Effects of childhood and adolescence physical activity patterns on psychosis risk-a general population cohort study

ARTICLE OPEN E ec s o childhood and adolescence physical ac i i y pa e ns on psychosis isk—a gene al popula ion coho s udy Elina So munen 1 , Maiju M. Saa inen 2 , Raimo K. R. Salokangas 1 , Ris o Telama 3 , Nina Hu i-Kähönen 4 , Tuija Tammelin 3 , Jo ma Viika i 5 , Olli Rai aka i 2,6 and Ja mo Hie ala 1,7,8 Schizoph enia spec um diso de s a e associa ed wi h high mo bidi y and mo ali y in soma ic diseases. The isk ac o s o his excess mo ali y include, e.g., obesi y, die a y ac o s, and physical inac i i y, especially a e he onse o psychosis, bu he e a e limi ed ea ly de elopmen al da a on hese ac o s in indi iduals who la e de elop psychosis. A popula ion-based coho s udy “Ca dio ascula Risk o Young Finns”s a ed in 1980 wi h 3596 child en and adolescen s om six di e en age g oups (3, 6, 9, 12, 15, and 18 yea s). Ca dio ascula heal h pa ame e s, including ques ionnai e o physical ac i i y be o e fi s hospi aliza ion (≤18 yea s), we e s udied in 1980, 1983, and 1986. All psychia ic diagnoses o he pa icipan s we e de i ed om he Finnish Hospi al Discha ge Regis e up o he yea 2012. We iden ified diagnos ic g oups o non-a ec i e psychosis (n= 68, including a schizoph enia subg oup, n= 41), pe sonali y diso de s (n= 43), a ec i e diso de s (n= 111), and subs ance- ela ed diso de s (n= 49), based on Diagnos ic and S a is ical Manual o Men al Diso de s, Fou h Edi ion (DSM-IV). G oups we e compa ed wi h con ols wi h no psychia ic diagnoses (n= 3325). Sex, age, body mass index, bi h weigh , non-p e e m bi h, and mo he ’s men al diso de s we e included in he s a is ical model. Low physical ac i i y in childhood and adolescence (9–18 yea s) independen ly p edic ed la e de elopmen o non-a ec i e psychosis. Lowe physical ac i i y index ( ela i e isk 1.26 [1.1–1.5]), lowe le el o common ac i i y du ing leisu e ime ( ela i e isk 1.71 [1.2–2.5]), and non-pa icipa ion in spo s compe i ions ( ela i e isk 2.58 [1.3–5.3]) we e associa ed wi h a highe isk o la e non-a ec i e psychosis (exp essed as inc ease in ela i e isk pe physical ac i i y uni ). The findings we e e en s onge o schizoph enia, bu no such link was obse ed o o he diagnoses. The cause o low physical ac i i y in p emo bid/p od omal phase is likely o be mul i ac o ial, including de ian mo o and cogni i e de elopmen . The esul s p o ide a a ionale o including exe cise and physical ac i i y in e en ions as a pa o psychosis p e en ion p og ams. npj Schizoph enia (2017) 3:5 ; doi:10.1038/s41537-016-0007-z INTRODUCTION Schizoph enia spec um diso de s a e consis en ly associa ed wi h high excess in mo ali y compa ed wi h he gene al popula ion. 1–3 Highe amoun o physical illnesses, e.g., ca dio ascula and pulmona y diseases, and me abolic diseases, 4 play a majo ole in he excessi e mo bidi y and mo ali y in schizoph enia. 2 Soma ic heal h p oblems in his popula ion a e pa ly due o he illness i sel as well as an ipsycho ic medica ion, bu unheal hy li es yle also plays a conside able ole. 5,6 Physical inac i i y is an independen isk ac o o ca dio ascula disease mo ali y. 7,8 An o e all dec ease in ca dio ascula mo ali y a es in he gene al popula ion has been obse ed, bu people wi h se ious men al illnesses a e no ully keeping up wi h he same de elopmen , 9 al hough in he No dic coun ies his gap has ac ually dimin- ished. 10 I seems ha he imp o ed heal h o he gene al popula ion does no ully benefi people wi h se ious men al illnesses. 1 People wi h psycho ic illnesses, on a e age, ha e an unheal hy li es yle, including low le els o physical ac i i y 11,12 and poo ca dio espi a o y fi ness. 13 Low le els o physical ac i i y a e p ominen ly obse ed in la e s ages o schizoph enia bu ha e also been epo ed in h ee s udies conce ning adolescen s who la e de elop psychosis. 14–16 The benefi s o physical ac i i y and exe cise in e en ions in ea ing dep ession a e ela i ely well es ablished. 17 Recen ly conduc ed me a-analyses show imp o e- men in ca dio espi a o y fi ness, 18,19 clinical symp oms, quali y o li e, global unc ioning and dep ession 20 wi h physical ac i i y p omo ion and exe cise in pa ien s wi h schizoph enia. Howe e , i seems ha pa ien s wi h psychosis migh be pa ly esis an o imp o emen in heal h, such as body mass index (BMI) o weigh educ ion due o exe cise. 19,21,22 Wide s uc u al b ain abno mali ies in schizoph enia ha e been well documen ed in ea lie s udies, 23–25 and a leas some o hese changes ha e a neu ode elopmen al o igin. Recen epo s sugges ha exe cise in e en ions in schizoph enia can induce s uc u al and unc ional b ain changes, such as he mo phology o hippocampus, 26 inc eased ce eb al g ay ma e , 27 and e en imp o emen in b ain connec i i y. 28 These findings lead us o hypo hesize ha physical ac i i y and exe cise will modi y b ain de elopmen and ma u a ion also in childhood and adolescence. Ea lie esea ch sugges s ha pa ien s wi h psychosis o schizoph enia, on a e age, ha e a his o y o delayed mo o Recei ed: 7 Sep embe 2016 Re ised: 18 No embe 2016 Accep ed: 2 Decembe 2016 1 Depa men o Psychia y, Uni e si y o Tu ku, Tu ku, Finland; 2 Resea ch Cen e o Applied and P e en i e Ca dio ascula Medicine, Uni e si y o Tu ku, Tu ku, Finland; 3 LIKES— Resea ch Cen e o Spo and Heal h Sciences, Jy äskylä, Finland; 4 Depa men o Pedia ics, Uni e si y o Tampe e and Tampe e Uni e si y Hospi al, Tampe e, Finland; 5 Depa men o Medicine, Uni e si y o Tu ku and Di ision o Medicine, Tu ku Uni e si y Hospi al, Tu ku, Finland; 6 Depa men o Clinical Physiology and Nuclea Medicine, Tu ku Uni e si y Hospi al, Tu ku, Finland; 7 Gene al Hospi al Psychia y Uni , Tu ku Uni e si y Hospi al, Tu ku, Finland and 8 Tu ku Psychia y, Tu ku, Finland Co espondence: Ja mo Hie ala (jahi@u u.fi) www.na u e.com/npjschz Published in pa ne ship wi h he Schizoph enia In e na ional Resea ch Socie y de elopmen in childhood and adolescence, 29–31 as well as delayed neu ological and cogni i e de elopmen . 