Work ability of employees in changing social services and health care organizations in Finland

151 ORIGINAL PAPERS In e na ional Jou nal o Occupa ional Medicine and En i onmen al Heal h 2012;25(2):151 – 165 DOI 10.2478/S13382-012-0014-x WORK ABILITY OF EMPLOYEES IN CHANGING SOCIAL SERVICES AND HEALTH CARE ORGANIZATIONS IN FINLAND LAURI KOKKINEN 1 and ANNE KONU 2 1 Finnish Ins i u e o Occupa ional Heal h, Tu ku, Finland Cen e o Expe ise o Wo k O ganiza ions 2 Uni e si y o Tampe e, Tampe e, Finland School o Heal h Sciences Abs ac Objec i es: In his s udy, we examined he connec ion be ween o ganiza ional changes and employees own e alua ions o hei wo k abili y. Ma e ials and Me hods: In ea ly 2010, we asked employees (n = 2429) wo king in he Finnish social se ices and heal h ca e indus y o iden i y all he o ganiza ional changes ha had occu ed a hei wo kplaces o e he p e ious wo yea s, and o e alua e hei own wo k abili y and whe he di e en s a emen s ela ed o he elemen s o wo k abili y we e ue o alse a he ime o he su ey. Fo ou me hod o analysis, we used logis ical eg ession analysis. Resul s: In models adjus ed o gende , age, ma i al s a us, p o essional educa ion and manage ial posi ion, he esponden s who had encoun e ed o ganiza ional changes we e a a highe isk o eeling ha hei wo k abili y had dec eased (OR = 1.49) han he esponden s whose wo kplaces had no been a ec ed by changes. Those esponden s who had encoun e ed o ga- niza ional changes we e also a a highe isk o eeling ha se e al elemen s ela ed o wo k abili y had de e io a ed. The isk o ha ing dec eased sel -e alua ed wo k abili y was in u n highe among he esponden s who s a ed hey could no unde s and he changes han among hose esponden s who unde s ood he changes (OR = 1.99). This was also he case among esponden s who el ha hei oppo uni ies o be in ol ed in he changes had been poo in compa ison o hose who el ha hey had had good oppo uni ies o be in ol ed in he p ocess (OR = 2.16). Conclusions: Ou indings sug- ges ha he o ganiza ional changes in social and heal h ca e may en ail, especially when poo ly execu ed, cos s o which li le a en ion has been paid un il now. When implemen ing o ganiza ional changes, i is i al o ensu e ha he employees unde s and why he changes a e being made, and ha hey a e gi en he oppo uni y o ake pa in he implemen a ion o hese changes. Key wo ds: O ganiza ional change, Wo k abili y, Social se ices, Heal h ca e Recei ed: No embe 12, 2011. Accep ed: Janua y 2, 2012. Add ess ep in eques o L. Kokkinen, Finnish Ins i u e o Occupa ional Heal h, Lemminkäisenka u 14-18, 20520 Tu ku, Finland (e-mail: [email p o ec ed]). INTRODUCTION The Finnish social and heal h ca e sys em was c ea ed main- ly du ing he 1960s and 1970s, and i s ins i u ional s uc u e has emained ela i ely unal e ed un il oday. On he le el o indi idual o ganiza ions, howe e , he e ha e been apid changes o qui e some ime now. Bo h he se ice p oduc- ion s uc u es and he o ganiza ion o labo ha e been unde cons an de elopmen . On he o ganiza ional le el, he changes implemen ed o e he las ew yea s ha e been ei he comp ehensi e and co e ed all ac i i ies, o ha e been di e en ia ed and p ojec -like in na u e, o indeed any hing be ween he wo. The p e ious esea ch li e a- u e shows e idence ha ce ain adical changes, pe sonnel cu s and p i a iza ion o ins ance, a e connec ed o wo k No e Ins i u e o Occupa ional Medicine, Łódź, Poland ORIGINAL PAPERS L. KOKKINEN and A. KONU IJOMEH 2012;25(2) 152 oppo uni ies employees had o be in ol ed in he plan- ning and implemen a ion o hese changes. Employees’ unde s anding o changes In he discussion and esea ch li e a u e on leade ship in o ganiza ional change, o he las h ee decades he cen- e s age has been aken by he manage ’s abili y o gain a sha ed unde s anding wi h his/he s a abou he o ga- niza ion’s need o change, i s goals, and mee ing hese