Jou nal o A ec i e Diso de s 358 (2024) 70–78
A ailable online 30 Ap il 2024
0165-0327/© 2024 The Au ho (s). Published by Else ie B.V. This is an open access a icle unde he CC BY license (h p://c ea i ecommons.o g/licenses/by/4.0/).
Resea ch pape
Associa ion be ween ajec o ies o he neighbo hood social exposome and
men al heal h in la e adolescence: A FinnTwin12 coho s udy
Zhiyang Wang
a
, Gabin D oua d
a
, Alyce M. Whipp
a
, Ma ja Heinonen-Guzeje
b
, Gab iele Bol e
c
,
Jaakko Kap io
a
,
b
,
*
a
Ins i u e o Molecula Medicine Finland, Helsinki Ins i u e o Li e Science, Uni e si y o Helsinki, Helsinki, Finland
b
Depa men o Public Heal h, Clinicum, Uni e si y o Helsinki, Helsinki, Finland
c
Depa men o Social Epidemiology, Ins i u e o Public Heal h and Nu sing Resea ch, Uni e si y o B emen, B emen, Ge many
ARTICLE INFO
Keywo ds:
Adolescence
Exposome
Men al heal h
T ajec o y
Social inequali y
ABSTRACT
Backg ound: Adolescen men al heal h p oblems impose a signi ican bu den. Explo ing e ol ing social en i-
onmen s could enhance comp ehension o hei impac on men al heal h. We aimed o depic he ajec o ies o
he neighbo hood social exposome om middle o la e adolescence and assess he in ica e ela ionship be ween
hem and la e adolescen men al heal h.
Me hods: Pa icipan s (n =3965) om he FinnTwin12 coho wi h comple ed ques ionnai es a age 17 we e
used. Nine men al heal h measu es we e assessed. The social exposome comp ised 28 neighbo hood social in-
dica o s. T ajec o ies o hese indica o s om ages 12 o 17 we e summa ized ia la en g ow h cu e modeling
in o g ow h ac o s, including baseline in e cep . Mix u e e ec s o all g ow h ac o s we e assessed h ough
quan ile-based g-compu a ion. Repea ed gene alized linea eg essions iden i ied signi ican g ow h ac o s. Sex
s a i ica ion was pe o med.
Resul s: The linea -quad a ic model was he mos op imal ajec o y model. No mix u e e ec was de ec ed.
Reg ession models showed some g ow h ac o s salien ly linked o he p- ac o , in e nalizing p oblems, anxie y,
hype ac i i y, and agg ession. The majo i y o hem we e baseline in e cep s. Quad a ic g ow h ac o s abou
mo he ongues co ela ed wi h anxie y among sex-combined pa icipan s and males. The linea g ow h ac o in
he p opo ion o households o couples wi hou child en was associa ed wi h in e nalizing p oblems in emales.
Limi a ions: We we e limi ed o including only neighbo hood-le el social exposu es, and he mul ile el con ex ual
exposome si ua ion in e e ed wi h ou assessmen .
Conclusions: T ajec o ies o he social neighbo hood exposome modes ly in luenced la e adolescen men al
heal h. Tackling oo causes o social inequali ies h ough a ge ed p og ams o li ing condi ions could imp o e
adolescen men al heal h.
1. In oduc ion
Men al heal h p oblems a e escala ing as a se e e public heal h
bu den, wi h a s iking inc easing end among younge indi iduals.
Empi ical s udies ac oss di e en coun ies ha e demons a ed ha
nowadays adolescen s and young adul s expe ience mo e men al heal h
p oblems han olde people (Goodwin e al., 2022; K oks ad e al.,
2022), and he pandemic has exace ba ed his d ama ic si ua ion
(COVID-19 Men al Diso de s Collabo a o s, 2021). As a c i ical de el-
opmen al s age, adolescen s men al heal h is suscep ible o he in luence
o he ex e nal en i onmen , o which he social en i onmen is an
impo an componen . Mul iple social de e minan s ha e been aised
such as po e y and employmen s a us (Aleg ía e al., 2018).
In social epidemiology, li ecou se esea ch ypically conside s he
ela ionship be ween exposu e and heal h in h ee ypes o isk models:
c i ical pe iod, accumula ion, and ajec o y (Glymou e al., 2014).
Dis inguishing he isk models helps o in es iga e he na u e o he
ela ionship and guide he in e en ion. Some s udies ha e de ec ed
signi ican ela ionships be ween c oss-sec ional social exposu es and
adolescen men al heal h (Xue e al., 2005; Keles e al., 2020; Huang
e al., 2023). Ne e heless, ou p e ious exposome-wide associa ion
s udy included a ious social indica o s a he pos al code o
* Co esponding au ho a : Ins i u e o Molecula Medicine, Uni e si y o Helsinki, PL 20 (Tukholmanka u 8), FI-00014 Helsinki, Finland.
E-mail add ess: [email p o ec ed] (J. Kap io).
Con en s lis s a ailable a ScienceDi ec
Jou nal o A ec i e Diso de s
jou nal homepage: www.else ie .com/loca e/jad
h ps://doi.o g/10.1016/j.jad.2024.04.096
Recei ed 18 Oc obe 2023; Recei ed in e ised o m 14 Ap il 2024; Accep ed 21 Ap il 2024
Jou nal o A ec i e Diso de s 358 (2024) 70–78
71
municipali y le el a h ee ime poin s and did no ind any signi ican
associa ion o hese neighbo hood-le el exposu es wi h dep essi e
symp oms in la e adolescence and young adul hood (Wang e al., 2023).
