A Cluster Randomized Clinical Trial Comparing FunctionalCapacity Evaluation and Functional Interviewing as Componentsof Occupational Rehabilitation Programs

Douglas P. Gross • Alexander K. Asante • Maxi Miciak • Michele C. Battie •

Linda J. Carroll • Ambrose Sun • Marti Mikalsky • Rene Huellstrung •

Riikka Niemelainen

� Springer Science+Business Media New York 2013

Abstract Purpose Functional capacity evaluations (FCE)

are used to identify work abilities and are commonly inte-

grated into rehabilitation programs. We studied whether

integrating FCE into rehabilitation leads to better outcomes

for injured workers. Methods A cluster randomised con-

trolled trial was conducted at a workers’ compensation

rehabilitation facility (registration ISRCTN61284905).

Clinicians were randomised into 2 groups: 1 group used

FCE while another conducted semi-structured functional

interviews. Outcomes included recommendations following

assessment, rehabilitation program outcomes including

functional work levels and pain intensity, as well as com-

pensation outcomes at 1, 3, and 6 months after assessment.

Analysis included Mann–Whitney U, Chi square and t tests.

Results Subjects included 225 claimants of whom 105 were

tested with FCE. Subjects were predominantly employed

(84 %) males (63 %) with sub-acute musculoskeletal con-

ditions (median duration 67 days). Claimants undergoing

FCE had *15 % higher average functional work levels

recommended at time of assessment (Mann–Whitney

U = 4,391.0, p \ 0.001) but differences at other follow-up

times were smaller (0–8 %), in favour of functional inter-

viewing, and not statistically significant. Clinically impor-

tant improvement during rehabilitation in functional

work level (0.9/4, SRM = 0.94), pain intensity (2.0/10,

SRM = 0.88) and self-reported disability (21.8/100,

SRM = 1.45) were only observed in those undergoing the

functional interview. Conclusions Performance-based FCE

integrated into occupational rehabilitation appears to lead to

higher baseline functional work levels compared to a semi-

structured functional interview, but not improved RTW

rates or functional work levels at follow-up. Functional

interviewing has potential for efficiency gains and higher

likelihood of clinically important improvement following

rehabilitation, however further research is needed.

Keywords Return-to-work � Compensation �Occupational injuries � Vocational rehabilitation �Disability insurance

Introduction

Given the substantial human, economic and societal burden

of musculoskeletal disorders, improved management

strategies are needed [1, 2]. Especially important are

strategies aimed at facilitating sustainable return-to-work

(RTW) for non-specific conditions (e.g. back, neck or

shoulder pain) where diagnostic imaging cannot be relied

Electronic supplementary material The online version of thisarticle (doi:10.1007/s10926-013-9491-4) contains supplementarymaterial, which is available to authorized users.

D. P. Gross (&) � M. C. Battie

Department of Physical Therapy, University of Alberta,

2-50 Corbett Hall, Edmonton, AB T6G 2G4, Canada

e-mail: dgross@ualberta.ca

D. P. Gross � A. Sun � M. Mikalsky � R. Huellstrung �R. Niemelainen

Workers’ Compensation Board-Alberta Millard Health,

Edmonton, AB, Canada

A. K. Asante

Alberta Health Services, Calgary Zone, Calgary, AB, Canada

M. Miciak

Faculty of Rehabilitation Medicine, University of Alberta,

Edmonton, AB, Canada

L. J. Carroll

School of Public Health, University of Alberta, Edmonton,

AB, Canada

123

J Occup Rehabil

DOI 10.1007/s10926-013-9491-4

on to determine resolution and readiness to return to

activity [3]. This has led to the development of perfor-

mance-based assessment tools such as functional capacity

evaluations (FCE) aimed at determining patients’ work

ability [4].

FCE has become an important component of many

functional restoration and other rehabilitation programs

[5–7]. It is often used to determine whether rehabilitation is

needed and then as an outcome measure before and after

rehabilitation. When compared to self-report functional

measures (i.e. tools relying on patients to indicate their

abilities using questionnaire, interview, clinical history,

etc.) performance-based FCE administered by a trained

therapist (i.e. actually having patients lift, carry, walk, etc.

in the clinic) has traditionally been considered more useful,

trustworthy and ‘‘objective’’ or free of bias [7–10]. How-

ever, such testing is time-consuming, requires specialized

equipment and is typically as expensive as advanced

diagnostic imaging procedures. Testing is also frequently

associated with pain reports [11, 12] and does not appear

entirely ‘‘objective’’ as previously supposed. Performance

appears to be determined not only by physiological

capacity, but by a variety of other psychological and social

factors including disability perceptions, self-efficacy

beliefs, pain intensity, and testing context [13–16]. More

critically, it has been demonstrated that FCE results are not

predictive of outcome following multidisciplinary reha-

bilitation after controlling for the number of pre-admission

health care visits [17] and is only modestly predictive of

future RTW [18–21]. Self-report measures (i.e. interview

or paper-based questionnaires) are typically less burden-

some for both patients and clinicians [22], but have been

criticized for relying totally on the patient’s ‘‘subjective’’

perspective and are less frequently relied on [23].

Whether FCE is superior to self-report measures and

actually worth the added burden has not been formally

evaluated via controlled clinical trial [24]. Several cross-

sectional studies have been performed looking at the corre-

lation between self-report and performance-based functional

measures in individuals with back pain [13, 15, 25–27].

Consistently, moderate correlations have been observed

indicating these different methods are measuring related

constructs. However, results from self-report measures have

been found to be systematically lower than performance

testing, with self-estimates of lifting ability during FCE

lower by 3–8 kg [15, 25]. The clinical implication of this

lower estimation is unknown, but potentially could lead to

limited or reduced RTW recommendations (i.e. return to

light duty as opposed to medium or heavy work). There is

also the possibility that lower functional levels limit reinte-

gration possibilities available to patients who do not have

jobs to return to.

Our objective was to examine whether performance-

based FCE enhances the occupational rehabilitation of

workers’ compensation claimants beyond information

gained from self-report measures. We hypothesized that:

(1) claimants undergoing performance-based FCE as part

of RTW assessment would demonstrate higher functional

work levels at time of baseline evaluation as compared to

claimants undergoing functional interviewing due to the

systematic lower estimations obtained with self-report

measures observed in previous cross-sectional studies; (2)

claimants undergoing either FCE or a self-report functional

interview would have comparable rates of overall RTW;

but (3) claimants undergoing FCE would RTW at higher

work levels due to the higher functional levels observed

with performance testing.

Methods

Design

A cluster randomised controlled trial design was used with

analysis at the level of individual claimant and an allocation

ratio of 1:1 [28]. Our clinical trial registration number was

ISRCTN61284905 and no important changes were made to

the methods after trial commencement. The design and

methodology were similar to that used successfully in eval-

uating a short-form Functional Capacity Evaluation (FCE)

protocol within the same jurisdiction [29]. The research was

conducted at the Alberta Workers’ Compensation Board’s

(WCB-Alberta) rehabilitation facility in Edmonton, Canada.

Data were extracted from WCB-Alberta administrative and

clinical databases, which have previously been used for

research purposes [5]. The University of Alberta’s Health

Research Ethics Board approved this study.

