Occupational and Environmental Medicine 1994;51:597-602

Management of occupational back pain: theSherbrooke model. Results of a pilot andfeasibility study

Patrick Loisel, Pierre Durand, Lucien Abenhaim, Lise Gosselin, Robert Simard,Jean Turcotte, John M Esdaile

Department ofSurgery (Division ofOrthopaedics)P LoiselDepartment ofHealthCommunity Sciences,University deSherbrooke, Quebec,CanadaP DurandL GosselinCenter for ClinicalEpidemiology andCommunity Studies,Jewish GeneralHospital, McGillUniversity, CanadaL Abenhaim.Department ofCommunity Health,Sherbrooke, Quebec,CanadaR SimardJ TurcotteDepartment ofMedicine (Division ofClinicalEpidemiology), TheMontreal GeneralHospital, McGillUniversity, CanadaJ EsdaileCorrespondence to:Dr P Loisel, Department ofSurgery (Division ofOrthopaedics), Universit6de Sherbrooke, Quebec,Canada.

Accepted 26 April 1994

AbstractObjectives-The aim was to combatoccurrence of chronic occupational backpain.Methods-A multidisciplinary model tomanage back pain that includes bothclinical and ergonomic approaches hasbeen developed. Early detection, earlyclinical and ergonomic evaluations, andearly active treatment make up the cor-nerstone of management. Detection ofcases starts after four weeks of absencefrom work. An ergonomic intervention isimplemented at six weeks. A medicalspecialist is involved at eight weeks. Ifreturn to work is not possible after 12weeks, a functional recovery therapy fol-lowed by a therapeutic return to work isimplemented. A multidisciplinary teamdecides if return to original or modifiedwork is possible or if vocational rehabili-tation is necessary. This model has beenimplemented by the investigators in theSherbrooke (Quebec, Canada) area, andis presently being evaluated through arandomised trial in 31 industrial settle-ments (about 20 000 workers). A clusterrandomization of industries and workerswill allow separate testing of ergonomicand clinical interventions.Results-One year after implementation,31 of 35 of the eligible industrial sitesparticipated in the study and 79 of 88 ofthe eligible workers affected by recentback pain had agreed to participate.Ergonomic and clinical interventionshave been implemented as planned. Onlythree workers dropped out. Hence thisglobal clinical and ergonomic manage-ment programme has been shown to befeasible in a general population.Conclusion-A global management pro-gramme of back pain joining ergonomicand clinical intervention with a multidis-ciplinary approach has not been testedyet. Linking these two strategies in asame multidisciplinary team represents asystemic approach to this multifactorialailment. During the first year of this trialwe did not find any conflict between thesetwo interventions from the employer's orworker's point ofview.

(Occup Environ Med 1994;51:597-602)

Occupational back pain in industrializedcountries is costly. In the United States direct

costs have been estimated to vary between 20and 55 billion dollars.'

Despite Nachemson's call for increasedresearch on occupational back pain,2 both thenumber of patients with chronic back painand the related costs continue to increase.Inadequate management of industrial backpain may be the major cause."- For this rea-son, an innovative model of management ofindustrial back pain that extends the recom-mendations of the Quebec task force on spinaldisorders in the workplace has been devel-oped.4 The model, described in this paper,includes scheduled interventions for theworkers sustaining back injuries and interven-tions at their worksites. It is presently beingevaluated through a randomised trial in thearea of Sherbrooke (Quebec, Canada).

Rationale for the modelOccupational episodes of back pain are oftenself limited. Seventy per cent of cases healspontaneously in less than four weeks, and95% in less than six months.7 The Quebectask force, reported that 75% of the totalcosts for occupational back pain in theprovince of Quebec (Canada) for the year1981 arose from 7% of workers with chronicdisease (absent from work for more than sixmonths). Moreover, recurrence rate ofepisodes of back pain has been reported to be36&3% over three years after a first episode.8Others have reported similar figures.910

Because both the costs and disability arisefrom the chronic cases, the severity of backpain is usually graded more on the basis ofsymptom duration than on a specific patho-logical abnormality.'3 4 1' 12 In fact, a precisediagnosis cannot be made in most chroniccases.3 4 11 13 Even if a precise diagnosis can bemade, treatment often remains aimed at reliefof symptoms. Trial and error is often theguide. As a result, even treatments that havebeen proved ineffective continue to be used,and few treatments have been proved effec-tive.4 Among treatments considered "proba-bly effective" by the Quebec task force, some,such as transcutaneous electronic nerve stimu-lation'4 15 or facet joint blocks'6 17 18 have sub-sequently been shown to be no more effectivethan a placebo.

