Motives for physical activity in older men and women : A twin study using accelerometer‐measured physical activity

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/ Mo i es o physical ac i i y in olde men and women : A win s udy using accele ome e ‐measu ed physical ac i i y © 2020 The Au ho s. Scandina ian Jou nal o Medicine & Science In Spo s published by John Wiley & Sons L d Accep ed e sion (Final d a ) Aal onen, S.; Walle , K.; Vähä‐Ypyä, H.; Rinne, J.; Sie änen, H.; Sil en oinen, K.; Kap io, J.; Kujala, U.M. Aal onen, S., Walle , K., Vähä‐Ypyä, H., Rinne, J., Sie änen, H., Sil en oinen, K., Kap io, J., & Kujala, U.M. (2020). Mo i es o physical ac i i y in olde men and women : A win s udy using accele ome e ‐measu ed physical ac i i y. Scandina ian Jou nal o Medicine and Science in Spo s, 30(8), 1409-1422. h ps://doi.o g/10.1111/sms.13673 2020 Scand J Med Sci Spo s. 2020;30:1409–1422. | 1409 wileyonlinelib a y.com/jou nal/sms Recei ed: 27 No embe 2019 | Re ised: 10 Ma ch 2020 | Accep ed: 25 Ma ch 2020 DOI: 10.1111/sms.13673 ORIGINAL ARTICLE Mo i es o physical ac i i y in olde men and women: A win s udy using accele ome e -measu ed physical ac i i y Sa iAal onen1 | Ka jaWalle 2 | Hen iVähä-Ypyä3 | JuhaRinne4 | Ha iSie änen3 | Ka iSil en oinen5 | JaakkoKap io1,6 | U ho M.Kujala2 This is an open access a icle unde he e ms o he C ea i e Commons A ibu ion License, which pe mi s use, dis ibu ion and ep oduc ion in any medium, p o ided he o iginal wo k is p ope ly ci ed. © 2020 The Au ho s. Scandina ian Jou nal o Medicine & Science In Spo s published by John Wiley & Sons L d 1Ins i u e o Molecula Medicine (FIMM), Uni e si y o Helsinki, Uni e si y o Helsinki, Helsinki, Finland 2Facul y o Spo and Heal h Sciences, Uni e si y o Jy äskylä, Uni e si y o Jy äskylä, Jy äskylä, Finland 3The UKK Ins i u e o Heal h P omo ion Resea ch, Tampe e, Finland 4Clinical Neu ology, Tu ku PET Cen e, Tu ku Uni e si y Hospi al, Tu ku, Finland 5Depa men o Social Resea ch, Uni e si y o Helsinki, Uni e si y o Helsinki, Helsinki, Finland 6Depa men o Public Heal h, Uni e si y o Helsinki, Uni e si y o Helsinki, Helsinki, Finland Co espondence Sa i Aal onen Ph.D., Ins i u e o Molecula Medicine (FIMM), Uni e si y o Helsinki, P.O. Box 20 (Tukholmanka u 8), FIN-00014 Uni e si y o Helsinki, Helsinki, Finland. Email: [email p o ec ed]i Funding in o ma ion Suomen Kul uu i ahas o; Juho Vainion Sää iö; The Academy o Finland, G an / Awa d Numbe : and 312073 , 266592, 100499, 205585, 141054, 265240, 263278 and 264146; The Finnish Minis y o Educa ion and Cul u e, G an /Awa d Numbe : OKM/56/626/2013 Mo i es o physical ac i i y may a y conside ably by age, sex, and he le el o physical ac i i y. We aimed o examine mo i es o physical ac i i y in olde men and women wi h di e en physical ac i i y le els as well as whe he gene ic and/o en i onmen al ac o s explain hose mo i es. Finnish wins (mean age 72.9yea s, 262 ull win pai s) sel - epo ed hei mo i es o physical ac i i y. Time spen on mode a e- o- igo ous physical ac i i y was moni o ed using a hip-wo n accele ome e . Compa isons be ween he di e en physical ac i i y g oups o olde wins (n=764-791/mo i e dimension) we e analyzed using he Wald es , and e ec sizes we e calcula ed as Cohen's d. Quan i a i e gene ic modeling was used o es ima e gene ic and en i onmen al con ibu ions. Fo bo h sexes, he mos equen ly epo ed mo i es o physical ac i i y we e physical i ness, heal h main enance, and psychological well-being. Con o ming o o he s’ expec a ions was mo e impo an o men han o women (P<.001, Cohen's d=0.38), while appea ance (P=.001 Cohen's d=−0.24) and psychological well-being (P=.02, Cohen's d=−0.17) we e highligh ed by women. Mos o he mo i e dimensions di e ed signi ican ly be ween he physically ac i e and inac i e indi iduals. I was es ima ed ha 5%-42% o he a ia ion in mo i es was con ibu ed by gene ic ac o s and 58%-95% by en i onmen al ac o s. The esul ha en i onmen al ac o s con ibu e in a g ea deal o mo i es indica es ha in e en ions o mo i a e physically inac i e olde indi iduals o be physically ac i e can be success ul. Howe e , pe sonalized in e en ions a e needed because sex and he le el o physical ac i i y we e ound o be associa ed wi h olde indi