Study Sample

Between November 28, 2011 and January 10, 2012, all

claimants undergoing work assessment at the WCB-

Alberta facility were eligible for entry into the study. Data

were routinely collected as part of program evaluation

procedures, with the exception of follow-up telephone

interviews, which were voluntary. Exclusion criteria

included participation in a shoulder rehabilitation pilot

program that was being undertaken concurrently with this

study, claimants diagnosed with serious hearing impair-

ment (sufficient to make an interview impossible or unre-

alistic), explicit instruction from the claimant’s case

manager to conduct FCE, or non-musculoskeletal disorders

such as traumatic head injury, tumour or infection. Other-

wise, the sample was made of all claimants assessed at the

J Occup Rehabil

123

facility during the study timeframe. As little within-cluster

dependence was anticipated (claimants are assigned to

therapists in a non-systematic, relatively random fashion

based on the referral question and next available assess-

ment opening in the computerized scheduling system),

sample size estimates were adjusted using a small inflation

factor (intracluster correlation = 0.05) [30, 31]. We aimed

for 100 claimants per group and enrollment was stopped

when we achieved these numbers to avoid disruption of

service delivery at the rehabilitation facility.

Clinical Trial and Randomisation Procedures

Within the WCB-Alberta system, claimants are systemat-

ically referred for RTW assessment when they have not yet

returned to work, have ongoing functional difficulty related

to their compensable injury, and are being considered for

multidisciplinary occupational rehabilitation [17, 32]. The

WCB-Alberta rehabilitation facility is contracted to con-

duct these assessments and currently all RTW assessments

contain a component of performance-based functional

assessment using FCE. The objective of these assessments

is to determine whether the claimant is capable of returning

to pre-accident work duties and, if not, to determine if

further rehabilitation is indicated.

To evaluate whether performance-based FCE enhances

the process, we created a comparison group of claimants

who underwent assessment without FCE. Due to feasibility

issues related to the current WCB-Alberta rehabilitation

facility’s admission process, instead of randomly allocating

individual claimants to study groups we randomised cli-

nicians administering work assessments. Claimants were

therefore considered to have entered clusters of individuals

assessed by the same therapist. Clinicians at the facility

who were trained in administering the WorkWell FCE

were randomly allocated to one of two groups. One group

was asked to continue regular procedures that included

FCE, while the second group was asked to stop using FCE

and were trained to conduct ‘self-report’ assessments using

semi-structured functional interviewing.

Randomization was done by two of the study research-

ers (DPG and MM) who were blind to the allocation

sequence using a computerized random number generator

(www.random.org). Clinicians were each assigned a ran-

dom number, with even numbers allocated to the FCE

group. Randomisation resulted in 15 clinicians in the FCE

group and 15 in the ‘self-report’ group. After training, one

clinician in the ‘self-report’ group dropped out for personal

reasons unrelated to the study. Average age and years

working at the workers’ compensation facility of clinicians

in the FCE group was 39.8 and 9.5 years respectively,

comparable to the ‘self-report’ group (mean age = 38.0

years, mean experience = 9.8 years, range 2–23 years).

All of these clinicians had previously been trained in the

WorkWell FCE protocol and were using it within their

current daily practice at the facility. Study researchers were

available to answer questions throughout the study and

held weekly status update meetings with clinicians, but

clinicians independently made all claimant level manage-

ment decisions.

Due to the nature of work assessment, neither clinicians

nor claimants were blind to group allocation. However,

claimants were not aware of the study and were therefore

blind to group membership, including at time of follow-up

telephone interviews. Additionally, outcome evaluation

was performed in a blinded fashion via obtaining infor-

mation on claims outcomes from WCB administrative

databases and keeping telephone interviewers blind to

group status (conducted by a independent polling firm who

had no other information about the subjects).

Measures

Intervention Variable

The primary comparison made was between performance-

based and ‘self-report’ functional assessment groups. A

dichotomous intervention variable was created indicating

group allocation by identifying the assessing clinician

using the facility database.

Apart from the form of functional assessment, all other

aspects of the RTW assessment process were comparable

across all claimants. Assessing clinicians reviewed claim-

ants’ files, took a detailed history, and performed a mus-

culoskeletal examination. Claimants were also asked to

complete the SF-36 [33], Pain Disability Index [34], a

10-point Pain Visual Analogue Scale (VAS) [35], and a

physical job demands questionnaire for comparison with

determined functional levels. Functional ability levels were

compared to physical demands of the claimants’ jobs for

determination of work ability and readiness to RTW.

Physical demands were determined in both study groups

using either the physical job demands questionnaire,

employer-reported job demands, or workplace evaluation

when this information was available prior to the functional

assessment.

Performance-Based FCE

Functional ability is currently assessed at the facility via

the proprietary WorkWell (Duluth, MN) FCE. Claimants

within this study had been referred for a basic WorkWell

1-day FCE protocol, which is done when the claimant is

being considered as a candidate for rehabilitation. The

comprehensive 2-day WorkWell protocol, conducted after

other rehabilitation options have been exhausted and claim

J Occup Rehabil

123

closure is being considered, was examined in a related but

separate study. The WorkWell FCE involves a series of

performance tests including manual handling, positional

testing, mobility and coordination tests [36, 37]. The pro-

tocol typically takes 2–4 h to complete. To make maxi-

mum effort and performance judgments, clinicians observe

subjects’ physiological and biomechanical response to

testing to determine when maximum physical ability has

been reached [23, 38]. Performance on the FCE items is

synthesized into an overall functional work level in the

categories of sedentary (up to 5 kg), light (5–10 kg),

medium (10–20 kg), or heavy/very heavy ([20 kg). These

categories are important measures in many workers’

compensation systems as they are frequently used in RTW

planning, vocational rehabilitation (i.e. supported job

searching or retraining), and for setting earnings loss sup-

plements in cases of permanent disability. Within the

WCB-Alberta jurisdiction these levels facilitate commu-

nication across stakeholders and are used to describe

overall functional work ability following RTW assessment,

results of worksite assessments and employer ratings of

physical job demands, and are used to guide many claim-

ant-level RTW and indemnity-related decisions. To make

specific RTW decisions, clinicians compare claimant

ability to required job demands. Claimants meeting or

exceeding job demands are deemed suitable for RTW,

while those not meeting job demands are typically referred

for multidisciplinary occupational rehabilitation or a single

service provider (i.e. physiotherapy or chiropractic). The

WorkWell FCE has been shown to have acceptable reli-

ability [37, 39], construct validity as a measure of work-

related ability [40], and modestly predicts future RTW [18,

19, 21]. Claimants in the FCE group who were selected to

undergo multidisciplinary rehabilitation following assess-

ment had exercise and activity prescriptions made based on

FCE results and also underwent FCE at time of discharge

from the program. For analysis, we identified whether

claimants had undergone FCE as well as the clinician-

recommended functional work level category.

Self-report Functional Assessment

Numerous self-report measures are available for informing

RTW decisions, including questionnaires and interviews.

[41, 42] For this study, we used a semi-structured func-

tional interview approach that was previously developed

based on items in the WorkWell FCE [15]. In this func-

tional interview respondents are questioned about their

ability for the various activities that are tested during the

FCE protocol. The functional interview working document

that was used by clinicians during the study shows the

various items in the assessment and is available in

Appendix 1 in supplementary material. Claimants are

asked to identify to what level they would be able to

maximally perform for each activity (i.e. lifting or carrying

in kilograms, bending or standing in minutes) and this was

synthesized into an overall functional work level using the

same categories of sedentary/light/medium/heavy. To

make RTW decisions based on information gained, clini-

cians compare reported function to physical demands of the

job the worker would be returning to. If clarification was

needed on the various activities in question, pictures of the

activities were shown. Additionally, clinicians have access

to a detailed list of objects and their respective weights that

could be referred to when completing the assessment to

assist claimants in making estimations.

