Impact of improved recording of work-relatedness in primary care visits at occupational health services on sickness absences: study protocol for a randomised controlled trial

STUDY PROTOCOL Open Access Impac o imp o ed eco ding o wo k- ela edness in p ima y ca e isi s a occupa ional heal h se ices on sickness absences: s udy p o ocol o a andomised con olled ial Salla A kins 1,4* , Ulla Ojajä i 2 , Nina Talola 1 , Me i Viljamaa 2 , Jaakko Ne alainen 3 and Jukka Ui i 1 Abs ac Backg ound: Employmen p o ec s and os e s heal h. Occupa ional heal h se ices, pa icula ly in Finland, ha e a cen al ole in p o ec ing employee heal h and p e en ing wo k abili y p oblems. Howe e , p ima y ca e wi hin occupa ional heal h se ices is cu en ly unde used in in o ming p e en i e ac i i ies. This s udy was designed o assess whe he he eco ding o wo k abili y p oblems and imp o emen o ollow-up o wo k- ela ed p ima y ca e isi s can educe sickness absences and wo k disabili y pensions a e 1 yea . Me hods/design: A p agma ic ial will be conduc ed using pa ien elec onic egis e s and egis e s o he cen al pensions agency in Finland. Twen y- wo occupa ional heal h cen es will be andomised o in e en ion and con ol g oups. In e en ion uni s will ecei e aining o imp o e eco ding o wo k abili y illnesses in he p ima y ca e se ing and imp o ed ollow-up p ocedu es. The in e en ion impac will be assessed h ough examining a es o sickness absence ac oss in e en ion and con ol clinics as well as be o e and a e he in e en ion. Discussion: The ial will de elop knowledge o he in e en ion po en ial o p ima y ca e o p e en ing wo k disabili y pensions and sickness absence. The use o ou ine pa ien egis e s and pensions egis e s o assess he ou comes o a andomised con olled ial will b ing o wa d ial me hodology, pa icula ly when using egis e -based da a. I success ul, he in e en ion will imp o e he quali y o occupa ional heal h ca e p ima y ca e and con ibu e o educing wo k disabili y. T ial egis a ion: ISRCTN Regis y e e ence numbe ISRCTN45728263. Regis e ed on 18 Ap il 2016. Keywo ds: Occupa ional heal h ca e, Sickness absence, Wo k abili y, Clus e andomised ial, Elec onic pa ien egis e s Backg ound The e is a s ong ela ionship be ween employmen and heal h. Full- ime employmen p o ec s and os e s heal h [1–3]. The bene icial e ec s o wo k on pe cei ed heal h s a us and quali y o li e can be seen e en a e e- employmen a e a pe iod o unemploymen [4]. Be- cause o he impo ance o wo k o bo h indi idual well-being and heal h, along wi h na ional p oduc i i y, i is impo an o ensu e ha indi iduals a e heal hy, main ain hei wo k abili y and a e able o pa icipa e in he wo k o ce. Main aining wo k abili y, howe e , is no a simple p ocess. The e a e se e al iden i ied isk ac o s o wo k disabili y, including wo k cha ac e is ics [5, 6]; uns able jobs and unemploymen [7]; a numbe o ch onic diseases [8]; sickness absence a es [9]; and some pe sonal cha ac e is ics, such as socioeconomic cha ac- e is ics [6], educa ion le el [10], gende [11] and age [10]. Mos o hese isk ac o s o wo k disabili y a e e- la ed o wo kplace cha ac e is ics [12], and employe s * Co espondence: [email p o ec ed] 1 Facul y o Medicine and Li e Sciences, Uni e si y o Tampe e, Tampe e, Finland 4 Depa men o Public Heal h Sciences, Ka olinska Ins i u e , S ockholm, Sweden Full lis o au ho in o ma ion is a ailable a he end o he a icle © The Au ho (s). 2017 Open Access This a icle is dis ibu ed unde he e ms o he C ea i e Commons A ibu ion 4.0 In e na ional License (h p://c ea i ecommons.o g/licenses/by/4.0/), which pe mi s un es ic ed use, dis ibu ion, and ep oduc ion in any medium, p o ided you gi e app op ia e c edi o he o iginal au ho (s) and he sou ce, p o ide a link o he C ea i e Commons license, and indica e i changes we e made. The C ea i e Commons Public Domain Dedica ion wai e (h p://c ea i ecommons.o g/publicdomain/ze o/1.0/) applies o he da a made a ailable in his a icle, unless o he wise s a ed. A kins e al. T ials (2017) 18:352 DOI 10.1186/s13063-017-2076-3 can in luence hese ac o s by leade ship, o ganisa ional changes and wo king