32,33 Non- pa icipa ion in physical ac i i ies among child en may pa ly be due o abe an de elopmen in mo o skills, seen in some pa ien s, o o he sub le p emo bid cogni i e o a ec i e symp oms. 34 Ye , no sys ema ic esea ch has p e iously been done conce ning physical ac i i y and exe cise pa e ns in child en and adolescen s who la e de elop psychosis. This s udy aimed o examine whe he physical ac i i y le els in childhood and adolescence independen ly p edic la e de elop- men o non-a ec i e psychosis. An ongoing, popula ion-based coho s udy enabled us o (1) ha e epe i i e measu es o physical ac i i y be o e and a e pube y, and (2) o link hese da a wi h hospi al discha ge egis e in o ma ion o non-a ec i e psychoses and o he psychia ic diagnoses ha led o one o mo e hospi al ea men pe iods. Changes in physical ac i i y o indi iduals who la e de elop psychosis migh al eady be seen in childhood, yea s be o e he onse o psychosis o p od omal symp oms. I so, he findings would sugges a a ionale o including sui able o ms o exe cise in ea men p og ams o ea ly psychosis. RESULTS Physical ac i i y le el was lowe among child en and adolescen s who la e de eloped non-a ec i e psychosis (Fig. 1a, b; Tables 1 and 2). One uni lowe physical ac i i y index (PAI) a he age o 9–18 yea s, adjus ed wi h co a ia es, was associa ed wi h a 26% highe isk o any non-a ec i e psychosis (Table 2). One uni lowe common ac i i y du ing leisu e ime was associa ed wi h a 71% highe isk, and non-pa icipa ion in spo s compe i ion wi h a 158% highe isk. In he g oup o pa ien s wi h schizoph enia, he esul s we e e en s onge . One uni lowe PAI in childhood and adolescence was associa ed wi h a 43% inc ease in he isk o la e de elopmen o schizoph enia. One uni lowe common ac i i y du ing leisu e ime was associa ed wi h a 76% highe isk, one uni lowe in ensi y o physical ac i i y wi h a 71% highe isk, and one uni lowe equency o pa icipa ion in o ganized aining wi h a 40% and non-pa icipa ion in spo s compe i ions wi h a 388% highe isk o la e de elopmen o schizoph enia. Physical ac i i y o ei he pa en was no associa ed wi h he isk o u u e psychosis o schizoph enia. PAI in childhood and adolescence was no associa ed wi h o he men al diso de s a adul age (p> 0.05 in all analyses, Supplemen a y Table 2). Mo he ’s men al diso de s we e associa ed wi h a 296% highe isk o la e de elopmen o psychosis and a 354% highe isk o la e schizoph enia in hei o sp ing (Table 2). Men al diso de o ei he o bo h pa en s was associa ed wi h a 269% highe isk o psychosis and a 365% highe isk o schizoph enia (Table 2). Childhood and adolescence BMI as a con inuous a iable, bi h weigh , o non-p e e m bi h had no significan associa ion wi h he isk o non-a ec i e psychosis la e in li e in iew o hese da a (p> 0.05). In he sensi i i y analyses, he associa ions o PAI and la e psychosis o schizoph enia emained simila , despi e he classifi- ca ion o BMI o unde weigh o o e weigh . As expec ed, unde weigh in childhood and/o adolescence inc eased he isk o psychosis o wo- old ( ela i e isk (RR) [95% CI] 2.1 [1.1, 4.0]; p= 0.026), bu he independen e ec o PAI emained unchanged (1.2 [1.1, 1.4]; p= 0.010). Ea ly unde weigh also seemed o inc ease he isk o schizoph enia bu his did no each s a is ical significance (p= 0.064). O e weigh was no associa ed wi h he isk o la e psychosis (p= 0.237). DISCUSSION The main finding o his s udy is ha low physical ac i i y le el in child en and adolescen s is an independen p edic o o de el- opmen o non-a ec i e psychosis. This pa e n o low physical ac i i y was e iden h oughou he 9–18 yea age pe iod wi h no majo di e ences be o e and a e pube y, which is conside ed o be one o he c i ical ime pe iods in he de elopmen o non- a ec i e psychoses. Low le el o common ac i i y du ing leisu e ime and non-pa icipa ion in spo s compe i ions we e isk ac o s o non-a ec i e psychoses. Also, low in ensi y o leisu e- ime ac i i y and low equency o pa icipa ion in o ganized aining we e isk ac o s o schizoph enia in pa icula . The obse a ion pe iod was confined o yea s 1980–1986 because he da a collec ion was comple e du ing his pe iod and only pa ial in he la e ollow-up poin s. Also, he ole o socie al e ec s needs o be conside ed in he in e p e a ion and ele ance o he esul s. The ole o social media as well as compu e / ideo games in leisu e- ime ac i i y in hese age g oups a e expec ed o be ela i ely small in 1980s compa ed wi h hose in he 2010s. Some epo s on he ime ends in 12–18-yea s-old you h’s physical ac i i y in Finland be ween 1979 and 2005 sugges ha he e is no majo change in o e all physical ac i i y, bu pa icipa ion in mode a e o igo ous physical ac i i y has sligh ly inc eased. 