goals and ollowing hem up. Poli ical esea che James McG ego Bu ns is one o he ini ia o s o his discussion. Bu ns b ough his iew on leade ship, de i ed om his analysis o poli ical leade s, in o he s udy o wo k o ga- niza ions [7]. He challenged he idea ha he ela ionship be ween he execu i es and he employees in a wo k o ga- niza ion mus always be based on a clean-cu , mechanical and sho ime-ho izon exchange ela ionship. Acco ding o him, he in eg al pa o leade ship in change was ha he leade s and s a we e able o sha e a common concep ion o he o ganiza ion’s u u e goals and he means o mee hese goals. Bu ns’ concep ion o leade ship has been wide- ly accep ed, and hese ema ks, o iginally de i ed om he analysis o poli ical leade s, ha e also been aised as he ide- al o leade ship in changing wo k o ganiza ions [8–10]. Re- sea ch li e a u e on social and heal h ca e om he las wo decades also indica es ha he way in which change is led is impo an o bo h he employees’ wo k mo i a ion and heal h. An especially impo an pa o his is ensu ing ha employees a e adequa ely awa e o he ad ance planning, implemen a ion and ollow-up s ages o change [11–15]. Employees’ oppo uni ies o be in ol ed in changes Du ing he las ew yea s, ‘ op-down’ o ganiza ional changes ha e been gi ing way o he so-called employee-d i en inno- a ions [16]. In he No dic con ex , employee-d i en inno a- ions can be iewed as a so o a con inuum o he mo e- men ha du ing he 1960s s o e o democ a ize wo k, in opposi ion o he Taylo is ic wo k o ganiza ion. This change, mo i a ion and heal h o employees [1–4]. Howe e , un il now, he e ha e been ew s udies on how o ganiza ional changes a e connec ed o employees’ expe iences when he changes in ol ed a e no b ough abou by a c isis si ua ion o a e no o he wise adical in na u e. In his pape , we endea o o ake in o accoun all he o - ganiza ional changes ha employees o he Finnish social and heal h ca e indus y ha e encoun e ed a hei wo k- places. Unlike mos p e ious esea ch on o ganiza ional change and employees’ expe iences, ou s is no only o- cused on he connec ion be ween he change and employ- ees’ heal h o mo i a ion. Fi s ly, we a e in e es ed in ind- ing ou whe he employees eel hey a e able o cope wi h hei wo k ega dless o hei po en ial illnesses and o he de ec s. We examine he connec ion be ween he o gani- za ional changes ha he employees encoun e and hei sel -assessed wo k abili y. Sel -es ima ed wo k abili y is in i sel an impo an ma e , bu acco ding o he p e ious esea ch i also indica es he employee’s likelihood o ak- ing sick lea e o e i ing ea ly due o heal h easons [5]. Secondly, we examine he connec ion be ween o gani- za ional changes and wo k abili y sepa a ely, h ough he di e en elemen s a ec ing wo k abili y. As we see i , wo k abili y and inabili y o wo k a e no solely abou he heal h and mo i a ion o he indi idual. Acco ding o Juhani Ilma inen’s heo y, wo k abili y consis s o a) he heal h o he indi idual, b) he p o essional compe ence o he indi idual, c) he alues, a i udes and commi men o he indi idual, and d) elemen s ela ed o wo k [6]. As we see i , wo k abili y is c ea ed by a well-balanced ela ion- ship be ween hese elemen s. In his pape , we aspi e o look beyond wo k abili y in o hese employees’ pe sonal esou ces and he elemen s ela ed o wo k. As we ana- lyze he connec ion be ween o ganiza ional changes and wo k abili y, and he di e en elemen s ela ed o wo k abili y, we pay special a en ion o wo ac o s in addi ion o he o ganiza ional changes hemsel es: he employee’s unde s anding o why he changes we e made, and he ORGANIZATIONAL CHANGE AND WORK ABILITY OF EMPLOYEES ORIGINAL PAPERS IJOMEH 2012;25(2) 153 ha e de e io a ed, in compa ison wi h employees whose wo kplaces ha e no unde gone any o ganiza ional chan- ges. In addi ion o his, we also ask whe he he employees who el ha hey did no unde s and he