Both individual and occupational riskfactors have been associated with back pain.Individual factors include previous episodesof back pain,13 19 20 physical signs andsymptoms,12 19 coping difficulties,"3 psycho-logical distress,2' physical fitness,22 personal

597

on May 28, 2020 by guest. P

rotected by copyright.http://oem

.bmj.com

/O

ccup Environ M

ed: first published as 10.1136/oem.51.9.597 on 1 S

eptember 1994. D

ownloaded from

Loisel, Durand, Abenhaim, Gosselin, Simard, Turcortte, Esdaile

habits,23 and education level.2425 Occupationalproblems include heavy work,2S28 specifictypes of work,26 jobsite arrangement andorganisation,29 exposure to vibration,'0 jobsatisfaction," and the adversarial nature ofthe workmen's compensation system.'23' Therole of these markers in an individual patient'sdisease is usually difficult to determine. Thustreating the 5-7% of chronic cases who do notheal by themselves is problematic. These diffi-culties explain the current management ofchronic back pain, which seems to be disor-ganised, irrational, delayed, and often of aconflicting nature.

Disorganisation comes from the many dif-ferent therapists (different medical and surgicalspecialists, acupuncturists, chiropractors,physiotherapists, and others) who maybecome involved over the course of a patient'sillness. Various treatments are offered (forexample, pills, herbs, injections, manipula-tions, physical therapies, surgery). Despite theavailability of numerous different treatmentsfrom various therapists, no single approachhas been proved more effective than the oth-ers. When back pain does not resolve, disabil-ity occurs, leading to prolonged absence fromwork. This has been named chronicity.3"33 Souncertainty in diagnosis leads to irrationaltreatments, and treatment by trial and error.Also, the compensation process is associatedwith chronicity.'4 Compensation for occupa-tional back pain is determined mainly by theregulations of workers' compensation boardssometimes involving legal actions. The regula-tions take precedence over the medical needsof the patient.32

Description ofthe Sherbrooke modelThe challenge of treating chronic back painhas resulted in novel responses. The Quebectask force proposed a diagnostic grid and an

organigram of desirable interventions to beimplemented at specific times. The secondincluded the advice of a medical specialistafter seven weeks of absence from work, activetreatment after eight weeks, and when noimprovement occurred, early vocational reha-bilitation.4 The usefulness of an ergonomicintervention was also suggested. Mayer et alhave claimed that the multidisciplinary reha-bilitation programmes that employ physicalfitness and work conditioning programmesalong with a cognitive-behavioural approachare effective.35 Wiesel showed that an appro-priate management in collaboration withpractising physicians was effective in a largeindustrial settlement, dramatically reducingthe duration of absence from work and thenumber of surgical interventions.6 Recentstudies have focused on the close relationbetween back pain and job satisfaction,'36 butpractical solutions are not yet available.Although jobsite problems have been associ-ated with back pain, no intervention studiesaimed at increasing the return to work ofinjured workers before chronicity is estab-lished, have been performed.The Sherbrooke model proposes to link

these different approaches to provide inte-grated interventions, directed at both theworker and the jobsite (fig 1). A specialisedmultidisciplinary back pain clinic comprisinga clinical and an ergonomic team, has earlyintervention to reduce chronicity as the mainobjective. Formal agreements have beennegotiated between the clinic and industrialpartners. To avoid unnecessary efforts andcosts for the 70% of workers who return towork before four weeks, recruitment ofworkersin the back pain clinic begins at the fourthweek of absence from work. Recurrences aretaken into account by summing the days ofabsence from work due to back pain duringthe past year. Also all cases with one day or

Figure 1 Scheme offullintervention representativeof the Sherbrooke model ofback pain management.Return to work is possibleat any time and will stopthis process of intervention.