35 I is well known ha pa en s’men al diso de s a e associa ed wi h a isk o psychosis. This was also he case in his s udy. In ou s udy, BMI was sligh ly lowe in indi iduals who would la e ha e a diagnosis o non-a ec i e psychosis. Unde weigh in childhood 36 9121518 6 7 8 9 10 11 9 121518 6 7 8 9 10 11 a b Fig. 1 Mean (95% CI) physical ac i i y index ( ange 5–14) in child en and adolescen s a 9–18 yea s o age. G ay line =indi iduals who la e de eloped any non-a ec i e psychosis (a) schizoph enia (b) and black line =con ols wi h no psychia ic diagnoses du ing he ollow-up E ec s o childhood and adolescence physical ac i i y pa e ns E So munen e al 2 npj Schizoph enia (2017) 5 Published in pa ne ship wi h he Schizoph enia In e na ional Resea ch Socie y and adolescence 37 is a known isk ac o o schizoph enia. In line wi h he p e ious s udies, ea ly unde weigh was associa ed wi h he isk o non-a ec i e psychosis in his sample. Howe e , he associa ion be ween low physical ac i i y and he isk o non- a ec i e psychosis emained unchanged when con inuous a i- able BMI was subs i u ed wi h a ca ego ized “BMI”, i.e., unde - weigh o o e weigh , as a co a ia e in he sensi i i y analysis. These esul s suppo he iew ha low p emo bid o p od omal physical ac i i y is linked o isk o non-a ec i e psychosis i espec i e o BMI. The causes o lowe physical ac i i y le els in he p emo bid o p od omal pe iod o non-a ec i e psychoses a e no ully unde s ood, bu a e likely o be mul i ac o ial. Ea lie s udies ha e shown ha a leas some pa ien s, who will la e de elop schizoph enia, ha e de ian mo o de elopmen in childhood. 38–40 In 1990, Walke e al. epo ed a case se ies sugges ing ha indi iduals, who la e wen on o de elop schizoph enia, could be di e en ia ed om hei heal hy siblings be o e he age o eigh based on obse ed beha io in home ideos. Child en, who would la e de elop schizoph enia, we e ound o be less esponsi e, ha e less eye con ac and posi i e a ec , as well as wo se fine and g oss mo o coo dina ion. 29 Schizoph enia is known o associa e wi h delayed mo o de elopmen , e.g., delayed walking, by he age o wo. 30,32 This has also been seen in di ec longi udinal Table 1. Physical ac i i y in childhood and adolescence (9–18 yea s, du ing yea s 1980–1986) in he g oups wi h la e de elopmen o schizoph enia o any non-a ec i e psychosis, and con ols wi h no psychia ic diagnoses du ing ollow-up yea s 1980–2012 Physical ac i i y in childhood and adolescence S udy popula ion Physical ac i i y ou comes in di e en g oups To al Pa ien s wi h non-a ec i e psychosis Pa ien s wi h schizoph enia Con ols Pa ien s wi h non-a ec i e psychosis Pa ien s wi h schizoph enia Con ols NN(%) N(%) N(%) Mean (SD) Mean (SD) Mean (SD) PAI ( ange 5–14) a 9 yea s 1501 25 (1.7) 16 (1.1) 1476 (98.3) 8.8 (1.6) 8.3 (1.3) 9.4 (1.6) 12 yea s 1559 30 (1.9) 21 (1.3) 1529 (98.0) 9.2 (1.9) 8.6 (1.7) 9.5 (1.8) 15 yea s 1493 27 (1.8) 17 (1.1) 1466 (98.2) 8.5 (1.6) 8.2 (1.3) 8.8 (1.9) 18 yea s 1291 21 (1.6) 15 (1.2) 1270 (98.4) 7.5 (1.2) 7.7 (1.1) 8.4 (2.0) Common ac i i y du ing leisu e ime ( ange 1–3) b 9 yea s 1538 27 (1.8) 16 (1.0) 1511 (98.2) 2.5 (0.6) 2.4 (0.7) 2.6 (0.6) 12 yea s 1599 31 (1.9) 21 (1.3) 1568 (98.1) 2.2 (0.8) 2.1 (0.8) 2.5 (0.7) 15 yea s 1533 29 (1.9) 18 (1.2) 1504 (98.1) 2.0 (0.6) 2.1 (0.6) 2.2 (0.6) 18 yea s 1324 23 (1.7) 16 (1.2) 1301 (98.3) 1.7 (0.7) 1.7 (0.7) 2.0 (0.7) F equency o leisu e- ime physical ac i i y ( ange 1–7) c 9 yea s 1089 24 (2.2) 11 (1.0) 1065 (97.8) 5.6 (1.4) 5.4 (1.6) 5.9 (1.2) 12 yea s 1146 27 (2.3) 9 (0.8) 1119 (97.6) 5.9 (1.3) 5.7 (1.8) 5.8 (1.3) 15 yea s 1672 35 (2.1) 13 (0.8) 1637 (97.9) 5.3 (1.3) 5.2 (1.6) 5.5 (1.5) 18 yea s 1541 29 (1.9) 9 (0.6) 1512 (98.1) 4.9 (2.0) 4.2 (2.3) 5.3 (1.6) In ensi y o physical ac i i y ( ange 1–3) d 9 yea s 1541 28 (1.8) 17 (1.1) 1513 (98.2) 2.0 (0.5) 1.8 (0.4) 1.9 (0.4) 12 yea s 1595 31 (1.9) 21 (1.3) 1564 (98.1) 2.0 (0.6) 2.0 (0.5) 2.0 (0.5) 15 yea s 1528 29 (1.9) 18 (1.2) 1499 (98.1) 2.1 (0.3) 2.1 (0.2) 2.1 (0.5) 18 yea s 1328 22 (1.7) 16 (1.2) 1306 (98.3) 2.0 (0.5) 2.1 (0.6) 2.2 (0.6) F equency o pa icipa ion in o ganized aining ( ange 1–3) e 9 yea s 535 5 (0.9) 2 (0.4) 530 (99.1) 1.6 (0.5) 2.0 (0.0) 2.5 (1.9) 12 yea s 987 15 (1.5) 7 (0.7) 972 (98.5) 2.7 (2.2) 1.1 (0.4) 2.9 (2.0) 15 yea s 1068 26 (2.4) 15 (1.4) 1042 (97.6) 2.6 (2.1) 2.0 (1.8) 2.8 (2.1) 18 yea s 949 12 (1.3) 9 (0.9) 937 (98.7) 1.8 (1.7) 2.0 (2.0) 2.4 (2.0) Pa icipa ion in spo s compe i ions N(%) N(%) N(%) 9 yea s 1137 17 (1.5) 8 (0.7) 1120 (98.5) 3 (17.6) 2 (25.0) 396 (35.4) 12 yea s 1187 26 (2.2) 11 (0.9) 1161 (97.8) 8 (30.8) 5 (45.5) 515 (44.4) 15 yea s 1112 21 (1.9) 7 (0.6) 1091 (98.1) 2 (9.5) 1 (14.3) 285 (26.1) 18 yea s 930 12 (1.3) 2 (0.2) 918 (98.7) 0 (0.0) 0 (0.0) 149 (16.2) DSM-IV diagnosis 295; DSM-IV diagnoses 295, 297, and 298 a PAI a ings anging om 5 o 14 b Common ac i i y du ing leisu e ime was asked by “Wha do you usually do in you leisu e ime?”:1=I am usually indoo s and ead o do some hing like ha , 2 =I spend my ime indoo s and ou doo s, ou doo s I usually walk o spend ime wi h my iends, 3 =I am usually ou doo s and exe cise a he much c F equency o leisu e- ime physical ac i i y was asked by “How o en do you engage in leisu e- ime physical ac i i y a leas hal an hou pe ime?”. The esponse al e na i es we e: 1 =no a all, 2 =less han once a mon h, 3 =once a mon h, 4 =2–3 imes a mon h, 5 =once a week, 6 =2–6 imes a week, 7 = e e y day d In ensi y o physical ac i i y was asked by “How much a e you b ea h- aking and swea ing when you engage in physical ac i i y and spo ?”:1=no a all, 2=mode a ely, 3 =a lo o e F equency o pa icipa ion in o ganized aining was asked by “Do you pa icipa e in o ganized physical ac i i y?”:1=no a all, occasionally o less han once a mon h, 2 = egula ly, once a mon h o mo e, o once a week, 3 =many hou s and imes a week Pa icipa ion in spo s compe i ions: 1 =no, 2 =yes E ec s o childhood and adolescence physical ac i i y pa e ns E So munen e al 3 Published in pa ne ship wi h he Schizoph enia In e na ional Resea ch Socie y npj Schizoph enia (2017) 5 measu emen s o mo o pe o mance in childhood, and mo o defici s seem o be specific o non-a ec i e psychoses. 