The person-oriented approach to burnout: A systematic review

Bu nou Resea ch 3 (2016) 11–23 Con en s lis s a ailable a ScienceDi ec Bu nou Resea ch jou nal homepage: www.else ie .com/loca e/bu n The pe son-o ien ed app oach o bu nou : A sys ema ic e iew Anne Mäkikangasa,∗, Ulla Kinnunenb aDepa men o Psychology, Uni e si y o Jy askyla, P.O. Box 35, Jy askyla 40014, Finland bSchool o Social Sciences and Humani ies (Psychology), Uni e si y o Tampe e, Tampe e 33014, Finland a icle in o A icle his o y: Recei ed 6 Oc obe 2015 Recei ed in e ised o m 11 Decembe 2015 Accep ed 12 Decembe 2015 Keywo ds: Bu nou Pe son-o ien ed app oach Sys ema ic e iew abs ac The a iable-o ien ed app oach has domina ed empi ical bu nou esea ch, bu du ing he las 10 yea s a pe son-o ien ed app oach o bu nou has also become common. The aim o his sys ema ic li e a u e e iew was o iden i y, ca ego ize and e alua e he empi ical esea ch o da e ha has adop ed a pe son- o ien ed app oach o bu nou . The esul s o hese s udies we e hen compa ed wi h hose gene a ed by a iable-o ien ed bu nou esea ch. An elec onic sea ch o se en da abases was conduc ed in sp ing 2015. Ini ially 470 publica ions we e iden ified, 24 o which me he selec ion c i e ia. The e iewed a icles we e ca ego ized in o h ee g oups based on hei esea ch a ge (s): (1) in a-indi idual pa e ns o bu nou symp oms (i.e., ypes o bu nou ) (42%), (2) in a-indi idual de elopmen o bu nou o e ime (i.e., bu nou ajec o ies) (33%), and (3) pa e ns o well-being indica o s wi hin indi iduals (i.e., well- being ypes) (33%). The ypical bu nou ypes and ajec o ies iden ified by pe son-o ien ed esea ch we e la gely pa allel wi h he in o ma ion p oduced by a iable-o ien ed esea ch, bu also b ough ou he he e ogenei y o he bu nou expe ience by e ealing a ypical bu nou and well-being ypes and indi idual de elopmen al ajec o ies. The ad an ages, along wi h he challenges, o aking a pe son- o ien ed app oach a e discussed. Based on he s udy designs, me hodologies, and main findings o he e iewed s udies, fi e a enues o u u e pe son-o ien ed bu nou s udies a e p oposed. © 2015 The Au ho s. Published by Else ie GmbH. This is an open access a icle unde he CC BY-NC-ND license (h p://c ea i ecommons.o g/licenses/by-nc-nd/4.0/). Con en s 1. In oduc ion.............................................................................................................................................12 1.1. Pe son e sus a iable-o ien ed app oach o bu nou .........................................................................................12 1.2. Mul idimensionali y o bu nou ................................................................................................................12 1.3. Longi udinal de elopmen o bu nou ..........................................................................................................13 1.4. Rela ionship wi h he o he well-being cons uc s.............................................................................................13 1.5. P esen e iew...................................................................................................................................14 2. Me hod ..................................................................................................................................................14 2.1. S udy selec ion: inclusion and exclusion c i e ia...............................................................................................14 2.2. Analysis..........................................................................................................................................15 3. Resul s...................................................................................................................................................15 3.1. Desc ip i e in o ma ion on he e iewed s udies ..............................................................................................15 3.2. Bu nou ypes ...................................................................................................................................17 3.3. Bu nou ajec o ies .............................................................................................................................17 3.4. Pa e ns o bu nou and o he well-being cons uc s ..........................................................................................18 4. Discussion...............................................................................................................................................18 4.1. Compa ison o pe son- and a iable-o ien ed bu nou esul s ................................................................................18 4.2. Ad an ages and challenges o a pe son-o ien ed app oach ....................................................................................19 4.3. Di ec ions o u u e pe son-o ien ed bu nou esea ch .......................................................................................20 4.4. Limi a ions and con ibu ions...................................................................................................................20 ∗Co esponding au ho . E-mail add ess: anne.makikangas@jyu.fi (A. Mäkikangas). h p://dx.doi.o g/10.1016/j.bu n.2015.12.002 2213-0586/© 2015 The Au ho s. Published by Else ie GmbH. This is an open access a icle unde he CC BY-NC-ND license (h p://c ea i ecommons.o g/licenses/ by-nc-nd/4.0/). 12 A. Mäkikangas, U. Kinnunen / Bu nou Resea ch 3 (2016) 11–23 Acknowledgemen s.....................................................................................................................................21 Appendix A .............................................................................................................................................21 Command lines o he sea ches used in he di e en da abases ......................................................................................21 Discipline-specific da abases................................................................................................................21 Mul idisciplina y da abases.................................................................................................................21 Re e ences ..............................................................................................................................................21 1. In oduc ion The se ious s ess synd ome o bu nou , which ep esen s a wo k- ela ed s a e o ill-being cha ac e ized by he symp oms o exhaus ion, cynicism, and educed p o essional e ficacy (Maslach, Jackson, & Lei e , 1996), has been in ensi ely s udied since i was fi s p esen ed in he psychological li e a u e in he mid-1970s (see Lei e , Bakke , & Maslach, 2014;Schau eli, Lei e , & Maslach, 2009). Mos o he empi ical wo k on bu nou has been a iable-o ien ed, i.e., bu nou symp oms ha e been aken as he uni o he analysis. Typical esea ch ques ions ep esen a i