Active commuting and depression symptoms in adults : a systematic review

In e na ional Jou nal o En i onmen al Resea ch and Public Heal h A icle Ac i e Commu ing and Dep ession Symp oms in Adul s: A Sys ema ic Re iew Adilson Ma ques 1,2,3,* , Miguel Pe al a 1,2 , Dua e Hen iques-Ne o 1, Diana F asquilho 4, Él io Rubio Gou ei a 5,6 and Diego Gomez-Baya 3,7 1CIPER, Faculdade de Mo icidade Humana, Uni e sidade de Lisboa, Es ada da Cos a, 1499-002 C uz Queb ada, Po ugal; [email p o ec ed] (M.P.); [email p o ec ed] (D.H.-N.) 2ISAMB, Faculdade de Medicina, Uni e sidade de Lisboa, 1649-028 Lisboa, Po ugal 3Escuela de Doc o ado, Uni e si y o Huel a, 21071 Huel a, Spain; [email p o ec ed] 4Champalimaud Clinical Cen e , Champalimaud Cen e o he Unknown, Champalimaud Founda ion, 1400-038 Lisbon, Po ugal; diana. asquilho@ esea ch. champalimaud.o g 5Depa amen o de Educação Física e Despo o, Uni e sidade da Madei a, 9020-105 Funchal, Po ugal; e ubiog@s a .uma.p 6In e ac i e Technologies Ins i u e, LARSyS, 9020-105 Funchal, Po ugal 7Depa men o Social, De elopmen al and Educa ional Psychology, Uni e si y o Huel a, 21007 Huel a, Spain *Co espondence: [email p o ec ed]; Tel.: (+351)-214149100 Recei ed: 8 Janua y 2020; Accep ed: 5 Feb ua y 2020; Published: 6 Feb ua y 2020   Abs ac : Physical ac i i y (PA) is sugges ed o ha e a p o ec i e e ec agains dep ession. One way o engaging in PA is h ough ac i e commu ing. This e iew summa ises he li e a u e ega ding he ela ionship be ween ac i e commu ing and dep ession among adul s and olde adul s. A sys ema ic e iew o s udies published up o Decembe 2019, pe o med in acco dance wi h he P e e ed Repo ing I ems o Sys ema ic Re iews and Me a-Analysis guidelines, was conduc ed using h ee da abases (PubMed, Scopus, and Web o Science). A o al o se en a icles we e iden i ied as ele an . The esul s om hese s udies we e inconsis en . Only wo p esen ed a signi ican ela ionship be ween ac i e commu ing and dep ession symp oms. In hose wo s udies, swi ching o mo e ac i e modes o a el and walking long dis ances we e nega i ely ela ed o he likelihood o de eloping new dep essi e symp oms. In he o he i e s udies, no signi ican associa ion be ween ac i e a el o ac i e commu ing and dep ession was ound. The ela ionship be ween ac i e commu ing and dep ession symp oms in adul s is no clea . Mo e s udies on his opic a e necessa y in o de o unde s and i ac i e commu ing can be used as a public heal h s a egy o ackle men al heal h issues such as dep ession. Keywo ds: ac i e a el; walking; cycling; men al heal h 1. In oduc ion Dep ession is a men al heal h p oblem ha a ec s mo e han 300 million adul s wo ldwide [ 1 ]. The p e alence o dep ession has been ising globally, and i was p ojec ed o be he second-la ges cause o he bu den o disease by 2020 [ 2 ]. Dep ession is al eady he g ea es non-communicable disease con ibu ing o he loss o heal h [ 1 ], due o i s associa ion wi h como bidi ies [ 3 ], isk o suicide [ 4 ], and p ema u e mo ali y [ 5 ]. Mos cases o dep ession a e ea ed using pha maco he apy and psycho he apy [ 6 , 7 ]. Howe e , bo h ypes o he apies a e expensi e and inc ease heal h cos s o heal h ca e sys ems [8]. Physical ac i i y (PA) is sugges ed o ha e a p o ec i e e ec agains dep ession and a posi i e e ec on he ea men o dep ession in non-clinical and clinical popula ions, ega dless o sex and In . J. En i on. Res. Public Heal h 2020,17, 1041; doi:10.3390/ije ph17031041 www.mdpi.com/jou nal/ije ph In . J. En i on. Res. Public Heal h 2020,17, 1041 2 o 11 age [ 9 , 10 ]. As non-pha macology he apy, PA does no show d ug side e ec s such as weigh gain, insomnia, d y mou h, o wi hd awal symp oms [ 11 ]. On he con a y, besides he posi i e bene i s on dep ession, i is also associa ed wi h posi i e public heal h gains such as physical pe o mance and ca dio ascula imp o emen s [11]. PA can be pe o med in se e al con ex s, including spo s ac i i ies, occupa ional con ex , leisu e- ime ac i i ies, domes ic ac i i ies, and in ac i e commu ing. Fo some adul s, walking o cycling while commu ing o wo k o o o he places seems o be impo an o enhancing hei daily le els o PA [ 12 , 13 ], because is likely o be sus ained when inco po a ed in o people’s daily ou ine [ 14 ]. As a esul , ac i e commu ing has been