Step 0

Detection of casesat risk of chronicity

4 weeks

Recruitmentof workers

Consent

Randomisation

Step 1

Worksite intervention

6-10 weeks

Occupationalphysician

Participativeergonomics

Step 2

Diagnostic andback school

8-12 weeks

*..- Back pain specialist

Back school

Step 3

Early rehabilitation

13-26 weeksFunctional rehabilitation therapy

+ therapeutic return to work

Management and prognosisby multidisciplinary team

End issue

Before 6 months

Return to previous work

Return to modified work

Vocational rehabilitation

598

on May 28, 2020 by guest. P

rotected by copyright.http://oem

.bmj.com

/O

ccup Environ M

ed: first published as 10.1136/oem.51.9.597 on 1 S

eptember 1994. D

ownloaded from

Management of occupational back pain: the Sherbrooke model

more of absence from work for back pain aredeclared to the research team. The industrialpartners refer cases directly to the back painclinic. Three successive management steps areused.

STEP 1The first step is mainly occupational. On thesixth week of absence from work, an occupa-tional physician examines the worker and anergonomist visits the jobsite. The occupa-tional physician is knowledgeable about backpain and the general characteristics of the dif-ferent participating industrial partners. Afterexamining the worker and obtaining worksiteinformation, appropriate recommendationsare submitted to the worker's general practi-tioner. Depending on the clinical state of theworker, the difficulty of the worker's tasks,and the availability of light duties, he may forinstance, suggest ways to improve diagnosis ortreatment, or propose a return to light or fullduties. At the same time, an interventioncalled "participative ergonomics" is imple-mented at the jobsite with the collaboration ofthe injured worker. Participative ergonomicsinvolves representatives of management andthe unions who, with the help of the ergono-mist of the back pain clinic, develop improve-ments in the work routine and evaluate jobsatisfaction.37 38 At that time an industrial part-ner is enrolled a two day orientation course isgiven to selected employees and managementpersonnel, who will subsequently serve as theon site back pain advisory group for that com-pany. The course includes teaching on occu-pational risk factors for back pain, ergonomicanalysis, and the methods used in participa-tive ergonomics. For each incident case theserepresentatives meet with the injured worker,their supervisor, and the ergonomist. Theergonomic process takes the following steps.Firstly, the precise nature of the worker'stasks is clarified from the descriptions madeseparately by the employer and the worker.Then the work tasks are observed by theergonomist, often in the presence of theinjured worker. Data collected include workprocess, characteristics of other jobs linked tothe tasks involved, features of equipment anddesign of the workplace, loads handled, preci-sion, quality, quantities handled, pace of thejob, postural requirements, and environmen-tal characteristics of the jobsite. After theseobservations, an "ergonomic diagnosis" ismade with regard to the back, and discussedwith the injured worker, the supervisor, andthe back pain group. Solutions are then dis-cussed and proposed to the management. Theemployer can choose to disregard these rec-ommendations. The ergonomist is madeaware of the employer's decision, and this caninfluence the timing of the injured worker'sreturn to work.The role of the back pain clinic ergonomist

is to help to find solutions to improve theworksite, as far as the back is concerned, andnot to implement recommended modifica-tions. This remains the employer's responsi-bility, but ergonomic problems are detected

and proposed solutions are aimed at an ear-lier, safer, and stable return to work.

STEP 2The second step focuses on diagnosis andinstruction at a back school. If return to workhas not occurred after seven weeks, theworker is examined by a medical specialistfrom the team to exclude a serious underlyingcondition (for example, malignancy). Anytests or consultations required are obtained.In the absence of a serious underlying condi-tion that explains the back pain, the worker isdirected to a back school. Back schools wererecognised as effective for subacute back painby the Quebec task force, but recently, doubthas been raised as to their value.39 TheSherbrooke back school is somewhat differentfrom the usual. It is an activity that lasts forone hour every day for four weeks.Throughout the 20 hours of sessions, backeducation, coaching, and practice of appro-priate exercises and counselling for daily lifeactivities are provided.