e o his app oach include how s ongly bu nou symp oms co ela e wi h each o he ( o e iews, see Kim & Ji, 2009;Wo ley, Vassa , Wheele , & Ba nes, 2008) and whe he bu nou is s able o e ime ei he ac oss he whole s udy popula ion o in ce ain p edefined subg oups ( o e iews, see Schau eli & Enzmann, 1998;Ta is, Le Blanc, Schau eli, & Sch eu s, 2005). Howe e , du ing he ea ly wen y-fi s cen u y, bu nou has alsobeen in es iga ed om ape son-o ien ed (also e med pe son- cen e ed) pe spec i e in which he indi idual is aken as he uni o analysis (Be gman & Lundh, 2015;Be gman, Magnusson, & El- Khou i, 2003). In he con ex o bu nou , his app oach is able o e eal in a-indi idual he e ogenei y in he bu nou synd ome and i sde elopmen o e ime.Mo e specifically, hismeansiden i ying po en ial ypes o pa e ns o bu nou symp oms wi hin indi idu- als and indi idual de elopmen al ajec o ies. This app oach also makes i possible o explo e he dis inc ion be ween bu nou and o he (job- ela ed) well-being a iables a he wi hin-pe son le el. This sys ema ic e iew was unde aken wi h he aim o iden i y- ing, ca ego izing and e alua ing he bu nou s udies conduc ed o da e ha ha eapplieda pe son-o ien ed app oach.Thei esul sa e hencompa edwi h hosegene a edby a iable-o ien ed esea ch. The e iew is o ganized and p esen ed in ou pa s. Fi s , he heo e ical and me hodological ace s o bo h he pe son- and a iable-o ien ed app oaches a e in oduced. Second, o compa - ison pu poses, a b ie o e iew o he exis ing knowledge on he mul idimensionali y o bu nou , i s long- e m de elopmen , and ela ionship wi h o he (job- ela ed) well-being cons uc s based on a iable-o ien ed app oach is p o ided. Thi d, he a ailable pe son-o ien ed bu nou s udies a e e iewed and e alua ed, and compa ed wi h he exis ing a iable-o ien ed findings. Finally, he ad an ages, along wi h he challenges, o aking a pe son-o ien ed app oach a e discussed, and di ec ions o u u e pe son-o ien ed bu nou esea ch sugges ed. 1.1. Pe son e sus a iable-o ien ed app oach o bu nou The mode n pe son-o ien ed app oach was in oduced by Block (1971) in he con ex o pe sonali y psychology, and u he elabo a ed in he milieu o Swedish de elopmen al psy- chology (Be gman e al., 2003; Magnusson, 1988). The pe son- and a iable-o ien ed app oaches di e undamen ally bo h heo e ically and me hodologically (Be gman & Lundh, 2015). Theo e ically, he pe son-o ien ed app oach is based on a holis ic- in e ac ionis ic pe spec i e on human de elopmen (Magnusson, 1988; Magnusson & Tö es ad, 1993), whe e he indi idual is seen as an o ganized whole, ha is, as he sum o in e ac i e componen s ope a ing oge he (Be gman & Lundh, 2015). Acco dingly, he pe son-o ien ed app oach aims o disco e he configu a ions o ac o s ha cha ac e ize a specific indi idual’s unc ioning(Magnusson,1988).Roo edinphilosophicalposi i ism, a a iable-o ien ed app oach pu sues he sea ch o gene alizable laws in line wi h he ideal o he na u al sciences and ega ds he a iable as a cen al concep ual and analy ical uni (see Be gman & Lundh, 2015). On he me hodological le el, bo h hese app oaches use quan i- a i e da a. The basic di e ence is ha pe son-o ien ed me hods aim a in es iga ing how a iables g oup wi hin indi iduals, whe eas he main ocus o in e es in a iable-o ien ed me hods is he ela ions be ween a iables (Be gman e al., 2003; Lau sen & Ho , 2006). Va iable-o ien ed me hods o analysis ocus on in e - indi idual a ia ion and ypically cap u e he (linea ) associa ions be ween a iables. Co ela ion- and eg ession-based me hods o analysis as well as mean le el compa isons ia analysis o a iance a e p o o ypically used in his app oach. The basic assump ion o hese s a is ical me hods is ha a popula ion is homogeneous wi h espec o he s udied phenomena, o example, a co ela ion pa - e n o de elopmen o e ime is assumed o be simila ac oss he whole s udy popula ion (Lau sen & Ho , 2006). Howe e , pe son-o ien ed analy ic me hods, p o o ypically class and clus e analysis, a e p edica ed on he assump ion ha he popula ion is he e ogeneous in e ms o he mean le els o and changes in he s udied phenomenon (Lau sen & Ho , 2006). Com- mon o he s a is ical me hods used in a pe son-o ien ed app oach is ha he numbe o classes is unknown and ha di e en class solu ions a e o med and compa ed based on s a is ical and he- o e ical conside a ions (Be gman e al., 2003). Thus, classes a e no o med on he basis o p edefined alues (e.g., cu -o alues o g oup means). Pe son-o ien ed analyses a e ypically used o wo pu poses, ha is, o iden i y ypes o indi iduals and indi- idual ajec o ies (Lau sen & Ho , 2006). In bo h cases, ypical and a ypical ypes and ajec o ies can be iden ified. The e o e, he pu pose is no ha e e y pe son o ms his/he own ype o ajec- o y; ins ead he in e es is in how indi iduals a e simila and how hey a e di e en om o he s and in wha espec s (Be gman e al., 2003). In bu nou esea ch, a pe son-o ien ed app oach as desc ibed abo e has he po en ial o p o ide answe s o, o a leas com- plemen he exis ing knowledge on, h ee key ques ions: he mul idimensionali y o bu nou (Schau eli & Ta is, 2005), long- e mde elopmen o bu nou (Schau eli,Maassen, Bakke , & Sixma, 2011) and i s ela ionship wi h o he (job- ela ed) well-being con- s uc s (Maslach, 2011; Schau eli & Enzmann, 1998). Nex , hese h ee ace s a e b iefly discussed on he basis o a iable-o ien ed bu nou esea ch. 1.2. Mul idimensionali y o bu nou Se e al concep ual and ope a ional defini ions, wi h a ying symp omology, ha e been p esen ed o job bu nou . Ne e he- less, he defini ion by Maslach e al. (1996) is mos o en used and widely accep ed in he psychological li e a u e. Acco ding o his h ee-dimensional defini ion,bu nou isa pe sis en synd ome cha ac e ized by exhaus ion, cynicism, and educed p o essional A. Mäkikangas, U. Kinnunen / Bu nou Resea ch 3 (2016) 11–23 13 e ficacy. Exhaus ion e e s o he d aining o emo ional esou ces, eelings o i edness, and ch onic a igue esul ing om wo k o e - load. Cynicism e e s o dis ancing onesel om one’s wo k and o he de elopmen o nega i e a i udes owa ds wo k. Reduced p o essional e ficacy is desc ibed by loss o compe ence and p o- duc i i y, and he endency o e alua e one’s pas and p esen accomplishmen s a wo k nega i ely. Al hough gene al ag eemen exis s on hese h ee symp oms, hei independency s. depen- dency has been deba ed (see Maslach, Schau eli, & Lei e , 2001; Shi om, Melamed, Toke , Be line , & Shapi a, 2005). Va iable-o ien ed s udies ha e indica ed ha exhaus ion and cynicism a e mo e s ongly associa ed wi h each o he han wi h educed p o essional e ficacy ( o e iews, see Kim & Ji, 2009; Wo ley e al., 2008). This weake associa ion o educed p o- essional e ficacy wi h he o he wo symp oms has e en led o ques ioning whe he educed p o essional e ficacy is in ac a cen al symp om o bu nou . Fo he same eason, exhaus- ion and cynicism a e o en seen as co e symp oms o job bu nou . Schau eli and