Previous research indicated that the manual handling

component of this functional interview correlates moder-

ately well (r = 0.50–0.73) with actual performance on the

corresponding WorkWell FCE items, supportive of its

construct validity as a measure of work-related function

[15]. However, consistent with other self-report measures,

results were systematically lower than FCE performance

testing with self-estimates of lifting ability lower by 3–8 kg

[15, 25]. Reliability of the tool has not previously been

examined, however prior to study implementation the full

semi-structured functional interview process was pilot

tested at a separate rehabilitation facility. Additionally,

during the course of the study the researchers held weekly

status update meetings with clinicians in the functional

interview group. This allowed us to monitor compliance

with the interview protocol, answer questions that arose,

and resolve any concerns expressed regarding the protocol.

There was also close communication between the study

clinicians and their facility ‘team leads’ that were also

members of the investigative team.

Prior to beginning the trial, clinicians assigned to the

‘self-report’ group were trained in how to conduct this

semi-structured functional interview. Training consisted of

a full-day session of instruction, including demonstration

and practice on use of functional interviewing. To assist

with conducting the interviews, clinicians were also

instructed on some of the basic principles of motivational

interviewing which aim to resolve ambivalence [43]. Cli-

nicians were instructed that the interviews should be col-

laborative, evocative (i.e. eliciting claimants’ own answers,

values and motivation), yet demonstrate respect for

claimant autonomy. For analysis, we identified whether

claimants had been assessed with the functional interview

or FCE and we obtained data on the clinician-recom-

mended functional work level category.

To avoid contamination, claimants in the self-report

functional interview group selected to undergo rehabilita-

tion were not exposed to FCE testing at any time during

their rehabilitation. Treatment clinicians were also trained

to perform semi-structured functional interviews and used

J Occup Rehabil

123

this assessment technique during rehabilitation and at dis-

charge assessment. Thus, exercise and activity prescrip-

tions along with RTW recommendations made during

rehabilitation were informed by functional interview

results.

Potential Confounders

To control for the possibility of unequal group formation,

information was obtained on a number of potential con-

founders. This included age, gender, number of days

between injury and assessment, job attachment status, pre-

accident annual salary, education level, marital status,

diagnosis, part of body injured, Human Resources and

Skills Development Canada’s National Occupational

Classification (NOC) [44], primary language and whether

an interpreter was involved in the assessment, and claimant

scores on the self-report pain, disability and health status

instruments completed during the assessment. These vari-

ables were chosen based on our previous studies within this

jurisdiction showing some predictive value or due to the-

oretical rationale for considering the variables as potential

confounders. We also determined the percentage of

claimants who underwent multidisciplinary rehabilitation

following assessment.

Outcome Measures

RTW outcomes included proxy indicators of work status

from the compensation database. We extracted information

on RTW and rehabilitation recommendations made by

clinicians at time of assessment. Using the compensation

administrative database, we also determined whether

claimants were receiving any wage replacement benefits.

Potential benefits included partial or total temporary dis-

ability benefits (received when claimants are off work for

part of or a complete day of work) or vocational rehabili-

tation benefits (received when claimants are undergoing

supported job search or retraining). Each of these was

extracted at time of assessment and at approximately the

same intervals as the follow-up interviews (30, 90, and

180 days post-assessment). These measures are commonly

used as indicators of RTW within compensation contexts,

and were available for 100 % of our sample [45]. After the

trial commenced, the decision was made to extract com-

pensation outcomes on any type of wage replacement

benefits (not just total temporary disability benefits) at the

same intervals as the telephone interviews (instead of

measuring time to benefit suspension), largely due to the

relatively small percentage of claimants receiving time loss

benefits at time of assessment.

Another important goal was to examine differences on

level of sustained work following assessment (i.e. heavy,

medium, light, sedentary or not working), with one group

judged superior if they returned to significantly higher lev-

els. As mentioned above, work levels are routinely used in

standardized occupational descriptions and are one outcome

of the work assessment at the rehabilitation facility [44].

Work levels described in the NOC are rated as sedentary (up

to 5 kg), light (5–10 kg), medium (10–20 kg), or heavy/

very heavy ([20 kg) [44]. To determine work levels after

the assessment, telephone contacts were made by an expe-

rienced polling firm (Leger Marketing). As work levels

frequently transition over time (due to job changes, pro-

gression of modified duties, etc.) follow-up contacts

occurred at multiple points in the subsequent year. Claim-

ants were contacted at 1, 3, and 6 months post-assessment.

Questions asked included whether the claimant was work-

ing, and if not working for what reason. In those who were

working, we asked claimants to rate their work level as

sedentary, light, medium, or heavy/very heavy [44]. We also

inquired about whether they were working full or part time

hours and whether they were working full or modified

duties.

For claimants who underwent multidisciplinary reha-

bilitation programs at the facility, we examined before and

after measures of functional work level category. We also

evaluated before and after measures of self-reported pain,

disability and health status and program completion,

duration and RTW outcomes. The decision was made to

evaluate these outcomes after the trial commenced, since

information was not available on claimant satisfaction rates

(as was originally planned) yet these are important clinical

outcomes following rehabilitation. All of these measures

were available within the facility’s clinical database.

Potential Harms

To monitor adverse or unintended effects, we tracked

serious negative outcomes experienced during or following

the assessments (i.e. new injuries during the assessment)

and complaints received about assessments from compen-

sation case managers.

Analysis

Initially, all data records were reviewed in order to ascer-

tain if any data issues, such as missing data, outliers or out

of range values, existed within the data set. Descriptive

statistics were then calculated. Related to cluster random-

ised trials, Donner and Klar have stated that an analysis at

the individual level that properly accounts for the effect of

clustering is equivalent to an appropriately weighted clus-

ter-level analysis [46]. To determine if a clustering effect

occurred, we examined differences across clusters on

claimant characteristics. We also examined intracluster

J Occup Rehabil

123

correlation using kappa and intraclass correlation coeffi-

cients for all of the primary and secondary outcomes. To

test our hypotheses and determine whether differences

existed between groups on the dichotomous proxy indictors

of RTW status (i.e. compensation benefit reception) at the

various follow-up intervals, Chi squared tests were per-

formed. To test our hypotheses and determine whether

differences existed between groups on the ordinal-level

functional work level categories at baseline and at the

various follow-up intervals, Mann–Whitney U tests were

performed. To determine absolute size of the effects,

functional level scores were also considered on a scale of

1–4 and mean differences were calculated. Groups were

statistically on all other outcomes using either non-para-

metric Mann–Whitney U, Chi square or t tests. Standard-

ized response means were also calculated on the pre-post

rehabilitation measures. Stratification and multivariable

analyses were used to evaluate whether variables that were

different between groups at baseline had an important

confounding effect. All analyses were conducted within the

originally assigned groups. A 0.05 alpha level was used to

judge statistical significance. All analyses were conducted

using SPSS 20 (Chicago, IL).