iduals’ mo i es o physical ac i i y. KEYWORDS aged indi iduals, de ice-measu ed physical ac i i y, exe cise, mo i a ion, wins 1410 | AALTONEN ET AL. 1 | INTRODUCTION Mo i es o physical ac i i y ha e been epo ed o be a c u- cial ac o o unde s and indi idual di e ences in physical ac i i y beha io .1-3 Because he le el o physical ac i i y ends o p og essi ely dec ease wi h age4 and because he p opo ion o he wo ld's popula ion o e 60yea s old will almos double wi hin he nex decades,5 an inc easing in e - es has a isen in in es iga ing he mo i es o physical ac- i i y in he elde ly popula ion. I has been shown ha olde indi iduals a e mo i a ed by many gene al ac o s, such as heal h,6 bu mo i a ional ac o s ha e also been e ealed o a y wi h age.7,8 Two sys ema ic e iews ha e e alua ed mo i es o phys- ical ac i i y in old indi iduals, bu hey did no include me- a-analyses o es ima e pooled e ec sizes.9,10 Bae a al. (2011) 9 epo ed physical heal h as a majo in ape sonal mo i e o physical ac i i y in se e al s udies in indi iduals in hei 80s, ollowed by psychological bene i s, such as en- joymen . Aspec s ela ed o social suppo we e men ioned as one o he mos impo an in e pe sonal mo i es o physical ac i i y. The e ec o p o essional ad ice was also men ioned in a ew pape s included in he e iew. In a e iew o e iews, Ca ill & Fos e (2018) 10 ocused on mo i es o pa icipa e in s eng h and balance aining ac i i ies in middle-aged and olde indi iduals. Some o he mo i es ound in his e iew we e e y gene al o all people o being physically ac i e, such as psychological heal h bene i s and social connec ion wi h o he s, while some mo i es we e mo e speci ic o olde indi iduals, such as imp o ing he abili y o comple e daily ac i i ies and educing he isk/ ea o alling. Despi e he easonable amoun o s udies on mo i es o physical ac i i y in old age, ewe s udies ha e examined whe he sex di e ences exis in mo i es o physical ac i - i y in old age. Recen ly, an U elen and colleagues (2017) 6 ocused on sex di e ences in mo i es o physical ac i i y among indi iduals in hei 60s. They ound ha he leading mo i es o physical ac i i y we e he same o bo h men and women, bu di e ences we e also ound—olde women we e mo i a ed mo e by appea ance, he social aspec o physical ac i i y and weigh con ol han olde men. Simila ly, clea sex di e ences in mo i es o physical ac i i y we e obse ed among younge Malaysian adul s om 20 o 64yea s o age: Men we e mo e mo i a ed by compe i ion and skill imp o e- men , while women highligh ed appea ance and physical condi ion as hei mo i a ional ac o s.11 Howe e , wo o he s udies, one o Finnish olde indi iduals 12 and he o he o I alian senio a hle es,13 did no ind any sex di e ences in mo i es o physical ac i i y. Only a ew s udies ha e paid a en ion o how mo i es o physical ac i i y di e by he le el o physical ac i i y among olde indi iduals, al hough he e is limi ed e idence ha he cu en physical ac i i y le el explains he di e ences in mo i es e en in olde indi iduals.1,3,14 When olde physically ac i e indi iduals ha e been compa ed o olde physically inac i e indi iduals, physical and psychological heal h,1,3,14 i ness,1,3,14 skill imp o emen ,3 and enjoymen 1,14 ha e been iden i ied as majo mo i es o physical ac i i y among olde physically ac i e indi iduals. Ac i e olde indi iduals ha e also gi en highe a ings o s ess managemen and social as- pec s as mo i a ional ac o s compa ed o hei inac i e coun- e pa s,1,14 while physically inac i e olde indi iduals ha e highligh ed o he people's expec a ions3 along wi h a need o pu pose ul and un physical ac i i y.14 The olde he age-g oup, he mo e mo i es may di e compa ed o younge age-g oups because olde indi iduals may ha e speci ic needs ha a ec hei daily li es and hei chances o be physically ac i e. Howe e , acco ding o ou knowledge, he e a e no s udies ha ha e exclusi ely exam- ined mo i es o physical ac i i y among communi y-dwell- ing indi iduals aged 70 and o e aking hei sex and cu en le el o physical ac i i y in o accoun . Fo his eason, he p esen s udy was launched. Mos o he human beha io al ai s a e complex and mul i ac o ial, wi h a likely con ibu- ion o many genes, each wi h small e ec (ie, “polygenic ai s”). These ai s a e no only