Ano he s udy also ejec ed he hypo hesis o he c i ical pe iod be ween
indi idual inancial ha dship in adolescence and men al heal h condi-
ions in he B i ish popula ion (Mo issey and Kinde man, 2020).
Fu he mo e, in he US, changes in neighbo hood economic disad an-
age om ea ly childhood o adolescence signi ican ly in luenced majo
dep essi e diso de symp oms in la e adolescence (Buu e al., 2009).
Du ing se e e economic ecession in Finland in he ea ly 1990s, he
educ ion in disposable amily income de e io a ed child en’s men al
heal h, and he e was a long- e m ecip ocal in luence be ween
comp omised pa en ing and child men al heal h, con ibu ing o chil-
d en’s in e nalizing and ex e nalizing symp oms (Solan aus e al.,
2004). The e o e, we deduced ha mul iple isk models o social ex-
posu es’ e ec on men al heal h may coexis . Longi udinal design wi h
g ow h modeling could be sui able o assessing ime- a ying mecha-
nisms (bo h baseline and change) unde lying he e ec s o he social
en i onmen on men al heal h, such as adap i e esponse ea ly in
de elopmen (Glymou e al., 2014). Howe e , p e ious in es iga ions
ha e mos ly ocused on single o limi ed numbe s o social exposu es.
The social exposome depic s a holis ic and comp ehensi e po ayal
o he social en i onmen wi h h ee co e p inciples: mul idimension-
ali y, ecip oci y, and iming and con inui y (Gudi-Minde mann e al.,
2023). Howe e , con en ional in es iga ions canno simul aneously
es ima e he complexi y, in eg i y, o spa io empo ali y o he social
exposome. Se e al p e ious exposome s udies a ge ing men al heal h
wi h a ange o social exposu es nei he employed he ajec o y model
no emphasized in e io complexi y (Choi e al., 2020; Lin e al., 2022;
Ni e al., 2020; P ies e al., 2022; an de Weije e al., 2022). To ul ill
his esea ch gap, ou s udy aimed o in es iga e he in ica e ela ion-
ship be ween ajec o ies o he neighbo hood social exposome and
men al heal h in la e adolescence wi h he ollowing objec i es: 1) o
cha ac e ize he ajec o y o each social indica o om mid o la e
adolescence, 2) o assess he mix u e e ec o g ow h ac o s o he social
exposome on men al heal h in la e adolescence, and 3) o iden i y
g ow h ac o s o he social exposome on men al heal h in la e
adolescence.
2. Ma e ial and me hods
2.1. Pa icipan s
This s udy was based on he longi udinal FinnTwin12 coho , which
is a popula ion-based p ospec i e coho o Finnish wins bo n be ween
1983 and 1987. A baseline, 5522 wins we e in i ed o pa icipa e, and
87 % o hei amilies ag eed o ake pa . Th ee ollow-ups we e con-
duc ed a age 14 (wa e 2), age 17 (wa e 3), and young adul hood (wa e
4). A baseline and a wa es 2 and 3, da a collec ion was conduc ed o e
a 5-yea pe iod, one bi h yea annually, wi h ques ionnai es sen ou
close o he win’s bi hdays. Fo he pu poses o he cu en s udy, only
da a om wins who pa icipa ed in wa e 3 we e used. O he 4594
wins who we e in i ed in his wa e, 4239 comple ed he ques ionnai es
a a mean age o 17.6 (s anda d de ia ion (SD): 0.3) and 2191 (51.7 %)
we e emale. An upda ed e iew o his coho is a ailable elsewhe e
(Rose e al., 2019).
2.2. Measu es
2.2.1. Indica o s o he neighbo hood social exposome
We de i ed social indica o s a he pos al code (2022’s e sion) le el
be ween 1987 and 2021 om S a is ics Finland and geocodes o wins’
esidences om bi h o 2021 om he Digi al and Popula ion Da a
Se ices Agency, Finland. Then, he map p o ided by Es i Finland,
con aining 2022’s pos al code in o ma ion, was u ilized o link he wins
o social indica o s o each yea om ages 12 (1995–1999) o 17
(2000–2004). Age 12 was selec ed as he s a ing poin o a oid he
impac o he ea ly 1990s se e e economic dep ession in Finland (Sol-
an aus e al., 2004). We combined se e al educa ion and age- ela ed
indica o s, esul ing in a o al o 28 social indica o s. They we e ca e-
go ized in o six domains: 1) age s uc u e, 2) educa ional le el, 3) un-
employmen , 4) eligion, language, and mig a ion, 5) household
s uc u e, and 6) household income. These social indica o s ep esen ed
he popula ion s uc u e a he pos al code, which is a good p oxy o
desc ibing a neighbo hood (Weck o h e al., 2022). The e we e o e
1100 pos al codes used in each age in he s udy, and he a e age numbe
o esiden s inside he pos al codes was 1802 (Kemppainen e al., 2023).
The a iable name, desc ip ion, and s a is ics o indica o s a e p esen ed
in Supplemen a y Table 1.