Results

Claimant Characteristics and Adequacy

of Randomisation Process

During the study timeframe, 225 workers’ compensation

claimants were assessed: 105 underwent FCE and 120 were

assessed using the functional interview. The higher number

of claimants assessed in the interview group was due to

increased efficiency of the assessment process (claimants

no longer needed to be cleared medically for maximal

performance testing). Twenty claimants were excluded

from the study (17 were participating in a pilot shoulder

rehabilitation program, and 3 had been referred explicitly

for performance-testing by their WCB case manager either

for comparison to previous FCE or for assessment with and

without a gait aide). A detailed study flow chart showing

the number of therapists enrolled and claimants at each step

of the study is available as supplementary material.

Claimant characteristics are detailed in Table 1. The

majority were employed (84 %) males (63 %) working in

the ‘trades and transport’ industry (49 %). Claimants in the

FCE and functional interview groups were similar on most

claimant characteristics examined (see Table 1), but a

statistically significant difference was observed on sex and

marital status. Claimants in the FCE group were less likely

to be male (55 vs. 71 %) and more likely to be married (56

vs. 43 %). No statistically significant differences were

Table 1 Characteristics of claimants referred for return-to-work

assessment

Entire

sample

(n = 225)

Interview

group

(n = 120)

FCE group

(n = 105)

Mean (SD) or percent

Age (years) 43.2 (13.1) 43.9 (13.0) 42.5 (13.3)

Accident to

admission (days)

167 (639) 180 (832) 151 (295)

Median = 67 Median = 62 Median = 72

Gross annual salary

($10 k CDN)

42.8 (29.3) 43.5 (29.6) 42.0 (29.1)

Sex (% male)* 63.1 70.0 55.2

Employed at time of

assessment (% yes)

83.6 81.1 86.7

Currently working 50.7 47.5 54.3

Language

(% English)

96.9 97.1 97.1

Interpreter involved

(% yes)

4.0 3.3 4.8

Education level

Grade 8 or less 4.4 4.2 4.8

Partial high school 11.6 14.2 8.6

High school

diploma

23.8 25.8 21.0

Partial technical

school

8.4 9.2 7.6

Technical diploma 12.4 12.5 12.4

Partial university 4.9 2.5 7.6

University degree 6.7 5.8 7.6

Not specified 28.0 25.8 30.5

Marital status*

Married/common

law

49.3 43.3 56.2

Single 24.9 33.3 15.2

Divorced/separated 8.9 9.2 8.6

Widowed 1.3 0.8 1.9

Not specified 15.6 13.3 18.1

Diagnosis

Sprain/strain/

non-specific

69.8 66.7 73.3

Fracture 11.1 12.5 9.5

Trauma 17.3 20.0 14.3

Dislocation 0.9 0.8 1.0

Carpal tunnel

syndrome

0.9 0 1.9

Part of body

Upper extremity/

shoulder

34.7 30.0 40.0

Back/trunk 31.6 29.2 34.3

Lower extremity 20.4 23.3 17.1

Neck 2.2 3.3 1.0

Other/multiple

body parts

11.1 14.2 7.6

J Occup Rehabil

123

observed between groups on the self-report pain, disability

or health-related quality of life questionnaires completed

during the work assessment. Scores indicated that claim-

ants experienced moderate levels of pain and disability,

and reported substantially reduced physical and mental

health compared with normative scores [47].

While no significant differences existed in claimant

characteristics between groups, there was a statistically

significant difference between groups in the discipline of

the assessing clinician. Clinicians conducting FCE were

more likely to be occupational therapists (72 vs. 21 %)

while those conducting functional interviews were more

likely to be physical therapists (24 vs. 14 %) or exercise

therapists/kinesiologists (55 vs. 14 %). However, there

were no statistically significant differences observed in any

study outcomes across therapist discipline indicating that

this difference likely did not impact results. When testing

for a clustering effect, the only claimant characteristic that

was statistically different across therapist clusters out of 23

individual and clinical assessment variables tested was

marital status (p = 0.02), indicating that a clustering effect

was unlikely. As expected, intracluster correlation coeffi-

cients for the various outcome measures were small,

ranging from 0.001 to 0.03.

Follow-Up Rates

There was complete ascertainment of outcomes assessed

through the facility database that included RTW recom-

mendations and functional levels at time of assessment and

discharge from rehabilitation, and all compensation wage

replacement benefit data. Some missing data was observed

on the self-report pain and disability questionnaires at time

from discharge from rehabilitation with 76 of 104 (73 %)

having complete data on these measures. No statistically

significant or clinically important differences were observed

on any characteristic or outcome measure between those

with or without missing data on these questionnaires.

Functional work level after assessment was ascertained

through telephone interviews. Overall rate of participation

in follow-up telephone interviews was 38 %, with follow-up

rates at 1, 3, and 6 months of 34, 24, and 22 % respectively.

Statistical testing indicated that compared to those who did

not participate, those interviewed were more likely to have

undergone a functional interview (45.0 vs. 29.5 %,

p = 0.02) and less likely to have back injuries (17 vs.

41 %), but were otherwise comparable on other descriptive

variables.

Work Assessment and Rehabilitation Outcomes

Approximately 48 % of claimants were recommended to

RTW following assessment, but 71 % were also recom-

mended for some form of rehabilitation (53 % underwent

multidisciplinary rehabilitation). Only 12 % had no other

interventions recommended, with all of these recommen-

dations similar between groups.

A statistically significant difference was observed

between groups on functional work level determined at

time of assessment (see Table 2). Claimants undergoing

FCE were less likely to be rated at a sedentary work level

(12 % in FCE group vs. 31 % for interview, Mann–

Table 1 continued

Entire

sample

(n = 225)

Interview

group

(n = 120)

FCE group

(n = 105)

National

occupational

classification

Management 1.8 0.8 2.9

Business, finance

and admin

10.7 16.7 3.8

Natural and applied

sciences

2.7 3.3 1.9

Health occupations 8.4 5.8 11.4

Education,

government, etc.

0.4 0 1.0

Art, culture,

recreation and

sport

0.4 0 1.0

Sales and service 19.1 19.2 19.0

Trades and

transport

48.9 46.7 51.4

Natural resources

and agriculture

3.1 3.3 2.9

Manufacturing and

utilities

4.4 4.2 4.8

Pain Visual

Analogue Scale

(out of 10,

n = 223)

4.6 (2.4) 4.6 (2.5) 4.5 (2.3)

Percent Pain

Disability Index

(n = 223)

45.7 (23.2) 47.0 (23.3) 44.2 (23.2)

SF36 raw scores

(n = 212)

Physical function 55.4 (23.6) 53.6 (25.2) 57.5 (21.6)

Role physical 31.6 (27.5) 30.0 (27.0) 33.3 (28.1)

Bodily pain 34.6 (19.0) 33.0 (19.9) 36.3 (18.6)

General health 67.5 (19.0) 67.6 (17.6) 67.2 (20.5)

Vitality 48.5 (20.8) 47.3 (20.9) 49.8 (20.5)

Social function 52.9 (27.8) 49.6 (28.5) 56.8 (26.6)

Role emotional 59.6 (31.8) 58.8 (32.9) 60.5 (30.7)

Mental health 63.6 (21.3) 62.2 (20.9) 65.2 (21.9)

* Statistically significant difference observed between groups

(p \ 0.05)

J Occup Rehabil

123

Table 2 Comparison of return-to-work assessment and rehabilitation program outcomes between groups

Entire sample (n = 225) Interview GROUP (n = 120) FCE group (n = 105)