a ec ed by gene ic ac o s, also en i onmen al ac o s (ie, all he ac o s ha make mem- be s o a monozygo ic win pai di e en om one ano he ) play a ole.15 Ne e heless, he unde lying oles o gene ic and en i onmen al ac o s behind indi idual di e ences in mo i es o physical ac i i y ha e no p e iously been s ud- ied in olde indi iduals, ye ou p e ious s udy indica ed he impo ance o hese ac o s behind mo i es o physical ac- i i y among adul s in hei mid-30s.16 In o de o ill in his knowledge gap, i is impo an o es ima e gene ic and en i- onmen al a ia ions o mo i es o physical ac i i y in olde indi iduals. This s udy p ima ily aims o examine wha he mos e- quen ly epo ed mo i es o physical ac i i y among men and women o e 70yea s old a e, and whe he he e a e di - e ences in mo i a ional ac o s be ween he sexes and be- ween he cu en le els o physical ac i i y. Fu he mo e, we examine o wha ex en gene ic and en i onmen al ac o s con ibu e o mo i es o physical ac i i y. Based on p e ious s udies,9,10 we hypo hesize ha bo h physical heal h and psy- chological heal h along wi h enjoymen o physical ac i i y a e he mos impo an ac o s o induce bo h olde men and women o be physically ac i e. Fu he , based on p e ious indings,6,11 we hypo hesize ha he e a e sex di e ences in mo i es o physical ac i i y among olde indi iduals—olde women a e assumed o be mo e likely mo i a ed by hei ap- pea ance and he social aspec o physical ac i i y han olde men. We also expec ha hose olde indi iduals who a e physically ac i e epo highe le els o in insic mo i es o physical ac i i y (such as physical and psychological heal h and enjoymen o ac i i y). Finally, we assume ha gene ic | 1411 AALTONEN ET AL. ac o s pa ly explain mo i es o physical ac i i y in old age, as seen in younge ages. Based on ou p e ious s udy,16 we belie e ha he con ibu ion o gene ic ac o s will ange be- ween 10% and 55%— he mo i e dimensions o enjoymen and socializing being he highes . 2 | METHODS 2.1 | Pa icipan s The pa icipan s o he MOBILETWIN s udy we e d awn om he Olde Finnish Twin Coho s udy ha was launched in 1974. I is a na ionwide sample o all same-sex Finnish win pai s bo n be o e 1958 wi h bo h co- wins ali e in 1975. A o al o 13888 win pai s ha e been iden i ied.17 O e he yea s, he wins ha e pa icipa ed in h ee la ge mail su eys in 1975, 1981, and 1990. Zygosi y (monozygo ic (MZ), dizy- go ic (DZ), o unce ain) has been de e mined using a well- alida ed ques ionnai e.18 The MOBILETWIN s udy was conduc ed be ween 2014 and 2016, and only win pa icipan s who we e bo n be ween 1940 and 1944 we e eligible o he s udy. Fu he , only he win indi iduals who had esponded o a leas one o he i s wo mail su eys (1975 o 1981) we e selec ed o he p esen s udy.19 In o al, 1632 win indi iduals ali e and con ac able (ie, 816 comple e same-sex win pai s, ou o which 256 we e MZ, 490 DZ and 70 wi h uncon i med zygosi y) we e in i ed o pa icipa e in a heal h and cogni ion elephone in e iew, use an accele ome e o moni o hei physical ac i i y beha - io , and espond o a ques ionnai e ela ed o physical unc- ioning. The a ge g oup was a selec ed communi y-dwelling g oup o olde indi iduals since hose less heal hy indi id- uals li ing in nu sing homes o long- e m ins i u ional ca e we e no con ac able. The de ails o he inclusion c i e ia ha e been published p e iously.19 In o al, 1012 (62%) win indi iduals pa icipa ed in he elephone in e iew, 791 used he hip-wo n accele ome e o he ime equi ed o he s udy, and he numbe o pa icipan s who epo ed hei mo i es o physical ac i i y anged om 819 o 837 pe mo i e dimension. Thus, we had da a a ailable on accele ome e -measu ed physical ac i i y and mo i es o physical ac i i y om 764 o 791 win indi iduals pe mo i e dimension (48%-51% women). 