ela ed o heal h and wellbeing imp o emen [ 15 , 16 ] and associa ed wi h ewe anspo - ela ed ca bon emissions [ 17 ]. Thus, i can be expec ed ha commu ing ac i ely migh also be ela ed o dep ession symp oms. Howe e , so a , he e is no clea e idence o he ela ionship be ween ac i e a el o ac i e commu ing and dep ession symp oms. The e o e, he pu pose o his s udy was o e iew he li e a u e ega ding he ela ionship be ween PA in ac i e commu ing and dep ession among adul s and olde adul s. 2. Ma e ials and Me hods This sys ema ic e iew was pe o med in acco dance wi h he P e e ed Repo ing I ems o Sys ema ic Re iews and Me a-Analysis (PRISMA) guidelines [18]. 2.1. Inclusion C i e ia A icles ha p esen ed a ela ionship be ween ac i e a el and ac i e commu ing and dep ession symp oms published in pee - e iewed jou nals up o Decembe 2019 we e eligible o inclusion. Eligibili y c i e ia included he ollowing: (1) c oss-sec ional, p ospec i e, and coho s udies (s udy design c i e ion); (2) ou comes included dep ession o dep essi e symp oms (ou come measu e c i e ion); (3) ac i e a el o ac i e commu ing ( ela ionship c i e ion); (4) adul s and olde adul s (pa icipan s c i e ion); (5) a icles published in English, Po uguese, o Spanish (language c i e ion); and (6) a icles ha did no mee he inclusion c i e ia o did no include indings ela ed o he inclusion c i e ia we e excluded (exclusion c i e ia). 2.2. Sea ch S a egy S udies published up o Decembe 2019 we e iden i ied by sea ching in elec onic da abases. The sea ch was unde aken in he PubMed, Scopus, and Web o Science da abases. A icles ha assessed he ela ionship be ween ac i e a el o ac i e commu ing and dep ession we e included in his e iew. The sea ch was pe o med using he ollowing combina ion o e ms: a el* OR anspo * OR commu e* OR cycle OR cycling OR bicycle* OR bike* OR walk* OR/AND dep ess* OR men al heal h OR psychological heal h OR anxie y OR psychological unc ion*. Sea ch e ms we e de ined p e iously and used in each da abase o iden i y a icles o e iew. Two e iewe s (A.M. and M.P.) independen ly sc eened i les and abs ac s o iden i y s udies ha me he inclusion c i e ia. Duplica e a icles we e emo ed. Rele an a icles we e e ie ed o a ull ead. Two au ho s (A.M. and M.P.) e iewed he ull ex o po en ial s udies, and decisions o include o exclude s udies in he e iew we e made by consensus. Disag eemen s we e sol ed by consensus and, when necessa y, a hi d e iewe se ed as a judge (D.H.-N. o D.F.). 2.3. Da a Ex ac ion and Ha monisa ion Based on PRISMA [ 18 ], a da a ex ac ion o m was de eloped. The ollowing in o ma ion was ex ac ed om each a icle: au ho s’ name and yea o publica ion; s udy design; coun y; sample cha ac e is ics (numbe o pa icipan s, sex, and age); he ins umen o assessing dep ession symp oms; he ins umen o assessing ac i e a el o ac i e commu ing; main esul s; and s udy quali y. The ex ac ion was pe o med by one au ho (A.M.), and coding was e i ied by wo au ho s (M.P. and D.H.-N.). In . J. En i on. Res. Public Heal h 2020,17, 1041 3 o 11 2.4. S udy Quali y and Risk o Bias S udy me hodological quali y was assessed using a checklis c i e ia om he Quali y Assessmen Tool o Quan i a i e S udies [ 19 ]. The checklis comp ises 19 i ems, assessing eigh me hodological domains: selec ion bias, s udy design, con ounde s, blinding, da a collec ion me hods, wi hd awals and d opou s, in e en ion in eg i y, and analyses. Each sec ion is classi ied as s ong, mode a e, and wi h weak me hodological quali y. Then, a global a ing is de e mined based on he sco es o each componen . Two esea che s a ed he a icles (M.P. and D.F.) in each domain and he s udies’ o e all quali y. Disc epancies we e esol ed by consensus. 2.5. Syn hesis o Resul s This e iew analysed he ela ionship be ween ac i e a el o ac i e commu ing and dep ession. The de ails o each s udy, including design, pa icipan cha ac e is ics and sample size, measu es, main esul s, and s udy quali y, a e p esen ed in a consis en manne . 