a angemen s [13]. Fo e ec i e ac- ion, he employe needs collabo a ion wi h occupa ional heal h se ices (OHS) ha can p o ide in o ma ion e- ga ding he ea ly signs o wo k disabili y. Collabo a ion wi h OHS can con inue om ea ly signs o wo k disabili y o managing ongoing wo k disabili y, such as by modi ying wo k asks o suppo ing e u n o wo k [14]. Wo k abili y is a cen al issue in Finland because 7.5% o he wo king popula ion e i e ea ly on disabili y pen- sions, which also lowe s he a e age e i emen age [5] and p oduc i i y. In Finland, OHS moni o and main ain employee wo k abili y and h ough a ious ac i i ies o p e en wo k disabili y and o he wo k- ela ed heal h p oblems [15]. The wide manda e o OHS is possible be- cause all employees a e equi ed o ha e p e en i e OHS by law, and app oxima ely 80% o he wo king popula ion also ecei e p ima y ca e se ices h ough OHS. In es men s in OHS ha e esul ed in bene i s o o ganisa ions and hei employees, such as educ ion in sick lea e [16] and inc eased p o i abili y [16, 17]. In e- cen yea s, disabili y pension a es ha e also declined [18]. In Finland, OHS a e an impo an p o ide o heal h ca e, pa allel o he p i a e and public sec o s [15]. OHS a e a a key posi ion o os e ing employees’ wo k abili y because hey see employees du ing bo h p e en i e and cu a i e ac i i ies. Though p og ess has been made in educing wo k dis- abili y, mo e wo k is needed in his a ea, and OHS po- en ial in his espec is unde used. The e is insu icien in o ma ion abou he ex en , impac and ac ions aken, based on assessmen s o wo k abili y o diagnoses’wo k- ela edness du ing hese isi s. O e hal o he causes o hese isi s a e wo k- ela ed (27% caused by wo k and 52% impai wo k abili y) [19], bu his in o ma ion is a ely used in p e en i e ac i i ies because elec onic e- co ds do no allow o p ecise collec ion and use o his in o ma ion. In e ening in wo king condi ions om he p ima y ca e se ing could con ibu e o imp o ing em- ployee wo k abili y and educing disabili y [20]. The ial desc ibed in his a icle is a mul i-cen e, p agma ic, pa allel-g oup, clus e andomised con olled ial using elec onic heal h eco ds and pension egis- ies o imp o e he eco ding o wo k- ela edness and impac on wo k abili y o p ima y ca e isi s and ul i- ma ely he ollow-up o hese cases a OHS uni s in Finland. We ollow he s anda dised S anda d P o ocol I ems: Recommenda ions o In e en ional T ials (SPIRIT) checklis in his epo (Addi ional ile 1). Aim 1 Ou i s aim in he p esen s udy is o assess he e ec o an in e en ion o imp o ing eco ding and ollow- up o pa ien p ima y ca e isi s in occupa ional heal h ca e on medium-leng h sickness absence. Ou hypo h- esis is ha imp o ed eco ding o wo k- ela edness and/ o e ec s on wo k abili y du ing p ima y ca e isi s and close ollow-up and ini ia ion o ac ions in ol ing he pa ien and employe a he wo kplace will educe medium-leng h (4–9 days) sickness absences. Aim 2 Ou second aim in he p esen s udy is o assess he e ec o an in e en ion o imp o e he eco ding and ollow-up o p ima y ca e isi s o OHS on wo k dis- abili y pensions as eco ded in he cen al pensions egis e and long- e m sickness absence (9–60 days) as eco ded in elec onic pa ien egis e s. We will also in- es iga e he impac o he in e en ion on sho - e m (1–3 days) sickness absences. Ou hypo hesis is ha close pa ien ollow-up will enable he occupa ional heal h (OH) and employe eams o add ess wo k abil- i y p oblems ea lie and hus educe sickness absences and disabili y pension a es. Resea ch ques ion Does an in e en ion o imp o e eco ding and ollow- up o p ima y ca e isi s’wo k- ela edness and hei impac on wo k abili y educe medium-leng h sickness absences (4–9 days) mo e han no in e en ion when measu ed a e 1 yea ? Me hods/design T ial design The ial is a mul i-cen e, pa allel-g oup, p agma ic su- pe io i y, clus e andomised con olled ial o 1 yea in du a ion. Each OH uni ep esen s one clus e . Because he in e en ion is p o ided o doc o s esponsible o mul iple pa ien s and mul iple employe o ganisa ions, a clus e design was chosen o dec ease con amina ion bias. The s udy design was de eloped