Original article
Cognitive and vocational rehabilitation after mild-to-moderate traumatic brain injury: A randomised controlled trial
Silje C.R. Fure
a,b,*, Emilie Isager Howe
a,c, Nada Andelic
a,b, Cathrine Brunborg
d, Unni Sveen
a,e, Cecilie Røe
a,b,c, Per-Ola Rike
f, Alexander Olsen
g,h, Øystein Spjelkavik
i, Helene Ugelstad
j, Juan Lu
b,k, Jennie Ponsford
l, Elizabeth W. Twamley
m,n,
Torgeir Hellstrøm
a, Marianne Løvstad
e,o,paDepartmentofPhysicalMedicineandRehabilitation,OsloUniversityHospital,Oslo,Norway
bResearchCentreforHabilitationandRehabilitationModelsandServices(CHARM),InstituteofHealthandSociety,UniversityofOslo,Oslo,Norway
cInstituteofClinicalMedicine,FacultyofMedicine,UniversityofOslo,Oslo,Norway
dOsloCentreforBiostatisticsandEpidemiology,ResearchSupportServices,OsloUniversityHospital,Oslo,Norway
eFacultyofHealthSciences,OsloMetropolitanUniversity,Oslo,Norway
fDepartmentofResearch,SunnaasRehabilitationHospitalTrust,Nesoddtangen,Norway
gDepartmentofPsychology,NorwegianUniversityofTechnologyandScience,Trondheim,Norway
hDepartmentofPhysicalMedicineandRehabilitation,St.OlavsHospital,TrondheimUniversityHospital,Trondheim,Norway
iWorkResearchInstitute,OsloMetropolitanUniversity,Oslo,Norway
jDepartmentofVocationalRehabilitation,NorwegianLaborandWelfareAdministration,Oslo,Norway
kDivisionofEpidemiology,DepartmentofFamilyMedicineandPopulationHealth,VirginiaCommonwealthUniversity,Richmond,VA,USA
lMonashEpworthRehabilitationResearchCentre,TurnerInstituteforBrainandMentalHealth,SchoolofPsychologicalSciences,MonashUniversity,Clayton, Victoria,Australia
mCenterofExcellenceforStressandMentalHealth,VeteransAffairs(VA)SanDiegoHealthcareSystem,SanDiego,CA,USA
nDepartmentofPsychiatry,UniversityofCalifornia,SanDiego,LaJolla,CA,USA
oDepartmentofResearch,SunnaasRehabilitationHospitalTrust,Nesoddtangen,Norway
pDepartmentofPsychology,UniversityofOslo,Oslo,Norway
ARTICLE INFO
Articlehistory:
Received18December2020 Accepted17March2021
Keywords:
Mild-to-moderatetraumaticbraininjury Concussion
Returntowork Workstability Vocationalrehabilitation Cognitiverehabilitation
ABSTRACT
Background:Returningtoworkisoftenaprimaryrehabilitationgoalaftertraumaticbraininjury(TBI).
However,theevidencebasefortreatmentoptionsregardingreturntowork(RTW)andstablework maintenanceremainsscarce.
Objective:Thisstudyaimedtoexaminetheeffectofacombinedcognitiveandvocationalinterventionon work-relatedoutcomesaftermild-to-moderateTBI.
Methods:Inthisstudy,wecompared6monthsofacombinedcompensatorycognitivetrainingandsupported employment(CCT-SE)interventionwith6monthsoftreatmentasusual(TAU)inarandomisedcontrolledtrial toexaminetheeffectontimetoRTW,workpercentage,hoursworkedperweekandworkstability.Eligible patientswerethosewithmild-to-moderateTBIwhowereemployed50%atthetimeofinjury,18to60years oldandsick-listed50%at8to12weeksafterinjuryduetopost-concussionsymptoms,assessedbythe RivermeadPostConcussionSymptomsQuestionnaire.BothtreatmentswereprovidedattheoutpatientTBI departmentatOsloUniversityHospital,andfollow-upswereconductedat3,6and12monthsafterinclusion.
Results:Weincluded116individuals,60randomisedtoCCT-SEand56toTAU.Thegroupsdidnotdiffer incharacteristics atthe 12-monthfollow-up. Overall,a highproportion hadreturned to workat 12months(CCT-SE,90%;TAU,84%,P=0.40),andallexcept3werestablyemployedaftertheRTW.
However,asignificantlyhigherproportionofparticipantsintheCCT-SEthanTAUgrouphadreturnedto stableemploymentat3months(81%vs.60%,P=0.02).
Conclusion:TheseresultssuggestthattheCCT-SEinterventionmighthelppatientswithmild-to-moderate TBIwhoarestillsick-listed8to12weeksafterinjuryinanearlierreturntostableemployment.However, theresultsshouldbereplicatedandacost-benefitanalysisperformedbeforeconcluding.
C2021TheAuthor(s).PublishedbyElsevierMassonSAS.ThisisanopenaccessarticleundertheCCBY license(http://creativecommons.org/licenses/by/4.0/).
* Correspondingauthorat:DepartmentofPhysicalMedicineandRehabilitation,OsloUniversityHospital,0407Oslo,Norway.
E-mailaddress:siljfu@ous-hf.no(S.C.R.Fure).
Availableonlineat
ScienceDirect
www.sciencedirect.com
https://doi.org/10.1016/j.rehab.2021.101538
1877-0657/C2021TheAuthor(s).PublishedbyElsevierMassonSAS.ThisisanopenaccessarticleundertheCCBYlicense(http://creativecommons.org/licenses/by/4.0/).