changes a e a a highe isk o su e ing om diminished wo k abili y, o de e io a ion in he elemen s ela ed o wo k abili y, han hose employees who el ha he changes we e unde - s andable. Thi dly, we ask whe he he employees who el ha hei oppo uni ies o be in ol ed in he change p o- cess had been poo a e a a highe isk o eeling ha hei wo k abili y and elemen s ela ed o i ha e de e io a ed han hose employees who s a ed ha hei oppo uni ies o be in ol ed we e good. METHODS S udy design and pa icipan s We employed a c oss-sec ional su ey o o de o ca y ou he esea ch. We asked he esponden s o iden i y he o ganiza ional changes ha had anspi ed a hei wo k- places o e he p e ious wo yea s, and in i ed hem o e alua e how well hey unde s ood hese changes, as well as o ell us whe he o no hey had had he oppo uni y o be in ol ed in he change p ocess. In addi ion o his, we asked he esponden s o e alua e hei wo k abili y and assess whe he ce ain s a emen s ela ed o he di e en elemen s o wo k abili y we e ue o alse a he ime o he su ey. A sample o 5000 Finnish social and heal h ca e wo ke s was selec ed o ake pa in he s udy om he wo k egis- y o S a is ics Finland. In ea ly 2010, when he ma e ial was collec ed, he egis e a ailable was om 2007. Thus, he people who en e ed he ield du ing he las wo yea s we e le ou o he sample, and all o he employees in he s udy had wo ked in he ield o social and heal h ca e o a leas wo yea s. The me hod we employed was s a i- ied andom sampling. The sample chosen was based on he ield o ope a ion o he wo kplace and geog aphical caused by he wo k democ a iza ion mo emen , has o his day con inued o p oduce a conside able amoun o esea ch li e a u e on he connec ion be ween he g ass oo s le el employees’ oppo uni ies o in luence hei own wo k and wo k mo i a ion, well-being, and heal h [17–21]. Thus, o decades now, we ha e no labo ed unde he alse p e ence ha all c ea i i y lies a he op o o ganiza ions and among consul an s. Howe e , o e he las yea s, he discussion on employee-d i en inno a ions has u he expanded he em- ployee’s ole o de elope o his/he o ganiza ion. In he social and heal h ca e indus y also, he g ass oo s le el em- ployees ha e ecen ly been challenged o inno a e sphe es beyond hei own wo k p ocesses wi hin hei espec i e o ganiza ions [22]. I may e y well be ha he la ge g oup o g ass oo s le el employees ha daily mee he clien s o hei o ganiza ion is an ex emely use ul ool o de eloping hei o ganiza ion in an inc easingly ac i e way. Howe e , un il now, li le a en ion has been paid o he expec a ions ha a e placed on employees in hei new oles as ac i e pa - icipan s in he change p ocess. In social and heal h ca e, he employee’s ole and au ho i y has been adi ionally clea ly ma ked by di e en o ganiza ional le els and he hie a chy exis ing be ween he di e en p o essionals o he indus y. Agains his backg ound, aking he “helicop e pe spec i e”, which is he p e- equisi e o employee-d i en inno a ion, may in p ac ice be e y di icul and a duous o g ass oo s le el employees. In addi ion, he employees o Finnish social and heal h ca e a e unde a g ea deal o s ain caused by la ge numbe s o clien s and he esponsibili y hey eel o hem [23]. I is he e o e also necessa y o examine whe he he oppo uni y o be in ol ed in he change p ocess a ec s he s a ’s wo k abili y in a ca ego ically posi i e way. OBJECTIVE In his pape we ocus on he isk ha employees who ha e expe ienced o ganiza ional change may eel ha hei wo k abili y and he di e en elemen s ela ed o i ORIGINAL PAPERS L. KOKKINEN and A. KONU IJOMEH 2012;25(2) 154 cu en ly unde going any changes ela ed o o ganiza- ional s uc u e o ope a ions. The answe al e na i es we e yes/no. Unde s anding and pa icipa ing in changes Fo ques ions ela ed o he unde s andabili y o he change and oppo uni ies o be in ol ed in he change p ocess, we applied he esea ch by Kinnunen and Lind- s