Heterogeneous trajectories of perceived stress and their associations with active leisure : a longitudinal study during the first year of COVID-19

This is a sel -a chi ed e sion o an o iginal a icle. This e sion may di e om he o iginal in pagina ion and ypog aphic de ails. Au ho (s): Ti le: Yea : Ve sion: Copy igh : Righ s: Righ s u l: Please ci e he o iginal e sion: CC BY 4.0 h ps://c ea i ecommons.o g/licenses/by/4.0/ He e ogeneous ajec o ies o pe cei ed s ess and hei associa ions wi h ac i e leisu e : a longi udinal s udy du ing he i s yea o COVID-19 © 2024 Kulbin, Jõgi, Pul e and Kask. Published e sion Kulbin, Ka el; Jõgi, Anna-Liisa; Pul e , Aleksande ; Kask, K is jan Kulbin, K., Jõgi, A.-L., Pul e , A., & Kask, K. (2024). He e ogeneous ajec o ies o pe cei ed s ess and hei associa ions wi h ac i e leisu e : a longi udinal s udy du ing he i s yea o COVID-19. F on ie s in Public Heal h, 12, A icle 1327966. h ps://doi.o g/10.3389/ pubh.2024.1327966 2024 F on ie s in Public Heal h 01 on ie sin.o g He e ogeneous ajec o ies o pe cei ed s ess and hei associa ions wi h ac i e leisu e: a longi udinal s udy du ing he i s yea o COVID-19 Ka elKulbin 1 *, Anna-LiisaJõgi 2,3, Aleksande Pul e 1 and K is janKask 1 1 School o Na u al Sciences and Heal h, Tallinn Uni e si y, Tallinn, Es onia, 2 School o Educa ional Sciences, Tallinn Uni e si y, Tallinn, Es onia, 3 Depa men o Teache Educa ion, Uni e si y o Jy äskylä, Jy äskylä, Finland In oduc ion: The e is a ple ho a o li e a u e on he dynamics o men al heal h indica o s h oughou he COVID-19 pandemic, ye esea ch is sca ce on he po en ial he e ogenei y in he de elopmen o pe cei ed s ess. Fu he mo e, he e is a pauci y o longi udinal esea ch on whe he ac i e leisu e engagemen , which ypically is bene icial in educing s ess, migh ha e simila bene i s du ing imes o majo dis up ion. He e we aimed o ex end p e ious wo k by explo ing he dynamics o change in s ess and coping, and he associa ions wi h ac i e leisu e engagemen o e he i s yea o COVID-19. Me hods: Da a om 439 adul s (Mage = 45, SD = 13) in Es onia who pa icipa ed in a longi udinal online s udy we e analyzed. The pa icipan s we e assessed a h ee imepoin s: Ap il–May 2020; No embe –Decembe 2020; and Ap il–May 2021. Resul s: Mean s ess and coping le els we e s able o e ime. Howe e , la en p o ile analysis iden i ied ou dis inc ajec o ies o change in s ess and coping, in ol ing esilien , s essed, eco e ing, and de e io a ing ends. Pa icipan s belonging o he posi i ely de eloping s ess ajec o ies epo ed highe ac i e leisu e engagemen han hose belonging o he nega i ely de eloping s ess ajec o ies. Discussion: These indings highligh he impo ance o adop ing pe son- cen e ed app oaches o unde s and he di e se expe iences o s ess, as well as sugges he p omo ion o ac i e leisu e as a po en ially bene icial coping esou ce, in u u e c ises. KEYWORDS pe cei ed s ess, coping, ac i e leisu e, pe son-cen e ed app oach, COVID-19 pandemic 1 In oduc ion I is accep ed ha COVID-19 and he ci cums ances su ounding he pandemic exace ba ed men al heal h a ound he wo ld. The COVID-19 pandemic sp ead in many wa es, and his was accompanied by a ying le els o social and economic es ic ions and he accumula ion o po en ially s ess ul li e ci cums ances (1). The pandemic ou b eak cons i u es an acu e, la ge-scale, and uncon ollable s esso wi h a long- e m impac . The de imen al OPEN ACCESS EDITED BY Gab iele Nibbio, Uni e si y o B escia, I aly REVIEWED BY Noemi Ta i-Ke esz es, Flinde s Uni e si y, Aus alia Anja Ch is ine Fenebe g, Uni e si y o Müns e , Ge many *CORRESPONDENCE Ka el Kulbin [email p o ec ed] RECEIVED 25 Oc obe 2023 ACCEPTED 24 Ap il 2024 PUBLISHED 09 May 2024 CITATION Kulbin K, Jõgi A-L, Pul e A and Kask K (2024) He e ogeneous ajec o ies o pe cei ed s ess and hei associa ions wi h ac i e leisu e: a longi udinal s udy du ing he i s yea o COVID-19. F on . Public Heal h 12:1327966. doi: 10.3389/ pubh.2024.1327966 COPYRIGHT © 2024 Kulbin, Jõgi, Pul e and Kask. This is an open-access a icle dis ibu ed unde he e ms o he C ea i e Commons A ibu ion License (CC BY). The use, dis ibu ion o ep oduc ion in o he o ums is pe mi ed, p o ided he o iginal au ho (s) and he copy igh owne (s) a e c edi ed and ha he o iginal publica ion in his jou nal is ci ed, in acco dance wi h accep ed academic p ac ice. No use, dis ibu ion o ep oduc ion is pe mi ed which does no comply wi h hese e ms. TYPE O iginal Resea ch PUBLISHED 09 May 2024 DOI 10.3389/ pubh.2024.1327966 Kulbin e al. 10.3389/ pubh.2024.1327966 F on ie s in Public Heal h 02 on ie sin.o g impac o he pandemic on men al heal h has p ima ily been documen ed h ough he popula ion-le el inc ease in dep ession and anxie y symp oms (2, 3). The o igin and he de elopmen o such men al heal h p oblems a e consis en ly ela ed o excessi e s ess [e.g., (4)], and hese associa ions a e aligned wi h he s ess- ulne abili y models