Science and Justice 49 (2009) 19–23

Contents lists available at ScienceDirect

Science and Justice

j ourna l homepage: www.e lsev ie r.com/ locate /sc i jus

The prevalence of drugs and alcohol found in road traffic fatalities: A comparativestudy of victims

Simon Elliott a,d,⁎, Helen Woolacott b,d, Robin Braithwaite c,d

a (ROAR) Forensics Ltd, Malvern Hills Science Park, Geraldine Road, Malvern, Worcestershire, WR14 3SZ, UKb Key Forensic Services Ltd, Unit 1 ATU1, University of Warwick Science Park, Sir William Lyons Road, Coventry, CV4 7EZ, UKc Department of Forensic Science & Drug Monitoring, King's College London, 150 Stamford Street, London, SE1 9NH, UKd Regional Laboratory for Toxicology Sandwell and West Birmingham Hospitals NHS Trust, City Hospital, Dudley Road, Birmingham B18 7QH, UK

⁎ Corresponding author. (ROAR) Forensics Ltd, MalverRoad, Malvern, Worcestershire, WR14 3SZ, UK. Tel.: +4585283.

E-mail address: simontox@yahoo.co.uk (S. Elliott).

1355-0306/$ – see front matter © 2008 Forensic Sciencdoi:10.1016/j.scijus.2008.06.001

a b s t r a c t

a r t i c l e i n f o

Article history:

Researchers have studied t Received 19 April 2008Received in revised form 15 June 2008Accepted 21 June 2008

Keywords:DrugsRTAAlcoholToxicologyDrivers

he involvement of drugs and alcohol in fatal road traffic incidents, but withparticular emphasis on the possible impairment of the driver. This paper describes a comparative study ofdrug and alcohol findings in various victim groups (drivers of cars, vans or lorries, car passengers,motorcyclists, motorcycle passengers, cyclists and pedestrians) between 2000 and 2006. Post-mortem bloodand urine specimens submitted were analysed by immunoassay, GC–NPD, GC–FID, GC–MS and HPLC–DAD.The results of 1047 cases indicated 54% of all victims were positive for drugs and/or alcohol, with the highestpercentage of positive findings occurring in pedestrians (63%). Males between the ages of 17–24 were mostlikely to be involved in a road traffic accident, whether being in control of a vehicle (driver) or involvedindirectly (car passenger, pedestrian, motorcycle passenger). A wide range of drugs were detected (e.g. drugsof abuse, anti-convulsants, anti-histamines, anti-inflammatories, anti-psychotics, cardiac drugs and over-the-counter products), but alcohol and cannabinoids were the most frequent substances across the victim groups.When detected, alcohol was commonly above the legal driving limit in blood and urine (N63% in those incontrol and N60% not in control). Overall, the presence of drugs and/or alcohol was of similar frequency inthose victims in control (55% of driver, 48% of motorcyclists, 33% of cyclists) and not in control of a vehicle(52% of car passengers, 63% of pedestrians). This degree of frequency strongly implicates the involvement ofdrugs and alcohol in road traffic incidents and infers an effect on driving ability and individual impairment.

© 2008 Forensic Science Society. Published by Elsevier Ireland Ltd. All rights reserved.

1. Introduction

Fatal road traffic accidents, incidents and collisions continue to be acommon occurrence. For many years, the involvement of alcohol,illicit and prescribed drugs has been studied, with particular emphasison the possible impairment of the driver. Due to the continuingchange in drugs available and use trends, the involvement ofsubstances can vary with each year or decade [1,2]. For example, astudy in Australia indicated an increase in drug prevalence in driverand rider fatalities between 1990 and 1999, particularly cannabis andopioids [1]. As investigation of road traffic cases can constitute asignificant portion of a forensic toxicology laboratory's workload,toxicological data is readily available. Epidemiological data can also becollated. This has resulted in numerous types of studies. Many focuson the potential role of alcohol, prescribed medication, drugs of abuse(e.g. cannabis) or specific drug classes (e.g. benzodiazepines andamphetamines) [3–13]. Other studies comprise comparisons between

n Hills Science Park, Geraldine4 1684 585282; fax: +44 1684

e Society. Published by Elsevier Irel

individuals assessed for possible driving under the influence (DUI) ofalcohol or drugs and victims of road traffic incidents have also beenpublished [2]. The increase in testing for DUI of drugs (DUID) hasprovided a number of evaluation methods, including roadside testingand use of commercially available on-site devices [14,15]. This has alsoinvolved the analysis of alternative matrices such as oral fluid [11,13].

