME) (Latin American and Caribbean Health Sciences) (1982 to Sept 2008)
bull ISI Web of Science Social Sciences Citation Index (SSCI) 1970 to Sept 2008
bull ERIC (searched 25 Sept 2008)
bull ZETOC (searched 25 Sept 2008)
bull SIGLE (searched 25 Sept 2008)
All of the databases were searched using the following text-word search strategy The strategy was adapted using either rsquorsquo or rsquo$rsquo as atruncation symbol
1 motor-cycl or motorcycl or motor-bik or motorbik or motor-scooter or motorscooter or scooter or moped or motocycl
or motocicl or moto
2 train or educat or instruct or program or course or qualif or learn or improv
3 1 and 2
H I S T O R Y
Protocol first published Issue 2 2005
Review first published Issue 10 2010
46Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4950
C O N T R I B U T I O N S O F A U T H O R S
Katina Kardamanidis was the first author of the protocol conducted the literature search extracted full studies assessed the studies
according to inclusion criteria assessed quality of the studies compared findings with the second author discussed disagreements with
the third author wrote the drafts and the final review
Rebecca Ivers was the second author of the protocol assessed the studies according to inclusion criteria assessed quality of the studies
compared findings with the first author edited the final draft and provided overall guidance for the process
Alexandra Martiniuk assessed quality of the studies compared findings with the first author discussed disagreements with the third
author and helped write the first draft and the final review
Mark Stevenson contributed to the protocol reviewed methods of studies and contributed to the final draft
Katrina Thistlethwaite conducted a second literature search extracted full studies and contributed to the final draft
D E C L A R A T I O N S O F I N T E R E S T
None known
S O U R C E S O F S U P P O R T
Internal sources
bull No sources of support supplied
External sources
bull National Health and Medical Research Council of Australia Australia
Salary support for Martiniuk Ivers and Stevenson
bull University of Sydney Sesqui Research Grant Australia
Direct research costs and salary support for Kardamanidis
D I F F E R E N C E S B E T W E E N P R O T O C O L A N D R E V I E W
There are no differences between the protocol and the review
I N D E X T E R M S
Medical Subject Headings (MeSH)
lowastMotorcycles Accident Prevention [lowastmethods] Accidents Traffic [lowastprevention amp control] Licensure Program Evaluation Wounds
and Injuries [mortality lowastprevention amp control]
47Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 5050
MeSH check words
Humans
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3350
Anderson 1980 (Continued)
Notes Study conducted in California USA
Billheimer 1998
Methods Cohort study
Participants Initial sample size
Intervention group Unclear numbers of CMSP graduates and ERC graduates
Control group Over 16000 untrained motorcyclists
Analysis sample size
1139 matched CMSP-pairs and 1182 ERC-pairs
Interventions No formal training vs the mandatory (for those younger than 18 years) pre-licence 16-hours basic California
Motorcyclist Safety Program (CMSP) and the 8-hour Experienced Rider Course (ERC)
Intervention integrity
No information given
Outcomes Police recorded crashes and offences
Length of follow up 2 years
Notes Study conducted in California USA
Cooper 1988
Methods Cohort study
Participants Initial sample size
Intervention group 418 graduates and 57 who failed
Control group 402
Analysis sample size
Intervention group unclear
Control group unclear
Total 863 (984)
Interventions No formaltrainingvs non-mandatory pre-licence rider trainingThe content of thetrainingwas notfurther explainedIntervention integrity
Participants had all completed the rider training
Outcomes Police recorded crashes and offences
Length of follow up 7 years
Notes Study conducted in British Columbia Canada
31Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3450
Davis 1997
Methods Retrospective cohort study
Participants Initial sample size
Intervention group 9320 graduates
Control group unclear
Analysis sample size
Intervention group 9320 graduates
Control group unclear
Interventions No formal training vs the non-mandatory Connecticut Motorcycle Operator Training Program The content of
training was not further explained
Intervention integrity Participants had all completed the rider training
Outcomes Police recorded crashes severity of crashes
Length of follow up 1 year
Notes Study conducted in Connecticut USA
Hall 1988
Methods Randomised trial
Participants Initial sample sizeIntervention group A 6248
Intervention group B 6616
Intervention group C 6996
Control group 6604
Total 26464 motorcycle licence applicants
Analysis sample size
Intervention group A 6012 (96)
Intervention group B 6428 (97)
Intervention group C 6778 (97)
Control group 6502 (98)
Interventions Standard licensing skill test vs the Motorcycle Operator Skill Test (MOST II) without training (A) MOST II with
3 hours training (B) or MOST II with 20 hours training (C) The 3 hours training comprised instructions aboutthe skills needed to pass the skill test (30 min theory and 2 frac12 hours practice on the motorcycle) The 20 hours
training consisted of the Motorcycle Safety Foundationrsquos Motorcycle Rider Course (MRC) comprising of 8 hours
theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-road practice
Intervention integrity
25 and 26 obtained their motorcycle licence in the lsquoMOST 3-hour trainingrsquo and lsquoMOST 20-hour trainingrsquo
intervention arm respectively In the lsquoMOST testrsquo intervention arm 29 obtained their licence and in the control
group 35
Outcomes Police recorded crashes and offences
Length of follow up 3 years
32Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3550
Hall 1988 (Continued)
Notes Study conducted in New York USA
Haworth 1997
Methods Population based case-control study
Participants Cases Motorcycle riders or pillion riders involved in motorcycle crash (on road) Recruited from hospital and police
records Not clear whether all cases (especially fatalities) were included
Community controls Multiple matched sample of motorcycle rider or pillion who passed the crash site at the same
time of day and day of week that the crash occurred
Initial sample size
Cases 336
Controls 1195
Analysis sample size
Cases 234 (70)
Controls 561 (47)
Interventions Basic handling skills course beginner intermediate or advanced course measured through self-report not validated
Outcomes Injuries with an Injury Severity Score of 5 or greater or admission for at least four hours and fatalities from a
motorcycle crash (from hospital and police records)
Length of follow up two years or more prior to the interview
Notes Study conducted in Victoria Australia
Hurt 1981
Methods Population based case-control study
Participants Cases Motorcycle riders who had a motorcycle crash that was reported to police (traffic accident reports) Not all
cases included
Community controls Multiple matched sample of motorcycle riders who passed the crash site at the same time of
day and day of week as the crash occurred but 2 years later
Initial sample size
Cases 4500 notifications of crashes to police (only 50 received in time research team able to respond to 1126 for
unreported reasons data collection completed for 900)
Controls Number approached unreported 2310 consented and interviewed
Analysis sample size
Cases 808 (808900 data on rider training) - 18 of initial sample
Controls 670 (6702310 data on rider training) - lt 29 of initial sample
Interventions Training experience self taught friends-family school-club formal-professional others measured through self-
report validated
Outcomes Injuries and fatalities from a motorcycle crash (from hospital records and coronerrsquos office)
33Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3650
Hurt 1981 (Continued)
Notes Study conducted in Los Angeles USA
Jonah 1982
Methods Cohort study
Participants Initial sample size
Intervention group
2310 licensed MTP graduates who are active riders
Control group
2029 licensed motorcyclists who are active riders
Non-licensed and non-active riders excluded
Analysis sample size
Intervention group
811 interviews (35) 533 driver records (23)
Control group 1080 interviews (71) 905 driver records (45)
Interventions No training vs the non-mandatory Canada Safety Council Motorcycle Training Program (MTP) comprising of 4
hours theory and 16 hours practise (6 hrs on-road) over 2 weekends
Intervention integrity
Participants had all successfully completed the MTP
Outcomes Self-reported and police reported crashes and offencesLength of follow up 4 years
Notes Study conducted in Ontario Canada
Kloeden 1994
Methods Cohort study
Participants Initial sample size
Intervention group 2533 licence applicants
Control group Unclear
Sample of riders obtaining probationary licenceIntervention group 1625 (64)
Control group 5015 (unclear)
Non-active riders excluded from control group (~66)
Analysis sample size
Intervention group 631 (39)
Control group 318 (6)
Interventions No formal training vs the mandatory pre-licence Ridersafe course consisting of two levelsthe first is a training course
in the basics of motorcycle riding where a pass entitles the rider to obtain a learnerrsquos permit The second is a skill
test where a pass entitles the rider to obtain a probationary motorcycle licence The training course consists of two 4
hour training sessions on separate days involving instruction videos and actual riding on motorcycles The skill test
consists of a 4 hour preparation course of 74 minutes of lecturediscussion 38 minutes of video and 75 minutes of
34Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3750
Kloeden 1994 (Continued)
on-course riding which is the testing phase
Intervention integrity
Unclear what percentage in the Ridersafe group and the control group obtained their probationary licence halfway
and their full motorcycle licence in the end
Outcomes Police reported crashes
Length of follow up 1 year
Notes Study conducted in SA Australia
Kraus 1975
Methods Population based case-control study
Participants Cases All motorcycle riders with a confirmed medically treated motorcycle injury (including fatalities) from all
hospitals in study base and peripheral area hospitals and all deaths due to motorcycle crashes in the study base All
serious injuries (leading to hospitalisation) and fatalities likely to be included
Community controls A random sample of registered motorcycle owners in the same temporary and geographical
study base as the cases arose from
Initial sample size Cases 1273
Controls 738
Analysis sample size Cases 741 (58)
Controls 434 (59)
Interventions Motorcycle rider training (not specified) measured through self-report
Outcomes Injuries leading to hospitalisation fatalities
Length of follow up not specified
Notes Study conducted in Sacramento county USA
Leung 1987
Methods Non-randomised trial
Participants Initial sample sizeIntervention group unknown number of MRC graduates
Control group unknown number of licensed motorcyclists
Analysis sample size
Intervention group 2941
Control group 43094
Interventions No formal training vs non-mandatory Wisconsin motorcycle rider course a 22 hours off-street programme that
teaches the basic rules of safe motorcycling
Intervention integrity
50 in the MRC group and 100 in the control group obtained their motorcycle licence in the end
35Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3850
Leung 1987 (Continued)
Outcomes Police recorded crashes and offences
Length of follow up up to 3 years
Notes Study conducted in Wisconsin USA
MAIDS 2004
Methods Population based case-control study
Participants Cases Motorcycle riders who had a motorcycle crash Unclear whether all cases included
Community controls Motorcycle riders identified at petrol stations Unclear whether matched to cases
Initial sample size
Cases not reported
Controls not reported
Analysis sample size
Cases 921 (Italy 200 Netherlands 200 Spain 121 Germany 250 France 150)
Controls 923 (Italy 200 Netherlands 200 Spain123 Germany 250 France 150)
Interventions Pre-license rider training or additional training (non-compulsory)
Outcomes Crashes
Notes Study in five sampling areas located in five European countries (France Germany Netherlands Spain and Italy)
McDavid 1989
Methods Cohort study
Participants Sample size after first match
Intervention group 346 male training graduates
Control group 346 male motorcycle riders licensed in same year
Sample size after second match
Intervention group 139 (40)
Control group 139 (40)
Analysis sample size
Intervention group 139 (40)
Control group 139 (40)
Interventions No formal training vs the non-mandatory pre-licence British Columbia Safety Councilrsquos 37-hour motorcycle safety
programme
Intervention integrity
Participants had all completed and passed the training
Outcomes Police reported crashes injurious crashes
Length of follow up 1-6 years
36Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3950
McDavid 1989 (Continued)
Notes Study conducted in British Columbia Canada
Mortimer 1984
Methods Cohort study
Participants Initial sample size
Intervention group 608 MRC graduates
Control group 360 motorcycle riders
Response rates
Intervention group 360 (59)
Control group 324 (90)
Non-active riders excluded
Analysis sample size
Intervention group 213 (35)
Control group 303 (84)
Interventions No formal training vs the non-mandatory Motorcycle Safety Foundationrsquos Motorcycle Rider Course comprising of
8 hours theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-road practice
Intervention integrity
Participants had all completed and passed the MRC
Outcomes Self-reported crashes and offencesLength of follow up 1 year
Notes Study conducted in Illinois USA
Mortimer 1988
Methods Cohort study
Participants Initial sample size
Intervention group 2307 MRC graduates
Control group 556 active licensed motorcycle riders
Response ratesIntervention group 1298 (56)
Control group 500 (90)
Non-active riders excluded
Analysis sample size
Intervention group 913 (40)
Control group 500 (90)
Interventions No formal training vs the non-mandatory Motorcycle Safety Foundationrsquos Motorcycle Rider Course comprising of
8 hours theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-road practice
Intervention integrity
Participants had all completed and passed the MRC
37Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4050
Mortimer 1988 (Continued)
Outcomes Self reported crashes and offences
Length of follow up 1 year
Notes Study conducted in Illinois USA
Osga 1980
Methods Cohort study
Participants Initial sample sizeIntervention group 1163 MSF MRC participants invited 818 (74) responded
Control group 329 motorcycle riders responded to recruitment strategies
Non-active riders excluded from analysis
Analysis sample size
Intervention group 595 (73)
Control group 329 (100)
Interventions No formal training vs the non-mandatory Motorcycle Safety Foundationrsquos Motorcycle Rider Course (MRC) com-
prising of 8 hours theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-road
practice
Intervention integrity
Participants had all successfully completed the MRC
Outcomes Self-reported crashes police recorded crash reports
Length of follow up 6 months - 3 12 years
Notes Study conducted in South Dakota USA
Perrino 2002
Methods Cohort study
Participants Initial sample size
Intervention group 2144 (unclear) MRC graduates (motorcycle licence applicants)
Control group 2601 (unclear) regular licensing test graduates (motorcycle licence applicants)Analysis sample size
Intervention group 1070 (unclear)
Control group 867 (unclear)
Interventions No formal training vs the mandatory (for those younger than 18 years) pre-licence Maryland Motorcycle Rider
Course (MRC) consisting of 8 hours theory and 12 hours practice
Intervention integrity
No information given
Outcomes Self-reported and police recorded crashes and offences
Length of follow up 1 year
38Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4150
Perrino 2002 (Continued)
Notes Study conducted in Maryland USA
Raymond 1979
Methods Cohort study
Participants Initial sample size
Intervention group
258 volunteers mainly (but not all) novice riders
Control group
1104 provisional licence applicants
Analysis sample size
Intervention group 190 (74)
Control group 627 (57)
Interventions No formal training vs the non-mandatory RACACU (Royal Automobile ClubAuto-Cycle Union) training scheme
The training program consisted of either 24 one-hour sessions or 13 two-hour sessions Twelve hours of the training
focused on theory about basic operational procedures traffic laws and highway code The practical session focused
on basic operational skills riding on a training area and riding in traffic with an instructor A proficiency test was
administered at the end of the course
Intervention integrity
No information given
Outcomes Self-reported crashes (including minor non-injury non-damage and off-road crashes)
Length of follow up 1 year
Notes Study conducted in the UK
Satten 1980
Methods Cohort study
Participants Initial sample size
Intervention group interviewers kept interviewing until they had 100 riders selected from a population of 891
enrollees in the MSF MRC course (unknown how many were asked and how many refused to participate)Control group unknown number of registered motorcycle riders
Non-active riders excluded from analysis
Analysis sample size
Intervention group 71
Control group 100
Interventions No formal training vs the non-mandatory Motorcycle Safety Foundationrsquos (MSF) Motorcycle Rider Course (MRC)
comprising of 8 hours theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-
road practice
Intervention integrity
Participants had all successfully completed the MRC
39Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4250
Satten 1980 (Continued)
Outcomes Self-reported crashes injuries and offences
Length of follow up 3 years
Notes Study conducted in Illinois USA
Savolainen 2007
Methods Cohort study
Participants Initial sample size
Intervention group 4000 riders that had conducted the MSF MRC course (passed or failed)
Control group 4000 motorcycle owners or permit or endorsement holders from the Indiana Bureau of Motor
Vehicles database
Analysis sample size
Intervention group 739 (18)
Control group 588 (15)
Interventions Intervention integrity
No information given
Outcomes Probability of crash involvement
Length of follow up up to 5 years
Notes Study conducted in Indiana USA
Swaddiwudhipong 1998
Methods Randomised trial
Participants Initial sample size
Intervention group
all 16 villages of 3 sub-districts
(participants were practising motorcyclists licensed and unlicensed)
Control group
18 villages of 3 other sub-districtsAnalysis sample size
Intervention group 3 districts (100)
Control group 3 districts (100)
Interventions No rider education vs non-mandatory community based rider education consisting of theory on motorcycle injury
risk helmet use traffic laws vehicle regulations and traffic signs and a written and skill test leading to a licence
Intervention integrity
No information given
Outcomes Injuries fatalities
Length of follow up 2 years
40Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4350
Swaddiwudhipong 1998 (Continued)
Notes Study conducted in Tak Province rural Thailand
Waller 1992
Methods Cohort study
Participants Initial sample size
Intervention group 349 riders who succesfully completed the Department of Public Safety motorcycle operator
training in 1988
Control group 349 motorcyclists who never completed the DPS motorcycle operator training
Analysis sample size
Intervention group 349
Control group 349
Interventions No DPS motorcycle operator training vs non-mandatory DPS motorcycle operator training The DPS training is
the Motorcycle Safety Foundationrsquos Motorcycle Rider Course comprising eight hours of theory (lectures discussions
films and multiple choice tests) a textbook and 12 hours off-road practice
Intervention integrity
All 349 riders succesfully completed the rider training
Outcomes Police recorded crashes
Length of follow up 3 years
Notes Study conducted in
Characteristics of excluded studies [ordered by study ID]
Study Reason for exclusion
Anderson 1980a Study type review and process evaluation of a rider course
Aupetit 2008 Outcome proposal of a rider training curriculum that combines traditional training and a motorcycle riding
simulator
Awane 1999 Study type review on use of simulators in rider and driver training
Baldi 2005 Intervention not rider training but intervention integrity of rider training availabilty of rider education
implementation of graduated licensing system etc on a state-wide level
Batchler 1988 Study type description of a training course
Blanchard 2006 Study type not a study assessing the association between rider training and outcomes
Bonnett 2005 Outcome qualitative data regarding satisfaction with the training programme
41Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4450
(Continued)
Braver 2007 Outcome licensure rates
Clayton 1980 Outcome knowledge skills attitude
Council 1978 Outcome crash reduction in ANY vehicle not motorcycles exclusively
Crowther 2005 Study type qualitative study
Daniello 2009 Study type review of the effectiveness of motorcycle training and licensing
Dewanti 2007 Intervention emotional characteristics behaviour (ie not motocycle rider training)
Du Plooy 1990 Study type review of types of rider training courses in South Africa
Engel 1989 Study type uncontrolled before-after study
Goldenbeld 2002 Outcome skills
Goldenbeld 2004 Outcome knowledge and skills
Haworth 2000 Intervention balance between attitudinal andvehicle control skills acrosscurriculla of motorcycle rider training
courses
Haworth 2005 Study type review
Haworth 2006 Study type cross-sectional survey
Hill 1983 Study type review and description of training system
Imatake 1980 Study type course descriptionreview
Ingebritsen 1990 Intervention risk factors for motorcycle crashes
Jain 2009 Study type cross-sectional descriptive study of motorcycle crashes over a 5 year period
Jonah 1979 Study type cross-sectional survey Outcome scores on motorcycle operator skill test
Jonah 1981 Intervention evaluation of a motorcycle rider skill test not motorcycle rider training
Kadar 1990 Study type review of components of rider training courses
Kelsey 1986 Intervention evaluation of a motorcycle rider skill test not motorcycle rider training
Koch 1980 Study type reviewcourse description
Koch 1990 Study type description of course content
42Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4550
(Continued)
Labbett 2005 Study type essay about possible explanations for motorcycle deaths
Law 2005 Intervention targeted motorcycle safety programme not specifically rider training
Lowes 1990 Study type review and descriptions of rider training
Mahon 1976 Study type review of motorcycle rider training methods
Maxwell 1985 Outcomesrecommendations forprogramme operatorsand otherson howthe motorcycle trainingprogramme
works and how it can be improved
Mayhew 1996 Study type review
McKnight 1987 Study type evaluation of methodological and administrative aspects of a motorcycle operator skill test
McPherson 1976 Outcome knowledge and skills
Mohan 2004 Study type review of evidence of interventions to reduce road traffic injuries
Morgan 1980 Study type reviewcourse description
Motoki 1990 Study type proposal and development of rider training curriculum
Nairn 1992 Study type literature review
Newland 1980 Study type description of development of motorcycle safety course
Newland 1982 Study type descriptionreview of a rider training course
Newland 1990 Study type uncontrolled BeforeAfter study
Ormston 2003 Outcome changes in attitude and self reported behaviour
Osga 1979 Outcome knowledge attitudes and skills
Panou 2007 Intervention new simulation tool to teach driverrider
Peterson 1990 Study factor demographic details of course participants completing the course vs those not completing the
course
Prem 1984 Study type experimental
Study factor examines association between skills and performance in tests
Radin Umar 2006 Intervention motorcycle safety programme not specifically rider training
Rockwell 1990 Study type methodologically flawed study type
43Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4650
(Continued)
Rothe 1987 Study type qualitative study of rider training in general
Rowden 2007 Study type qualitative study
RSC Victoria 1993 Study type review of motorcycle rider training and education in Victoria
Russam 1979 Study type review of training programmes in Japan
Saffron 1981 Study type literature review
Schulz 1990 Study factor predictors of motorcycle crashes
Sexton 2004 Study type cross-sectional analytic analysis
Shephard 1986 Study factor reviewdescription of programme
Simpson 1990 Study type review of motorcycle rider trainingeducation
Sokolov 1990 Study factor review of motorcycle riding courses in USSR
Steffens 1988 Outcome self-reported behaviour experiences and researcher observed riding behaviour
Sudlow 2003 Study type review of motorcycle training in the UK
Tasmania 1995 Study type cross-sectional descriptiveprocess evaluation
Tetard 1985 Study type experimental
Outcome detection of indices
Thackray 1980 Outcome instructional effectiveness user acceptance administrative feasibility
Thompson 1994 Study type cross-sectional survey
Outcome attitudes and knowledge
Tobler 1990 Study type review of course curricula
Transport 1989 Intervention syllabus for rider training
Tuan 2007 Study type cross-sectional study (survey) which does not assess the effects of an intervention but examines
motocyclistsrsquo behaviour
Victor 1980 Study type review
Waterer 1990 Study type review of UK rider training issues
Weller 1989 Study type review of studies evaluating motorcycle safety and education measures
44Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4750
(Continued)
Wells 1981 Study type description of different methods of training
Winn 1986 Study type review of rsquohistoryrsquo of rider training
Wisher 1988 Study type cross-sectional survey
Wood 1987 Study type Interrupted Time Series but did not meet the criteria for an ITS as published by the Cochrane
Effective Practice and Organisation of Care Group
Yamazaki 1990 Study type cross-sectional survey
Outcome riding skills skills retention and attitude
Characteristics of ongoing studies [ordered by study ID]
Burgess 2010
Trial name or title UKs National Ride program
Methods Cohort study
Participants Motorcycle offenders
Interventions Police led diversion program
Outcomes Self reported attitudinal and behavioural measures
Starting date
Contact information
Notes Expected to be completed in early 2010
45Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4850
D A T A A N D A N A L Y S E S
This review has no analyses
A P P E N D I C E S
Appendix 1 Search strategy
We searched the following electronic databases
bull Cochrane Injuries Group Specialised Register (searched 22 Sept 2008)
bull CENTRAL (The Cochrane Library 2008 Issue 3)
bull TRANSPORT 1988 to 200706 (latest date available)
bull ATRI Australian Transport Index (to Sept 2008)
bull MEDLINE (Ovid SP) 1950 to Sept 2008
bull EMBASE (Ovid SP) 1980 to Sept 2008
bull CINAHL (EBSCO) 1982 to Sept 2008
bull WHOLIS (World Health Organization Library Information System) (to Sept 2008)
bull PsycINFO (Ovid SP) 1806 to Sept 2008
bull LILACS (BIREME) (Latin American and Caribbean Health Sciences) (1982 to Sept 2008)
bull ISI Web of Science Social Sciences Citation Index (SSCI) 1970 to Sept 2008
bull ERIC (searched 25 Sept 2008)
bull ZETOC (searched 25 Sept 2008)
bull SIGLE (searched 25 Sept 2008)
All of the databases were searched using the following text-word search strategy The strategy was adapted using either rsquorsquo or rsquo$rsquo as atruncation symbol
1 motor-cycl or motorcycl or motor-bik or motorbik or motor-scooter or motorscooter or scooter or moped or motocycl
or motocicl or moto
2 train or educat or instruct or program or course or qualif or learn or improv
3 1 and 2
H I S T O R Y
Protocol first published Issue 2 2005
Review first published Issue 10 2010
46Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4950
C O N T R I B U T I O N S O F A U T H O R S
Katina Kardamanidis was the first author of the protocol conducted the literature search extracted full studies assessed the studies
according to inclusion criteria assessed quality of the studies compared findings with the second author discussed disagreements with
the third author wrote the drafts and the final review
Rebecca Ivers was the second author of the protocol assessed the studies according to inclusion criteria assessed quality of the studies
compared findings with the first author edited the final draft and provided overall guidance for the process
Alexandra Martiniuk assessed quality of the studies compared findings with the first author discussed disagreements with the third
author and helped write the first draft and the final review
Mark Stevenson contributed to the protocol reviewed methods of studies and contributed to the final draft
Katrina Thistlethwaite conducted a second literature search extracted full studies and contributed to the final draft
D E C L A R A T I O N S O F I N T E R E S T
None known
S O U R C E S O F S U P P O R T
Internal sources
bull No sources of support supplied
External sources
bull National Health and Medical Research Council of Australia Australia
Salary support for Martiniuk Ivers and Stevenson
bull University of Sydney Sesqui Research Grant Australia
Direct research costs and salary support for Kardamanidis
D I F F E R E N C E S B E T W E E N P R O T O C O L A N D R E V I E W
There are no differences between the protocol and the review
I N D E X T E R M S
Medical Subject Headings (MeSH)
lowastMotorcycles Accident Prevention [lowastmethods] Accidents Traffic [lowastprevention amp control] Licensure Program Evaluation Wounds
and Injuries [mortality lowastprevention amp control]
47Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 5050
MeSH check words
Humans
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3450
Davis 1997
Methods Retrospective cohort study
Participants Initial sample size
Intervention group 9320 graduates
Control group unclear
Analysis sample size
Intervention group 9320 graduates
Control group unclear
Interventions No formal training vs the non-mandatory Connecticut Motorcycle Operator Training Program The content of
training was not further explained