Salano a (2007) ha e sugges ed ha his may be an e ec o di e ences in he wo ding o he scale i ems, i.e., exhaus ion and cynicism a e wo ded nega i ely and educed p o essional e ficacy posi i ely. Ne e heless, Mäkikangas, Hä inen, Kinnunen, and Pekkonen (2011) ha e demons a ed ha educed p o essional e ficacy co ela es mo e s ongly wi h he o he wo symp oms among bu ned-ou han non-bu ned-ou indi iduals. Hence, he di e ing associa ions be ween he h ee symp oms o bu nou may eflec hei di e en de elopmen al o de (see Ta is e al., 2005). Tha is, he bu nou p ocess is assumed o ollow specific s ages s a ing om exhaus ion and ending in educed p o- essional e ficacy (Lei e & Maslach, 1988). In ligh o his iew, i may be ha educed p o essional e ficacy has no been mani es ed as pa o he bu nou synd ome simply because he majo i y o he p e ious esea ch samples ha e consis ed o ela i ely heal hy wo ke s (see Kim & Ji, 2009; Wo ley e al., 2008). Howe e , al e na- i e de elopmen al o de s o bu nou symp oms ha e been also p oposed (e.g., Golembiewski, Munzen ide , & S e enson, 1986). Fo example, in he model o Golembiewski e al. (1986) he de el- opmen o exhaus ion is he final phase. The ela ionship be ween he h ee bu nou symp oms can be pa ially cla ified by using a pe son-o ien ed me hodology, which e eals how he symp oms combine oge he a he in a- indi idual le el by o ming di e en bu nou ypes o pa e ns. Such bu nou ypes ha e he po en ial o e eal mo e abou he ela ionship be ween he bu nou symp oms han me e co ela ion coe ficien s. Fo example, al hough bu nou symp oms, especially exhaus ion and cynicism, co ela e highly, i is possible ha some indi iduals only su e om some o hese symp oms. In addi ion, e enin he caseo se e e bu nou , he meanle els o he symp oms may s ill a y ac oss indi iduals, he eby e ealing indi idualized expe iences. 1.3. Longi udinal de elopmen o bu nou Despi e decades o bu nou esea ch, he empo al and de elop- men al aspec s o bu nou emain con o e sial (see Schau eli e al., 2011). O iginally, bu nou was heo ized as a dynamic ongoing p o- cess in ol ing employees’ psychological esponses o un ea ed long-las ing wo k s ess (Maslach, 1982). Nowadays, based on a iable-o ien ed s udies, i ends o be concei ed as an endu ing and ch onic s a e. The ank-o de s abili ies o bu nou ha e been s udied o e di e ing ime anges, ypically a six-mon h o one- yea in e als o e wo o h ee yea s ( o a e iew, see Ta is e al., 2005), bu s udies wi h up o h ee- o fi e-yea in e als o e 10- yea ollow-ups ha e also been epo ed (Schau eli e al., 2011). These s udies, ega dless o he in e al be ween measu emen s, ha e ne e heless ound a he simila ank-o de s abili ies: he a e age ank-o de es ima e o bu nou was .56 o e six mon hs and .57 o e one yea ( o a e iew, see Mäkikangas, Kinnunen, Feld , & Schau eli, in p ess). When longe ime pe iods we e used, o example fi e (Schau eli e al., 2011) o eigh yea s (Toppinen- Tanne , Kalimo, & Mu anen, 2002), he s abili y coe ficien s we e ound o be oughly 60. These findings indica e ha om a ank-o de s abili y pe - spec i e (i.e., he ex en o which he o de o indi iduals emains simila o e ime)bu nou seems obea ela i ely con inuous s a e. Howe e , on he basis o he s abili ies ob ained, i is clea ha he ini ial measu emen o bu nou explained unde hal o he a i- ance o he subsequen measu emen . The e o e, he e is oom o change o e ime. In addi ion, i should be no ed ha ank-o de s abili y es ima es do no e eal wha happens a he mean le el; ha is, despi e high ank-o de s abili y, he mean le el may ne - e heless inc ease o dec ease o e ime (see Mäkikangas e al., in p ess). In o de o ob ain a de ailed unde s anding o he de elopmen o bu nou o e ime, absolu e s abili y (i.e., he ex en o which bu nou sco es change o e ime) and ipsa i e s abili y (i.e., he deg ee o con inui y in bu nou pa e ns) also need o be in es i- ga ed along wi h ank-o de s abili y (see Caspi & Robe s, 1999). Pe son-o ien ed me hods o analysis, by e ealing bo h absolu e and ipsa i e s abili ies, enable indi idual de elopmen al ajec o- iesand he ajec o ies o s abili yandchange o di e en bu nou ypes o be b ough o ligh . 1.4. Rela ionship wi h he o he well-being cons uc s The ela ionship be ween bu nou and o he (job- ela ed) well- being cons uc s has been a opic o esea ch om he beginning o he academic concep ualiza ion o bu nou . The fi s – s ill deba ed issue – is he ela ion be ween bu nou and dep es- sion (Lei e & Du up, 1994; Schau eli & Enzmann, 1998) due o hei simila symp omology and high co-occu ence (Ahola & Hakanen, 2014;Bianchi, Bo y, Hing ay, T ucho , & Lau en , 2013). C oss-sec ional s udies ha e confi med a ela i ely high co ela ion (i.e., ypically abo e .60) be ween dep essi e symp- oms and bu nou ( o e iews, see Bianchi, Schon eld, & Lau en , 2015a;Glass & McKnigh , 1996). In addi ion, longi udinal s ud- ies ha e shown bo h a ecip ocal ela ion be ween dep ession and bu nou (Ahola & Hakanen, 2007) and a p edic i e ela ion om bu nou o dep ession (Hakanen & Schau eli, 2012). How- e e , he e a e also s udies ha ha e ailed o find his p edic i e ela ion (Bianchi, Schon eld, & Lau en , 2015c). Al hough bu nou and dep ession ha e been di e en ia ed a he s a is ical and con- ex ual le els, he dis inc ion be ween bu nou and dep ession is agile as se e e bu nou di e s only sligh ly om clinical dep es- sion (see Bianchi e al., 2015a). The e o e, ag eemen on whe he bu nou and dep ession a e dis inc cons uc s has no ye been eached. Du ing ecen decades, bu nou esea che s ha e paid inc eas- ing a en ion o he ela ion be ween bu nou and i s opposi e, namely wo k engagemen (Maslach & Lei e , 1997;Schau eli, Salano a, González-Romá, & Bakke , 2002). Wo k engagemen is composed o he dimensions o ene gy ( igo ), in ol emen (ded- ica ion) and e ficacy, which a e seen as he opposi es o he h ee symp oms o bu nou (Maslach & Lei e , 1997). Howe e , in Eu o- pean esea ch, he hi d dimension o wo k engagemen is no e ficacy bu abso p ion (Schau eli e al., 2002). This iew o bu nou as he e osion o engagemen (Maslach & Lei e , 1997; Schau eli e al., 2009) has gained suppo om a iable-o ien ed s udies: in wo me a-analyses (see Cole, Wal e , Bedeian, & O ´ Boyle, 2012; C aw o d, LePine, & Rich, 2010), bu nou and wo k engagemen showed a nega i e co ela ion ( =−.55 and −.48). Howe e , co e- 14 A. Mäkikangas, U. Kinnunen / Bu nou Resea ch 3 (2016) 11–23 la ions ha e also a ied widely, om −.20 o −.90 (see Mäkikangas, Feld , Kinnunen, & Tol anen, 2012). These a ia ions be ween s ud- ies challenge he idea ha bu nou and wo k engagemen a e opposi e ends o he same con inuum and ins ead sugges ha bu nou and wo k engagemen can occu simul aneously, a leas among some employees. Fu he mo e, ecen esea ch p oposes ha wo bipola dimen- sions unde lie bo h job bu nou and engagemen : ene gy, comp ising he exhaus ion- igo con inuum, and iden ifica ion, comp ising he cynicism-in ol emen con inuum (Deme ou i, Mos