1. Introduction
Approximately50millionpeoplegloballysustainatraumatic braininjury(TBI)eachyear[1].Ofthese,roughly90%areclassified as mild TBI (mTBI) [2], with about 15% to 20% experiencing symptoms for morethan 3 months[3].Physical, cognitiveand emotionalsymptomsaffectpatients,theirfamiliesandtheirability toremaincompetitivelyemployed[4,5].
Anestimated18%to60%ofpatientshadareturntowork(RTW) [6] after TBI. The vast variability of RTW rates is due to the inclusionofdifferentTBIseverities,follow-uptimes,samplesizes anddefinitionsofRTW.OnestudyofpatientswithmTBIwithout structuredrehabilitationreportedanRTWrateof62%at1year after injury [7].Several factors complicatetheprocess ofRTW.
Some of the factors most often assessed are post-concussion symptoms, demographic factors,pre-injury occupational status, previouspsychiatrichistoryandinjuryseverity[4,5].
ThesamefactorsthatcomplicateRTWafterTBImayalsoaffect theabilitytoretainastableworkattachment.Concerningwork stability,studiesincludingindividualswithintracranialinjuriesof allseveritylevelshavereportedthat34%to55%foundstablework afterTBI[8,9].WithfewstudiesreportingworkstabilityafterTBI andaninconsistentmethodofdefiningworkstability,thereisa definitelackofdataconcerningworkstability,particularlyafter mild-to-moderateTBI.
A systematic review from 2016 found strong evidence supporting work-directed interventions combined with educa- tion/coachingforimprovingRTWoutcomesafteracquiredbrain injury [10]. Other systematic reviews examined the effect of cognitive rehabilitation on RTW after TBI; one supported the treatment methods, withparticular emphasis on compensatory strategies[4],butothersfoundnoevidenceofeffect[11,12].
The diverging results concerning the effect of cognitive rehabilitationonRTWhasledtoanincreasedfocusonvocational rehabilitationinterventionsprovidedattheworkplace.In2015,a multicentre,randomisedcontrolledtrial(RCT)of1193participants foundthatwork-focusedcognitivebehaviouraltherapycombined withindividualjobsupportimprovedRTWproportionstosome extentinpatientswithcommonmentaldisorders[13].Likewise, there is some preliminary evidence supporting the use of supportedemploymentinvocationalrehabilitationafterTBI[14].
In2015,Twamleyetal.[15]published1-yearfollow-upresults fromapilotRCTcombiningcompensatorycognitiverehabilitation and supportedemployment inveteranswithahistory ofmTBI.
They observed no group differences in the attainment of competitive workbut someimprovement regarding symptoms andqualityoflife.Theseresultsrequirereplicationinlarger-scale studiesusingaciviliansample.
ThecurrentstudyincorporatedthisknowledgeinanRCTusing a combined cognitiveandvocational intervention toassessthe effect on RTW and work stability in patients after mild-to- moderateTBI. The3-and6-monthinterimresultsofthisstudy havebeenpublished[16].Wehypothesizedthattheintervention wouldresultinahigherproportionofpatientsreturningtostable competitiveemploymentbythe12-monthfollow-upinaddition tohavingahigherworkpercentageandmoreworkhoursperweek ascomparedwiththecontrolgroup.
2. Methods
2.1. Studydesign
ThestudyisaprospectiveRCT.Eligiblepatientswererecruited from a specialised TBI-rehabilitation outpatient clinic at Oslo University Hospital and were randomised to the combined
compensatory cognitive training and supported employment (CCT-SE) intervention or treatment as usual (TAU). The results ofafeasibilitystudyhavebeenpublishedpreviously[17].Physia- tristsattheDepartmentofPhysicalMedicineandRehabilitationof thehospitalinformed eligiblepatients aboutthestudy, and all participantsprovidedwrittenconsent.Baselineassessmentswere performed8to12weeksafterinjury,withfollow-upassessments at3,6and12monthsafterinclusion.TheRegionalCommitteefor MedicalandHealthEthicsinSouth-EastNorwayapprovedthetrial (2016/2038),andtheprotocolwasregisteredatClinicalTrials.gov (NCT03092713)[18].ThisstudyfollowstheCONSORTstatement [19]andtheethicalprinciplesoftheHelsinkideclaration.
2.2. Participants
Eligibleparticipantshadsustainedamild-to-moderateTBI8to 12weekspreviously,livedinOsloorAkershuscounty(approxi- mately 1.3 million inhabitants; one-fourth of the Norwegian population),wereofworkingage(18–60years),wereemployed 50%atthetimeofinjuryandweresick-listed50%atinclusion duetopost-concussionsymptomsassessedwiththeRivermead PostConcussionSymptomsQuestionnaire(RPQ)[20].Theywere deemed as having post-concussion symptoms if at least one symptomwasratedas2.MeantotalRPQscoreatbaselinewas 28(range5-54).ClassificationofmTBIinvolvedusingtheAmerican CongressofRehabilitationMedicinecriteria[21].Mild-to-moder- ateTBIwasdefinedasGlasgowComaScale([22])10–15,lossof consciousness<24 hr and post-traumatic amnesia<7 days.
Exclusion criteria included progressive neurological disease, ongoing substance abuse and/or inability to speak or write Norwegian.