Intervention integrity Participants had all completed the rider training
Outcomes Police recorded crashes severity of crashes
Length of follow up 1 year
Notes Study conducted in Connecticut USA
Hall 1988
Methods Randomised trial
Participants Initial sample sizeIntervention group A 6248
Intervention group B 6616
Intervention group C 6996
Control group 6604
Total 26464 motorcycle licence applicants
Analysis sample size
Intervention group A 6012 (96)
Intervention group B 6428 (97)
Intervention group C 6778 (97)
Control group 6502 (98)
Interventions Standard licensing skill test vs the Motorcycle Operator Skill Test (MOST II) without training (A) MOST II with
3 hours training (B) or MOST II with 20 hours training (C) The 3 hours training comprised instructions aboutthe skills needed to pass the skill test (30 min theory and 2 frac12 hours practice on the motorcycle) The 20 hours
training consisted of the Motorcycle Safety Foundationrsquos Motorcycle Rider Course (MRC) comprising of 8 hours
theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-road practice
Intervention integrity
25 and 26 obtained their motorcycle licence in the lsquoMOST 3-hour trainingrsquo and lsquoMOST 20-hour trainingrsquo
intervention arm respectively In the lsquoMOST testrsquo intervention arm 29 obtained their licence and in the control
group 35
Outcomes Police recorded crashes and offences
Length of follow up 3 years
32Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3550
Hall 1988 (Continued)
Notes Study conducted in New York USA
Haworth 1997
Methods Population based case-control study
Participants Cases Motorcycle riders or pillion riders involved in motorcycle crash (on road) Recruited from hospital and police
records Not clear whether all cases (especially fatalities) were included
Community controls Multiple matched sample of motorcycle rider or pillion who passed the crash site at the same
time of day and day of week that the crash occurred
Initial sample size
Cases 336
Controls 1195
Analysis sample size
Cases 234 (70)
Controls 561 (47)
Interventions Basic handling skills course beginner intermediate or advanced course measured through self-report not validated
Outcomes Injuries with an Injury Severity Score of 5 or greater or admission for at least four hours and fatalities from a
motorcycle crash (from hospital and police records)
Length of follow up two years or more prior to the interview
Notes Study conducted in Victoria Australia
Hurt 1981
Methods Population based case-control study
Participants Cases Motorcycle riders who had a motorcycle crash that was reported to police (traffic accident reports) Not all
cases included
Community controls Multiple matched sample of motorcycle riders who passed the crash site at the same time of
day and day of week as the crash occurred but 2 years later
Initial sample size
Cases 4500 notifications of crashes to police (only 50 received in time research team able to respond to 1126 for
unreported reasons data collection completed for 900)
Controls Number approached unreported 2310 consented and interviewed
Analysis sample size
Cases 808 (808900 data on rider training) - 18 of initial sample
Controls 670 (6702310 data on rider training) - lt 29 of initial sample
Interventions Training experience self taught friends-family school-club formal-professional others measured through self-
report validated
Outcomes Injuries and fatalities from a motorcycle crash (from hospital records and coronerrsquos office)
33Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3650
Hurt 1981 (Continued)
Notes Study conducted in Los Angeles USA
Jonah 1982
Methods Cohort study
Participants Initial sample size
Intervention group
2310 licensed MTP graduates who are active riders
Control group
2029 licensed motorcyclists who are active riders
Non-licensed and non-active riders excluded
Analysis sample size
Intervention group
811 interviews (35) 533 driver records (23)
Control group 1080 interviews (71) 905 driver records (45)
Interventions No training vs the non-mandatory Canada Safety Council Motorcycle Training Program (MTP) comprising of 4
hours theory and 16 hours practise (6 hrs on-road) over 2 weekends
Intervention integrity
Participants had all successfully completed the MTP
Outcomes Self-reported and police reported crashes and offencesLength of follow up 4 years
Notes Study conducted in Ontario Canada
Kloeden 1994
Methods Cohort study
Participants Initial sample size
Intervention group 2533 licence applicants
Control group Unclear
Sample of riders obtaining probationary licenceIntervention group 1625 (64)
Control group 5015 (unclear)
Non-active riders excluded from control group (~66)
Analysis sample size
Intervention group 631 (39)
Control group 318 (6)
Interventions No formal training vs the mandatory pre-licence Ridersafe course consisting of two levelsthe first is a training course
in the basics of motorcycle riding where a pass entitles the rider to obtain a learnerrsquos permit The second is a skill
test where a pass entitles the rider to obtain a probationary motorcycle licence The training course consists of two 4
hour training sessions on separate days involving instruction videos and actual riding on motorcycles The skill test
consists of a 4 hour preparation course of 74 minutes of lecturediscussion 38 minutes of video and 75 minutes of
34Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3750
Kloeden 1994 (Continued)
on-course riding which is the testing phase
Intervention integrity
Unclear what percentage in the Ridersafe group and the control group obtained their probationary licence halfway
and their full motorcycle licence in the end
Outcomes Police reported crashes
Length of follow up 1 year
Notes Study conducted in SA Australia
Kraus 1975
Methods Population based case-control study
Participants Cases All motorcycle riders with a confirmed medically treated motorcycle injury (including fatalities) from all
hospitals in study base and peripheral area hospitals and all deaths due to motorcycle crashes in the study base All
serious injuries (leading to hospitalisation) and fatalities likely to be included
Community controls A random sample of registered motorcycle owners in the same temporary and geographical
study base as the cases arose from
Initial sample size Cases 1273
Controls 738
Analysis sample size Cases 741 (58)
Controls 434 (59)
Interventions Motorcycle rider training (not specified) measured through self-report
Outcomes Injuries leading to hospitalisation fatalities
Length of follow up not specified
Notes Study conducted in Sacramento county USA
Leung 1987
Methods Non-randomised trial
Participants Initial sample sizeIntervention group unknown number of MRC graduates
Control group unknown number of licensed motorcyclists
Analysis sample size
Intervention group 2941
Control group 43094
Interventions No formal training vs non-mandatory Wisconsin motorcycle rider course a 22 hours off-street programme that
teaches the basic rules of safe motorcycling
Intervention integrity
50 in the MRC group and 100 in the control group obtained their motorcycle licence in the end
35Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3850
Leung 1987 (Continued)
Outcomes Police recorded crashes and offences
Length of follow up up to 3 years
Notes Study conducted in Wisconsin USA
MAIDS 2004
Methods Population based case-control study
Participants Cases Motorcycle riders who had a motorcycle crash Unclear whether all cases included
Community controls Motorcycle riders identified at petrol stations Unclear whether matched to cases
Initial sample size
Cases not reported
Controls not reported
Analysis sample size
Cases 921 (Italy 200 Netherlands 200 Spain 121 Germany 250 France 150)
Controls 923 (Italy 200 Netherlands 200 Spain123 Germany 250 France 150)
Interventions Pre-license rider training or additional training (non-compulsory)
Outcomes Crashes
Notes Study in five sampling areas located in five European countries (France Germany Netherlands Spain and Italy)
McDavid 1989
Methods Cohort study
Participants Sample size after first match
Intervention group 346 male training graduates
Control group 346 male motorcycle riders licensed in same year
Sample size after second match
Intervention group 139 (40)
Control group 139 (40)
Analysis sample size
Intervention group 139 (40)
Control group 139 (40)
Interventions No formal training vs the non-mandatory pre-licence British Columbia Safety Councilrsquos 37-hour motorcycle safety
programme
Intervention integrity
Participants had all completed and passed the training
Outcomes Police reported crashes injurious crashes
Length of follow up 1-6 years
36Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3950
McDavid 1989 (Continued)
Notes Study conducted in British Columbia Canada
Mortimer 1984
Methods Cohort study
Participants Initial sample size
Intervention group 608 MRC graduates
Control group 360 motorcycle riders
Response rates
Intervention group 360 (59)
Control group 324 (90)
Non-active riders excluded
Analysis sample size
Intervention group 213 (35)
Control group 303 (84)
Interventions No formal training vs the non-mandatory Motorcycle Safety Foundationrsquos Motorcycle Rider Course comprising of
8 hours theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-road practice
Intervention integrity
Participants had all completed and passed the MRC
Outcomes Self-reported crashes and offencesLength of follow up 1 year
Notes Study conducted in Illinois USA
Mortimer 1988
Methods Cohort study
Participants Initial sample size
Intervention group 2307 MRC graduates
Control group 556 active licensed motorcycle riders
Response ratesIntervention group 1298 (56)
Control group 500 (90)
Non-active riders excluded
Analysis sample size
Intervention group 913 (40)
Control group 500 (90)
Interventions No formal training vs the non-mandatory Motorcycle Safety Foundationrsquos Motorcycle Rider Course comprising of
8 hours theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-road practice
Intervention integrity
Participants had all completed and passed the MRC
37Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4050
Mortimer 1988 (Continued)
Outcomes Self reported crashes and offences
Length of follow up 1 year
Notes Study conducted in Illinois USA
Osga 1980
Methods Cohort study
Participants Initial sample sizeIntervention group 1163 MSF MRC participants invited 818 (74) responded
Control group 329 motorcycle riders responded to recruitment strategies
Non-active riders excluded from analysis
Analysis sample size
Intervention group 595 (73)
Control group 329 (100)
Interventions No formal training vs the non-mandatory Motorcycle Safety Foundationrsquos Motorcycle Rider Course (MRC) com-
prising of 8 hours theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-road
practice
Intervention integrity
Participants had all successfully completed the MRC
Outcomes Self-reported crashes police recorded crash reports
Length of follow up 6 months - 3 12 years
Notes Study conducted in South Dakota USA
Perrino 2002
Methods Cohort study
Participants Initial sample size
Intervention group 2144 (unclear) MRC graduates (motorcycle licence applicants)
Control group 2601 (unclear) regular licensing test graduates (motorcycle licence applicants)Analysis sample size
Intervention group 1070 (unclear)
Control group 867 (unclear)
Interventions No formal training vs the mandatory (for those younger than 18 years) pre-licence Maryland Motorcycle Rider
Course (MRC) consisting of 8 hours theory and 12 hours practice
Intervention integrity
No information given
Outcomes Self-reported and police recorded crashes and offences
Length of follow up 1 year
38Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4150
Perrino 2002 (Continued)
Notes Study conducted in Maryland USA
Raymond 1979
Methods Cohort study
Participants Initial sample size
Intervention group
258 volunteers mainly (but not all) novice riders
Control group
1104 provisional licence applicants
Analysis sample size
Intervention group 190 (74)
Control group 627 (57)
Interventions No formal training vs the non-mandatory RACACU (Royal Automobile ClubAuto-Cycle Union) training scheme
The training program consisted of either 24 one-hour sessions or 13 two-hour sessions Twelve hours of the training
focused on theory about basic operational procedures traffic laws and highway code The practical session focused
on basic operational skills riding on a training area and riding in traffic with an instructor A proficiency test was
administered at the end of the course
Intervention integrity
No information given
Outcomes Self-reported crashes (including minor non-injury non-damage and off-road crashes)
Length of follow up 1 year
Notes Study conducted in the UK
Satten 1980
Methods Cohort study
Participants Initial sample size
Intervention group interviewers kept interviewing until they had 100 riders selected from a population of 891
enrollees in the MSF MRC course (unknown how many were asked and how many refused to participate)Control group unknown number of registered motorcycle riders
Non-active riders excluded from analysis
Analysis sample size
Intervention group 71
Control group 100
Interventions No formal training vs the non-mandatory Motorcycle Safety Foundationrsquos (MSF) Motorcycle Rider Course (MRC)
comprising of 8 hours theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-
road practice
Intervention integrity
Participants had all successfully completed the MRC
39Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4250
Satten 1980 (Continued)
Outcomes Self-reported crashes injuries and offences
Length of follow up 3 years
Notes Study conducted in Illinois USA
Savolainen 2007
Methods Cohort study
Participants Initial sample size
Intervention group 4000 riders that had conducted the MSF MRC course (passed or failed)
Control group 4000 motorcycle owners or permit or endorsement holders from the Indiana Bureau of Motor
Vehicles database
Analysis sample size
Intervention group 739 (18)
Control group 588 (15)
Interventions Intervention integrity
No information given
Outcomes Probability of crash involvement
Length of follow up up to 5 years
Notes Study conducted in Indiana USA
Swaddiwudhipong 1998
Methods Randomised trial
Participants Initial sample size
Intervention group
all 16 villages of 3 sub-districts
(participants were practising motorcyclists licensed and unlicensed)
Control group
18 villages of 3 other sub-districtsAnalysis sample size
Intervention group 3 districts (100)
Control group 3 districts (100)
Interventions No rider education vs non-mandatory community based rider education consisting of theory on motorcycle injury
risk helmet use traffic laws vehicle regulations and traffic signs and a written and skill test leading to a licence
Intervention integrity
No information given
Outcomes Injuries fatalities
Length of follow up 2 years
40Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4350
Swaddiwudhipong 1998 (Continued)
Notes Study conducted in Tak Province rural Thailand
Waller 1992
Methods Cohort study
Participants Initial sample size
Intervention group 349 riders who succesfully completed the Department of Public Safety motorcycle operator
training in 1988
Control group 349 motorcyclists who never completed the DPS motorcycle operator training
Analysis sample size
Intervention group 349
Control group 349
Interventions No DPS motorcycle operator training vs non-mandatory DPS motorcycle operator training The DPS training is
the Motorcycle Safety Foundationrsquos Motorcycle Rider Course comprising eight hours of theory (lectures discussions
films and multiple choice tests) a textbook and 12 hours off-road practice
Intervention integrity
All 349 riders succesfully completed the rider training
Outcomes Police recorded crashes
Length of follow up 3 years
Notes Study conducted in
Characteristics of excluded studies [ordered by study ID]
Study Reason for exclusion
Anderson 1980a Study type review and process evaluation of a rider course
Aupetit 2008 Outcome proposal of a rider training curriculum that combines traditional training and a motorcycle riding
simulator
Awane 1999 Study type review on use of simulators in rider and driver training
Baldi 2005 Intervention not rider training but intervention integrity of rider training availabilty of rider education
implementation of graduated licensing system etc on a state-wide level
Batchler 1988 Study type description of a training course
Blanchard 2006 Study type not a study assessing the association between rider training and outcomes
Bonnett 2005 Outcome qualitative data regarding satisfaction with the training programme
41Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4450
(Continued)
Braver 2007 Outcome licensure rates
Clayton 1980 Outcome knowledge skills attitude
Council 1978 Outcome crash reduction in ANY vehicle not motorcycles exclusively
Crowther 2005 Study type qualitative study
Daniello 2009 Study type review of the effectiveness of motorcycle training and licensing
Dewanti 2007 Intervention emotional characteristics behaviour (ie not motocycle rider training)
Du Plooy 1990 Study type review of types of rider training courses in South Africa
Engel 1989 Study type uncontrolled before-after study
Goldenbeld 2002 Outcome skills
Goldenbeld 2004 Outcome knowledge and skills
Haworth 2000 Intervention balance between attitudinal andvehicle control skills acrosscurriculla of motorcycle rider training
courses
Haworth 2005 Study type review
Haworth 2006 Study type cross-sectional survey
Hill 1983 Study type review and description of training system
Imatake 1980 Study type course descriptionreview
Ingebritsen 1990 Intervention risk factors for motorcycle crashes
Jain 2009 Study type cross-sectional descriptive study of motorcycle crashes over a 5 year period
Jonah 1979 Study type cross-sectional survey Outcome scores on motorcycle operator skill test
Jonah 1981 Intervention evaluation of a motorcycle rider skill test not motorcycle rider training
Kadar 1990 Study type review of components of rider training courses
Kelsey 1986 Intervention evaluation of a motorcycle rider skill test not motorcycle rider training
Koch 1980 Study type reviewcourse description
Koch 1990 Study type description of course content
42Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4550
(Continued)
Labbett 2005 Study type essay about possible explanations for motorcycle deaths
Law 2005 Intervention targeted motorcycle safety programme not specifically rider training
Lowes 1990 Study type review and descriptions of rider training
Mahon 1976 Study type review of motorcycle rider training methods
Maxwell 1985 Outcomesrecommendations forprogramme operatorsand otherson howthe motorcycle trainingprogramme
works and how it can be improved
Mayhew 1996 Study type review
McKnight 1987 Study type evaluation of methodological and administrative aspects of a motorcycle operator skill test
McPherson 1976 Outcome knowledge and skills
Mohan 2004 Study type review of evidence of interventions to reduce road traffic injuries
Morgan 1980 Study type reviewcourse description
Motoki 1990 Study type proposal and development of rider training curriculum
Nairn 1992 Study type literature review
Newland 1980 Study type description of development of motorcycle safety course
Newland 1982 Study type descriptionreview of a rider training course
Newland 1990 Study type uncontrolled BeforeAfter study
Ormston 2003 Outcome changes in attitude and self reported behaviour
Osga 1979 Outcome knowledge attitudes and skills
Panou 2007 Intervention new simulation tool to teach driverrider
Peterson 1990 Study factor demographic details of course participants completing the course vs those not completing the
course
Prem 1984 Study type experimental
Study factor examines association between skills and performance in tests
Radin Umar 2006 Intervention motorcycle safety programme not specifically rider training
Rockwell 1990 Study type methodologically flawed study type
43Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4650
(Continued)
Rothe 1987 Study type qualitative study of rider training in general
Rowden 2007 Study type qualitative study
RSC Victoria 1993 Study type review of motorcycle rider training and education in Victoria
Russam 1979 Study type review of training programmes in Japan
Saffron 1981 Study type literature review
Schulz 1990 Study factor predictors of motorcycle crashes
Sexton 2004 Study type cross-sectional analytic analysis
Shephard 1986 Study factor reviewdescription of programme
Simpson 1990 Study type review of motorcycle rider trainingeducation
Sokolov 1990 Study factor review of motorcycle riding courses in USSR
Steffens 1988 Outcome self-reported behaviour experiences and researcher observed riding behaviour
Sudlow 2003 Study type review of motorcycle training in the UK
Tasmania 1995 Study type cross-sectional descriptiveprocess evaluation
Tetard 1985 Study type experimental
Outcome detection of indices
Thackray 1980 Outcome instructional effectiveness user acceptance administrative feasibility
Thompson 1994 Study type cross-sectional survey
Outcome attitudes and knowledge
Tobler 1990 Study type review of course curricula
Transport 1989 Intervention syllabus for rider training
Tuan 2007 Study type cross-sectional study (survey) which does not assess the effects of an intervention but examines
motocyclistsrsquo behaviour
Victor 1980 Study type review
Waterer 1990 Study type review of UK rider training issues
Weller 1989 Study type review of studies evaluating motorcycle safety and education measures
44Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4750
(Continued)
Wells 1981 Study type description of different methods of training
Winn 1986 Study type review of rsquohistoryrsquo of rider training
Wisher 1988 Study type cross-sectional survey
Wood 1987 Study type Interrupted Time Series but did not meet the criteria for an ITS as published by the Cochrane
Effective Practice and Organisation of Care Group
Yamazaki 1990 Study type cross-sectional survey
Outcome riding skills skills retention and attitude
Characteristics of ongoing studies [ordered by study ID]
Burgess 2010
Trial name or title UKs National Ride program
Methods Cohort study
Participants Motorcycle offenders
Interventions Police led diversion program
Outcomes Self reported attitudinal and behavioural measures
Starting date
Contact information
Notes Expected to be completed in early 2010
45Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4850
D A T A A N D A N A L Y S E S
This review has no analyses
A P P E N D I C E S
Appendix 1 Search strategy
We searched the following electronic databases
bull Cochrane Injuries Group Specialised Register (searched 22 Sept 2008)
bull CENTRAL (The Cochrane Library 2008 Issue 3)
bull TRANSPORT 1988 to 200706 (latest date available)
bull ATRI Australian Transport Index (to Sept 2008)
bull MEDLINE (Ovid SP) 1950 to Sept 2008
bull EMBASE (Ovid SP) 1980 to Sept 2008
bull CINAHL (EBSCO) 1982 to Sept 2008
bull WHOLIS (World Health Organization Library Information System) (to Sept 2008)
bull PsycINFO (Ovid SP) 1806 to Sept 2008
bull LILACS (BIREME) (Latin American and Caribbean Health Sciences) (1982 to Sept 2008)
bull ISI Web of Science Social Sciences Citation Index (SSCI) 1970 to Sept 2008
bull ERIC (searched 25 Sept 2008)
bull ZETOC (searched 25 Sept 2008)
bull SIGLE (searched 25 Sept 2008)
All of the databases were searched using the following text-word search strategy The strategy was adapted using either rsquorsquo or rsquo$rsquo as atruncation symbol
1 motor-cycl or motorcycl or motor-bik or motorbik or motor-scooter or motorscooter or scooter or moped or motocycl
or motocicl or moto
2 train or educat or instruct or program or course or qualif or learn or improv
3 1 and 2
H I S T O R Y
Protocol first published Issue 2 2005
Review first published Issue 10 2010
46Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4950
C O N T R I B U T I O N S O F A U T H O R S
Katina Kardamanidis was the first author of the protocol conducted the literature search extracted full studies assessed the studies
according to inclusion criteria assessed quality of the studies compared findings with the second author discussed disagreements with
the third author wrote the drafts and the final review
Rebecca Ivers was the second author of the protocol assessed the studies according to inclusion criteria assessed quality of the studies
compared findings with the first author edited the final draft and provided overall guidance for the process
Alexandra Martiniuk assessed quality of the studies compared findings with the first author discussed disagreements with the third
author and helped write the first draft and the final review
Mark Stevenson contributed to the protocol reviewed methods of studies and contributed to the final draft
Katrina Thistlethwaite conducted a second literature search extracted full studies and contributed to the final draft
D E C L A R A T I O N S O F I N T E R E S T
None known
S O U R C E S O F S U P P O R T
Internal sources
bull No sources of support supplied
External sources
bull National Health and Medical Research Council of Australia Australia
Salary support for Martiniuk Ivers and Stevenson
bull University of Sydney Sesqui Research Grant Australia
Direct research costs and salary support for Kardamanidis
D I F F E R E N C E S B E T W E E N P R O T O C O L A N D R E V I E W
There are no differences between the protocol and the review
I N D E X T E R M S
Medical Subject Headings (MeSH)
lowastMotorcycles Accident Prevention [lowastmethods] Accidents Traffic [lowastprevention amp control] Licensure Program Evaluation Wounds
and Injuries [mortality lowastprevention amp control]
47Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 5050
MeSH check words
Humans
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3550
Hall 1988 (Continued)
Notes Study conducted in New York USA
Haworth 1997
Methods Population based case-control study
Participants Cases Motorcycle riders or pillion riders involved in motorcycle crash (on road) Recruited from hospital and police
records Not clear whether all cases (especially fatalities) were included
Community controls Multiple matched sample of motorcycle rider or pillion who passed the crash site at the same
time of day and day of week that the crash occurred
Initial sample size
Cases 336
Controls 1195
Analysis sample size
Cases 234 (70)
Controls 561 (47)
Interventions Basic handling skills course beginner intermediate or advanced course measured through self-report not validated
Outcomes Injuries with an Injury Severity Score of 5 or greater or admission for at least four hours and fatalities from a
motorcycle crash (from hospital and police records)
Length of follow up two years or more prior to the interview
Notes Study conducted in Victoria Australia
Hurt 1981
Methods Population based case-control study
Participants Cases Motorcycle riders who had a motorcycle crash that was reported to police (traffic accident reports) Not all
cases included
Community controls Multiple matched sample of motorcycle riders who passed the crash site at the same time of
day and day of week as the crash occurred but 2 years later
Initial sample size
Cases 4500 notifications of crashes to police (only 50 received in time research team able to respond to 1126 for
unreported reasons data collection completed for 900)
Controls Number approached unreported 2310 consented and interviewed
Analysis sample size
Cases 808 (808900 data on rider training) - 18 of initial sample
Controls 670 (6702310 data on rider training) - lt 29 of initial sample
Interventions Training experience self taught friends-family school-club formal-professional others measured through self-
report validated
Outcomes Injuries and fatalities from a motorcycle crash (from hospital records and coronerrsquos office)
33Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3650
Hurt 1981 (Continued)
Notes Study conducted in Los Angeles USA
Jonah 1982
Methods Cohort study
Participants Initial sample size
Intervention group
2310 licensed MTP graduates who are active riders
Control group
2029 licensed motorcyclists who are active riders
Non-licensed and non-active riders excluded
Analysis sample size
Intervention group
811 interviews (35) 533 driver records (23)
Control group 1080 interviews (71) 905 driver records (45)
Interventions No training vs the non-mandatory Canada Safety Council Motorcycle Training Program (MTP) comprising of 4
hours theory and 16 hours practise (6 hrs on-road) over 2 weekends
Intervention integrity
Participants had all successfully completed the MTP
Outcomes Self-reported and police reported crashes and offencesLength of follow up 4 years
Notes Study conducted in Ontario Canada
Kloeden 1994
Methods Cohort study
Participants Initial sample size
Intervention group 2533 licence applicants
Control group Unclear
Sample of riders obtaining probationary licenceIntervention group 1625 (64)
Control group 5015 (unclear)
Non-active riders excluded from control group (~66)
Analysis sample size
Intervention group 631 (39)
Control group 318 (6)
Interventions No formal training vs the mandatory pre-licence Ridersafe course consisting of two levelsthe first is a training course
in the basics of motorcycle riding where a pass entitles the rider to obtain a learnerrsquos permit The second is a skill
test where a pass entitles the rider to obtain a probationary motorcycle licence The training course consists of two 4
hour training sessions on separate days involving instruction videos and actual riding on motorcycles The skill test
consists of a 4 hour preparation course of 74 minutes of lecturediscussion 38 minutes of video and 75 minutes of
34Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3750
Kloeden 1994 (Continued)
on-course riding which is the testing phase
Intervention integrity
Unclear what percentage in the Ridersafe group and the control group obtained their probationary licence halfway
and their full motorcycle licence in the end
Outcomes Police reported crashes
Length of follow up 1 year
Notes Study conducted in SA Australia
Kraus 1975
Methods Population based case-control study
Participants Cases All motorcycle riders with a confirmed medically treated motorcycle injury (including fatalities) from all
hospitals in study base and peripheral area hospitals and all deaths due to motorcycle crashes in the study base All
serious injuries (leading to hospitalisation) and fatalities likely to be included
Community controls A random sample of registered motorcycle owners in the same temporary and geographical
study base as the cases arose from