e , & Bakke , 2010;González-Romá, Schau eli, Bakke , & Llo e , 2006). The iden ifica ion con inuum has ecei ed mo e suppo han he ene gy con inuum (Deme ou i e al., 2010), which u he con uses unde s anding o he ela ionship be ween engagemen and bu nou and whe he hei co-occu ence is possible. To conclude, he mixed esul s and a ying co ela ion pa - e ns p esen ed abo e can be a ibu ed o he exis ence o se e al bu nou —(job- ela ed) well-being subg oups, which a iable- o ien ed s udies end o igno e. The ela ionship be ween bu nou and o he (job- ela ed) well-being indica o s can be cla ified by aking in o conside a ion he exis ence o well-being sub ypes by iden i ying in a-indi idual pa e ns o well-being a iables. In addi ion o dep ession and wo k engagemen , wo o he employee well-being s a es ha e been associa ed wi h bu nou , and may he e o e co-exis wi h i , namely wo kaholism (Schau eli, Ta is, & an Rhenen, 2008) and job dissa is ac ion (Maslach & Schau eli, 1993). These a e also aken in o accoun in his e iew. Bu nou , wo k engagemen , job sa is ac ion and wo ka- holism comp ise he ou quad an s o he ci cumplex model o emo ions applied in he wo k con ex (Bakke & Oe lemans, 2011). Thus, a he indi idual le el, he ela ion be ween bu nou and he o he h ee job- ela ed well-being s a es is heo e ically meaning ul. 1.5. P esen e iew The aim o he p esen e iew was o find, e alua e and syn he- size he e idence on quan i a i e job bu nou yielded by adop ing a pe son-o ien ed me hodology. This e iew is bo h quali a i e and sys ema ic. A quali a i e e iew o esea ch is especially beneficial in e iewing a body o ela i ely sca ce, bu g owing li e a u e on a specific opic (Su i & Cla ke, 2009). A sys ema ic quali a i e e iew u he mo e pe mi s he inclusion o di e en kinds o s udies, unlike o example,a me a-analysiswhich onlyallows heinclusion o simila s udies (G an & Boo h, 2009). The e o e, he p esen sys- ema icquali a i e e iew enables examina ion o he di e sea ay o pe son-o ien ed s udies and hei con ibu ion o he bu nou li e a u e. The esea ch ques ions posed in he e iew we e: (1) o wha specific esea ch pu poses has a pe son-o ien ed app oach been applied in bu nou esea ch? (2) Wha a e he main findings o pe son-o ien ed bu nou esea ch? (3) How does bu nou knowl- edge p oduced by pe son-o ien ed esea ch combine wi h ha p oduced by a iable-o ien ed esea ch? To answe hese ques- ions, he pe son-o ien ed s udies a e fi s ca ego ized, a e which hei findings a e syn he ized and e alua ed, and, finally, compa ed wi h hose o he a iable-o ien ed s udies. 2. Me hod 2.1. S udy selec ion: inclusion and exclusion c i e ia S udyselec ionand epo ing was conduc edinacco dancewi h he PRISMA guidelines (Mohe , Libe a i, Te zla , & Al man, 2009). A sys ema ic elec onic sea ch was conduc ed o find all he a i- cles aking a pe son-o ien ed app oach o bu nou . The elec onic sea ch was conduc ed in se en di e en da abases in Ap il–May 2015. An ale o new a icles mee ing he sea ch c i e ia was se o un un il he end o May 2015. The sea ched da abases com- p ised h ee discipline-specific (PsycIn o, Medline, ABI In o m) and ou mul idisciplina y da abases (Academic Sea ch Eli e EBSCO, P oQues , Science Di ec and Google Schola ). The comp ehen- si e sea ch s a egies included sea ch e ms pe aining o bo h bu nou (bu nou , exhaus ion, cynicism, educed pe sonal e ficacy) and pe son-o ien ed esea ch and i s me hodology (e.g., pe son- o ien ed, pe son-cen e ed, clus e analysis, p ofile analysis). The sea ch e ms could be ound in keywo ds, i les, abs ac s o subjec headings. The command lines o he sea ches used in he di e en da abases a e p esen ed in he Appendix A. A icles we e selec ed in h ee phases. Fi s , po en ial s udies yielded by he sea ch we e chosen on he basis o hei i le and abs ac , using he inclusion c i e ia. Second, he whole ex s we e e ie ed and ead using he exclusion c i e ia. Thi d, a manual sea ch was conduc ed by checking e e ence lis s and consul ing expe s o each all he po en ial a icles. These phases, ollowing he PRISMA flow diag am guidelines, a e demons a ed in Fig. 1 (Mohe e al., 2009). In he fi s phase, he ini ial elec onic sea ch yielded a o al o 589 a icles. A e emo al o duplica es, he o al a icle coun was 470. The a icles we e equi ed o mee ou inclusion c i e ia. Fi s , he chosen s udies we e quan i a i e empi ical a icles w i en in English. Second, he s udy employed a pe son-o ien ed me hod o analysis. Thi d, he s udy pa icipan s we e g ouped acco ding o bu nou o one o i s symp oms. Fou h, s udies ha g ouped pa icipan s by using bu nou (o one i s symp oms), along wi h ei he dep ession o o he job- ela ed a ec i e well-being con- s uc s based on he ci cumplex model (i.e., wo k engagemen , job sa is ac ion, and wo kaholism) (Bakke & Oe lemans, 2011), we e included. No es ic ions we e se on he publica ion da e, sample o bu nou measu e used, as he aim o he e iew was o ob ain an ex ensi e and di e se iew o he pe son-o ien ed app oach o bu nou esea ch. The i les and abs ac s o hese 470 a icles we e sc eened agains he inclusion c i e ia abo e, yielding al oge he 75 a icles. In he second phase, he ull ex s o hese 75 a icles we e e iewed by wo independen e iewe s ( he fi s au ho and a esea ch assis an ). In his phase, he ollowing exclusion c i e ia we e used. Fi s , s udies whe e bu nou g oups we e o med on he basis o p edefined c i e ia we e excluded (i.e., high/low g oups o bu nou based on medians, mean bu nou sco es o o he cu -o alues). Second, s udies ha u ilized p ede e mined g oups based on ano he a iable(s) han bu nou we e also elimina ed. Thi d, s udies whe e bu nou was one g ouping a iable among a iables ha we e o he han (job- ela ed) well-being indica o s co e ed by he ci cumplex model we e excluded. Fou h, s udies whe e he bu nou i ems we e exp essed as a sum sco e oge he wi h o he well-being i ems we e excluded, since i was no possible in hese s udies o e alua e he unique impac o bu nou as a g ouping ac o and i s ela ion wi h o he cons uc s. Applica ion o hese ou exclusion c i e ia on he emaining 75 a icles u he educed he a icle coun o 21. The consensus a e be ween he e iewe s was 98.6%. Dispa i ies in judgmen we e esol ed by discussion, esul ing in ull consensus on he inclusion and exclusion o 21 s udies. In he hi d phase, a manual sea ch was conduc ed in o de o check whe he he sea ch p ocess had ound all he ele an a icles. The e e ence lis s o he a icles we e also c osschecked and expe s con ac ed o ensu e ha all he ele an a icles had been ound.Theseaddi ionalsea chesyielded h ee u he a icles, making a g and o al o 24 e iewable a icles. A. Mäkikangas, U. Kinnunen / Bu nou Resea ch 3 (2016) 11–23 15 Fig. 1. S ages o s udy selec ion. 2.2. Analysis The e iewed pe son-o ien ed bu nou a icles we e ca ego- ized based on hei esea ch pu poses in o h ee ca ego ies: (1) in a-indi idualpa e nso bu nou symp oms,(2)in a-indi idual de elopmen o bu