STEP 3If return to work has not occurred after theback school programme (after about 12 weeksof absence from work), we consider the backpain to be chronic. The third step of manage-ment is instituted. This step, named func-tional rehabilitation therapy, is a modifiedMayer's intervention,'5 including fitnessdevelopment and conditioning for work with acognitive behavioural approach. It is carriedout by the multidisciplinary team of the backpain clinic. The functional rehabilitation ther-apy begins with a one week evaluation ofphysical fitness, functional capacity, and psy-chological state. New protocols have beendeveloped for this evaluation. Physical fitnessis evaluated by the physical educator of theteam, through measurements of cardiorespira-tory capacity, muscular strength andendurance, joint flexibility, and life habits.The occupational therapist performs a fourday evaluation. This includes a speciallydeveloped set of isoinertial tests that repro-duce common work situations, such as carry-ing loads at different heights or performingtasks in different postures. Time and weightsare measured. Precise knowledge of theworker's normal tasks on the job are providedby the ergonomist. Psychological state is eval-uated by a psychologist through question-naires (MMPI, Beck) and a structuredinterview. A specific physical education, workhardening, and cognitive-behavioural supportprogramme is then implemented for three tofive weeks. It allows for a development of theglobal condition of the worker, and forimproving specific skills and endurancerequired by the specific worker's tasks. Morerealistic expectations concerning the backcondition and pain management skills aretaught. This development of global fictionalcapacities is followed by a progressive returnto work, named therapeutic return to work.This involves alternating days of work andtreatment. Work is progressively increased.

599

on May 28, 2020 by guest. P

rotected by copyright.http://oem

.bmj.com

/O

ccup Environ M

ed: first published as 10.1136/oem.51.9.597 on 1 S

eptember 1994. D

ownloaded from

Loisel, Durand, Abenhaim, Gosselin, Simard, Turcortte, Esdaile

Therapeutic return to work allows for a morecomplete recovery before returning to workon a full time basis, and helps to lessen thestress of return without recourse to specialisedcounselling. Therapeutic return to work isspecifically designed for each individual case.The occupational therapist and the worker'ssupervisor agree on an appropriate and realisticprogression of tasks.

Finally, taking into account the functionalcapacities of the worker during the functionalrehabilitation therapy, if return to the originaljob seems unlikely to the multidisciplinaryteam (including the ergonomist), the workeris rapidly referred for a vocational rehabilita-tion programme. The purpose of the promptreferral is to avoid unnecessary, harmful, andcostly delays. So, one of three courses is fol-lowed in the first six months in all cases:return to usual work, return to the usual jobtasks but improved through the ergonomicintervention, or if not possible, because theback disease is not compatible with the joband because the job tasks cannot be improvedthrough ergonomics, vocational rehabilita-tion.

VALIDATION OF THE MODELThe proposed approach advocates the inte-grated management of subacute occupationalback pain by an independent back pain clinicwith the aim of preventing chronicity. Theapproach is innovative, especially in combin-ing a clinical and ergonomic approach in asingle integrated process. It has to be testedfor its efficacy and cost effectiveness. A ran-

Figure 2 Cluster randomisation design used to compare the effectiveness of the worksite(ergonomic) intervention alone, the clinical intervention alone, or both worksite andclinical interventions combined, to standard care as provided by the worker's physician.Industrial partners have been randomised to receive the worksite intervention or not, toreduce contamination between cases for this intervention.

domised trial, including both clinical andergonomic components, has been initiated.The study population includes all the workersfrom the industries and government agencieswith more than 175 workers within a radius of30 km of the back pain clinic (located at theSherbrooke University Hospital). A clusterrandomisation of establishments and workerswas made to allow the testing of multiplehypotheses. Specifically, the value of theergonomic intervention alone, the clinicalintervention alone, and the combined inter-ventions in preventing chronicity will be evalu-ated (fig 2). Based on data from the worker'scompensation board, 40 establishments metthe inclusion criteria. Each of the 40 was firstrandomised to receive or not receive theergonomic intervention. The establishmentswere randomised according to the number ofemployees (< 500 or > 500) and the indus-trial sector (services, hospitals, manufactur-ing). This approach ensured that the twointervention groups would be comparable insize and type of production. Also, given thedifferent incidence rates of work accidents inthe various industrial sectors, this would helpin obtaining comparable rates in the twointervention groups. Each was then visited.The planned intervention was explained andproposed to the appropriate representatives ofthe management and the unions. After thesevisits, five establishments were excluded asmore detailed inquiry showed that they failedto meet the inclusion criteria (all had < 175full time employees). Of the 35 eligible part-ners, four (11%) refused to participate. Theremaining 31 establishments signed a formalagreement of participation.