31 Ano he coho s udy epo ed ha pa ien s wi h la e de elopmen o schizoph enia eached all de elopmen al miles ones, pa icula ly smiling, li ing head, si ing, c awling and walking, la e han heal hy con ols o indi iduals who la e de eloped a psychia ic diso de o he han schizoph enia. 33 Non-pa icipa ion in physical ac i i ies among child en could pa ly be due o limi ed de elopmen o mo o skills and a oidance o exe cise. 34 I is, howe e , likely ha a complex combina ion o de ian mo o de elopmen , mo i a ional and ewa d defici s, sub le a ec i e p oblems, and di ficul ies in social in e ac ion all con ibu e o lowe in e es in physical ac i i ies and in pa icula hose o ms equi ing social skills. 41 Physical ac i i y as a pa o he ea ly in e en ions o psychoses Rega dless o he causes o lowe physical ac i i y le els in he p emo bid/p od omal phases o non-a ec i e psychoses, ou esul s ha e ele ance o p e-emp i e psychia y, and p o ide a ionale o including exe cise in ea ly in e en ions o psychosis. Physical ac i i y and exe cise he apy in ea ing psychia ic diso de s ha e been s udied in ensi ely du ing ecen yea s. Fo example, physical ac i i y and exe cise in ea ing majo dep es- sion is al eady suppo ed by conside able e idence. A la ge Coch ane e iew shows ha exe cise has mode a ely g ea e e ec in educing symp oms o dep ession when compa ed wi h no ea men , placebo o ac i e con ol in e en ions, such as elaxa ion o medi a ion. 17 Fo example, he UK Na ional Ins i u e o Heal h and Clinical Excellence ecommends s uc u ed exe cise o he ea men o mild o mode a e dep ession. Howe e , esul s on exe cise in e en ion in ea ing schizo- ph enia ha e been inconsis en . Exe cise in e en ions can educe bo h posi i e and nega i e symp oms 19 and imp o e clinical symp oms, quali y o li e, global unc ioning, dep ession, 20 and e en cogni i e unc ioning 42 in adul schizoph enia pa ien s. One me a-analysis did no find imp o emen s in nega i e o posi i e symp oms o schizoph enia, o he indi idual’s quali y o li e. 21 Imp o emen in physical fi ness by exe cise in e en ion o p omo ion is shown by ew ecen me a-analyses. 18,19 Ne e - heless, exe cise in e en ions we e no e ec i e in educing BMI 19,21 o weigh 21,22 in people wi h schizoph enia. I seems ha exe cise in e en ion is less e ec i e o pa ien s wi h psychosis compa ed wi h pa ien s wi h dep ession. Howe e , he benefi s o physical ac i i y in ea ing psychosis has been shown and one me a-analysis al eady p esen s p ac ical s a egies o physical ac i i y p omo ion as a pa o ea men . 13 I is cu en ly no known whe he hese exe cise in e en ions could be mo e e ec i e in ea lie p emo bid o p od omal phases o he illness. Physical exe cise and b ain mo phology S uc u al b ain abno mali ies ha e been conclusi ely documen- ed in schizoph enia, such as la ge la e al and hi d en icles as well as widesp ead educ ion o g ay ma e in he neoco ex and limbic a eas. 23–25 The e a e ecen epo s on he e ec s o exe cise on b ain mo phology in schizoph enia. The esul s show ha exe cise he apy o ca dio espi a o y fi ness imp o emen is ela ed o hippocampal enla gemen , 26 inc eased ce eb al g ay ma e olume, and dec eased olume in la e al and hi d en icle in adul pa ien s wi h schizoph enia. 27 In addi ion, a ecen di usion enso imaging s udy by S a ko a e al. showed ha whi e ma e in eg i y, in pa icula hose ac s in ol ed in mo o unc ioning, was imp o ed by an exe cise in e en ion o 6 mon hs. 28 An in iguing, and also likely, possibili y is ha physical ac i i y and exe cise a ec he de elopmen and ma u a ion o he cen al ne ous sys em. We now know ha exe cise inc eases neu ogen- esis in he limbic a eas, especially in he den a e gy us o he hippocampus. 43,44 Neu ogenesis also akes place in he sub en- icula zone whe e he newly o med neu ons, a leas pa ly, mig a e o s ia um, 45,46 which is cen ally in ol ed in ne wo ks egula ing mo emen and cogni ion. Clea ly, mo e esea ch is Table 2. Childhood and adolescen cha ac e is ics measu ed a he age o 9–18 and hei associa ions wi h he isk o la e de elopmen o any non- a ec i e psychosis o schizoph enia in 1980–2012 Childhood and adolescen cha ac e is ics Risk o any non-a ec i e psychosis Risk o schizoph enia Uni a ia e Mul i a ia e a Uni a ia e Mul i a ia e a RR (95% CI) PRR (95% CI) PRR (95% CI) PRR (95% CI) P 1-uni lowe PAI ( ange 5–14) 1.17 (1.02–1.3) 0.021 1.26 (1.1–1.5) 0.005 1.36 (1.2–1.6) <0.001 1.43 (1.2–1.7) <0.001 1-uni lowe common ac i i y du ing leisu e ime ( ange 1–3) 1.54 (1.1–2.2) 0.014 1.71 (1.2–2.5) 0.008 1.71 (1.1–2.8) 0.029 1.76 (1.02–3.0) 0.042 1-uni lowe equency o leisu e- ime physical ac i i y ( ange 1–7) 1.13 (0.98–1.3) 0.083 1.12 (0.96–1.3) 0.150 1.07 (0.9–1.3) 0.426 1.14 (0.96–1.4) 0.145 1-uni lowe in ensi y o physical ac i i y ( ange 1–3) 1.14 (0.7–1.8) 0.593 1.13 (0.7–1.9) 0.631 1.49 (0.9–2.4) 0.100 1.71 (1.1–2.8) 0.030 1-uni lowe equency o pa icipa ion in o ganized aining ( ange 1–3) 1.15 (1.01–1.3) 0.039 1.15 (0.99–1.3) 0.074 1.38 (1.1–1.7) 0.003 1.40 (1.1–1.8) 0.005 Pa icipa ion in spo s compe i ions (no s. yes) 2.42 (1.3–4.6) 0.007 2.58 (1.3–5.3) 0.009 4.11 (1.4–12.0) 0.01 4.88 (1.4–17.0) 0.013 1-uni highe BMI 1.00 (0.9–1.1) 0.909 0.96 (0.8–1.1) 0.584 1.03 (0.9–1.2) 0.709 0.99 (0.8–1.2) 0.942 Mo he ’s men al diso de s (yes s. no) 6.88 (3.1–15.2) <0.001 3.96 (1.4–11.1) 0.009 7.65 (2.8–20.6) <0.001 4.54 (1.4–14.9) 0.012 Men al diso de s o ei he pa en (yes s. no) 4.63 (2.2–9.8) <0.001 3.69 (1.5–9.3) 0.005 