in acco dance wi h he SPIRIT guidelines checklis . Se ing The s udy was de eloped oge he wi h Pihlajalinna Työ e eys, an OHS p o ide ha is pa o a p i a e conso ium. A he ime o ial ini ia ion, he p o ide has 22 OH uni s in di e en ci ies and owns ac oss Finland, which oge he had app oxima ely 67,000 em- ployees on hei clien lis s a he end o 2015. In Finland, app oxima ely 91% o all employees a e wi hin he emi o OHS [21], app oxima ely 86% o whom ha e company-p o ided p ima y ca e se ices, wi h he es ecei ing legisla i e p e en i e se ices. Pa icipan s and ec ui men Each OH uni wi hin Pihlajalinna Työ e eys had be- ween 500 and 9000 employees on hei egis e . In he A kins e al. T ials (2017) 18:352 Page 2 o 8 s udy, each clus e is conside ed an independen uni . All OH uni s in he conso ium wi h ull- ime OH s a we e selec ed o he s udy. Two cen es wi hin he same own as a la ge cen e wi h pa - ime OH s a we e le ou o he s udy. The in e en ion aining was consid- e ed ou ine s a aining. Employees on he egis e o he OHS in 2015–2017 be ween he ages o 18 and 65 yea s will be included in he analysis. The coho will be open, and employees a e eligible i hey join he OH egis e a any ime du ing he s udy pe iod. An em- ployee isi ing any o he 22 OH uni s we e conside ed eligible o analysis. Randomisa ion and alloca ion To alloca e OH uni s in o in e en ion and con ol g oups, a andomisa ion p ocess using dynamic alloca- ion (speci ically minimisa ion) was used. This me hod was i s p esen ed by Ta es [22] and la e by Pocock and Simon [23]. The use o minimisa ion has inc eased in ecen yea s [24], and i has been sugges ed as he ‘pla inum s anda d’o ials [25] because i allows o balance be ween ea men and con ol a ms. In his ial, we conside ed he ollowing as po en ial con- ounde s and included hem in he minimisa ion: (1) he clien olume o he OHS sepa a ed in o h ee en e p i- ses—small, medium and la ge, (2) he main sec o o he OHS clien s (indus y s o he s) and (3) whe he one o he OHS uni ’s en e p ise clien s was a majo pape ac- o y o no ( he pape ac o y had ins i u ed la ge-scale OH in e en ions in ecen yea s). We began he andomisa ion wi h simple andomisa- ion conduc ed by he eam s a is ician (NT). We allo- ca ed he i s ou uni s h ough simple andomisa ion (1:1) using he andom numbe gene a o in he R p o- g am (R Founda ion o S a is ical Compu ing, Vienna, Aus ia). Fo he emaining 18 uni s, we used minimisa- ion o alloca e each OHS uni o ei he he ea men g oup o he con ol g oup o minimise he o al imbal- ance o po en ial con ounde s, con inuing wi h a 1:1 a- io. Because he in e en ion was embedded in ou ine p ac ice, alloca ion was concealed om pa icipan s bu no om p o ide s o in es iga o s. The conce n o lack o blinding du ing he minimisa ion p ocess did no apply in his case, owing o he clus e -based design and he ac ha bo h he in e en ion and con ol clus e s a e pa o Pihlajalinna Työ e eys. Assessmen poin s Baseline in o ma ion will be assessed o 1 yea p e- ceding he in e en ion, om 1 May 2015 o 1 May 2016. App oxima ely 72,000 spells o sickness absence we e egis e ed among 25,000 clien s a Pihlajalinna Työ e eys in 2015. Fo he p ima y ou come (mean sickness absence o medium leng h a e he in e en ion compa ed wi h baseline), assessmen will be conduc ed by NT a e 1 yea o he in e en ion, using da a up o 1 May 2017 (see Fig. 1 o ial low and Fig. 2 o SPIRIT diag am). Fo mean disabili y pensions pe cen e, he ime will be inc eased o 31 Decembe 2017, wi h subse- quen analysis hoped o a e mo e unding is ecei ed. In e en ion All uni s we e in o med o he ongoing esea ch and i s aims be o e he s a o aining. In he in e en ion g oup, all doc o s and nu ses o o he OH p o essionals who a e pa o in e en ion uni s ook pa in aining p o ided by Pihlajalinna Työ e eys. The in e en ion aining s a ed on 1 May 2016. The aining las ed ap- p oxima ely 1 h. Du ing he aining, he p inciple o he in e en ion was explained, and all pa icipan s ecei ed w i en ins uc ions on he eco ding p ocedu e and in e en ion p ocesses. When physicians o nu ses we e absen , a ollow-up isi o phone call was made o im- pa he