Workers with low back pain of four weeksin duration are referred to the back painclinic. The purpose of the study is explained.Consenting subjects are randomised to theclinical intervention of the back pain clinic orto standard care arranged by their primarycare physician (they have no access to theback pain clinic). This design enables the sep-arate testing of hypotheses on the contribu-tion of the ergonomic intervention and on theclinical intervention. As shown in fig 2, weobtain four different groups, one with mini-mal ergonomic and clinical interventions, onewith maximal ergonomic and clinical inter-ventions, and two groups with either one ofthe two types of interventions. Group identityof participating workers is unknown to themedical evaluators. The main dependent vari-able is the duration of absence from work,which should be shortened in the group withmaximal interventions. The Kaplan-Meiermethod is used to obtain proportional curvesfor each group (survival curves). Sample sizeis estimated at 50 workers per group if the aerror is set at 0-05 and the fi error at 0-20;however, about 30 workers per group wouldprobably suffice to verify the main hypothe-SiS.40 Thus survival analysis and the Cox pro-portionate model will be used to verify themain hypothesis. No comparative analysis willbe made before the end of the trial. The studyreceived approval by the ethics committee ofthe Sherbrooke University Hospital (Centre

600

on May 28, 2020 by guest. P

rotected by copyright.http://oem

.bmj.com

/O

ccup Environ M

ed: first published as 10.1136/oem.51.9.597 on 1 S

eptember 1994. D

ownloaded from

Management of occupational back pain: the Sherbrooke model

Table Results offunctional rehabilitation therapy for the 15 workers of the pilot study

Duration Duration Durationof absence ofFRT of TRW State after

Age Type ofwork Diagnosis (weeks) (weeks) (weeks) 1 year

39 Assistant cook LBP 59 4 8 Return to previous work34* Nurse LBP 49 9 4 Return to previous work33 Assistant nurse Lumbosciatic pain 98 6 3 Return to previous work34 Assistant nurse Cervicodorsal pain 55 7 5 Return to previous work38 Welder Lumbosciatic pain 35 1 7 Return to previous work59 Metal inspector LBP 31 6 4 Return to previous work25 Mechanic LBP 22 2 0 Return to previous work48 Press operator LBP 21 5 0 Return to previous work39 Assistant nurse Lumbosciatic pain 31 3 7 Return to previous work37 Secretary Cervicodorsal pain 36 10 7 Return to previous work58 Mechanic LBP 32 5 4 Vocational rehabilitation29 Assistant nurse LBP 90 13 0 Vocational rehabilitation44 Packaging employee LBP 102 5 0 Vocational rehabilitation45 Mechanic Lumbosciatic pain 28 2 0 Surgery31 Assistant cook Lumbosciatic pain 28 3 3 SurgeryMean 48 5 340

*Only one three day recurrence. LBP = Lower back pain; FRT = functional rehabilitation therapy; TRW = therapeutic return to work.

Hospitalier Universitaire de Sherbrooke,Sherbrooke, Quebec, Canada).

RESULTS OF THE PILOT AND FEASIBILITY STUDYA trial such as that initiated required consid-erable preparation. Regional and provincialrepresentatives of employers and unions weremet to obtain their acceptance of the need forthe study. Representatives of both manage-ment and unions of each potential industrialpartner were met to explain the objectives ofthe study and the practical implications forthe companies and the workers. Knowledge ofthe project was disseminated through infor-mation meetings for practising physicians inthe study region, as well as posters, leaflets,and interviews in the local media.