These and additional studies of fatalities focus on victimcharacteristics such as age and sex or separate victim groups such asdrivers (various vehicles), riders (motorcyclists, pedal cyclists) orpedestrians [16]. In the literature, although there has been acomparison of injuries between victim groups, this does not appearto have occurred for toxicological findings [17,18]. This paper describesa comparative study of drug and alcohol findings in victim groups overa 6 year period. The results of victims in control of a vehicle (driver orrider) are compared against those not in control of a vehicle(passenger or pedestrian), in order to assess any difference in thetype of drugs involved and/or the involvement of alcohol.

2. Materials and methods

The presence of drugs and alcohol was determined as part ofroutine casework at the request of HM Coroner and the Police from

and Ltd. All rights reserved.

Table 1Age and sex demographics for the group categories

aSample numbers exclude individuals with unknown demographic data. Numbers of unknown individuals shown in brackets.bMost frequent age range in each category is highlighted.

20 S. Elliott et al. / Science and Justice 49 (2009) 19–23

various areas of England andWales. Specimens submitted were bloodand/or urine. Analysis was performed using immunoassay, GC withnitrogen–phosphorus detection (GC–NPD), GC with mass-spectro-metry (GC–MS), HPLC with diode-array UV detection (HPLC–DAD)and GC with flame-ionisation detection (GC–FID). This allowed thepotential detection of over 700 drugs and drug metabolites based onin-house and commercially available libraries and previously pub-lished methods. Many different drugs types were included, forexample, anti-depressants, anti-histamines, anti-psychotics, drugs ofabuse, anti-inflammatories, calcium channel blockers, β-blockers andanti-convulsants. Alcohol (ethanol) analysis was performed usingHeadspace GC–FID. The cases were grouped into 6 categories; drivers(of cars, vans or lorries), car passengers, motorcyclists, motorcyclepassengers, cyclists and pedestrians. All victims were deceased. Theresults of 1047 cases are presented from a 6 year period between 2000and 2006. As the number of regions included increased over theperiod used, it was not appropriate to show the trend in fatalities yearon year.

3. Results

Of the 1047 cases, 603 victims were drivers, 104 were carpassengers, 193 were motorcyclists, 4 were motorcycle passengers,18 were cyclists and 125 were pedestrians. Specific data for age andsex for each category are shown in Table 1. There was a wide range ofages for all categories (6–92 years old overall), but drivers, carpassengers, pedestrians and motorcycle passengers were mostcommon in the 17–24 year old age group, however only 4 motorcyclepassengers were involved. Motorcyclists and cyclists were mostfrequent in the 35–44 age group. In all categories, victims werepredominantly male (83% male, 17% female). Of particularly note, 96%and 94% of motorcycle and pedal cycle riders, respectively, were male.

562 (54%) of all victims were positive for drugs and/or alcohol.Table 2 shows the highest percentage of positive findings occurred inpedestrians (63%); conversely, no drugs or alcohol were detected inthe few motorcycle passengers analysed. 33% of cyclists and 48% of

Table 2Overall prevalence of drugs and/or alcohol in various victim groups in road traffic fatalities

Category % Positive (fordrugs and/oralcohol)

Positivesample no.(n)

% Negative (fordrugs and/oralcohol)

Negativesample no.(n)

Drivers 55 330 45 273Car passengers 52 54 48 50Motorcyclists 48 93 52 100Motorcycle

passengers0 0 100 4

Cyclists 33 6 67 12Pedestrians 63 79 37 46

motorcyclists were positive. Drugs and/or alcohol were found in 55%of drivers, compared to 52% of car passengers.