Initial sample size Cases 1273
Controls 738
Analysis sample size Cases 741 (58)
Controls 434 (59)
Interventions Motorcycle rider training (not specified) measured through self-report
Outcomes Injuries leading to hospitalisation fatalities
Length of follow up not specified
Notes Study conducted in Sacramento county USA
Leung 1987
Methods Non-randomised trial
Participants Initial sample sizeIntervention group unknown number of MRC graduates
Control group unknown number of licensed motorcyclists
Analysis sample size
Intervention group 2941
Control group 43094
Interventions No formal training vs non-mandatory Wisconsin motorcycle rider course a 22 hours off-street programme that
teaches the basic rules of safe motorcycling
Intervention integrity
50 in the MRC group and 100 in the control group obtained their motorcycle licence in the end
35Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3850
Leung 1987 (Continued)
Outcomes Police recorded crashes and offences
Length of follow up up to 3 years
Notes Study conducted in Wisconsin USA
MAIDS 2004
Methods Population based case-control study
Participants Cases Motorcycle riders who had a motorcycle crash Unclear whether all cases included
Community controls Motorcycle riders identified at petrol stations Unclear whether matched to cases
Initial sample size
Cases not reported
Controls not reported
Analysis sample size
Cases 921 (Italy 200 Netherlands 200 Spain 121 Germany 250 France 150)
Controls 923 (Italy 200 Netherlands 200 Spain123 Germany 250 France 150)
Interventions Pre-license rider training or additional training (non-compulsory)
Outcomes Crashes
Notes Study in five sampling areas located in five European countries (France Germany Netherlands Spain and Italy)
McDavid 1989
Methods Cohort study
Participants Sample size after first match
Intervention group 346 male training graduates
Control group 346 male motorcycle riders licensed in same year
Sample size after second match
Intervention group 139 (40)
Control group 139 (40)
Analysis sample size
Intervention group 139 (40)
Control group 139 (40)
Interventions No formal training vs the non-mandatory pre-licence British Columbia Safety Councilrsquos 37-hour motorcycle safety
programme
Intervention integrity
Participants had all completed and passed the training
Outcomes Police reported crashes injurious crashes
Length of follow up 1-6 years
36Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3950
McDavid 1989 (Continued)
Notes Study conducted in British Columbia Canada
Mortimer 1984
Methods Cohort study
Participants Initial sample size
Intervention group 608 MRC graduates
Control group 360 motorcycle riders
Response rates
Intervention group 360 (59)
Control group 324 (90)
Non-active riders excluded
Analysis sample size
Intervention group 213 (35)
Control group 303 (84)
Interventions No formal training vs the non-mandatory Motorcycle Safety Foundationrsquos Motorcycle Rider Course comprising of
8 hours theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-road practice
Intervention integrity
Participants had all completed and passed the MRC
Outcomes Self-reported crashes and offencesLength of follow up 1 year
Notes Study conducted in Illinois USA
Mortimer 1988
Methods Cohort study
Participants Initial sample size
Intervention group 2307 MRC graduates
Control group 556 active licensed motorcycle riders
Response ratesIntervention group 1298 (56)
Control group 500 (90)
Non-active riders excluded
Analysis sample size
Intervention group 913 (40)
Control group 500 (90)
Interventions No formal training vs the non-mandatory Motorcycle Safety Foundationrsquos Motorcycle Rider Course comprising of
8 hours theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-road practice
Intervention integrity
Participants had all completed and passed the MRC
37Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4050
Mortimer 1988 (Continued)
Outcomes Self reported crashes and offences
Length of follow up 1 year
Notes Study conducted in Illinois USA
Osga 1980
Methods Cohort study
Participants Initial sample sizeIntervention group 1163 MSF MRC participants invited 818 (74) responded
Control group 329 motorcycle riders responded to recruitment strategies
Non-active riders excluded from analysis
Analysis sample size
Intervention group 595 (73)
Control group 329 (100)
Interventions No formal training vs the non-mandatory Motorcycle Safety Foundationrsquos Motorcycle Rider Course (MRC) com-
prising of 8 hours theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-road
practice
Intervention integrity
Participants had all successfully completed the MRC
Outcomes Self-reported crashes police recorded crash reports
Length of follow up 6 months - 3 12 years
Notes Study conducted in South Dakota USA
Perrino 2002
Methods Cohort study
Participants Initial sample size
Intervention group 2144 (unclear) MRC graduates (motorcycle licence applicants)
Control group 2601 (unclear) regular licensing test graduates (motorcycle licence applicants)Analysis sample size
Intervention group 1070 (unclear)
Control group 867 (unclear)
Interventions No formal training vs the mandatory (for those younger than 18 years) pre-licence Maryland Motorcycle Rider
Course (MRC) consisting of 8 hours theory and 12 hours practice
Intervention integrity
No information given
Outcomes Self-reported and police recorded crashes and offences
Length of follow up 1 year
38Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4150
Perrino 2002 (Continued)
Notes Study conducted in Maryland USA
Raymond 1979
Methods Cohort study
Participants Initial sample size
Intervention group
258 volunteers mainly (but not all) novice riders
Control group
1104 provisional licence applicants
Analysis sample size
Intervention group 190 (74)
Control group 627 (57)
Interventions No formal training vs the non-mandatory RACACU (Royal Automobile ClubAuto-Cycle Union) training scheme
The training program consisted of either 24 one-hour sessions or 13 two-hour sessions Twelve hours of the training
focused on theory about basic operational procedures traffic laws and highway code The practical session focused
on basic operational skills riding on a training area and riding in traffic with an instructor A proficiency test was
administered at the end of the course
Intervention integrity
No information given
Outcomes Self-reported crashes (including minor non-injury non-damage and off-road crashes)
Length of follow up 1 year
Notes Study conducted in the UK
Satten 1980
Methods Cohort study
Participants Initial sample size
Intervention group interviewers kept interviewing until they had 100 riders selected from a population of 891
enrollees in the MSF MRC course (unknown how many were asked and how many refused to participate)Control group unknown number of registered motorcycle riders
Non-active riders excluded from analysis
Analysis sample size
Intervention group 71
Control group 100
Interventions No formal training vs the non-mandatory Motorcycle Safety Foundationrsquos (MSF) Motorcycle Rider Course (MRC)
comprising of 8 hours theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-
road practice
Intervention integrity
Participants had all successfully completed the MRC
39Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4250
Satten 1980 (Continued)
Outcomes Self-reported crashes injuries and offences
Length of follow up 3 years
Notes Study conducted in Illinois USA
Savolainen 2007
Methods Cohort study
Participants Initial sample size
Intervention group 4000 riders that had conducted the MSF MRC course (passed or failed)
Control group 4000 motorcycle owners or permit or endorsement holders from the Indiana Bureau of Motor
Vehicles database
Analysis sample size
Intervention group 739 (18)
Control group 588 (15)
Interventions Intervention integrity
No information given
Outcomes Probability of crash involvement
Length of follow up up to 5 years
Notes Study conducted in Indiana USA
Swaddiwudhipong 1998
Methods Randomised trial
Participants Initial sample size
Intervention group
all 16 villages of 3 sub-districts
(participants were practising motorcyclists licensed and unlicensed)
Control group
18 villages of 3 other sub-districtsAnalysis sample size
Intervention group 3 districts (100)
Control group 3 districts (100)
Interventions No rider education vs non-mandatory community based rider education consisting of theory on motorcycle injury
risk helmet use traffic laws vehicle regulations and traffic signs and a written and skill test leading to a licence
Intervention integrity
No information given
Outcomes Injuries fatalities
Length of follow up 2 years
40Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4350
Swaddiwudhipong 1998 (Continued)
Notes Study conducted in Tak Province rural Thailand
Waller 1992
Methods Cohort study
Participants Initial sample size
Intervention group 349 riders who succesfully completed the Department of Public Safety motorcycle operator
training in 1988
Control group 349 motorcyclists who never completed the DPS motorcycle operator training
Analysis sample size
Intervention group 349
Control group 349
Interventions No DPS motorcycle operator training vs non-mandatory DPS motorcycle operator training The DPS training is
the Motorcycle Safety Foundationrsquos Motorcycle Rider Course comprising eight hours of theory (lectures discussions
films and multiple choice tests) a textbook and 12 hours off-road practice
Intervention integrity
All 349 riders succesfully completed the rider training
Outcomes Police recorded crashes
Length of follow up 3 years
Notes Study conducted in
Characteristics of excluded studies [ordered by study ID]
Study Reason for exclusion
Anderson 1980a Study type review and process evaluation of a rider course
Aupetit 2008 Outcome proposal of a rider training curriculum that combines traditional training and a motorcycle riding
simulator
Awane 1999 Study type review on use of simulators in rider and driver training
Baldi 2005 Intervention not rider training but intervention integrity of rider training availabilty of rider education
implementation of graduated licensing system etc on a state-wide level
Batchler 1988 Study type description of a training course
Blanchard 2006 Study type not a study assessing the association between rider training and outcomes
Bonnett 2005 Outcome qualitative data regarding satisfaction with the training programme
41Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4450
(Continued)
Braver 2007 Outcome licensure rates
Clayton 1980 Outcome knowledge skills attitude
Council 1978 Outcome crash reduction in ANY vehicle not motorcycles exclusively
Crowther 2005 Study type qualitative study
Daniello 2009 Study type review of the effectiveness of motorcycle training and licensing
Dewanti 2007 Intervention emotional characteristics behaviour (ie not motocycle rider training)
Du Plooy 1990 Study type review of types of rider training courses in South Africa
Engel 1989 Study type uncontrolled before-after study
Goldenbeld 2002 Outcome skills
Goldenbeld 2004 Outcome knowledge and skills
Haworth 2000 Intervention balance between attitudinal andvehicle control skills acrosscurriculla of motorcycle rider training
courses
Haworth 2005 Study type review
Haworth 2006 Study type cross-sectional survey
Hill 1983 Study type review and description of training system
Imatake 1980 Study type course descriptionreview
Ingebritsen 1990 Intervention risk factors for motorcycle crashes
Jain 2009 Study type cross-sectional descriptive study of motorcycle crashes over a 5 year period
Jonah 1979 Study type cross-sectional survey Outcome scores on motorcycle operator skill test
Jonah 1981 Intervention evaluation of a motorcycle rider skill test not motorcycle rider training
Kadar 1990 Study type review of components of rider training courses
Kelsey 1986 Intervention evaluation of a motorcycle rider skill test not motorcycle rider training
Koch 1980 Study type reviewcourse description
Koch 1990 Study type description of course content
42Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4550
(Continued)
Labbett 2005 Study type essay about possible explanations for motorcycle deaths
Law 2005 Intervention targeted motorcycle safety programme not specifically rider training
Lowes 1990 Study type review and descriptions of rider training
Mahon 1976 Study type review of motorcycle rider training methods
Maxwell 1985 Outcomesrecommendations forprogramme operatorsand otherson howthe motorcycle trainingprogramme
works and how it can be improved
Mayhew 1996 Study type review
McKnight 1987 Study type evaluation of methodological and administrative aspects of a motorcycle operator skill test
McPherson 1976 Outcome knowledge and skills
Mohan 2004 Study type review of evidence of interventions to reduce road traffic injuries
Morgan 1980 Study type reviewcourse description
Motoki 1990 Study type proposal and development of rider training curriculum
Nairn 1992 Study type literature review
Newland 1980 Study type description of development of motorcycle safety course
Newland 1982 Study type descriptionreview of a rider training course
Newland 1990 Study type uncontrolled BeforeAfter study
Ormston 2003 Outcome changes in attitude and self reported behaviour
Osga 1979 Outcome knowledge attitudes and skills
Panou 2007 Intervention new simulation tool to teach driverrider
Peterson 1990 Study factor demographic details of course participants completing the course vs those not completing the
course
Prem 1984 Study type experimental
Study factor examines association between skills and performance in tests
Radin Umar 2006 Intervention motorcycle safety programme not specifically rider training
Rockwell 1990 Study type methodologically flawed study type
43Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4650
(Continued)
Rothe 1987 Study type qualitative study of rider training in general
Rowden 2007 Study type qualitative study
RSC Victoria 1993 Study type review of motorcycle rider training and education in Victoria
Russam 1979 Study type review of training programmes in Japan
Saffron 1981 Study type literature review
Schulz 1990 Study factor predictors of motorcycle crashes
Sexton 2004 Study type cross-sectional analytic analysis
Shephard 1986 Study factor reviewdescription of programme
Simpson 1990 Study type review of motorcycle rider trainingeducation
Sokolov 1990 Study factor review of motorcycle riding courses in USSR
Steffens 1988 Outcome self-reported behaviour experiences and researcher observed riding behaviour
Sudlow 2003 Study type review of motorcycle training in the UK
Tasmania 1995 Study type cross-sectional descriptiveprocess evaluation
Tetard 1985 Study type experimental
Outcome detection of indices
Thackray 1980 Outcome instructional effectiveness user acceptance administrative feasibility
Thompson 1994 Study type cross-sectional survey
Outcome attitudes and knowledge
Tobler 1990 Study type review of course curricula
Transport 1989 Intervention syllabus for rider training
Tuan 2007 Study type cross-sectional study (survey) which does not assess the effects of an intervention but examines
motocyclistsrsquo behaviour
Victor 1980 Study type review
Waterer 1990 Study type review of UK rider training issues
Weller 1989 Study type review of studies evaluating motorcycle safety and education measures
44Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4750
(Continued)
Wells 1981 Study type description of different methods of training
Winn 1986 Study type review of rsquohistoryrsquo of rider training
Wisher 1988 Study type cross-sectional survey
Wood 1987 Study type Interrupted Time Series but did not meet the criteria for an ITS as published by the Cochrane
Effective Practice and Organisation of Care Group
Yamazaki 1990 Study type cross-sectional survey
Outcome riding skills skills retention and attitude
Characteristics of ongoing studies [ordered by study ID]
Burgess 2010
Trial name or title UKs National Ride program
Methods Cohort study
Participants Motorcycle offenders
Interventions Police led diversion program
Outcomes Self reported attitudinal and behavioural measures
Starting date
Contact information
Notes Expected to be completed in early 2010
45Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4850
D A T A A N D A N A L Y S E S
This review has no analyses
A P P E N D I C E S
Appendix 1 Search strategy
We searched the following electronic databases
bull Cochrane Injuries Group Specialised Register (searched 22 Sept 2008)
bull CENTRAL (The Cochrane Library 2008 Issue 3)
bull TRANSPORT 1988 to 200706 (latest date available)
bull ATRI Australian Transport Index (to Sept 2008)
bull MEDLINE (Ovid SP) 1950 to Sept 2008
bull EMBASE (Ovid SP) 1980 to Sept 2008
bull CINAHL (EBSCO) 1982 to Sept 2008
bull WHOLIS (World Health Organization Library Information System) (to Sept 2008)
bull PsycINFO (Ovid SP) 1806 to Sept 2008
bull LILACS (BIREME) (Latin American and Caribbean Health Sciences) (1982 to Sept 2008)
bull ISI Web of Science Social Sciences Citation Index (SSCI) 1970 to Sept 2008
bull ERIC (searched 25 Sept 2008)
bull ZETOC (searched 25 Sept 2008)
bull SIGLE (searched 25 Sept 2008)
All of the databases were searched using the following text-word search strategy The strategy was adapted using either rsquorsquo or rsquo$rsquo as atruncation symbol
1 motor-cycl or motorcycl or motor-bik or motorbik or motor-scooter or motorscooter or scooter or moped or motocycl
or motocicl or moto
2 train or educat or instruct or program or course or qualif or learn or improv
3 1 and 2
H I S T O R Y
Protocol first published Issue 2 2005
Review first published Issue 10 2010
46Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4950
C O N T R I B U T I O N S O F A U T H O R S
Katina Kardamanidis was the first author of the protocol conducted the literature search extracted full studies assessed the studies
according to inclusion criteria assessed quality of the studies compared findings with the second author discussed disagreements with
the third author wrote the drafts and the final review
Rebecca Ivers was the second author of the protocol assessed the studies according to inclusion criteria assessed quality of the studies
compared findings with the first author edited the final draft and provided overall guidance for the process
Alexandra Martiniuk assessed quality of the studies compared findings with the first author discussed disagreements with the third
author and helped write the first draft and the final review
Mark Stevenson contributed to the protocol reviewed methods of studies and contributed to the final draft
Katrina Thistlethwaite conducted a second literature search extracted full studies and contributed to the final draft
D E C L A R A T I O N S O F I N T E R E S T
None known
S O U R C E S O F S U P P O R T
Internal sources
bull No sources of support supplied
External sources
bull National Health and Medical Research Council of Australia Australia
Salary support for Martiniuk Ivers and Stevenson
bull University of Sydney Sesqui Research Grant Australia
Direct research costs and salary support for Kardamanidis
D I F F E R E N C E S B E T W E E N P R O T O C O L A N D R E V I E W
There are no differences between the protocol and the review
I N D E X T E R M S
Medical Subject Headings (MeSH)
lowastMotorcycles Accident Prevention [lowastmethods] Accidents Traffic [lowastprevention amp control] Licensure Program Evaluation Wounds
and Injuries [mortality lowastprevention amp control]
47Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 5050
MeSH check words
Humans
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3650
Hurt 1981 (Continued)
Notes Study conducted in Los Angeles USA
Jonah 1982
Methods Cohort study
Participants Initial sample size
Intervention group
2310 licensed MTP graduates who are active riders
Control group
2029 licensed motorcyclists who are active riders
Non-licensed and non-active riders excluded
Analysis sample size
Intervention group
811 interviews (35) 533 driver records (23)
Control group 1080 interviews (71) 905 driver records (45)
Interventions No training vs the non-mandatory Canada Safety Council Motorcycle Training Program (MTP) comprising of 4
hours theory and 16 hours practise (6 hrs on-road) over 2 weekends
Intervention integrity
Participants had all successfully completed the MTP
Outcomes Self-reported and police reported crashes and offencesLength of follow up 4 years
Notes Study conducted in Ontario Canada
Kloeden 1994
Methods Cohort study
Participants Initial sample size
Intervention group 2533 licence applicants
Control group Unclear
Sample of riders obtaining probationary licenceIntervention group 1625 (64)
Control group 5015 (unclear)
Non-active riders excluded from control group (~66)
Analysis sample size
Intervention group 631 (39)
Control group 318 (6)
Interventions No formal training vs the mandatory pre-licence Ridersafe course consisting of two levelsthe first is a training course
in the basics of motorcycle riding where a pass entitles the rider to obtain a learnerrsquos permit The second is a skill
test where a pass entitles the rider to obtain a probationary motorcycle licence The training course consists of two 4
hour training sessions on separate days involving instruction videos and actual riding on motorcycles The skill test
consists of a 4 hour preparation course of 74 minutes of lecturediscussion 38 minutes of video and 75 minutes of
34Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3750
Kloeden 1994 (Continued)
on-course riding which is the testing phase
Intervention integrity
Unclear what percentage in the Ridersafe group and the control group obtained their probationary licence halfway
and their full motorcycle licence in the end
Outcomes Police reported crashes
Length of follow up 1 year
Notes Study conducted in SA Australia
Kraus 1975
Methods Population based case-control study
Participants Cases All motorcycle riders with a confirmed medically treated motorcycle injury (including fatalities) from all
hospitals in study base and peripheral area hospitals and all deaths due to motorcycle crashes in the study base All
serious injuries (leading to hospitalisation) and fatalities likely to be included
Community controls A random sample of registered motorcycle owners in the same temporary and geographical
study base as the cases arose from
Initial sample size Cases 1273
Controls 738
Analysis sample size Cases 741 (58)
Controls 434 (59)
Interventions Motorcycle rider training (not specified) measured through self-report
Outcomes Injuries leading to hospitalisation fatalities
Length of follow up not specified
Notes Study conducted in Sacramento county USA
Leung 1987
Methods Non-randomised trial
Participants Initial sample sizeIntervention group unknown number of MRC graduates
Control group unknown number of licensed motorcyclists
Analysis sample size
Intervention group 2941
Control group 43094
Interventions No formal training vs non-mandatory Wisconsin motorcycle rider course a 22 hours off-street programme that
teaches the basic rules of safe motorcycling
Intervention integrity
50 in the MRC group and 100 in the control group obtained their motorcycle licence in the end
35Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3850
Leung 1987 (Continued)
Outcomes Police recorded crashes and offences
Length of follow up up to 3 years
Notes Study conducted in Wisconsin USA
MAIDS 2004
Methods Population based case-control study
Participants Cases Motorcycle riders who had a motorcycle crash Unclear whether all cases included
Community controls Motorcycle riders identified at petrol stations Unclear whether matched to cases
Initial sample size
Cases not reported
Controls not reported
Analysis sample size
Cases 921 (Italy 200 Netherlands 200 Spain 121 Germany 250 France 150)
Controls 923 (Italy 200 Netherlands 200 Spain123 Germany 250 France 150)
Interventions Pre-license rider training or additional training (non-compulsory)
Outcomes Crashes
Notes Study in five sampling areas located in five European countries (France Germany Netherlands Spain and Italy)
McDavid 1989
Methods Cohort study
Participants Sample size after first match
Intervention group 346 male training graduates
Control group 346 male motorcycle riders licensed in same year
Sample size after second match
Intervention group 139 (40)
Control group 139 (40)
Analysis sample size
Intervention group 139 (40)
Control group 139 (40)
Interventions No formal training vs the non-mandatory pre-licence British Columbia Safety Councilrsquos 37-hour motorcycle safety
programme
Intervention integrity
Participants had all completed and passed the training
Outcomes Police reported crashes injurious crashes
Length of follow up 1-6 years
36Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3950
McDavid 1989 (Continued)
Notes Study conducted in British Columbia Canada
Mortimer 1984
Methods Cohort study
Participants Initial sample size
Intervention group 608 MRC graduates
Control group 360 motorcycle riders
Response rates
Intervention group 360 (59)
Control group 324 (90)
Non-active riders excluded
Analysis sample size
Intervention group 213 (35)
Control group 303 (84)
Interventions No formal training vs the non-mandatory Motorcycle Safety Foundationrsquos Motorcycle Rider Course comprising of
8 hours theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-road practice
Intervention integrity
Participants had all completed and passed the MRC
Outcomes Self-reported crashes and offencesLength of follow up 1 year
Notes Study conducted in Illinois USA
Mortimer 1988
Methods Cohort study
Participants Initial sample size
Intervention group 2307 MRC graduates
Control group 556 active licensed motorcycle riders
Response ratesIntervention group 1298 (56)
Control group 500 (90)
Non-active riders excluded
Analysis sample size
Intervention group 913 (40)
Control group 500 (90)
Interventions No formal training vs the non-mandatory Motorcycle Safety Foundationrsquos Motorcycle Rider Course comprising of
8 hours theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-road practice
Intervention integrity
Participants had all completed and passed the MRC
37Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4050
Mortimer 1988 (Continued)
Outcomes Self reported crashes and offences
Length of follow up 1 year
Notes Study conducted in Illinois USA
Osga 1980
Methods Cohort study
Participants Initial sample sizeIntervention group 1163 MSF MRC participants invited 818 (74) responded
Control group 329 motorcycle riders responded to recruitment strategies
Non-active riders excluded from analysis
Analysis sample size
Intervention group 595 (73)
Control group 329 (100)
Interventions No formal training vs the non-mandatory Motorcycle Safety Foundationrsquos Motorcycle Rider Course (MRC) com-
prising of 8 hours theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-road
practice
Intervention integrity
Participants had all successfully completed the MRC
Outcomes Self-reported crashes police recorded crash reports
Length of follow up 6 months - 3 12 years
Notes Study conducted in South Dakota USA
Perrino 2002
Methods Cohort study
Participants Initial sample size
Intervention group 2144 (unclear) MRC graduates (motorcycle licence applicants)
Control group 2601 (unclear) regular licensing test graduates (motorcycle licence applicants)Analysis sample size
Intervention group 1070 (unclear)
Control group 867 (unclear)
Interventions No formal training vs the mandatory (for those younger than 18 years) pre-licence Maryland Motorcycle Rider
Course (MRC) consisting of 8 hours theory and 12 hours practice
Intervention integrity
No information given
Outcomes Self-reported and police recorded crashes and offences
Length of follow up 1 year
38Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4150
Perrino 2002 (Continued)
Notes Study conducted in Maryland USA
Raymond 1979
Methods Cohort study
Participants Initial sample size
Intervention group
258 volunteers mainly (but not all) novice riders
Control group
1104 provisional licence applicants
Analysis sample size
Intervention group 190 (74)
Control group 627 (57)
Interventions No formal training vs the non-mandatory RACACU (Royal Automobile ClubAuto-Cycle Union) training scheme
The training program consisted of either 24 one-hour sessions or 13 two-hour sessions Twelve hours of the training
focused on theory about basic operational procedures traffic laws and highway code The practical session focused
on basic operational skills riding on a training area and riding in traffic with an instructor A proficiency test was
administered at the end of the course
Intervention integrity
No information given
Outcomes Self-reported crashes (including minor non-injury non-damage and off-road crashes)
Length of follow up 1 year
Notes Study conducted in the UK
Satten 1980
Methods Cohort study
Participants Initial sample size
Intervention group interviewers kept interviewing until they had 100 riders selected from a population of 891
enrollees in the MSF MRC course (unknown how many were asked and how many refused to participate)Control group unknown number of registered motorcycle riders
Non-active riders excluded from analysis
Analysis sample size
Intervention group 71
Control group 100
Interventions No formal training vs the non-mandatory Motorcycle Safety Foundationrsquos (MSF) Motorcycle Rider Course (MRC)
comprising of 8 hours theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-
road practice
Intervention integrity
Participants had all successfully completed the MRC
39Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4250
Satten 1980 (Continued)
Outcomes Self-reported crashes injuries and offences
Length of follow up 3 years
Notes Study conducted in Illinois USA
Savolainen 2007
Methods Cohort study
Participants Initial sample size
Intervention group 4000 riders that had conducted the MSF MRC course (passed or failed)
Control group 4000 motorcycle owners or permit or endorsement holders from the Indiana Bureau of Motor
Vehicles database
Analysis sample size
Intervention group 739 (18)
Control group 588 (15)
Interventions Intervention integrity
No information given
Outcomes Probability of crash involvement
Length of follow up up to 5 years
Notes Study conducted in Indiana USA
Swaddiwudhipong 1998
Methods Randomised trial
Participants Initial sample size
Intervention group
all 16 villages of 3 sub-districts
(participants were practising motorcyclists licensed and unlicensed)
Control group
18 villages of 3 other sub-districtsAnalysis sample size
Intervention group 3 districts (100)
Control group 3 districts (100)