Mean (SD) or percent

Assessment outcomes

Functional level determined at assessment*

Sedentary (\5 kg) 22.0 30.8 12.4

Light (6–10 kg) 25.8 27.5 23.8

Medium (10–20 kg) 43.1 38.3 48.6

Heavy/very heavy ([20 kg) 8.9 3.3 15.2

Assessment discharge outcome

Return-to-work 47.8 46.7 47.6

Fit-to-work 51.8 52.5 51.4

Not fit-to-work/other 0.4 0.8 1.0

Reason for not returning to work

Referred for further rehab services 70.6 76.2 66.7

Not job attached 12.6 12.7 13.0

Looking for employment 0.8 1.5 0

Retirement 0.9 1.5 0

Other reason 13.4 7.9 20.4

Rehabilitation recommendation

Multidisciplinary rehabilitation program 52.5 47.5 58.2

No intervention recommended 12.4 16.7 7.6

Single service provider (i.e. PT or Chiro) 35.1 35.8 34.3

Anticipated duration of rehabilitation (weeks)* 4.0 (1.3) 4.3 (1.5) 3.8 (0.9)

Actually underwent rehabilitation 46.2 40.8 52.4

(n = 104) (n = 49) (n = 55)

Underwent multidisciplinary rehabilitation

Type of program

Functional restoration 95.0 93.9 96.1

Complex chronic pain 5.0 6.1 3.9

Actual program duration (weeks) 2.7 (1.1) 2.9 (0.9) 2.6 (1.2)

Completed program 89.4 93.9 85.5

Admission functional work levels*

Sedentary (\5 kg) 19.2 24.5 14.5

Light (6–10 kg) 26.0 32.7 21.8

Medium (10–20 kg) 47.1 42.9 50.9

Heavy/very heavy ([20 kg) 7.7 0 12.7

Admission Pain Visual Analogue Scale (out of 10) 4.6 (2.3) 4.8 (2.6) 4.5 (2.0)

Admission Percent Pain Disability Index 48.1 (21.4) 49.6 (21.8) 46.8 (21.2)

Admission SF36 Raw Scores

Physical function 53.6 (22.3) 50.0 (24.7) 56.8 (21.2)

Role physical 28.1 (26.8) 24.7(25.8) 31.1 (27.5)

Bodily pain* 32.9 (19.7) 28.2 (19.6) 37.1 (19.0)

General health 67.7 (19.5) 70.2 (16.4) 65.6 (21.7)

Vitality 48.5 (20.4) 47.3 (21.9) 49.7 (19.2)

Social function 50.9 (27.4) 47.2 (28.6) 54.2 (26.2)

Role emotional 57.0 (32.6) 57.1 (36.2) 56.9 (29.4)

Mental health 61.5 (21.9) 60.3 (22.8) 62.6 (21.3)

SF36 Physical Health Composite Score* 33.8 (7.9) 32.1 (7.5) 35.3 (7.9)

J Occup Rehabil

123

Whitney U = 4,391.0, p \ 0.001). On average, claimants

undergoing FCE were rated as capable of performing

heavier work (mean 2.7/4 for FCE vs. 2.1/4 out of 4 for

functional interview, p \ 0.001), representing a mean dif-

ference of 0.6/4 (15 %). Differences between groups on

functional work levels recommended by the assessing cli-

nician are shown graphically in Fig. 1. Due to the imbal-

ance between groups on male/female ratio, a stratified

analysis was conducted based on sex. Differences between

groups remained statistically significant in both males and

females, with claimants undergoing FCE consistently rated

higher. Despite the differences observed on functional

work levels, no statistically significant differences were

observed on clinician RTW or rehabilitation recommen-

dations after assessment (see Table 2). However, claimants

undergoing the functional interview had significantly

higher average anticipated lengths of rehabilitation (4.3 vs.

3.8 weeks).

Among claimants who underwent multidisciplinary

rehabilitation, a statistically significant between-group

difference on baseline assessment functional work levels

was observed, however, differences on discharge func-

tional work levels were not statistically significant.

Functional work level change scores between admission

assessment and discharge were significantly lower in the

FCE group (Mann–Whitney U = 1,078.5, p \ 0.001).

Claimants in the FCE group experienced a lower mean

improvement of 0.3/4 (SRM = 0.35) compared with the

interview group that experienced a mean improvement of

0.9/4 (SRM = 0.94). These results remained statistically

significant (p \ 0.001) after controlling for sex and marital

status. Claimants who underwent the ‘self-report’ func-

tional interview also had significantly higher anticipated

durations of rehabilitation (4.3 vs. 3.8 weeks) indicating

clinicians thought these claimants would take longer to

achieve their functional goals. Differences between groups

on actual program durations were small and not statistically

significant (2.9 weeks for interview group vs. 2.6 weeks

for FCE group, p = 0.07). Other rehabilitation program

outcomes were comparable between groups, however,

claimants who underwent performance-based FCE reported

significantly higher pain and disability levels at time of

discharge from rehabilitation (see Table 2). Clinically

important improvements in pain and reported disability

following rehabilitation were only observed in the inter-

view group which experienced a mean improvement on the

Table 2 continued

(n = 104) (n = 49) (n = 55)

Sf36 Mental Health Composite Score 41.7 (13.2) 41.4 (14.5) 41.9 (12.1)

Rehabilitation discharge outcome

Return-to-work 83.7 83.7 83.6

Fit-to-work 16.3 16.3 16.4

Not fit-to-work/other 0 0 0

Discharge functional work levels

Sedentary (\5 kg) 8.1 10.2 6.0

Light (6–10 kg) 13.1 12.2 14.0

Medium (10–20 kg) 46.5 40.8 52.0

Heavy/very heavy ([20 kg) 32.3 36.7 28.0

Discharge pain VAS (out of 10, n = 76)* 3.0 (2.3) 2.5 (2.2) 3.6 (2.2)

Discharge Percent Pain Disability Index (n = 76)* 30.5 (23.7) 25.3 (23.0) 36.1 (23.5)

Discharge SF36 Raw Scores (n = 76)

Physical function 68.1 (24.0) 69.8 (25.2) 66.3 (22.9)

Role physical 47.2 (28.3) 49.5 (30.1) 44.7 (26.4)

Bodily pain 49.5 (22.1) 53.1 (24.5) 45.6 (18.9)

General health 69.2 (20.3) 72.0 (16.8) 66.1 (23.3)

Vitality 55.5 (20.0) 57.0 (20.3) 53.8 (19.9)

Social function 64.7 (29.1) 69.2 (27.1) 59.9 (30.8)

Role emotional 64.8 (29.2) 66.2 (30.5) 63.3 (28.1)

Mental health 66.0 (21.8) 69.9 (18.9) 61.9 (24.1)

SF36 Physical Health Composite Score 40.6 (9.1) 41.4 (10.0) 39.8 (8.0)

Sf36 Mental Health Composite Score 44.0 (12.3) 45.7 (11.4) 42.1 (13.2)

* Statistically significant difference observed between groups (p \ 0.05)

J Occup Rehabil

123

Pain VAS = 2.0/10 (SRM = 0.88) and PDI = 21.8/100

(SRM = 1.45) compared with mean improvement in the

FCE group on the Pain VAS = 0.6/10 (SRM = 0.30) and

PDI = 9.5/100 (SRM = 0.37). These results remained

statistically significant (p \ 0.001) after controlling for

baseline pain and disability reports as well as sex and

marital status. None of the other variables tested had an

important confounding effect.

Compensation Outcomes

No statistically significant differences were observed

between groups on any of the compensation outcomes (see

Table 3). Outcomes were quite positive, with 92 % of

claimants off all wage replacement benefits by 180 days

after assessment in both groups.