o psychopa hology (5, 6). These models explain he possible ways in which s ess ul expe iences may igge he onse o a men al heal h diso de , whe he an indi idual is p edisposed (i.e., ulne able) o a men al heal h condi ion, and wha ole p o ec i e ac o s may play in hese in e ac ions. Nume ous esea ch e idence ha e linked high pe cei ed s ess no only o emo ional dis u bances such as anxie y (7) bu also o physical heal h [e.g., hype ension, ca dio ascula diseases (8)]. Iden i ica ion o sub-popula ions wi h high isk o s ess and in e en ions o educe s ess le els can po en ially help o p e en la e men al diso de s (9). The ansac ional s ess model (10, 11) posi s ha a pe son’s capaci y o cope and adjus o li e challenges is a consequence o in e ac ions ha occu be ween a pe son and hei en i onmen . The abili y o cope wi h s ess depends on how an indi idual e alua es he ele ance o he s esso s (p ima y app aisal) and whe he a pe son belie es o hold su icien esou ces o elie e o emo e he s esso (seconda y app aisal). In line wi h he ansac ional s ess model, Cohen e al. (12) a gue ha a psychological s a e o pe cei ed s ess (he ea e s ess) occu s when a si ua ion in a pe son’s li e is app aised as h ea ening o demanding and a he same ime esou ces a e insu icien o cope wi h he si ua ion. Howe e , his app oach does no assume ha ce ain li e si ua ions a e inhe en ly s ess ul bu e e s o he cogni i e app aisal p ocess whe e he cogni i ely media ed emo ional esponse is gi en o he si ua ion [e.g., (13)]. Se e al longi udinal s udies among di e en age g oups ha e in es iga ed how s ess le els may ha e changed du ing he pandemic. Mos o hese s udies demons a ed a s able cou se o s ess le els i espec i e o he pandemic si ua ion (14–16). Such indings ha e been explained conside ing he signi ican social and economic challenges (e.g., inancial insecu i ies, changes in he wo king modali ies, dis up ions in he social li e) ha he pandemic b ough in addi ion o he heal h c isis, and which oge he p olonged he isk o ch onic s ess. Sal i e al. (17) epo ed ha s ess le els e en inc eased a e he i s lockdown pe iod in he sp ing o 2020 and u he pla eaued by he second wa e o he pandemic. They sugges ed ha , in addi ion o a con inuous socie al and economic c isis, he li ing o es ic ions in be ween he wa es aised he pe cep ion o isk and he eby a ec ed s ess le els. Con o e sially, Gallaghe e al. (18) demons a ed dec easing s ess le els as he cou se o he pandemic con inued and asc ibed such indings o he p esupposi ion ha indi iduals become mo e esilien o he epe cussions o he pandemic o e ime [see (19)]. Mos o he p e ious longi udinal s udies on he de elopmen o s ess among adul s du ing he COVID-19 pandemic ha e ocused on he a e age changes (i.e., a iable-cen e ed app oach), bu he dis inc cou ses o he s ess o e ime (e.g., inc easing o some, while dec easing o o he s) may bias he esul s and can obscu e he e ogeneous pa e ns (i.e., pe son-cen e ed app oach) o expe iences. The e is no eason o doub ha all he scena ios explained by he abo e-ci ed s udies may ha e pa ly a ec ed s ess esponse h oughou he pandemic bu depended on many con ex ual and pe son-cen e ed ac o s. A me a- analysis o he impac o COVID-19 on men al heal h indica o s showed subs an ial he e ogenei y among he indings o longi udinal s udies (20), which sugges s ha he e we e no ubiqui ous e ec s on men al heal h. Se e al longi udinal s udies (21–23) ha e sc u inized he possibili y o dis inc cou ses o he change in symp oms o men al diso de s (e.g., dep ession and anxie y). These s udies ound he e ogeneous ajec o ies o symp oms du ing he pandemic, showing ha app oxima ely 70–80% o he popula ion consis en ly epo ed no symp oms o men al diso de s. They concluded ha o smalle g oups in he popula ion symp oms o men al diso de s inc eased, o in con as dec eased, as he pandemic con inued i s cou se. These indings add o e idence ha lockdowns did no ha e e enly de imen al e ec s on men al heal h and ha a ce ain p opo ion o people we e psychologically esilien o he ci cums ances, o some migh ha e e en bene i ed om he new wo k and li e pa e ns. A ew s udies ha e also employed a pe son-cen e ed app oach o examining pe cei ed s ess and s esso exposu e based on c oss-sec ional da a om he beginning o COVID-19. These s udies ha e iden i ied dis inc p o iles o pandemic- ela ed exposu e o s esso s in adul s (24) and he e ogeneous p o iles o s ess and coping le els among p egnan women (25). Howe e , such c oss-sec ional s udies do no allow he examina ion o po en ially dis inc ajec o y g oups o s ess de elopmen s o e ime. To ou knowledge, he only published longi udinal in es iga ion employing a pe son-cen e ed app oach o he examina ion o changes in pe cei ed s ess le els du ing he pandemic has been conduc ed among adolescen s [age 12–15 yea s, (16)]. This s udy ound no suppo o dis inc ajec o ies o pe cei ed s ess. Adolescen s (ages 12–15 yea s) we e cha ac e ized by homogeneously s able and mode a e s ess le els du ing he i s yea o he pandemic. The au ho s explained his inding by assuming ha adolescen s commonly expe ience s ess ega dless o he pandemic, and hus pandemic- ela ed s esso s did no g ea ly a ec hei no ma i e s ess le els. To he bes o ou knowledge, no s udies ha e sc u inized in adul s he po en ial he e ogenei y o ajec o ies (i.e., change o e ime) o pe cei ed s ess du ing he pandemic, and he indings o he p e ious a iable-cen e ed longi udinal s udies on pe cei ed s ess a e inconsis en (14, 15, 17, 18). An all-emb acing socio-his o ical e en such as he COVID-19 pandemic p o ides a unique occasion o iden i y di e en pa hs o adap a ion o maladap a ion o pe sis en s esso s and o examine coping esou ces ha help indi iduals manage he e ec s o such d as ic ci cums ances. Engagemen in leisu e ac i i ies is one such beha io al coping esou ce. Explo ing how leisu e con ibu es o elie ing and coun e ac ing s ess has been s udied o decades. Coleman and Iso-Ahola (26) i s p oposed in hei heo y ha leisu e acili a es social suppo and gene a es endu ing belie s o sel -de e mina ion, which bu e s he nega i e impac o s ess on men al and physical heal h. In addi ion, Iwasaki and Mannell (27) desc ibed how leisu e may ac also as a s a egy o pallia i e coping (i.e., empo a ily di e ing om s ess ul e en s o eg oup and gain pe spec i e) and mood enhancemen (i.e., educing nega i e mood and enhancing posi i e mood). Empi ical s udies ha e shown e idence ha when people unde s ess ul ci cums ances a e engaged in leisu e ac i i ies, he s ess is educed and he e o e he nega i e impac o he s ess on heal h is also educed (28–31). Zawadzki e al. (32) ha e also iden i ied he eal- ime wi hin-pe son p ocesses such ha when indi iduals epo ed engaging in leisu e, hey had lowe s ess compa ed o when no engaged in leisu e ac i i y. Iwasaki (33, 34) has shown ha leisu e coping p edic ed posi i e coping ou comes e en beyond he e ec s o gene al coping s a egies (e.g., p oblem- ocused coping un ela ed o leisu e). Al hough no consensus de ini ion o leisu e engagemen is imposed, p io esea ch has mos ly ea ed i as a beha io al Kulbin e al. 10.3389/ pubh.2024.1327966 F on ie s in Public Heal h 03 on ie sin.o g concep —de ined as he equency o he amoun o ime in which one pa icipa es in leisu e ac i i ies ou side wo k du ies, pe sonal main enance, and o he obliga ions (35). The classi ica ion o leisu e ac i i ies has nei he been consis en in he li e a u e. Leisu e ac i i ies ha e been di ided ei he as passi e (also e e ed o as “low-demand” o “ ime-ou ” leisu e) o ac i e (also e e ed o as “high-demand” o “achie emen ” leisu e) (36–38). P io esea ch has shown ha engagemen in ac i e leisu e ac i i ies (e.g., hobbies, physical, and na u e-based ac i i ies) is mo e consis en ly linked wi h he bene i s o s ess educ ion (30, 39, 40). Cal abiano (41) iden i ied ha ou doo ac i i ies/spo s and hobbies we e he mos signi ican leisu e ac i i ies o educe s ess. Such ac i i ies o en in ol e using bo h physical and men al ene gy and o en happen wi h o he people. Iwasaki e al. (42) ha e emphasized ha ac i e leisu e is mo e han jus physical ac i i ies, and less physically ac i e o ms o leisu e should no beunde alued in leisu e coping p ocesses. I can beassumed ha ac i e leisu e ac i i ies in ol e ing edien s (e.g., social in e ac ion, c ea i e exp ession, cogni i e s imula ion) o s imula e a wide ange o mechanisms (e.g., psychological, biological, social) which may simul aneously play a ole in alle ia ing s ess [see (43)]. Howe e , i has been shown ha pa adoxically people end o educe hei pa icipa ion in ac i e leisu e when hey a e s essed, which can becaused by an in ui i e p e e ence o passi e leisu e du ing hec ic imes o by a no delibe a e eac ion o he le els o s ess (44). Thus, he ela ionship be ween ac i e leisu e and s ess could bebidi ec ional, wi h s ess also a ec ing mo i a ion o engage in ac i e leisu e. A he same ime, he op ions o ac i e leisu e we e o en es ic ed du ing he pandemic, possibly u he limi ing he engagemen in ac i e leisu e. P e ious s udies ha e epo ed ha he numbe o leisu e ac i i ies people engaged in dec eased (45), and engagemen in physical and ou doo ac i i ies was educed (46) du ing he i s yea o COVID-19. Se e al s udies ha e examined leisu e engagemen as a po en ial coping esou ce also du ing COVID-19. Based on he ecological momen a y assessmen da a, i has been shown ha engaging in ee ime was associa ed wi h lowe s ess le els du ing he pandemic (47). Exis ing indings also sugges ha changes in leisu e engagemen (compa ed o p e-COVID) we e ela ed o poo e men al heal h (46, 48) and people who el hei cu en leisu e engagemen le el ell below hei desi ed le el epo ed lowe men al well-being (46). Takiguchi e al. (45) ha e shown in hei longi udinal s udy ha engaging in a la ge numbe