5.70 (2.3–14.2) <0.001 4.65 (1.7–13.0) 0.004 RR isk a io, CI confidence in e al DSM-IV diagnoses 295, 297, and 298; DSM-IV diagnoses 295 a All mul i a ia e analyses include sex, age, BMI, PAI, bi h weigh , and non-p e e m bi h. Mo he ’s men al diso de s we e included in all analyses excep men al diso de s o ei he pa en E ec s o childhood and adolescence physical ac i i y pa e ns E So munen e al 4 npj Schizoph enia (2017) 5 Published in pa ne ship wi h he Schizoph enia In e na ional Resea ch Socie y needed on he mechanisms o how exe cise a ec s he b ain in di e en phases o human de elopmen . S eng h and limi a ions The subjec s in his andomly selec ed popula ion coho we e p ospec i ely ollowed up om ea ly childhood o young adul hood be o e e idence o any psycho ic diso de . The psychia ic diagnoses in his obse a ional s udy we e de i ed om he hospi al discha ge egis e in Finland. The diagnos ic alidi y o schizoph enia spec um in egis e -based s udies has been epo ed o be good, 47,48 whe eas he alidi y o o he diagnoses has no been well s udied. I is also clea ha pa ien s equi ing hospi al ea men in he o he diagnos ic g oups ep esen mo e se e e o ms o hese diso de s. In ou s udy, he numbe o pa ien s who will de elop psychosis, and especially schizoph enia, is ela i ely low. This is a limi a ion bu he longi udinal design s ill makes his sample aluable. Physical ac i i y was measu ed by a sel - epo ques ionnai e. We ha e p e iously alida ed he physical ac i i y ques ionnai e in an independen popula ion by s udying he link be ween ele an componen s o PAI (i.e., in ensi y, equency), wi h he olume o mo emen assessed wi h accele ome e s and he numbe o s eps measu ed wi h pedome e s. These s udies show significan posi i e co ela ions, 49 which is well in line wi h o he simila s udies. 50 In addi ion, we ha e collec ed s ep da a using alida ed pedome e s in 1934 indi iduals om he Young Finns s udy popula ion. Pa icipan s wo e an Om on Walking S yle One (Model HJ-152R-E) s ep coun e o a pe iod o 1 week. Simila ly, as in he small alida ion s udy done in an independen popula ion, significan co ela ions we e seen be ween he numbe o s eps and indi idual componen s o he PAI. 49 Telama e al. showed a significan co ela ion be ween PAI and he bicycle e gome e es , ca ied ou in a subsample o 102 subjec s. Al hough i was shown in adul subjec s, PAI in 1980, when subjec s we e 9–18 yea s old, also co ela ed significan ly wi h he fi ness es in 2001. 51 Ex ensi e and consis en ollow-up da a we e a ailable on exe cise le els bu also on se e al possible isk ac o s o schizoph enia, enabling co a ia e adjus ed s a is ical analyses. The e ec o exe cise le el on he isk o non-a ec i e psychosis was s ill highly significan a e adjus men o o he isk ac o s, bu i is s ill possible ha he link may be explained by a ac o no measu ed in his coho s udy. METHODS S udy sample The pa icipan s we e de i ed om a popula ion-based, epidemiologic ollow-up s udy Ca dio ascula Risk o Young Finns (YFS). 52 A o al o 4320 heal hy Finnish child en and adolescen s in age coho s o 3, 6, 9, 12, 15, and 18 we e in i ed and 3596 (83%) pa icipa ed in he fi s c oss-sec ional su ey in 1980. The pa icipan s we e andomly selec ed om he na ional egis e om fi e Finnish popula ion cen e s (Helsinki, Tu ku, Tampe e, Kuopio, and Oulu) and hei u al su oundings. A ull-scale ollow-up o he o iginal p o ocol was possible o 6 yea s, yielding ollow-up da a om he yea s 1980, 1983, and 1986 o child en and adolescen s aged 3–18 yea s, 6–21 yea s, and 9–24 yea s, espec i ely. The ea e , he e is a gap o 15 yea s in he 3-yea ollow-up schedule, excep o mino subs udies conce ning only selec ed pa icipan g oups. The nex ollow-up o all pa icipan s ook place in 2001, when e en he younges o pa icipan s we e 24 yea s old. In he p esen s udy, only measu emen s om he fi s h ee ollow-ups, om 1980 o 1986 up o he pa icipan s’age o 18 yea s, we e included. Thus, none o he subjec da a a e comple e om age 3 o 18, and o hose bo n p io o 1968, he e canno be da a om mo e han one o wo isi s due o he design o he s udy. As he numbe o pa icipan s is high and we ha e no eason o assume any ema kable di e ences be ween he bi h coho s, he ollow-up se ies om he h ee s udy isi s (1980–1986) including child en and adolescen s om six age poin s we e combined o he analyses. F om 50% o he 3596 pa icipan s, ull da a om all h ee s udy isi s (1980–1986) we e a ailable. O 25% o pa icipan s, da a we e a ailable om wo isi s, and o 25% om one isi . Psychia ic diagnoses o he pa icipan s (yea s 1980–2012) we e ob ained om he Finnish Na ional Hospi al Discha ge Regis e , which is main ained by he Na ional Ins i u e o Heal h and Wel a e in Finland. The egis e co e s all gene al and men al hospi als in Finland since 1969, wi hou gaps. ICD-diagnoses we e con e ed o DSM-IV diagnoses (Supplemen a y Table 1). Diagnos ic g oups we e o med, and subjec s who had se e al psychia ic diagnoses we e ca ego ized unde only one o he g oups in he ollowing o de o p io i y: schizoph enia (DSM-IV 295) and all non-a ec i e psychoses (DSM-IV 295, 297, 298), pe sonali y diso de s (DSM-IV 301), a ec i e diso de s (mood and anxie y diso de s, DSM-IV 296, 300, 311), and subs ance- ela ed diso de s (DSM-IV 291, 303, 292, 304, 305). Non-a ec i e psychosis was diagnosed in 68 o he 3596 pa icipan s, 40 (59%) men and 28 (41%) women, esul ing in a 1.9% p e alence o his class o psychoses in his popula ion. In he g oup o non-a ec i e psychosis, 41 (60%) o he subjec s we e diagnosed as ha ing schizoph enia, schizoph eni o m diso de o schizoa ec i e