öm on he connec ion be ween s uc u al and ope a- ional changes and he well-being o pe sonnel [14]. We buil ou own ba e y o ques ions based on hei ques- ions, which we e hus scien i ically es ed. To measu e he unde s andabili y o change, we used a g oup o ques ions comp ising ou s a emen s. These s a emen s we e: a) I ha e ecei ed an adequa e amoun o in o ma ion on how he changes will a ec my wo k, b) The easons o he change a e no clea o me (in ollow-up wo k his s a emen was e ised o ma ch he o he s a emen s), c) The goals o he change a e clea o me, d) I ha e been well awa e o he p og ess and di e - en s ages o he change p ocess. The indica o we used o measu e he oppo uni y o be in ol ed in he change p ocess comp ised wo s a emen s: a) S a ha e been su - icien ly in ol ed in he planning o he change, b) I ha e been able o in luence he implemen a ion o change a my wo kplace. The di e en answe s we e gi en on a i e-poin Like scale: (1 = ully ag ee, 2 = ag ee o a poin , 3 = nei he ag ee no disag ee, 4 = disag ee o a poin , 5 = ully dis- ag ee). The o iginal six subques ions we e uni ed o o m wo sum a iables: ‘ he unde s andabili y o he change’ and ‘ he oppo uni y o be in ol ed in he change’. The C onbach’s alpha coe icien s measu ing he in e nal con- sis ency o he sum a iables we e 0.76 and 0.79. P io o he analysis, bo h sum a iables we e dicho omized by combining al e na i e answe s 1 and 2, and al e na- i es [3–5] o o m hei own ca ego ies. loca ion. In 2010, we used he pan-Eu opean classi ica ion o ield o ope a ion (NACE), which was i s in oduced in 2008 and di ides he ield o social and heal h ca e in o 11 di e en sample sec o s. These sample sec o s a e: he social se ice sec o , child en’s day ca e, home ca e, nu sing homes, suppo ed housing, se ices o he disabled, he men al heal h and d ug abuse sec o , ou - pa ien ca e, heal h cen e s wa ds, hospi als, and o he s. The sample o his s udy was collec ed so ha e en om he smalles sample sec ion we ob ained a sample o a leas 200 people. We also ook geog aphical ep esen a- ion in o conside a ion when collec ing he sample. The sample was chosen so ha all 19 p o inces o Finland we e ep esen ed in p opo ion o hei ela i e sizes. The su ey o ms we e mailed o he esponden s oge- he wi h a p e-paid e u n en elope. The esponse a e o he su ey was 61.8%, which means ha a o al o 3090 esponden s answe ed he su ey. Be o e con- duc ing he analysis, 397 esponden s who no longe wo ked in he ield o ope a ion du ing he ime o he su ey, bu who we e chosen in he egis e -based selec- ion, we e emo ed om he da a. We also emo ed 264 esponden s om he da a who con inued o wo k in he ield, bu had no answe ed he ques ions conce n- ing o ganiza ional changes. Thus, 2429 esponden s we e accep ed in o he inal esea ch da a. Ou o hese esponden s, 76% wo ked in he public sec o , 17% in p i a e en e p ises, 3% in non-p o i o ganiza ions ( he so-called hi d sec o ), and 3% we e sel -employed. This dis ibu ion co esponds well o he eal si ua ion in he Finnish social and heal h ca e indus y [24]. MEASURES O ganiza ional change We asked he esponden s whe he any o ganiza ion- al changes had occu ed a hei wo kplaces wi hin he p e ious wo yea s, o i hei wo kplaces we e ORGANIZATIONAL CHANGE AND WORK ABILITY OF EMPLOYEES ORIGINAL PAPERS IJOMEH 2012;25(2) 155 asks co espond well o my cu en abili ies, 3 = I can handle mo e complica ed wo k asks han he ones as- signed o me a he momen ). The p e ious s udies had shown ha wo k asks ha a e app op ia ely challenging; no oo easy, bu no oo demanding ei he , a e connec ed o wo k abili y [32]. Thus, he indica o was dicho omized by combing answe al e na i es 1 and 3 o o m he ‘capa- bili ies do no co espond wi h wo k asks’ ca ego y and al- e na i e 2 became he ‘capabili ies co espond wi h wo k asks’ ca ego y. We measu ed a i udes and commi men by p esen ing he esponden s wi h