Telephone Interview Outcomes

No statistically significant differences were observed

between groups on any of the key outcomes obtained

during telephone interviews (see Table 3). Our hypothesis

related to higher functional work levels after undergoing

FCE was not supported as reported working levels were

similar between groups (mean = 2.6, 2.6, 2.7 for FCE and

2.6, 2.8, 3.0 for interview at 1, 3, and 6 months respec-

tively, [ 0.05). This represents mean differences between

groups ranging from 0/4 to 0.3/4 in favour of the interview

(0–8 %). At the 3- and 6-month follow-up interview

Fig. 1 Comparison of functional work levels recommended by

assessing clinician at time of return-to-work assessment. Differences

are statistically significant (Mann–Whitney test p \ 0.001, n = 225)

Table 3 Comparison of workers’ compensation and telephone

interview outcomes between groups

Variable Interview

group

FCE group

(n = 120)

p value

(n = 105)

Compensation benefit

outcomes

Number/

percent

Receiving benefits at

assessment

Total temporary 40 (33.3) 30 (28.6) 0.44

Partial temporary 15 (12.5) 21 (21.0) 0.09

Total vocational 0 (0) 0 (0)

Partial vocational 0 (0) 0 (0)

Any compensation benefits 55 (45.8) 51 (48.6) 0.68

Benefits at 30-days post

assessment

Total temporary 26 (21.7) 13 (12.4) 0.07

Partial temporary 22 (18.3) 22 (21.0) 0.62

Total vocational 0 (0) 0 (0)

Partial vocational 0 (0) 0 (0)

Any compensation benefits 48 (40.0) 35 (33.3) 0.30

Benefits at 90-days post

assessment

Total temporary 7 (5.8) 2 (1.9) 0.13

Partial temporary 4 (3.3) 5 (4.8) 0.59

Total vocational 0 (0) 3 (2.9) 0.06

Partial vocational 2 (1.7) 1 (1.0) 0.64

Any compensation benefits 13 (10.8) 11 (10.5) 0.93

Benefits at 180-days post

assessment

Total temporary 2 (1.7) 1 (1.0) 0.64

Partial temporary 3 (2.5) 2 (1.9) 0.76

Total vocational 0 (0) 1 (1.0) 0.28

Partial vocational 3 (2.5) 1 (1.0) 0.38

Any compensation benefits 8 (6.7) 5 (4.8) 0.54

Telephone interview outcomes Number/percent or mean (SD)

Participated in interview 54 (45.0) 31 (29.5) 0.02

Reported pre-injury demands

(n = 76)

Sedentary 8 (16.3) 4 (14.8) 0.61

Light 4 (8.2) 2 (7.4)

Medium 19 (38.8) 7 (25.9)

Heavy/very heavy 18 (36.7) 14 (51.9)

Self-rated work ability

compared to lifetime best

(out of 10)

1 month (n = 76) 6.0 (3.1) 6.9 (2.7) 0.20

3 months (n = 54) 7.5 (2.4) 7.4 (2.8) 0.91

6 months (n = 49) 6.5 (2.9) 6.1 (2.7) 0.66

Numeric pain rating (out

of 10)

1 month 4.1 (2.6) 3.3 (2.3) 0.20

3 months 3.2 (2.4) 2.5 (2.6) 0.36

6 months 3.3 (2.7) 4.0 (2.4) 0.38

J Occup Rehabil

123

intervals, a higher percentage of claimants in the functional

interview group reported working at a heavy/very heavy

level (33 and 40 vs. 27 and 31 % respectively) although

this difference was not statistically different (p = 0.9 and

0.09). Due to the imbalance in telephone interview follow-

up rates between claimants with back and non-back related

injuries, we conducted a stratified analysis on all key out-

come measures. This analysis indicated part of body was

not an important confounding variable.

Adverse or Unintended Effects

No new or re-injuries associated with assessments were

reported during the trial, however we did receive 5 com-

plaints from case managers. These all questioned the use of

functional interviews instead of performance-based FCE.

Case managers consistently considered the functional

interviews to be less ‘objective’ than FCE.

Discussion

Performance-based FCE does not appear to enhance out-

comes when integrated into the process of occupational

rehabilitation. While claimants undergoing performance-

based FCE were rated as capable of performing substan-

tially higher functional work levels (*15 % higher with

FCE) at baseline as compared to those who underwent a

semi-structured functional interview, no statistically sig-

nificant or clinically meaningful differences were observed

on the functional work levels or compensation outcomes at

any other follow-up period. Claimants who underwent the

‘self-report’ functional interview also had significantly

higher anticipated durations of rehabilitation indicating

Table 3 continued

Variable Interview

group

FCE group

(n = 120)

p value

(n = 105)

1-month work outcomes

Work status

Regular duties full time 8 (16.3) 8 (29.6) 0.27

Regular duties part time 3 (6.1) 1 (3.7)

Modified duty full time 11 (22.4) 6 (22.2)

Modified duty part time 10 (20.4) 1 (3.7)

Not working 17 (34.7) 11 (40.7)

Work lifting requirements at 1

month

Sedentary 9 (28.1) 5 (31.2) 0.85

Light 3 (9.4) 2 (12.5)

Medium 12 (37.5) 4 (25.0)

Heavy/very heavy 8 (25.0) 5 (31.2)

Why not working at 1 month?

Unable due to injury 9 (52.9) 1 (9.1) 0.01

Unemployed but seeking

work

2 (11.8) 4 (36.4)

In retraining due to injury 5 (29.4) 2 (18.2)

Not working for other

reasons

0 (0) 4 (36.4)

New injury 1 (5.9) 0 (0)

3-month outcomes

Work status

Regular duties full time 23 (62.2) 8 (47.1) 0.61

Regular duties part time 2 (5.4) 1 (5.9)

Modified duty full time 4 (10.8) 1 (5.9)

Modified duty part time 2 (5.4) 1 (5.9)

Not working 6 (16.2) 6 (35.3)

Work lifting requirements at

3 months

Sedentary 6 (20.0) 2 (18.2) 0.90

Light 5 (16.7) 3 (27.3)

Medium 9 (30.0) 3 (27.3)

Heavy/very heavy 10 (33.3) 3 (27.3)

Why not working at

3 months?

Unable due to injury 3 (50.0) 3 (50.0) 0.94

Unemployed but seeking

work

1 (16.7) 0 (0)

In retraining due to injury 1 (16.7) 2 (33.3)

Not working for other

reasons

1 (16.7) 1 (16.7)

6-month outcomes

Work status

Regular duties full time 17 (51.5) 9 (56.2) 0.97

Regular duties part time 4 (12.1) 2 (12.5)

Modified duty full time 3 (9.1) 1 (6.2)

Modified duty part time 1 (3.0) 1 (6.2)

Not working 8 (24.2) 3 (18.8)

Table 3 continued

Variable Interview

group

FCE group

(n = 120)

p value

(n = 105)

Work lifting requirements at

6 months

Sedentary 2 (8.0) 4 (30.8) 0.09

Light 6 (24.0) 0 (0)

Medium 7 (28.0) 5 (38.5)

Heavy/very heavy 10 (40.0) 4 (30.8)

Why not working at

6 months?