diso de (DSM- IV 295), 5 (7%) had delusional diso de (DSM-IV 297), and 22 (32%) had b ie psycho ic diso de o psycho ic diso de NOS (DSM-IV 298). The p e alence o schizoph enia was 1.1%. The younges subjec ha ing he fi s hospi al ea men o psycho ic diso de was 18, which was used as a cu -o age o analysis o he p emo bid/p od omal phase o psychosis. The mean (SD) age o ecei ing a hospi al- ela ed diagnosis o non- a ec i e psychosis was 28.4 (7.0) yea s, being 28.2 (6.6) yea s among men and 28.7 (7.8) yea s among women. The p e alences o o he psychia ic Table 3. The assessmen o physical ac i i y and c ea ion o he PAI in 1980–1986 Ques ion in he ques ionnai e Code o PAI How o en do you engage in leisu e- ime physical ac i i y a leas hal an hou pe ime? No a all 1 Less han once a mon h 1 Once a mon h 1 2–3 imes a mon h 1 Once a week 2 2–6 imes a week 2 E e y day 3 How much a e you b ea h- aking and swea ing when you engage in physical ac i i y and spo ? No a all 1 Mode a ely 2 A lo o 3 How many imes a week do you usually engage in he aining sessions o a spo s club? No a all 1 Occasionally 1 Less han once a mon h 1 Once a mon h o mo e 2 Once a week 2 Many hou s and imes a week 3 Do you pa icipa e in egional o spo clubs-le el compe i ions? No 1 Yes 2 Wha do you usually do in you leisu e ime? I am usually indoo s and ead o do some hing like ha 1 I spend my ime indoo s and ou doo s, ou doo s I usually walk o spend ime wi h my iends 2 I am usually ou doo s and exe cise a he much 3 PAI o al, ange 5–14 E ec s o childhood and adolescence physical ac i i y pa e ns E So munen e al 5 Published in pa ne ship wi h he Schizoph enia In e na ional Resea ch Socie y npj Schizoph enia (2017) 5 hospi al- ela ed diagnoses we e 1.2% (n= 43) o pe sonali y diso de s, 3.1% (n= 111) o a ec i e diso de s, and 1.4% (n= 49) o subs ance- ela ed diso de s. The pe missions o use egis e da a and link diagnos ic da a o YFS da a we e acqui ed om he espec i e o ganiza ions. The p o ocol was app o ed by he E hics Commi ee o he Hospi al Dis ic o Sou hwes Finland. Physical ac i i y in childhood and adolescence Physical ac i i y was assessed wi h a sel - epo ques ionnai e o subjec s aged 9, 12, 15, and 18. The ques ionnai e (Table 3) was adminis e ed indi idually du ing he s udy isi s ha included medical examina ions. The ques ions included he equency and in ensi y o leisu e- ime physical ac i i y, pa icipa ion in spo s club aining, pa icipa ion in compe i i e spo e en s, and common ac i i y du ing leisu e ime. The answe s we e coded om 1 o 3, wi h 1 ep esen ing inac i i y o e y low ac i i y, 2 mode a ely in ensi e o equen ac i i y, and 3 equen o igo ous ac i i y. Answe s o pa icipa ion in compe i i e spo e en s we e coded only om 1 o 2. The PAI was calcula ed as a sum o measu emen s in he a o emen ioned ques ions, wi h he a ings anging om 5 o 14. 53 Clinical cha ac e is ics o child en and adolescen s Heigh and weigh o he child en and adolescen s we e measu ed and BMI was calcula ed as kg/m 2 . BMI was u he dicho omized using he classifica ion p o ided by Cole e al. o unde weigh s. no unde weigh ep esen ing adul BMI ≤18.5 s. highe 54 and o e weigh s. no o e weigh ep esen ing adul BMI ≥25 s. lowe . 55 Bi h weigh was asked in 1983 and 1986 in a ques ionnai e o he pa icipan s’pa en s. Pa en al cha ac e is ics Pa icipan s’mo he s we e asked abou pa en s’men al diso de s o p oblems, diagnosed by a doc o , wi h a sel - epo ques ionnai e in 1980 and 1983. Pa en s’physical ac i i y was asked abou in he yea s 1986 and 1989 by asking whe he a pa en is engaged in egula physical ac i i y ( he answe s we e coded om 1 o 3, 1 = a ely o no a all, 2 = some imes o wi h o he hobbies, 3 = egula ly). I a pa en answe ed wi h op ion 3, he equency o physical ac i i y was also asked (answe s we e coded om 1 o 5, 1 = once a mon h o less, 2 = 2–3 imes a mon h, 3 = once a week, 4 = 2–6 imes a week, 5 = e e y day). S a is ical me hods The desc ip i e s a is ics a e gi en as n(%) and mean (SD). Associa ions o childhood and adolescen physical ac i i y wi h he isk o adul age psychosis a e gi en as isk a ios wi h 95% confidence in e als (RR [95% CI]) om uni a ia e and mul i a iable modified Poisson eg ession models. 56 Gene alized es ima ing equa ion es ima ion was used in analyses o epea ed measu es. 57 The mul i a iable models included sex, age, BMI in childhood and adolescence, mo he ’s men al diso de s, bi h weigh , and non-p e e m bi h as co a ia es. BMI and physical ac i i y da a we e used om all a ailable ime poin s and analyzed longi udinally. Mo he ’s men al diso de s we e excluded om he model, including men al diso de s o ei he pa en . The po en ial con ounding e ec o BMI on he associa ion o PAI and isk o psychosis o schizoph enia was u he checked wi h sensi i i y analyses, subs i u ing BMI wi h unde - weigh s. no unde weigh and o e weigh s. no o e weigh in he uni a ia e and mul i a ia e models. S a is ical analyses we e done using SAS® e sion 9.4 (SAS Ins i u e, Ca y, NC, USA). CONCLUSIONS We ound ha physical ac i i y in childhood and adolescence is an independen isk ac o o la e de elopmen o non-a ec i e psychosis. Fu he esea ch is needed o assess he ole and possibili ies o ea ly exe cise and physical ac i i y in e en ion as a pa o psychosis p e en ion. ACKNOWLEDGEMENTS We would like o hank Ville Aal o and I ina Lisinen o da a managemen , he pa icipan s in he YFS s udy and YFS s udy esea che s. The Young Finns S udy has been financially suppo ed by he Academy o Finland: g an s 134309 (Eye), 126925, 121584, 124282, 129378 (Sal e), 117797 (Gendi), and 41071 (Skidi), he