ou s a emen s. These we e: a) My wo k is ewa ding and I wo k because I enjoy i , b) My cu en job is an essen ial pa o my li e, c) I ega d my p o es- sion highly, d) I am pe sonally commi ed o my cu en wo k. The esponden s es ima ed hei alues, a i udes and commi men on a i e-poin Like scale (1 = ully ag ee, 2 = ag ee o a poin , 3 = nei he ag ee no dis- ag ee, 4 = disag ee o a poin , 5 = ully disag ee). The o iginal ou subques ions we e combined o o m one sum a iable: ‘a i udes and commi men ’. The C onbach’s alpha coe icien measu ing he in e nal consis ency o he sum a iable was 0.79. P io o he analysis, he sum a i- able was dicho omized by combining al e na i e answe s 1 and 2, and al e na i es 3–5 o o m hei own ca ego ies. We examined he elemen s ela ed o wo k acco ding o he Ilma inen’s model. He di ides hese elemen s in o ou sub ac o s: manage ial wo k and leade ship, wo k communi y and o ganiza ion, wo k con en and demands, and wo king condi ions. Fi s ly, manage ial wo k and leade ship we e measu ed by asking he esponden s o e alua e he jus ice o he decisions made a hei wo kplaces. The p e ious s udies had shown ha he employees’ iew on how jus decision- making is, is one o he mos essen ial ac o s de e mining manage ial wo k and leade ship, and is associa ed no only wi h he well-being o s a , bu also wi h he pe o mance o he wo kplace in gene al [33,34]. Fo he ques ions Wo k abili y and elemen s o wo k abili y We asked he esponden s o e alua e hei wo k abili y. Fo he analysis o he ques ion measu ing he espon- den s’ wo k abili y, we u ilized a wo k abili y index de e- loped by Finnish esea che s [25]. We only used he i s and mos impo an ques ion o he index. In he p e io us s udies, his ques ion had e ec i ely p edic ed he ou - come o he wo k abili y index, which comp ises se en subques ions [26]. The exac ques ion posed o he espon- den s is as ollows: Le ’s assume ha you wo k abili y a i s bes is wo h 10 poin s. How many poin s would you gi e you cu en wo k abili y? The esponden s es ima ed hei wo k abili y on an 11-poin scale (0 = en i ely inca- pable o wo king, o 10 = a he heigh o wo k abili y). The wo k abili y es ima ions, as in he p e ious s udies, we e dicho omized by combining answe al e na i es 0–7 and al e na i es 8–10 o o m hei own ca ego ies [27]. Fo he elemen s ha ha e an impac on wo k abili y, we u ilized Juhani Ilma inen’s wo k [6]. Acco ding o his model, wo k abili y consis s o a) he heal h o he indi- idual, b) he p o essional compe ence o he indi idual, c) he alues, a i udes and mo i a ion o he indi idual, and d) elemen s ela ed o wo k. We measu ed he heal h o he esponden s ia hei own es- ima ions, on a i e-poin Like scale (1 = good, 2 = ai ly good, 3 = mode a e, 4 = ai ly poo , 5 = poo ). Sel -e alu- a ion o heal h is one o he mos o en used measu es o heal h, and low sel -e alua ed heal h has been shown o be connec ed o se e al illnesses [28,29]. Simila ly o he p e ious s udies, he indica o o sel -e alua ed s a e o heal h was dicho omized be o e he analysis by combining answe al e na i es 1 and 2, and al e na i es 3–5 o o m hei own ca ego ies [30,31]. To elici he esponden s’ p o essional compe ence, we asked abou hei abili y o pe o m hei wo k asks. The esponden s we e asked o e alua e hei capabili- ies on a h ee-poin scale (1 = I need u he aining o be able o cope well wi h my wo k asks, 2 = My wo k ORIGINAL PAPERS L. KOKKINEN and A. KONU IJOMEH 2012;25(2) 156 Fou hly, we measu ed wo king condi ions h ough he oc- cu ence o ime p essu e. The p e ious s udies had shown ha he p e alence o ime p essu e o en p o ed o be a ac o ha s ained he s a o he social and heal h ca e indus y [24]. We asked he esponden s how o en wi hin he p e ious 12 mon hs hey had been bo he ed, wo ied o s essed o e he ollowing issues: a) Cons an ush and p essu e caused by wo k ha has no been done, b) Inade- qua e numbe o s a , c) Too li le ime o ake b eaks du ing he wo k day. The esponden s ga e hei es ima- ion on a i e-poin Like scale (1 = e y a ely o ne e , 2 = qui e seldom, 3 = e e y now and again, 4 = qui e o en, 5 = e y o en o con inuously). All subques ions we e combined o o m one sum a iable: ‘exposu e o ime p essu e’. The C onbach’s alpha coe icien measu - ing he in e nal consis ency o he sum a iable was 0.84. P io o he analysis, all h ee sum a iables measu ing wo k- ela ed elemen s and he ques ion measu ing he s ess ac o ela ed o obscu i y o job desc ip ion we e dicho omized by combining answe al e na i es 1 and 2, and al e na i es 3–5 o o m hei own ca ego ies. Backg ound cha ac e is ics We mapped ou esponden s’ backg ound cha ac e is ics by asking hem abou hei gende , age, ma i al s a us, p o essional educa ion and whe he o no hey wo ked in a manage ial posi ion. S a is ical Analysis We o med logis ical eg ession models o he eigh p e io- usly desc ibed esponse a iables de ining sel -e alua ed wo k abili y and he di e en elemen s ela ed o i . We di ided ou eg ession models in o h ee g oups acco ding o hei explana o y a iables. In he i s g oup, we examined o ganiza ional changes. In his case, he con ol g oup o he esponden s who epo ed ha ing unde gone changes was o med by he esponden s whose wo kplaces had encoun e ed no measu ing jus ice, we u ilized he wo k by Mooi man [35]. The s a emen s we e as ollows: a) A ou wo kplace, deci- sions a e made based on eal in o ma ion, b) A ou wo k- place, unsuccess ul decisions can be e ac ed o changed, c) All pa ies conce ned a e ep esen ed in decision-mak- ing, d) All decisions made a ou wo kplace a e consis- en , e) E e ybody has he igh o s a e hei opinion on ma e s ha conce n hem, ) The e ec s o decisions a e ollowed up and a e communica ed o he s a , g) Mo e in o ma ion on he g ounds o decisions is a ailable o hose who wan i . The esponden s es ima ed he jus ice o decision-making a hei wo kplace on a i e-poin Lik- e scale (1 = ully ag ee, 2 = ag ee o a poin , 3 = nei he ag ee no disag ee, 4 = disag ee o a poin , 5 = ully dis- ag ee). The se en o iginal subques ions we e combined o o m one sum a iable: ‘ he jus ice o decision-making’. The C onbach’s alpha coe icien measu ing he in e nal consis ency o he sum a iable was 0.91. Secondly, we measu ed he unc ionali y o wo k com- muni y h ough ou s a emen s: a) Ou wo k communi y is lexible, b) Ou wo k communi y ope a es e icien ly, c) The co-ope a ion be ween he di e en membe s o ou wo k communi y wo ks well, d) The di ision o labo in ou wo k communi y is success ul. The esponden s es i- ma ed he unc ionali y o hei wo kplace on a i e-poin Like scale (1 = ully ag ee, 2 = ag ee o a poin , 3 = nei- he ag ee no disag ee, 4 = disag ee o a poin , 5 = ully disag ee). The o iginal ou subques ions we e combined o o m one sum a iable: ‘ unc ionali y o wo k commu- ni y’. The C onbach’s alpha coe icien measu ing he in- e nal consis ency o he sum a iable was 0.85. Thi dly, we measu ed wo k con en s and demands by asking he esponden s how o en wi hin he p e io- us 12 mon hs hey had been bo he ed, wo ied o s essed abou obscu i y in hei job desc ip ion. The esponden s ga e hei es ima ion on a i e-poin Like scale (1 = e y a ely o ne e , 2 = qui e seldom, 3 = e e y now and again, 4 = qui e o en, 5 = e y o en o con inuously). ORGANIZATIONAL CHANGE AND WORK ABILITY OF EMPLOYEES ORIGINAL PAPERS IJOMEH 2012;25(2) 157 was no s a is ically signi ican di e ence be ween any o he g oups in ela ion o any o he backg ound a iables we examined. Table 2 demons a es he isk ha employees who ha e encoun e ed o ganiza ional change may eel ha hei wo k abili y and he di e en elemen s ela ed o i ha e de e io a ed in compa ison o he esponden s whose o - ganiza ions ha e unde gone no changes. This able shows ha he esponden s who had expe ienced o ganiza ional changes wi hin he p e