2.2 | Assessmen o he mo i es o physical ac i i y To e alua e he olde indi iduals’ mo i es o physical ac i i y, we used a ques ionnai e ha was based on he Rec ea ional Exe cise Mo i a ion Measu e (REMM).20 The o iginal REMM measu e was designed o assess adul s’ leisu e- ime physical ac i i y mo i es, and i is based on he Sel -De e mina ion Theo y.21 This heo y p o ides a s ong ounda ion o unde s anding he goals and mo i es o be- ha io and ocuses on he impo ance o in insic mo i es in d i ing human beha io . The o iginal e sion o he REMM measu e consis s o 73 i ems ep esen ing bo h in insic and ex insic mo i es o physical ac i i y. These 73 i ems o m 8 dimensions o 8 o 13 i ems. In he MOBILETWIN s udy, hese 8 dimensions we e used o assess mo i es o physical ac i i y due o he space es ic ions o he mail su ey ques ionnai e bu also o make he ques ionnai e easie and as e o olde indi iduals o comple e. We used one gene al i em pe dimension o ep- esen ex ensi ely he con en o he dimension— his is, one i em in he modi ied REMM equals o one dimension o he o iginal REMM. Fu he , because he MOBILETWIN s udy ocused on olde indi iduals, we sligh ly adap ed he con en s o mo i a ion measu e o be e mee he equi emen s o he olde indi iduals (ie, o imp o e unc ional capaci y ha enables independen li ing) due o he ac ha he o iginal measu e was designed o assess middle-aged adul s’ mo i es o physical ac i i y. The dimension ha is ela ed o skill im- p o emen and willingness o ge be e a an exe cise (mas- e y) is e y goal and igo ous physical ac i i y-o ien ed. Howe e , goal-o ien ed igo ous physical ac i i y is e y low in his age-g oup,22,23 and in he p esen sample,24 and, hus, his dimension was d opped om his s udy. Fu he , a dimension ela ed o he compe i i e/ego-o ien ed side o he physical ac i i y (ie, be i e and/o look be e han o he s) has also no been ound as a majo con ibu o o olde indi- iduals’ physical ac i i y beha io .11 Due o his, we adap ed his dimension as well o be e i he measu e o olde in- di iduals. Ob iously, hese modi ica ions may ha e conse- quences o he alidi y and he eliabili y o he measu e. The dimensions o he mo i a ion measu e in he MOBILETWIN s udy we e ( ep esen a i e i em shown in pa en heses): (a) enjoymen (“ha e a good ime and enjoy ex- e cising”), (b) socializing (“be wi h iends and/o do ac i i y wi h o he s”), (c) heal h (“be heal hie ”), (d) o he s’ expec a- ions (“con o m o o he s’ expec a ions”), (e) physical i ness (“be physically i ”), ( ) psychological well-being (“im- p o e psychological heal h/well-being”), and (g) appea ance (“main ain/imp o e appea ance”). The dimension o “enjoy- men ” ep esen s in insic mo i es, whe eas all he o he s ep esen aspec s o ex insic mo i a ion. Pa icipan s we e asked o a e each i em on a 5-poin Like scale anging om s ongly disag ee (a) o s ongly ag ee (e). All he i ems we e in oduced by he sen ence “I am physically ac i e o…”. The de elope s o he REMM ha e alida ed he o igi- nal REMM measu e.20 The Finnish e sion o he 73-i em REMM has also been alida ed,25 and he in e nal consis en- cies o he dimensions we e ound o be simila o hose ci ed by he de elope s o he o iginal measu e. 1412 | AALTONEN ET AL. 2.3 | Assessmen o physical ac i i y A ligh i-axial hip-wo n accele ome e was used o moni o he physical ac i i y beha io o he olde indi iduals (Hookie AM20, T axmee L d, Espoo).19 The pa icipan s ca ied he de- ices o se en consecu i e days du ing waking hou s (excep ba hing, aking a showe , and swimming). The ime c i e ion o adequa e accele ome e da a collec ion was ha he accel- e ome e s we e ca ied o a leas 10hou s pe day o 4days (non-wea ime was de ined as a sum o a leas 30minu es o consecu i e ze o accele a ion), which was me by 791 win in- di iduals. Ou o hese indi iduals, 80% ca ied he de ices o 7days and 14% o 6days. The a e age ime he pa icipan s ca ied he de ices was 14:01:44h:min:s/d. The analysis o aw accele a ion da a was based on al- go i hms ha employ he mean ampli ude de ia ion (MAD) o he aw accele a ion signal and he angle o pos u e es i- ma ion (APE) o he body. Toge he , hese me ics p o ide abou 90% accu acy in assessing he in ensi y, olume and dis ibu ion o daily physical ac i i y sepa a ing seden a y and s a iona y beha io s (ie, lying, si ing and s anding) om any physical ac i i y beha io .26-28 Fo he analysis, MAD and APE alues we e de e mined o each 6second epoch. The epoch-wise MAD alues we e exp essed in me abolic equi alen s (MET), and a 1-minu e exponen ial mo ing a - e age o he MET alues was calcula ed o es ima e inciden ene gy consump ion. The alidi y o MAD alues has been p o ed o be high 27—a co ela ion be ween he MAD alues o he esul an accele a ion was ound o be s ong wi h di ec