in e en ion in o ma ion. Each physician and nu se esponsible o a company’s OHS ook esponsibi- li y o hei companies’employees’ ollow-up. The im- plemen a ion o he in e en ion was ollowed by he Pihlajalinna Työ e eys aine , who ollowed up wi h nu ses and physicians and in i ed hem o pe sonal dis- cussions on he p ocess as necessa y. Figu e 3 de ails he in e en ion and pa icipan low. The in e en ion was designed o i s es ima e he ela- ionship o he main o addi ional diagnosis o he isi eason o wo k. Du ing each p ima y ca e isi , he a - ending physician would assess i s ly whe he he diag- nosis was ela ed o wo k. Secondly, he clien ’s isko applying o a wo k disabili y pension was es ima ed acco ding o ou ca ego ies: (1) no isk o wo k disabili y pension, (2) less han 50% chance o wo k disabili y pen- sion in he nea u u e (2–5 yea s), (3) mo e han 50% chance o wo k disabili y pension in he nea u u e (2–5 yea s) and (4) immedia e isk o wo k disabili y pension. This assessmen is en e ed in Pihlajalinna Työ e eys’s elec onic pa ien egis e . Once weekly, all pa ien s a e ecalled on he elec onic sys em o whom a eco d o wo k- ela edness was made. Each case is hen assigned o he OH nu se esponsible o he pa ien ’semploye ela- ions. The esponsible OH nu se hen ini ia es ollow-up isi s as app op ia e. These can include wo ksi e isi s; ipa i e nego ia ions be ween he OH physician, em- ployee and employe ; occupa ional physio he apy o he employee; occupa ional psychologis s’assessmen s; o di - e en kinds o ehabili a ion p ocesses a an indi idual le el. When a clien is deemed as a g ea e han 50% isk o wo k disabili y, a sepa a e ehabili a ion plan will be comple ed wi h he clien p esen . This plan will include de ails on he clien ’s heal h s a us, planned ac ions o ehabili a ion, and ollow-up by he OH p o essionals. A kins e al. T ials (2017) 18:352 Page 3 o 8 Fig. 2 SPIRIT igu e: Summa izes he alloca ion, in e en ions, and ou comes o he s udy Fig. 1 T ial low diag am. OHS Occupa ion heal h se ices A kins e al. T ials (2017) 18:352 Page 4 o 8 The con ol g oup did no ecei e addi ional aining on eco ding wo k- ela edness. I is possible ha Con ol g oup physicians migh hea o he in e en ion and migh pay mo e a en ion o assessing po en ial h ea s o wo k abili y du ing he in e en ion. Howe e , no cen al in o ma ion is p o ided ega ding he in e en- ion in he conso ium, and no o malised in o ma ion is p o ided o con ol uni s, which may educe con ound- ing. A e he end o he in e en ion, con ol uni s will be gi en a chance o pa icipa e in aining and imple- men he in e en ion i he esul s o he in e en ion we e posi i e. Compliance is checked ia egula con- ac s by he in e en ion aine wi h he OH uni s, enqui ing abou implemen a ion and challenges. Ou come measu es P ima y ou come The p ima y ou come is educ ion om baseline o he mean numbe o medium e m (4–9 days) sickness ab- sences om he wo kplace pe in e en ion and con ol cen e a e 1 yea o ollow-up as measu ed by OHS pa- ien eco ds. We chose medium-leng h sickness ab- sences as he p ima y ou come o ensu e ha we ha e a su icien numbe o cases o assess impac . Wi hin he Pihlajalinna Työ e eys coho in 2015, he e we e 17,794 ins ances o sickness absences be ween 4 and 9 days in du a ion. This ou come is also suppo ed by a la ge Finnish s udy on o e 6000 municipal wo ke s [26] which indica ed ha sickness absences o be ween 4 and 9 days we e mo e p e alen (85.7 and 59.0 pe 100 pe son-yea s o women and men, espec i ely) han hose o e 2 weeks (10+ days) in du a ion. The s udy also indica ed ha sel -ce i ied (1–3 days) absences we e he mos common (156.4 and 96.1 pe 100 pe son- yea s o women and men, espec i ely), bu hese a e no eliably a ailable om he elec onic pa ien egis e held by Pihlajalinna Työ e eys. Longe sickness ab- sences (be ween 2 weeks and 60 days) a e less p e alen and had a es o 22.8 and 18.0 pe 100 pe son-yea s o women and men, espec i ely, wi h long absences o Fig. 3 Pa icipan low diag am A kins e al. T ials (2017) 18:352 Page 5 o 8 o e 60 days being ela i ely in equen (4.7 and 4.4 pe 100 pe son-yea s o women and men, espec i ely) [26]. Sho