o leisu e ac i i ies du ing he pandemic educed dep essi e symp oms h ough esilience. Howe e , longi udinal esea ch is sca ce on whe he ac i e leisu e engagemen , which is usually bene icial o s ess educ ion, migh ha e simila bene i s in imes o majo dis up ions o he pandemic. I can beassumed ha he e ogeneous ajec o ies (i hey eme ged as such) o pe cei ed s ess du ing he pandemic we e cha ac e ized by dis inc le els o ac i e leisu e engagemen . As engagemen in ac i e leisu e is linked wi h he bene i s o s ess educ ion (30, 39–41), i can be u he assumed ha highe engagemen in ac i e leisu e was associa ed wi h posi i ely de eloping (i.e., dec easing) s ess ajec o ies. I can beexpec ed ha lowe engagemen in ac i e leisu e was ela ed o nega i ely de eloping (i.e., inc easing) s ess ajec o ies. The p esen s udy aims o explo e he dynamics o change in s ess and i s associa ions wi h ac i e leisu e engagemen as a s ess coping esou ce o e he i s yea o he COVID-19 pandemic. The s udy seeks o expand p e ious esea ch by examining a ying ajec o ies o change in s ess (i.e., di e ences in he le el, and he di ec ion o change) and he in e play be ween he changes in s ess and ac i e leisu e engagemen o e ime. By doing his, wecould gain a mo e di e en ia ed unde s anding o he pandemic’s complex impac on s ess le els and con ibu e o o mula ing guidance o s ess- elie ing beha io s in po en ial u u e lockdowns and pandemics. As his s udy is explo a o y by na u e, o achie e he aim o he s udy, he ollowing esea ch ques ions a e examined: (1) How did pe cei ed s ess change o e he i s yea o COVID-19? (2) Can dis inc ajec o ies beiden i ied based on pe cei ed s ess le els o e he i s yea o COVID-19? (3) How was ac i e leisu e engagemen ela ed o belonging o a ce ain s ess ajec o y o e he i s yea o COVID-19? 2 Ma e ials and me hods 2.1 P ocedu e and sample This s udy is pa o a longi udinal in es iga ion ha ocuses on he dynamics o men al heal h and well-being du ing he COVID-19 pandemic in Es onia. App o al o conduc ing he esea ch was g an ed by he Tallinn Uni e si y e hics commi ee (Ap il 15, 2020; decision no 6). Volun a y pa icipan s we e ec ui ed o he su ey ia online ads (wi h he link o he su ey) in news po als (e.g., Del i. ee), and social media channels (e.g., Facebook). The en i e s udy was conduc ed online using he Su eyMonkey pla o m. Es onian- speaking adul s aged 18 o olde cu en ly esiding in Es onia we e eligible o pa icipa e. No compensa ion was o e ed as an incen i e o pa icipa e. A e eading an in o ma ion page and con i ming hei in o med consen , pa icipan s comple ed he su ey. The da ase s ac oss h ee assessmen s we e me ged based on unique anonymized iden i ica ion numbe s (using SPSS). The da a was collec ed o e h ee imepoin s ac oss he i s yea o he COVID-19 pandemic: a Time 1 (T1 – Ap il 20 h un il May 11 h, 2020); a Time 2 (T2 – No embe 9 h un il Decembe 6 h, 2020); and a Time 3 (T3 – Ap il 27 h un il May 23 d, 2021). A T1, 530 pa icipan s we e ec ui ed o he longi udinal s udy, o whom 257 esponded a T2 and 249 esponded a T3. An addi ional 212 pa icipan s we e ec ui ed a T2 ( ia a simila s a egy as a T1), o whom 142 esponded also a T3. Two hund ed pa icipan s esponded o he su ey a all h ee imepoin s. To beable o analyze po en ial changes, hose who had esponded o he su ey a leas wice we e included in he da a analysis. This s a egy esul ed in a sample size o 448, which was p edominan ly composed o emales (92.4%). Pa icipan s’ ages anged om 18 o 81 (M = 45.37, SD = 12.97). 98.2% o he pa icipan s epo ed hei na i e language as Es onian. In e ms o ela ionship s a us, 31% we e single (including widowed, di o ced) and 69% we e in a ela ionship (including ma ied, cohabi a ion, ci il pa ne ship). 82.1% o he pa icipan s we e employed, and 17.9% we e no employed (including s uden s, and pensione s). Figu e1 shows he pandemic si ua ion in Es onia du ing he h ee da a collec ion pe iods. In sp ing 2020, while he i s measu emen (T1) occu ed, he S a e o Eme gency was in e ec in Es onia, which Kulbin e al. 10.3389/ pubh.2024.1327966 F on ie s in Public Heal h 04 on ie sin.o g mean ha he a ailabili y o medical se ices was dec eased. S uden s we e ans e ed o dis ance lea ning; public ga he ings we e banned. Along wi h es ic ions in a eling, all leisu e acili ies we e closed, excluding pa ks and ec ea ional ails i ollowing he “2 + 2 ule” (i.e., a maximum o wo people oge he a one ime, keeping a minimum dis ance o wo me e s apa om o he s). In au umn 2020 (T2), a e a ela i ely i us- and es ic ion- ee summe , he second wa e o he i us a i ed, and he numbe o new cases was ising apidly. Howe e , by ha ime, ligh e es ic ions (compa ed o T1) we e only being g adually e-in oduced—schools we e s ill open and leisu e acili ies we e so a mos ly a ailable. The hi d da a collec ion, in sp ing 2021 (T3), ollowed a pe iod in which he numbe s o new cases and hospi aliza ions had been he highes obse ed h oughou he pandemic, and he go e nmen had e-imposed s ic e es ic ions las ing un il May 2021. Widesp ead accina ion agains COVID-19in he gene al popula ion (age g oups below 60 yea s) did no s a un il mid-May 2021in Es onia (49) when ou hi d da a collec ion (T3) was ending. 