ly measu ed inciden oxygen up ake (VO2) du ing walking o unning a an indoo ack ( >.9).27 This s ong associa ion allowed o con- e sion o MAD alues o inciden ene gy consump ion (MET alue). Acco ding o ecommended use,29 he ollowing cu - o poin s o di e en physical ac i i ies we e se : 1.5-3 MET o ligh ac i i ies, 3-6 MET o mode a e ac i i ies, and o e 6 MET o igo ous ac i i ies, and co esponding mean daily o al imes we e de e mined. Mode a e- o- igo ous physical ac i i y was cons uc ed by summing up he ime o mode a e and igo ous physical ac i i ies. We assigned ime spen on mode a e- o- igo ous physical ac i i y in o quin iles, because he olde pa icipan s in his s udy had di icul ies o each ha le el o physical ac i i y. Mo eo e , he ex emes o he dis ibu ion we e used in he analyses o compa e di e ences in mo i es o physical ac i i y be ween mode a ely o igo - ously physically inac i e (being he lowes i h) and ac i e (being he highes i h) olde indi iduals. 2.4 | S a is ical me hods We gene a ed he dis ibu ions, means, s anda d de ia ions, s anda d e o s, and/o con idence in e als o he basic cha ac e is ics and mo i es o physical ac i i y in olde men and women using S a a 14.1 so wa e (S a aCo p, College S a ion). The le el o signi icance was se a P<.05 in all analyses. The Wald es was used o assess whe he mo i es o physical ac i i y signi ican ly di e ed be ween he quin iles o accele ome e -measu ed ime spen on mode a e- o- ig- o ous physical ac i i y. Pa icipan s who had comple e in- o ma ion on bo h a iables we e selec ed o analyses. A squa e- oo - ans o ma ion o mode a e- o- igo ous physi- cal ac i i y was applied, wi h a success ul esul , o co ec p oblems wi h he no mali y assump ion. Because he pa icipan s o his s udy a e wins, he obse - a ions and hei e o e ms be ween membe s o a win pai may be co ela ed. The e o e, we adjus ed he s anda d e o s o win clus e ing. To assess he magni ude o he obse ed e ec , we calcula ed Cohen's d (ie, he di e ence be ween means di ided by he s anda d de ia ion). Because ou analy- ses included olde indi iduals wi h and wi hou physician-di- agnosed diseases, and because hose physician-diagnosed diseases ha e been shown o educe he le el o physical ac- i i y in his s udy popula ion,30 we conduc ed a sensi i i y analysis o examine whe he hese wo g oups o olde indi- iduals di e ed in e ms o mo i a ional ac o s. Because we had in o ma ion on bo h MZ and DZ wins, we we e able o es ima e gene ic and en i onmen al a ia- ions o mo i es o physical ac i i y in olde indi iduals by using in o ma ion on hei di e en gene ic ela edness. MZ co- wins ha e i ually he same DNA sequence, while DZ co- wins sha e, on a e age, 50% o he gene ic a ia ion.31 The simila i ies o MZ and DZ wins can be es ima ed by cal- cula ing in a-class co ela ion coe icien s (ICC). Familial in luences a e in e ed i he ICCs o DZ win pai s (gene - ically ull-siblings) di e om ze o. I he co ela ions o MZ wins a e signi ican ly highe han hose o DZ wins, gene ic in luences a e indica ed. I he co ela ions o DZ wins do no di e signi ican ly om hose o MZ wins bu bo h di e om ze o, hen sha ed en i onmen al in luences a e indica ed. Sha ed en i onmen al e ec s e e o all en- i onmen al in luences ha make co- wins alike (expec ed co ela ion 1.0 o bo h MZ and DZ wins). Unique en i on- men al e ec s deno e all en i onmen al in luences ha make membe s o a win pai unlike, since hey a e unco ela ed in bo h MZ and DZ wins, and include measu emen e o . Quan i a i e gene ic modeling was u he used o ex- amine he ex en o which a ia ion in mo i es o physi- cal ac i i y is accoun ed o by gene ic and en i onmen al in luences. Based on he p inciples desc ibed abo e, we decomposed he ai a ia ion in mo i e dimensions in o h ee componen s: addi i e gene ic a ia ion, sha ed en i- onmen al a ia ion, and unique en i onmen al a ia ion.32 The p opo ion o a ia ion accoun ed o by gene ic in- luences is called he i abili y. High he i abili y es ima es | 1413 AALTONEN ET AL. indica e a mino ole o en i onmen al in luences, whe eas low he i abili y es ima es a e sugges i e o a g ea e ole o sha ed o unsha ed en i onmen al in luences, on he di - e ences be ween indi iduals. Gene ic modeling began by compa ing di e en uni a