3. Resul s 3.1. Sea ch Resul s Figu e 1p esen s he low ci a ion h ough he sys ema ic e iew p ocess. The sys ema ic sea ch yielded 3938 publica ions. A e excluding he duplica es (n=2246), he i le and abs ac o 1692 a icles we e sc eened. A e elimina ing 1606 a icles a he i le and abs ac le el, 86 a icles emained and we e subsequen ly ead. F om he 86 a icles, 35 we e elimina ed o ha ing a di e en ou come, 40 we e elimina ed because hey we e no ocused on ac i e a el o ac i e commu ing, h ee we e elimina ed o no being empi ical s udies, and one was dele ed because i was w i en in Manda in. Thus, se en a icles we e iden i ied as ele an . In . J. En i on. Res. Public Heal h 2020, 17, x FOR PEER REVIEW 3 o 12 2.4. S udy Quali y and Risk o Bias S udy me hodological quali y was assessed using a checklis c i e ia om he Quali y Assessmen Tool o Quan i a i e S udies [19]. The checklis comp ises 19 i ems, assessing eigh me hodological domains: selec ion bias, s udy design, con ounde s, blinding, da a collec ion me hods, wi hd awals and d opou s, in e en ion in eg i y, and analyses. Each sec ion is classi ied as s ong, mode a e, and wi h weak me hodological quali y. Then, a global a ing is de e mined based on he sco es o each componen . Two esea che s a ed he a icles (M.P. and D.F.) in each domain and he s udies’ o e all quali y. Disc epancies we e esol ed by consensus. 2.5. Syn hesis o Resul s This e iew analysed he ela ionship be ween ac i e a el o ac i e commu ing and dep ession. The de ails o each s udy, including design, pa icipan cha ac e is ics and sample size, measu es, main esul s, and s udy quali y, a e p esen ed in a consis en manne . 3. Resul s 3.1. Sea ch Resul s Figu e 1 p esen s he low ci a ion h ough he sys ema ic e iew p ocess. The sys ema ic sea ch yielded 3938 publica ions. A e excluding he duplica es (n = 2246), he i le and abs ac o 1692 a icles we e sc eened. A e elimina ing 1606 a icles a he i le and abs ac le el, 86 a icles emained and we e subsequen ly ead. F om he 86 a icles, 35 we e elimina ed o ha ing a di e en ou come, 40 we e elimina ed because hey we e no ocused on ac i e a el o ac i e commu ing, h ee we e elimina ed o no being empi ical s udies, and one was dele ed because i was w i en in Manda in. Thus, se en a icles we e iden i ied as ele an . Figu e 1. Flow diag am o s udy selec ion. 3.2. S udy Cha ac e is ics Table 1 p esen s he s udies’ cha ac e is ics. F om he se en a icles, i e had p ospec i e designs and wo had c oss-sec ional designs. In o al, all s udies combined included 47,300 adul s aged o e 18 yea s. Geog aphically, wo we e pe o med in he Uni ed S a es; wo in Japan; one in Canada; one in he Uni ed Kingdom; and one in se e al La in Ame ican coun ies (A gen ina, Boli ia, B azil, Figu e 1. Flow diag am o s udy selec ion. 3.2. S udy Cha ac e is ics Table 1p esen s he s udies’ cha ac e is ics. F om he se en a icles, i e had p ospec i e designs and wo had c oss-sec ional designs. In o al, all s udies combined included 47,300 adul s aged o e 18 yea s. Geog aphically, wo we e pe o med in he Uni ed S a es; wo in Japan; one in Canada; one in he Uni ed Kingdom; and one in se e al La in Ame ican coun ies (A gen ina, Boli ia, B azil, In . J. En i on. Res. Public Heal h 2020,17, 1041 4 o 11 Colombia, Ecuado , M é xico, Panam á , Pe u, U uguay, and Venezuela). The Cen e o Epidemiologic S udies Dep ession Scale (CES-D) was he mos used scale o assess dep ession symp oms (in i e a icles). In mos a icles, he walking dis ance was sel - epo ed and included leisu e ime and ac i e commu ing. The se en s udies we e conside ed o mode a e quali y. The olde s udy included in he sys ema ic e iew was published in 2010, and i examined he e ec o walking on inciden dep essi e symp oms in Japanese-Ame ican olde men wi h and wi hou ch onic disease [ 20 ]. The au ho s used a coho o men om Hawaii bo n be ween 1900 and 1919. Se e al ounds o examina ions we e pe o med; he se en h (which was used in he s udy) was in 1999–2000. The CES-D 11 scale was used o sc een o dep essi e symp oms, and walking ac i i y was sel - epo ed. I was ound ha olde men, wi hou ch onic diseases, in he in e media e (odds a io (OR) =0.52, 95% con idence in e al (CI): 0.32–0.83) and highes (OR =0.61, 95% CI: 0.39–0.97) walking g oups had signi ican ly lowe odds o de eloping eigh -yea inciden dep essi e symp oms. Pos e io ly, using a one-yea ollow-up longi udinal su ey, Kai e al. [ 21 ] examined he p ospec i e associa ion o walking o wo k wi h dep essi e symp oms among 634 Japanese wo ke s. Dep ession symp oms we e sc