2.2.2. Men al heal h measu e
The sel - epo ed Mul idimensional Pee Nomina ion In en o y
(MPNI) was used o assess wins’ men al heal h a age 17 (Pulkkinen
e al., 1999). I is a 37-i em in en o y o ep esen a model o emo ional
and beha io al egula ion wi h a 4-poin scale om 0 o 3. The e we e
h ee classes o men al heal h measu es, a di e en le els o agg ega-
ion o i ems. In he i s class, we used he i e subscales: dep ession
( wo i ems), social anxie y (anxie y he ea e , h ee i ems),
hype ac i i y-impulsi i y (hype ac i i y he ea e , six i ems), agg essi e
beha io s (agg ession he ea e , six i ems), and ina en ion ( ou i ems).
Mean sco es, which we e he a e age ac oss all he esponse i ems in
each subscale, we e calcula ed as ou comes, and no missing i em was
allowed in any subscale. In he second class, he e we e h ee-
dimensional measu es: in e nalizing p oblems, ex e nalizing p oblems,
and p osocial beha io s. In e nalizing p oblems consis ed o dep ession
and anxie y subscale i ems as well as an i em on ic imiza ion. Ex e -
nalizing p oblems consis ed o hype ac i i y, agg ession, and ina en-
ion subscale i ems. Twel e i ems composed he measu e o p osocial
beha io s. Two missing i ems we e allowed o hese measu es in he
second class. The psychopa hology ac o (p- ac o ) was he hi d class’s
measu e, which was composed o wo-dimensional componen s in he
second class (in e nalizing p oblems and ex e nalizing p oblems). Fou
missing i ems we e allowed o he p- ac o . Fo bo h second and hi d
classes, we used mean sco es o measu es. Thus, he e a e a o al o nine
MPNI measu es. A highe sco e implied mo e occu ences o he p ob-
lem/beha io . The MPNI scale has been alida ed and showed p edic-
i e powe o psychia ic diso de diagnoses in ea ly adul hood
(Pulkkinen, 2017; Whipp e al., 2019).
2.2.3. Co a ia es
We included se e al co a ia es a p io i in analyses: sex (male, e-
male), zygosi y (monozygo ic, dizygo ic, unknown), pa en al educa ion,
smoking (ne e , o me , occasional, cu en ), and s udy and wo k s a us
(nei he s udy no wo k, only s udy, only wo k). Sex was assigned a
bi h. I bo h pa en s had low educa ion (less han 12 yea s) o high
educa ion (12 o mo e yea s), pa en al educa ion was ca ego ized as
“limi ed” o “high,” espec i ely. In cases o di e ing pa en al educa ion
le els, i was classi ied as “in e media e” (Huppe z e al., 2017). Then,
he e was a co a ia e used o e lec he physical en i onmen :
No malized Di e ence Vege a ion Index (NDVI). I desc ibes he di -
e ence be ween isible and nea -in a ed e lec ance o ege a ion
co e o quan i y ege a ion g eenness (Schinasi e al., 2018). The NDVI
was measu ed a he geocode o wins’ esidences a age 17, wi hin a
100 m bu e du ing he whole yea . Because we emphasized con ex ual
aspec s o he social exposome a he neighbo hood le el in his s udy,
we conside ed sex and pa en al educa ion as co a ia es, al hough hey
a e also impo an social de e minan s.
2.3. S a is ical analysis
2.3.1. La en g ow h cu e modeling o ajec o ies
We employed la en g ow h cu e models o summa ize ajec o ies
Z. Wang e al.
Jou nal o A ec i e Diso de s 358 (2024) 70–78
72
o social indica o s in o la en a iables, which allows good co espon-
dence be ween he s a is ical model and he heo y (P eache e al.,
2008). These la en a iables, commonly e e ed o as g ow h ac o s,
we e calcula ed wi h pa hways in luencing he social indica o s a age
12, 13, 14, 15, 16, and 17. The pa hway coe icien s we e se acco ding
o he cha ac e is ics o each g ow h ac o . We speci ied ou models o
each social indica o a i s : 1) linea model, 2) quad a ic model, 3)
linea -quad a ic model, and 4) loga i hmic model. A diag am o he
linea -quad a ic model is shown in Fig. 1. Bayesian in o ma ion c i e-
ion (BIC), Compa a i e Fi Index (CFI), and s anda dized oo mean
squa e esidual (SRMR) we e used o e alua e model i ness and selec
he mos -op imis ic model (Xia and Yang, 2019). G ow h ac o s om
he mos -op imis ic model we e hen gene a ed and s anda dized (i.e.,
mean ze o and 1-uni a iance) o ollowing analyses. CFI o e 0.9 and
SRMR below 0.09 we e conside ed good model-da a i (Xia and Yang,
2019).
We included all wins wi h alid geocodes en olled a baseline (in-
di idual win n =5089) o la ge gains in s a is ical powe . The
maximum likelihood and obus s anda d e o s we e used o allow
missing alues du ing model i ing and o con ol o clus e ing in he
da a due o amilial ela edness, espec i ely. A small numbe o he
unemploymen indica o s had no been collec ed o some pos al codes
o a pe iod, and we eplaced hem wi h hei coun e pa s a he mu-
nicipali y le el. Fo example, 5.5 % o indi idual wins we e missing
in o ma ion o he a iable “unemp55” in all six yea s. The R package
“la aan” ( e sion 0.6-15) was used (Rosseel, 2012).