ia e models o selec he bes - i - ing models o explain indi idual di e ences in mo i es o physical ac i i y (Table S2). Fi s , we de e mined whe he sha ed en i onmen al ac o s we e p esen o explain he a ia ion in mo i e dimensions. Uni a ia e model- i ing esul s e ealed ha d opping he sha ed en i onmen al pa h coe icien s did no lead o a signi ican de e io a ion o model i (P- alues anged om .16 o .96). Thus, he sha ed en i onmen al componen s we e d opped om he inal models. TABLE 1 Cha ac e is ics o s udy pa icipan s a he ime o accele ome e -based physical ac i i y assessmen Cha ac e is ics Men Women P- alue Age (yea s; mean (95% CI)) 72.9 (72.8-73.0) 72.8 (72.7-72.9) .20  n=413 n=434  BMI (kg/m2; mean (95% CI)) 26.2 (25.8-26.5) 26.1 (25.7-26.6) .91  n=412 n=427  MVPA pe day (minu es; mean (95% CI)) 47.9 (46.1-49.8) 43.1 (41.4-44.7) .005  n=385 n=406  Subjec i e heal h s a us   .40 Ve y good 11.1% 6.7%  Good 46.8% 48.9%  Fai 36.6% 41.0%  Poo 5.3% 3.5%  Ve y poo 0.2% 0%   n=415 n=434  Subjec i e physical i ness   .51 Ve y good 18.3% 12.3%  Good 46.6% 51.3%  Fai 28.1% 32.5%  Poo 6.5% 3.9%  Ve y poo 0.5% 0%   n=416 n=431  Gene al physical ac i i y le el   <.001 Mode a e/ igo ous 19.9% 16.6%  Ligh 68.1% 74.1%  Seden a y 12.0% 9.4%  Mainly in bed 0% 0%   n=408 n=428  Diseases ha hinde physical ac i i y   .92 Yes 24.8% 24.5%  No 75.2% 75.5%   n=411 n=428  Walking 2 kilome e s   .32 Manage o walk 80,0% 77.2%  Manage o walk, bu ha e mino di icul ies 14.2% 15.4%  Manage o walk, bu ha e majo di icul ies 3.9% 4.9%  Canno walk wi hou help 0.3% 0.5%  Canno walk e en wi h help 1.7% 2.1%   n=409 n=429  Abb e ia ions: BMI, body mass index; CI, con idence in e als; MVPA, mode a e- o- igo ous physical ac i i y. 1414 | AALTONEN ET AL. Fu he mo e, we es ed whe he he e we e di e ences in he absolu e and ela i e gene ic and en i onmen al a i- ances in mo i e dimensions be ween men and women. The esul s o absolu e and ela i e gene ic and en i onmen al a iances in mo i e dimensions did no show signi ican di e ences be ween men and women (P- alues anged om .92 o 1.00), excep ela i e gene ic and en i on- men al a iances in he mo i e dimensions o enjoymen (P=.002) and o he s’ expec a ions (P=.04). Howe e , he sepa a e analyses o men and women showed insu icien s a is ical powe ega ding hese mo i e dimensions; hus, bo h sexes we e analyzed oge he in gene ic modeling wi h ega d o he mo i e dimensions o enjoymen and o he s’ expec a ions. Gene ic modeling was ca ied ou wi h OpenMx ( e sion 2.0.1) so wa e, which is a package o ex ended s uc u al equa ion modeling on he R s a is ical pla o m. All he mod- els we e i ed o he aw da a using maximum-likelihood algo i hms (allowing inclusion also o wins wi hou in o - ma ion on hei co- wins) and ea ing unobse ed da a as missing-a - andom.33 2.5 | E hics o he s udy The E hics Commi ee o he Hospi al Dis ic o Sou hwes Finland app o ed he MOBILETWIN s udy p o ocol on May 20, 2014. The pa icipan s p o ided w i en in o med TABLE 2 Dis ibu ion and means o mo i e dimensions among s udy pa icipan s Mo i e dimension Like scale esponse op ions and assigned sco es in pa en heses  S ongly disag ee (1) Disag ee (2) Neu al (3) Ag ee (4) S ongly ag ee (5) Mean (95% CI) Bo h men and women       Enjoymen (n=830) 22 29 135 319 325 4.08 (4.01-4.15) Socializing (n=825) 9 10 52 262 504 3.59 (3.51-3.66) Heal h (n=837) 9 10 52 262 504 4.48 (4.43-4.54) O he s’ expec a ions (n=819) 375 172 168 75 29 2.04 (1.96-2.12) Physical i ness (n=835) 5 4 48 229 549 4.57 (4.53-4.62) Psychological well-being (n=836) 14 6 66 287 463 4.41 (4.26-4.46) Appea ance (n=830) 253 135 243 139 60 2.54 (2.45-2.63) Men       Enjoymen (n=408) 14 16 63 154 161 4.06 (3.95-4.16) Socializing (n=407) 30 45 106 141 85 3.51 (3.39-3.62) Heal h (n=408) 4 5 30 134 235 4.45 (4.37-4.52) O he s’ expec a ions (n=401) 150 87 95 49 20 2.26 (2.14-2.38) Physical i ness (n=408) 3 2 20 120 263 4.56 (4.50-4.63) Psychological well-being (n=409) 9 3 39 147 211 4.34 (4.26-4.42) Appea ance (n=408) 142 66 122 58 20 2.38 (2.26-2.50) Women       Enjoymen (n=422) 8 13 72 165 164 4.10 (4.01-4.19) Socializing (n=418) 23 40 102 142 111 3.67 (3.56-3.77) Heal h (n=429) 5 5 22 128 269 4.52 (4.45-4.59) O he s’ expec a ions (n=418) 225 85 73 26 9 