e (1–3 days) sickness absences we e no included as a p ima y ou come, because hey a e no eliably a ailable om he pa ien egis e , and hey ha e been ound no o be associa ed wi h disabili y pensions, excep h ough hei associa ion wi h longe - e m (3+ days) sick- ness absences [9]. Long (3+ days) sickness absence spells ha e been ound o be p edic i e o wo k disabili y be o e he age o 55 [9] and ega dless o heal h s a us [27]. Seconda y ou comes We will assess he ollowing seconda y ou comes: 1. Reduc ion in mean numbe o sho e m (1–3 days) sickness absences om he wo kplace pe clus e om baseline a e 1 yea om he s a o he in e en ion as measu ed by sel - epo eco ded on OHS eco ds o OHS eco ds o sickness absence 2. Reduc ion o mean numbe o any o m o wo k disabili y pensions as measu ed by an employee egis e ing as ecei ing a wo k disabili y pension on he cen al pensions egis e om baseline o up o 2 yea s om he in e en ion as measu ed by he en y on he cen al pensions egis e 3. Reduc ion o mean numbe o long- e m (9+ days) sickness absences om he wo kplace pe clus e om baseline o 1 yea a e he in e en ion as measu ed by OHS eco ds Powe calcula ion The p e-s udy powe and sample size calcula ions o ou main ou come we e based on da a collec ed in 2015 wi h es ima es o he indi iduals be ween he ages o 18 and 65 om he 22 OH uni s (n= 24,892). The 22 clus- e s had an a e age size o 2300 indi iduals. We es i- ma ed he in a-clus e co ela ion coe icien o be 0.05. We used he n4means unc ion in he CRTSize package in R [28], which akes in o accoun ha he ou come o indi iduals canno be assumed o be independen . We used a wo-sided alpha o 5% o ou p ima y ou come, he sickness absences o medium (4–9) days, wi h an SD o 1.5. Wi h 11 clus e s andomised o each a m (assum- ing 2300 indi iduals pe OHS), he s udy had 91% powe o de ec a leas a 10% di e ence be ween indi iduals in he in e en ion and con ol OHS g oups. These calcula- ions showed ha e en wi h a 20% loss o ollow-up o missing da a, we would s ill ha e 90% powe o he p i- ma y analysis. Da a collec ion Da a on pa ien isi s o he OHS uni we e collec ed om a cen al pa ien egis e . Sel - epo ed sickness ab- sences o be ween 1-3 days a e epo ed by he employe o he OHS uni . P e-in e en ion da a we e ex ac ed by he in o ma ion echnology se ice p o ide o Pihlaja- linna Työ e eys, who p o ided coded and anonymised pa ien da a, a sepa a e lis o pe sonal iden i ie s and a code o combining he pa ien da a wi h he pe sonal iden i ie . The lis o pe sonal iden i ie s was sen o he Finnish Cen e o Pensions, which holds a egis y o pensions, including disabili y pensions, o he en i e popula ion. The Finnish Cen e o Pensions ex ac ed da a on pensions o he en i e lis o pe sonal iden i ie s and coded i wi h he same code. The pseudonymised pa- ien da a and coded pension da a we e sen o he Uni e - si y o Tampe e. The same p ocess will be epea ed o da a collec ed du ing and a e he in e en ion. Da a managemen The da a om he Finnish Cen e o Pensions and Pihlajalinna Työ e eys we e combined on he basis o pa- ien codes a he Uni e si y o Tampe e using R so wa e. Da a we e kep on a secu e se e , passwo d-accessible only o he esea ch eam. Da a accu acy will be checked by e e ing back o he OHS uni s o de ails whe e he e a e suspec ed e o s in da a en y. The Finnish Cen e o Pensions egis e sys em checks he da a au oma ically, and he da a a e ou inely e iewed and co ec ed. Da a analysis Da a will be analysed on he in en ion- o- ea p inciple. All employees egis e ed a a clus e will be analysed as pa o ha clus e . We will use bo h R and IBM SPSS S a is ics so wa e (IBM, A monk, NY, USA) o analyse he da a. We will i s examine he da a using desc ip i e s a is ics. Be ween-g oup analysis will be conduc ed using Pea son’s chi-squa e es o Fishe ’s exac es o ca ego ical da a and S uden ’s es o he Mann- Whi ney U es o con inuous a iables, depending on no mali y o a dis ibu ion. The no mali y o he dis i- bu ion o he a iable will be examined wi h he Kolmogo o -Smi no es . The in e en ion e ec s o p ima y and seconda y ou comes will be analysed using eg ession analysis, logis