eened by he CES-D scale, and walking ac i i y was sel - epo ed. The au ho s epo ed ha baseline mean walking o wo k ime was 29.2 ± 18.0 min pe day. Walking o wo k e iles we e calcula ed; mean alues o he low, medium, and high walking o wo k ime e iles we e 13.8 ± 7.0, 29.6 ± 1.9, and 49.7 ± 16.1, espec i ely. No signi ican associa ion was ound be ween he dep essi e symp oms and e iles o walking o wo k du a ion. In an e o o de e mine whe he walking o dep essi e symp oms we e he s onge p edic o o each o he , Julien e al. [ 22 ] examined he longi udinal associa ions be ween walking and dep essi e symp oms in a popula ion-based sample o 498 Canadian u ban-dwelling olde adul s. Fou epea ed measu es o e a i e-yea pe iod we e pe o med. The ge ia ic dep ession scale was used o assess dep essi e symp oms, and walking was sel - epo ed. Al hough dep essi e symp oms p edic ed walking equency, walking equency did no p edic dep essi e symp oms a subsequen ime poin s. La e , a c oss-sec ional s udy based on he Na ional Su ey o Ame ican Li e in es iga ed he odds o ha ing dep essi e symp oms acco dingly o walking equency in 2978 A ican-Ame ican adul s [ 23 ]. Walking was measu ed by sel - epo ed equency, and dep essi e symp oms we e measu ed wi h modi ied e sions o he CES-D scale. In his s udy, women who epo ed walking o en p esen ed lowe odds o dep essi e symp oms han women who epo ed ne e walking (OR =0.56, 95% CI: 0.38–0.82); howe e , no signi ican esul s we e ound o men. In he same yea , Kuwaha a e al. [ 24 ] published a coho s udy o 29,082 Japanese wo ke s, examining he p ospec i e associa ions o physical ac i i y du ing commu ing wi h he isk o dep essi e symp oms. Du ing a mean ollow-up o 4.7 yea s, 6177 adul s de eloped dep essi e symp oms. Physical ac i i y and dep essi e symp oms (simila o he CES-D scale) we e assessed by a sel - epo ed ques ionnai e. The haza d a io (HR) o indi iduals who s and o walk du ing wo k was 0.86 (95% CI: 0.81–0.92) compa ed wi h seden a y wo ke s. Howe e , walking o and om wo k was no ound o be associa ed wi h he isk o dep essi e symp oms. Mo e ecen ly, Kno e al. [ 25 ] examined whe he changes in commu e mode we e associa ed wi h di e ences in he se e i y o dep essi e symp oms in 5474 wo king adul s om he Uni ed Kingdom in a popula ion-base p ospec i e coho s udy (mean ollow-up, 4.65 yea s). Mode, equency, and dis ance o commu ing o wo k we e sel - epo ed. The se e i y o dep essi e symp oms was assessed using he wo-i em Pa ien Heal h Ques ionnai e. Baseline asymp oma ic adul s who al e ed om inac i e o ac i e commu ing o wo k epo ed less se e e symp oms a ollow-up han hose who emained inac i e ( β = − 0.10, 95% CI: − 0.20–0.00). Howe e , among baseline symp oma ic adul s, longe ips o wo k we e associa ed wi h wo se symp oms a ollow-up ( β =0.64, 95% CI: 0.13–1.16). The mos ecen s udy examined he c oss-sec ional associa ions be ween commu e pa e ns and men al heal h in 5438 adul s using su ey da a om 11 La in Ame ican ci ies [ 26 ]. Commu ing was sel - epo ed, and dep ession was measu ed using he CES-D scale. No associa ion was ound be ween non-mo o ised modes o a el and dep ession. In . J. En i on. Res. Public Heal h 2020,17, 1041 5 o 11 Table 1. Cha ac e is ics o he s udies. Au ho , Yea S udy Design Sample and Coun y Dep ession Measu e Ac i e Commu ing Measu e Obse a ions Main Resul s Me hodological Quali y Smi h e al., 2010 P ospec i e coho . n=3196 Japanese-Ame ican men who we e in ol ed in he Honolulu Hea P og am aged 71–93 yea s; mean age 77 yea s. Uni ed S a es. Cen e o Epidemiologic S udies Dep ession Scale (CES-D 11) Pa icipan s epo ed how many ci y blocks hey walked each day. Blocks we e con e ed in o miles using 12 blocks pe mile as a con e sion ac o . This assessmen was de eloped om he Ha a d Alumni Su ey. Walking dis ance included leisu e- ime and ac i e commu ing. Those who walked mo e had signi ican ly lowe a es o p e alen dep essi e symp oms in c oss-sec ional analyses. Elde ly men, wi hou ch onic diseases, who walked longe dis ances pe day we e less likely o de elop new dep essi e symp oms o e eigh yea s o ollow-up. Mode a e Kai e al., 2011 P ospec i e (1-yea ollow-up). n=634: 536 men, 98 women aged 20–60 yea s; 36.7 ±9.2 yea s. Japan. Cen e o Epidemiologic S udies