2.2 Measu es 2.2.1 Pe cei ed s ess Pe cei ed s ess was assessed using he Es onian e sion o he 10-i em Pe cei ed S ess Scale [PSS-10; (50)]. Pa icipan s we e asked o indica e on a i e-poin scale (0 = ne e , 4 = e y o en) how o en hey el o hough a ce ain way du ing he las 4 weeks (e.g., “How o en did you eel unable o con ol he impo an hings in li e?”). O iginally, his sel - epo ed ques ionnai e was designed o measu e “ he deg ee o which si ua ions in one’s li e a e app aised as s ess ul” [(12), p.385], consis ing o six posi i ely (i ems 1, 2, 3, 6, 9, 10) and ou nega i ely (i ems 4, 5, 7, 8) wo ded i ems. Al hough he scale was de eloped o cap u e s ess as a single la en ac o , ollowing empi ical s udies in di e en con ex s and languages using con i ma o y ac o analysis (CFA) echniques ha e p edominan ly shown ha a wo- ac o model i s he da a be e han a unidimensional model (51–53). These wo ela ed ac o s ha e been desc ibed as (a) pe cei ed s ess (o helplessness; nega i ely wo ded i ems) and (b) pe cei ed coping (o sel -e icacy; posi i ely wo ded i ems). In a o o he wo- ac o solu ion, au ho s ha e poin ed ou ha he con en o posi i ely ph ased and nega i ely ph ased i ems do no coincide (54); and he wo ac o s ha e shown dis inc p edic i e quali ies (55). Fo he Es onian e sion o he PSS-10, only he p elimina y psychome ic p ope ies ha e been p e iously epo ed, based on p incipal componen analysis and in e nal eliabili y coe icien s o he unidimensional solu ion o he scale (56). Thus, CFA was conduc ed o he PSS-10 o examine whe he a one- o wo- ac o solu ion i s he da a bes . Ou da a suppo ed he wo- ac o model o he Es onian e sion o he PSS-10. Hence, he cu en s udy ea ed pe cei ed s ess as a wo-dimensional cons uc o s ess and coping. Longi udinal measu emen in a iance (MI) analysis was also conduc ed o ensu e whe he compa isons o s ess and coping sco es ac oss he h ee imepoin s we e meaning ul (57). Ou da a showed con igu al in a iance and pa ial scala and me ic in a iance in h ee imepoin s, as ac o loadings and i em in e cep s we e allowed o a y o wo i ems. The de ailed esul s o CFA and MI a e p o ided in Supplemen a y ma e ial. C onbach’s alphas showed good in e nal consis ency o bo h he s ess and coping i ems a each ime poin (α = 0.83–0.88). Mean alues we e calcula ed o bo h scales a each imepoin and used in u he analyses. FIGURE1 Si ua ion du ing h ee da a collec ion ime windows (T1–T3): he numbe o COVID-19 dea hs, pa ien s hospi alized, and new cases pe day. Sou ce: Compiled by au ho s based on da a p o ided by TEHIK (49). Kulbin e al. 10.3389/ pubh.2024.1327966 F on ie s in Public Heal h 05 on ie sin.o g 2.2.2 Ac i e leisu e Ac i e leisu e engagemen was measu ed wi h a o ma i e scale, comp ising he equency o esponden s’ pa icipa ion in h ee leisu e ac i i ies: (1) engaging in physical ac i i ies (e.g., spo s, walking); (2) spending ime in na u al se ings (e.g., pa ks, o es s); (3) pa icipa ing in main hobby/pu sui . A simila agg ega ion app oach has been used by nume ous p e ious s udies when he goal has been o cap u e a b oade leisu e ac i i y domain wi h one indica o [e.g., (36, 46, 58)]. The h ee ac i e leisu e ac i i ies we e selec ed based on li e a u e: hei s ess-alle ia ing quali ies ha e been widely desc ibed (30, 39, 40); and hey ha e been consis en ly linked wi h be e men al heal h, bo h be o e (59) and du ing he pandemic (48). Al hough ou choice o leisu e ac i i y i ems was no all-inclusi e, i apped majo ac i e leisu e engagemen ace s ele an o his s udy (39–43). Pa icipan s we e asked o a e how o en hey spen ime doing each o he ac i i ies du ing he las mon h. Response op ions we e: 1 = “less han once a week o ne e ”; 2 = “1–2 imes a week”; 3 = “3–4 imes a week”; 4 = “5–6 imes a week”; 5 = “e e y day”; and 6 = “2 o mo e imes a day.” The mean agg ega ion o he h ee ac i i ies was used, which weigh s each ac i i y equally. Highe sco es indica e highe ac i e leisu e engagemen . 