ic eg ession will be used o dicho omous ou come a iables, and Poisson eg ession will be used when he ou come a iable is a coun . In bo h analyses, we will examine he impac o , and con- ol o , possible con ounde s. The eg ession analysis will esul in ORs and a e a ios wi h 95% CIs. The e- sul s will be conside ed signi ican a p< 0.05. Fo ana- lysing he di e ences in du a ion o sickness absences, we will use Kaplan-Meye su i al cu es and he log- ank es . Da a moni o ing Because he s udy did no include ulne able popula- ions, was conside ed no o ep esen a ha m o A kins e al. T ials (2017) 18:352 Page 6 o 8 pa icipan s, and was conside ed o ha e po en ial o ou ine p ac ice, we deemed a da a moni o ing commi - ee unnecessa y. E hical app o al E hical app o al was ecei ed om he Pi kanmaa Hos- pi al Dis ic e iew boa d (Pi kanmaa Hospi al Dis ic e iew boa d 10/03/2016, e e ence R16041). Discussion The weal h o egis e s, especially in No dic coun ies, allows o inno a i e esea ch o be conduc ed wi h minimal need o addi ional da a collec ion o ouble o he pa icipan s. Recen au ho s ha e a gued ha egis y ials can ha e a weal h o bene i s, bu quali y issues need o be aken in o accoun [29]. Ou p agma ic ial uses he combina ion o elec onic heal h eco ds wi h egis ies and es s a eal-wo ld in e en ion wi h a good s udy design and maximum ex e nal alidi y [30] wi h li le dis up ion o he physicians implemen ing he s udy [31]. The ial is p agma ic in na u e, and hus, while yielding esul s highly ans e able o o he se - ings, he e a e s a is ical challenges in design. These in- clude, o example, he high a ia ion in clus e size, which may educe he powe . Rega dless o his, we ex- pec he clus e and o al sample sizes o be su icien o de ec a ue di e ence, and he impac o he clus e size a ia ion can be de e mined du ing analysis. While in e es ing om a me hodological s andpoin , he p ac ical impac o quali y imp o emen in occupa- ional heal h ca e has implica ions beyond his ial and his pa icula conso ium. Th ough assessmen o wo k- ela edness o each isi , p ima y ca e isi s can be used e ec i ely as a ool o seconda y p e en ion and ea ly in e en ion. These app oaches can be used also in se - ings whe e occupa ional heal h ca e is no s a u o y. S udies ha e indica ed ha he eco ding o wo k- ela edness o isi s is no op imal [32], and app oaches such as his buil in o se ices could po en ially imp o e epo ing. Imp o ed epo ing and ecogni ion can also con ibu e o collabo a ion be ween OHS and employe s and gene al p ac i ione s [33]. These issues a e impo - an pa icula ly in Eu ope, whe e he popula ion is age- ing and li e expec ancy is inc easing and whe e economic de elopmen depends on a heal hy wo k o ce. Challenges o he success o he ial can be he scale o he in e en ion as well as ollow-up ime, which is sho wi h espec o he ou come o disabili y pensions. Fu he unding will be sough o ollow his coho o a longe pe iod. In addi ion, no all sickness absences can be ex ac ed o all clien s, because clien s may be gi en sickness absence ce i ica es om he public sec o o specialised se ices. Changing beha iou on he basis o one aining isi may be oo op imis ic; physicians may con inue no o eco d wo k- ela edness o impac on wo k abili y op imally; in e nal s uc u es may no suppo collec ing and ollowing up employe isi s; o employe s may no be eage o in es in employee wo k abili y. Th ough his andomised ial, we will be able o assess whe he ou app oach wo ks, and we will conduc impo an me hodological wo k on using elec onic eg- is e s o p agma ic ials. Th ough he planned p ocess e alua ion done alongside he ial, we will be able o as- sess wha wo ked and wha did no wo k, as well as how we should modi y he in e en ion in u u e o ensu e ha suppo ing wo k abili y is a he cen e o p ima y ca e isi s a occupa ional heal h ca e cen es. T ial s a us The ial began on 1 May 2016 and ended on 1 May 2017. Addi ional ile Addi ional ile 1: SPIRIT checklis . (PDF 168 kb) Abb e ia ions OH: Occupa ional heal h; OHS: Occupa ional heal h se ices; SPIRIT: S anda d P o ocol I ems: Recommenda ions o In e en ional T ials Acknowledgemen