2.3. Interventions 2.3.1. CCT-SE
Participantsintheinterventiongroupreceivedacombination of compensatory cognitive training (CCT [23]) and supported employment(SE[24]).CCTisa10-week,group-based,manualized interventionwithweeklysessionsof2hrprovidedbya clinical psychologistand a physician. CCTaimedat teaching theparti- cipantscompensatorystrategiestohelpmanagepost-concussion symptoms,specificallyfocusingonstrategiestoalleviatecognitive symptoms. Topics in the sessions included headache, fatigue, difficulties with sleep, concentration, memory and executive function.
The vocational partof the intervention wasbased on SE in which a ‘‘place-and-train’’methodis adapted [25]: participants weresupportedbyanemploymentspecialistinreturningtotheir currentjobsby workingattheiractual,competitive,workplace.
This part of the intervention was delivered individually, for a maximumof6monthsperparticipant,andadministeredbythe DepartmentofVocationalRehabilitation,NorwegianLabourand WelfareAdministration.Theemploymentspecialistsattendedall sessionsofoneCCTgrouptoimprovetheintegrationofconcepts fromtheCCTintotheRTWprocess.
Monthlymeetings wereheldduring theintervention period and were attended by the CCT interventionists, employment specialistsand atleastonesenior researchertoensureoptimal trans-sectoral collaboration and a shared understanding of the individualparticipants.
2.3.2. TAU
ThecontrolgroupreceivedTAUforamaximumof6months afterinclusion.AtOsloUniversityHospital,TAUentailstreatment and follow-up froma specialised multidisciplinaryTBI teamat the TBI outpatient department. The participants received a
consultationwithaphysiatristandwerereferredtoaphysiatrist, physicaltherapist,occupationaltherapist,neuropsychologistora social worker as required.Some participantswere alsooffered participation in aneducational groupthat focusedon common problemsafterTBIandlastedfor2hrperweekfor4weeks.
A detailed description of interventions in both treatment groupsisinthestudyprotocol[18].
2.4. Studyoutcomes
Theprimaryoutcomewasself-reportedworkparticipationat the12-monthfollow-upmeasuredbytheproportionofpatients whohadreturnedtowork(0–100%).Furthermore,workinghours
perweek(0–37.5hr),workpercentage(0–100%),workstability andself-reported timefrominjurytoreturntopre-injurywork levels(in days)were secondaryoutcomes. Datawerecollected duringappointmentsattheTBIoutpatientclinicatinclusionand follow-upat3,6and12monthsafterinclusion.Workinghoursper weekwerecalculatedfromworkpercentage[(workpercentage* 37.5)/100].Tooperationaliseworkstability,eachparticipantwas assigneda workcategoryateach follow-updependingontheir currentworkpercentage(0%,50%,50–79%or80–100%).Patients whomovedtoalowerworkcategoryfromanyfollow-uptothe next were classified as ‘‘unstably employed’’. Patients who maintained or improved their level of workparticipation were classifiedas‘‘stablyemployed’’.
Fig.1.Flowchartofinclusionandfollow-up.CCT-SE,compensatorycognitivetrainingandsupportedemployment;GCS,GlasgowComaScale;TAU,treatmentasusual.
2.5. Samplesize
The samplesize wascalculated basedon theproportions of RTW,aimingfora33%absolutedifferenceinRTWstatusbetween the 2 treatment groups at the 12-monthfollow-up [18]. From studiesofoccupationalhealthcareonRTW,weassumedthatan oddsratioof2.0wasthesmallestclinicalandsocietalrelevantratio [26]. This indicates that participantsin theintervention group returnedtoworktwiceasquicklyasparticipantsin thecontrol group.Assumingthattwo-thirdsoftheparticipantswouldachieve RTW during the follow-up, the sample size calculated with
G*Power resulted in 110 patients, with 55 patients in each treatmentgroup(
a
=0.05,powerlevel80%).Withanexpectedloss tofollow-upof15%,125participantswererequired.2.6. Randomisationandblinding
Allincludedpatientswererandomisedina1:1ratiotooneof the2treatmentgroupsafterbaselineassessment.Anindependent statistician produced a computer-generated permuted block sequencewithrandomisedblocksizes(2,4,6or8)beforestarting inclusion.Theresearcherwhowasresponsiblefortheallocationof patients to the treatment groups was not involved in patient recruitmentor assessment. Outcomeassessors wereblinded to patient allocation. Blinding of rehabilitation specialists and patientswasnotpossible.
2.7. Statisticalanalysis
Data analyses were performed with Stata 16. Descriptive methods were used to describe baseline and injury-related characteristics. A mixed-effects logistic regression was applied toevaluatetheproportionofparticipantswho hadreturned to work.Linearmixed-effectsmodelswerefittedtoanalyseworking hoursperweekandworkpercentagebetweengroupsandwithin groups.Timeandtime-by-treatmentinteractionwerefixedeffects inallmodels,allowingarandominterceptandrandomeffectof time.Themaineffectoftreatmentwasremovedfromthemodels toadjustforpotentialbaselinedifferences.Differencesbetween the groups in days to returning to pre-injury work levels wereanalysed byKaplan–Meiercurvesand alog-ranktest. The Kaplan–Meiercurveswereadjustedforthepossibleconfounding effectofthepresenceoftraumaticintracranialinjuryonCT/MRIor Table1
Baselinecharacteristicsofindividualswithmild-to-moderatetraumaticbraininjuryat8to12weekspost-injurybystudygroupandforthetotalsample.