Agreement with individual industrial part-ners was obtained in the summer and autumn1991, and the enrolment of workers wasstarted on September 2 1991. Thirty oneestablishments of various activities (indus-tries, hospitals, and services), private andpublic, agreed and continue to participateafter one year. By 15 September 1992,88 eligible workers had been referred to thestudy. Seventy nine (90%) agreed to partici-pate. They have been randomised to the spe-cial clinical intervention at the back pain clinicor to standard care. Three (4%) workers havedropped out. Ergonomic interventions havebeen performed in all the relevant worksites.The compliance of the industrial partners hasgenerally been excellent, with participation ofthe representatives of the management and ofthe unions. Attending physicians have beencompliant in applying the recommendationsof the multidisciplinary team of the back painclinic. The physicians consider that theirpatients have benefited from the prompt insti-tution of specialised care.

Functional rehabilitation treatments andtherapeutic return to work have been used in15 patients who have gone on to prolongedback pain (more than six months duration), toensure the quality and the feasibility of thatprocess. The results of this pilot study (beforethe beginning of the randomised study) ofcombined functional rehabilitation therapyand therapeutic return to work have been sat-isfactory (table). Hence it has been applied tothe randomised workers of the study.

Results of the efficacy study will not be

available until both enrolment and follow upare completed.

DiscussionDisability and monetary costs arising fromchronic back pain have motivated the designof our treatment model and its evaluation in arandomised trial. Based on a literature review,we believe that this model represents the firstapplication of the clinical recommendations ofthe Quebec task force report. It combines anergonomic intervention with advanced clinicalmanagement of workers with back pain for thefirst time.The inability to predict which individual

worker with low back pain will be among the5% to 7% that go on to have sustained backpain justifies the early (after four weeks ofdisability) interventions used in our study.Whereas one could have relied on theWorkers' Compensation Board to identifyworkers with back pain of four weeks dura-tion, this is not permitted under Quebec law,even for a research study. Hence, we havedeveloped consensus among both industriesand unions for the need for the proposedstudy. This additional effort required is likelyresponsible for the ongoing complete partici-pation of all the industrial partners and thehigh participation rate (90%).

Ergonomic assessment is rarely used fordetermining the timing of return to work forworkers with back pain. When an ergonomicintervention is offered, it is usually belated,when vocational rehabilitation is considered.In the model we are testing the ergonomicprocess is implemented early. Both theergonomic assessment and the application ofthe resultant recommendations take time. Apotential additional benefit is that jobsiteimprovements may improve work conditionsfor other workers doing the same job, and soprevent episodes of back pain in a more gen-eralised way.

Functional rehabilitation therapy is usuallynot offered until chronicity has developed. Inthe Sherbrooke model this intervention isused early (after three months). We know thatby three months, most of those still off workwill become chronic cases. We expect earlierintervention with functional rehabilitationtherapy will be beneficial.

Therapeutic return to work is a new con-

601

on May 28, 2020 by guest. P

rotected by copyright.http://oem

.bmj.com

/O

ccup Environ M

ed: first published as 10.1136/oem.51.9.597 on 1 S

eptember 1994. D

ownloaded from

Loisel, Durand, Abenhaim, Gosselin, Simard, Turcortte, Esdaile

cept, derived from assignment to light duties,but supervised by a clinical and ergonomicteam. Implementation of therapeutic return towork requires the close collaboration of theoccupational therapist, the worker, and theemployer.

Ergonomic and clinical interventions areclearly very different strategies of intervention.The first is aimed at improving the workplaceand the second at improving the functionalcapacities of the worker. Linking these twostrategies in the same multidisciplinary teamrepresents a systemic approach to this multi-factorial ailment. During the first year of thistrial we did not see any conflict between thesetwo interventions from the employer's or fromthe worker's points of view. When functionalrehabilitation therapy is necessary, the clinicalteam is helped by the advice of the ergono-mist.

Ergonomics and functional rehabilitationtherapy are costly interventions. The costs ofincome compensation, however, greatlyexceed the costs of the treatments for backpain.' For this reason, we anticipate that thecosts incurred by these interventions will beless than the savings obtained by having par-ticipants return to work earlier.