s The au ho s acknowledge he pa icipa ion o he occupa ional heal h s a in he s udy and all he indi idual clien s who a e pa o his s udy. Funding The p ojec wi hin which his s udy is conduc ed is unded by he Eu opean Social Fund ( e e ence numbe S20659). The unde has aken no ac i e ole in he design, implemen a ion o epo ing o his s udy. A ailabili y o da a and ma e ials No applicable. Au ho s’con ibu ions JU concep ualised he s udy. MV, NT, JN, UO and SA pa icipa ed in he design o he s udy. SA, MV and UO coo dina ed implemen a ion o he s udy. SA w o e he i s d a o he manusc ip and led subsequen e isions. UO, MV and JN edi ed he manusc ip o in ellec ual con en . NT w o e he sampling and sample size sec ion wi h con ibu ion by JN. All au ho s con ibu ed o subsequen d a s o he manusc ip , and all au ho s ead and app o ed he inal manusc ip . E hics app o al and consen o pa icipa e E hics app o al was ecei ed om he Pi kanmaa Hospi al Dis ic E hical Re iew Commi ee ( e e ence numbe R16041). The need o indi idual in o med consen was wai ed due o he la ge numbe o pa icipan s and he non-in asi e na u e o he s udy. Consen o publica ion No applicable. Compe ing in e es s The au ho s decla e ha hey ha e no compe ing in e es s. Publishe ’sNo e Sp inge Na u e emains neu al wi h ega d o ju isdic ional claims in published maps and ins i u ional a ilia ions. A kins e al. T ials (2017) 18:352 Page 7 o 8 Au ho de ails 1 Facul y o Medicine and Li e Sciences, Uni e si y o Tampe e, Tampe e, Finland. 2 Pihlajalinna Työ e eys, Tampe e, Finland. 3 Facul y o Social Sciences, Uni e si y o Tampe e, Tampe e, Finland. 4 Depa men o Public Heal h Sciences, Ka olinska Ins i u e , S ockholm, Sweden. Recei ed: 6 Decembe 2016 Accep ed: 30 June 2017 Re e ences 1. Ross CE, Mi owsky J. Does employmen a ec heal h? J Heal h Soc Beha . 1995;36:230–43. 2. Els ad JI. Inequali ies in heal h ela ed o women’s ma i al, pa en al, and employmen s a us—a compa ison be ween he ea ly 70s and he la e 80s. Soc Sci Med. 1996;42:75–89. 3. an de Noo d M, Ijzelenbe g H, D oome s M, P ope KI. Heal h e ec s o employmen : a sys ema ic e iew o p ospec i e s udies. Occup En i on Med. 2014;71:730–6. 4. Ca lie BE, Schu ing M, Lö e s FJB, Bakke B, Bo ge s N, Bu do A. The in luence o e-employmen on quali y o li e and sel - a ed heal h, a longi udinal s udy among unemployed pe sons in he Ne he lands. BMC Public Heal h. 2013;13:503. 5. Lahelma E, Laaksonen M, Lallukka T, Ma ikainen P, Pie iläinen O, Saas amoinen P, e al. Wo king condi ions as isk ac o s o disabili y e i emen : a longi udinal egis e linkage s udy. BMC Public Heal h. 2012;12:309. 6. Pol inen A, Gould R, Lahelma E, Ma ikainen P. Socioeconomic di e ences in disabili y e i emen in Finland: he con ibu ion o ill-heal h, heal h beha iou s and wo king condi ions. Scand J Public Heal h. 2013;41:470–8. 7. Gus a sson K, A onsson G, Ma klund S, Wikman A, Flode us B. Pe iphe al labou ma ke posi ion and isk o disabili y pension: a p ospec i e popula ion-based s udy. BMJ Open. 2014;4, e005230. 8. Män yniemi A, Oksanen T, Salo P, Vi anen M, Sjös en N, Pen i J, e al. Job s ain and he isk o disabili y pension due o musculoskele al diso de s, dep ession o co ona y hea disease: a p ospec i e coho s udy o 69,842 employees. Occup En i on Med. 2012;69:574–81. 9. Ki imäki M, Fo ma P, Wiks öm J, Halmeenmäki T, Pen i J, Elo ainio M, e al. Sickness absence as a isk ma ke o u u e disabili y pension: he 10- own s udy. J Epidemiol Communi y Heal h. 2004;58:710–1. 10. Olo sson T, Pe e sson IF, E iksson JK, Englund M, Sima d JF, Nilsson JÅ, e al. P edic o s o wo k disabili y du ing he i s 3 yea s a e diagnosis in a na ional heuma oid a h i is incep ion coho . Ann Rheum Dis. 2014;73:845–53. 11. Albe sen K, Lund T, Ch is ensen KB, K is ensen TS, Villadsen E. P edic o s o disabili y pension o e a 10-yea pe iod o men and women. Scand J Public Heal h. 2007;35:78–85. 12. Joling CI, Bla e BM, Ybema JF, Bonge s PM. Can a o able psychosocial wo k condi ions and high wo k dedica ion p o ec agains he occu ence o wo k- ela ed musculoskele al diso de s? Scand J Wo k En i on Heal h. 2008;34:345–55. 