n CCT-SE(n=60) TAU(n=56) Totalsample(n=116)
Sociodemographicfactors
Age,years,median(range) 60/56 42(24-60) 44(27-60) 43(24-60)
Sex,female 60/56 33(55) 36(64) 69(59)
Education,years,mean(SD) 60/56 16(2) 16(3) 16(3)
Married/cohabitating 60/56 43(72) 34(61) 77(66)
Injury-relatedfactors
Causeofinjury 60/56
Falls 19(31) 30(54) 49(42)
Trafficaccidents 12(20) 11(20) 23(20)
Sports 10(17) 4(7) 14(12)
Violence 3(5) 3(5) 6(5)
Exposuretoinanimateobjects 15(25) 8(14) 23(20)
Unknown 1(2) 0(0) 1(1)
CT/MRIfindings,traumaticintracranial 60/56 11(18) 16(29) 27(23)
InjuryseveritybyACRMcriteria 60/56
Mild 58(97) 51(91) 109(94)
Moderate 2(3) 5(9) 7(6)
Lossofconsciousness(LOC) 60/56
<30min 21(35) 16(29) 37(32)
30min–24hr 1(1) 2(4) 3(3)
NoLOC 31(52) 30(53) 61(52)
Notregistered 7(12) 8(14) 15(13)
Post-traumaticamnesia(PTA) 60/56
<1hr 18(30) 17(31) 35(30)
1–24hr 7(12) 9(16) 16(14)
25hours–7days 0(0) 2(4) 2(2)
NoPTA 25(42) 26(47) 51(44)
Notregistered 10(16) 2(2) 12(10)
Work-relatedfactors
Occupation,whitecollar 60/56 53(88) 50(89) 103(89)
Permanentposition 60/56 56(93) 49(88) 105(91)
Full-timeposition 60/56 55(92) 48(86) 103(89)
Privatesector 60/56 36(60) 28(50) 64(55)
Durationofemployment,months,median(range) 59/55 54(0-408) 42(0-480) 51(0-480)
Dataaren(%)unlessotherwiseindicated.CCT-SE:compensatorycognitivetrainingandsupportedemployment;TAU:treatmentasusual.
Fig.2.Proportionofpatientswhoreturnedtoworkaftermild-to-moderateTBI.
CCT-SE, compensatory cognitive training and supported employment; TAU, treatmentasusual.
whethertheparticipantswereworkingatbaseline.Analyseswere carried out on an intention-to-treat basis by an independent statistician who was blinded to group allocation. The level of significancewassetatP<0.05.
3. Results
Becauseoflower-than-expectedlosstofollow-up(6%)andthe timelimitofthestudy,thestudyinclusion,whichbeganinJuly 2017,wasterminatedinApril2019afterinclusionof121patients (Fig.1). Treatmentin both groupswas performed fromAugust 2017toNovember2019.Fivepatientswhoinitiallyconsentedto participatewithdrewtheirconsentbeforerandomisation.Conse- quently,116participantswereincludedintheanalyses,60ran- domisedtotheCCT-SEgroupand56toTAU.Participantsinthe CCT-SEgroupwereincludedatamean(SD)of77(3)daysafter injury, and those in theTAU groupat68 (3) days after injury.
AdherencetotheCCTinterventionwashigh.Threepatientswere absent from a total of 6 sessions, which resulted in a 99%
attendancerateforthegroup[16].
Manyincludedpatientswerefemale(59%),mostwerehighly educated,andmosthadanmTBI(94%)(Table1).Thegroupsdid not differ in baseline characteristics. A more comprehensive descriptionofbaselinecharacteristicsisreportedelsewhere,anda detailed descriptionofthetreatmentreceivedinboth groupsis reportedinotherpublicationsfromtheproject[16–18].
3.1. Proportionofpatientsreturningtowork
Theproportionofpatientsreturningtoworkat3monthswas higherin theCCT-SEthanTAUgroup(mean81%vs.60%,mean between-group difference from baseline to3 months 14%, 95%
confidenceinterval[CI]5;32,P=0.02)(Fig.2).Thecontrolgroup hadcaughtupwiththetreatmentgroupbythe6-and12-month follow-ups, and the mean between-group differences were no longersignificant( 9%and 6%).Inlinewiththefindingthatthe RTWprocessmainlyoccurredwithinthefirst3monthsintheCCT- SE group,thewithin-groupdifferencewassignificantonlyfrom baselineto3monthsforthisgroupbutwassignificantbetweenall time points for the TAU group (see Table 2 for between- and within-groupdifferences).
3.2. Workinghoursperweekandworkpercentage
Linear mixed-effects models showed that the number of working hours per week and work percentage increased over timebutdidnotsignificantlydifferbetweengroups.
3.3. Daysuntilpre-injuryworklevels
Overall,39(65%)participantsintheCCT-SEgroupand30(54%) in theTAUgroupreturnedtopre-injuryworklevelsduringthe studyperiod.Halfofthepatientswerebacktopre-injurylevels within365daysafterinjuryintheCCT-SEgroupandby415daysin the TAU group. The50-day difference wasnot significant. The presenceoftraumaticintracranialabnormalitiesconfoundedthe associationbetweentreatmentgroupsanddaysbefore reaching pre-injuryworklevelsandwasadjustedfor(Fig.3).Adjustmentfor whetherthepatientswereworkingatbaselinedidnotaffectthis association(datanotshown).