In order to establish the usage of drugs and alcohol, the positivecases were divided into 3 categories (drugs only, drugs and alcoholand alcohol only) as shown in Fig. 1. For drivers, 32% (n=105) involveddrugs only, 26% (n=87) involved drugs and alcohol and 42% (n=138)involved only alcohol, therefore 68% of victims had consumed alcoholand 58% had drugs detected. Of the car passengers, 85% (n=46) ofpositive instances involved alcohol (with and without drugs). Alcoholonly was also the most common finding in cyclists (67%) andpedestrians (55%), whereas drugs alone was the commonest findingin motorcyclists (44%). For all victim categories, the drugs detectedwere grouped as opiates, cannabinoids (as THC-carboxylic acid),cocaine, anti-depressants, benzodiazepines, amphetamines, barbitu-rates and “others” (Fig. 2). “Other” drugs included a wide range ofprescription medications (e.g. anti-convulsants, anti-histamines, anti-inflammatories, anti-psychotics, cardiac drugs and diabetic drugs),over-the-counter products (e.g. paracetamol) and some other drugs ofabuse (e.g. benzylpiperazine and ketamine). These are shown inTable 3. Across all victims, over-the-counter drugs were found in 55instances, up to 6 instances involved the other drug types andmiscellaneous drugs of various types were found in 30 cases. Drugsknown to be administered during hospitalisation or emergencytreatment were not included (midazolam, amiodarone, thiopental,lignocaine, atracurium, metoclopramide and metronidazole). Inter-rogation of the drug data showed cannabinoids (as THC-carboxylicacid) to be the most detected drug in motorcyclists and was generallythe most detected overall for all victims except cyclists (Fig. 2). Ofparticular note, was the high frequency of benzodiazepines detectedin pedestrians, but the apparent 100% frequency of anti-depressants incyclists related to only 2 individuals where fluoxetine was detected inboth. The abused anaesthetic drug, ketamine, was detected in 2drivers. Ketamine may be used in emergency and pre-hospitalsituations, specifically by helicopter medical teams when transferring

Fig. 1. Percentage frequency of drugs, alcohol and drugs and alcohol in “positive”victims.

Fig. 2. Percentage frequency of drug types detected in the victim groups.

Table 3Summary of “other” drugs detected in victim groups (overall frequency shown for allgroups due to reduced sample numbers for some drugs/classes)

Drugs/classes (examples detected) Frequency

Over-the-counter 55ParacetamolIbuprofenSalicylatePseudoephedrine

Anti-convulsants 6ValproatePhenytoin

Anti-histamines 6LoratidineDiphenhydramine

Anti-inflammatory 3CelecoxibNaproxen

Anti-psychotics 3ProcyclidineClozapineOlanzapine

Other drugs of abuse 6GHBBZP, TFMPPKetamine

Diabetic drugs 3MetforminGliclazide

Heart drugs 6β-blockersDigoxinCarvedilol

Miscellaneous 30WarfarinQuinineChloroquineSildenafilOmeprazoleLansoprazoleFlucloxacillinRopiniroleFurosemide

21S. Elliott et al. / Science and Justice 49 (2009) 19–23

victims to hospital [19]. But in both cases the circumstances did notsupport this and it was not known to have been administered as partof medical treatment. Benzylpiperazine (BZP) was also detected inboth these cases and this aspect has been published elsewhere [20].GHB (gamma-hydroxybutyrate) was detected in 1 motorcyclist at aconcentration of 167 mg/L in ante-mortem serum, who died afteradmission to hospital.

A wide range of alcohol concentrations were found in the bloodand urine of all victims (Table 4). Drivers (blood 10–515 mg/dL, urine13–528 mg/dL), car passengers (blood 11–269 mg/dL, urine 33–459mg/dL), motorcyclists (blood 10–375mg/dL, urine 11–450mg/dL),cyclists (blood 14–379 mg/dL, urine 259 mg/dL) and pedestrians(blood 17–603 mg/dL, urine 10–335 mg/dL). No alcohol was found inthe 4 motorcycle passengers and the interpretation of cyclist data isvery difficult as only 4 out of the 18 cyclists analysed had consumedalcohol. Blood alcohol concentrations were most commonly between101 and 200 mg/dL for all victim groups and corresponded to N33% ofinstances (Table 4). For car passengers, blood concentrations between51 and 100mg/dLwere equally common (33%). In urine, the frequencyof urine concentration ranges was variable between all victims, butpedestrians tended to have higher urine concentrations (N200 mg/dL)but concentrations up to 200 mg/dL were most frequently detected inthe other groups. Overall, distribution of the alcohol concentrationranges showed that concentrations in blood and urine were highestfor drivers and pedestrians. Comparison of all victim categoriesagainst the current UK legal driving limit of 80 mg/dL in in-life wholeblood and 107 mg/dL in urine, is also shown in Table 4. In allcategories, when detected, the blood alcohol concentrationwas higherthan the legal driving limit in over 60% of victims. In particular, 67%and 77% of drivers where alcohol was detected were over the legalblood and urine driving limit, respectively. High frequencies ofpedestrians were also over the limit — 80% (blood) and 76% (urine).