Unable due to injury 3 (37.5) 1 (33.3) 0.78

Unemployed but seeking

work

2 (25.0) 1 (33.3)

Not working for other

reasons

2 (25.0) 0 (0)

New injury 1 (12.5) 1 (33.3)

J Occup Rehabil

123

clinicians thought these claimants would take longer to

achieve their functional goals. However, by 1-month after

assessment all functional and RTW outcomes were com-

parable between groups, even slightly favouring the func-

tional interview group.

Within the subgroup of claimants who underwent mul-

tidisciplinary rehabilitation, the performance-based FCE

group was assessed as having significantly higher func-

tional work levels at admission assessment when compared

to the functional interview group, but discharge functional

levels were comparable. Indeed, claimants in the functional

interview group improved their functional work levels to a

greater extent than the FCE group. Unexpectedly, only

claimants in the functional interview group experienced

clinically important improvements in reported pain inten-

sity [48] and disability ratings [49] over the course of their

rehabilitation despite having worse bodily pain and phys-

ical health scores at admission. This was despite non-sig-

nificant, fairly small (*2.5 days) differences in overall

program duration between the two groups and could

potentially be due to pain exacerbation (i.e. flare-up) that is

frequently associated with performance-based FCE testing

due to the maximal physical effort required [11, 12].

Findings indicate that a semi-structured functional inter-

view incorporating some of the principles of motivational

interviewing used by clinicians who are trained and expe-

rienced in FCE may be as clinically useful as performance-

based FCE, with potential for efficiency gains and likeli-

hood of greater improvement in some clinical outcomes.

The observation that function demonstrated during

performance-based FCE is higher than self-reported ability

is consistent with previous research [15, 25]. Prior studies

have indicated that outcomes from self-report measures are

systematically lower than performance testing, with self-

estimates of lifting ability lower by 3–8 kg. However, this

is the first study to attempt to rigorously evaluate whether

these differences are sustainable and translate into

improved clinical and RTW outcomes. Results suggest that

these differences do not lead into improved outcomes.

However, further research is needed to determine if our

findings are repeatable in other jurisdictions and contexts.

The fact that we received complaints from compensation

case managers during the study indicates there is some

perceived value of performance-based assessment among

relevant stakeholders. We are currently examining the issue

of stakeholder perceptions of FCE within this jurisdiction

in an associated qualitative study.

Strengths of this study include the use of cluster ran-

domized controlled trial methods incorporated into a

pragmatic clinical context. Data were gathered in an actual

clinical setting as part of routine patient care and few

restrictions were placed on our sample, indicating it should

be fairly representative of this population. We were able to

obtain 100 % follow-up on the compensation outcomes,

and randomisation appeared adequate with the exception of

more OTs randomized to conduct functional interviews.

However, clinician discipline was not associated with

outcomes and little evidence of a clustering effect was

observed indicating the unequal clinical disciplines did not

influence outcomes. Other limitations include low partici-

pation in follow-up interviews (38 %). This is consistent

with research involving workers’ compensation claimants

[50]. However, few systematic differences were observed

between those who did and not participate in follow-up

contacts and stratified analysis showed there was no con-

founding by body part affected which differed between

groups. Of interest, we found that claimants undergoing

functional interview instead of FCE were much more likely

to participate in follow-up interviews. It may be that asking

claimants to thoroughly discuss their functional difficulties

at assessment made them more open to participate in

subsequent discussions. While reliability of the functional

interview has not been previously tested, there is some

prior evidence of construct validity of the manual handling

component and this study provides further validity evi-

dence for its utility as a measure of work ability. Further

research on the functional interview is recommended

including evaluation of its reliability and other psycho-

metric properties.

Conclusion

Performance-based FCE integrated into occupational

rehabilitation appears to lead to higher baseline functional

levels compared to a semi-structured functional interview,

but not improved RTW rates or functional work levels at

follow-up. Functional interviewing has potential for effi-

ciency gains and higher likelihood of improved functional

work levels, pain and disability following rehabilitation.

Further research is indicated on the reliability and validity

of the functional interview tool.

Acknowledgments This research was supported with funds from

WorkSafeBC through the Focus on Tomorrow program. WCB-

Alberta Millard Health assisted with data acquisition and study

implementation. The clinical trial registration number was IS-

RCTN61284905 and the protocol can be accessed at the following

website: http://www.controlled-trials.com/ISRCTN61284905/. The

authors have no competing interests or financial arrangements that

would represent a conflict of interest for this research.

References

1. Martin BI, Deyo RA, Mirza SK, Turner JA, Comstock BA,

Hollingworth W, et al. Expenditures and health status among

J Occup Rehabil

123

adults with back and neck problems. JAMA. 2008;

299(6):656–64.

2. Murray CJ, Vos T, Lozano R, Naghavi M, Flaxman AD, Michaud

C, et al. Disability-adjusted life years (DALYs) for 291 diseases

and injuries in 21 regions, 1990–2010: a systematic analysis for

the Global Burden of Disease Study 2010. Lancet. 2012;

380(9859):2197–223.

3. Hadler NM. Occupational musculoskeletal disorders. 3rd ed.

Philadelphia: Lippincott Williams & Wilkins; 2005.

4. Innes E, Straker L. A clinician’s guide to work-related assess-

ments: 1-purposes and problems. WORK. 1998;11:183–9.

5. Gross DP, Haws C, Niemelainen R. What is the rate of functional

improvement during occupational rehabilitation in workers’

compensation claimants? J Occup Rehabil. 2012;22(3):292–300.

6. Rainville J, Kim RS, Katz JN. A review of 1985 Volvo Award

winner in clinical science: objective assessment of spine function

following industrial injury: a prospective study with comparison

group and 1-year follow-up. Spine. 2007;32(18):2031–4.

7. Mayer TG, Gatchel RJ, Kishino N, Keeley J, Capra P, Mayer H,

et al. Objective assessment of spine function following industrial

injury. A prospective study with comparison group and one-year

follow-up. Spine. 1985;10(6):482–93.

8. Isernhagen SJ. The comprehensive guide to work injury man-

agement. Gaithersburg, MD: Aspen; 1995.

9. Wind H, Gouttebarge V, Kuijer PP, Sluiter JK, Frings-Dresen

MH. The utility of functional capacity evaluation: the opinion of

physicians and other experts in the field of return to work and

disability claims. Int Arch Occup Environ Health. 2006;

79(6):528–34.

10. Menard MR, Hoens AM. Objective evaluation of functional

capacity: medical, occupational, and legal settings. JOSPT.

1994;19(5):249–60.

11. Soer R, Groothoff JW, Geertzen JH, van der Schans CP, Reesink

DD, Reneman MF. Pain response of healthy workers following a

functional capacity evaluation and implications for clinical

interpretation. J Occup Rehabil. 2008;18(3):290–8.

12. Reneman MF, Kuijer W, Brouwer S, Preuper HR, Groothoff JW,

Geertzen JH, et al. Symptom increase following a functional

capacity evaluation in patients with chronic low back pain: an

explorative study of safety. J Occup Rehabil. 2006;16(2):

197–205.

13. Gross DP, Battie MC. Factors influencing results of functional

capacity evaluations in workers’ compensation claimants with

low back pain. Phys Ther. 2005;85(4):315–22.

14. Reneman MF, Kool J, Oesch P, Geertzen JH, Battie MC, Gross

DP. Material handling performance of patients with chronic low

back pain during functional capacity evaluation: a comparison

between three countries. Disabil Rehabil. 2006;28(18):1143–9.

15. Asante AK, Brintnell ES, Gross DP. Functional self-efficacy

beliefs influence functional capacity evaluation. J Occup Rehabil.