Dep ession Scale (CES-D) The du a ion o leisu e- ime physical ac i i y and commu ing by walking we e measu ed using a sel - epo ques ionnai e. Walking dis ance included leisu e- ime physical ac i i y and commu ing o wo k. The adjus ed odds a io (OR) o dep essi e symp oms in he highes e ile o leisu e- ime physical ac i i y was 50% lowe (OR =0.50, 95% CI: 0.26–0.97) han hose in he lowes e ile. In con as , no signi ican associa ion was ound be ween he isk o dep essi e symp oms and du a ion o commu ing by walking. Mode a e Julien e al., 2013 P ospec i e (5-yea s ollow-up). VoisiNuAge S udy. n=498: 236 men, 262 women aged 68–84 yea s (74.86 ±4.18 men and 74.90 ±3.97 women). Canada. Ge ia ic Dep ession Scale (GDS) Physical Ac i i y Scale o he Elde ly (PASE). Walking dis ance included leisu e- ime and ac i e commu ing. Dep essi e symp oms p edic ed walking equency (highe dep essi e symp oms we e ela ed o ewe walking days), bu walking equency did no p edic dep essi e symp oms a subsequen ime poin s. Mode a e To es a el., 2015 C oss-sec ional. Na ional Su ey o Ame ican Li e (NSAL). n=2978: 1903 men, 1075 women aged ≥18 yea s. Uni ed s a es. Cen e o Epidemiologic S udies Dep ession Scale (CES-D) Walking was measu ed wi h esponses o one ques ion om he Ame icans’ Changing Li es ques ionnai e. Walking dis ance included leisu e- ime and ac i e commu ing. Women who epo ed o en walking had lowe odds o dep essi e symp oms han women who epo ed ne e walking (OR = 0.56, 95% CI: 0.38–0.82). Walking equency was no ela ed o dep essi e symp oms in men. Mode a e In . J. En i on. Res. Public Heal h 2020,17, 1041 6 o 11 Table 1. Con . Au ho , Yea S udy Design Sample and Coun y Dep ession Measu e Ac i e Commu ing Measu e Obse a ions Main Resul s Me hodological Quali y Kuwaha a e al., 2015 P ospec i e (5-yea ollow-up). Japan Epidemiology Collabo a ion on Occupa ional Heal h (J-ECOH) S udy. n=29,082 wo ke s: 24,676 men, 4406 women aged 20–64 yea s; mean age 42.7 yea s. Japan. Epidemiologic S udies Dep ession Scale (CES-D), Sel -Ra ing Dep ession Scale (SDS) Pa icipan s we e asked whe he hey egula ly engaged in any physical ac i i y du ing leisu e. Du a ion o walking o and om wo k was sel - epo ed and ca ego ised as <20minu es, 20–40 min, and ≥ 40 min. I assessed leisu e- ime physical ac i i y, physical ac i i y a wo k, and ac i e commu ing. Leisu e exe cise showed a U-shaped associa ion wi h he isk o dep essi e symp oms. Walking o and om wo k was no associa ed wi h dep essi e symp oms. Mode a e Kno e al., 2018 Popula ion-based p ospec i e coho . UK Biobank. n=5474, aged 37–73 yea s. Uni ed Kingdom. Pa ien Heal h Ques ionnai e (PHQ-2) Pa icipan s epo ed he equency o ips om home o wo k ( ips/week); he dis ance a elled (miles); and he mode o anspo used (“ca o mo o ehicle” (he ea e ”ca ” o simplici y), ”public anspo ”, and ”walk” and/o ”cycle”). Pape was ocused on modes o a el o wo k. Pa icipan s who we e asymp oma ic a baseline and swi ched o mo e ac i e modes o commu ing ended o epo a lowe se e i y o symp oms a ollow-up han hose who con inued o a el inac i ely (β= −0.10, 95% CI: −0.20–0.00). Among commu e s who we e symp oma ic a baseline, longe jou neys we e associa ed wi h wo se symp oms a ollow-up (β=0.64, 95% CI: 0.13–1.16). Shi ing om exclusi e ca use owa ds mo e ac i e commu ing may help p e en and a enua e dep essi e symp oms in wo king adul s. Mode a e Wang e al., 2019 C oss-sec ional. La Encues a CAF 2016. n=5438. A gen ina, Boli ia, B azil, Colombia, Ecuado , M é xico, Panam á , Pe u, U uguay, and Venezuela. Cen e o Epidemiologic S udies Dep ession Scale (CES-D) Pa icipan s epo ed commu ing ime du ing a “no mal day”, unconges ed commu ing ime, and a el mode commonly used o he commu e. Di e en modes o a el we e assessed. E e y 10 mo e minu es o commu ing ime is associa ed wi h a 0.5% (p=0.011) highe p obabili y o sc eening posi i ely o dep ession. The e we e no ound any signi ican associa ions be ween non-mo o ised modes o a el and dep ession. Mode a e In . J. En i on. Res. Public Heal h 2020,17, 1041 7 o 11 4. Discussion This e iew summa ises s udies, published up o Decembe 2019, which me he de ined c i e ia. Se en s udies we e sys ema ically e iewed o add ess he ela ionship be ween ac i e commu ing and dep ession symp oms. In