nou o e ime, and (3) pa e ns o well-being indica o s wi hin indi iduals. The fi s ca ego y, “in a-indi idual pa e ns o bu nou symp oms”, in es iga es he di e en ways he bu nou symp oms o exhaus ion, cynicism and educed p o es- sional e ficacy a e combined in indi iduals o o m di e en ypes o bu nou . The second ca ego y, “in a-indi idual de elopmen o e ime”, examines s abili y and/o change in bu nou o some o i s symp oms and maps ou possible ajec o ies in he de el- opmen o bu nou . The hi d ca ego y, “pa e ns o well-being indica o s”, s udies he di e en ypes o well-being measu ed by bu nou (o one o i s symp oms), dep ession and h ee o he job- ela ed well-being indica o (s): wo k engagemen , job sa is- ac ion and wo kaholism. Thus his ca ego y examines he possible co-occu ence o o e lap be ween bu nou and hese ou o he well-being s a es. Table 1 shows he ca ego iza ion o each s udy. The Table also summa izes he e iewed s udies, p o iding in o ma ion on (1) he desc ip ion o he da a (pa icipan s, sample size used ei he in analysis o a baseline, age, gende dis ibu ion), (2) he bu nou scale used, and (3) he pe son-o ien ed me hod o analysis applied. The esul s sec ion summa izes and e alua es he main esul s o each ca ego y in he abo e o de o p esen a ion. 3. Resul s 3.1. Desc ip i e in o ma ion on he e iewed s udies The e iewed s udies (ma ked wi h * in he e e ences) had appea ed in 20 jou nals and we e published du ing he pe iod 2004–2015 (Table 1). Job bu nou was in es iga ed in 19 and school bu nou in ou s udies, and one s udy ocused on a hle e bu nou (Isoa d-Gau heu , Guille -Descas, & Duda, 2013). The majo i y o he s udies (10 ou o 24, 42%) in es iga ed bu nou ypes based on in a-indi idual pa e ns o bu nou symp oms, and one- hi d ocusedonei he bu nou ajec o ies(8ou o 24)o on he ela ion be ween bu nou and o he well-being cons uc s by in es iga ing di e en ypes o well-being s a es (8 ou o 24). In wo s udies, wo o hese ca ego ies we e simul aneously in ocus (see Table 1) (Boe sma & Lindblom, 2009; Mäkikangas e al., 2012). O e hal o he s udies (58%) used samples collec ed in one o wo No dic coun ies, namely Finland and Sweden, and only fi e s udies had been conduc ed ou side Eu ope. O he analyzed job bu nou a icles, he majo i y (12 ou o 24) used occupa ion- o o ganiza ion-specific samples, whe eas he e ogeneous samples comp ising employees om a ious occupa ions we e used in ou s udies. In h ee s udies, in e en ion da a we e u ilized. Hal o he s udies used some e sion o he Maslach Bu nou In en o y (MBI) (i.e., ei he GS, ES, HS) (see Maslach e al., 1996). An occupa ion- ailo ed bu nou scale was used in wo s udies (Deme ou i, Ve beke, & Bakke , 2005;Lee, Cho, Kissinge , & Ogle, 16 A. Mäkikangas, U. Kinnunen / Bu nou Resea ch 3 (2016) 11–23 Table 1 Summa y o he pe son-o ien ed bu nou s udies used in he li e a u e e iew in alphabe ical o de (N=24). Au ho s Desc ip ion o da a Bu nou scale Pe son-o ien ed me hod o analysis Ca ego iza ion o he s udiesb Ahola e al. (2014) Finnish den is s (n=1964), mean age 44.4 (SD= 7.9), 76% women MBI human se ice su ey; o al bu nou aLa en class analysis 3 G ow h mix u e modeling Bianchi e al. (2015b) F ench school eache s (n =627), mean age 41 (SD= 9), 73% women MBI gene al su ey; emo ional exhaus ion, depe sonaliza ion; o al sco e Clus e analysis 3 Boe sma and Lindblom (2009) He e ogeneous sample o Swedish employees (n=1118), mean age 43 (SD= 10.8), 52% women MBI gene al su ey; exhaus ion, cynicism, p o essional e ficacy Clus e analysis 1, 2 B udnik (2004) Polish eache s and s uden s o physical aining (n=256), mean age 36.9 ( ange= 23–57), 56% women MBI educa o s su ey; emo ional exhaus ion, depe sonaliza ion, pe sonal accomplishmen Clus e analysis 1 Deme ou i e al. (2005) Th ee subsamples o Du ch accoun manage s (n=616), mean age 35.7–37.4 (SD= 9.0–10.4), 78.4–96% men Face bu nou scale (Singh, Goolsby, & Rhoads, 1994) based on MBI; emo ional exhaus ion, depe sonaliza ion, pe sonal accomplishmen Clus e analysis 1 E olah i e al. (2013) Swedish middle-aged women employees (n=116), age ange 49–53c SMBQ (Shi om-Melamed Bu nou Ques ionnai e); o al bu nou a Clus e analysis 2 Hul ell e al. (2013) Swedish eache s (n=816), mean age 34.95 (SD= 7.87), 85% women SWEBO (Scale o Wo k Engagemen and Bu nou ); o al bu nou a Clus e analysis 2 Hä inen e al. (2009) Finnish ehabili a ion clien s (n=85), agec, gende d MBI gene al su ey; exhaus ion, cynicism, educed p o essional e ficacy G ow h mix u e modeling 2 Hä inen e al. (2004) Finnish ehabili a ion clien s (n=135), mean age 51.2 (SD= 5.5), 60% women MBI gene al su ey; exhaus ion, cynicism, educed p o essional e ficacy Clus e analysis 1 Hä inen e al. (2013) Finnish ehabili a ion clien s (n=85), mean age 49.34 (SD= 6.61), 75.3% women MBI gene al su ey, exhaus ion, cynicism, p o essional e ficacy La en p ofile analysis 2 Innanen e al. (2014) He e ogeneous sample o Finnish employees (n=161), age ange 32–39c, 75% women MBI gene al su ey; exhaus ion, cynicism, p o essional e ficacy La en p ofile analysis 3 Isoa d-Gau heu e al. (2013) F ench handball playe s (n=309)e, mean age 15.4 (SD= 0.9), 50.8% women ABQ (A hele Bu nou Ques ionnai e); physical and emo ional exhaus ion, spo de alua ion, educed sense o accomplishmen Clus e analysis 1 Lee e al. (2010) Ame ican counselo s (n=132), mean age 46.2 (SD= 11.37), 83.3% women CBI (Counselo Bu nou In en o y); exhaus ion, incompe ence, nega i e wo k en i onmen , de aluing clien , de e io a ion in pe sonal li e Clus e analysis 1 Lee e al. (2010) Ko ean middle and high school s uden s (n=338), agec, 52.7% women MBI s uden su ey; emo ional exhaus ion, cynicism, academic e ficacy Clus e analysis 1 Loo (2004) Canadian male police manage s (n=135), agec MBI human se ices su ey; emo ional exhaus ion, depe sonaliza ion, pe sonal accomplishmen Clus e analysis 1 Mäkikangas e al. (2012) Finnish manage s (n=433), mean age 31 ( ange 24–36), 83% men BBI-15 (Be gen Bu nou Indica o ); exhaus ion, cynicism G ow h mix u e modeling 2, 3 Mäkikangas e al. (2014) Finnish heal h and social ca e, and se ice wo ke s (n=256), mean age 42.91 (SD= 10.8), 90% women MBI gene al su ey; exhaus ion G ow h mix u e modeling 3 Mäkikangas e al. (2015) Finnish 50-yea -old employees (n=183), 50.8% men MBI gene al su ey; exhaus ion La en p ofile analysis 3 Rudman and Gus a sson (2011) Swedish nu ses (n=997), mean age 30.5 ( ange= 21–52), 89% women OLBI (Oldenbu g bu nou in en o y); o al bu nou a Clus e analysis 2 Salmela-A o and Upadyaya (2014) Two samples o Finnish adolescen s (n=614/575)e, mean ages 15–17.05 (SD= 0.34–0.27), 52.9% men/gende d SBI (School Bu nou In en o y), o al bu nou a G ow h mix u e modeling 2 A. Mäkikangas, U. Kinnunen / Bu nou Resea ch 3 (2016) 11–23 17 Table 1 (Con inued) Au ho s Desc ip ion o da a Bu nou scale Pe son-o ien ed me hod o analysis Ca ego iza ion o he s udiesb Timms e al. (2012) Aus alian school employees (n=953), mean age 