4. Discussion

Demographic data indicated that males between the ages of 17–24were most likely to be involved in a road traffic accident, whetherbeing in control of a vehicle (driver) or involved indirectly (carpassenger, pedestrian, motorcycle passenger). Only 1 driver was foundto be driving illegally (under 17 years old). Interestingly, motorcyclistsand cyclists tended to be older (35–44) with over 94% of victims beingmale. Although it could be argued that young males may be moreassociated with risk taking, thereby affecting driving behaviour andpossibly speed, younger drivers will also have less driving experiencebut conversely may be able to react better than much older drivers.The age of the victim may also influence the nature and type of drugsused (if any), from drugs of abuse to prescription drugs. For example,

older victims may be more likely to be prescribed anti-diabetic drugsfor late-onset diabetes.

Of the 3 categories of victims in control of a vehicle (car/van/lorry,motorcycle or pedal cycle), 53% of all drivers and riders had drugs and/or alcohol present. Of the 3 comparative categories, not directly incontrol of a vehicle, drugs and/or alcohol were also present in 58% ofall car passengers and pedestrians but no drugs or alcohol were foundin the 4 motorcycle passengers. Specific comparison between driversand car passengers showed for drivers, the results indicated in positivecases, most (42%) involved alcohol only, with the majority (67%) overthe drink–drive limit. When drugs were detected in drivers,cannabinoids and “other” drugs weremost common. This is comparedto the car passengers, where victims were also most frequentlypositive for alcohol only (54%), again with the majority (60%) over thelegal limit in blood. Cannabinoids (as THC-carboxylic acid) was themost frequently detected drug. Therefore, the findings in thiscomparative group are very similar with passengers (52%) as likelyto be positive for drugs and/or alcohol as drivers (55%). This hasparticular implications for investigations; firstly, sometimes it is notalways clear who the driver of the vehicle was and secondly, theactions of a passenger whilst under the influence of drugs/alcoholmayhave caused driver distraction resulting in a fatal collision. In a socialsetting, car passengers may bemore likely to use drugs or alcohol thanthe “designated driver” but information regarding these situations isinvariably not known. Likewise, from the case information (includingwitness evidence) provided for pedestrians, some are suspected ofbeing under the influence having been to a social gathering or publichouse. Conversely, some pedestrian fatalities involve suspected

Table 4Summary of alcohol findings in blood and urine showing the range of concentrations and the percentage number above the current UK legal driving limit

aMost frequent age range in each category is highlighted (concentration units: mg/dL=mg/100 mL=mg %).bFrequency of instances where the concentrations were above the legal driving limit are shown as a percentage of total number where alcohol was detected.

22 S. Elliott et al. / Science and Justice 49 (2009) 19–23

dangerous driving of a vehicle. Nonetheless, of all the victim groups,pedestrians showed a distinct positive bias for drugs/alcohol withmore victims positive (usually for alcohol only) than negative (63%positive against 37% negative). 76–80% of cases had blood/urinealcohol concentrations over the corresponding drink–drive limit andtherefore may have been under the influence of alcohol at the time ofthe incident. When drugs were detected they constituted “other”drugs based on the drug groupings used (in particular, anti-psychoticsand anti-convulsants).