gene al, al hough some s udies demons a e ha ac i e commu ing migh be nega i ely ela ed o dep ession symp oms, he e idence sugges s ha he ac i e commu ing and dep ession ela ionship is unclea . F om he se en s udies sc eened, only wo p esen ed a signi ican ela ionship be ween ac i e commu ing and dep ession symp oms. In hose wo s udies [ 20 , 25 ], swi ching o mo e ac i e modes o a el and walking long dis ances we e nega i ely ela ed o he likelihood o de eloping new dep essi e symp oms. When examining he e ec o walking on inciden dep essi e symp oms in olde men, Smi h e al. [ 20 ] obse ed ha hose who walked mo e had lowe a es o p e alen dep essi e symp oms and ha hose wi h ch onic diseases who walked longe dis ances pe day we e less likely o de elop eigh -yea inciden dep essi e symp oms. Fu he mo e, he au ho s sugges ed ha he e seems o be a h eshold e ec in he p o ec i e e ec o walking on he de elopmen o inciden dep essi e symp oms, as ew di e ences we e ound be ween he in e media e and high walking g oups. This inding is o impo ance, especially o olde adul s who ha e mo e di icul ies in engaging in high olumes o physical ac i i y. Mo e ecen ly, using da a om he UK Biobank, a s udy in es iga ed whe he changes in commu e mode we e associa ed wi h di e ences in he se e i y o dep essi e symp oms [ 25 ]. I was ound ha adul s who al e ed om inac i e o ac i e commu ing o wo k epo ed less se e e symp oms a ollow-up han hose who emained inac i e. This sugges s ha commu ing beha iou s change ma e s, and hus, ac i e commu ing may be used as an in e en ion s a egy o p e en he de elopmen o dep essi e symp oms. P e ious in es iga ions in physical ac i i y ha e shown ha , o heal h pu poses, p esen beha iou s may be mo e impo an han pas beha iou s [ 27 ]. Ac i e a el o e s a sui able way o in eg a ing PA in o daily li e, and hose who use ac i e modes o commu ing ha e highe le els o PA [ 12 , 13 ]. As a esul , people who o en use mo e ac i e ways o a el migh ha e be e heal h s a us [ 15 – 17 ]. Hypo he ically, i means ha ac i e commu ing can also ha e a p o ec i e e ec on dep ession symp omology o educe dep ession symp oms [9,10]. Howe e , di e en ypes o PA ha e di e en e ec s on men al heal h s a us. Among adul s, he e is e idence ha leisu e- ime PA is mo e ela ed o men al heal h p oblems, and speci ically o dep ession, han o he o ms o PA [ 28 – 30 ]. In i e ou o se en a icles, a signi ican ela ionship be ween ac i e commu ing and dep ession symp oms was no ound [ 21 – 24 , 26 ]. C oss-sec ional e idence om wo s udies showed ha he associa ion be ween ac i e commu ing and dep ession is no consis en . In one s udy, pe o med among La in Ame ican adul s, i was ound ha non-mo o ised modes o a el we e no associa ed wi h dep ession [ 26 ]. On he o he hand, a s udy among A ican-Ame ican adul s epo ed ha , while women who epo ed walking o en p esen ed a lowe p obabili y o dep essi e symp oms han women who epo ed ne e walking, he same pa e n was no obse ed in men [ 23 ]. In o de o p e en an inc ease in dep essi e symp oms, physical ac i i y may need o imp o e i ness [ 31 ]. Howe e , walking in many cases is p obably no pe o med a a high enough in ensi y o imp o e i ness. This di e ence be ween sexes may be due o a la ge pe cen age o A ican-Ame ican women being obese and, hus, ha ing lowe exe cise capaci y [ 23 ]. This can lead o walking being a mo e in ense physical ac i i y among A ican-Ame ican women compa ed o A ican-Ame ican men. The o he h ee s udies had a longi udinal design and, in gene al, concluded ha walking was no associa ed wi h dep essi e symp oms a ollow-up [ 21 , 22 , 24 ]. In a one-yea ollow-up longi udinal su ey pe o med among Japanese wo ke s, i was ound ha dep essi e symp oms we e no associa ed wi h e iles o walking o wo k du a ions [ 21 ]. Simila ly, in a coho s udy wi h a mean ollow-up o 4.7 yea s also among Japanese wo ke s, walking o and om wo k was no ela ed o he isk o dep essi e symp oms [ 24 ]. Las ly, a s udy among Canadian olde adul s e i ied ha , while dep essi e symp oms p edic ed walking equency a subsequen ime poin s, walking equency did no p edic dep essi e symp oms [ 22 ]. The lack o a signi ican ela ionship be ween ac i e commu ing and dep ession symp oms in hese a icles a e in ag eemen