Interventions No rider education vs non-mandatory community based rider education consisting of theory on motorcycle injury
risk helmet use traffic laws vehicle regulations and traffic signs and a written and skill test leading to a licence
Intervention integrity
No information given
Outcomes Injuries fatalities
Length of follow up 2 years
40Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4350
Swaddiwudhipong 1998 (Continued)
Notes Study conducted in Tak Province rural Thailand
Waller 1992
Methods Cohort study
Participants Initial sample size
Intervention group 349 riders who succesfully completed the Department of Public Safety motorcycle operator
training in 1988
Control group 349 motorcyclists who never completed the DPS motorcycle operator training
Analysis sample size
Intervention group 349
Control group 349
Interventions No DPS motorcycle operator training vs non-mandatory DPS motorcycle operator training The DPS training is
the Motorcycle Safety Foundationrsquos Motorcycle Rider Course comprising eight hours of theory (lectures discussions
films and multiple choice tests) a textbook and 12 hours off-road practice
Intervention integrity
All 349 riders succesfully completed the rider training
Outcomes Police recorded crashes
Length of follow up 3 years
Notes Study conducted in
Characteristics of excluded studies [ordered by study ID]
Study Reason for exclusion
Anderson 1980a Study type review and process evaluation of a rider course
Aupetit 2008 Outcome proposal of a rider training curriculum that combines traditional training and a motorcycle riding
simulator
Awane 1999 Study type review on use of simulators in rider and driver training
Baldi 2005 Intervention not rider training but intervention integrity of rider training availabilty of rider education
implementation of graduated licensing system etc on a state-wide level
Batchler 1988 Study type description of a training course
Blanchard 2006 Study type not a study assessing the association between rider training and outcomes
Bonnett 2005 Outcome qualitative data regarding satisfaction with the training programme
41Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4450
(Continued)
Braver 2007 Outcome licensure rates
Clayton 1980 Outcome knowledge skills attitude
Council 1978 Outcome crash reduction in ANY vehicle not motorcycles exclusively
Crowther 2005 Study type qualitative study
Daniello 2009 Study type review of the effectiveness of motorcycle training and licensing
Dewanti 2007 Intervention emotional characteristics behaviour (ie not motocycle rider training)
Du Plooy 1990 Study type review of types of rider training courses in South Africa
Engel 1989 Study type uncontrolled before-after study
Goldenbeld 2002 Outcome skills
Goldenbeld 2004 Outcome knowledge and skills
Haworth 2000 Intervention balance between attitudinal andvehicle control skills acrosscurriculla of motorcycle rider training
courses
Haworth 2005 Study type review
Haworth 2006 Study type cross-sectional survey
Hill 1983 Study type review and description of training system
Imatake 1980 Study type course descriptionreview
Ingebritsen 1990 Intervention risk factors for motorcycle crashes
Jain 2009 Study type cross-sectional descriptive study of motorcycle crashes over a 5 year period
Jonah 1979 Study type cross-sectional survey Outcome scores on motorcycle operator skill test
Jonah 1981 Intervention evaluation of a motorcycle rider skill test not motorcycle rider training
Kadar 1990 Study type review of components of rider training courses
Kelsey 1986 Intervention evaluation of a motorcycle rider skill test not motorcycle rider training
Koch 1980 Study type reviewcourse description
Koch 1990 Study type description of course content
42Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4550
(Continued)
Labbett 2005 Study type essay about possible explanations for motorcycle deaths
Law 2005 Intervention targeted motorcycle safety programme not specifically rider training
Lowes 1990 Study type review and descriptions of rider training
Mahon 1976 Study type review of motorcycle rider training methods
Maxwell 1985 Outcomesrecommendations forprogramme operatorsand otherson howthe motorcycle trainingprogramme
works and how it can be improved
Mayhew 1996 Study type review
McKnight 1987 Study type evaluation of methodological and administrative aspects of a motorcycle operator skill test
McPherson 1976 Outcome knowledge and skills
Mohan 2004 Study type review of evidence of interventions to reduce road traffic injuries
Morgan 1980 Study type reviewcourse description
Motoki 1990 Study type proposal and development of rider training curriculum
Nairn 1992 Study type literature review
Newland 1980 Study type description of development of motorcycle safety course
Newland 1982 Study type descriptionreview of a rider training course
Newland 1990 Study type uncontrolled BeforeAfter study
Ormston 2003 Outcome changes in attitude and self reported behaviour
Osga 1979 Outcome knowledge attitudes and skills
Panou 2007 Intervention new simulation tool to teach driverrider
Peterson 1990 Study factor demographic details of course participants completing the course vs those not completing the
course
Prem 1984 Study type experimental
Study factor examines association between skills and performance in tests
Radin Umar 2006 Intervention motorcycle safety programme not specifically rider training
Rockwell 1990 Study type methodologically flawed study type
43Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4650
(Continued)
Rothe 1987 Study type qualitative study of rider training in general
Rowden 2007 Study type qualitative study
RSC Victoria 1993 Study type review of motorcycle rider training and education in Victoria
Russam 1979 Study type review of training programmes in Japan
Saffron 1981 Study type literature review
Schulz 1990 Study factor predictors of motorcycle crashes
Sexton 2004 Study type cross-sectional analytic analysis
Shephard 1986 Study factor reviewdescription of programme
Simpson 1990 Study type review of motorcycle rider trainingeducation
Sokolov 1990 Study factor review of motorcycle riding courses in USSR
Steffens 1988 Outcome self-reported behaviour experiences and researcher observed riding behaviour
Sudlow 2003 Study type review of motorcycle training in the UK
Tasmania 1995 Study type cross-sectional descriptiveprocess evaluation
Tetard 1985 Study type experimental
Outcome detection of indices
Thackray 1980 Outcome instructional effectiveness user acceptance administrative feasibility
Thompson 1994 Study type cross-sectional survey
Outcome attitudes and knowledge
Tobler 1990 Study type review of course curricula
Transport 1989 Intervention syllabus for rider training
Tuan 2007 Study type cross-sectional study (survey) which does not assess the effects of an intervention but examines
motocyclistsrsquo behaviour
Victor 1980 Study type review
Waterer 1990 Study type review of UK rider training issues
Weller 1989 Study type review of studies evaluating motorcycle safety and education measures
44Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4750
(Continued)
Wells 1981 Study type description of different methods of training
Winn 1986 Study type review of rsquohistoryrsquo of rider training
Wisher 1988 Study type cross-sectional survey
Wood 1987 Study type Interrupted Time Series but did not meet the criteria for an ITS as published by the Cochrane
Effective Practice and Organisation of Care Group
Yamazaki 1990 Study type cross-sectional survey
Outcome riding skills skills retention and attitude
Characteristics of ongoing studies [ordered by study ID]
Burgess 2010
Trial name or title UKs National Ride program
Methods Cohort study
Participants Motorcycle offenders
Interventions Police led diversion program
Outcomes Self reported attitudinal and behavioural measures
Starting date
Contact information
Notes Expected to be completed in early 2010
45Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4850
D A T A A N D A N A L Y S E S
This review has no analyses
A P P E N D I C E S
Appendix 1 Search strategy
We searched the following electronic databases
bull Cochrane Injuries Group Specialised Register (searched 22 Sept 2008)
bull CENTRAL (The Cochrane Library 2008 Issue 3)
bull TRANSPORT 1988 to 200706 (latest date available)
bull ATRI Australian Transport Index (to Sept 2008)
bull MEDLINE (Ovid SP) 1950 to Sept 2008
bull EMBASE (Ovid SP) 1980 to Sept 2008
bull CINAHL (EBSCO) 1982 to Sept 2008
bull WHOLIS (World Health Organization Library Information System) (to Sept 2008)
bull PsycINFO (Ovid SP) 1806 to Sept 2008
bull LILACS (BIREME) (Latin American and Caribbean Health Sciences) (1982 to Sept 2008)
bull ISI Web of Science Social Sciences Citation Index (SSCI) 1970 to Sept 2008
bull ERIC (searched 25 Sept 2008)
bull ZETOC (searched 25 Sept 2008)
bull SIGLE (searched 25 Sept 2008)
All of the databases were searched using the following text-word search strategy The strategy was adapted using either rsquorsquo or rsquo$rsquo as atruncation symbol
1 motor-cycl or motorcycl or motor-bik or motorbik or motor-scooter or motorscooter or scooter or moped or motocycl
or motocicl or moto
2 train or educat or instruct or program or course or qualif or learn or improv
3 1 and 2
H I S T O R Y
Protocol first published Issue 2 2005
Review first published Issue 10 2010
46Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4950
C O N T R I B U T I O N S O F A U T H O R S
Katina Kardamanidis was the first author of the protocol conducted the literature search extracted full studies assessed the studies
according to inclusion criteria assessed quality of the studies compared findings with the second author discussed disagreements with
the third author wrote the drafts and the final review
Rebecca Ivers was the second author of the protocol assessed the studies according to inclusion criteria assessed quality of the studies
compared findings with the first author edited the final draft and provided overall guidance for the process
Alexandra Martiniuk assessed quality of the studies compared findings with the first author discussed disagreements with the third
author and helped write the first draft and the final review
Mark Stevenson contributed to the protocol reviewed methods of studies and contributed to the final draft
Katrina Thistlethwaite conducted a second literature search extracted full studies and contributed to the final draft
D E C L A R A T I O N S O F I N T E R E S T
None known
S O U R C E S O F S U P P O R T
Internal sources
bull No sources of support supplied
External sources
bull National Health and Medical Research Council of Australia Australia
Salary support for Martiniuk Ivers and Stevenson
bull University of Sydney Sesqui Research Grant Australia
Direct research costs and salary support for Kardamanidis
D I F F E R E N C E S B E T W E E N P R O T O C O L A N D R E V I E W
There are no differences between the protocol and the review
I N D E X T E R M S
Medical Subject Headings (MeSH)
lowastMotorcycles Accident Prevention [lowastmethods] Accidents Traffic [lowastprevention amp control] Licensure Program Evaluation Wounds
and Injuries [mortality lowastprevention amp control]
47Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 5050
MeSH check words
Humans
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3750
Kloeden 1994 (Continued)
on-course riding which is the testing phase
Intervention integrity
Unclear what percentage in the Ridersafe group and the control group obtained their probationary licence halfway
and their full motorcycle licence in the end
Outcomes Police reported crashes
Length of follow up 1 year
Notes Study conducted in SA Australia
Kraus 1975
Methods Population based case-control study
Participants Cases All motorcycle riders with a confirmed medically treated motorcycle injury (including fatalities) from all
hospitals in study base and peripheral area hospitals and all deaths due to motorcycle crashes in the study base All
serious injuries (leading to hospitalisation) and fatalities likely to be included
Community controls A random sample of registered motorcycle owners in the same temporary and geographical
study base as the cases arose from
Initial sample size Cases 1273
Controls 738
Analysis sample size Cases 741 (58)
Controls 434 (59)
Interventions Motorcycle rider training (not specified) measured through self-report
Outcomes Injuries leading to hospitalisation fatalities
Length of follow up not specified
Notes Study conducted in Sacramento county USA
Leung 1987
Methods Non-randomised trial
Participants Initial sample sizeIntervention group unknown number of MRC graduates
Control group unknown number of licensed motorcyclists
Analysis sample size
Intervention group 2941
Control group 43094
Interventions No formal training vs non-mandatory Wisconsin motorcycle rider course a 22 hours off-street programme that
teaches the basic rules of safe motorcycling
Intervention integrity
50 in the MRC group and 100 in the control group obtained their motorcycle licence in the end
35Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3850
Leung 1987 (Continued)
Outcomes Police recorded crashes and offences
Length of follow up up to 3 years
Notes Study conducted in Wisconsin USA
MAIDS 2004
Methods Population based case-control study
Participants Cases Motorcycle riders who had a motorcycle crash Unclear whether all cases included
Community controls Motorcycle riders identified at petrol stations Unclear whether matched to cases
Initial sample size
Cases not reported
Controls not reported
Analysis sample size
Cases 921 (Italy 200 Netherlands 200 Spain 121 Germany 250 France 150)
Controls 923 (Italy 200 Netherlands 200 Spain123 Germany 250 France 150)
Interventions Pre-license rider training or additional training (non-compulsory)
Outcomes Crashes
Notes Study in five sampling areas located in five European countries (France Germany Netherlands Spain and Italy)
McDavid 1989
Methods Cohort study
Participants Sample size after first match
Intervention group 346 male training graduates
Control group 346 male motorcycle riders licensed in same year
Sample size after second match
Intervention group 139 (40)
Control group 139 (40)
Analysis sample size
Intervention group 139 (40)
Control group 139 (40)
Interventions No formal training vs the non-mandatory pre-licence British Columbia Safety Councilrsquos 37-hour motorcycle safety
programme
Intervention integrity
Participants had all completed and passed the training
Outcomes Police reported crashes injurious crashes
Length of follow up 1-6 years
36Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3950
McDavid 1989 (Continued)
Notes Study conducted in British Columbia Canada
Mortimer 1984
Methods Cohort study
Participants Initial sample size
Intervention group 608 MRC graduates
Control group 360 motorcycle riders
Response rates
Intervention group 360 (59)
Control group 324 (90)
Non-active riders excluded
Analysis sample size
Intervention group 213 (35)
Control group 303 (84)
Interventions No formal training vs the non-mandatory Motorcycle Safety Foundationrsquos Motorcycle Rider Course comprising of
8 hours theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-road practice
Intervention integrity
Participants had all completed and passed the MRC
Outcomes Self-reported crashes and offencesLength of follow up 1 year
Notes Study conducted in Illinois USA
Mortimer 1988
Methods Cohort study
Participants Initial sample size
Intervention group 2307 MRC graduates
Control group 556 active licensed motorcycle riders
Response ratesIntervention group 1298 (56)
Control group 500 (90)
Non-active riders excluded
Analysis sample size
Intervention group 913 (40)
Control group 500 (90)
Interventions No formal training vs the non-mandatory Motorcycle Safety Foundationrsquos Motorcycle Rider Course comprising of
8 hours theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-road practice
Intervention integrity
Participants had all completed and passed the MRC
37Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4050
Mortimer 1988 (Continued)
Outcomes Self reported crashes and offences
Length of follow up 1 year
Notes Study conducted in Illinois USA
Osga 1980
Methods Cohort study
Participants Initial sample sizeIntervention group 1163 MSF MRC participants invited 818 (74) responded
Control group 329 motorcycle riders responded to recruitment strategies
Non-active riders excluded from analysis
Analysis sample size
Intervention group 595 (73)
Control group 329 (100)
Interventions No formal training vs the non-mandatory Motorcycle Safety Foundationrsquos Motorcycle Rider Course (MRC) com-
prising of 8 hours theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-road
practice
Intervention integrity
Participants had all successfully completed the MRC
Outcomes Self-reported crashes police recorded crash reports
Length of follow up 6 months - 3 12 years
Notes Study conducted in South Dakota USA
Perrino 2002
Methods Cohort study
Participants Initial sample size
Intervention group 2144 (unclear) MRC graduates (motorcycle licence applicants)
Control group 2601 (unclear) regular licensing test graduates (motorcycle licence applicants)Analysis sample size
Intervention group 1070 (unclear)
Control group 867 (unclear)
Interventions No formal training vs the mandatory (for those younger than 18 years) pre-licence Maryland Motorcycle Rider
Course (MRC) consisting of 8 hours theory and 12 hours practice
Intervention integrity
No information given
Outcomes Self-reported and police recorded crashes and offences
Length of follow up 1 year
38Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4150
Perrino 2002 (Continued)
Notes Study conducted in Maryland USA
Raymond 1979
Methods Cohort study
Participants Initial sample size
Intervention group
258 volunteers mainly (but not all) novice riders
Control group
1104 provisional licence applicants
Analysis sample size
Intervention group 190 (74)
Control group 627 (57)
Interventions No formal training vs the non-mandatory RACACU (Royal Automobile ClubAuto-Cycle Union) training scheme
The training program consisted of either 24 one-hour sessions or 13 two-hour sessions Twelve hours of the training
focused on theory about basic operational procedures traffic laws and highway code The practical session focused
on basic operational skills riding on a training area and riding in traffic with an instructor A proficiency test was
administered at the end of the course
Intervention integrity
No information given
Outcomes Self-reported crashes (including minor non-injury non-damage and off-road crashes)
Length of follow up 1 year
Notes Study conducted in the UK
Satten 1980
Methods Cohort study
Participants Initial sample size
Intervention group interviewers kept interviewing until they had 100 riders selected from a population of 891
enrollees in the MSF MRC course (unknown how many were asked and how many refused to participate)Control group unknown number of registered motorcycle riders
Non-active riders excluded from analysis
Analysis sample size
Intervention group 71
Control group 100
Interventions No formal training vs the non-mandatory Motorcycle Safety Foundationrsquos (MSF) Motorcycle Rider Course (MRC)
comprising of 8 hours theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-
road practice
Intervention integrity
Participants had all successfully completed the MRC
39Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4250
Satten 1980 (Continued)
Outcomes Self-reported crashes injuries and offences
Length of follow up 3 years
Notes Study conducted in Illinois USA
Savolainen 2007
Methods Cohort study
Participants Initial sample size
Intervention group 4000 riders that had conducted the MSF MRC course (passed or failed)
Control group 4000 motorcycle owners or permit or endorsement holders from the Indiana Bureau of Motor
Vehicles database
Analysis sample size
Intervention group 739 (18)
Control group 588 (15)
Interventions Intervention integrity
No information given
Outcomes Probability of crash involvement
Length of follow up up to 5 years
Notes Study conducted in Indiana USA
Swaddiwudhipong 1998
Methods Randomised trial
Participants Initial sample size
Intervention group
all 16 villages of 3 sub-districts
(participants were practising motorcyclists licensed and unlicensed)
Control group
18 villages of 3 other sub-districtsAnalysis sample size
Intervention group 3 districts (100)
Control group 3 districts (100)
Interventions No rider education vs non-mandatory community based rider education consisting of theory on motorcycle injury
risk helmet use traffic laws vehicle regulations and traffic signs and a written and skill test leading to a licence
Intervention integrity
No information given
Outcomes Injuries fatalities
Length of follow up 2 years
40Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4350
Swaddiwudhipong 1998 (Continued)
Notes Study conducted in Tak Province rural Thailand
Waller 1992
Methods Cohort study
Participants Initial sample size
Intervention group 349 riders who succesfully completed the Department of Public Safety motorcycle operator
training in 1988
Control group 349 motorcyclists who never completed the DPS motorcycle operator training
Analysis sample size
Intervention group 349
Control group 349
Interventions No DPS motorcycle operator training vs non-mandatory DPS motorcycle operator training The DPS training is
the Motorcycle Safety Foundationrsquos Motorcycle Rider Course comprising eight hours of theory (lectures discussions
films and multiple choice tests) a textbook and 12 hours off-road practice
Intervention integrity
All 349 riders succesfully completed the rider training
Outcomes Police recorded crashes
Length of follow up 3 years
Notes Study conducted in
Characteristics of excluded studies [ordered by study ID]
Study Reason for exclusion
Anderson 1980a Study type review and process evaluation of a rider course
Aupetit 2008 Outcome proposal of a rider training curriculum that combines traditional training and a motorcycle riding
simulator
Awane 1999 Study type review on use of simulators in rider and driver training
Baldi 2005 Intervention not rider training but intervention integrity of rider training availabilty of rider education
implementation of graduated licensing system etc on a state-wide level
Batchler 1988 Study type description of a training course
Blanchard 2006 Study type not a study assessing the association between rider training and outcomes
Bonnett 2005 Outcome qualitative data regarding satisfaction with the training programme
41Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4450
(Continued)
Braver 2007 Outcome licensure rates
Clayton 1980 Outcome knowledge skills attitude
Council 1978 Outcome crash reduction in ANY vehicle not motorcycles exclusively
Crowther 2005 Study type qualitative study
Daniello 2009 Study type review of the effectiveness of motorcycle training and licensing
Dewanti 2007 Intervention emotional characteristics behaviour (ie not motocycle rider training)
Du Plooy 1990 Study type review of types of rider training courses in South Africa
Engel 1989 Study type uncontrolled before-after study
Goldenbeld 2002 Outcome skills
Goldenbeld 2004 Outcome knowledge and skills
Haworth 2000 Intervention balance between attitudinal andvehicle control skills acrosscurriculla of motorcycle rider training
courses
Haworth 2005 Study type review
Haworth 2006 Study type cross-sectional survey
Hill 1983 Study type review and description of training system
Imatake 1980 Study type course descriptionreview
Ingebritsen 1990 Intervention risk factors for motorcycle crashes
Jain 2009 Study type cross-sectional descriptive study of motorcycle crashes over a 5 year period
Jonah 1979 Study type cross-sectional survey Outcome scores on motorcycle operator skill test
Jonah 1981 Intervention evaluation of a motorcycle rider skill test not motorcycle rider training
Kadar 1990 Study type review of components of rider training courses
Kelsey 1986 Intervention evaluation of a motorcycle rider skill test not motorcycle rider training
Koch 1980 Study type reviewcourse description
Koch 1990 Study type description of course content
42Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4550
(Continued)
Labbett 2005 Study type essay about possible explanations for motorcycle deaths
Law 2005 Intervention targeted motorcycle safety programme not specifically rider training
Lowes 1990 Study type review and descriptions of rider training
Mahon 1976 Study type review of motorcycle rider training methods
Maxwell 1985 Outcomesrecommendations forprogramme operatorsand otherson howthe motorcycle trainingprogramme
works and how it can be improved
Mayhew 1996 Study type review
McKnight 1987 Study type evaluation of methodological and administrative aspects of a motorcycle operator skill test
McPherson 1976 Outcome knowledge and skills
Mohan 2004 Study type review of evidence of interventions to reduce road traffic injuries
Morgan 1980 Study type reviewcourse description
Motoki 1990 Study type proposal and development of rider training curriculum
Nairn 1992 Study type literature review
Newland 1980 Study type description of development of motorcycle safety course
Newland 1982 Study type descriptionreview of a rider training course
Newland 1990 Study type uncontrolled BeforeAfter study
Ormston 2003 Outcome changes in attitude and self reported behaviour
Osga 1979 Outcome knowledge attitudes and skills
Panou 2007 Intervention new simulation tool to teach driverrider
Peterson 1990 Study factor demographic details of course participants completing the course vs those not completing the
course
Prem 1984 Study type experimental
Study factor examines association between skills and performance in tests
Radin Umar 2006 Intervention motorcycle safety programme not specifically rider training
Rockwell 1990 Study type methodologically flawed study type
43Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4650
(Continued)
Rothe 1987 Study type qualitative study of rider training in general
Rowden 2007 Study type qualitative study
RSC Victoria 1993 Study type review of motorcycle rider training and education in Victoria
Russam 1979 Study type review of training programmes in Japan
Saffron 1981 Study type literature review
Schulz 1990 Study factor predictors of motorcycle crashes
Sexton 2004 Study type cross-sectional analytic analysis
Shephard 1986 Study factor reviewdescription of programme
Simpson 1990 Study type review of motorcycle rider trainingeducation
Sokolov 1990 Study factor review of motorcycle riding courses in USSR
Steffens 1988 Outcome self-reported behaviour experiences and researcher observed riding behaviour
Sudlow 2003 Study type review of motorcycle training in the UK
Tasmania 1995 Study type cross-sectional descriptiveprocess evaluation
Tetard 1985 Study type experimental
Outcome detection of indices
Thackray 1980 Outcome instructional effectiveness user acceptance administrative feasibility
Thompson 1994 Study type cross-sectional survey
Outcome attitudes and knowledge
Tobler 1990 Study type review of course curricula
Transport 1989 Intervention syllabus for rider training
Tuan 2007 Study type cross-sectional study (survey) which does not assess the effects of an intervention but examines
motocyclistsrsquo behaviour
Victor 1980 Study type review
Waterer 1990 Study type review of UK rider training issues
Weller 1989 Study type review of studies evaluating motorcycle safety and education measures
44Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4750
(Continued)
Wells 1981 Study type description of different methods of training
Winn 1986 Study type review of rsquohistoryrsquo of rider training
Wisher 1988 Study type cross-sectional survey
Wood 1987 Study type Interrupted Time Series but did not meet the criteria for an ITS as published by the Cochrane
Effective Practice and Organisation of Care Group
Yamazaki 1990 Study type cross-sectional survey
Outcome riding skills skills retention and attitude
Characteristics of ongoing studies [ordered by study ID]
Burgess 2010
Trial name or title UKs National Ride program
Methods Cohort study
Participants Motorcycle offenders
Interventions Police led diversion program
Outcomes Self reported attitudinal and behavioural measures
Starting date
Contact information
Notes Expected to be completed in early 2010
45Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4850
D A T A A N D A N A L Y S E S
This review has no analyses
A P P E N D I C E S
Appendix 1 Search strategy
We searched the following electronic databases
bull Cochrane Injuries Group Specialised Register (searched 22 Sept 2008)
bull CENTRAL (The Cochrane Library 2008 Issue 3)
bull TRANSPORT 1988 to 200706 (latest date available)
bull ATRI Australian Transport Index (to Sept 2008)
bull MEDLINE (Ovid SP) 1950 to Sept 2008
bull EMBASE (Ovid SP) 1980 to Sept 2008
bull CINAHL (EBSCO) 1982 to Sept 2008
bull WHOLIS (World Health Organization Library Information System) (to Sept 2008)
bull PsycINFO (Ovid SP) 1806 to Sept 2008
bull LILACS (BIREME) (Latin American and Caribbean Health Sciences) (1982 to Sept 2008)
bull ISI Web of Science Social Sciences Citation Index (SSCI) 1970 to Sept 2008
bull ERIC (searched 25 Sept 2008)
bull ZETOC (searched 25 Sept 2008)
bull SIGLE (searched 25 Sept 2008)
All of the databases were searched using the following text-word search strategy The strategy was adapted using either rsquorsquo or rsquo$rsquo as atruncation symbol
1 motor-cycl or motorcycl or motor-bik or motorbik or motor-scooter or motorscooter or scooter or moped or motocycl
or motocicl or moto
2 train or educat or instruct or program or course or qualif or learn or improv
3 1 and 2
H I S T O R Y
Protocol first published Issue 2 2005
Review first published Issue 10 2010
46Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4950
C O N T R I B U T I O N S O F A U T H O R S
Katina Kardamanidis was the first author of the protocol conducted the literature search extracted full studies assessed the studies
according to inclusion criteria assessed quality of the studies compared findings with the second author discussed disagreements with
the third author wrote the drafts and the final review
Rebecca Ivers was the second author of the protocol assessed the studies according to inclusion criteria assessed quality of the studies
compared findings with the first author edited the final draft and provided overall guidance for the process
Alexandra Martiniuk assessed quality of the studies compared findings with the first author discussed disagreements with the third
author and helped write the first draft and the final review