2.3.2. Exposome plu alis ic analysis
Pa icipan s wi hou in o ma ion on men al heal h measu es and
co a ia es we e excluded, hus 3965 indi idual wins we e included in
he inal analysis. The pa e ns o he included and excluded pa icipan s
on g ow h ac o s we e assessed by uni a ia e linea eg ession. A
Pea son co ela ion ma ix accoun ing o he clus e ing in he da a due
o amilial ela edness was c ea ed. The no mali y o men al heal h
measu es was isually assessed h ough his og ams. Mos men al heal h
measu es di e ed be ween sexes, so sex s a i ica ion was pe o med in
he ollowing plu alis ic analyses (Pulkkinen e al., 1999).
Ini ially, we se an a p io i hypo hesis: “E e y ac o ma e s”.
Quan ile g-compu a ion, le e aging he co ela ion be ween g ow h
ac o s, was used o gene a e unbiased es ima es o he o e all e ec o
he mix u e o g ow h ac o s on each men al heal h measu e wi h in-
di idual ac o weigh s in bo h di ec ions (Keil e al., 2020). I is based
on he con en ional weigh ed quan ile sum eg ession and is an e icien
app oach o assess mix u e e ec s (Keil e al., 2020). Clus e e ec s
induced by amily ela edness was con olled o . The R package
“qgcomp” ( e sion 2.10.1) was used (Keil e al., 2020).
When he hypo hesis o he mix u e e ec was null, we i ed
gene alized linea eg ession models epea edly o each g ow h ac o
wi h each men al heal h measu es o selec he mos impo an g ow h
ac o s. Models we e adjus ed o co a ia es, and obus s anda d e o s
we e calcula ed o con ol o amily ela edness. We a p io i speci ied a
P- alue o 0.001 as a signi ican h eshold o mul iple es ing. The R
package “ exposome” ( e sion 1.25.2) was used (He nandez-Fe e e al.,
2019).
3. Resul s
3.1. Cha ac e is ics o included pa icipan s and g ow h ac o s
O he 3965 included indi idual wins (Table 1), he majo i y we e
emale (52 %) and dizygo ic wins (62 %), and 58 % o pa en s had a
limi ed le el o educa ion. A age 17, 25 % o indi idual wins epo ed
being cu en smoke s and 83 % we e ull- ime s uden s and no wo k-
ing. The cha ac e is ic o men al heal h measu es by sex a e shown in
Supplemen a y Table 2. Th ough uni a ia e linea eg ession, emales
had signi ican ly highe mean sco es o he p- ac o , in e nalizing
p oblems, dep ession, anxie y, and hype ac i i y, while he mean sco e
Fig. 1. Pa h diag am o linea -quad a ic la en g ow h cu e modeling.
Table 1
Cha ac e is ics o wins included in he plu alis ic analysis.
Cha ac e is ics N (%) / Mean (SD)
Sex-combined
(indi idual win n
=3965)
Male
(indi idual win
n =1898)
Female
(indi idual win
n =2067)
Zygosi y
Monozygo ic 1317 (33.2) 593 (31.2) 724 (35.0)
Dizygo ic 2464 (62.1) 1226 (64.6) 1238 (59.9)
Unknown 184 (4.6) 79 (4.2) 105 (5.1)
Pa en al educa ion
Limi ed 2290 (57.8) 1082 (57.0) 1208 (58.4)
In e media e 914 (23.1) 453 (23.9) 461 (22.3)
High 761 (19.2) 363 (19.1) 398 (19.3)
Smoking
Ne e 1178 (29.7) 589 (31.0) 589 (28.5)
Fo me 1363 (34.4) 655 (34.5) 708 (34.3)
Occasional 430 (10.8) 173 (9.1) 257 (12.4)
Cu en 994 (25.1) 481 (25.3) 513 (24.8)
S udy and wo k
s a us
Nei he s udy no
wo k
144 (3.6) 54 (2.9) 90 (4.4)
Only s udy 3275 (82.6) 1631 (85.9) 1644 (79.5)
Bo h s udy and
wo k
546 (13.8) 213 (11.2) 333 (16.1)
No malized
di e ence
ege a ion index
0.5 (0.1) 0.5 (0.1) 0.5 (0.1)
Z. Wang e al.
Jou nal o A ec i e Diso de s 358 (2024) 70–78
73
o ina en ion was signi ican ly highe in males. The la ges co ela ion
was be ween he p- ac o and ex e nalizing p oblems in sex-combined
and sex-s a i ied wins (Supplemen a y Fig. 1). P osocial beha io
was nega i ely co ela ed wi h all o he measu es.
Fo all 28 social indica o s, he linea -quad a ic model was he mos
op imis ic and bes i ed he da a (Supplemen a y Table 3). A o al o 84
g ow h ac o s we e de i ed om he linea -quad a ic model, and all
showed signi ican a ia ions be ween indi iduals. Mos g ow h ac o s
did no di e be ween pa icipan s included o excluded in he inal
analysis, sugges ing a low isk o selec ion bias (Supplemen a y Table 4).
We obse ed s ong co ela ions wi hin he domains o social indica o s
and be ween domains: “household s uc u e” and “age s uc u e”,
“household income” and “unemploymen ”, and “household income” and
“ eligion, language, and mig a ion” in he analyzed neighbo hood
(Supplemen a y Fig. 2).