In . J. En i on. Res. Public Heal h 2020,17, 1041 8 o 11 wi h p e ious s udies among men and women, which showed ha walking dis ance was no ela ed o dep ession symp oms [ 32 – 34 ]. Fo hose who use an ac i e mode o anspo a ion, ansi connec i i y ma e s mo e o men al heal h han ac i e a el [ 26 , 34 ]. Fu he mo e, he e is a di e ence be ween walking and cycling as a mode o ac i e anspo a ion. Mos s udies ocus on walking as a way o ac i e anspo a ion. Those who usually walk as an ac i e mode o anspo a ion unde ake a a lowe in ensi y han cycling [ 34 , 35 ]. This is o pa icula impo ance, because PA in ensi y is a de e minan o heal h. Thus, pe haps in some s udies, he le el o PA in ensi y while walking was oo low o cause e ec s on men al heal h and dep ession symp oms. Inconsis en indings iden i ied in his sys ema ic e iew can be a e lec ion o he di e en me hodologies used and he di e en geog aphical a eas and social con ex s examined. Measu es o ac i e commu ing widely a ied om s udy o s udy, al hough all we e sel - epo ed. Some s udies ocused on olume [ 20 , 21 , 24 ], o he s on equency [ 22 , 23 ]; one s udy ocused only on he mode o commu ing [ 26 ] and ano he one examined he ansi ion om non-ac i e commu ing o ac i e commu ing [ 25 ]. The ype o ac i e commu ing also di e ed among s udies. Mos in es iga ed only walking [ 20 – 24 ], while wo s udies in es iga ed ac i e commu ing wi hou speci ying he ype [ 25 , 26 ]. Addi ionally, he con ex o he commu ing was di e en among s udies. Th ee s udies e e ed o commu ing o and om wo k [ 21 , 24 , 25 ]. The o he s did no speci y he con ex o commu ing. Due o his, compa abili y among s udies is e y limi ed. Rega ding physical ac i i y and ac i e commu ing olume and equency a e impo an in o ma ion. Resul s may di e based on hese a iables. Al hough some s udies examined olume o equency, hey used di e en ime ames. S udies a ely conside bo h olume and equency oge he . Ano he impo an aspec o compa abili y o he esul s is he ype o ac i e commu ing, mainly because di e en ypes o commu ing (e.g., walking e sus cycling) ha e di e en in ensi ies and olumes associa ed o hem. Fu u e s udies should include equency and olume and be as speci ic as possible when analysing ac i e commu ing. P e e ably, simila me hods should be used among s udies o imp o e compa abili y. In addi ion o he s udies being unde aken in di e en geog aphic a eas and using di e en ways o measu e ac i e anspo , i is impo an o men ion ha di e en scales we e used o assess symp oms o dep ession. The use o di e en scales be ween he a icles may also ha e con ibu ed o he inconsis ency o he esul s ha we e obse ed. The social and buil en i onmen is an impo an ac o o be conside ed in men al heal h. The neighbou hood socioeconomic le el and buil -en i onmen cha ac e is ics a e associa ed wi h a highe p obabili y o sc eening posi i ely o dep ession [ 36 ]. Walking o cycling in some buil -en i onmen s can con ibu e o he de elopmen o dep ession symp oms [ 26 , 36 ]. Fu he mo e, o some people, ac i e anspo a ion is he only mode o anspo a ion, because hey do no ha e a ca o access o mass ansi op ions. In hese cases, an ac i e mode o anspo a ion is no an op ion, i is a manda o y condi ion. As a esul , anspo exclusion is associa ed wi h social exclusion, which migh inc emen po en ial men al heal h p oblems such as dep ession [37]. These s udy e iew indings should be conside ed in ligh o some limi a ions. Fi s ly, in all sc eened a icles, he in o ma ion on commu ing and dep ession was sel - epo ed, which may be subjec o bias. Mo eo e , sel - epo ed PA may ha e led o o e es ima es o o al PA. Secondly, se e al a icles do no make a clea dis inc ion be ween ac i e anspo as a o m o displacemen and ac i e anspo as a o m o leisu e (e.g., walking o leisu e). This lack o dis inc ion p ecludes a mo e igo ous analysis o he ela ionship be ween ac i e anspo and dep ession symp oms. Thi dly, some s udies we e ocused on speci ic popula ions, such as Japanese-Ame ican olde men in Hawaii, A ican-Ame ican adul s, o wo ke s. This may comp omise he gene alisabili y o he esul s o he gene al popula ion. Finally, wo s udies we e c oss-sec ional, which makes i impossible o in e causali y. Finally, sea ch e ms we e selec ed o iden i y a icles ha associa