Mark Stevenson contributed to the protocol reviewed methods of studies and contributed to the final draft
Katrina Thistlethwaite conducted a second literature search extracted full studies and contributed to the final draft
D E C L A R A T I O N S O F I N T E R E S T
None known
S O U R C E S O F S U P P O R T
Internal sources
bull No sources of support supplied
External sources
bull National Health and Medical Research Council of Australia Australia
Salary support for Martiniuk Ivers and Stevenson
bull University of Sydney Sesqui Research Grant Australia
Direct research costs and salary support for Kardamanidis
D I F F E R E N C E S B E T W E E N P R O T O C O L A N D R E V I E W
There are no differences between the protocol and the review
I N D E X T E R M S
Medical Subject Headings (MeSH)
lowastMotorcycles Accident Prevention [lowastmethods] Accidents Traffic [lowastprevention amp control] Licensure Program Evaluation Wounds
and Injuries [mortality lowastprevention amp control]
47Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 5050
MeSH check words
Humans
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3850
Leung 1987 (Continued)
Outcomes Police recorded crashes and offences
Length of follow up up to 3 years
Notes Study conducted in Wisconsin USA
MAIDS 2004
Methods Population based case-control study
Participants Cases Motorcycle riders who had a motorcycle crash Unclear whether all cases included
Community controls Motorcycle riders identified at petrol stations Unclear whether matched to cases
Initial sample size
Cases not reported
Controls not reported
Analysis sample size
Cases 921 (Italy 200 Netherlands 200 Spain 121 Germany 250 France 150)
Controls 923 (Italy 200 Netherlands 200 Spain123 Germany 250 France 150)
Interventions Pre-license rider training or additional training (non-compulsory)
Outcomes Crashes
Notes Study in five sampling areas located in five European countries (France Germany Netherlands Spain and Italy)
McDavid 1989
Methods Cohort study
Participants Sample size after first match
Intervention group 346 male training graduates
Control group 346 male motorcycle riders licensed in same year
Sample size after second match
Intervention group 139 (40)
Control group 139 (40)
Analysis sample size
Intervention group 139 (40)
Control group 139 (40)
Interventions No formal training vs the non-mandatory pre-licence British Columbia Safety Councilrsquos 37-hour motorcycle safety
programme
Intervention integrity
Participants had all completed and passed the training
Outcomes Police reported crashes injurious crashes
Length of follow up 1-6 years
36Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3950
McDavid 1989 (Continued)
Notes Study conducted in British Columbia Canada
Mortimer 1984
Methods Cohort study
Participants Initial sample size
Intervention group 608 MRC graduates
Control group 360 motorcycle riders
Response rates
Intervention group 360 (59)
Control group 324 (90)
Non-active riders excluded
Analysis sample size
Intervention group 213 (35)
Control group 303 (84)
Interventions No formal training vs the non-mandatory Motorcycle Safety Foundationrsquos Motorcycle Rider Course comprising of
8 hours theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-road practice
Intervention integrity
Participants had all completed and passed the MRC
Outcomes Self-reported crashes and offencesLength of follow up 1 year
Notes Study conducted in Illinois USA
Mortimer 1988
Methods Cohort study
Participants Initial sample size
Intervention group 2307 MRC graduates
Control group 556 active licensed motorcycle riders
Response ratesIntervention group 1298 (56)
Control group 500 (90)
Non-active riders excluded
Analysis sample size
Intervention group 913 (40)
Control group 500 (90)
Interventions No formal training vs the non-mandatory Motorcycle Safety Foundationrsquos Motorcycle Rider Course comprising of
8 hours theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-road practice
Intervention integrity
Participants had all completed and passed the MRC
37Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4050
Mortimer 1988 (Continued)
Outcomes Self reported crashes and offences
Length of follow up 1 year
Notes Study conducted in Illinois USA
Osga 1980
Methods Cohort study
Participants Initial sample sizeIntervention group 1163 MSF MRC participants invited 818 (74) responded
Control group 329 motorcycle riders responded to recruitment strategies
Non-active riders excluded from analysis
Analysis sample size
Intervention group 595 (73)
Control group 329 (100)
Interventions No formal training vs the non-mandatory Motorcycle Safety Foundationrsquos Motorcycle Rider Course (MRC) com-
prising of 8 hours theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-road
practice
Intervention integrity
Participants had all successfully completed the MRC
Outcomes Self-reported crashes police recorded crash reports
Length of follow up 6 months - 3 12 years
Notes Study conducted in South Dakota USA
Perrino 2002
Methods Cohort study
Participants Initial sample size
Intervention group 2144 (unclear) MRC graduates (motorcycle licence applicants)
Control group 2601 (unclear) regular licensing test graduates (motorcycle licence applicants)Analysis sample size
Intervention group 1070 (unclear)
Control group 867 (unclear)
Interventions No formal training vs the mandatory (for those younger than 18 years) pre-licence Maryland Motorcycle Rider
Course (MRC) consisting of 8 hours theory and 12 hours practice
Intervention integrity
No information given
Outcomes Self-reported and police recorded crashes and offences
Length of follow up 1 year
38Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4150
Perrino 2002 (Continued)
Notes Study conducted in Maryland USA
Raymond 1979
Methods Cohort study
Participants Initial sample size
Intervention group
258 volunteers mainly (but not all) novice riders
Control group
1104 provisional licence applicants
Analysis sample size
Intervention group 190 (74)
Control group 627 (57)
Interventions No formal training vs the non-mandatory RACACU (Royal Automobile ClubAuto-Cycle Union) training scheme
The training program consisted of either 24 one-hour sessions or 13 two-hour sessions Twelve hours of the training
focused on theory about basic operational procedures traffic laws and highway code The practical session focused
on basic operational skills riding on a training area and riding in traffic with an instructor A proficiency test was
administered at the end of the course
Intervention integrity
No information given
Outcomes Self-reported crashes (including minor non-injury non-damage and off-road crashes)
Length of follow up 1 year
Notes Study conducted in the UK
Satten 1980
Methods Cohort study
Participants Initial sample size
Intervention group interviewers kept interviewing until they had 100 riders selected from a population of 891
enrollees in the MSF MRC course (unknown how many were asked and how many refused to participate)Control group unknown number of registered motorcycle riders
Non-active riders excluded from analysis
Analysis sample size
Intervention group 71
Control group 100
Interventions No formal training vs the non-mandatory Motorcycle Safety Foundationrsquos (MSF) Motorcycle Rider Course (MRC)
comprising of 8 hours theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-
road practice
Intervention integrity
Participants had all successfully completed the MRC
39Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4250
Satten 1980 (Continued)
Outcomes Self-reported crashes injuries and offences
Length of follow up 3 years
Notes Study conducted in Illinois USA
Savolainen 2007
Methods Cohort study
Participants Initial sample size
Intervention group 4000 riders that had conducted the MSF MRC course (passed or failed)
Control group 4000 motorcycle owners or permit or endorsement holders from the Indiana Bureau of Motor
Vehicles database
Analysis sample size
Intervention group 739 (18)
Control group 588 (15)
Interventions Intervention integrity
No information given
Outcomes Probability of crash involvement
Length of follow up up to 5 years
Notes Study conducted in Indiana USA
Swaddiwudhipong 1998
Methods Randomised trial
Participants Initial sample size
Intervention group
all 16 villages of 3 sub-districts
(participants were practising motorcyclists licensed and unlicensed)
Control group
18 villages of 3 other sub-districtsAnalysis sample size
Intervention group 3 districts (100)
Control group 3 districts (100)
Interventions No rider education vs non-mandatory community based rider education consisting of theory on motorcycle injury
risk helmet use traffic laws vehicle regulations and traffic signs and a written and skill test leading to a licence
Intervention integrity
No information given
Outcomes Injuries fatalities
Length of follow up 2 years
40Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4350
Swaddiwudhipong 1998 (Continued)
Notes Study conducted in Tak Province rural Thailand
Waller 1992
Methods Cohort study
Participants Initial sample size
Intervention group 349 riders who succesfully completed the Department of Public Safety motorcycle operator
training in 1988
Control group 349 motorcyclists who never completed the DPS motorcycle operator training
Analysis sample size
Intervention group 349
Control group 349
Interventions No DPS motorcycle operator training vs non-mandatory DPS motorcycle operator training The DPS training is
the Motorcycle Safety Foundationrsquos Motorcycle Rider Course comprising eight hours of theory (lectures discussions
films and multiple choice tests) a textbook and 12 hours off-road practice
Intervention integrity
All 349 riders succesfully completed the rider training
Outcomes Police recorded crashes
Length of follow up 3 years
Notes Study conducted in
Characteristics of excluded studies [ordered by study ID]
Study Reason for exclusion
Anderson 1980a Study type review and process evaluation of a rider course
Aupetit 2008 Outcome proposal of a rider training curriculum that combines traditional training and a motorcycle riding
simulator
Awane 1999 Study type review on use of simulators in rider and driver training
Baldi 2005 Intervention not rider training but intervention integrity of rider training availabilty of rider education
implementation of graduated licensing system etc on a state-wide level
Batchler 1988 Study type description of a training course
Blanchard 2006 Study type not a study assessing the association between rider training and outcomes
Bonnett 2005 Outcome qualitative data regarding satisfaction with the training programme
41Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4450
(Continued)
Braver 2007 Outcome licensure rates
Clayton 1980 Outcome knowledge skills attitude
Council 1978 Outcome crash reduction in ANY vehicle not motorcycles exclusively
Crowther 2005 Study type qualitative study
Daniello 2009 Study type review of the effectiveness of motorcycle training and licensing
Dewanti 2007 Intervention emotional characteristics behaviour (ie not motocycle rider training)
Du Plooy 1990 Study type review of types of rider training courses in South Africa
Engel 1989 Study type uncontrolled before-after study
Goldenbeld 2002 Outcome skills
Goldenbeld 2004 Outcome knowledge and skills
Haworth 2000 Intervention balance between attitudinal andvehicle control skills acrosscurriculla of motorcycle rider training
courses
Haworth 2005 Study type review
Haworth 2006 Study type cross-sectional survey
Hill 1983 Study type review and description of training system
Imatake 1980 Study type course descriptionreview
Ingebritsen 1990 Intervention risk factors for motorcycle crashes
Jain 2009 Study type cross-sectional descriptive study of motorcycle crashes over a 5 year period
Jonah 1979 Study type cross-sectional survey Outcome scores on motorcycle operator skill test
Jonah 1981 Intervention evaluation of a motorcycle rider skill test not motorcycle rider training
Kadar 1990 Study type review of components of rider training courses
Kelsey 1986 Intervention evaluation of a motorcycle rider skill test not motorcycle rider training
Koch 1980 Study type reviewcourse description
Koch 1990 Study type description of course content
42Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4550
(Continued)
Labbett 2005 Study type essay about possible explanations for motorcycle deaths
Law 2005 Intervention targeted motorcycle safety programme not specifically rider training
Lowes 1990 Study type review and descriptions of rider training
Mahon 1976 Study type review of motorcycle rider training methods
Maxwell 1985 Outcomesrecommendations forprogramme operatorsand otherson howthe motorcycle trainingprogramme
works and how it can be improved
Mayhew 1996 Study type review
McKnight 1987 Study type evaluation of methodological and administrative aspects of a motorcycle operator skill test
McPherson 1976 Outcome knowledge and skills
Mohan 2004 Study type review of evidence of interventions to reduce road traffic injuries
Morgan 1980 Study type reviewcourse description
Motoki 1990 Study type proposal and development of rider training curriculum
Nairn 1992 Study type literature review
Newland 1980 Study type description of development of motorcycle safety course
Newland 1982 Study type descriptionreview of a rider training course
Newland 1990 Study type uncontrolled BeforeAfter study
Ormston 2003 Outcome changes in attitude and self reported behaviour
Osga 1979 Outcome knowledge attitudes and skills
Panou 2007 Intervention new simulation tool to teach driverrider
Peterson 1990 Study factor demographic details of course participants completing the course vs those not completing the
course
Prem 1984 Study type experimental
Study factor examines association between skills and performance in tests
Radin Umar 2006 Intervention motorcycle safety programme not specifically rider training
Rockwell 1990 Study type methodologically flawed study type
43Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4650
(Continued)
Rothe 1987 Study type qualitative study of rider training in general
Rowden 2007 Study type qualitative study
RSC Victoria 1993 Study type review of motorcycle rider training and education in Victoria
Russam 1979 Study type review of training programmes in Japan
Saffron 1981 Study type literature review
Schulz 1990 Study factor predictors of motorcycle crashes
Sexton 2004 Study type cross-sectional analytic analysis
Shephard 1986 Study factor reviewdescription of programme
Simpson 1990 Study type review of motorcycle rider trainingeducation
Sokolov 1990 Study factor review of motorcycle riding courses in USSR
Steffens 1988 Outcome self-reported behaviour experiences and researcher observed riding behaviour
Sudlow 2003 Study type review of motorcycle training in the UK
Tasmania 1995 Study type cross-sectional descriptiveprocess evaluation
Tetard 1985 Study type experimental
Outcome detection of indices
Thackray 1980 Outcome instructional effectiveness user acceptance administrative feasibility
Thompson 1994 Study type cross-sectional survey
Outcome attitudes and knowledge
Tobler 1990 Study type review of course curricula
Transport 1989 Intervention syllabus for rider training
Tuan 2007 Study type cross-sectional study (survey) which does not assess the effects of an intervention but examines
motocyclistsrsquo behaviour
Victor 1980 Study type review
Waterer 1990 Study type review of UK rider training issues
Weller 1989 Study type review of studies evaluating motorcycle safety and education measures
44Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4750
(Continued)
Wells 1981 Study type description of different methods of training
Winn 1986 Study type review of rsquohistoryrsquo of rider training
Wisher 1988 Study type cross-sectional survey
Wood 1987 Study type Interrupted Time Series but did not meet the criteria for an ITS as published by the Cochrane
Effective Practice and Organisation of Care Group
Yamazaki 1990 Study type cross-sectional survey
Outcome riding skills skills retention and attitude
Characteristics of ongoing studies [ordered by study ID]
Burgess 2010
Trial name or title UKs National Ride program
Methods Cohort study
Participants Motorcycle offenders
Interventions Police led diversion program
Outcomes Self reported attitudinal and behavioural measures
Starting date
Contact information
Notes Expected to be completed in early 2010
45Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4850
D A T A A N D A N A L Y S E S
This review has no analyses
A P P E N D I C E S
Appendix 1 Search strategy
We searched the following electronic databases
bull Cochrane Injuries Group Specialised Register (searched 22 Sept 2008)
bull CENTRAL (The Cochrane Library 2008 Issue 3)
bull TRANSPORT 1988 to 200706 (latest date available)
bull ATRI Australian Transport Index (to Sept 2008)
bull MEDLINE (Ovid SP) 1950 to Sept 2008
bull EMBASE (Ovid SP) 1980 to Sept 2008
bull CINAHL (EBSCO) 1982 to Sept 2008
bull WHOLIS (World Health Organization Library Information System) (to Sept 2008)
bull PsycINFO (Ovid SP) 1806 to Sept 2008
bull LILACS (BIREME) (Latin American and Caribbean Health Sciences) (1982 to Sept 2008)
bull ISI Web of Science Social Sciences Citation Index (SSCI) 1970 to Sept 2008
bull ERIC (searched 25 Sept 2008)
bull ZETOC (searched 25 Sept 2008)
bull SIGLE (searched 25 Sept 2008)
All of the databases were searched using the following text-word search strategy The strategy was adapted using either rsquorsquo or rsquo$rsquo as atruncation symbol
1 motor-cycl or motorcycl or motor-bik or motorbik or motor-scooter or motorscooter or scooter or moped or motocycl
or motocicl or moto
2 train or educat or instruct or program or course or qualif or learn or improv
3 1 and 2
H I S T O R Y
Protocol first published Issue 2 2005
Review first published Issue 10 2010
46Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4950
C O N T R I B U T I O N S O F A U T H O R S
Katina Kardamanidis was the first author of the protocol conducted the literature search extracted full studies assessed the studies
according to inclusion criteria assessed quality of the studies compared findings with the second author discussed disagreements with
the third author wrote the drafts and the final review
Rebecca Ivers was the second author of the protocol assessed the studies according to inclusion criteria assessed quality of the studies
compared findings with the first author edited the final draft and provided overall guidance for the process
Alexandra Martiniuk assessed quality of the studies compared findings with the first author discussed disagreements with the third
author and helped write the first draft and the final review
Mark Stevenson contributed to the protocol reviewed methods of studies and contributed to the final draft
Katrina Thistlethwaite conducted a second literature search extracted full studies and contributed to the final draft
D E C L A R A T I O N S O F I N T E R E S T
None known
S O U R C E S O F S U P P O R T
Internal sources
bull No sources of support supplied
External sources
bull National Health and Medical Research Council of Australia Australia
Salary support for Martiniuk Ivers and Stevenson
bull University of Sydney Sesqui Research Grant Australia
Direct research costs and salary support for Kardamanidis
D I F F E R E N C E S B E T W E E N P R O T O C O L A N D R E V I E W
There are no differences between the protocol and the review
I N D E X T E R M S
Medical Subject Headings (MeSH)
lowastMotorcycles Accident Prevention [lowastmethods] Accidents Traffic [lowastprevention amp control] Licensure Program Evaluation Wounds
and Injuries [mortality lowastprevention amp control]
47Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 5050
MeSH check words
Humans
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 3950
McDavid 1989 (Continued)
Notes Study conducted in British Columbia Canada
Mortimer 1984
Methods Cohort study
Participants Initial sample size
Intervention group 608 MRC graduates
Control group 360 motorcycle riders
Response rates
Intervention group 360 (59)
Control group 324 (90)
Non-active riders excluded
Analysis sample size
Intervention group 213 (35)
Control group 303 (84)
Interventions No formal training vs the non-mandatory Motorcycle Safety Foundationrsquos Motorcycle Rider Course comprising of
8 hours theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-road practice
Intervention integrity
Participants had all completed and passed the MRC
Outcomes Self-reported crashes and offencesLength of follow up 1 year
Notes Study conducted in Illinois USA
Mortimer 1988
Methods Cohort study
Participants Initial sample size
Intervention group 2307 MRC graduates
Control group 556 active licensed motorcycle riders
Response ratesIntervention group 1298 (56)
Control group 500 (90)
Non-active riders excluded
Analysis sample size
Intervention group 913 (40)
Control group 500 (90)
Interventions No formal training vs the non-mandatory Motorcycle Safety Foundationrsquos Motorcycle Rider Course comprising of
8 hours theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-road practice
Intervention integrity
Participants had all completed and passed the MRC
37Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4050
Mortimer 1988 (Continued)
Outcomes Self reported crashes and offences
Length of follow up 1 year
Notes Study conducted in Illinois USA
Osga 1980
Methods Cohort study
Participants Initial sample sizeIntervention group 1163 MSF MRC participants invited 818 (74) responded
Control group 329 motorcycle riders responded to recruitment strategies
Non-active riders excluded from analysis
Analysis sample size
Intervention group 595 (73)
Control group 329 (100)
Interventions No formal training vs the non-mandatory Motorcycle Safety Foundationrsquos Motorcycle Rider Course (MRC) com-
prising of 8 hours theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-road
practice
Intervention integrity
Participants had all successfully completed the MRC
Outcomes Self-reported crashes police recorded crash reports
Length of follow up 6 months - 3 12 years
Notes Study conducted in South Dakota USA
Perrino 2002
Methods Cohort study
Participants Initial sample size
Intervention group 2144 (unclear) MRC graduates (motorcycle licence applicants)
Control group 2601 (unclear) regular licensing test graduates (motorcycle licence applicants)Analysis sample size
Intervention group 1070 (unclear)
Control group 867 (unclear)
Interventions No formal training vs the mandatory (for those younger than 18 years) pre-licence Maryland Motorcycle Rider
Course (MRC) consisting of 8 hours theory and 12 hours practice
Intervention integrity
No information given
Outcomes Self-reported and police recorded crashes and offences
Length of follow up 1 year
38Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4150
Perrino 2002 (Continued)
Notes Study conducted in Maryland USA
Raymond 1979
Methods Cohort study
Participants Initial sample size
Intervention group
258 volunteers mainly (but not all) novice riders
Control group
1104 provisional licence applicants
Analysis sample size
Intervention group 190 (74)
Control group 627 (57)
Interventions No formal training vs the non-mandatory RACACU (Royal Automobile ClubAuto-Cycle Union) training scheme
The training program consisted of either 24 one-hour sessions or 13 two-hour sessions Twelve hours of the training
focused on theory about basic operational procedures traffic laws and highway code The practical session focused
on basic operational skills riding on a training area and riding in traffic with an instructor A proficiency test was
administered at the end of the course
Intervention integrity
No information given
Outcomes Self-reported crashes (including minor non-injury non-damage and off-road crashes)
Length of follow up 1 year
Notes Study conducted in the UK
Satten 1980
Methods Cohort study
Participants Initial sample size
Intervention group interviewers kept interviewing until they had 100 riders selected from a population of 891
enrollees in the MSF MRC course (unknown how many were asked and how many refused to participate)Control group unknown number of registered motorcycle riders
Non-active riders excluded from analysis
Analysis sample size
Intervention group 71
Control group 100
Interventions No formal training vs the non-mandatory Motorcycle Safety Foundationrsquos (MSF) Motorcycle Rider Course (MRC)
comprising of 8 hours theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-
road practice
Intervention integrity
Participants had all successfully completed the MRC
39Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4250
Satten 1980 (Continued)
Outcomes Self-reported crashes injuries and offences
Length of follow up 3 years
Notes Study conducted in Illinois USA
Savolainen 2007
Methods Cohort study
Participants Initial sample size
Intervention group 4000 riders that had conducted the MSF MRC course (passed or failed)
Control group 4000 motorcycle owners or permit or endorsement holders from the Indiana Bureau of Motor
Vehicles database
Analysis sample size
Intervention group 739 (18)
Control group 588 (15)
Interventions Intervention integrity
No information given
Outcomes Probability of crash involvement
Length of follow up up to 5 years
Notes Study conducted in Indiana USA
Swaddiwudhipong 1998
Methods Randomised trial
Participants Initial sample size
Intervention group
all 16 villages of 3 sub-districts
(participants were practising motorcyclists licensed and unlicensed)
Control group
18 villages of 3 other sub-districtsAnalysis sample size
Intervention group 3 districts (100)
Control group 3 districts (100)
Interventions No rider education vs non-mandatory community based rider education consisting of theory on motorcycle injury
risk helmet use traffic laws vehicle regulations and traffic signs and a written and skill test leading to a licence
Intervention integrity
No information given
Outcomes Injuries fatalities
Length of follow up 2 years
40Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4350
Swaddiwudhipong 1998 (Continued)
Notes Study conducted in Tak Province rural Thailand
Waller 1992
Methods Cohort study
Participants Initial sample size
Intervention group 349 riders who succesfully completed the Department of Public Safety motorcycle operator
training in 1988
Control group 349 motorcyclists who never completed the DPS motorcycle operator training
Analysis sample size
Intervention group 349
Control group 349
Interventions No DPS motorcycle operator training vs non-mandatory DPS motorcycle operator training The DPS training is
the Motorcycle Safety Foundationrsquos Motorcycle Rider Course comprising eight hours of theory (lectures discussions
films and multiple choice tests) a textbook and 12 hours off-road practice
Intervention integrity
All 349 riders succesfully completed the rider training
Outcomes Police recorded crashes
Length of follow up 3 years
Notes Study conducted in
Characteristics of excluded studies [ordered by study ID]
Study Reason for exclusion
Anderson 1980a Study type review and process evaluation of a rider course
Aupetit 2008 Outcome proposal of a rider training curriculum that combines traditional training and a motorcycle riding
simulator
Awane 1999 Study type review on use of simulators in rider and driver training
Baldi 2005 Intervention not rider training but intervention integrity of rider training availabilty of rider education
implementation of graduated licensing system etc on a state-wide level
Batchler 1988 Study type description of a training course
Blanchard 2006 Study type not a study assessing the association between rider training and outcomes
Bonnett 2005 Outcome qualitative data regarding satisfaction with the training programme
41Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4450
(Continued)
Braver 2007 Outcome licensure rates
Clayton 1980 Outcome knowledge skills attitude
Council 1978 Outcome crash reduction in ANY vehicle not motorcycles exclusively
Crowther 2005 Study type qualitative study
Daniello 2009 Study type review of the effectiveness of motorcycle training and licensing
Dewanti 2007 Intervention emotional characteristics behaviour (ie not motocycle rider training)
Du Plooy 1990 Study type review of types of rider training courses in South Africa
Engel 1989 Study type uncontrolled before-after study
Goldenbeld 2002 Outcome skills
Goldenbeld 2004 Outcome knowledge and skills
Haworth 2000 Intervention balance between attitudinal andvehicle control skills acrosscurriculla of motorcycle rider training
courses
Haworth 2005 Study type review
Haworth 2006 Study type cross-sectional survey
Hill 1983 Study type review and description of training system
Imatake 1980 Study type course descriptionreview
Ingebritsen 1990 Intervention risk factors for motorcycle crashes
Jain 2009 Study type cross-sectional descriptive study of motorcycle crashes over a 5 year period
Jonah 1979 Study type cross-sectional survey Outcome scores on motorcycle operator skill test
Jonah 1981 Intervention evaluation of a motorcycle rider skill test not motorcycle rider training
Kadar 1990 Study type review of components of rider training courses
Kelsey 1986 Intervention evaluation of a motorcycle rider skill test not motorcycle rider training
Koch 1980 Study type reviewcourse description
Koch 1990 Study type description of course content
42Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4550
(Continued)
Labbett 2005 Study type essay about possible explanations for motorcycle deaths
Law 2005 Intervention targeted motorcycle safety programme not specifically rider training
Lowes 1990 Study type review and descriptions of rider training
Mahon 1976 Study type review of motorcycle rider training methods
Maxwell 1985 Outcomesrecommendations forprogramme operatorsand otherson howthe motorcycle trainingprogramme
works and how it can be improved
Mayhew 1996 Study type review
McKnight 1987 Study type evaluation of methodological and administrative aspects of a motorcycle operator skill test
McPherson 1976 Outcome knowledge and skills
Mohan 2004 Study type review of evidence of interventions to reduce road traffic injuries
Morgan 1980 Study type reviewcourse description
Motoki 1990 Study type proposal and development of rider training curriculum
Nairn 1992 Study type literature review
Newland 1980 Study type description of development of motorcycle safety course
Newland 1982 Study type descriptionreview of a rider training course
Newland 1990 Study type uncontrolled BeforeAfter study
Ormston 2003 Outcome changes in attitude and self reported behaviour
Osga 1979 Outcome knowledge attitudes and skills