3.2. Mix u e e ec by quan ile g-compu a ion
A e adjus ing o all co a ia es, we did no ind any signi ican
associa ions be ween he mix u e o g ow h ac o s and any o he nine
men al heal h measu es in sex-combined wins (Table 2). A e sex
s a i ica ion, s ill, no signi ican associa ion was iden i ied (Table 2).
3.3. Associa ions be ween g ow h ac o s and men al heal h measu es
using gene alized linea eg ession
The esul s among sex-combined wins a e p esen ed in Fig. 2 A e
adjus ing o co a ia es, he highe in e cep o edu1 (p opo ion o
unknown o lowe seconda y educa ion le el o lowe in he neighbo -
hood) was signi ican ly associa ed wi h highe mean sco es o in e -
nalizing p oblems (be a: 0.03, 95 % con idence in e al (CI): 0.01, 0.04)
(Fig. 2B) and anxie y (be a: 0.04, 95 % CI: 0.02, 0.06) (Fig. 2D). The
quad a ic g ow h ac o s o insa (p opo ion wi h Finnish as mo he
ongue) (be a: -0.03, 95 % CI: −0.05, −0.01) and swe (p opo ion wi h
Swedish as mo he ongue) (be a: 0.03, 95 % CI: 0.01, 0.05) we e
signi ican ly associa ed wi h anxie y (Fig. 2D). Fu he mo e, he in e -
cep o household1 (p opo ion o single households) (be a: -0.02, 95 %
CI: −0.04, −0.01) was signi ican ly associa ed wi h hype ac i i y
(Fig. 2F).
In males only, he highe in e cep o household1 was associa ed
wi h lowe mean sco es o he p- ac o (be a: -0.02, 95 % CI: −0.04,
−0.01) (Fig. 3A). The in e cep s o edu3 (p opo ion o educa ion le el
o bachelo o abo e) (be a: -0.05, 95 % CI: −0.08, −0.02) and e lu h
(belonging o Lu he an Chu ch communi y) (be a: 0.05, 95 % CI: 0.03,
0.08) we e signi ican ly associa ed wi h anxie y (Fig. 3D).
In emales only, he highe linea g ow h ac o s o household3
(p opo ion o households o couples wi hou child en) and swe we e
signi ican ly associa ed wi h smalle mean sco es o in e nalizing
p oblems (be a: -0.04, 95 % CI: −0.06, −0.02) (Fig. 4B) and anxie y
(be a: -0.05, 95 % CI: −0.09, −0.02) (Fig. 4D), espec i ely. The in-
e cep s o unemp1824 (unemploymen a e be ween age 18 and 24)
(be a: 0.04, 95 % CI: 0.02, 0.06) and umep55 (unemploymen a e o e
age 55) (be a: 0.04, 95 % CI: 0.02, 0.06) we e posi i ely associa ed wi h
agg ession, bu he in e cep o swe was nega i ely associa ed wi h
agg ession (be a: -0.03, 95 % CI: −0.05, −0.01) (Fig. 4G).
4. Discussion
In he FinnTwin12 coho , we explo ed he in ica e e ec s o a-
jec o ies o indica o s o he neighbo hood social exposome on a ange o
men al heal h measu es in la e adolescence. We did no de ec any
mix u e e ec in sex-combined o -s a i ied analyses. Assuming some
g ow h ac o s a e po en ially mo e impo an , we pe o med a ac o
selec ion ia epea ed eg ession models. G ow h ac o s om he do-
mains o “educa ional le el,” “unemploymen ,” “ eligion, language, and
mig a ion,” and “household s uc u e” we e associa ed wi h men al
heal h in sex-combined o sex-s a i ied analyses, while hei e ec sizes
we e ela i ely small. None heless, some associa ions, o example be-
ween he ajec o y o he unemploymen s a us o he esiden ial a ea
and agg ession, we e seen e en a e accoun ing o he e ec s o
pa en al educa ion and own s udy and wo k s a us.
The neighbo hood social exposome and i s ajec o y modes ly
a ec ed adolescen men al heal h. Al hough mos g ow h ac o s iden-
i ied we e in e cep (baseline) ac o s, some ac o s indica ing changes
we e also ound o be ela ed o language and household. Findings wi h
baseline ac o s we e consis en wi h p e ious s udies ha men al heal h
and well-being we e equen ly linked o neighbo hood dep i a ion
including aspec s o po e y, income, unemploymen , and educa ion
among young people (Ji aj e al., 2020; Visse e al., 2021; Weck o h
e al., 2022). Al hough ew s udies ha e di ec ly b idged he ajec o ies
o neighbo hood household and mo he ongue s uc u es o men al
heal h, hese ajec o ies we e usually co ela ed wi h he change o
socioeconomic s a us (Ande s´
en e al., 2021; Hyypp¨
a and M¨
aki, 2001;
Koops e al., 2021). F om a mo e holis ic pe spec i e, ajec o ies ep-
esen ed he changes unde gone in he social en i onmen , d i ing so-
cial seg ega ion. Seg ega ion was shown o induce egional and
socioeconomic g adien s and dispa i ies in heal hca e esou ces in
Finland (Kajan ie e al., 2006). A UK na u al expe imen showed ha
neighbo hood-le el egene a ion p ojec s, amelio a ing social
inequali y, imp o ed esiden s’ men al heal h, which pa ially suppo s
he ajec o y e ec (Whi e e al., 2017). Addi ionally, sex-speci ic e -
ec s o he ajec o ies o he social exposome we e obse ed in ou
indings. B azil and Cla k ound ha he ela ionship be ween changes
in men al heal h and neighbo hood po e y was s onge in emales han
males (B azil and Cla k, 2017), howe e , he e is e idence o he con-
a y (Ba , 2018; Humph ey and Roo , 2017). This may be due o sex
di e ences in suscep ibili y and he capaci y o espond o a social
con ex (Pe son Waye e al., 2023; S inson, 1985). O e all, he li ecou se
Table 2
Adjus ed es ima es o associa ions be ween he mix u e o s anda dized g ow h ac o s o he neighbo hood social indica o s and men al heal h measu es using
quan ile g-compu a ion.