45–49c, 73.5% women OLBI (Oldenbu g Bu nou In en o y); exhaus ion, disengagemen Clus e analysis 3 Tuominen-Soini and Salmela-A o (2014) Finnish high school s uden s and young adul s (n=979)e, mean age 18.14 (SD= 1.11), 60% women SBI (School Bu nou In en o y); exhaus ion, cynicism, inadequacy La en p ofile analysis 3 Zhang e al. (2013) Chinese middle and seconda y school s uden s (n=730), agec, 44% women MBI s uden su ey; exhaus ion, cynicism, e ficacy Clus e analysis 1 Önde and Basim (2008) Tu kish nu ses (n=248), mean age 31.18 ( ange= 19–50), gende d MBI human se ices su ey; emo ional exhaus ion, depe sonaliza ion, pe sonal accomplishmen Clus e analysis 1 aTo al bu nou indica es sum squa ed o all bu nou dimensions. bCa ego ies: 1 = In a-indi idual pa e ns o bu nou dimensions; 2 = In a-indi idual de elopmen o bu nou o e ime; 3 = C oss-sec ional o longi udinal pa e ns o well-being indica o s. cMean age o i s s anda d de ia ion no p o ided. dGende dis ibu ion no p o ided. eBaseline sample size. 2010), while he es used a ious measu es, such as he OLBI (Old- enbu g Bu nou In en o y) o BBI-15 (Be gen Bu nou Indica o ). The majo i y o he s udies used clus e analysis (15 ou o 24, 62.5%), and he emainde mo e ad anced fini e mix u e models such as la en p ofile analysis o g ow h mix u e modeling. 3.2. Bu nou ypes Indi idual pa e ns o bu nou o med by i s symp oms (i.e., bu nou ypes) we e in es iga ed in en s udies (see Table 1). In a non-wo k con ex , ypes o bu nou we e explo ed in wo s ud- ies among Asian s uden s (Lee, Puig, Kim, Shin, Lee, & Lee, 2010; Zhang,Klassen, &Wang, 2013)and adolescen -agea hle es (Isoa d- Gau heu e al., 2013). Job bu nou ypes we e iden ified in six s udies (Boe sma & Lindblom, 2009; B udnik, 2004; Deme ou i e al., 2005; Lee, Cho e al., 2010; Loo, 2004; Önde & Basim, 2008), and in one s udy ypes we e explo ed among ehabili a ion clien s (Hä inen, Kinnunen, Pekkonen, & A o, 2004). The e iewed en s udies iden ified a a ying numbe – h ee o six – o bu nou ypes. Howe e , o hese, h ee (n=3) and ou (n=4)bu nou - ypesolu ionswe e he mos ypically epo ed.The mos ypically ound ypes we echa ac e izedbyei he low (n=10) o highle els(n= 7)o all hebu nou symp oms, ha is,lowo high le els o exhaus ion, cynicism and educed p o essional e ficacy. In addi ion, one commonly iden ified bu nou ype sco ed high on exhaus ion and cynicism, bu simul aneously low on educed p o essional e ficacy (n=7). Ano he commonly epo ed bu nou ype showed he opposi e pa e n: high sco es on educed p o es- sional e ficacy and low sco es on exhaus ion and cynicism (n=6). Hal o he s udies(n=5) iden ified heexhaus ion-domina ed ype. In addi ion o hese, single ypes ep esen ing, o example, high le els o depe sonaliza ion alone (Deme ou i e al., 2005) o simul- aneously high le els o educed p o essional e ficacy and cynicism (Isoa d-Gau heu e al., 2013), we e epo ed. Al hough he e iewed s udies used di e en samples (i.e., s u- den s, a hle es, heal hy and bu ned-ou employees) and bu nou scales, he ypical bu nou ypes ound showed synch ony be ween he bu nou symp oms, i.e., all he symp oms we e simul aneously high o low, o de-synch ony, whe e he ype was cha ac e ized by ei he high le els o exhaus ion and cynicism o educed p o es- sional e ficacy o solely high exhaus ion le els. 3.3. Bu nou ajec o ies The longi udinal de elopmen o bu nou was in es iga ed in eigh s udies (see Table 1): ajec o ies o school bu nou du - ing adolescence we e in es iga ed in one s udy (Salmela-A o & Upadyaya, 2014), bu nou de elopmen du ing a ehabili a ion in e en ion was examined in wo s udies (Hä inen e al., 2009; Hä inen, Mäkikangas, Kinnunen, & Pekkonen, 2013) and he es (n=5) ocused on job bu nou among employees. He e we ocus mo ecloselyon he fi es udies conduc ed amongwo kingemploy- ees, as hei esul s can be compa ed o a iable-o ien ed esul s. O hese fi e s udies, ou in es iga ed bu nou ajec o ies o e ime(E olah i,Hul ell,&Collins,2013;Hul ell,Melin,&Gus a sson, 2013;Mäkikangas e al., 2012; Rudman & Gus a sson, 2011) and one ocused on s abili y and change pa e ns be ween bu nou ypes o e ime (Boe sma & Lindblom, 2009). All hese s udies we e based on ei he wo (n=2) o h ee measu emen s (n= 3), and he leng h o he ime-lag be ween measu emen s was ypically one o wo yea s. Th ee ou o ou ajec o y s udies ocused on ea ly ca ee s ages, i.e., newly g adua ed nu ses (Rudman & Gus a sson, 2011), eache s (Hul ell e al., 2013) and manage s (Mäkikangas e al., 2012) we e ollowed du ing hei fi s yea s o employmen . All he ou ajec o y s udies epo ed conside able he e o- genei y in bu nou de elopmen o e ime. Typically ajec o ies indica ing low, mode a e o high s able le els o bu nou syn- d ome o i s single symp oms we e ound, bu linea (inc easing o dec easing) and cu ilinea (i.e., U-shaped o e e se U-shaped) ajec o ies we e also epo ed. The numbe o ajec o ies iden i- fied was ela ed o he numbe o measu emen s. This is expec ed as he g ea e he numbe o measu emen s, he g ea e he possibili ies o unco e ing di e en ajec o ies. Fo example, Mäkikangase al. (2012) ound ou ajec o ies o bo h exhaus ion and cynicism using wo-wa e da a (i.e., s able low, s able mode - a e, dec easing and inc easing), whe eas in h ee-wa e da a six (E olah i e al., 2013), se en (Hul ell e al., 2013) o e en eigh (Rudman & Gus a sson, 2011) ajec o ies o bu nou synd ome (based on o al bu nou sco e) we e ound. The ajec o ies ound in he h ee-wa e da a a ied widely ac oss s udies. E olah i e al. (2013) epo ed wo ajec o ies wi h ini ially high bu nou le els ollowed by eco e y, wo ajec o ies desc ibinginc easing bu nou le els, one ajec o y wi h inc easing and diminishing le els, and one ajec o y wi h low s able le els. In u n, Hul ell e al. (2013) epo ed h ee s able ajec o ies (i.e., low, 18 A. Mäkikangas, U. Kinnunen / Bu nou Resea ch 3 (2016) 11–23 mode a e, high), wo linea ly changing ajec o ies (i.e., inc eas- ing and dec easing), and wo cu ilinea ajec o ies (U-shaped o e e se U-shaped). Rudman and Gus a sson (2011) ound one a- jec o y wi h s able low le els, wo linea ly changing ajec o ies (i.e., inc easing and dec easing), ou cu ilinea ajec o ies and wo changing, bu s abilizing ajec o ies. To conclude, he findings o hese pe son-o ien ed s udies ques- ion he s abili y o bu nou epo ed in a iable-o ien ed s udies. In ac , hey highligh he mul i ace ed de elopmen al pa hs o bu nou . A he same ime, hese s udies also e eal ha ajec- o ies ep esen ing s able le els o bu nou o main enance o he same ype o bu nou o e ime we e consis en ly p edominan compa ed o he p opo ion o change ajec o ies o ansi ions be ween he bu nou ypes. Thus, hese findings sugges ha , bu nou ypically de elops along a s able pa h, whe eas changes, i.e.,whe he inc easing,dec easing o cu ilinea , a ea ypical long- e m de elopmen pa hs. 3.4. Pa e ns o bu nou and o he well-being cons uc s The ela ionship be ween bu nou and o he well-being con- s uc swe ein es