2.3 S a is ical analyses Since he inal sample also included hose pa icipan s who had missed one o he da a collec ion poin s, he da ase had missing alues o pe cei ed s ess, pe cei ed coping, and ac i e leisu e engagemen a di e en imepoin s (31.7% o cases a T1; 10.9% a T2; 12.7% a T3). Reg ession impu a ion was used o p ese e all cases and o ill in he missing alues (60). Fo he impu a ion models o s ess and coping, a ailable sco es o bo h cons uc s o he o he wo imepoin s we e used as p edic o s. In he eg ession impu a ion models o ac i e leisu e engagemen , a ailable sco es o he o he wo imepoin s o he same cons uc we e used as p edic o s. Nex , he s ess and coping a iables we e sc u inized o he absence o mul i a ia e ou lie s. Nine cases we e elimina ed as mul i a ia e ou lie s, which esul ed in he inal sample size o 439. Fu he , epea ed measu es analysis o a iance (ANOVA) con olling o co a ia es (age and gende ) was used o examine changes in s ess and coping o e ime. Viola ions o sphe ici y we e add essed using G eenhouse–Geisse co ec ions. Nex , explo a o y la en p o ile analysis (LPA) was used o iden i y dis inc ajec o ies o pe cei ed s ess ac oss he h ee imepoin s. LPA as a pe son-cen e ed echnique allows he iden i ica ion o he e ogeneous subpopula ions comp ising dis inc esponse pa e ns ac oss ime. Deciding he numbe o subg oups (i.e., ajec o ies) is based on he g ouping p ecision and he compa a i e i indices, as well as he in e p e abili y o subg oups (61). Bo h s ess and coping ac o s we e modeled in one LPA wi h he a iances allowed o a y be ween g oups. Also, he co a iance be ween s ess in h ee imepoin s and he co a iance be ween coping in h ee ime poin s we e allowed o a y be ween g oups. The i o models wi h he di e en numbe o p o iles was compa ed using he Akaike in o ma ion c i e ion (AIC), he Bayesian in o ma ion c i e ion (BIC), he sample-size adjus ed Bayesian in o ma ion c i e ion (aBIC), he Vuong-Lo-Mendell-Rubin likelihood a io es (VMLR), boo s apped likelihood a io es (BLRT), a measu e o en opy, in e p e abili y o he obse ed ajec o ies, and he size o he p o iles (61). A e model selec ion, pa icipan s we e classi ied acco ding o hei mos likely p o ile membe ship. Finally, a mixed ANOVA model con olling o age was un o examine he in e ac ion be ween changes in ac i e leisu e engagemen ( ime as a wi hin-subjec s ac o ) and ajec o ies o s ess and coping (as a be ween-subjec s ac o ). Gende was no included as a co a ia e due o he low numbe o men (<5) in some o he s ess ajec o y g oups ound wi h LPA. The assump ion o homogenei y o a iances was es ed by Box’s M es . Viola ions o sphe ici y we e co ec ed by applying a G eenhouse–Geisse co ec ion. Fo pos hoc mul iple compa isons, Bon e oni adjus men was used. CFA and in a iance es s we e pe o med in R e sion 4.1.3 (62), using la aan package (63). Reg ession impu a ions we e pe o med in R package mice (64). LPA was conduc ed using Mplus 8.8 (65). ANOVAs we e pe o med in SPSS e sion 28. 3 Resul s The means, s anda d de ia ions, anges, C onbach’s alphas, and bi a ia e co ela ions o all he s udy a iables a e shown in Table1. 3.1 Changes in s ess and coping: a iable-cen e ed app oach Fi s , changes in a e age pe cei ed s ess and pe cei ed coping du ing he i s yea o he pandemic we e in es iga ed using epea ed measu es ANOVA. The e we e no signi ican changes ound ac oss h ee imepoin s in mean sco es o s ess, F (1.88, 818.60) = 0.83, p = 0.43, η p2 = 0.002, no in mean sco es o coping, F (2, 872) = 0.84, p = 0.43, ηp2 = 0.002. 3.2 Dis inc ajec o ies o s ess and coping: pe son-cen e ed app oach To iden i y po en ial dis inc ajec o ies o s ess du ing he i s yea o he pandemic, a la en p o ile analysis was conduc ed on pe cei ed s ess and coping sco es measu ed a h ee imepoin s. Six se s o LPA-s we e compa ed. The d op o AIC and aBIC alues decele a ed, and BIC alue did no u he dec ease, a e he ou - ajec o y solu ion (see Table2 o he i indices, en opy, and g oup sizes). The i e- ajec o y solu ion did no e eal any new pa e ns o change, and he mo e pa simonious ou - ajec o y model was chosen as i had he bes in e p e abili y. Figu e2 p esen s he s ess and coping ajec o ies o e h ee imepoin s o ou g oups iden i ied in he LPA model. The i s ajec o y, labeled as ‘S essed’ (15%, N = 66), was cha ac e ized by a high s ess le el and a low coping le el h oughou he s udy. In he second ajec o y, labeled as “De e io a ing” (27%, N = 120), he pa icipan s had ela i ely low s ess and high coping a he beginning o he pandemic, bu i was ollowed by a sus ained incline in s ess and decline in coping h oughou he i s yea o he pandemic. The la ges p opo ion o pa icipan s (33%, N = 144) belonged o he hi d ajec o y labeled as “Resilien .” The pa icipan s in his g oup had consis en ly low s ess and high coping ac oss he i s yea o he Kulbin e al. 10.3389/ pubh.2024.1327966 F on ie s in Public Heal h 06 on ie sin.o g pandemic. In he ou h ajec o y, labeled as “Reco e ing” (25%, N = 109), he pa icipan s epo ed ela i ely high le els o s ess a he beginning o he pandemic. Howe e , hese pa icipan s “bounced back” o e ime, as indica ed by a decline in s ess and an incline in coping h oughou he nex wo imepoin s. Nex , we es ed i he ou g oups iden i ied in he LPA we e cha ac e ized by di e ences in age, ela ionship s a us, o wo k s a us. A one-way ANOVA was used o compa e he mean age be ween ou s ess and coping g oups. The e was no s a is ically signi ican di e ence in age be ween he ou g oups [F (3, 435) = 2.36, p = 0.07]. Chi-squa e es s we e used o examine i he g oup membe ship was ela ed o ela ionship s a us o wo k s a us. Rela ionship s a us was dicho omized in o “single” (incl. Widowed, di o ced) and “in a ela ionship” (incl. Ma ied, cohabi a ion, ci il pa ne ship). Wo k s a us was dicho omized in o “employed” and “no employed” (incl. S uden , pensione ). G oup membe ship was nei he ela ed o ela ionship s a us [X2 (3, N = 439) = 1.39, p = 0.71] no o wo k s a us [X2 (3, 439) = 3.57, p = 0.31]. 