Social Insu ance Ins i u ion o Finland, Kuopio, Tampe e and Tu ku Uni e si y Hospi al Medical Funds, Juho Vainio Founda ion, Sig id Juselius Founda ion, Y jö Jahnsson Founda ion, Paa o Nu mi Founda ion, Finnish Founda ion o Ca dio ascula Resea ch and Finnish Cul u al Founda ion, Tampe e Tube culosis Founda ion, and Emil Aal onen Founda ion. AUTHOR CONTRIBUTIONS J.H. and O.R. had ull access o all o he da a in he s udy and ake esponsibili y o he in eg i y o he da a and he accu acy o he da a analysis. J.H., R.K.R.S., J.V., and O.R. we e esponsible o he s udy design. T.T., R.T., N.H.K. we e esponsible o he physical ac i i y/exe cise da a. E.S. d a ed he fi s e sion o he manusc ip and did he analyses. M.M.S. was esponsible o he s a is ical me hods. All au ho s c i ically e ised he manusc ip o impo an in ellec ual con en . All au ho s app o ed he final e sion o publica ion. COMPETING INTERESTS The au ho s epo no biomedical financial in e es s o po en ial conflic s o in e es . REFERENCES 1. Saha, S., Chan , D. & Mcg a h, J. A sys ema ic e iew o mo ali y in schizoph enia. A ch. Gen. Psychia y 64, 1123–1131 (2007). 2. Ol son, M., Ge ha d, T., Huang, C., C ys al, S. & S oup, T. S. P ema u e mo ali y among adul s wi h schizoph enia in he Uni ed S a es. JAMA Psychia y 72, 1172–1181 (2015). 3. 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. JAMA Psychia y 72, 334 (2015). 4. Leuch , S., Bu ka d, T., Hende son, J., Maj, M. & Sa o ius, N. Physical illness and schizoph enia: a e iew o he li e a u e. Ac a Psychia . Scand. 116,317–333 (2007). 5. Co ell, C. U. e al. Ca diome abolic isk in pa ien s wi h fi s -episode schizoph enia spec um diso de s baseline. JAMA Psychia y 71,1350–1363 (2014). 6. Salokangas, R. K. R. Medical p oblems in schizoph enia pa ien s li ing in he communi y (al e na i e acili ies). Cu . Opin. Psychia y 20, 402–405 (2007). 7. Blai , S. N., Cheng, Y. & Holde , J. S. Is physical ac i i y o physical fi ness mo e impo an in defining heal h benefi s? Med. Sci. Spo s Exe c. 33, S379–S420 (2001). 8. Lee, I. M. & Ske e , P. J. Physical ac i i y and all-cause mo ali y: wha is he dose- esponse ela ion? Med. Sci. Spo s Exe c. 33, S459–S494 (2001). 9. Nol e, E. & McKee, C. M. Measu ing he heal h o na ions: upda ing an ea lie analysis. Heal h A . 27,58–71 (2008). 10. Wahlbeck, K., Wes man, J., No den o , M., Gissle , M. & Lau sen, T. M. Ou comes o No dic men al heal h sys ems: li e expec ancy o pa ien s wi h men al dis- o de s. B . J. Psychia y 199, 453–458 (2011). 11. Daumi , G. L. e al. Physical ac i i y pa e ns in adul s wi h se e e men al illness. J. Ne . Men . Dis. 193, 641–646 (2005). 12. Roick, C. e al. Heal h habi s o pa ien s wi h schizoph enia. Soc. Psychia y Psy- chia . Epidemiol. 42, 268–276 (2007). 13. Vancamp o , D. e al. P omo ion o ca dio espi a o y fi ness in schizoph enia: a clinical o e iew and me a-analysis. Ac a Psychia . Scand. 132, 131–143 (2015). 14. Okkenhaug, A. e al. Physical ac i i y in adolescen s who la e de eloped schi- zoph enia: a p ospec i e case-con ol s udy om he young-HUNT. No d. J. Psy- chia y 70, 111–115 (2016). 15. Koi ukangas, J. e al. Physical ac i i y and fi ness in adolescen s a isk o psy- chosis wi hin he No he n Finland 1986 bi h coho . Schizoph . Res. 116, 152–158 (2010). 16. Da idson, M. e al. Beha io al and in ellec ual ma ke s o schizoph enia in appa en ly heal hy male adolescen s. Am. J. Psychia y 156, 1328–1335 (1999). 17. Cooney, G. e al. Exe cise o dep ession ( e iew). Coch ane Da abase Sys Re . Issue 5. A . No.:CD004366 (2013). 18. Vancamp o , D., Rosenbaum, S., Wa d, P. B. & S ubbs, B. Exe cise imp o es ca - dio espi a o y fi ness in people wi h schizoph enia: a sys ema ic e iew and me a-analysis. Schizoph . Res. 169, 453–457 (2015). 19. Fi h, J., Co e , J., Ellio , R., F ench, P. & Yung, A R. A sys ema ic e iew and me a- analysis o exe cise in e en ions in schizoph enia pa ien s. Psychol. Med. 45, 1343–1361 (2015). 20. Dauwan, M., Begemann, M. J. H., He inga, S. M. & Somme , I. E. Exe cise imp o es clinical symp oms, quali y o li e, global unc ioning, and dep ession in schizo- ph enia: a sys ema ic e iew and me a-analysis. Schizoph . Bull. 42, 588–599 (2015). 21. Pea sall, R., Smi h, D. J., Pelosi, A. & Geddes, J. Exe cise he apy in adul s wi h se ious men al illness: a sys ema ic e iew and me a-analysis. BMC Psychia y 14, 117 (2014). E ec s o childhood and adolescence physical ac i i y pa e ns E So munen e al 6 npj Schizoph enia (2017) 5 Published in pa ne ship wi h he Schizoph enia In e na ional Resea ch Socie y 22. K ogh, J. e al. Can exe cise inc ease fi ness and educe weigh in pa ien s wi h schizoph enia and dep ession? F on . Psychia y 5, 89 (2014). 23. Hulsho Pol, H. E. & Kahn, R. S. Wha happens a e he fi s episode? A e iew o p og essi e b ain changes in ch onically ill pa ien s wi h schizoph enia. Schi- zoph . Bull. 34, 354–366 (2007). 24. Shen on, M. E., Dickey, C. C., F umin, M. & McCa ley, R. W. A e iew o MRI findings in schizoph enia. Schizoph . Res. 49,1–52 (2001). 25. W igh , I. C. e al. Me a-analysis o egional b ain olumes in schizoph enia. Am. J. Psychia y 157,16–25 (2000). 26. Pajonk, F.-G. e al. Hippocampal plas ici y in esponse o exe cise in schizo- ph enia. A ch. Gen. Psychia y 67, 133–143 (2010). 27. Scheewe, T. W. e al. Exe cise he apy, ca dio espi a o y fi ness and hei e ec on b ain olumes: a andomised con olled ial in pa ien s wi h schizoph enia and heal hy con ols. Eu . Neu opsychopha macol. 23, 675–685 (2013). 28. S a ko a, A. e al. Physical exe cise keeps he b ain connec ed: biking inc eases whi e ma e in eg i y in pa ien s wi h schizoph enia and heal hy con ols. Schizoph . Bull. 41, 869–878 (2015). 