Men al heal h measu e Adjus ed coe icien (95 % CI)
a
Sex-combined Male Female
P- ac o −0.01 (−0.21, 0.18) −0.11 (−0.39, 0.17) 0.12 (−0.16, 0.39)
In e nalizing p oblems 0.15 (−0.15, 0.46) −0.11 (−0.54, 0.32) 0.33 (−0.11, 0.77)
Dep ession 0.13 (−0.28, 0.55) −0.23 (−0.81, 0.34) 0.37 (−0.23, 0.97)
Anxie y 0.18 (−0.20, 0.57) −0.05 (−0.60, 0.49) 0.39 (−0.16, 0.94)
Ex e nalizing p oblems −0.07 (−0.30, 0.16) −0.11 (−0.44, 0.22) 0.04 (−0.29, 0.37)
Hype ac i i y −0.24 (−0.55, 0.07) −0.41 (−0.85, 0.04) 0.01 (−0.43, 0.45)
Agg ession −0.03 (−0.32, 0.26) 0.10 (−0.32, 0.51) 0.03 (−0.38, 0.44)
Ina en ion 0.11 (−0.18, 0.41) 0.03 (−0.39, 0.44) 0.11 (−0.31, 0.53)
P osocial beha io 0.02 (−0.19, 0.24) −0.02 (−0.34, 0.31) 0.01 (−0.28, 0.30)
a
Adjus ed o sex, zygosi y, pa en al educa ion, smoking a age 17, s udy and wo k s a us a age 17, and no malized di e ence ege a ion index wi h in a 100 m
bu e a age 17.
Z. Wang e al.
Jou nal o A ec i e Diso de s 358 (2024) 70–78
74
app oach, ocusing on he his o y o social con ex s, assis s us o illu-
mina e he empo al impac o he exposome on adolescen men al
heal h (Whea on and Cla ke, 2003). As an implica ion, we sugges ha
policymake s s i e o educe di e ing condi ions and p omo e be e
neighbo hoods om an in eg a ed and dynamic pe spec i e.
O no e, since social indica o s e lec he ex e nal social en i on-
men a he han he indi idual le el, ou in e p e a ion p ima ily o-
cuses on he neighbo hood e ec s. Only a small numbe o g ow h
ac o s we e selec ed, o which e ec sizes we e small, and some e ec s’
di ec ions ( o example, a highe p opo ion o single households a
baseline was associa ed wi h lowe hype ac i i y symp oms) we e no
s aigh o wa d o unde s and. The e o e, we in e ed ha he e ec s o
ajec o ies o he social exposome on men al heal h we e ei he no
subs an ial o we e “con ounded” by some hing else. A mul ile el
con ex ual exposome si ua ion was p oposed, bu only one le el
(neighbo hood) was conside ed in ou analyses (Hu e al., 2023).
Imagining a Swedish-speaking amily esiding in he p edominan ly
Finnish-speaking eas e n pa o Finland, no able di e ences be ween
amilial and neighbo hood en i onmen s po en ially exis in he do-
mains o eligion, language, and mig a ion, which may ecip ocally
mi iga e o exace ba e hei e ec s on men al heal h. In Aus alia,
second-gene a ion immig an s om a non-English-speaking backg ound
had a highe p e alence o men al diso de s han hose om an English-
speaking backg ound (Liddell e al., 2016). Flou i e al. (2016) also
poin ed ou an in e ac ion be ween he neighbo hood single-pa en a e
and amily ma i al s a us on child emo ional symp oms. Weck o h e al.
(2022) ound ha men al heal h and loneliness, as co a ia es, a enu-
a ed he impac o li ing in he inne u ban a ea and unemploymen a e
on he quali y o li e in Finland. In he adi ional assessmen ,
neighbo hood-, amilial- o indi idual-le el exposu es mu ually ac like
con ounde s, bu we a he e e o his si ua ion as a mul ile el s uc-
u e wi hin he exposome. In addi ion, exposu es used in his s udy we e
based on indi idual’s geocodes a he han sel - epo s, which again
equi es conside a ion o in o ma ion g anula i y. A s udy in
Swi ze land sugges ed ha pe cei ed en i onmen al s esso s p edic
men al heal h be e han objec i e measu es among esiden s o e 17
yea s (Gomm and Be naue , 2023). In he u u e, i is impe a i e o
inco po a e mul ile el exposu es om di e en sou ces o cap u e all
dimensions o he social exposome (Gudi-Minde mann e al., 2023) and
conside hei possible cascading e ec s on men al heal h o enhance
ou unde s anding o his in ica e ela ionship.