Acceptance of the programme by theworker and his or her practising physician,and acceptance of the case as related to theoccupation by the workers' compensationboard, are necessary to allow participation ofthe worker. Even if the case is managed by theback pain clinic, the worker's family physicianstays in charge and is continuously advised bythe multidisciplinary team. In accordancewith Quebec work regulations, it is mandatorythat the physician chosen by the worker staysin charge and has the final decision on thetreatment recommendations.

Most studies of occupational back painhave been implemented in a single industrialsettlement with specific conditions, and theresults may not be generalisable to otheremployers and workers. The present studyinvolves most of the employers with morethan 175 workers in a defined geographicarea. Implementation of the Sherbrooke inter-vention study of occupational back pain hasbeen made possible due to active collabora-tion between employers, unions, and theworkers' compensation board.

Although the final results for efficacy andcost effectiveness are not yet available, theglobal clinical and ergonomic managementprogramme has been shown to be feasible in ageneral population.

We gratefully acknowledge the careful help of Francine Lemayand Claude Alie, research assistants, and Sylvie Martineau,who typed the manuscript.

This study has been funded by a grant from the Institut deRecherche en Sante et Skcurite du Travail du Quebec(IRSST).

I Cats-Baril WL, Frymoyer JW. The economics of spinal disor-der: The adult spine-principles and practice. New York:Raven Press 1991.

2 Nachemson A. The lumbar spine, An orthopaedic chal-lenge. Spine 1976;1:59-71.

3 Nachemson A. Newest knowledge of low back pain. A crit-ical look. Clin Orthop 1992;279:8-20.

4 Spitzer WO, Leblanc FE, Dupuis M, et al. Scientificapproach to the assessment and management of activity-related spinal disorders: a monograph for clinicians.Report of the Quebec Task Force on Spinal Disorders.Spine 1987;12:59.

5 Bigos SJ, Battie M. Acute care to prevent back disability.Ten years of progress. Clin Orthop 1987;221: 121 -30.

6 Wiesel SW, Feffer HL, Rothman RH. Industrial low backpain--A prospective evaluation of a standardized diag-nostic and treatment protocol. Spine 1984;9:199.

7 Andersson GBJ, Pope MH, Frymoyer JW, Snook S.Epidemiology and cost. In: Pope MH, Frymoyer JW,Andersson GBJ, Chaffin DB, eds. Occupational low backpain: assessment, Wreatment and prevention. St Louis:Mosby year book, 1991:95- 116.

8 Spengler DM, Bigos SJ, Martin NA, Zeh J. Fisher L,Nachemson A. Back injuries in industry: a retrospectivestudy, I. Overview and cost analysis. Spine 1986;11:241-5.

9 Abenhaim L, Suissa S. Importance and economic burdenof occupational back pain: a study of 2,500 cases repre-sentative of Quebec.] Occup Med 1987;29:670-4.

10 Abenhaim L, Suissa S, Rossignol M. Risk of recurrence ofoccupational back pain over three vear follow up. Br 7Ind Med 1988;45:829-33.

11 Wood OHN, Baddley L. Epidemiology otf back pain. In:Tayson MIV, ed. The lumbar spine and back pain. Crd ed.New York: Churchill Livingstone, 1987:1 15.

12 Frvmoyer JW. Predicting disability from low back pain.Chn Orthop 1992;279:101 --9.

i3 Lancourt J. Kettelhut M. Predicting return to work torlower back pain patients receiving worker's compensa-tion. Spine 1992;17:629 40.

14 Lehmann TR, Russell DW, Spratt KF, et al. Efficacv ofelectroacupuncture and TENS in the rehabilitation ofchronic low back pain patients. Pain 1986;26:277-90.

15 Deyo RA, Walsh N, Martin D, Schoenfield L,Ramamurthy S. A controlled trial of transcutaneouselectronic nerve stimulation (TENS) and exercise forchronic low back pain. N EnglyMed 1990;322: 1627-34.

16 Jackson RP, Jacobs RR, Montesavo PX. Facet joint injec-tion in low back pain: a prospective statistical studySpine 1988;13:966-71.