3.4. Workstability
Only 3 participants showed decreased work percentage
category from 6 to 12 months, so most patients had stably Table2 Proportionworking,workpercentagesandhoursworkedperweek.Resultsfrommixed-modelanalyses. Baseline3months6months12monthsWithin-group difference 0–3months Between-group difference 0–3months Within-group difference 3–6months Between-group difference 3–6months Within-group difference 6–12months
Between-group difference 6–12months Proportionworking CTT-SE38(25;52)81(71;92)85(75;95)90(82;98)43(29;57) P<0.00114(-5;32) P=0.022
4(-6;14) P=0.434-9(-24;7) P=0.164
5(-3;13) P=0.236-6(-20;7) P=0.397TAU31(20;41)60(48;73)73(61;84)84(72;96)29(18;42) P<0.00113(1;24) P<0.00111(0;22) P<0.001 Workpercentage CTT-SE12(6;18)34(27;40)51(43;59)77(67;88)22(14;29) P<0.0012(-9;12) P=0.474
17(10;24) P<0.0010(-11;11) P=0.518
26(19;35) P<0.0016(-5;18) P=0.194TAU10(4;16)30(23;37)47(39;55)67(56;78)20(13;28), P<0.00117(9;24) P<0.00120(12;29) P<0.001 Hoursworkedperweek CTT-SE4.5(2.3;6.8)12.6(10.0;15.2)18.9(15.9;21.9)28.9(25.0;32.8)8.0(5.3;10.8) P<0.0010.5(-3.4;4.5) P=0.474
6.4(3.6;9.1) P<0.0010.1(-3.9;4.1) P=0.518 10.0(7.0;12.9) P<0.0012.3(-2.0;6.6) P=0.194TAU3.7(1.3;6.0)11.2(8.6;13.9)17.5(14.4;20.6)25.1(21.1;29.2)7.5(4.7;10.4) P<0.0016.3(3.4;9.2) P<0.0017.6(4.5;10.7) P<0.001 Dataaremean(95%confidenceinterval).CCT-SE:compensatorycognitivetrainingandsupportedemployment;TAU:treatmentasusual.
returnedtowork,regardlessofgroup.Frombaselineto3-month follow-up, 4 patients (CCT-SE: n=2,TAU: n=2)wereunstably employed,andfrom3to6-monthfollow-up,6(CCT-SE:n=2,TAU:
n=4) were unstably employed. The group difference was not significant.
4. Discussion
Thisstudyexamined theeffectivenessofa pragmatic,cross- sectoralandinnovativecomplexintervention(CCT-SE)onRTWin patientswithmild-to-moderateTBIwhowerestillsymptomatic and onsick leave8to12weeksafterinjury.In contrasttoour hypotheses,wefoundnodifferencesinworkoutcomesbetween theCCT-SEandTAUgroupsatthe12-monthfollow-up.However,a significantlyhigherproportionoftheCCT-SEgrouphadreturnedto work after3 monthsas compared with theTAU group, which suggests an early effect of the CCT-SE intervention on return to competitive work after mild-to-moderate TBI. We foundno significantgroupdifferencesintheotherwork-relatedoutcomes.
However,themediandifferenceintimefrominjurytoreturnto pre-injury worklevelswas50 days,whichsupportsaccelerated RTWintheCCT-SEgroup.Overall,thewithin-groupdifferences showed an improvement in all outcomesover time, and most patientsinbothgroupswerestablyemployedafteraninitialRTW.
Returning toworkisa primary rehabilitationgoalafter TBI.
Vocational rehabilitation may be challenging because of the heterogeneityofhealth-relatedTBIconsequencesandpre-morbid andcontextualfactors(i.e.,personalandenvironmentalfactors).
The literaturehassuggested focusingon both health and work factors,theinvolvementofthepatientandemployer,acombina- tion ofwork-directedinterventions[10],and theintegrationof thesefactorsintoearlyrehabilitationafterTBI[27].
Atthestudyplanningtime,evidencewaslackingtosupportthe effectivenessofvocationalrehabilitationforpeoplewithmild-to- moderateTBI[5,28].Anovelapproachtovocationalrehabilitation, the ‘‘place-and-train’’ principles, involving SE, gained empirical
supportintheNorwegiancontext,withpositiveresultsforboth work-and non-work-relatedoutcomes for people withmental illness[25].ThepresentstudywasfurtherinspiredbyTwamley et al. [15], who conducted a pilot study using the original CogSMARTinterventioncombinedwithSEinveteranswithmild- to-moderate TBI. The authors found improvement in affective post-concussionsymptomsandqualityoflifebutno significant improvementinRTW.ThepresentstudyandTwamleyetal.[15]
useddifferentinclusioncriteria,suchastimesinceinjury(8–12 weeksvs.>4years),toolsusedtodetermineimpairment(theRPQ inthecurrentsamplevs.neuropsychologicalperformanceinthe pilot study) and duration of SE support (6 vs. 12 months).
Furthermore,oursampleusedthecriterionofemploymentatthe time of injury, whereas participants in the pilot study were unemployedbutweremotivatedtoreturntowork.Additionally, the sample in the current study was civilian; our study was conductedwithina differentgovernmental welfaresystemand includedmorethantwiceasmanyparticipants.
Ofnote,TAUinthisstudywasrelativelycomprehensive.Vikane etal.[29]assessedtheeffectoftheprogramconstitutingTAUin thecurrentstudycomparedtofollow-upbyageneralpractitioner forpatientsatriskorsick-listedwithpost-concussionsymptoms at2monthsaftermTBI.Thegroupreceivingfollow-upcarebya generalpractitioneralsohadamultidisciplinaryexaminationwith subsequentadvice.TheauthorsfoundthatparticipantsintheTAU program showeddecreased symptom burden on theRPQ after 1year,butthegroupsdidnotdifferindaystosustainableRTW,so TAUwasnoteffectiveforRTW.