Unde he social exposome amewo k, we need sophis ica ed and
ad anced me hodologies o deepen ou unde s anding o he ela ion-
ship be ween he exposome and men al heal h. An ABCD coho s udy
ound ha exposome sco es calcula ed wi h ac o analysis, including
neighbo hood exposu es, we e signi ican ly associa ed wi h adolescen
p- ac o sco es (P ies e al., 2022). Many machine lea ning me hods
ha e been employed in o he ypes o exposome analyses and ou comes
(Ohanyan e al., 2022), ecommending he applica ion o plu alis ic
analysis. Neu cou e al. (2023) aised he conside a ion o agnos ics in
social exposome esea ch, owa ds es ing all possible hypo heses
wi hou p ejudice. In he con ex o he exposome amewo k ha in-
eg a es a plu alis ic analysis, we p opose a “semi-agnos ic” pa adigm
whe ein hypo heses a e o mula ed a he exposome le el a he han on
indi idual exposu es. Thus, we made a hypo hesis on he mix u e e ec
in his s udy ini ially, which desc ibes he co-occu ence o mul iple
exposu es in o one la en mix u e, pa icula ly om simila sou ces
(T aini e al., 2022). The mix u e e ec is usually hypo hesized in he
Fig. 2. Volcano plo s o associa ions, ia gene alized linea eg ession, o g ow h ac o s wi h p- ac o (A), in e nalizing p oblems (B), dep ession (C), anxie y (D),
ex e nalizing p oblems (E), hype ac i i y (F), agg ession (G), ina en ion (H), and p osocial beha io (I) among sex-combined pa icipan s (indi idual wins n =
3965). The x-axis ep esen s eg ession coe icien s and he Y-axis ep esen s –log
10
(P- alue) o eg essions. The p e ix i means he in e cep , he p e ix s means he
linea g ow h ac o , and he p e ix q means quad a ic he g ow h ac o .
Z. Wang e al.
Jou nal o A ec i e Diso de s 358 (2024) 70–78
75
ield o oxicoepidemiology, and we employed i o social indica o s
since hey a e om he same sou ce and ela i ely a he same le el o
s uc u e (Hu e al., 2023). Al hough we did no iden i y any signi ican
indings ela ed o he mix u e e ec , we ex ended ou in es iga ion o
iden i y c ucial ac o s and explo e al e na i e assump ions. The “semi-
agnos ic” pa adigm o e s he ad an age o p o iding bo h heo e ical
guidance o u he mechanism explana ion and a b oade pe spec i e
o unco e unknowns. P ac ical applica ion should also conside he
dimensionali y, exposu e s uc u e, esea ch ques ion, and u u e im-
plica ions o cau iously choose pa adigms.
This s udy was s eng hened by i s longi udinal design wi h eliable
esiden ial his o y, enabling he accu a e assessmen o ajec o ies o
indica o s o he social exposome a he neighbo hood le el. The
inco po a ion o he MPNI and a ange o social indica o s allowed us o
comp ehensi ely app aise he in luence o social en i onmen al ends
on men al heal h in la e adolescence. Howe e , he e a e also some
limi a ions. Fi s , he lack o o he dimensions o he social exposome
including indi idual exposu es may ha e a ec ed ou assessmen .
The e o e, one could conside no el models such as neu al ne wo ks o
disen angle he mul ile el con ex ual exposome si ua ion in he u u e.
Second, we only included c oss-sec ional men al heal h measu es, which
we e unable o assess he e ec on he de elopmen o men al heal h.
Thi d, as a coho s udy, we did no ha e he coun e ac ual condi ion,
so we did no imply any causal in e ence and sugges ed being mo e
cau ious on in e p e a ion. A las , we used an a bi a y signi icance
h eshold, which may limi he powe . Fu he s udies a e welcomed o
e i y ou indings.
5. Conclusion
This s udy explo ed he in ica e ela ionship be ween ajec o ies o
he neighbo hood social exposome and adolescen men al heal h.
Al hough we did no iden i y mix u e e ec s be ween ajec o ies and
men al heal h, a small numbe o neighbo hood social exposu es’ a-
jec o ies showed a no iceable e ec on men al heal h in la e adolescence
a e g ow h ac o selec ion. Howe e , he mul ile el con ex ual
exposome si ua ion may obs uc ou assessmen , and in e p e a ion
should be cau ious. O he s udies a e o be encou aged o eplica e and
ex end ou indings. Addi ionally, we ad oca e a “semi-agnos ic” pa a-
digm in social exposome esea ch o be e na iga e and comp ehend i s
inhe en complexi y.
Funding
This esea ch was pa ly unded by he Eu opean Union’s Ho izon
2020 esea ch and inno a ion p og am unde g an ag eemen No
874724 (Equal-Li e). Equal-Li e is pa o he Eu opean Human Expo-
some Ne wo k. Da a collec ion in FinnTwin12 has been suppo ed by he
Na ional Ins i u e on Alcohol Abuse and Alcoholism (g an s AA-12502,
AA-00145, and AA-09203 o Richa d J. Rose) and he Academy o
Finland (g an s 100499, 205585, 118555, 141054, 264146, 308248,
312073, 336823, and 352792 o Jaakko Kap io). Jaakko Kap io ac-
knowledges suppo by he Academy o Finland (g an s 265240,
263278).