13. Kaspin LC, Go man KM, Mille RM. Sys ema ic e iew o employe -sponso ed wellness s a egies and hei economic and heal h- ela ed ou comes. Popul Heal h Manag. 2013;16:14–21. 14. an Oos om SH, D iessen MT, de Ve HCW, F anche RL, Schons ein E, Loisel P, e al. Wo kplace in e en ions o p e en ing wo k disabili y. Coch ane Da abase Sys Re . 2009;2, CD006955. 15. Ma imo JP, Mäki alo J; Finnish Ins i u e o Occupa ional Heal h. The s a us o occupa ional heal h se ices in Finland and he ole o he Finnish Ins i u e o Occupa ional Heal h in he de elopmen o occupa ional heal h se ices: epo o he in e na ional e alua ion o he Finnish Ins i u e o Occupa ional Heal h (FIOH). Helsinki: Työ e eyslai os; 2014 16. Baicke K, Cu le D, Song Z. Wo kplace wellness p og ams can gene a e sa ings. Heal h A (Millwood). 2010;29:304–11. 17. Mille P, Whynes D, Reid A. An economic e alua ion o occupa ional heal h. Occup Med (Lond). 2000;50:159–63. 18. Sauni R, Ui i J. Uude yöky y ömyyseläkkee o a ähen ynee neljänneksen. Suom Laaka il. 2015;45:3056–7. 19. Ikonen A, Räsänen K, Manninen P, Rau io M, Husman P, Ojajä i A, e al. Wo k- ela ed p ima y ca e in occupa ional heal h physician’s p ac ice. J Occup Rehabil. 2012;22:88–96. 20. Ma imo KP, Va onen H, Husman K, Viika i-Jun u a E. Fac o s associa ed wi h sel -assessed wo k abili y. Occup Med (Lond). 2007;57:380–2. 21. Kauppinen T, Ma ila-Holappa P, Pe kiö-Mäkelä M, Saalo A, Toikkanen J, Tuomi aa a S, e al. Työ ja e eys suomessa 2012: seu an a ie oa yöolois a ja yöhy in oinnis a. Työ e eyslai os: Helsinki; 2013. 22. Ta es DR. Minimiza ion: a new me hod o assigning pa ien s o ea men and con ol g oups. Clin Pha macol The . 1974;15:443–53. 23. Pocock SJ, Simon R. Sequen ial ea men assignmen wi h balancing o p ognos ic ac o s in he con olled clinical ial. Biome ics. 1975;31:103–15. 24. Ta es DR. The use o minimiza ion in clinical ials. Con emp Clin T ials. 2010;31:180–4. 25. T easu e T, MacRae KD. Minimisa ion: he pla inum s anda d o ials? Randomisa ion doesn’ gua an ee simila i y o g oups; minimisa ion does. BMJ. 1998;317:362–3. 26. Laaksonen M, Ma ikainen P, Rahkonen O, Lahelma E. Explana ions o gende di e ences in sickness absence: e idence om middle-aged municipal employees om Finland. Occup En i on Med. 2008;65:325–30. 27. Hul in H, Lindholm C, Mölle J. Is he e an associa ion be ween long- e m sick lea e and disabili y pension and unemploymen beyond he e ec o heal h s a us? –A coho s udy. PLoS One. 2012;7, e35614. 28. Ro ondi MA. CRTSize: sample es ima ion unc ions o clus e andomized ials. 2011. A ailable om: h p://c an. -p ojec .o g/web/packages/CRTSize/ index.h ml. Accessed 15 Aug 2016. 29. Laue MS, D’Agos ino S RB. The andomized egis y ial— he nex dis up i e echnology in clinical esea ch? N Engl J Med. 2013;369:1579–81. 30. S aa TP, Goldac e B, Gulli o d M, Cassell J, Pi mohamed M, Taweel A, e al. P agma ic andomised ials using ou ine elec onic heal h eco ds: pu ing hem o he es . BMJ. 2012;344, e55. 31. Gliklich RE, D eye NA, Lea y MB, edi o s. In e acing egis ies wi h elec onic heal h eco ds. In: Regis ies o e alua ing pa ien ou comes: a use ’s guide. Vol. 2. 3 d ed. Rock ille, MD: Agency o Heal hca e Resea ch and Quali y; 2014. p. 3–22 32. A naud S, Cabu S, Viau A, Sou ille M, Ve ge P. Di e en epo ing pa e ns o occupa ional diseases among physicians: a s udy o F ench gene al p ac i ione s, pulmonologis s and heuma ologis s. In A ch Occup En i on Heal h. 2010;83:251–8. 33. Moßhamme D, Na anzon I, Manske I, G u schkowski P, Riege MA. Coope a ion be ween gene al p ac i ione s and occupa ional heal h physicians in Ge many: how can i be op imised? A quali a i e s udy. In A ch Occup En i on Heal h. 2014;87:137–46. • We accep p e-submission inqui ies • Ou selec o ool helps you o ind he mos ele an jou nal • We p o ide ound he clock cus ome suppo • Con enien online submission • Tho ough pee e iew • Inclusion in PubMed and all majo indexing se ices • Maximum isibili y o you esea ch Submi you manusc ip a www.biomedcen al.com/submi Submi you nex manusc ip o BioMed Cen al and we will help you a e e y s ep: A kins e al. T ials (2017) 18:352 Page 8 o 8