29. Walke , E. & Lewine, R. J. P edic ion o adul -onse schizoph enia om childhood home mo ies o he pa ien s. Am. J. Psychia y 147, 1052–1056 (1990). 30. Jones, P., Rodge s, B., Mu ay, R. & Ma mo , M. Child de elopmen isk ac o s o adul schizoph enia in he B i ish 1946 bi h coho . Lance 19, 1398–1402 (1994). 31. Cannon, M. e al. E idence o ea ly-childhood, pan-de elopmen al impai men specific o schizoph eni o m diso de : esul s om a longi udinal bi h coho . A ch. Gen. Psychia y 59, 449–456 (2002). 32. Isohanni, M. e al. Ea ly de elopmen al miles ones in adul schizoph enia and o he psychoses: a 31-yea ollow-up o he No he n Finland 1966 bi h coho . Schizoph . Res. 52,1–19 (2001). 33. Sø ensen, H. J. e al. Ea ly de elopmen al miles ones and isk o schizoph enia: a 45-yea ollow-up o he copenhagen pe ina al coho . Schizoph . Res. 118,41–47 (2010). 34. Gould, D. & Weiss, M. R. in Ad anced in Pedia ic Spo Sciences: Beha iou al Issues (eds Gould, D., Weiss, M.) (Human Kine ics, 1987). 35. Nupponen, H., Laakso, L., Rimpelä, A., Pe e, L. & Telama, R. Ques ionnai e-assessed mode a e o igo ous physical ac i i y o he Finnish you h in 1979–2005. Scand. J. Med. Sci. Spo 20,1–7 (2010). 36. Wahlbeck, K., Fo sén, T., Osmond, C., Ba ke , D. J. & E iksson, J. G. Associa ion o schizoph enia wi h low ma e nal body mass index, small size a bi h, and hinness du ing childhood. A ch. Gen. Psychia y 58,48–52 (2001). 37. Weise , M. e al. Body mass index and u u e schizoph enia in Is aeli male ado- lescen s. J. Clin. Psychia y 65, 1546–1549 (2004). 38. Sande s, R., Kesha an, M. & Schoole , N. Neu ological examina ion abno mali ies in neu olep ic-nai e pa ien s wi h fi s -b eak schizoph enia: p elimina y esul s. Am. J. Psychia y 151, 1231–1233 (1994). 39. Gup a, S. e al. Neu ological so signs in neu olep ic-nai e and neu olep ic- ea ed schizoph enic pa ien s and in no mal compa ison subjec s. Am. J. Psy- chia y 152, 191–196 (1995). 40. Pe al a, V. e al. Risk ac o s, p e-mo bid unc ioning and episode co ela es o neu ological so signs in d ug-nai e pa ien s wi h schizoph enia-spec um dis- o de s. Psychol. Med. 41, 1279–1289 (2011). 41. G een, M. F. & Lei man, D. I. Social cogni ion in schizoph enia. Schizoph . Bull. 34, 670–672 (2008). 42. Fi h, J. e al. Ae obic exe cise imp o es cogni i e unc ioning in people wi h schizoph enia: a sys ema ic e iew and me a-analysis. Schizoph . Bull; doi: 10.1093/schbul/sbw115 (2016). 43. an P aag, H. Exe cise enhances lea ning and hippocampal neu ogenesis in aged mice. J. Neu osci. 25, 8680–8685 (2005). 44. E iksson, P. S. e al. Neu ogenesis in he adul human hippocampus. Na . Med. 4, 1313–1317 (1998). 45. E ns , A. e al. Neu ogenesis in he s ia um o he adul human b ain. Cell 156, 1072–1083 (2014). 46. In a, D., Lang, U. E., Bo gwa d , S., Meye -Lindenbe g, A. & Gass, P. Adul neu o- genesis in he human s ia um: possible implica ions o psychia ic diso de s. Mol. Psychia y 21, 446–447 (2016). 47. Mäkiky ö, T. e al. Accu acy o egis e -based schizoph enia diagnoses in a gene ic s udy. Eu . Psychia y 13,57–62 (1998). 48. Su isaa i, J. M., Haukka, J. K., Tanskanen, A. J. & Lönnq is , J. K. Decline in he incidence o schizoph enia in Finnish coho s bo n om 1954 o 1965. A ch. Gen. Psychia y 56, 733–740 (1999). 49. Mansikkaniemi, K. e al. C oss-sec ional associa ions be ween physical ac i i y and selec ed co ona y hea disease isk ac o s in young adul s: he ca dio as- cula isk in young Finns s udy. Ann. Med. 44, 733–744 (2012). 50. Tudo -Locke, C., Williams, J., Reis, J. & Plu o, D. U ili y o pedome e s o assessing physical ac i i y: cons uc alidi y. Spo s Med. 34, 281–291 (2004). 51. Telama, R. e al. Physical ac i i y om childhood o adul hood: a 21-yea acking s udy. Am. J. P e . Med. 28, 267–273 (2005). 52. Rai aka i, O. T. e al. Coho p ofile: he ca dio ascula isk in young Finns s udy. In . J. Epidemiol. 37, 1220–1226 (2008). 53. Telama, R., Leskinen, E. & Yang, X. S abili y o habi ual physical ac i i y and spo pa icipa ion: a longi udinal acking s udy. Scand. J. Med. Sci. Spo s 6, 371–378 (1996). 54. Cole, T. J., Flegal, K. M., Nicholls, D. & Jackson, A. A. Body mass index cu o s o define hinness in child en and adolescen s: in e na ional su ey. B . Med. J. 335, 1–8 (2007). 55. Cole, T. J., Bellizzi, M. C., Flegal, K. M. & Die z, W. H. Es ablishing a s anda d defini ion o child o e weigh and obesi y wo ldwide: in e na ional su ey. B . Med. J. 320, 1240–1243 (2000). 56. Zou, G. A modified poisson eg ession app oach o p ospec i e s udies wi h bina y da a. Am. J. Epidemiol. 159, 702–706 (2004). 57. Zou, G. Y. & Donne , A. Ex ension o he modified poisson eg ession model o p ospec i e s udies wi h co ela ed bina y da a. S a . Me hods Med. Res.22, 661–670 (2011). This wo k is licensed unde a C ea i e Commons A ibu ion 4.0 In e na ional License. The images o o he hi d pa y ma e ial in his a icle a e included in he a icle’s C ea i e Commons license, unless indica ed o he wise in he c edi line; i he ma e ial is no included unde he C ea i e Commons license, use s will need o ob ain pe mission om he license holde o ep oduce he ma e ial. To iew a copy o his license, isi h p://c ea i ecommons.o g/licenses/by/ 4.0/ © The Au ho (s) 2017 Supplemen a y In o ma ion accompanies he pape on he npj Schizoph enia websi e (doi:10.1038/s41537-016-0007-z). E ec s o childhood and adolescence physical ac i i y pa e ns E So munen e al 7 Published in pa ne ship wi h he Schizoph enia In e na ional Resea ch Socie y npj Schizoph enia (2017) 5