ious wo yea s we e almos one and a hal imes mo e likely o eel ha hei wo k abili y was diminished. The e is, howe e , no signi ican di e ence in e ms o sel -e alua ed s a e o heal h. The isk ha he esponden s’ abili ies do no co espond wi h hei wo k asks was 1.27 imes highe among he esponden s who had expe ienced changes, as was he isk o hem ha ing nega i e a i udes and lack o commi men ha a ec wo k abili y. In compa ison o he esponden s who had no expe ienced changes, he esponden s who had been ouched by changes we e mo e likely o label wo k- ela- ed issues nega i ely. The isk ha an employee el ha : (1) he decisions made had been unjus was 1.76 i mes highe ; (2) he wo k communi y unc ioned ine icien - ly – 1.5 imes highe ; (3) job desc ip ions we e obscu e and caused s ain – 1.42 highe , and (4) hu y caused s ess – mo e han wice as high. Ou o 1659 esponden s who had expe ienced chang- es, 81.9%, o 1359 esponden s, s a ed ha hey did no unde s and he change. As much as 90%, i.e. 1493 espon- den s, es ima ed ha hei oppo uni ies o be in ol ed in he change p ocess we e lacking. The esponden s’ back- g ound cha ac e is ics in ela ion o how well hey unde - s ood he changes and how highly hey anked hei op- po uni ies o be in ol ed in he changes a e p esen ed in Table 3. The esponden s who el hey did no unde s and he changes and he esponden s who el ha hei op- po uni ies o be in ol ed in he change p ocess had been poo we e mo e o en no in a manage ial posi ion, had changes. In he second g oup, we s udied he unde s and- abili y o he changes epo ed by ou esponden s. In his case, he con ol g oup o he esponden s who el ha hey did no unde s and he changes was o med by hose esponden s who el he changes we e unde s andable. In he hi d g oup, we s udied he esponden s’ oppo uni ies o be in ol ed in he change p ocess, and he e he con ol g oup o he esponden s who el hei oppo uni ies o pa icipa e had been poo we e hose esponden s who el ha hey had good oppo uni ies o be in ol ed in he change p ocess. Thus, we o med 24 logis ical eg ession models o s udy he eigh esponse a iables in h ee di e en g oups. Fo ou me hod o analysis, we deployed logis ical eg ession analysis, because i enabled us o simul aneously s udy he dependence o he dicho omized esponse a iables on mo e han one explana o y a iable. Logis ical eg ession analysis does no explain he ealized dis ibu ions, bu a he aims a p edic ing p obabili ies. In his s udy, we used he analysis me hod o s udy he ex en o he isk ha he s udied g oups ( ha we e o med based on ex- posu e) belonged o ce ain o he g oups ( o med based on esponse) when he o he ela ed ac o s we e adjus ed o . Values abo e one deno e ha he isk is highe han among he con ol g oup, and alues below one indica e a isk ha is lowe han among he con ol g oup. In his pape , we adjus ed o gende , age, ma i al s a us, p o es- sional educa ion, and manage ial posi ion in each logis i- cal eg ession model. A o al o 0–1.5% o he answe s pe ques ion was missing. RESULTS Ou o all he 2429 esponden s, 68.3%, ha is 1659 e- sponden s, had encoun e ed o ganiza ional changes wi hin he p e ious wo yea s. Table 1 p esen s he esponden s’ backg ound cha ac e is ics in ela ion o he o ganiza ion- al changes ha had anspi ed a hei wo kplaces. The e ORIGINAL PAPERS L. KOKKINEN and A. KONU IJOMEH 2012;25(2) 158 Table 1. Responden s acco ding o hei backg ound a iables in ela ion o he o ganiza ional changes hey had expe ienced o e he p e ious wo yea s Response a iables O ganiza ional change o e he p e ious wo yea s all no yes Gende male 204 (8.5) 65 (8.4) 139 (8.4) emale 2 225 (91.6) 705 (91.6) 1 520 (91.6) Age (yea s) < 45 929 (38.2) 307 (39.9) 622 (37.5) 45–54 841 (34.6) 251 (32.6) 590 (35.6) > 55 659 (27.1) 212 (27.5) 447 (26.9) Ma i al s a us single 213 (8.8) 68 (8.8) 145 (8.7) cohabi a ing o ma ied 1 887 (77.7) 605 (78.6) 1 282 (77.3) di o ced, sepa a ed o widowed 328 (13.5) 97 (12.6) 231 (13.9) P o essional educa ion no educa ion o oca ional cou se 201 (8.3) 76 (9.9.) 