2007;17(1):73–82.

16. van Abbema R, Lakke SE, Reneman MF, van der Schans CP, van

Haastert CJ, Geertzen JH, et al. Factors associated with functional

capacity test results in patients with non-specific chronic low

back pain: a systematic review. J Occup Rehabil. 2011;. doi:10.

1007/s10926-011-9306-4.

17. Gross DP, Battie MC. Predicting timely recovery and recurrence

following multidisciplinary rehabilitation in patients with com-

pensated low back pain. Spine. 2005;30(2):235–40.

18. Gross DP, Battie MC, Cassidy JD. The prognostic value of

functional capacity evaluation in patients with chronic low back

pain: part 1: timely return to work. Spine. 2004;29(8):914–9.

19. Matheson LN, Isernhagen SJ, Hart DL. Relationships among

lifting ability, grip force, and return to work. Phys Ther.

2002;82(3):249–56.

20. Kuijer PP, Gouttebarge V, Brouwer S, Reneman MF, Frings-

Dresen MH. Are performance-based measures predictive of work

participation in patients with musculoskeletal disorders? A sys-

tematic review. Int Arch Occup Environ Health. 2012;

85(2):109–23.

21. Streibelt M, Blume C, Thren K, Reneman MF, Mueller-Fahrnow

W. Value of functional capacity evaluation information in a

clinical setting for predicting return to work. Arch Phys Med

Rehabil. 2009;90(3):429–34.

22. Matheson LN, Matheson ML, Grant J. Development of a measure

of perceived functional ability. J Occup Rehabil. 1993;3:15–30.

23. Isernhagen SJ. Functional capacity evaluation: rationale, proce-

dure, utility of the kinesiophysical approach. J Occup Rehabil.

1992;2(3):157–68.

24. Reneman MF, Soer R, Gross DP. Developing research on per-

formance-based functional work assessment: report on the first

international functional capacity evaluation research meeting.

J Occup Rehabil. 2013;. doi:10.1007/s10926-013-9425-1.

25. Brouwer S, Dijkstra PU, Stewart RE, Goeken LN, Groothoff JW,

Geertzen JH. Comparing self-report, clinical examination and

functional testing in the assessment of work-related limitations in

patients with chronic low back pain. Disabil Rehabil. 2005;

27(17):999–1005.

26. Reneman MF, Jorritsma W, Schellekens JM, Goeken LN. Con-

current validity of questionnaire and performance-based disabil-

ity measurements in patients with chronic nonspecific low back

pain. J Occup Rehabil. 2002;12(3):119–29.

27. Wittink H, Rogers W, Sukiennik A, Carr DB. Physical func-

tioning: self-report and performance measures are related but

distinct. Spine. 2003;28(20):2407–13.

28. Elbourne DR, Campbell MK. Extending the CONSORT state-

ment to cluster randomized trials: for discussion. Stat Med.

2001;20(3):489–96.

29. Gross DP, Battie MC, Asante AK. Evaluation of a short-form

functional capacity evaluation: less may be best. J Occup Reha-

bil. 2007;17(3):422–35.

30. Klar N, Donner A. Current and future challenges in the design

and analysis of cluster randomization trials. Stat Med.

2001;20(24):3729–40.

31. Eldridge S, Cryer C, Feder G, Underwood M. Sample size cal-

culations for intervention trials in primary care randomizing by

primary care group: an empirical illustration from one proposed

intervention trial. Stat Med. 2001;20(3):367–76.

32. Stephens B, Gross DP. The influence of a continuum of care

model on the rehabilitation of compensation claimants with soft

tissue disorders. Spine. 2007;32(25):2898–904.

33. McHorney CA, Ware JE Jr, Raczek AE. The MOS 36-Item Short-

Form Health Survey (SF-36): II. Psychometric and clinical tests

of validity in measuring physical and mental health constructs.

Med Care. 1993;31(3):247–63.

34. Pollard CA. Preliminary validity study of the pain disability

index. Percept Mot Skills. 1984;59(3):974.

35. Finch E, Brooks D, Stratford P, Mayo N. Physical rehabilitation

outcome measures: a guide to enhanced clinical decision making.

2nd ed. Toronto: Canadian Physiotherapy Association; 2002.

36. Soer R, Poels BJ, Geertzen JH, Reneman MF. A comparison of

two lifting assessment approaches in patients with chronic low

back pain. J Occup Rehabil. 2006;16(4):639–46.

37. Brouwer S, Reneman MF, Dijkstra PU, Groothoff JW, Schelle-

kens JMH, Goeken LNH. Test-retest reliability of the Isernhagen

Work Systems Functional Capacity Evaluation in patients with

chronic low back pain. J Occup Rehabil. 2003;13:207–18.

38. Reneman MF, Jaegers S, Westmaas M, Goeken L. The reliability

of determining effort level of lifting and carrying in a functional

capacity evaluation. Work. 2002;18:23–7.

J Occup Rehabil

123

39. Reneman MF, Dijkstra PU, Westmaas M, Goeken LN. Test-retest

reliability of lifting and carrying in a 2-day functional capacity

evaluation. J Occup Rehabil. 2002;12(4):269–75.

40. Gross DP, Battie MC. Construct validity of a kinesiophysical

functional capacity evaluation administered within a worker’s

compensation environment. J Occup Rehabil. 2003;13(4):

287–95.

41. Durand MJ, Loisel P, Hong QN, Charpentier N. Helping clini-

cians in work disability prevention: the work disability diagnosis

interview. J Occup Rehabil. 2002;12(3):191–204.

42. Andrew Walsh D, Jane Kelly S, Sebastian Johnson P, Rajkumar

S, Bennetts K. Performance problems of patients with chronic

low-back pain and the measurement of patient-centered outcome.

Spine. 2004;29(1):87–93.

43. Shannon R, Hillsdon M. Motivational interviewing in musculo-

skeletal care. Musculoskeletal Care. 2007;5(4):206–15.

44. National Occupational Classification: Occupational descriptions.

Employment and Social Development Canada; 2011. Online

resource accessed October 30, 2013. Website: www.hrsdc.gc.ca/

eng/jobs/lmi/noc/index.shtml.

45. Krause N, Dasinger LK, Deegan LJ, Brand RJ, Rudolph L.

Alternative approaches for measuring duration of work disability

after low back injury based on administrative workers’ com-

pensation data. Am J Ind Med. 1999;35(6):604–18.

46. Donner A, Klar N. Pitfalls of and controversies in cluster ran-

domization trials. Am J Public Health. 2004;94(3):416–22.

47. Hopman WM, Towheed T, Anastassiades T, Tenenhouse A,

Poliquin S, Berger C, et al. Canadian normative data for the SF-

36 health survey. Canadian Multicentre Osteoporosis Study

Research Group. CMAJ. 2000;163(3):265–71.

48. Farrar JT, Portenoy RK, Berlin JA, Kinman JL, Strom BL.

Defining the clinically important difference in pain outcome

measures. Pain. 2000;88(3):287–94.

49. Soer R, Reneman MF, Vroomen PC, Stegeman P, Coppes MH.

Responsiveness and minimal clinically important change of the

pain disability index in patients with chronic back pain. Spine.

2012;12(11):1035–9.

50. Gross DP, Battie MC. Functional capacity evaluation perfor-

mance does not predict sustained return to work in claimants with

chronic back pain. J Occup Rehabil. 2005;15(3):285–94.

J Occup Rehabil

123