Panou 2007 Intervention new simulation tool to teach driverrider
Peterson 1990 Study factor demographic details of course participants completing the course vs those not completing the
course
Prem 1984 Study type experimental
Study factor examines association between skills and performance in tests
Radin Umar 2006 Intervention motorcycle safety programme not specifically rider training
Rockwell 1990 Study type methodologically flawed study type
43Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4650
(Continued)
Rothe 1987 Study type qualitative study of rider training in general
Rowden 2007 Study type qualitative study
RSC Victoria 1993 Study type review of motorcycle rider training and education in Victoria
Russam 1979 Study type review of training programmes in Japan
Saffron 1981 Study type literature review
Schulz 1990 Study factor predictors of motorcycle crashes
Sexton 2004 Study type cross-sectional analytic analysis
Shephard 1986 Study factor reviewdescription of programme
Simpson 1990 Study type review of motorcycle rider trainingeducation
Sokolov 1990 Study factor review of motorcycle riding courses in USSR
Steffens 1988 Outcome self-reported behaviour experiences and researcher observed riding behaviour
Sudlow 2003 Study type review of motorcycle training in the UK
Tasmania 1995 Study type cross-sectional descriptiveprocess evaluation
Tetard 1985 Study type experimental
Outcome detection of indices
Thackray 1980 Outcome instructional effectiveness user acceptance administrative feasibility
Thompson 1994 Study type cross-sectional survey
Outcome attitudes and knowledge
Tobler 1990 Study type review of course curricula
Transport 1989 Intervention syllabus for rider training
Tuan 2007 Study type cross-sectional study (survey) which does not assess the effects of an intervention but examines
motocyclistsrsquo behaviour
Victor 1980 Study type review
Waterer 1990 Study type review of UK rider training issues
Weller 1989 Study type review of studies evaluating motorcycle safety and education measures
44Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4750
(Continued)
Wells 1981 Study type description of different methods of training
Winn 1986 Study type review of rsquohistoryrsquo of rider training
Wisher 1988 Study type cross-sectional survey
Wood 1987 Study type Interrupted Time Series but did not meet the criteria for an ITS as published by the Cochrane
Effective Practice and Organisation of Care Group
Yamazaki 1990 Study type cross-sectional survey
Outcome riding skills skills retention and attitude
Characteristics of ongoing studies [ordered by study ID]
Burgess 2010
Trial name or title UKs National Ride program
Methods Cohort study
Participants Motorcycle offenders
Interventions Police led diversion program
Outcomes Self reported attitudinal and behavioural measures
Starting date
Contact information
Notes Expected to be completed in early 2010
45Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4850
D A T A A N D A N A L Y S E S
This review has no analyses
A P P E N D I C E S
Appendix 1 Search strategy
We searched the following electronic databases
bull Cochrane Injuries Group Specialised Register (searched 22 Sept 2008)
bull CENTRAL (The Cochrane Library 2008 Issue 3)
bull TRANSPORT 1988 to 200706 (latest date available)
bull ATRI Australian Transport Index (to Sept 2008)
bull MEDLINE (Ovid SP) 1950 to Sept 2008
bull EMBASE (Ovid SP) 1980 to Sept 2008
bull CINAHL (EBSCO) 1982 to Sept 2008
bull WHOLIS (World Health Organization Library Information System) (to Sept 2008)
bull PsycINFO (Ovid SP) 1806 to Sept 2008
bull LILACS (BIREME) (Latin American and Caribbean Health Sciences) (1982 to Sept 2008)
bull ISI Web of Science Social Sciences Citation Index (SSCI) 1970 to Sept 2008
bull ERIC (searched 25 Sept 2008)
bull ZETOC (searched 25 Sept 2008)
bull SIGLE (searched 25 Sept 2008)
All of the databases were searched using the following text-word search strategy The strategy was adapted using either rsquorsquo or rsquo$rsquo as atruncation symbol
1 motor-cycl or motorcycl or motor-bik or motorbik or motor-scooter or motorscooter or scooter or moped or motocycl
or motocicl or moto
2 train or educat or instruct or program or course or qualif or learn or improv
3 1 and 2
H I S T O R Y
Protocol first published Issue 2 2005
Review first published Issue 10 2010
46Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4950
C O N T R I B U T I O N S O F A U T H O R S
Katina Kardamanidis was the first author of the protocol conducted the literature search extracted full studies assessed the studies
according to inclusion criteria assessed quality of the studies compared findings with the second author discussed disagreements with
the third author wrote the drafts and the final review
Rebecca Ivers was the second author of the protocol assessed the studies according to inclusion criteria assessed quality of the studies
compared findings with the first author edited the final draft and provided overall guidance for the process
Alexandra Martiniuk assessed quality of the studies compared findings with the first author discussed disagreements with the third
author and helped write the first draft and the final review
Mark Stevenson contributed to the protocol reviewed methods of studies and contributed to the final draft
Katrina Thistlethwaite conducted a second literature search extracted full studies and contributed to the final draft
D E C L A R A T I O N S O F I N T E R E S T
None known
S O U R C E S O F S U P P O R T
Internal sources
bull No sources of support supplied
External sources
bull National Health and Medical Research Council of Australia Australia
Salary support for Martiniuk Ivers and Stevenson
bull University of Sydney Sesqui Research Grant Australia
Direct research costs and salary support for Kardamanidis
D I F F E R E N C E S B E T W E E N P R O T O C O L A N D R E V I E W
There are no differences between the protocol and the review
I N D E X T E R M S
Medical Subject Headings (MeSH)
lowastMotorcycles Accident Prevention [lowastmethods] Accidents Traffic [lowastprevention amp control] Licensure Program Evaluation Wounds
and Injuries [mortality lowastprevention amp control]
47Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 5050
MeSH check words
Humans
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4050
Mortimer 1988 (Continued)
Outcomes Self reported crashes and offences
Length of follow up 1 year
Notes Study conducted in Illinois USA
Osga 1980
Methods Cohort study
Participants Initial sample sizeIntervention group 1163 MSF MRC participants invited 818 (74) responded
Control group 329 motorcycle riders responded to recruitment strategies
Non-active riders excluded from analysis
Analysis sample size
Intervention group 595 (73)
Control group 329 (100)
Interventions No formal training vs the non-mandatory Motorcycle Safety Foundationrsquos Motorcycle Rider Course (MRC) com-
prising of 8 hours theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-road
practice
Intervention integrity
Participants had all successfully completed the MRC
Outcomes Self-reported crashes police recorded crash reports
Length of follow up 6 months - 3 12 years
Notes Study conducted in South Dakota USA
Perrino 2002
Methods Cohort study
Participants Initial sample size
Intervention group 2144 (unclear) MRC graduates (motorcycle licence applicants)
Control group 2601 (unclear) regular licensing test graduates (motorcycle licence applicants)Analysis sample size
Intervention group 1070 (unclear)
Control group 867 (unclear)
Interventions No formal training vs the mandatory (for those younger than 18 years) pre-licence Maryland Motorcycle Rider
Course (MRC) consisting of 8 hours theory and 12 hours practice
Intervention integrity
No information given
Outcomes Self-reported and police recorded crashes and offences
Length of follow up 1 year
38Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4150
Perrino 2002 (Continued)
Notes Study conducted in Maryland USA
Raymond 1979
Methods Cohort study
Participants Initial sample size
Intervention group
258 volunteers mainly (but not all) novice riders
Control group
1104 provisional licence applicants
Analysis sample size
Intervention group 190 (74)
Control group 627 (57)
Interventions No formal training vs the non-mandatory RACACU (Royal Automobile ClubAuto-Cycle Union) training scheme
The training program consisted of either 24 one-hour sessions or 13 two-hour sessions Twelve hours of the training
focused on theory about basic operational procedures traffic laws and highway code The practical session focused
on basic operational skills riding on a training area and riding in traffic with an instructor A proficiency test was
administered at the end of the course
Intervention integrity
No information given
Outcomes Self-reported crashes (including minor non-injury non-damage and off-road crashes)
Length of follow up 1 year
Notes Study conducted in the UK
Satten 1980
Methods Cohort study
Participants Initial sample size
Intervention group interviewers kept interviewing until they had 100 riders selected from a population of 891
enrollees in the MSF MRC course (unknown how many were asked and how many refused to participate)Control group unknown number of registered motorcycle riders
Non-active riders excluded from analysis
Analysis sample size
Intervention group 71
Control group 100
Interventions No formal training vs the non-mandatory Motorcycle Safety Foundationrsquos (MSF) Motorcycle Rider Course (MRC)
comprising of 8 hours theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-
road practice
Intervention integrity
Participants had all successfully completed the MRC
39Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4250
Satten 1980 (Continued)
Outcomes Self-reported crashes injuries and offences
Length of follow up 3 years
Notes Study conducted in Illinois USA
Savolainen 2007
Methods Cohort study
Participants Initial sample size
Intervention group 4000 riders that had conducted the MSF MRC course (passed or failed)
Control group 4000 motorcycle owners or permit or endorsement holders from the Indiana Bureau of Motor
Vehicles database
Analysis sample size
Intervention group 739 (18)
Control group 588 (15)
Interventions Intervention integrity
No information given
Outcomes Probability of crash involvement
Length of follow up up to 5 years
Notes Study conducted in Indiana USA
Swaddiwudhipong 1998
Methods Randomised trial
Participants Initial sample size
Intervention group
all 16 villages of 3 sub-districts
(participants were practising motorcyclists licensed and unlicensed)
Control group
18 villages of 3 other sub-districtsAnalysis sample size
Intervention group 3 districts (100)
Control group 3 districts (100)
Interventions No rider education vs non-mandatory community based rider education consisting of theory on motorcycle injury
risk helmet use traffic laws vehicle regulations and traffic signs and a written and skill test leading to a licence
Intervention integrity
No information given
Outcomes Injuries fatalities
Length of follow up 2 years
40Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4350
Swaddiwudhipong 1998 (Continued)
Notes Study conducted in Tak Province rural Thailand
Waller 1992
Methods Cohort study
Participants Initial sample size
Intervention group 349 riders who succesfully completed the Department of Public Safety motorcycle operator
training in 1988
Control group 349 motorcyclists who never completed the DPS motorcycle operator training
Analysis sample size
Intervention group 349
Control group 349
Interventions No DPS motorcycle operator training vs non-mandatory DPS motorcycle operator training The DPS training is
the Motorcycle Safety Foundationrsquos Motorcycle Rider Course comprising eight hours of theory (lectures discussions
films and multiple choice tests) a textbook and 12 hours off-road practice
Intervention integrity
All 349 riders succesfully completed the rider training
Outcomes Police recorded crashes
Length of follow up 3 years
Notes Study conducted in
Characteristics of excluded studies [ordered by study ID]
Study Reason for exclusion
Anderson 1980a Study type review and process evaluation of a rider course
Aupetit 2008 Outcome proposal of a rider training curriculum that combines traditional training and a motorcycle riding
simulator
Awane 1999 Study type review on use of simulators in rider and driver training
Baldi 2005 Intervention not rider training but intervention integrity of rider training availabilty of rider education
implementation of graduated licensing system etc on a state-wide level
Batchler 1988 Study type description of a training course
Blanchard 2006 Study type not a study assessing the association between rider training and outcomes
Bonnett 2005 Outcome qualitative data regarding satisfaction with the training programme
41Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4450
(Continued)
Braver 2007 Outcome licensure rates
Clayton 1980 Outcome knowledge skills attitude
Council 1978 Outcome crash reduction in ANY vehicle not motorcycles exclusively
Crowther 2005 Study type qualitative study
Daniello 2009 Study type review of the effectiveness of motorcycle training and licensing
Dewanti 2007 Intervention emotional characteristics behaviour (ie not motocycle rider training)
Du Plooy 1990 Study type review of types of rider training courses in South Africa
Engel 1989 Study type uncontrolled before-after study
Goldenbeld 2002 Outcome skills
Goldenbeld 2004 Outcome knowledge and skills
Haworth 2000 Intervention balance between attitudinal andvehicle control skills acrosscurriculla of motorcycle rider training
courses
Haworth 2005 Study type review
Haworth 2006 Study type cross-sectional survey
Hill 1983 Study type review and description of training system
Imatake 1980 Study type course descriptionreview
Ingebritsen 1990 Intervention risk factors for motorcycle crashes
Jain 2009 Study type cross-sectional descriptive study of motorcycle crashes over a 5 year period
Jonah 1979 Study type cross-sectional survey Outcome scores on motorcycle operator skill test
Jonah 1981 Intervention evaluation of a motorcycle rider skill test not motorcycle rider training
Kadar 1990 Study type review of components of rider training courses
Kelsey 1986 Intervention evaluation of a motorcycle rider skill test not motorcycle rider training
Koch 1980 Study type reviewcourse description
Koch 1990 Study type description of course content
42Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4550
(Continued)
Labbett 2005 Study type essay about possible explanations for motorcycle deaths
Law 2005 Intervention targeted motorcycle safety programme not specifically rider training
Lowes 1990 Study type review and descriptions of rider training
Mahon 1976 Study type review of motorcycle rider training methods
Maxwell 1985 Outcomesrecommendations forprogramme operatorsand otherson howthe motorcycle trainingprogramme
works and how it can be improved
Mayhew 1996 Study type review
McKnight 1987 Study type evaluation of methodological and administrative aspects of a motorcycle operator skill test
McPherson 1976 Outcome knowledge and skills
Mohan 2004 Study type review of evidence of interventions to reduce road traffic injuries
Morgan 1980 Study type reviewcourse description
Motoki 1990 Study type proposal and development of rider training curriculum
Nairn 1992 Study type literature review
Newland 1980 Study type description of development of motorcycle safety course
Newland 1982 Study type descriptionreview of a rider training course
Newland 1990 Study type uncontrolled BeforeAfter study
Ormston 2003 Outcome changes in attitude and self reported behaviour
Osga 1979 Outcome knowledge attitudes and skills
Panou 2007 Intervention new simulation tool to teach driverrider
Peterson 1990 Study factor demographic details of course participants completing the course vs those not completing the
course
Prem 1984 Study type experimental
Study factor examines association between skills and performance in tests
Radin Umar 2006 Intervention motorcycle safety programme not specifically rider training
Rockwell 1990 Study type methodologically flawed study type
43Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4650
(Continued)
Rothe 1987 Study type qualitative study of rider training in general
Rowden 2007 Study type qualitative study
RSC Victoria 1993 Study type review of motorcycle rider training and education in Victoria
Russam 1979 Study type review of training programmes in Japan
Saffron 1981 Study type literature review
Schulz 1990 Study factor predictors of motorcycle crashes
Sexton 2004 Study type cross-sectional analytic analysis
Shephard 1986 Study factor reviewdescription of programme
Simpson 1990 Study type review of motorcycle rider trainingeducation
Sokolov 1990 Study factor review of motorcycle riding courses in USSR
Steffens 1988 Outcome self-reported behaviour experiences and researcher observed riding behaviour
Sudlow 2003 Study type review of motorcycle training in the UK
Tasmania 1995 Study type cross-sectional descriptiveprocess evaluation
Tetard 1985 Study type experimental
Outcome detection of indices
Thackray 1980 Outcome instructional effectiveness user acceptance administrative feasibility
Thompson 1994 Study type cross-sectional survey
Outcome attitudes and knowledge
Tobler 1990 Study type review of course curricula
Transport 1989 Intervention syllabus for rider training
Tuan 2007 Study type cross-sectional study (survey) which does not assess the effects of an intervention but examines
motocyclistsrsquo behaviour
Victor 1980 Study type review
Waterer 1990 Study type review of UK rider training issues
Weller 1989 Study type review of studies evaluating motorcycle safety and education measures
44Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4750
(Continued)
Wells 1981 Study type description of different methods of training
Winn 1986 Study type review of rsquohistoryrsquo of rider training
Wisher 1988 Study type cross-sectional survey
Wood 1987 Study type Interrupted Time Series but did not meet the criteria for an ITS as published by the Cochrane
Effective Practice and Organisation of Care Group
Yamazaki 1990 Study type cross-sectional survey
Outcome riding skills skills retention and attitude
Characteristics of ongoing studies [ordered by study ID]
Burgess 2010
Trial name or title UKs National Ride program
Methods Cohort study
Participants Motorcycle offenders
Interventions Police led diversion program
Outcomes Self reported attitudinal and behavioural measures
Starting date
Contact information
Notes Expected to be completed in early 2010
45Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4850
D A T A A N D A N A L Y S E S
This review has no analyses
A P P E N D I C E S
Appendix 1 Search strategy
We searched the following electronic databases
bull Cochrane Injuries Group Specialised Register (searched 22 Sept 2008)
bull CENTRAL (The Cochrane Library 2008 Issue 3)
bull TRANSPORT 1988 to 200706 (latest date available)
bull ATRI Australian Transport Index (to Sept 2008)
bull MEDLINE (Ovid SP) 1950 to Sept 2008
bull EMBASE (Ovid SP) 1980 to Sept 2008
bull CINAHL (EBSCO) 1982 to Sept 2008
bull WHOLIS (World Health Organization Library Information System) (to Sept 2008)
bull PsycINFO (Ovid SP) 1806 to Sept 2008
bull LILACS (BIREME) (Latin American and Caribbean Health Sciences) (1982 to Sept 2008)
bull ISI Web of Science Social Sciences Citation Index (SSCI) 1970 to Sept 2008
bull ERIC (searched 25 Sept 2008)
bull ZETOC (searched 25 Sept 2008)
bull SIGLE (searched 25 Sept 2008)
All of the databases were searched using the following text-word search strategy The strategy was adapted using either rsquorsquo or rsquo$rsquo as atruncation symbol
1 motor-cycl or motorcycl or motor-bik or motorbik or motor-scooter or motorscooter or scooter or moped or motocycl
or motocicl or moto
2 train or educat or instruct or program or course or qualif or learn or improv
3 1 and 2
H I S T O R Y
Protocol first published Issue 2 2005
Review first published Issue 10 2010
46Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4950
C O N T R I B U T I O N S O F A U T H O R S
Katina Kardamanidis was the first author of the protocol conducted the literature search extracted full studies assessed the studies
according to inclusion criteria assessed quality of the studies compared findings with the second author discussed disagreements with
the third author wrote the drafts and the final review
Rebecca Ivers was the second author of the protocol assessed the studies according to inclusion criteria assessed quality of the studies
compared findings with the first author edited the final draft and provided overall guidance for the process
Alexandra Martiniuk assessed quality of the studies compared findings with the first author discussed disagreements with the third
author and helped write the first draft and the final review
Mark Stevenson contributed to the protocol reviewed methods of studies and contributed to the final draft
Katrina Thistlethwaite conducted a second literature search extracted full studies and contributed to the final draft
D E C L A R A T I O N S O F I N T E R E S T
None known
S O U R C E S O F S U P P O R T
Internal sources
bull No sources of support supplied
External sources
bull National Health and Medical Research Council of Australia Australia
Salary support for Martiniuk Ivers and Stevenson
bull University of Sydney Sesqui Research Grant Australia
Direct research costs and salary support for Kardamanidis
D I F F E R E N C E S B E T W E E N P R O T O C O L A N D R E V I E W
There are no differences between the protocol and the review
I N D E X T E R M S
Medical Subject Headings (MeSH)
lowastMotorcycles Accident Prevention [lowastmethods] Accidents Traffic [lowastprevention amp control] Licensure Program Evaluation Wounds
and Injuries [mortality lowastprevention amp control]
47Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 5050
MeSH check words
Humans
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4150
Perrino 2002 (Continued)
Notes Study conducted in Maryland USA
Raymond 1979
Methods Cohort study
Participants Initial sample size
Intervention group
258 volunteers mainly (but not all) novice riders
Control group
1104 provisional licence applicants
Analysis sample size
Intervention group 190 (74)
Control group 627 (57)
Interventions No formal training vs the non-mandatory RACACU (Royal Automobile ClubAuto-Cycle Union) training scheme
The training program consisted of either 24 one-hour sessions or 13 two-hour sessions Twelve hours of the training
focused on theory about basic operational procedures traffic laws and highway code The practical session focused
on basic operational skills riding on a training area and riding in traffic with an instructor A proficiency test was
administered at the end of the course
Intervention integrity
No information given
Outcomes Self-reported crashes (including minor non-injury non-damage and off-road crashes)
Length of follow up 1 year
Notes Study conducted in the UK
Satten 1980
Methods Cohort study
Participants Initial sample size
Intervention group interviewers kept interviewing until they had 100 riders selected from a population of 891
enrollees in the MSF MRC course (unknown how many were asked and how many refused to participate)Control group unknown number of registered motorcycle riders
Non-active riders excluded from analysis
Analysis sample size
Intervention group 71
Control group 100
Interventions No formal training vs the non-mandatory Motorcycle Safety Foundationrsquos (MSF) Motorcycle Rider Course (MRC)
comprising of 8 hours theory (lectures discussions films and multiple choice tests) a textbook and 12 hours off-
road practice
Intervention integrity
Participants had all successfully completed the MRC
39Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4250
Satten 1980 (Continued)
Outcomes Self-reported crashes injuries and offences
Length of follow up 3 years
Notes Study conducted in Illinois USA
Savolainen 2007
Methods Cohort study
Participants Initial sample size
Intervention group 4000 riders that had conducted the MSF MRC course (passed or failed)
Control group 4000 motorcycle owners or permit or endorsement holders from the Indiana Bureau of Motor
Vehicles database
Analysis sample size
Intervention group 739 (18)
Control group 588 (15)
Interventions Intervention integrity
No information given
Outcomes Probability of crash involvement
Length of follow up up to 5 years
Notes Study conducted in Indiana USA
Swaddiwudhipong 1998
Methods Randomised trial
Participants Initial sample size
Intervention group
all 16 villages of 3 sub-districts
(participants were practising motorcyclists licensed and unlicensed)
Control group
18 villages of 3 other sub-districtsAnalysis sample size
Intervention group 3 districts (100)
Control group 3 districts (100)
Interventions No rider education vs non-mandatory community based rider education consisting of theory on motorcycle injury
risk helmet use traffic laws vehicle regulations and traffic signs and a written and skill test leading to a licence
Intervention integrity
No information given
Outcomes Injuries fatalities
Length of follow up 2 years
40Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4350
Swaddiwudhipong 1998 (Continued)
Notes Study conducted in Tak Province rural Thailand
Waller 1992
Methods Cohort study
Participants Initial sample size
Intervention group 349 riders who succesfully completed the Department of Public Safety motorcycle operator
training in 1988
Control group 349 motorcyclists who never completed the DPS motorcycle operator training
Analysis sample size
Intervention group 349
Control group 349
Interventions No DPS motorcycle operator training vs non-mandatory DPS motorcycle operator training The DPS training is
the Motorcycle Safety Foundationrsquos Motorcycle Rider Course comprising eight hours of theory (lectures discussions
films and multiple choice tests) a textbook and 12 hours off-road practice
Intervention integrity
All 349 riders succesfully completed the rider training
Outcomes Police recorded crashes
Length of follow up 3 years
Notes Study conducted in
Characteristics of excluded studies [ordered by study ID]
Study Reason for exclusion
Anderson 1980a Study type review and process evaluation of a rider course
Aupetit 2008 Outcome proposal of a rider training curriculum that combines traditional training and a motorcycle riding
simulator
Awane 1999 Study type review on use of simulators in rider and driver training
Baldi 2005 Intervention not rider training but intervention integrity of rider training availabilty of rider education
implementation of graduated licensing system etc on a state-wide level
Batchler 1988 Study type description of a training course
Blanchard 2006 Study type not a study assessing the association between rider training and outcomes
Bonnett 2005 Outcome qualitative data regarding satisfaction with the training programme
41Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4450
(Continued)
Braver 2007 Outcome licensure rates
Clayton 1980 Outcome knowledge skills attitude
Council 1978 Outcome crash reduction in ANY vehicle not motorcycles exclusively
Crowther 2005 Study type qualitative study
Daniello 2009 Study type review of the effectiveness of motorcycle training and licensing
Dewanti 2007 Intervention emotional characteristics behaviour (ie not motocycle rider training)
Du Plooy 1990 Study type review of types of rider training courses in South Africa
Engel 1989 Study type uncontrolled before-after study
Goldenbeld 2002 Outcome skills
Goldenbeld 2004 Outcome knowledge and skills
Haworth 2000 Intervention balance between attitudinal andvehicle control skills acrosscurriculla of motorcycle rider training
courses
Haworth 2005 Study type review
Haworth 2006 Study type cross-sectional survey
Hill 1983 Study type review and description of training system
Imatake 1980 Study type course descriptionreview
Ingebritsen 1990 Intervention risk factors for motorcycle crashes
Jain 2009 Study type cross-sectional descriptive study of motorcycle crashes over a 5 year period
Jonah 1979 Study type cross-sectional survey Outcome scores on motorcycle operator skill test
Jonah 1981 Intervention evaluation of a motorcycle rider skill test not motorcycle rider training
Kadar 1990 Study type review of components of rider training courses
Kelsey 1986 Intervention evaluation of a motorcycle rider skill test not motorcycle rider training
Koch 1980 Study type reviewcourse description
Koch 1990 Study type description of course content
42Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4550
(Continued)
Labbett 2005 Study type essay about possible explanations for motorcycle deaths
Law 2005 Intervention targeted motorcycle safety programme not specifically rider training
Lowes 1990 Study type review and descriptions of rider training
Mahon 1976 Study type review of motorcycle rider training methods
Maxwell 1985 Outcomesrecommendations forprogramme operatorsand otherson howthe motorcycle trainingprogramme
works and how it can be improved
Mayhew 1996 Study type review
McKnight 1987 Study type evaluation of methodological and administrative aspects of a motorcycle operator skill test
McPherson 1976 Outcome knowledge and skills
Mohan 2004 Study type review of evidence of interventions to reduce road traffic injuries
Morgan 1980 Study type reviewcourse description
Motoki 1990 Study type proposal and development of rider training curriculum
Nairn 1992 Study type literature review
Newland 1980 Study type description of development of motorcycle safety course
Newland 1982 Study type descriptionreview of a rider training course
Newland 1990 Study type uncontrolled BeforeAfter study
Ormston 2003 Outcome changes in attitude and self reported behaviour
Osga 1979 Outcome knowledge attitudes and skills
Panou 2007 Intervention new simulation tool to teach driverrider
Peterson 1990 Study factor demographic details of course participants completing the course vs those not completing the
course
Prem 1984 Study type experimental
Study factor examines association between skills and performance in tests
Radin Umar 2006 Intervention motorcycle safety programme not specifically rider training
Rockwell 1990 Study type methodologically flawed study type
43Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4650
(Continued)
Rothe 1987 Study type qualitative study of rider training in general
Rowden 2007 Study type qualitative study
RSC Victoria 1993 Study type review of motorcycle rider training and education in Victoria