17 Carette S, Marcoux S, Truchon R, Grondin C, Gagnon J,Allard Y, Latulippe M. A controlled trial of cortico-steroid injections into facet joints for chronic low backpain. N EnglJ Med 199 1;325:14.

18 Jackson RP. The facet syndrome myth or reality? (linOrthop 1992;279:110-21.

19 Frymoyer JW. Back pain and sciatica. N Engl Y Mied 1988;318:291-300.

20 Svensson HO, Andersson GBJ. Low back pain in 40 to 47year old men: work history and work environment fac-tors. Spine 1983;8:272-6.

21 Waddell G, Main CJ, Morris EW, DiPaola M, Gray 1CM.Chronic low-back pain, psychologic distress, and illnessbehaviour. Spine 1984;9:209-- 13.

22 Cady LD Jr, Thomas PC, Kawarsky RJ. Program forincreasing health and physical fitness of fire fighters. 7Occup Med 1985;27:110-4.

23 Deyo RA, Edward Bass J. Lifestyle and low back pain. Theinfluence of smoking and obesity Spine 1989;14:501-6.

24 Vallfors B. Acute, subacute and chronic low back pain:clinical symptoms, absenteeism and working environ-ment. Scand 7RehabilMed 1985;1 1 (suppl):1-98.

25 Magora A, Taustein I. An investigation of the problem ofsick leave in the patient suffering from low-back pain.Industnal Medicine and Surgery 1969;38:398-408.

26 Pope MH, Andersson GBJ. Chaffin DB. The workplace.In: Occupational low back pain: assessment, treatment andprevention: St Louis: Mosby year book, 1991:117- 1 31.

27 Nordin M, Frankel VH. Evaluation of the workplace---anintroduction. Clin Orthop 1987;221:85-8.

28 Troup JDG. Relation of lumbar spine disorders to heavymanual work and lifting. Lancet 1965;1:857-61.

29 Snook S. Approaches to the control of back pain in indus-try, lob design, job placement and education training.Occup Med 1988;3:45-59.

30 Pope MH, Hansson TH. Vibration of the spine and lowback pain. Clin Orthop 1992;279:49-59.

31 Bigos SJ, Battie MC, Fisher LD, Fordyce WE, HanssonTH, Nachemson AL, Spengler DM. A prospective studyof work perceptions and psychosocial factors affectingthe report of back injury. Spine 1990;16: -6.

32 Greenough CG, Fraser RD. The effects of compensationon recovery from low back injury. Spine 1989;14:947-55.

33 Trief P, Stein N. Pending litigation and rehabilitation out-come of chronic back pain. Arch Phys Med Rehabil1985;66:95-9.

34 Ontario Task Force on Vocational Rehabilitation: 'Aninjury to one is an injurV to all": towards dignity andindependence for the injured worker. Report submittedto the Minister of Labour, 2 September, 1987.

35 Mayer TG, Gatchel RJ, Kishino N, Keeley J, Capra P.Mayer H, et al. Objective assessment of spine functionfollowing industrial injury: a prospective study withcomparison group and one-year follow-up. Spine1985;10:482-93.

36 Bigos SJ, Spingler DM, Martin NA, et al. Back injuries inindustry: a retrospective study. III Employee relatedfactors. Spine 1986;111:252-6.

37 Straker LM. Work-associated back problems: collaborativesolutions. Journal of the Society of Occupational Medicine1990;40:75-9.

38 Costa MJ, Patry L, Kuorinka J, Baril-Gingras G.Participatory ergonomics, the motor of an action-research in health safety. In: Queinnec Y, Daniellou F,ed. Project management: field studies. Paris: Taylor andFrancis, 1991:1383-5.

39 Berwick DM, Budman S, Feldstein M. No clinical effectof back schools in an HMO a randomized prospectivetrial. Spine 1989;14:338-44.

40 Lachin JM. Introduction to sample size determination andpower analysis for clinical trials. Controlled clinical trials1981;2:93-113.

602

on May 28, 2020 by guest. P

rotected by copyright.http://oem

.bmj.com

/O

ccup Environ M

ed: first published as 10.1136/oem.51.9.597 on 1 S

eptember 1994. D

ownloaded from