However, the results of the 2 studies are not directly comparable. In the current study, TAU constituted the control group.Furthermore,thedifferencesininclusioncriteriabetween thestudieshampercomparisons,suchasdifferentseveritylevels (mild-to-moderateTBIvs.mTBI),timeofinclusion(8–12vs.6–8 weeksafterinjury),ageofsample(18–60vs. 18–55years)and whetherthepatients had beenhospitalised(not necessarilyvs.
5 hr).Additionally,Vikaneetal. usedadifferentdefinitionof stableRTWthanthecurrentstudyandcollectedsickleavedata Fig.3.Daystoreachpre-injuryworklevelbytreatmentgroup:unadjustedandadjustedforthepresenceoftraumaticintracranialinjuryonCT/MRI.CCT-SE,compensatory cognitivetrainingandsupportedemployment;TAU,treatmentasusual.
froma nationalregistry [29].Considering thesedifferences,the high level of care received in TAU, with highattendance rate andlowlosstofollow-up,mightstillhaveaffectedtheresultsand reducedthedifferencebetweenthe2treatmentgroups.Usinga control group receiving a less-comprehensive follow-up might haveresultedinalargerprimary-outcomedifferencebetweenthe groups. A qualitative processevaluation that explores patients’
experiencewiththeRTWprocesswillbepublished,inadditionto theevaluationofclinicaloutcomes.
Overallratesofreturntocompetitiveemployment(partorfull time)at12monthswerehighinbothCCT-SEandTAUgroups(90%
and84%).Thisfindingmaybeexplainedinpartbythecontextof the study, in addition to expected spontaneous recovery. The Norwegian welfare systemincludes measuresto ensure a low unemploymentrate,inadditiontouniversallyaccessible,afford- able and high-quality health care services. Furthermore, all patientsinthisstudywereemployed50%atthetimeofinjury, which increases their likelihoodof regaining employmentafter injury as compared withunemployed patients [5]. Conversely, only65%ofpatientsintheCCT-SEgroupand54%intheTAUgroup hadreturnedtotheirpre-injuryworklevelat12months’follow- up. The Norwegian welfare system also includes a generous workers’compensationprogramthatcovers100%oflostincome for thefirst yearof sick-listing and approximately66% beyond the first year. The Organization for Economic Cooperation and Development has previously revealed that, of its member countries,Norwayhasthehighestlevelofsick-listingsandcosts relatedtolostlabour[13].Reimbursementforthelossofincome whensick-listed(i.e.,upto12months)mighthaveaffectedthe patients’motivationtoreturnquicklytofull-timelabour[5]and may,ingeneral,hampertheefficacyofwork-relatedinterventions.
4.1. Strengthsandlimitations
The current study is a well-designed, innovative and cross- sectoralRCTexaminingRTWinaspecificsubsampleofTBIpatients with persistent symptoms. The risk of bias was minimised by thelowlosstofollow-up[30].Becauseoftheciviliansample,the resultsaremoregeneralisablethanthoseobtainedfromasample ofmilitaryveterans.Thestudywasconductedin thecontextof generousincomecompensationduringsickleave,thuspotentially decreasing its generalisability to countries with other welfare systems because this may influence motivation for RTW and consequently RTWrates [5,31].The generalisabilityshouldalso beconsideredinlightofthecomprehensivemultidisciplinarycare receivedinTAU,whichisnotrepresentativeofthestandardofcare receivedatmostothernationalorinternationalfacilities.
Atypically for the general TBI population, the sample was predominantlywomen,inwhitecollaroccupations,andfull-time employees. However, this sample represents the patients after mild-to-moderateTBIwithprolongedsymptomswhoareseeking treatment and residein an urban area. Datafrom the Quality RegistryattheTBIoutpatientclinicshowthat10%morefemale than male patients are referred for multidisciplinary follow-up (personalcommunicationwithQualityRegistrystaff).
Themainoutcomeswerebasedon self-reporteddata,which couldbeconsideredalimitationifrespondentsreportfalsevalues ordonotremembercorrectly.However,thestudyparticipantshad sustainedmild-to-moderateTBI,andtheirknowledgeofpersonal work-relateddatawasnotsuspectedtobenotablyaffected.
Theoriginalsamplesizecalculationwasdesignedtodetecta 33% absolute difference between the treatment groups in the proportionthatreturnedtoworkatthe12-monthfollow-up.In previouswork-relatedinterventionstudiesonoccupationalback pain andmental disord,both30% and20% differencesbetween groupsinRTWwereused[26,32].Inourstudy,wefoundonlya6%
differencebetweengroupsatthe12-monthfollow-up.Thisfinding couldberelatedtothepragmaticcontextofthestudy(inclusionof themultidisciplinaryfollow-upastheTAUgroup)andthenatural recoveryprocess ofmild-to-moderateTBI.However, the50-day mediandifferenceintimefrominjurytoreturntopre-injurywork levelsmightindicateanimportanteffectoftheCCT-SEinterven- tion.Thisfindingwillbeexploredfurtherinastudyonthecost- effectivenessofthisintervention.
5. Conclusions
The study results suggest that the combined cognitive and vocational intervention improved the early return to stable employment inpatients withmild-to-moderate TBI. Expediting astableRTWmaysubstantiallyreducecostsrelatedtolostlabour aftermild-to-moderateTBI,inadditiontohelpingpatientsreturn totheirpre-injurylevelsoffunctioning.Theresultsofthisstudy require replication, and a cost-benefit analysis should be performedbeforedrawingafirmconclusion.
Funding
ThisworkwassupportedbytheResearchCouncilofNorway(256689/
H10),whichhadnoroleinthestudydesign;inthecollection,analysisand interpretationofdata;inthewritingofthereport;orinthedecisionto submitthearticleforpublication.