Russam 1979 Study type review of training programmes in Japan
Saffron 1981 Study type literature review
Schulz 1990 Study factor predictors of motorcycle crashes
Sexton 2004 Study type cross-sectional analytic analysis
Shephard 1986 Study factor reviewdescription of programme
Simpson 1990 Study type review of motorcycle rider trainingeducation
Sokolov 1990 Study factor review of motorcycle riding courses in USSR
Steffens 1988 Outcome self-reported behaviour experiences and researcher observed riding behaviour
Sudlow 2003 Study type review of motorcycle training in the UK
Tasmania 1995 Study type cross-sectional descriptiveprocess evaluation
Tetard 1985 Study type experimental
Outcome detection of indices
Thackray 1980 Outcome instructional effectiveness user acceptance administrative feasibility
Thompson 1994 Study type cross-sectional survey
Outcome attitudes and knowledge
Tobler 1990 Study type review of course curricula
Transport 1989 Intervention syllabus for rider training
Tuan 2007 Study type cross-sectional study (survey) which does not assess the effects of an intervention but examines
motocyclistsrsquo behaviour
Victor 1980 Study type review
Waterer 1990 Study type review of UK rider training issues
Weller 1989 Study type review of studies evaluating motorcycle safety and education measures
44Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4750
(Continued)
Wells 1981 Study type description of different methods of training
Winn 1986 Study type review of rsquohistoryrsquo of rider training
Wisher 1988 Study type cross-sectional survey
Wood 1987 Study type Interrupted Time Series but did not meet the criteria for an ITS as published by the Cochrane
Effective Practice and Organisation of Care Group
Yamazaki 1990 Study type cross-sectional survey
Outcome riding skills skills retention and attitude
Characteristics of ongoing studies [ordered by study ID]
Burgess 2010
Trial name or title UKs National Ride program
Methods Cohort study
Participants Motorcycle offenders
Interventions Police led diversion program
Outcomes Self reported attitudinal and behavioural measures
Starting date
Contact information
Notes Expected to be completed in early 2010
45Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4850
D A T A A N D A N A L Y S E S
This review has no analyses
A P P E N D I C E S
Appendix 1 Search strategy
We searched the following electronic databases
bull Cochrane Injuries Group Specialised Register (searched 22 Sept 2008)
bull CENTRAL (The Cochrane Library 2008 Issue 3)
bull TRANSPORT 1988 to 200706 (latest date available)
bull ATRI Australian Transport Index (to Sept 2008)
bull MEDLINE (Ovid SP) 1950 to Sept 2008
bull EMBASE (Ovid SP) 1980 to Sept 2008
bull CINAHL (EBSCO) 1982 to Sept 2008
bull WHOLIS (World Health Organization Library Information System) (to Sept 2008)
bull PsycINFO (Ovid SP) 1806 to Sept 2008
bull LILACS (BIREME) (Latin American and Caribbean Health Sciences) (1982 to Sept 2008)
bull ISI Web of Science Social Sciences Citation Index (SSCI) 1970 to Sept 2008
bull ERIC (searched 25 Sept 2008)
bull ZETOC (searched 25 Sept 2008)
bull SIGLE (searched 25 Sept 2008)
All of the databases were searched using the following text-word search strategy The strategy was adapted using either rsquorsquo or rsquo$rsquo as atruncation symbol
1 motor-cycl or motorcycl or motor-bik or motorbik or motor-scooter or motorscooter or scooter or moped or motocycl
or motocicl or moto
2 train or educat or instruct or program or course or qualif or learn or improv
3 1 and 2
H I S T O R Y
Protocol first published Issue 2 2005
Review first published Issue 10 2010
46Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4950
C O N T R I B U T I O N S O F A U T H O R S
Katina Kardamanidis was the first author of the protocol conducted the literature search extracted full studies assessed the studies
according to inclusion criteria assessed quality of the studies compared findings with the second author discussed disagreements with
the third author wrote the drafts and the final review
Rebecca Ivers was the second author of the protocol assessed the studies according to inclusion criteria assessed quality of the studies
compared findings with the first author edited the final draft and provided overall guidance for the process
Alexandra Martiniuk assessed quality of the studies compared findings with the first author discussed disagreements with the third
author and helped write the first draft and the final review
Mark Stevenson contributed to the protocol reviewed methods of studies and contributed to the final draft
Katrina Thistlethwaite conducted a second literature search extracted full studies and contributed to the final draft
D E C L A R A T I O N S O F I N T E R E S T
None known
S O U R C E S O F S U P P O R T
Internal sources
bull No sources of support supplied
External sources
bull National Health and Medical Research Council of Australia Australia
Salary support for Martiniuk Ivers and Stevenson
bull University of Sydney Sesqui Research Grant Australia
Direct research costs and salary support for Kardamanidis
D I F F E R E N C E S B E T W E E N P R O T O C O L A N D R E V I E W
There are no differences between the protocol and the review
I N D E X T E R M S
Medical Subject Headings (MeSH)
lowastMotorcycles Accident Prevention [lowastmethods] Accidents Traffic [lowastprevention amp control] Licensure Program Evaluation Wounds
and Injuries [mortality lowastprevention amp control]
47Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 5050
MeSH check words
Humans
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4250
Satten 1980 (Continued)
Outcomes Self-reported crashes injuries and offences
Length of follow up 3 years
Notes Study conducted in Illinois USA
Savolainen 2007
Methods Cohort study
Participants Initial sample size
Intervention group 4000 riders that had conducted the MSF MRC course (passed or failed)
Control group 4000 motorcycle owners or permit or endorsement holders from the Indiana Bureau of Motor
Vehicles database
Analysis sample size
Intervention group 739 (18)
Control group 588 (15)
Interventions Intervention integrity
No information given
Outcomes Probability of crash involvement
Length of follow up up to 5 years
Notes Study conducted in Indiana USA
Swaddiwudhipong 1998
Methods Randomised trial
Participants Initial sample size
Intervention group
all 16 villages of 3 sub-districts
(participants were practising motorcyclists licensed and unlicensed)
Control group
18 villages of 3 other sub-districtsAnalysis sample size
Intervention group 3 districts (100)
Control group 3 districts (100)
Interventions No rider education vs non-mandatory community based rider education consisting of theory on motorcycle injury
risk helmet use traffic laws vehicle regulations and traffic signs and a written and skill test leading to a licence
Intervention integrity
No information given
Outcomes Injuries fatalities
Length of follow up 2 years
40Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4350
Swaddiwudhipong 1998 (Continued)
Notes Study conducted in Tak Province rural Thailand
Waller 1992
Methods Cohort study
Participants Initial sample size
Intervention group 349 riders who succesfully completed the Department of Public Safety motorcycle operator
training in 1988
Control group 349 motorcyclists who never completed the DPS motorcycle operator training
Analysis sample size
Intervention group 349
Control group 349
Interventions No DPS motorcycle operator training vs non-mandatory DPS motorcycle operator training The DPS training is
the Motorcycle Safety Foundationrsquos Motorcycle Rider Course comprising eight hours of theory (lectures discussions
films and multiple choice tests) a textbook and 12 hours off-road practice
Intervention integrity
All 349 riders succesfully completed the rider training
Outcomes Police recorded crashes
Length of follow up 3 years
Notes Study conducted in
Characteristics of excluded studies [ordered by study ID]
Study Reason for exclusion
Anderson 1980a Study type review and process evaluation of a rider course
Aupetit 2008 Outcome proposal of a rider training curriculum that combines traditional training and a motorcycle riding
simulator
Awane 1999 Study type review on use of simulators in rider and driver training
Baldi 2005 Intervention not rider training but intervention integrity of rider training availabilty of rider education
implementation of graduated licensing system etc on a state-wide level
Batchler 1988 Study type description of a training course
Blanchard 2006 Study type not a study assessing the association between rider training and outcomes
Bonnett 2005 Outcome qualitative data regarding satisfaction with the training programme
41Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4450
(Continued)
Braver 2007 Outcome licensure rates
Clayton 1980 Outcome knowledge skills attitude
Council 1978 Outcome crash reduction in ANY vehicle not motorcycles exclusively
Crowther 2005 Study type qualitative study
Daniello 2009 Study type review of the effectiveness of motorcycle training and licensing
Dewanti 2007 Intervention emotional characteristics behaviour (ie not motocycle rider training)
Du Plooy 1990 Study type review of types of rider training courses in South Africa
Engel 1989 Study type uncontrolled before-after study
Goldenbeld 2002 Outcome skills
Goldenbeld 2004 Outcome knowledge and skills
Haworth 2000 Intervention balance between attitudinal andvehicle control skills acrosscurriculla of motorcycle rider training
courses
Haworth 2005 Study type review
Haworth 2006 Study type cross-sectional survey
Hill 1983 Study type review and description of training system
Imatake 1980 Study type course descriptionreview
Ingebritsen 1990 Intervention risk factors for motorcycle crashes
Jain 2009 Study type cross-sectional descriptive study of motorcycle crashes over a 5 year period
Jonah 1979 Study type cross-sectional survey Outcome scores on motorcycle operator skill test
Jonah 1981 Intervention evaluation of a motorcycle rider skill test not motorcycle rider training
Kadar 1990 Study type review of components of rider training courses
Kelsey 1986 Intervention evaluation of a motorcycle rider skill test not motorcycle rider training
Koch 1980 Study type reviewcourse description
Koch 1990 Study type description of course content
42Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4550
(Continued)
Labbett 2005 Study type essay about possible explanations for motorcycle deaths
Law 2005 Intervention targeted motorcycle safety programme not specifically rider training
Lowes 1990 Study type review and descriptions of rider training
Mahon 1976 Study type review of motorcycle rider training methods
Maxwell 1985 Outcomesrecommendations forprogramme operatorsand otherson howthe motorcycle trainingprogramme
works and how it can be improved
Mayhew 1996 Study type review
McKnight 1987 Study type evaluation of methodological and administrative aspects of a motorcycle operator skill test
McPherson 1976 Outcome knowledge and skills
Mohan 2004 Study type review of evidence of interventions to reduce road traffic injuries
Morgan 1980 Study type reviewcourse description
Motoki 1990 Study type proposal and development of rider training curriculum
Nairn 1992 Study type literature review
Newland 1980 Study type description of development of motorcycle safety course
Newland 1982 Study type descriptionreview of a rider training course
Newland 1990 Study type uncontrolled BeforeAfter study
Ormston 2003 Outcome changes in attitude and self reported behaviour
Osga 1979 Outcome knowledge attitudes and skills
Panou 2007 Intervention new simulation tool to teach driverrider
Peterson 1990 Study factor demographic details of course participants completing the course vs those not completing the
course
Prem 1984 Study type experimental
Study factor examines association between skills and performance in tests
Radin Umar 2006 Intervention motorcycle safety programme not specifically rider training
Rockwell 1990 Study type methodologically flawed study type
43Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4650
(Continued)
Rothe 1987 Study type qualitative study of rider training in general
Rowden 2007 Study type qualitative study
RSC Victoria 1993 Study type review of motorcycle rider training and education in Victoria
Russam 1979 Study type review of training programmes in Japan
Saffron 1981 Study type literature review
Schulz 1990 Study factor predictors of motorcycle crashes
Sexton 2004 Study type cross-sectional analytic analysis
Shephard 1986 Study factor reviewdescription of programme
Simpson 1990 Study type review of motorcycle rider trainingeducation
Sokolov 1990 Study factor review of motorcycle riding courses in USSR
Steffens 1988 Outcome self-reported behaviour experiences and researcher observed riding behaviour
Sudlow 2003 Study type review of motorcycle training in the UK
Tasmania 1995 Study type cross-sectional descriptiveprocess evaluation
Tetard 1985 Study type experimental
Outcome detection of indices
Thackray 1980 Outcome instructional effectiveness user acceptance administrative feasibility
Thompson 1994 Study type cross-sectional survey
Outcome attitudes and knowledge
Tobler 1990 Study type review of course curricula
Transport 1989 Intervention syllabus for rider training
Tuan 2007 Study type cross-sectional study (survey) which does not assess the effects of an intervention but examines
motocyclistsrsquo behaviour
Victor 1980 Study type review
Waterer 1990 Study type review of UK rider training issues
Weller 1989 Study type review of studies evaluating motorcycle safety and education measures
44Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4750
(Continued)
Wells 1981 Study type description of different methods of training
Winn 1986 Study type review of rsquohistoryrsquo of rider training
Wisher 1988 Study type cross-sectional survey
Wood 1987 Study type Interrupted Time Series but did not meet the criteria for an ITS as published by the Cochrane
Effective Practice and Organisation of Care Group
Yamazaki 1990 Study type cross-sectional survey
Outcome riding skills skills retention and attitude
Characteristics of ongoing studies [ordered by study ID]
Burgess 2010
Trial name or title UKs National Ride program
Methods Cohort study
Participants Motorcycle offenders
Interventions Police led diversion program
Outcomes Self reported attitudinal and behavioural measures
Starting date
Contact information
Notes Expected to be completed in early 2010
45Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4850
D A T A A N D A N A L Y S E S
This review has no analyses
A P P E N D I C E S
Appendix 1 Search strategy
We searched the following electronic databases
bull Cochrane Injuries Group Specialised Register (searched 22 Sept 2008)
bull CENTRAL (The Cochrane Library 2008 Issue 3)
bull TRANSPORT 1988 to 200706 (latest date available)
bull ATRI Australian Transport Index (to Sept 2008)
bull MEDLINE (Ovid SP) 1950 to Sept 2008
bull EMBASE (Ovid SP) 1980 to Sept 2008
bull CINAHL (EBSCO) 1982 to Sept 2008
bull WHOLIS (World Health Organization Library Information System) (to Sept 2008)
bull PsycINFO (Ovid SP) 1806 to Sept 2008
bull LILACS (BIREME) (Latin American and Caribbean Health Sciences) (1982 to Sept 2008)
bull ISI Web of Science Social Sciences Citation Index (SSCI) 1970 to Sept 2008
bull ERIC (searched 25 Sept 2008)
bull ZETOC (searched 25 Sept 2008)
bull SIGLE (searched 25 Sept 2008)
All of the databases were searched using the following text-word search strategy The strategy was adapted using either rsquorsquo or rsquo$rsquo as atruncation symbol
1 motor-cycl or motorcycl or motor-bik or motorbik or motor-scooter or motorscooter or scooter or moped or motocycl
or motocicl or moto
2 train or educat or instruct or program or course or qualif or learn or improv
3 1 and 2
H I S T O R Y
Protocol first published Issue 2 2005
Review first published Issue 10 2010
46Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4950
C O N T R I B U T I O N S O F A U T H O R S
Katina Kardamanidis was the first author of the protocol conducted the literature search extracted full studies assessed the studies
according to inclusion criteria assessed quality of the studies compared findings with the second author discussed disagreements with
the third author wrote the drafts and the final review
Rebecca Ivers was the second author of the protocol assessed the studies according to inclusion criteria assessed quality of the studies
compared findings with the first author edited the final draft and provided overall guidance for the process
Alexandra Martiniuk assessed quality of the studies compared findings with the first author discussed disagreements with the third
author and helped write the first draft and the final review
Mark Stevenson contributed to the protocol reviewed methods of studies and contributed to the final draft
Katrina Thistlethwaite conducted a second literature search extracted full studies and contributed to the final draft
D E C L A R A T I O N S O F I N T E R E S T
None known
S O U R C E S O F S U P P O R T
Internal sources
bull No sources of support supplied
External sources
bull National Health and Medical Research Council of Australia Australia
Salary support for Martiniuk Ivers and Stevenson
bull University of Sydney Sesqui Research Grant Australia
Direct research costs and salary support for Kardamanidis
D I F F E R E N C E S B E T W E E N P R O T O C O L A N D R E V I E W
There are no differences between the protocol and the review
I N D E X T E R M S
Medical Subject Headings (MeSH)
lowastMotorcycles Accident Prevention [lowastmethods] Accidents Traffic [lowastprevention amp control] Licensure Program Evaluation Wounds
and Injuries [mortality lowastprevention amp control]
47Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 5050
MeSH check words
Humans
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4350
Swaddiwudhipong 1998 (Continued)
Notes Study conducted in Tak Province rural Thailand
Waller 1992
Methods Cohort study
Participants Initial sample size
Intervention group 349 riders who succesfully completed the Department of Public Safety motorcycle operator
training in 1988
Control group 349 motorcyclists who never completed the DPS motorcycle operator training
Analysis sample size
Intervention group 349
Control group 349
Interventions No DPS motorcycle operator training vs non-mandatory DPS motorcycle operator training The DPS training is
the Motorcycle Safety Foundationrsquos Motorcycle Rider Course comprising eight hours of theory (lectures discussions
films and multiple choice tests) a textbook and 12 hours off-road practice
Intervention integrity
All 349 riders succesfully completed the rider training
Outcomes Police recorded crashes
Length of follow up 3 years
Notes Study conducted in
Characteristics of excluded studies [ordered by study ID]
Study Reason for exclusion
Anderson 1980a Study type review and process evaluation of a rider course
Aupetit 2008 Outcome proposal of a rider training curriculum that combines traditional training and a motorcycle riding
simulator
Awane 1999 Study type review on use of simulators in rider and driver training
Baldi 2005 Intervention not rider training but intervention integrity of rider training availabilty of rider education
implementation of graduated licensing system etc on a state-wide level
Batchler 1988 Study type description of a training course
Blanchard 2006 Study type not a study assessing the association between rider training and outcomes
Bonnett 2005 Outcome qualitative data regarding satisfaction with the training programme
41Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4450
(Continued)
Braver 2007 Outcome licensure rates
Clayton 1980 Outcome knowledge skills attitude
Council 1978 Outcome crash reduction in ANY vehicle not motorcycles exclusively
Crowther 2005 Study type qualitative study
Daniello 2009 Study type review of the effectiveness of motorcycle training and licensing
Dewanti 2007 Intervention emotional characteristics behaviour (ie not motocycle rider training)
Du Plooy 1990 Study type review of types of rider training courses in South Africa
Engel 1989 Study type uncontrolled before-after study
Goldenbeld 2002 Outcome skills
Goldenbeld 2004 Outcome knowledge and skills
Haworth 2000 Intervention balance between attitudinal andvehicle control skills acrosscurriculla of motorcycle rider training
courses
Haworth 2005 Study type review
Haworth 2006 Study type cross-sectional survey
Hill 1983 Study type review and description of training system
Imatake 1980 Study type course descriptionreview
Ingebritsen 1990 Intervention risk factors for motorcycle crashes
Jain 2009 Study type cross-sectional descriptive study of motorcycle crashes over a 5 year period
Jonah 1979 Study type cross-sectional survey Outcome scores on motorcycle operator skill test
Jonah 1981 Intervention evaluation of a motorcycle rider skill test not motorcycle rider training
Kadar 1990 Study type review of components of rider training courses
Kelsey 1986 Intervention evaluation of a motorcycle rider skill test not motorcycle rider training
Koch 1980 Study type reviewcourse description
Koch 1990 Study type description of course content
42Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4550
(Continued)
Labbett 2005 Study type essay about possible explanations for motorcycle deaths
Law 2005 Intervention targeted motorcycle safety programme not specifically rider training
Lowes 1990 Study type review and descriptions of rider training
Mahon 1976 Study type review of motorcycle rider training methods
Maxwell 1985 Outcomesrecommendations forprogramme operatorsand otherson howthe motorcycle trainingprogramme
works and how it can be improved
Mayhew 1996 Study type review
McKnight 1987 Study type evaluation of methodological and administrative aspects of a motorcycle operator skill test
McPherson 1976 Outcome knowledge and skills
Mohan 2004 Study type review of evidence of interventions to reduce road traffic injuries
Morgan 1980 Study type reviewcourse description
Motoki 1990 Study type proposal and development of rider training curriculum
Nairn 1992 Study type literature review
Newland 1980 Study type description of development of motorcycle safety course
Newland 1982 Study type descriptionreview of a rider training course
Newland 1990 Study type uncontrolled BeforeAfter study
Ormston 2003 Outcome changes in attitude and self reported behaviour
Osga 1979 Outcome knowledge attitudes and skills
Panou 2007 Intervention new simulation tool to teach driverrider
Peterson 1990 Study factor demographic details of course participants completing the course vs those not completing the
course
Prem 1984 Study type experimental
Study factor examines association between skills and performance in tests
Radin Umar 2006 Intervention motorcycle safety programme not specifically rider training
Rockwell 1990 Study type methodologically flawed study type
43Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4650
(Continued)
Rothe 1987 Study type qualitative study of rider training in general
Rowden 2007 Study type qualitative study
RSC Victoria 1993 Study type review of motorcycle rider training and education in Victoria
Russam 1979 Study type review of training programmes in Japan
Saffron 1981 Study type literature review
Schulz 1990 Study factor predictors of motorcycle crashes
Sexton 2004 Study type cross-sectional analytic analysis
Shephard 1986 Study factor reviewdescription of programme
Simpson 1990 Study type review of motorcycle rider trainingeducation
Sokolov 1990 Study factor review of motorcycle riding courses in USSR
Steffens 1988 Outcome self-reported behaviour experiences and researcher observed riding behaviour
Sudlow 2003 Study type review of motorcycle training in the UK
Tasmania 1995 Study type cross-sectional descriptiveprocess evaluation
Tetard 1985 Study type experimental
Outcome detection of indices
Thackray 1980 Outcome instructional effectiveness user acceptance administrative feasibility
Thompson 1994 Study type cross-sectional survey
Outcome attitudes and knowledge
Tobler 1990 Study type review of course curricula
Transport 1989 Intervention syllabus for rider training
Tuan 2007 Study type cross-sectional study (survey) which does not assess the effects of an intervention but examines
motocyclistsrsquo behaviour
Victor 1980 Study type review
Waterer 1990 Study type review of UK rider training issues
Weller 1989 Study type review of studies evaluating motorcycle safety and education measures
44Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4750
(Continued)
Wells 1981 Study type description of different methods of training
Winn 1986 Study type review of rsquohistoryrsquo of rider training
Wisher 1988 Study type cross-sectional survey
Wood 1987 Study type Interrupted Time Series but did not meet the criteria for an ITS as published by the Cochrane
Effective Practice and Organisation of Care Group
Yamazaki 1990 Study type cross-sectional survey
Outcome riding skills skills retention and attitude
Characteristics of ongoing studies [ordered by study ID]
Burgess 2010
Trial name or title UKs National Ride program
Methods Cohort study
Participants Motorcycle offenders
Interventions Police led diversion program
Outcomes Self reported attitudinal and behavioural measures
Starting date
Contact information
Notes Expected to be completed in early 2010
45Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4850
D A T A A N D A N A L Y S E S
This review has no analyses
A P P E N D I C E S
Appendix 1 Search strategy
We searched the following electronic databases
bull Cochrane Injuries Group Specialised Register (searched 22 Sept 2008)
bull CENTRAL (The Cochrane Library 2008 Issue 3)
bull TRANSPORT 1988 to 200706 (latest date available)
bull ATRI Australian Transport Index (to Sept 2008)
bull MEDLINE (Ovid SP) 1950 to Sept 2008
bull EMBASE (Ovid SP) 1980 to Sept 2008
bull CINAHL (EBSCO) 1982 to Sept 2008
bull WHOLIS (World Health Organization Library Information System) (to Sept 2008)
bull PsycINFO (Ovid SP) 1806 to Sept 2008
bull LILACS (BIREME) (Latin American and Caribbean Health Sciences) (1982 to Sept 2008)
bull ISI Web of Science Social Sciences Citation Index (SSCI) 1970 to Sept 2008
bull ERIC (searched 25 Sept 2008)
bull ZETOC (searched 25 Sept 2008)
bull SIGLE (searched 25 Sept 2008)
All of the databases were searched using the following text-word search strategy The strategy was adapted using either rsquorsquo or rsquo$rsquo as atruncation symbol
1 motor-cycl or motorcycl or motor-bik or motorbik or motor-scooter or motorscooter or scooter or moped or motocycl
or motocicl or moto
2 train or educat or instruct or program or course or qualif or learn or improv
3 1 and 2
H I S T O R Y
Protocol first published Issue 2 2005
Review first published Issue 10 2010
46Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4950
C O N T R I B U T I O N S O F A U T H O R S
Katina Kardamanidis was the first author of the protocol conducted the literature search extracted full studies assessed the studies
according to inclusion criteria assessed quality of the studies compared findings with the second author discussed disagreements with
the third author wrote the drafts and the final review
Rebecca Ivers was the second author of the protocol assessed the studies according to inclusion criteria assessed quality of the studies
compared findings with the first author edited the final draft and provided overall guidance for the process
Alexandra Martiniuk assessed quality of the studies compared findings with the first author discussed disagreements with the third
author and helped write the first draft and the final review
Mark Stevenson contributed to the protocol reviewed methods of studies and contributed to the final draft
Katrina Thistlethwaite conducted a second literature search extracted full studies and contributed to the final draft
D E C L A R A T I O N S O F I N T E R E S T
None known
S O U R C E S O F S U P P O R T
Internal sources
bull No sources of support supplied
External sources
bull National Health and Medical Research Council of Australia Australia
Salary support for Martiniuk Ivers and Stevenson
bull University of Sydney Sesqui Research Grant Australia
Direct research costs and salary support for Kardamanidis
D I F F E R E N C E S B E T W E E N P R O T O C O L A N D R E V I E W
There are no differences between the protocol and the review
I N D E X T E R M S
Medical Subject Headings (MeSH)
lowastMotorcycles Accident Prevention [lowastmethods] Accidents Traffic [lowastprevention amp control] Licensure Program Evaluation Wounds
and Injuries [mortality lowastprevention amp control]
47Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 5050
MeSH check words
Humans
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4450
(Continued)
Braver 2007 Outcome licensure rates
Clayton 1980 Outcome knowledge skills attitude
Council 1978 Outcome crash reduction in ANY vehicle not motorcycles exclusively
Crowther 2005 Study type qualitative study
Daniello 2009 Study type review of the effectiveness of motorcycle training and licensing
Dewanti 2007 Intervention emotional characteristics behaviour (ie not motocycle rider training)
Du Plooy 1990 Study type review of types of rider training courses in South Africa
Engel 1989 Study type uncontrolled before-after study
Goldenbeld 2002 Outcome skills
Goldenbeld 2004 Outcome knowledge and skills
Haworth 2000 Intervention balance between attitudinal andvehicle control skills acrosscurriculla of motorcycle rider training
courses
Haworth 2005 Study type review
Haworth 2006 Study type cross-sectional survey
Hill 1983 Study type review and description of training system
Imatake 1980 Study type course descriptionreview
Ingebritsen 1990 Intervention risk factors for motorcycle crashes
Jain 2009 Study type cross-sectional descriptive study of motorcycle crashes over a 5 year period
Jonah 1979 Study type cross-sectional survey Outcome scores on motorcycle operator skill test
Jonah 1981 Intervention evaluation of a motorcycle rider skill test not motorcycle rider training
Kadar 1990 Study type review of components of rider training courses
Kelsey 1986 Intervention evaluation of a motorcycle rider skill test not motorcycle rider training
Koch 1980 Study type reviewcourse description
Koch 1990 Study type description of course content
42Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4550
(Continued)
Labbett 2005 Study type essay about possible explanations for motorcycle deaths