3.3 Changes in ac i e leisu e engagemen in ela ion o dis inc ajec o ies o s ess and coping Changes in ac i e leisu e engagemen we e in es iga ed in ela ion o dis inc ajec o ies o s ess and coping. Speci ically, a 4 ( ajec o ies) X 3 ( imepoin s) mixed ANOVA model con olling o age was un o examine he in e ac ion e ec be ween ajec o ies o s ess and coping (g oup membe ship as a be ween-subjec s ac o ) and ime (as a wi hin-subjec s ac o ) on ac i e leisu e engagemen . The mixed ANOVA esul s a e illus a ed in Figu e3. The e was a main e ec o ime on ac i e leisu e engagemen , F (1.96, 850.85) = 9.68, p < 0.001, η p2 = 0.02. Bon e oni adjus ed pai wise compa isons showed a quad a ic e ec such ha ac i e leisu e engagemen dec eased (p = 0.001) om sp ing 2020 o au umn 2020, and hen inc eased (p = 0.001) om au umn 2020 o sp ing 2021 (see Table1 o means and SDs). The e was a main e ec o s ess and coping ajec o y membe ship on ac i e leisu e engagemen , F (3, 434) = 12.18, p < 0.001, η p2 = 0.08. Bon e oni adjus ed pai wise compa isons e ealed ha pa icipan s belonging o he “Resilien ” (M = 3.46, SD = 0.94) ajec o y epo ed highe ac i e leisu e engagemen han hose in he “S essed” (M = 2.93, SD = 0.93) and “De e io a ing” (M = 2.83, SD = 0.93) ajec o ies (bo h compa isons p < 0.001). In addi ion, pa icipan s belonging o he “Reco e ing” (M = 3.30, SD = 0.93) ajec o y epo ed highe ac i e leisu e engagemen han hose in he “De e io a ing” (M = 2.83, SD = 0.93) ajec o y (p < 0.001). The in e ac ion be ween he s ess ajec o ies and changes in ac i e leisu e engagemen was no ound, F (5.88, 850.85) = 1.30, p = 0.26, η p2 = 0.009, ailing o p o e ha changes in ac i e leisu e engagemen we e ela ed o dis inc ajec o ies o s ess and coping. 4 Discussion The p esen s udy aimed o explo e he dynamics o change in s ess and i s associa ions wi h ac i e leisu e engagemen as a s ess coping esou ce du ing he i s yea o he COVID-19 pandemic. Ou TABLE1 Means (M), s anda d de ia ions (SD), anges, C onbach’s alphas (α), and co ela ions be ween s udy a iables. MSD Range α1 2 3 4 5 6 7 8 1. Pe cei ed s ess T1a1.75 0.76 0.17–4 0.86 – 2. Pe cei ed s ess T2a1.74 0.79 0–4 0.88 0.54 – 3. Pe cei ed s ess T3a1.73 0.80 0–4 0.88 0.52 0.70 – 4. Pe cei ed coping T1a2.53 0.66 0–4 0.87 −0.71 −0.50 −0.50 – 5. Pe cei ed coping T2a2.49 0.71 0–4 0.83 −0.42 −0.72 −0.59 0.61 – 6. Pe cei ed coping T3a2.48 0.74 0–4 0.86 −0.37 −0.55 −0.71 0.66 0.70 – 7. Ac i e leisu e engagemen T1b3.34 1.07 1–6 – −0.19 −0.22 −0.21 0.23 0.19 0.20 – 8. Ac i e leisu e engagemen T2b3.00 1.12 1–6 – −0.13 −0.23 −0.25 0.17 0.21 0.22 0.70 – 9. Ac i e leisu e engagemen T3b3.17 1.15 1–6 – −0.14 −0.26 −0.27 0.17 0.23 0.28 0.64 0.69 N = 439; all co ela ions we e s a is ically signi ican a p < 0.01. aScale om 0 o 4. bScale om 1 o 6. TABLE2 Fi s a is ics o compa ison o di e en longi udinal la en p o ile models o pe cei ed s ess and coping. Numbe o p o iles NAIC BIC aBIC En opy VLMR p- alue BLRT p- alue 1 439 4,886 4,959 4,902 – – – 2 191/248 4,624 4,775 4,658 0.74 0.04 <0.01 3 215/120/104 4,522 4,751 4,573 0.76 0.16 <0.01 4 66/120/144/109 4,426 4,732 4,494 0.78 0.03 <0.01 5 104/70/52/85/128 4,363 4,747 4,448 0.81 0.27 <0.01 6 83/105/94/28/73/56 4,339 4,800 4,442 0.82 0.76 0.14 N, g oup sizes; AIC, Akaike in o ma ion c i e ion; BIC, Bayesian in o ma ion c i e ion; aBIC, sample-size adjus ed Bayesian in o ma ion c i e ion; VLMR, Vuong-Lo-Mendell-Rubin likelihood a io es ; BLRT, boo s apped likelihood a io es . Es ima es o he chosen ou - ajec o y solu ion a e bolded. Kulbin e al. 10.3389/ pubh.2024.1327966 F on ie s in Public Heal h 07 on ie sin.o g pe son-cen e ed analy ical app oach wi h longi udinal da a adds o p e ious esea ch by iden i ying he e ogeneous ajec o ies o change in s ess among adul s. In addi ion, he cu en s udy ex ends p e ious esea ch by demons a ing how s ess ajec o ies we e cha ac e ized by dis inc le els o ac i e leisu e engagemen in imes o majo social and economic dis up ions o he pandemic. Add essing he i s esea ch ques ion, he esul s om a iable- cen e ed analyses indica ed ha pe cei ed s ess and coping le els we e s able i espec i e o he si ua ion o e he i s yea o he pandemic. Such inding coincides wi h many o he longi udinal s udies on s ess le els du ing he pandemic (14–16). Howe e , as ou subsequen pe son-cen e ed analyses showed, he depic ion ob ained h ough he con en ional a iable-cen e ed app oach ailed o cap u e he complexi y o he si ua ion. Ou second esea