iga edineigh s udies (see Table1).O hese, wo ocused on bu nou and dep ession (Ahola, Hakanen, Pe honiemi, & Mu anen, 2014;Bianchi, Schon eld, & Lau en , 2015b) and ou on he ela ion be ween bu nou and wo k engagemen (Mäkikangas e al.,2012,2014;Timms, B ough,&G aham,2012;Tuominen-Soini & Salmela-A o, 2014). Two s udies examined employee well- being ypes mo e b oadly by simul aneously analyzing h ee (i.e., bu nou , wo k engagemen , wo kaholism; Innanen, Tol anen, & Salmela-A o, 2014) o ou (i.e., bu nou , wo k engagemen , wo ka- holism and job sa is ac ion; Mäkikangas, Ran anen e al., 2015) job- ela ed well-being cons uc s. In bo h he bu nou -dep ession s udies (Ahola e al., 2014; Bianchi e al., 2015b), simila findings we e ob ained: g oups we e o med by he simila le els o bu nou and dep ession (i.e., low, a e age, high) and hei long- e m de elopmen also occu ed in andem, i.e., ei he emaining s able o changing in synch ony. Thus, on he indi idual le el, he le el and changes o bu nou and dep ession showed high cong uence o e ime. Se e al bu nou -wo k engagemen ypes ( a ying om h ee o fi e) we e iden ified ac oss he ou e iewed s udies. The co-exis ence o di e gence o bu nou and wo k engagemen expe- iences wi hin indi iduals was de e mined by mean le els; high le els o bu nou seemed o p ohibi expe iences o wo k engage- men , as hei co-occu ence was obse ed only a a e age le els in e e y s udy (Mäkikangas e al., 2012, 2014; Timms e al., 2012;Tuominen-Soini&Salmela-A o,2014).Inspec iono con inua e ealed ha a he in a-indi idual le el, cynicism and dedica ion we e ypically mu ually exclusi e expe iences, whe eas o some ex en he co-occu ence o exhaus ion and igo was possible (Mäkikangas e al., 2012). Howe e , a si ua ion whe e he le els and changes in exhaus ion and igo we e in opposi e di ec ions ep esen ed a ypical employee ype o which 63% o he s udied employees belonged a he ai le el (Mäkikangas e al., 2012), and 71% a he s a e (i.e., day) le el (Mäkikangas e al., 2014). Innanen e al. (2014) and Mäkikangas, Ran anen e al. (2015) also showed simila esul s o he bu nou -wo k engagemen con inuum: high le els o bu nou and wo k engagemen we e mu ually exclusi e a he wi hin-pe son le el. In hese wo s ud- ies, he well-being ypes we e addi ionally based on wo kaholism (Innanen e al., 2014) as well as on job sa is ac ion (Mäkikangas, Ran anen e al., 2015). In e es ingly, he a e age le els o wo ka- holismappea edamong hewell-being ype epo inghighle elso bu nou as well as among he ype wi h high le els o wo k engage- men (Innanen e al., 2014). Low le els o job sa is ac ion appea ed wi h high le els o exhaus ion (Mäkikangas, Ran anen e al., 2015). Inclusion o job sa is ac ion also p oduced a new well-being ype, namely Bo ed-ou , wi h ela i ely high le els o exhaus ion, low le els o wo k engagemen and ex emely low le els o job sa is- ac ion, which di e en ia ed his ype om he pu e Bu ned-ou ype (Mäkikangas, Ran anen e al., 2015). To sum up, i seems ha a he in a-indi idual le el, symp- oms o bu nou and dep ession a e insepa able and ha hey also de elop conjoin ly. Fu he , cynicism and dedica ion a e mu u- ally exclusi e expe iences, and hus ep esen he opposi e ends o he same con inuum, ha is, employees do no simul aneously ques ion he meaning o hei wo k and eel a s ong sense o in ol emen in and en husiasm o i . Howe e , he simul aneous p esence o exhaus ion and igo was obse ed among one- hi d o he employees bu only a a e age le els. Along wi h dep ession and wo k engagemen , o he well-being cons uc s, such as wo ka- holism and job dissa is ac ion, we e in e wined wi h he bu nou expe ience a he wi hin-pe son le el, hus poin ing o he need o mo e esea ch a en ion. 4. Discussion The aim o his sys ema ic e iew was o summa ize, ca ego ize and e alua e he esea ch findings o bu nou s udies conduc ed wi h a pe son-o ien ed app oach, and o compa e hese findings wi h hose ob ained wi h he mo e commonly used a iable- o ien ed app oach. The ca ego iza ion o he pe son-o ien ed bu nou s udies (n=24) showed ha he esea ch in e es hus a has been in ypes o bu nou (42%) and, mo e b oadly, in ypes o well-being (33%) and in long- e m ajec o ies (33%). The in o - ma ion p oduced by hese h ee ca ego ies o esea ch will now be compa ed wi h hose gene a ed by a iable-o ien ed bu nou esea ch. 4.1. Compa ison o pe son- and a iable-o ien ed bu nou esul s Basedon he e ieweda icles, i isclea ha bu nou symp oms do no mani es o de elop in he same way in all indi iduals. How- e e , despi e his in e -indi idual he e ogenei y, ce ain ypical bu nou ypes and ajec o ies we e iden ified. Fi s , he esul s o s udies in es iga ing ypes o bu nou e ealed ha , mos ypically, in line wi h he synd ome iew o bu nou (Maslach e al., 2001), all h ee bu nou symp oms co-occu ed a he in a-indi idual le el (a ei he low o high le els). In some cases, each symp om ei he mani es ed on i s own o in symp om dyads, a possibili y which has also been acknowledged in he bu nou li e a u e (Deme ou i, Bakke , Nach eine , & Schau eli, 2001). Pa allel wi h he co ela- i ein o ma ionp oducedby he a iable-o ien ed bu nou s udies ( o e iews, see Kim & Ji, 2009; Wo ley e al., 2008), indi id- uals ypically expe ienced exhaus ion and cynicism a he same ime, whe eas educed p o essional e ficacy mani es ed la gely as a symp om on i s own. Howe e , a ypical bu nou ypes we e also iden ified, e.g., depe sonaliza ion (i.e., cynicism) ei he s anding alone (Deme ou i e al., 2005) o oge he wi h educed p o es- sional e ficacy (Isoa d-Gau heu e al., 2013). Despi e he g ea di e si y in he e iewed s udies on ypes o bu nou , e.g., in he s udy con ex and he bu nou scale used, bu nou ypes ypically mani es ed as a he cons an . Second, mul i ace ed de elopmen al ajec o ies o bu nou , i.e., s able, linea ly inc easing o dec easing, o cu ilinea , we e epo ed ac oss he s udies. This is a finding ha ques ions he findings o he a iable-o ien ed s udies ha bu nou is s able, bu which is in line wi h he heo e ical iew ha desc ibes bu nou de elopmen as a dynamic ongoing p ocess (Maslach, 1982).Ne e heless, he ypical de elopmen al ajec o iesshowed high absolu e s abili y, as he mean le els emained ei he low A. Mäkikangas, U. Kinnunen / Bu nou Resea ch 3 (2016) 11–23 19 o a e age o e ime, hus pa alleling he s able end ound in he a iable-o ien ed s udies ( o a e iew, see Ta is e al., 2005). The p opo ions we e consis en ly highe in he s able ajec o- ies han in hose indica ing change; ha is, indi iduals ended o main ain he same le el o bu nou o e ime. Be ween one- hi d (E olah i e al., 2013) and h ee-qua e s o he indi iduals s ud- ied (Mäkikangas e al., 2012) belonged o hese s able ajec o ies. A simila obse a ion was made wi h ega d o ipsa i e s abili y: Employees ypically belonged o he same bu nou ype o e ime (Boe sma & Lindblom, 2009). Thi d, he ela ion be ween bu nou and he o he (job- ela ed) well-being indica o s a he indi idual le el was mo