Disclosureofinterest
Theauthorsdeclarethattheyhavenocompetinginterest.
Acknowledgements
Wethankthepatientsfortheirparticipationandacknowledge thehelpprovidedbyKnut-PetterS.Langloandresearchassistants whoaidedinconductingfollow-upsandenteringdata.
References
[1]MaasAIR,MenonDK,AdelsonPD,AndelicN,BellMJ,BelliA,etal.Traumatic braininjury:integratedapproachestoimproveprevention,clinicalcare,and research. LancetNeurol 2017;16(12):987–1048. http://dx.doi.org/10.1016/
s1474-4422(17)30371–x.
[2]CassidyJD,CarrollLJ,PelosoPM,BorgJ,vonHolstH,HolmL,etal.Incidence, riskfactorsandpreventionofmildtraumaticbraininjury:resultsoftheWHO CollaboratingCentreTaskForceonMildTraumaticBrainInjury.Journalof rehabilitationmedicine2004;(43Suppl):28–60.
[3]CancelliereC,KristmanVL,CassidyJD,HincapieCA,CoteP,BoyleE,etal.
Systematicreviewofreturntoworkaftermildtraumaticbraininjury:results oftheInternationalCollaborationonMildTraumaticBrainInjuryPrognosis.
ArchPhysMedRehabil2014;95(3Suppl):S201–9.http://dx.doi.org/10.1016/
j.apmr.2013.10.010.
[4]ManiK,CaterB,HudlikarA.Cognitionandreturntoworkaftermild/moderate traumaticbraininjury:Asystematicreview.Work2017;58(1):51–62.http://
dx.doi.org/10.3233/wor–172597.
[5]ShamesJ,TregerI,RingH,GiaquintoS.Returntoworkfollowingtraumatic braininjury:trendsandchallenges.DisabilRehabil2007;29(17):1387–95.
http://dx.doi.org/10.1080/09638280701315011.
[6]GormleyM,DevanaboyinaM,AndelicN,RøeC, SeelRT,LuJ.Long-term employmentoutcomesfollowingmoderatetoseveretraumaticbraininjury:
asystematicreviewandmeta-analysis.BrainInj2019;33(13–14):1567–80.
http://dx.doi.org/10.1080/02699052.2019.1658222.
[7]RøeC,SveenU,Alvsa˚kerK,Bautz–HolterE.Post-concussionsymptomsafter mildtraumaticbraininjury:influenceofdemographic factorsandinjury severity ina 1–year cohortstudy. DisabilRehabil 2009;31(15):1235–43.
http://dx.doi.org/10.1080/09638280802532720.
[8]KreutzerJS,MarwitzJH,Walker W,SanderA,ShererM,BognerJ, etal.
Moderatingfactorsinreturntoworkandjobstabilityaftertraumaticbrain injury.JHeadTraumaRehabil2003;18(2):128–38.http://dx.doi.org/10.1097/
00001199-200303000-00004.
[9]Dillahunt–AspillagaC,PughMJ,CotnerBA,SilvaMA,HaskinA,TangX,etal.
EmploymentStabilityinVeteransandServiceMembersWithTraumaticBrain Injury:AVeteransAdministration TraumaticBrainInjuryModelSystems Study. Arch Phys Med Rehabil 2018;99(2s):S23–32. http://dx.doi.org/
10.1016/j.apmr.2017.05.012.
[10]Donker–CoolsBH,DaamsJG,WindH,Frings-DresenMH.Effectivereturn–to–
workinterventionsafteracquiredbraininjury:Asystematicreview.BrainInj 2016;30(2):113–31.http://dx.doi.org/10.3109/02699052.2015.1090014.
[11]ThomasRE,AlvesJ,VaskaMlisMM,MagalhaesR.Therapyandrehabilitationof mild braininjury/concussion: Systematicreview.Restor Neurol Neurosci 2017;35(6):643–66.http://dx.doi.org/10.3233/rnn–170761.
[12]KumarKS,SamuelkamaleshkumarS,ViswanathanA,MacadenAS.Cognitive rehabilitationforadultswithtraumaticbraininjurytoimproveoccupational outcomes.CochraneDatabaseSystRev2017;6(6).http://dx.doi.org/10.1002/
14651858.CD007935.pub2[Cd007935].
[13]RemeSE,GrasdalAL,LøvvikC,LieSA,ØverlandS.Work-focusedcognitive- behaviouraltherapyandindividualjobsupporttoincreaseworkparticipation in commonmental disorders: arandomised controlledmulticentre trial.
OccupEnvironMed2015;72(10):745–52.http://dx.doi.org/10.1136/oemed–
2014-102700.
[14]FadylJK,McPhersonKM.Approachestovocationalrehabilitationaftertrau- matic brain injury: a review of the evidence. J Head Trauma Rehabil 2009;24(3):195–212.http://dx.doi.org/10.1097/HTR.0b013e3181a0d458.
[15]TwamleyEW,ThomasKR,GregoryAM,JakAJ,BondiMW,DelisDC,etal.
CogSMART Compensatory Cognitive Training forTraumatic Brain Injury:
Effects Over1Year.J HeadTraumaRehabil 2015;30(6):391–401. http://
dx.doi.org/10.1097/htr.0000000000000076.