1.83 (1.72-1.93) Physical i ness (n=427) 2 2 28 109 286 4.58 (4.52-4.65) Psychological well-being (n=427) 5 3 27 140 252 4.48 (4.41-4.55) Appea ance (n=422) 111 69 121 81 40 2.69 (2.57-2.82) Abb e ia ion: CI, con idence in e als. | 1415 AALTONEN ET AL. consen . All s udy me hods we e ca ied ou in acco dance wi h he app o ed guidelines and he Helsinki Decla a ion. 3 | RESULTS The basic desc ip i e cha ac e is ics o he s udy pa icipan s a e shown in Table1. Men and women did no di e wi h ega d o hei body mass index (P=.91). Men we e mo e physically ac i e based on accele ome e -measu ed ime spen on mode a e- o- igo ous physical ac i i y han women (48% s 43%, P=.005). Based on he su ey ques ions, a ela i ely highe amoun o men han women epo ed engag- ing in mode a e- o- igo ous physical ac i i y (20% s 17%, P<.001), which is in line wi h he accele ome e da a. On he o he hand, men epo ed mo e seden a y beha io (si ing ime) han women (12% s 9%), while women epo ed mo e ligh physical ac i i y han men (74% s 68%). Howe e , he e we e no di e ences in sel - epo ed physical i ness o subjec i e heal h s a us be ween olde men and women (P=.51 and P=.40, espec i ely), no did he exis ence o physician-diagnosed diseases ha hinde physical ac i i y o he abili y o walk 2 kilome e s di e be ween sexes (P=.92 and P=.32, espec i ely). 3.1 | Mo i es o physical ac i i y We analyzed he mo i es o physical ac i i y bo h in he pooled g oups o men and women and sepa a ely o men and women (Table2). When men and women we e analyzed oge he , he mos impo an mo i e o physical ac i i y wi h he highes mean sco e was physical i ness, ollowed by heal h main enance and psychological well-being (Table2). The leas impo an mo i es wi h he lowes mean sco es o physical ac i i y among all indi iduals we e ela ed o con- o ming o o he s’ expec a ions and appea ance. Despi e he ac ha olde men and women epo ed he same mo i es o physical ac i i y o be impo an sex-speci ic analyses e- ealed ha he e we e some di e ences in mo i es o physi- cal ac i i y (Table3). Con o ming o he people's expec a ions was signi ican ly mo e impo an o men (P<.001, Cohen's d=0.38), while appea ance (P=.001 Cohen's d=−0.24) and psychological well-being (P=.02, Cohen's d=−0.17) we e highligh ed by women. 3.2 | Mo i es o physical ac i i y and accele ome e -measu ed physical ac i i y Mos o he mo i e dimensions we e s a is ically signi i- can ly and posi i ely co ela ed wi h each o he (Table4). Fu he analyses e ealed ha accele ome e -measu ed ime spen on mode a e- o- igo ous physical ac i i y was also s a is ically signi ican ly and posi i ely co ela ed wi h i e mo i a ional ac o s ou o se en (Table4). The s onges associa ion was ound be ween mode a e- o- igo ous physi- cal ac i i y and enjoymen ( =.38) ollowed by physical i ness ( =.29), heal h main enance ( =.25), and psycho- logical well-being ( =.25). Though he associa ion wi h he mo i e dimension o socializing was signi ican , i was qui e weak ( =.09, P<.05). Howe e , mo i e dimensions e- la ed o con o ming o o he s’ expec a ions and appea ance we e no ound o be associa ed wi h accele ome e -meas- u ed ime spen on mode a e- o- igo ous physical ac i i y in olde indi iduals. Di e ences in he mo i e dimensions be ween he quin- iles o accele ome e -measu ed ime spen on mode - a e- o- igo ous physical ac i i y by sex a e p esen ed in Table5. The able shows ha he e a e s a is ically signi ican di e ences in he mo i e dimensions o enjoymen , heal h main enance, physical i ness, and psychological well-being be ween he di e en quin ile g oups o accele ome e -mea- su ed ime spen on mode a e- o- igo ous physical ac i i y. This was ue o bo h sexes. When we compa ed hose in he lowes and highes quin iles o accele ome e -measu ed ime TABLE 3 Di e ences in he mo i e dimensions o physical ac i i y be ween olde men and women Mo i e dimension Men Mean (95% CI) Women Mean (95% CI) Mean di e ence (95% CI) p- alue E ec size Cohen's d Enjoymen 4.06 (3.95-4.17) 4.10 (4.01-4.19) −0.04 (−0.18-0.10) .57 −0.04 Socializing 3.51 (3.39-3.63) 3.67 (3.55-3.78) −0.16 (−0.32-0.00) .06 −0.14 Heal h 4.45 (4.37-4.53) 4.52 (4.44-4.59) −0.07 (−0.18-0.04) .22 −0.09 O he s’ expec a ions 2.26 (2.13-2.39) 1.83 (1.71-1.94) 0.43 (0.26-0.60) <.001 