ch ques ion aimed a iden i ying po en ially dis inc s ess ajec o ies. The pe son-cen e ed (la en p o ile) analyses, based on pe cei ed s ess and coping sco es measu ed a h ee imepoin s, e ealed a mo e nuanced unde s anding o empo al s ess dynamics du ing he i s yea o he pandemic among adul s. Fou he e ogeneous ajec o ies o change in s ess and coping we e iden i ied. The la ges p opo ion o he sample belonged o he Resilien g oup (33%), wi h consis en ly s able low s ess and high coping ac oss he yea . This g oup was composed o indi iduals who ended o app aise he ci cums ances as no ha m ul o hem and/o pe cei ed hei esou ces as su icien o cope wi h he demands, ega dless o he a ying condi ions h oughou he i s yea o he pandemic (11, 66). The clea eme gence o such a g oup also suppo s Bonanno’s (67) wo k on a guing how a subs an ial p opo ion o indi iduals endu e a e si e e en s wi h mino e ec s on hei heal hy unc ioning. One-qua e o he sample consis ed o Reco e ing indi iduals, who expe ienced ela i ely high le els o s ess du ing he i s sp ing o he pandemic, bu “bounced back” du ing he ollowing yea . This a o able adap a ion ajec o y could be asc ibed o no el y, unp edic abili y, and ini ial di icul ies wi h new obliga ions ha caused acu e s ess du ing he i s wa e o he i us, bu o e ime adap a ion o he condi ions occu ed and he si ua ion was app aised as less h ea ening [see (19, 68)]. O e a qua e o ou sample belonged o he De e io a ing ajec o y, wi h ela i ely low s ess and high coping a he beginning o he pandemic which was ollowed by a sus ained incline in s ess and decline in coping o e he s udy pe iod. A con inuous socie al and FIGURE2 Es ima ed mean pe cei ed s ess (le ) and pe cei ed coping ( igh ) sco es om he ou - ajec o y solu ion o he la en p o ile analysis ac oss h ee imepoin s. Each g oup indica es a dis inc ajec o y du ing he i s yea o he pandemic. Bo h scales om 0 o 4. E o ba s ep esen 95% con idence in e als. Kulbin e al. 10.3389/ pubh.2024.1327966 F on ie s in Public Heal h 08 on ie sin.o g economic c isis, loss o hope o a quick end o he pandemic, and a possible inc ease in pe cep ion o heal h isks (17) may ha e played a ole o he indi iduals in he de e io a ing ajec o y. The smalles p opo ion o ou sample belonged o he S essed g oup, who expe ienced high s ess and low coping le els h oughou he s udy pe iod. Since wedid no possess p e-pandemic da a on ou sample, i is no possible o c edibly a ibu e high s ess le els o he pandemic. Ne e heless, hese pa e ns o inc easing o pe sis en ly excessi e s ess le els call o pa icula a en ion. Ou analyses demons a ed ha ocusing only on he a e age changes (i.e., a iable-cen e ed app oach) obscu es he a iabili y o he empo al changes in s ess du ing he pandemic and could lead o o e simpli ied in e ences. As opposed o he assump ion o uni o m e ec s o he a ying ci cums ances o he pandemic on s ess le els (14, 17, 18), ou s udy highligh s ha he e was a clea he e ogenei y o empo al changes in pe cei ed s ess ac oss he i s yea o he pandemic. Mo e gene ally, his means ha he iden i ica ion o di e en subg oups in he empo al p ocess o s ess p o ides an oppo uni y o desc ibe di e ences in he de ails o e ec i e coping wi h s ess. Con a y o ou esul s, a s udy conduc ed among adolescen s ound no e idence o he e ogenei y in s ess ajec o ies du ing he i s yea o he pandemic (16). Weassume ha he di e en a ge popula ions o hese s udies explain he disc epancy in indings. One possible explana ion is ha changes in he daily ou ine o adul s we e mo e he e ogeneous compa ed o adolescen s (e.g., in e up ions in he ypical school ou ines we e simila o all s uden s). Among adul s, p e ious s udies on men al diso de symp oms du ing he pandemic ha employed a pe son- cen e ed app oach, ha e consis en ly shown dis inc ajec o ies o he symp oms’ de elopmen (21–23) and hus, suppo ou indings conside ing he link be ween s ess and psychopa hology (6). In e es ingly, he ou ajec o ies also o e lap wi h he p o o ypical ou come ajec o ies o human s ess esponses a e po en ially auma ic li e e en s [see (68)]. I seems ha con inuous and po en ially s ess ul condi ions o he pandemic (i.e., ch onic e en s) we e ollowed by a simila he e ogenei y o s ess esponses ac oss ime, as ha e been obse ed a e sho - e m a e si e li e e en s (i.e., acu e e en s). When conside ing he socio-demog aphics po en ially associa ed wi h he ou s ess ajec o ies, ou analysis indica ed ha he dis inc ajec o ies could no bea ibu ed o age, being single ( s. in a ela ionship), o being employed ( s. no employed). This pa ially con adic s p e ious indings which ha e consis en ly shown ha younge age is ela ed o a highe isk o nega i ely de eloping men al heal h ajec o ies (21–23, 68). FIGURE3 Es ima ed mean ac i e leisu e engagemen sco es ac oss h ee imepoin s acco ding o ou dis inc s ess ajec o ies. Scale om 1 o 6. E o ba s ep esen 95% con idence in e als.