Law 2005 Intervention targeted motorcycle safety programme not specifically rider training
Lowes 1990 Study type review and descriptions of rider training
Mahon 1976 Study type review of motorcycle rider training methods
Maxwell 1985 Outcomesrecommendations forprogramme operatorsand otherson howthe motorcycle trainingprogramme
works and how it can be improved
Mayhew 1996 Study type review
McKnight 1987 Study type evaluation of methodological and administrative aspects of a motorcycle operator skill test
McPherson 1976 Outcome knowledge and skills
Mohan 2004 Study type review of evidence of interventions to reduce road traffic injuries
Morgan 1980 Study type reviewcourse description
Motoki 1990 Study type proposal and development of rider training curriculum
Nairn 1992 Study type literature review
Newland 1980 Study type description of development of motorcycle safety course
Newland 1982 Study type descriptionreview of a rider training course
Newland 1990 Study type uncontrolled BeforeAfter study
Ormston 2003 Outcome changes in attitude and self reported behaviour
Osga 1979 Outcome knowledge attitudes and skills
Panou 2007 Intervention new simulation tool to teach driverrider
Peterson 1990 Study factor demographic details of course participants completing the course vs those not completing the
course
Prem 1984 Study type experimental
Study factor examines association between skills and performance in tests
Radin Umar 2006 Intervention motorcycle safety programme not specifically rider training
Rockwell 1990 Study type methodologically flawed study type
43Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4650
(Continued)
Rothe 1987 Study type qualitative study of rider training in general
Rowden 2007 Study type qualitative study
RSC Victoria 1993 Study type review of motorcycle rider training and education in Victoria
Russam 1979 Study type review of training programmes in Japan
Saffron 1981 Study type literature review
Schulz 1990 Study factor predictors of motorcycle crashes
Sexton 2004 Study type cross-sectional analytic analysis
Shephard 1986 Study factor reviewdescription of programme
Simpson 1990 Study type review of motorcycle rider trainingeducation
Sokolov 1990 Study factor review of motorcycle riding courses in USSR
Steffens 1988 Outcome self-reported behaviour experiences and researcher observed riding behaviour
Sudlow 2003 Study type review of motorcycle training in the UK
Tasmania 1995 Study type cross-sectional descriptiveprocess evaluation
Tetard 1985 Study type experimental
Outcome detection of indices
Thackray 1980 Outcome instructional effectiveness user acceptance administrative feasibility
Thompson 1994 Study type cross-sectional survey
Outcome attitudes and knowledge
Tobler 1990 Study type review of course curricula
Transport 1989 Intervention syllabus for rider training
Tuan 2007 Study type cross-sectional study (survey) which does not assess the effects of an intervention but examines
motocyclistsrsquo behaviour
Victor 1980 Study type review
Waterer 1990 Study type review of UK rider training issues
Weller 1989 Study type review of studies evaluating motorcycle safety and education measures
44Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4750
(Continued)
Wells 1981 Study type description of different methods of training
Winn 1986 Study type review of rsquohistoryrsquo of rider training
Wisher 1988 Study type cross-sectional survey
Wood 1987 Study type Interrupted Time Series but did not meet the criteria for an ITS as published by the Cochrane
Effective Practice and Organisation of Care Group
Yamazaki 1990 Study type cross-sectional survey
Outcome riding skills skills retention and attitude
Characteristics of ongoing studies [ordered by study ID]
Burgess 2010
Trial name or title UKs National Ride program
Methods Cohort study
Participants Motorcycle offenders
Interventions Police led diversion program
Outcomes Self reported attitudinal and behavioural measures
Starting date
Contact information
Notes Expected to be completed in early 2010
45Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4850
D A T A A N D A N A L Y S E S
This review has no analyses
A P P E N D I C E S
Appendix 1 Search strategy
We searched the following electronic databases
bull Cochrane Injuries Group Specialised Register (searched 22 Sept 2008)
bull CENTRAL (The Cochrane Library 2008 Issue 3)
bull TRANSPORT 1988 to 200706 (latest date available)
bull ATRI Australian Transport Index (to Sept 2008)
bull MEDLINE (Ovid SP) 1950 to Sept 2008
bull EMBASE (Ovid SP) 1980 to Sept 2008
bull CINAHL (EBSCO) 1982 to Sept 2008
bull WHOLIS (World Health Organization Library Information System) (to Sept 2008)
bull PsycINFO (Ovid SP) 1806 to Sept 2008
bull LILACS (BIREME) (Latin American and Caribbean Health Sciences) (1982 to Sept 2008)
bull ISI Web of Science Social Sciences Citation Index (SSCI) 1970 to Sept 2008
bull ERIC (searched 25 Sept 2008)
bull ZETOC (searched 25 Sept 2008)
bull SIGLE (searched 25 Sept 2008)
All of the databases were searched using the following text-word search strategy The strategy was adapted using either rsquorsquo or rsquo$rsquo as atruncation symbol
1 motor-cycl or motorcycl or motor-bik or motorbik or motor-scooter or motorscooter or scooter or moped or motocycl
or motocicl or moto
2 train or educat or instruct or program or course or qualif or learn or improv
3 1 and 2
H I S T O R Y
Protocol first published Issue 2 2005
Review first published Issue 10 2010
46Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4950
C O N T R I B U T I O N S O F A U T H O R S
Katina Kardamanidis was the first author of the protocol conducted the literature search extracted full studies assessed the studies
according to inclusion criteria assessed quality of the studies compared findings with the second author discussed disagreements with
the third author wrote the drafts and the final review
Rebecca Ivers was the second author of the protocol assessed the studies according to inclusion criteria assessed quality of the studies
compared findings with the first author edited the final draft and provided overall guidance for the process
Alexandra Martiniuk assessed quality of the studies compared findings with the first author discussed disagreements with the third
author and helped write the first draft and the final review
Mark Stevenson contributed to the protocol reviewed methods of studies and contributed to the final draft
Katrina Thistlethwaite conducted a second literature search extracted full studies and contributed to the final draft
D E C L A R A T I O N S O F I N T E R E S T
None known
S O U R C E S O F S U P P O R T
Internal sources
bull No sources of support supplied
External sources
bull National Health and Medical Research Council of Australia Australia
Salary support for Martiniuk Ivers and Stevenson
bull University of Sydney Sesqui Research Grant Australia
Direct research costs and salary support for Kardamanidis
D I F F E R E N C E S B E T W E E N P R O T O C O L A N D R E V I E W
There are no differences between the protocol and the review
I N D E X T E R M S
Medical Subject Headings (MeSH)
lowastMotorcycles Accident Prevention [lowastmethods] Accidents Traffic [lowastprevention amp control] Licensure Program Evaluation Wounds
and Injuries [mortality lowastprevention amp control]
47Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 5050
MeSH check words
Humans
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4550
(Continued)
Labbett 2005 Study type essay about possible explanations for motorcycle deaths
Law 2005 Intervention targeted motorcycle safety programme not specifically rider training
Lowes 1990 Study type review and descriptions of rider training
Mahon 1976 Study type review of motorcycle rider training methods
Maxwell 1985 Outcomesrecommendations forprogramme operatorsand otherson howthe motorcycle trainingprogramme
works and how it can be improved
Mayhew 1996 Study type review
McKnight 1987 Study type evaluation of methodological and administrative aspects of a motorcycle operator skill test
McPherson 1976 Outcome knowledge and skills
Mohan 2004 Study type review of evidence of interventions to reduce road traffic injuries
Morgan 1980 Study type reviewcourse description
Motoki 1990 Study type proposal and development of rider training curriculum
Nairn 1992 Study type literature review
Newland 1980 Study type description of development of motorcycle safety course
Newland 1982 Study type descriptionreview of a rider training course
Newland 1990 Study type uncontrolled BeforeAfter study
Ormston 2003 Outcome changes in attitude and self reported behaviour
Osga 1979 Outcome knowledge attitudes and skills
Panou 2007 Intervention new simulation tool to teach driverrider
Peterson 1990 Study factor demographic details of course participants completing the course vs those not completing the
course
Prem 1984 Study type experimental
Study factor examines association between skills and performance in tests
Radin Umar 2006 Intervention motorcycle safety programme not specifically rider training
Rockwell 1990 Study type methodologically flawed study type
43Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4650
(Continued)
Rothe 1987 Study type qualitative study of rider training in general
Rowden 2007 Study type qualitative study
RSC Victoria 1993 Study type review of motorcycle rider training and education in Victoria
Russam 1979 Study type review of training programmes in Japan
Saffron 1981 Study type literature review
Schulz 1990 Study factor predictors of motorcycle crashes
Sexton 2004 Study type cross-sectional analytic analysis
Shephard 1986 Study factor reviewdescription of programme
Simpson 1990 Study type review of motorcycle rider trainingeducation
Sokolov 1990 Study factor review of motorcycle riding courses in USSR
Steffens 1988 Outcome self-reported behaviour experiences and researcher observed riding behaviour
Sudlow 2003 Study type review of motorcycle training in the UK
Tasmania 1995 Study type cross-sectional descriptiveprocess evaluation
Tetard 1985 Study type experimental
Outcome detection of indices
Thackray 1980 Outcome instructional effectiveness user acceptance administrative feasibility
Thompson 1994 Study type cross-sectional survey
Outcome attitudes and knowledge
Tobler 1990 Study type review of course curricula
Transport 1989 Intervention syllabus for rider training
Tuan 2007 Study type cross-sectional study (survey) which does not assess the effects of an intervention but examines
motocyclistsrsquo behaviour
Victor 1980 Study type review
Waterer 1990 Study type review of UK rider training issues
Weller 1989 Study type review of studies evaluating motorcycle safety and education measures
44Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4750
(Continued)
Wells 1981 Study type description of different methods of training
Winn 1986 Study type review of rsquohistoryrsquo of rider training
Wisher 1988 Study type cross-sectional survey
Wood 1987 Study type Interrupted Time Series but did not meet the criteria for an ITS as published by the Cochrane
Effective Practice and Organisation of Care Group
Yamazaki 1990 Study type cross-sectional survey
Outcome riding skills skills retention and attitude
Characteristics of ongoing studies [ordered by study ID]
Burgess 2010
Trial name or title UKs National Ride program
Methods Cohort study
Participants Motorcycle offenders
Interventions Police led diversion program
Outcomes Self reported attitudinal and behavioural measures
Starting date
Contact information
Notes Expected to be completed in early 2010
45Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4850
D A T A A N D A N A L Y S E S
This review has no analyses
A P P E N D I C E S
Appendix 1 Search strategy
We searched the following electronic databases
bull Cochrane Injuries Group Specialised Register (searched 22 Sept 2008)
bull CENTRAL (The Cochrane Library 2008 Issue 3)
bull TRANSPORT 1988 to 200706 (latest date available)
bull ATRI Australian Transport Index (to Sept 2008)
bull MEDLINE (Ovid SP) 1950 to Sept 2008
bull EMBASE (Ovid SP) 1980 to Sept 2008
bull CINAHL (EBSCO) 1982 to Sept 2008
bull WHOLIS (World Health Organization Library Information System) (to Sept 2008)
bull PsycINFO (Ovid SP) 1806 to Sept 2008
bull LILACS (BIREME) (Latin American and Caribbean Health Sciences) (1982 to Sept 2008)
bull ISI Web of Science Social Sciences Citation Index (SSCI) 1970 to Sept 2008
bull ERIC (searched 25 Sept 2008)
bull ZETOC (searched 25 Sept 2008)
bull SIGLE (searched 25 Sept 2008)
All of the databases were searched using the following text-word search strategy The strategy was adapted using either rsquorsquo or rsquo$rsquo as atruncation symbol
1 motor-cycl or motorcycl or motor-bik or motorbik or motor-scooter or motorscooter or scooter or moped or motocycl
or motocicl or moto
2 train or educat or instruct or program or course or qualif or learn or improv
3 1 and 2
H I S T O R Y
Protocol first published Issue 2 2005
Review first published Issue 10 2010
46Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4950
C O N T R I B U T I O N S O F A U T H O R S
Katina Kardamanidis was the first author of the protocol conducted the literature search extracted full studies assessed the studies
according to inclusion criteria assessed quality of the studies compared findings with the second author discussed disagreements with
the third author wrote the drafts and the final review
Rebecca Ivers was the second author of the protocol assessed the studies according to inclusion criteria assessed quality of the studies
compared findings with the first author edited the final draft and provided overall guidance for the process
Alexandra Martiniuk assessed quality of the studies compared findings with the first author discussed disagreements with the third
author and helped write the first draft and the final review
Mark Stevenson contributed to the protocol reviewed methods of studies and contributed to the final draft
Katrina Thistlethwaite conducted a second literature search extracted full studies and contributed to the final draft
D E C L A R A T I O N S O F I N T E R E S T
None known
S O U R C E S O F S U P P O R T
Internal sources
bull No sources of support supplied
External sources
bull National Health and Medical Research Council of Australia Australia
Salary support for Martiniuk Ivers and Stevenson
bull University of Sydney Sesqui Research Grant Australia
Direct research costs and salary support for Kardamanidis
D I F F E R E N C E S B E T W E E N P R O T O C O L A N D R E V I E W
There are no differences between the protocol and the review
I N D E X T E R M S
Medical Subject Headings (MeSH)
lowastMotorcycles Accident Prevention [lowastmethods] Accidents Traffic [lowastprevention amp control] Licensure Program Evaluation Wounds
and Injuries [mortality lowastprevention amp control]
47Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 5050
MeSH check words
Humans
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4650
(Continued)
Rothe 1987 Study type qualitative study of rider training in general
Rowden 2007 Study type qualitative study
RSC Victoria 1993 Study type review of motorcycle rider training and education in Victoria
Russam 1979 Study type review of training programmes in Japan
Saffron 1981 Study type literature review
Schulz 1990 Study factor predictors of motorcycle crashes
Sexton 2004 Study type cross-sectional analytic analysis
Shephard 1986 Study factor reviewdescription of programme
Simpson 1990 Study type review of motorcycle rider trainingeducation
Sokolov 1990 Study factor review of motorcycle riding courses in USSR
Steffens 1988 Outcome self-reported behaviour experiences and researcher observed riding behaviour
Sudlow 2003 Study type review of motorcycle training in the UK
Tasmania 1995 Study type cross-sectional descriptiveprocess evaluation
Tetard 1985 Study type experimental
Outcome detection of indices
Thackray 1980 Outcome instructional effectiveness user acceptance administrative feasibility
Thompson 1994 Study type cross-sectional survey
Outcome attitudes and knowledge
Tobler 1990 Study type review of course curricula
Transport 1989 Intervention syllabus for rider training
Tuan 2007 Study type cross-sectional study (survey) which does not assess the effects of an intervention but examines
motocyclistsrsquo behaviour
Victor 1980 Study type review
Waterer 1990 Study type review of UK rider training issues
Weller 1989 Study type review of studies evaluating motorcycle safety and education measures
44Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4750
(Continued)
Wells 1981 Study type description of different methods of training
Winn 1986 Study type review of rsquohistoryrsquo of rider training
Wisher 1988 Study type cross-sectional survey
Wood 1987 Study type Interrupted Time Series but did not meet the criteria for an ITS as published by the Cochrane
Effective Practice and Organisation of Care Group
Yamazaki 1990 Study type cross-sectional survey
Outcome riding skills skills retention and attitude
Characteristics of ongoing studies [ordered by study ID]
Burgess 2010
Trial name or title UKs National Ride program
Methods Cohort study
Participants Motorcycle offenders
Interventions Police led diversion program
Outcomes Self reported attitudinal and behavioural measures
Starting date
Contact information
Notes Expected to be completed in early 2010
45Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4850
D A T A A N D A N A L Y S E S
This review has no analyses
A P P E N D I C E S
Appendix 1 Search strategy
We searched the following electronic databases
bull Cochrane Injuries Group Specialised Register (searched 22 Sept 2008)
bull CENTRAL (The Cochrane Library 2008 Issue 3)
bull TRANSPORT 1988 to 200706 (latest date available)
bull ATRI Australian Transport Index (to Sept 2008)
bull MEDLINE (Ovid SP) 1950 to Sept 2008
bull EMBASE (Ovid SP) 1980 to Sept 2008
bull CINAHL (EBSCO) 1982 to Sept 2008
bull WHOLIS (World Health Organization Library Information System) (to Sept 2008)
bull PsycINFO (Ovid SP) 1806 to Sept 2008
bull LILACS (BIREME) (Latin American and Caribbean Health Sciences) (1982 to Sept 2008)
bull ISI Web of Science Social Sciences Citation Index (SSCI) 1970 to Sept 2008
bull ERIC (searched 25 Sept 2008)
bull ZETOC (searched 25 Sept 2008)
bull SIGLE (searched 25 Sept 2008)
All of the databases were searched using the following text-word search strategy The strategy was adapted using either rsquorsquo or rsquo$rsquo as atruncation symbol
1 motor-cycl or motorcycl or motor-bik or motorbik or motor-scooter or motorscooter or scooter or moped or motocycl
or motocicl or moto
2 train or educat or instruct or program or course or qualif or learn or improv
3 1 and 2
H I S T O R Y
Protocol first published Issue 2 2005
Review first published Issue 10 2010
46Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4950
C O N T R I B U T I O N S O F A U T H O R S
Katina Kardamanidis was the first author of the protocol conducted the literature search extracted full studies assessed the studies
according to inclusion criteria assessed quality of the studies compared findings with the second author discussed disagreements with
the third author wrote the drafts and the final review
Rebecca Ivers was the second author of the protocol assessed the studies according to inclusion criteria assessed quality of the studies
compared findings with the first author edited the final draft and provided overall guidance for the process
Alexandra Martiniuk assessed quality of the studies compared findings with the first author discussed disagreements with the third
author and helped write the first draft and the final review
Mark Stevenson contributed to the protocol reviewed methods of studies and contributed to the final draft
Katrina Thistlethwaite conducted a second literature search extracted full studies and contributed to the final draft
D E C L A R A T I O N S O F I N T E R E S T
None known
S O U R C E S O F S U P P O R T
Internal sources
bull No sources of support supplied
External sources
bull National Health and Medical Research Council of Australia Australia
Salary support for Martiniuk Ivers and Stevenson
bull University of Sydney Sesqui Research Grant Australia
Direct research costs and salary support for Kardamanidis
D I F F E R E N C E S B E T W E E N P R O T O C O L A N D R E V I E W
There are no differences between the protocol and the review
I N D E X T E R M S
Medical Subject Headings (MeSH)
lowastMotorcycles Accident Prevention [lowastmethods] Accidents Traffic [lowastprevention amp control] Licensure Program Evaluation Wounds
and Injuries [mortality lowastprevention amp control]
47Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 5050
MeSH check words
Humans
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4750
(Continued)
Wells 1981 Study type description of different methods of training
Winn 1986 Study type review of rsquohistoryrsquo of rider training
Wisher 1988 Study type cross-sectional survey
Wood 1987 Study type Interrupted Time Series but did not meet the criteria for an ITS as published by the Cochrane
Effective Practice and Organisation of Care Group
Yamazaki 1990 Study type cross-sectional survey
Outcome riding skills skills retention and attitude
Characteristics of ongoing studies [ordered by study ID]
Burgess 2010
Trial name or title UKs National Ride program
Methods Cohort study
Participants Motorcycle offenders
Interventions Police led diversion program
Outcomes Self reported attitudinal and behavioural measures
Starting date
Contact information
Notes Expected to be completed in early 2010
45Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4850
D A T A A N D A N A L Y S E S
This review has no analyses
A P P E N D I C E S
Appendix 1 Search strategy
We searched the following electronic databases
bull Cochrane Injuries Group Specialised Register (searched 22 Sept 2008)
bull CENTRAL (The Cochrane Library 2008 Issue 3)
bull TRANSPORT 1988 to 200706 (latest date available)
bull ATRI Australian Transport Index (to Sept 2008)
bull MEDLINE (Ovid SP) 1950 to Sept 2008
bull EMBASE (Ovid SP) 1980 to Sept 2008
bull CINAHL (EBSCO) 1982 to Sept 2008
bull WHOLIS (World Health Organization Library Information System) (to Sept 2008)
bull PsycINFO (Ovid SP) 1806 to Sept 2008
bull LILACS (BIREME) (Latin American and Caribbean Health Sciences) (1982 to Sept 2008)
bull ISI Web of Science Social Sciences Citation Index (SSCI) 1970 to Sept 2008
bull ERIC (searched 25 Sept 2008)
bull ZETOC (searched 25 Sept 2008)
bull SIGLE (searched 25 Sept 2008)
All of the databases were searched using the following text-word search strategy The strategy was adapted using either rsquorsquo or rsquo$rsquo as atruncation symbol
1 motor-cycl or motorcycl or motor-bik or motorbik or motor-scooter or motorscooter or scooter or moped or motocycl
or motocicl or moto
2 train or educat or instruct or program or course or qualif or learn or improv
3 1 and 2
H I S T O R Y
Protocol first published Issue 2 2005
Review first published Issue 10 2010
46Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4950
C O N T R I B U T I O N S O F A U T H O R S
Katina Kardamanidis was the first author of the protocol conducted the literature search extracted full studies assessed the studies
according to inclusion criteria assessed quality of the studies compared findings with the second author discussed disagreements with
the third author wrote the drafts and the final review
Rebecca Ivers was the second author of the protocol assessed the studies according to inclusion criteria assessed quality of the studies
compared findings with the first author edited the final draft and provided overall guidance for the process
Alexandra Martiniuk assessed quality of the studies compared findings with the first author discussed disagreements with the third
author and helped write the first draft and the final review
Mark Stevenson contributed to the protocol reviewed methods of studies and contributed to the final draft
Katrina Thistlethwaite conducted a second literature search extracted full studies and contributed to the final draft
D E C L A R A T I O N S O F I N T E R E S T
None known
S O U R C E S O F S U P P O R T
Internal sources
bull No sources of support supplied
External sources
bull National Health and Medical Research Council of Australia Australia
Salary support for Martiniuk Ivers and Stevenson
bull University of Sydney Sesqui Research Grant Australia
Direct research costs and salary support for Kardamanidis
D I F F E R E N C E S B E T W E E N P R O T O C O L A N D R E V I E W
There are no differences between the protocol and the review
I N D E X T E R M S
Medical Subject Headings (MeSH)
lowastMotorcycles Accident Prevention [lowastmethods] Accidents Traffic [lowastprevention amp control] Licensure Program Evaluation Wounds
and Injuries [mortality lowastprevention amp control]
47Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 5050
MeSH check words
Humans
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4850
D A T A A N D A N A L Y S E S
This review has no analyses
A P P E N D I C E S
Appendix 1 Search strategy
We searched the following electronic databases
bull Cochrane Injuries Group Specialised Register (searched 22 Sept 2008)
bull CENTRAL (The Cochrane Library 2008 Issue 3)
bull TRANSPORT 1988 to 200706 (latest date available)
bull ATRI Australian Transport Index (to Sept 2008)
bull MEDLINE (Ovid SP) 1950 to Sept 2008
bull EMBASE (Ovid SP) 1980 to Sept 2008
bull CINAHL (EBSCO) 1982 to Sept 2008
bull WHOLIS (World Health Organization Library Information System) (to Sept 2008)
bull PsycINFO (Ovid SP) 1806 to Sept 2008
bull LILACS (BIREME) (Latin American and Caribbean Health Sciences) (1982 to Sept 2008)
bull ISI Web of Science Social Sciences Citation Index (SSCI) 1970 to Sept 2008
bull ERIC (searched 25 Sept 2008)
bull ZETOC (searched 25 Sept 2008)
bull SIGLE (searched 25 Sept 2008)
All of the databases were searched using the following text-word search strategy The strategy was adapted using either rsquorsquo or rsquo$rsquo as atruncation symbol
1 motor-cycl or motorcycl or motor-bik or motorbik or motor-scooter or motorscooter or scooter or moped or motocycl
or motocicl or moto
2 train or educat or instruct or program or course or qualif or learn or improv
3 1 and 2
H I S T O R Y
Protocol first published Issue 2 2005
Review first published Issue 10 2010
46Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4950
C O N T R I B U T I O N S O F A U T H O R S
Katina Kardamanidis was the first author of the protocol conducted the literature search extracted full studies assessed the studies
according to inclusion criteria assessed quality of the studies compared findings with the second author discussed disagreements with
the third author wrote the drafts and the final review
Rebecca Ivers was the second author of the protocol assessed the studies according to inclusion criteria assessed quality of the studies
compared findings with the first author edited the final draft and provided overall guidance for the process
Alexandra Martiniuk assessed quality of the studies compared findings with the first author discussed disagreements with the third
author and helped write the first draft and the final review
Mark Stevenson contributed to the protocol reviewed methods of studies and contributed to the final draft
Katrina Thistlethwaite conducted a second literature search extracted full studies and contributed to the final draft
D E C L A R A T I O N S O F I N T E R E S T
None known
S O U R C E S O F S U P P O R T
Internal sources
bull No sources of support supplied
External sources
bull National Health and Medical Research Council of Australia Australia
Salary support for Martiniuk Ivers and Stevenson
bull University of Sydney Sesqui Research Grant Australia
Direct research costs and salary support for Kardamanidis
D I F F E R E N C E S B E T W E E N P R O T O C O L A N D R E V I E W
There are no differences between the protocol and the review
I N D E X T E R M S
Medical Subject Headings (MeSH)
lowastMotorcycles Accident Prevention [lowastmethods] Accidents Traffic [lowastprevention amp control] Licensure Program Evaluation Wounds
and Injuries [mortality lowastprevention amp control]
47Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 5050
MeSH check words
Humans
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 4950
C O N T R I B U T I O N S O F A U T H O R S
Katina Kardamanidis was the first author of the protocol conducted the literature search extracted full studies assessed the studies
according to inclusion criteria assessed quality of the studies compared findings with the second author discussed disagreements with
the third author wrote the drafts and the final review
Rebecca Ivers was the second author of the protocol assessed the studies according to inclusion criteria assessed quality of the studies
compared findings with the first author edited the final draft and provided overall guidance for the process
Alexandra Martiniuk assessed quality of the studies compared findings with the first author discussed disagreements with the third
author and helped write the first draft and the final review
Mark Stevenson contributed to the protocol reviewed methods of studies and contributed to the final draft
Katrina Thistlethwaite conducted a second literature search extracted full studies and contributed to the final draft
D E C L A R A T I O N S O F I N T E R E S T
None known
S O U R C E S O F S U P P O R T
Internal sources
bull No sources of support supplied
External sources
bull National Health and Medical Research Council of Australia Australia
Salary support for Martiniuk Ivers and Stevenson
bull University of Sydney Sesqui Research Grant Australia
Direct research costs and salary support for Kardamanidis
D I F F E R E N C E S B E T W E E N P R O T O C O L A N D R E V I E W
There are no differences between the protocol and the review
I N D E X T E R M S
Medical Subject Headings (MeSH)
lowastMotorcycles Accident Prevention [lowastmethods] Accidents Traffic [lowastprevention amp control] Licensure Program Evaluation Wounds
and Injuries [mortality lowastprevention amp control]
47Motorcycle rider training for the prevention of road traffic crashes (Review)
Copyright copy 2010 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
872019 Kardamanidis et al (2010) - Motorcycle rider training for the prevention of road traffic crashes
httpslidepdfcomreaderfullkardamanidis-et-al-2010-motorcycle-rider-training-for-the-prevention-of 5050
MeSH check words
Humans