e ac i e anspo a ion and dep ession symp oms. Ne e heless, se e al a icles we e subsequen ly excluded, because nei he he i le no abs ac con ained e ms ha could be associa ed wi h ac i e a el o ac i e commu ing. In . J. En i on. Res. Public Heal h 2020,17, 1041 9 o 11 5. Conclusions This sys ema ic e iew shows ha he ela ionship be ween ac i e commu ing and dep ession symp oms in adul s and olde adul s is inconsis en . O e all, ou o se en s udies, only wo ound bene icial e ec s o ac i e commu ing on dep ession symp oms. Ac i e commu ing imp o es daily PA le els, which is known o be ela ed o dep ession symp oms. Thus, hypo he ically, ac i e commu ing could be ela ed o dep ession. Howe e , i e s udies did no ind a signi ican ela ionship be ween ac i e commu ing and dep ession symp oms. Mo e s udies on his opic a e necessa y in o de o unde s and i ac i e commu ing can be used as a public heal h s a egy o ackle men al heal h p oblems such as dep ession. Fu u e s udies should include equency and olume and be as speci ic as possible when analysing ac i e commu ing. P e e ably, simila me hods should be used among s udies o imp o e compa abili y. Fu he mo e, u u e s udies should be mo e speci ic ega ding he con ex o commu ing. Au ho Con ibu ions: Concep ualiza ion, A.M., M.P. and D.H.-N.; me hodology, A.M., M.P. and D.H.-N.; so wa e, M.P. and D.H.-N.; alida ion, A.M. and D.G.-B.; o mal analysis, A.M. and D.F.; in es iga ion, A.M. and M.P.; da a cu a ion, A.M., M.P. and D.H.-N.; w i ing—o iginal d a p epa a ion, A.M. and D.F.; w i ing— e iew and edi ing, D.F., É .R.G. and D.G.-B.; and supe ision, É .R.G. and D.G.-B. All au ho s ha e ead and ag eed o he published e sion o he manusc ip . Funding: This esea ch ecei ed no ex e nal unding. Con lic s o In e es : The au ho s decla e no con lic o in e es . Re e ences 1. WHO. Dep ession and O he Common Men al Diso de s. Global Heal h Es ima es; Wo ld Heal h O ganiza ion: Gene a, Swi ze land, 2017. 2. Chapman, D.P.; Pe y, G.S. Dep ession as a majo componen o public heal h o olde adul s. P e . Ch onic Dis. 2008,5, A22. [PubMed] 3. Vancamp o , D.; Co ell, C.U.; Galling, B.; P obs , M.; De He , M.; Wa d, P.B.; Rosenbaum, S.; Gaugh an, F.; Lally, J.; S ubbs, B. Diabe es melli us in people wi h schizoph enia, bipola diso de and majo dep essi e diso de : A sys ema ic e iew and la ge scale me a-analysis. Wo ld Psychia y 2016 ,15, 166–174. [C ossRe ] [PubMed] 4. Fe a i, A.J.; Cha lson, F.J.; No man, R.E.; Pa en, S.B.; F eedman, G.; Mu ay, C.J.; Vos, T.; Whi e o d, H.A. Bu den o dep essi e diso de s by coun y, sex, age, and yea : Findings om he global bu den o disease s udy 2010. PLoS Med. 2013,10, e1001547. [C ossRe ] [PubMed] 5. Walke , E.R.; McGee, R.E.; D uss, B.G. Mo ali y in men al diso de s and global disease bu den implica ions: A sys ema ic e iew and me a-analysis. JAMA Psychia y 2015,72, 334–341. [C ossRe ] [PubMed] 6. Ka yo aki, E.; Smi , Y.; Hold Henningsen, K.; Huibe s, M.J.; Robays, J.; de Beu s, D.; Cuijpe s, P. Combining pha maco he apy and psycho he apy o mono he apy o majo dep ession? A me a-analysis on he long- e m e ec s. J. A ec . Diso d. 2016,194, 144–152. [C ossRe ] [PubMed] 7. Khan, A.; Fauce , J.; Lich enbe g, P.; Ki sch, I.; B own, W.A. A sys ema ic e iew o compa a i e e icacy o ea men s and con ols o dep ession. PLoS ONE 2012,7, e41778. [C ossRe ] 8. Ol son, M.; Amos, T.B.; Benson, C.; McRae, J.; Ma cus, S.C. P ospec i e se ice use and heal h ca e cos s o Medicaid bene icia ies wi h ea men - esis an dep ession. J. Manag. Ca e Spec. Pha m. 2018 ,24, 226–236. [C ossRe ] 9. Schuch, F.B.; Vancamp o , D.; Fi h, J.; Rosenbaum, S.; Wa d, P.B.; Sil a, E.S.; Hallg en, M.; Ponce De Leon, A.; Dunn, A.L.; Deslandes, A.C.; e al. Physical ac i i y and inciden dep ession: A me a-analysis o p ospec i e coho s udies. Am. J. Psychia y 2018,175, 631–648. [C ossRe ] 10. Reba , A.L.; S an on, R.; Gea d, D.; Sho , C.; Duncan, M.J.; Vandelano e, C. A me a-me a-analysis o he e ec o physical ac i i y on dep ession and anxie y in non-clinical adul popula ions. Heal h Psychol. Re . 2015,9, 366–378. [C ossRe ] 11. Tasci, G.; Bayka a, S.; Gu ok, M.G.; A maca, M. E ec o exe cise on he apeu ic esponse in dep ession ea men . Psychia y Clin. Psychopha macol. 2019,29, 137–143. [C ossRe ]