Unlike vehicle drivers and their passengers, motorcyclists were farmore likely to have taken drugs or alcohol compared to theirpassengers. But as previously mentioned, only 4 pillion passengerscould be analysed, with all 4 negative for drugs/alcohol. Compared todrivers, fewer motorcycle riders were positive for drugs and/or alcohol(55% of drivers, 48% of riders). For motorcyclists, findings were stillmostly (52%) negative for drugs and/or alcohol but if detected, casestypically involved drugs only (most frequently, cannabinoids). If alcoholwas present, both the blood and urine concentrations were above thelegal limit. Drugs and alcohol aside, tiredness, distraction/loss ofconcentration, weather/road conditions, driving/riding experience,time of day and mechanical state of the vehicle can all increase therisk of accident. Vehicle speed is also amajor factor in incidents [21] andis a possible reason for the similarity in the percentage of fatalitieswhere no drugs or alcohol were detected compared to positive cases(Table 2). But this applies more to cars and motorcycles than pedalcycles. Most pedal cyclists (67%) were negative for drugs and/or alcohol.If positive, alcohol only was most common with 75% of blood alcoholconcentrations above the legal limit, but the overall number of cyclistcases was low (n=18). As such no conclusion can bemade regarding thepresence of fluoxetine in the 2 cases where drugs were found.

Across all victim groups (with the exception of motorcyclepassengers), alcohol was a very common finding and continues tobe a significant factor in road traffic fatalities. This also includesvictims not in control of a vehicle. Drugs were detected with similarfrequency, with a wide range of drug types found, which supports therequirement for appropriate toxicological analysis in casework [9].However, cannabinoids was particularly common amongst the victimgroups and reflects findings published by other researchers [1,2,4,10].Unsurprisingly, this is likely to be associated with the increase incannabis use, particularly in younger victims [10] and is of particularimportance in those in control of a vehicle due to increasing publisheddata indicating a significant impairing effect when used alone andespecially with alcohol [22].

5. Conclusion

This paper presents new data for the comparison of toxicologicalfindings in road traffic fatalities between various victim groups.Demographic data indicated that males between the ages of 17–24were most likely to be involved in a road traffic accident, whetherbeing in control of a vehicle (driver) or involved indirectly (carpassenger, pedestrian, motorcycle passenger). The presence of drugsand/or alcohol was of similar frequency (~50%) in those victims incontrol and not in control of a vehicle. When detected, alcohol wasalso found at concentrations above the legal driving limit in the bloodand/or urine. Given this degree of frequency for drug/alcoholdetection in such cases, these studies particularly infer a possibleimpairment of those directly involved in an incident (driver/riders) aswell as indirect victims (pedestrians). Overall, alcohol and cannabi-noids were commonly found in all victims (with the exception of thefew motorcycle passengers analysed) and support the findings ofother researchers but have now been demonstrated in victim groupsother than those traditionally monitored (e.g. drivers).

Acknowledgments

The authors would like to thank HM Coroners for their permissionin publishing the data.

References

[1] O.H. Drummer, D. Gerostamoulos, H. Batziris, M. Chu, J.R.M. Caplehorn, M.D.Robertson, P. Swann, The incidence of drugs in drivers killed in Australian roadtraffic crashes, Forensic Sci. Int. 134 (2003) 154–162.

[2] A. Seymour, J.S. Oliver, Role of drugs and alcohol in impaired drivers and fatallyinjured drivers in the Strathclyde police region of Scotland, Forensic Sci. Int. 103(1999) 89–100.

[3] A. Holmgren, P. Holmgren, F.C. Kugelberg, A.W. Jones, J. Ahlner, Predominance ofillicit drugs and poly-drug use among drug-impaired drivers in Sweden, Traffic Inj.Prev. 8 (4) (2007) 361–367.

[4] B.M. Appenzeller, S. Schneider, M. Yegles, A. Maul, R. Wennig, Drugs and chronicalcohol abuse in drivers, Forensic Sci Int. 155 (2005) 83–90.

[5] M. Concheiro, A. de Castro, O. Quintela, M. López-Rivadulla, A. Cruz, Determinationof drugs of abuse and their metabolites in human plasma by liquid chromato-graphy–mass spectrometry. An application to 156 road fatalities, J. Chromatogr. BAnalyt. Technol. Biomed. Life Sci. 832 (1) (2006) 81–89.