125 (7.6.) oca ional diploma o o he in e media e deg ee 1 492 (61.7) 463 (60.5) 1 029 (62.3) highe oca ional diploma o uni e si y deg ee 724 (30.0) 226 (29.5) 498 (30.1) Manage ial posi ion no 2 204 (92.1) 700 (92.6) 1 504 (91.9) yes 189 (7.9) 56 (7.4) 133 (8.1) Table 2. Risk o esponden s who had expe ienced o ganiza ional changes ha ing dec eased sel -e alua ed wo k abili y and elemen s ela ed o wo k abili y Response a iables Responden s who had unde gone changes (OR) Dec eased sel -e alua ed wo k abili y 1.49*** Sel -e alua ed heal h a e age o poo 1.18 Abili y does no co espond o wo k asks 1.27** A i udes and commi men do no p omo e wo k abili y 1.27* Fac o s ela ed o wo k decisions made a wo kplace a e unjus 1.76*** unc ionali y o wo k communi y de e io a ed 1.54*** obscu e job desc ip ion s ains 1.42*** ime p essu e causes s ess 2.04*** Gende , age, ma i al s a us, p o essional educa ion and manage ial posi ion we e adjus ed o in each logis ic eg ession model. OR – adds a io. p * < 0.05, ** < 0.01, *** < 0.001. ORGANIZATIONAL CHANGE AND WORK ABILITY OF EMPLOYEES ORIGINAL PAPERS IJOMEH 2012;25(2) 159 his/he wo k asks. Howe e , he g oup who did no un- de s and he changes was almos wice as likely o ha e a nega i e a i ude and lack o commi men ha a ec s wo k abili y. The isk ha he esponden s el ha : (1) he decisions made had been unjus was mo e han ou imes as high; (2) he wo k communi y unc ioned ine icien ly and job desc ip ions we e obscu e and caused s ain – mo e han wice as high; and (3) hu y caused s ess – al- mos wice as high. Table 4 also p esen s he isk ha hose esponden s who el ha hei oppo uni ies o be in ol ed had been poo would also eel ha hei wo k abili y and i s di e en ele- men s had de e io a ed, in compa ison o he esponden s who el ha hei oppo uni ies o be in ol ed in he change p ocess had been good. The able shows ha hose been educa ed in a oca ional ins i u e o in ano he in e - media e school, and we e women. Table 4 p esen s he isk ha he esponden s who s a ed hey did no unde s and he changes migh also eel ha hei wo k abili y and he di e en elemen s a ec ing i had de e io a ed, in compa ison o hose esponden s who el ha he changes hey had encoun e ed we e unde - s andable. The able shows ha he esponden s who el ha he changes we e no unde s andable we e almos wice as likely o eel ha hei wo k abili y was poo in compa ison o he esponden s who s a ed ha hey un- de s ood he changes. The isk was mo e han 1.5 imes highe as ega ds sel -e alua ed heal h. The e was no sig- ni ican di e ence be ween he wo g oups ega ding he isk ha he employee’s abili y does no co espond o Table 3. Responden s acco ding o hei backg ound a iables in ela ion o unde s andabili y o o ganiza ional changes and oppo uni ies o be in ol ed in hem Response a iables Change unde s andable Good oppo uni ies o be in ol ed yes no yes no Gende * *** male 35 (11.6) 107 (7.9) 30 (17.9) 112 (7.5) emale 265 (88.4) 1252 (92.1) 136 (82.1) 1,381 (92.5) Age (yea s) < 45 113 (37.8) 514 (37.8) 56 (34) 575 (38.5) 45–54 109 (36.4) 478 (35.2) 64 (38.3) 524 (35.1) > 55 78 (25.8) 367 (27.0) 46 (27.7) 394 (26.4) Ma i al s a us single 23 (7.8) 121 (8.9) 12 (7.4) 131 (8.8.) cohabi a ing o ma ied 234 (77.9) 1,051 (77.4) 123 (74.1) 1,161 (77.8) di o ced, sepa a ed o widowed 43 (14.3) 186 (13.7) 31 (18.5) 200 (13.4) P o essional educa ion ** *** no educa ion o oca ional cou se 28 (9.2) 99 (7.3) 13 (8.1) 113 (7.6) oca ional diploma o o he in e media e deg ee 158 (52.9) 867 (64.1) 80 (48.4) 946 (63.6) highe oca ional diploma o uni e si y deg ee 113 (37.9) 87 (28.6.) 72 (43.5) 428 (28.8) Manage ial posi ion *** *** no 246 (83.2) 1,258 (93.8) 126 (76.7) 1,376 (93.4) yes 50 (16.8) 83 (6.2) 38 (23.3) 97 (6.6) p * < 0.05, ** < 0.01, *** < 0.001.