[16]HoweEI,FureSCR,LøvstadM,EnehaugH,SagstadK,HellstrømT,etal.Effec- tivenessofCombiningCompensatoryCognitiveTrainingandVocationalInter- ventionvs.TreatmentasUsualonReturntoWorkFollowingMild-to-Moderate TraumaticBrainInjury:InterimAnalysisat3and6MonthFollow-Up.Frontiersin Neurology2020;11(1414).http://dx.doi.org/10.3389/fneur.2020.561400.
[17]HoweEI,LøvstadM,LangloK–PS,HellstrømT,SpjelkavikØ,UgelstadH,etal.
Feasibilityofacognitiverehabilitationprogramforindividualswithmild–to–
moderate traumaticbrain injury:Participants’engagement andsatisfaction.
CogentMedicine2019;6(1).http://dx.doi.org/10.1080/2331205X.2019.1565614 [1565614].
[18]HoweEI,LangloKS,TerjesenHCA, RoeC,SchankeAK,Soberg HL,etal.
Combinedcognitive andvocational interventionsafter mildto moderate traumaticbraininjury:studyprotocolforarandomizedcontrolledtrial.Trials 2017;18(1):483.http://dx.doi.org/10.1186/s13063-017-2218-7.
[19]SchulzKF,AltmanDG,MoherD.CONSORT2010Statement:updatedguide- linesforreportingparallelgrouprandomisedtrials.BMJ2010;340.http://
dx.doi.org/10.1136/bmj.c332[c332].
[20]KingNS,CrawfordS,WendenFJ,MossNE,WadeDT.TheRivermeadPost ConcussionSymptomsQuestionnaire:ameasureofsymptomscommonly experiencedafterheadinjuryanditsreliability.JNeurol1995;242(9):587–
92.http://dx.doi.org/10.1007/bf00868811.
[21]ACRMMildTraumaticBrainInjuryCommitee.Definitionofmildtraumatic braininjury.JHeadTraumaRehabil1993;8(3):86–7.
[22]TeasdaleG,JennettB.Assessmentofcomaandimpairedconsciousness.A practicalscale.Lancet1974;2(7872):81–4.http://dx.doi.org/10.1016/s0140- 6736(74)91639-0.
[23]StorzbachD,TwamleyEW,RoostMS,GolshanS,WilliamsRM,O&NeilM,etal.
CompensatoryCognitiveTrainingforOperationEnduringFreedom/Operation IraqiFreedom/OperationNewDawnVeteransWithMildTraumaticBrain Injury.JHeadTraumaRehabil2017;32(1):16–24.http://dx.doi.org/10.1097/
htr.0000000000000228.
[24]Becker,DeborahR,Drake,RobertE.–AWorkingLifeforPeoplewithSevere MentalIllness.DOI:10.1093/acprof:oso/9780195131215.001.0001.
[25]SveinsdottirV,LøvvikC,FyhnT,MonstadK,LudvigsenK,ØverlandS,etal.
ProtocolfortheeffectevaluationofIndividualPlacementandSupport(IPS):a randomizedcontrolledmulticentertrialofIPSversustreatmentasusualfor patientswithmoderatetoseverementalillnessinNorway.BMCPsychiatry 2014;14:307.http://dx.doi.org/10.1186/s12888-014-0307-7.
[26]vanOostromSH,AnemaJR,TerluinB,deVetHCW,KnolDL,vanMechelenW.
Cost–effectivenessofaworkplaceinterventionforsick–listedemployeeswith commonmental disorders:designofarandomizedcontrolledtrial.BMC PublicHealth2008;8(1):12.http://dx.doi.org/10.1186/1471-2458-8–12.
[27]vanVelzenJM,vanBennekomCA,vanDormolenM,SluiterJK,Frings–Dresen MH. Evaluation of the implementation of the protocol of an early vocational rehabilitation intervention for people with acquired brain injury. Disabil Rehabil 2016;38(1):62–70. http://dx.doi.org/10.3109/
09638288.2015.1017057.
[28]GrahamCW,WestMD,BourdonJL,IngeKJ,SewardHE.EmploymentInter- ventionsforReturntoWorkinWorkingAgedAdultsFollowingTraumatic Brain Injury (TBI): A Systematic Review. Campbell Systematic Reviews 2016;12(1).http://dx.doi.org/10.4073/csr.2016.6[i–133].
[29]VikaneE,HellstrømT,RøeC,Bautz–HolterE,AßmusJ,SkouenJS.Multidisci- plinaryoutpatienttreatmentinpatientswithmildtraumaticbraininjury:A randomised controlled intervention study. Brain Inj 2017;31(4):475–84.
http://dx.doi.org/10.1080/02699052.2017.1280852.
[30]CorriganJD,Harrison-FelixC,BognerJ,DijkersM,TerrillMS,WhiteneckG.
Systematicbiasintraumaticbraininjuryoutcomestudiesbecauseoflossto follow-up. Arch PhysMed Rehabil 2003;84(2):153–60. http://dx.doi.org/
10.1053/apmr.2003.50093.
[31]ReynoldsS,PaniakC,Toller–LobeG,NagyJ.Alongitudinalstudyofcompen- sation–seekingandreturntoworkinatreatedmildtraumaticbraininjury sample. J Head Trauma Rehabil 2003;18(2):139–47. http://dx.doi.org/
10.1097/00001199-200303000-00005.
[32]SteenstraIA,Anema JR,Bongers PM,de VetHC, vanMechelenW.Cost effectivenessofamulti–stagereturntoworkprogramforworkersonsick leaveduetolowbackpain,designofapopulationbasedcontrolledtrial [ISRCTN60233560]. BMC musculoskeletal disorders 2003;4:26. http://
dx.doi.org/10.1186/1471-2474-4–26.