[6] K. Heng, S. Hargarten, P. Layde, A. Craven, S. Zhu, Moderate alcohol intake andmotor vehicle crashes: the conflict between health advantage and at-risk use,Alcohol Alcohol. 41 (4) (2006) 451–456.

[7] A.C. Hansen, I.B. Kristensen, C. Dragsholt, J.P.B. Hansen, Alcohol and drugs (medicaland illicit) in fatal road accidents in a city of 300000 inhabitants, Forensic Sci. Int.79 (1996) 49–52.

23S. Elliott et al. / Science and Justice 49 (2009) 19–23

[8] A. Engeland, S. Skurtveit, J. Mørland, Risk of road traffic accidents associated withthe prescription of drugs: a registry-based cohort study, Ann. Epidemiol. 17 (8)(2007) 597–602.

[9 L.J. Farrell, S. Kerrigan, B.K. Logan, Recommendations for toxicological investigationof drug impaired driving, J. Forensic Sci. 52 (5) (2007) 1214–1218.

[10] P. Mura, C. Chatelain, V. Dumestre, J.M. Gaulier, M.H. Ghysel, C. Lacroix, M.F.Kergueris, M. Lhermitte, M. Moulsma, G. Pepin, F. Vincent, P. Kintz, Use of drugs ofabuse in less than 30-year-old drivers killed in a road crash in France: a spectacularincrease for cannabis, cocaine and amphetamines, Forensic Sci. Int. 160 (2006)168–172.

[11] S.W. Toennes, G.F. Kauert, S. Steinmeyer, M.R. Moeller, Driving under the influenceof drugs— evaluation of analytical data of drugs in oral fluid, serum and urine, andcorrelation with impairment symptoms, Forensic Sci. Int. 152 (2005) 149–155.

[12] F. Barbone, A.D. McMahon, P.G. Davey, A.D. Morris, I.C. Reid, D.G. McDevitt, T.M.MacDonald, Association of road-traffic accidents with benzodiazepine use, Lancet352 (9137) (1998) 1331–1336.

[13] C. Engblom, T. Gunnar, A. Rantanen, P. Lillsunde, Driving under the influence ofdrugs — amphetamine concentrations in oral fluid and whole blood samples,J. Anal. Toxicol. 31 (5) (2007) 276–280.

[14] D.J. Crouch, R.K. Hersch, R.F. Cook, J.F. Frank, J.M. Walsh, A field evaluation of fiveon-site drug-testing devices, J. Anal. Toxicol. 26 (7) (2002) 493–499.

[15] E. Raes, A.G. Verstraete, Usefulness of roadside urine drug screening in driverssuspected of driving under the influence of drugs (DUID), J. Anal. Toxicol. 29 (7)(2005) 590–598.

[16] W.Y. Lee, P.A. Cameron, M.J. Bailey, Road traffic injuries in the elderly, Emerg. Med.J. 23 (4) (2006) 327.

[17] K. Toro, M. Hubay, P. Sotonyi, E. Keller, Fatal traffic injuries among pedestrians,bicyclists and motor vehicle occupants, Forensic Sci. Int. 151 (2005) 151–156.

[18] H. Markogiannakis, E. Sanidas, E. Messaris, D. Koutentakis, K. Alpantaki, A.Kafetzakis, D. Tsiftsis, Motor vehicle trauma: analysis of injury profiles by road-user category, Emerg. Med. J. 23 (1) (2006) 27–31.

[19] J.E. Svenson, M.K. Abernathy, Ketamine for prehospital use: new look at an olddrug, Am. J. Emerg. Med. 25 (8) (2007) 977–980.

[20] S. Elliott, C. Smith, Investigation of the first deaths in the UK involving thedetection and quantitation of the piperazines BZP and 3-TFMPP, J. Anal. Toxicol. 32(2) (2008) 172–177.

[21] A. Vorko-Jović, J. Kern, Z. Biloglav, Risk factors in urban road traffic accidents, J. Saf.Res 37 (2006) 93–98.

[22] C.J. O'Kane, D.C. Tutt, L.A. Bauer, Cannabis and driving: a new perspective, Emerg.Med. Aust. 14 (3) (2002) 296–303.