e complica ed han he co ela ions sugges ed. Pe son-o ien ed esea ch esul s based on he ene gy and iden ifica ion con inua we e mos ly in line wi h hose o he a iable-o ien ed s udies (Deme ou i e al., 2010; González-Romá e al., 2006), bu specified hem u he . Tha is, a he indi idual le el cynicism and dedica ion we e yp- ically no expe ienced a he same ime, whe eas co-occu ence o exhaus ion and igo was possible, bu only a he a e age le els (Mäkikangas e al., 2012, 2014). These a ypical ypes wi h co-occu ing a e age le els o exhaus ion and igo we e ound among app oxima ely one- hi d o he employees. Thus, in line wi h a iable-o ien ed s udies (Deme ou i e al., 2010; González- Romá e al., 2006), hese findings sugges ha igo and exhaus ion ep esen wo somewha sepa a e cons uc s, while dedica ion and cynicism ep esen opposi es. Hence, hese con inua di e ences should be aken in o accoun in u u e s udies. The well-being ypes became mo e mul i a ious when se e al well-being concep s we e in es iga ed simul aneously (Innanen e al., 2014; Mäkikangas, Ran anen e al., 2015). Typically, low and high well-being ypes we e ound ega dless o he numbe o con- s uc s included. Ne e heless, a he same ime, he numbe o small, a ypical well-being ypes inc eased. O e all, i seemed ha bu nou was dis inc i e om he o he well-being cons uc s in he job con ex , meaning ha bu nou had a unique impac on he g ouping o he well-being ypes, hus di e ing om he o he job- ela ed well-being cons uc s o he ci cumplex model (Bakke & Oe lemans, 2011). Howe e , he symp oms o bu nou and dep es- sion seemed o be insepa able a he in a-indi idual le el, as hei le els and changes mani es ed in andem (Ahola e al., 2014; Bianchi e al., 2015b). Thus, o he indi idual, he symp oms o bu nou and dep ession we e alike. Howe e , on he basis o hese wo s udies, i would be p ema u e o conclude ha bu nou and dep ession a e iden ical, as in bo h s udies he employees in es i- ga ed we e ela i e heal hy (i.e., he mean alues o bo h bu nou and dep ession we e low) and dimensionali y o bu nou was no aken in o accoun . To sum up, al hough he pe son-o ien ed bu nou s udies e ealed indi idual bu nou ypes as well as ajec o ies, he esul s we e la gely in line wi h hose p oduced by he a iable-o ien ed s udies. Ye because he pe son-o ien ed pe spec i e e eals ypi- cal as well as a ypical pa e ns o indi idual expe iences, i o e s an impo an , i no essen ial, complemen o he a iable-o ien ed s udies ha cu en ly domina e bu nou esea ch. The ad an ages and challenges o he pe son-o ien ed app oach will be discussed nex . 4.2. Ad an ages and challenges o a pe son-o ien ed app oach The pe son-o ien ed app oach o bu nou has a leas h ee majo ad an ages: (1) i enables a comp ehensi e unde s anding o he bu nou synd ome, wi h i s h ee symp oms, a he indi id- ual le el; (2) i acili a es he applicabili y o he esea ch findings in p ac ice; and (3) i yields an accu a e analysis o ypes and ajec- o ies based on no el analy ical me hods. Fi s , he cen al aspec o a pe son-o ien ed app oach is emphasis i places on unde s anding he indi idual as a unc ioning whole, a he han emphasiz- ing indi idual a iables pe se (Be gman e al., 2003). The e o e, a pe son-o ien ed app oach in es iga es bu nou symp oms and hei de elopmen holis ically wi hin a pe son, no in isola ion om each o he as is he case in a iable-o ien ed app oach. Conse- quen ly mul i- ace ed and de ailed in o ma ion on he phenomena o in e es can be p oduced ha is needed a bo h he heo e ical and p ac ical le el. Second, as he ypes and ajec o ies p oduced by he pe son- o ien ed app oach pack in o ma ion e ficien ly, hey a e easy o unde s and and also in ui i ely appealing (see Robins & T acy, 2003). F om he p ac ical poin o iew, he pe son-o ien ed app oach o bu nou helps o iden i y hose who su e om se e e bu nou o a e in isk o de eloping bu nou . As di e en domains o he wo k en i onmen a e known o be di e en ly associ- a ed wi h he symp oms o bu nou (see Maslach & Lei e , 2008; Maslach e al., 2001), iden i ying bu nou ypes can be used as a ool o de eloping in e en ions ailo ed o indi idual needs. Mo eo e , ia knowledge p oduced by pe son-o ien ed bu nou esea ch, p ac i ione s wo king in he field o occupa ional heal h psychology may be be e placed o deal wi h he bu nou phe- nomenon in a cos -e ec i e manne and ea di e en ypes o bu nou mo e e ficien ly. Fu he mo e, a pe son-o ien ed me hod- ology o e s impo an p ac ical ools o in es iga ing and u he de eloping bu nou in e en ions, e.g., o in es iga e o whom and in wha way in e en ions a e beneficial (c . Hä inen e al., 2009; Hä inen e al., 2013). Thi d, he no el pe son-o ien ed me hodology has se e al ad an ages. Fini e mix u e models accu a ely cap u e he na u- ally exis ing bu nou ypes and ajec o ies. Fo example, g ow h mix u e modelling es ima es indi idual ajec o ies by simul ane- ously aking in o accoun bo h le el and change o e ime (Jung & Wick ama, 2008). In addi ion, classifica ion unce ain y can be es i- ma ed based on he pos e io p obabili y o indi iduals belonging o a specific ype o ajec o y (Wang & Zhou, 2013). Fu he mo e, he di e en ype and ajec o y solu ions ound can be compa ed o each o he wi h igo ous es s (e.g., boo s apped likelihood a io es ) (Lubke & Mu hén, 2005). The e o e, nowadays, mode n pe son-o ien ed me hods can be used in a heo y-d i en confi ma- o y way in o de o a oid dus bowl empi icism (see Wang & Zhou, 2013). In iew o he ob ious ad an ages o he pe son-o ien ed app oach, why i is s ill a ely used in bu nou s udies o , mo e b oadly, in he con ex o occupa ional heal h psychology pe se (see Mäkikangas e al., in p ess)? The e migh be se e al expla- na ions. One is ha his app oach is no well-known ou side he field o pe sonali y and de elopmen al psychology. Ano he is he possibili y o (seman ic) misunde s andings o he pe son- o ien ed app oach (Lau sen, 2015; Lau sen & Ho , 2006), ha is, his app oach is some imes con used wi h he s udy o pe sonal- i y o quali a i e s udy o single pe sons. Some analy ical me hods, such as g ow h cu e modelling, which es ima e indi idual a i- a ion, may also inco ec ly be seen as pe son-o ien ed me hods. Howe e , we belie e – in line wi h Lau sen (2015) – ha he main eason o he a e sion o he pe son-o ien edapp oachis heclaim ha i is only explo a o y and desc ip i e, lacks a heo e ical basis, and is no p edic i e. In he posi i is ic pa adigm o quan i a i e psychological esea ch, causali y, i.e., which an eceden s p edic ou comes, is eage ly pu sued. Howe e , in he heo e ical p oposi ions guiding pe son-o ien ed esea ch, he ques ion o causali y is gi en less weigh han i is in a iable-o ien ed s udies (Be gman & Lundh, 2015). This is because de elopmen is seen as a complex p ocess ha is pa ly indi idual-specific: “I is o en no possible o e en concei e how one componen could be manipula ed wi hou a he same ime a ec ing o he componen s”(Be gman & Lundh, 2015,