0.38 Physical i ness 4.56 (4.49-4.63) 4.58 (4.51-4.65) −0.02 (−0.11-0.08) .73 −0.03 Psychological well-being 4.34 (4.25-4.43) 4.48 (4.41-4.55) −0.14 (−0.25-−0.03) .02 −0.17 Appea ance 2.38 (2.26-2.51) 2.69 (2.56-2.83) −0.31 (−0.49-−0.13) .001 −0.24 No e: The numbe o olde win indi iduals in he mo i e dimensions anged be ween 401 and 409 in men and be ween 418 and 429 in women. Abb e ia ion: CI, con idence in e al. 1416 | AALTONEN ET AL. spen on mode a e- o- igo ous physical ac i i y (Table6), we ound ha mos o he mo i e dimensions we e signi ican ly mo e impo an o mode a ely o igo ously physically ac- i e olde indi iduals ( he highes quin ile) han o inac i e ( he lowes quin ile) olde indi iduals (P- alues anged om <.001 o .03). Howe e , he social aspec o physical ac i i y (P=.22, Cohen's d=0.84) and o he s’ expec a ions o be physically ac i e (P=.58, Cohen's d=−0.10) along wi h ap- pea ance (P=.63, Cohen's d=0.08) did no di e be ween mode a ely o igo ously ac i e and inac i e olde men. In women, con o ming o o he s’ expec a ions (P=.03, Cohen's d=−0.36) was mo e impo an o mode a ely o igo ously physically inac i e olde women and only he mo i e dimen- sion o appea ance (P=.94, Cohen's d=−0.01) did no di - e be ween mode a ely o igo ously physically ac i e and inac i e olde women. The accele ome e -measu ed ime o mode a e- o- igo ous physical ac i i ies o ac i e olde in- di iduals anged be ween 61 and 98min/d and o inac i e coun e pa s be ween 2 and 30min/d. Because ou analyses included olde indi iduals wi h and wi hou physician-diagnosed diseases hinde ing physical ac- i i y, we conduc ed a sensi i i y analysis o examine whe he hese wo g oups o olde indi iduals di e ed in e ms o mo i a ional ac o s (Table S1). The esul s showed no sig- ni ican di e ences in mo i a ional ac o s be ween olde indi iduals wi h and wi hou physician-diagnosed diseases hinde ing physical ac i i y excep he mo i e dimension e- la ed o he enjoymen o physical ac i i y—olde indi iduals who epo ed ha hey had physician-diagnosed diseases ha hinde hei physical ac i i y we e mo i a ed signi ican ly mo e by enjoymen han olde indi iduals wi h no such dis- eases (P=.001). 3.3 | Gene ic and en i onmen al con ibu ions o mo i es o physical ac i i y The in a-class co ela ion coe icien s and quan i a i e gene ic modeling o mo i es o physical ac i i y we e p o ided o be e unde s and he ole o gene ic and en i- onmen al in luences in explaining indi idual di e ences in mo i es o physical ac i i y. The in a-class co ela ion coe icien s we e sys ema ically highe , hough no all we e s a is ically signi ican ly highe , among MZ pai s ( om 0.03 o 0.60) han among DZ pai s ( om −0.01 o 0.32) wi h a ew excep ions (Table7) indica ing ha gene ic ac o s may be impo an in hese mo i es o physical ac i i y in olde indi- iduals. Mos DZ win co ela ions ha we e lowe han MZ win co ela ions we e, howe e , less han hal he MZ co - ela ions, which sugges s ha en i onmen al ac o s sha ed by co- wins play a less impo an ole in mo i es o physical ac i i y. This sugges ion was con i med by he ac ha he bes - i ing gene ic models we e ound o be models wi hou TABLE 4 Polycho ic co ela ions be ween mo i e dimensions and polyse ial co ela ions be ween mo i e dimensions and accele ome e -measu ed ime spen on mode a e- o- igo ous physical ac i i y among olde indi iduals  Enjoymen Socializing To main ain heal h O he s’ expec a ions Physical i ness Psychological well-being Appea ance Accele ome e - measu ed MVPA Enjoymen 1 (n=830)        Socializing 0.57*** (n=820) 1 (n=825)       Heal h 0.67*** (n=829) 0.42*** (n=825) 1 (n=837)      O he s’ expec a ions 0.06 (n=815) 0.23*** (n=812) 0.01 (n=817) 1 (n=819)     Physical i ness 0.63*** (n=828) 0.35*** (n=823) 0.86*** (n=832) −0.02 (n=817) 1 (n=835)    Psychological well-being 0.60*** (n=829) 0.43*** (n=825) 0.71*** (n=834) 0.04 (n=818) 0.84*** (n=832) 1 (n=836)   Appea ance 0.29*** (n=825) 0.31*** (n=823) 0.34*** (n=828) 0.46*** (n=818) 0.29*** (n=827) 0.41*** (n=829) 1 (n=830)  Accele ome e -measu ed MVPA 0.38*** (n=773) 0.09* (n=771) 0.25*** (n=780) −0.06 (n=764) 0.29*** (n=779) 0.25*** (n=779) 0.02 (n=774) 1 (n=791) Abb e ia ion: MVPA, mode a e- o- igo ous physical ac i i y. *P<.05. ***P<.001.