Fig. 3. Volcano plo s o associa ions, ia gene alized linea eg ession, o g ow h ac o s wi h p- ac o (A), in e nalizing p oblems (B), dep ession (C), anxie y (D),
ex e nalizing p oblems (E), hype ac i i y (F), agg ession (G), ina en ion (H), and p osocial beha io (I) among males (indi idual wins n =1898). The x-axis
ep esen s eg ession coe icien s and he Y-axis ep esen s –log
10
(P- alue) o eg essions. The p e ix i means he in e cep , he p e ix s means he linea g ow h
ac o , and he p e ix q means quad a ic he g ow h ac o .
Z. Wang e al.
Jou nal o A ec i e Diso de s 358 (2024) 70–78
76
E hical s a emen
The e hics commi ee o he Depa men o Public Heal h o he
Uni e si y o Helsinki (Helsinki, Finland), he e hics commi ee o he
Helsinki Uni e si y Cen al Hospi al Dis ic (Helsinki, Finland), and he
Ins i u ional Re iew Boa d o Indiana Uni e si y (Blooming on, Indiana,
USA) app o ed he FinnTwin12 s udy p o ocol. All pa icipan s and
hei pa en s/legal gua dians ga e in o med w i en consen o pa ici-
pa e in he s udy. The au ho s asse ha all p ocedu es con ibu ing o
his wo k comply wi h he e hical s anda ds o he ele an na ional and
ins i u ional commi ees on human expe imen a ion and wi h he Hel-
sinki Decla a ion o 1975, as e ised in 2008.
CRediT au ho ship con ibu ion s a emen
Zhiyang Wang: Concep ualiza ion, Da a cu a ion, Fo mal analysis,
Me hodology, W i ing – o iginal d a . Gabin D oua d: Concep uali-
za ion, Me hodology, W i ing – e iew & edi ing. Alyce M. Whipp:
Concep ualiza ion, Da a cu a ion, Me hodology, W i ing – e iew &
edi ing. Ma ja Heinonen-Guzeje : Concep ualiza ion, Me hodology,
W i ing – e iew & edi ing. Gab iele Bol e: Concep ualiza ion, W i ing
– e iew & edi ing. Jaakko Kap io: Concep ualiza ion, Da a cu a ion,
Funding acquisi ion, Me hodology, P ojec adminis a ion, Supe ision,
W i ing – e iew & edi ing.
Decla a ion o compe ing in e es
The au ho s decla e ha hey ha e no known compe ing inancial
in e es s o pe sonal ela ionships ha could ha e appea ed o in luence
he wo k epo ed in his pape .
Da a a ailabili y
The FinnTwin12 da a is no publicly a ailable due o he es ic ions
o in o med consen . Howe e , he FinnTwin12 da a is a ailable
h ough he Ins i u e o Molecula Medicine Finland (FIMM) Da a Ac-
cess Commi ee (DAC) ([email p o ec ed]) o au ho ized e-
sea che s who ha e IRB/e hics app o al and an ins i u ionally app o ed
s udy plan. To ensu e he p o ec ion o p i acy and compliance wi h
na ional da a p o ec ion legisla ion, a da a use/ ans e ag eemen is
needed, he con en and speci ic clauses o which will depend on he
na u e o he eques ed da a.
Appendix A. Supplemen a y da a
Supplemen a y da a o his a icle can be ound online a h ps://doi.
o g/10.1016/j.jad.2024.04.096.
Re e ences
Aleg ía, M., NeMoye , A., Falg`
as Bagu´
e, I., Wang, Y., Al a ez, K., 2018. Social
de e minan s o men al heal h: whe e we a e and whe e we need o go. Cu .
Psychia y Rep. 20, 95. h ps://doi.o g/10.1007/s11920-018-0969-9.
Ande s´
en, H., Kankaan an a, H., Tuomis o, L.E., Pii il¨
a, P., So ij¨
a i, A.,
Langhamme , A., Backman, H., Lundb¨
ack, B., R¨
onma k, E., Leh im¨
aki, L.,
Ilma inen, P., 2021. Mul imo bidi y in Finnish and Swedish speaking Finns;
associa ion wi h daily habi s and socioeconomic s a us – No dic EpiLung c oss-
sec ional s udy. P e . Med. Rep. 22, 101338 h ps://doi.o g/10.1016/j.
pmed .2021.101338.
Fig. 4. Volcano plo s o associa ions, ia gene alized linea eg ession, o g ow h ac o s wi h p- ac o (A), in e nalizing p oblems (B), dep ession (C), anxie y (D),
ex e nalizing p oblems (E), hype ac i i y (F), agg ession (G), ina en ion (H), and p osocial beha io (I) among emales (indi idual wins n =2067). The x-axis
ep esen s eg ession coe icien s and he Y-axis ep esen s –log
10
(P- alue) o eg essions. The p e ix i means he in e cep , he p e ix s means he linea g ow h
ac o , and he p e ix q means quad a ic he g ow h ac o .
Z. Wang e al.
Jou nal o A ec i e Diso de s 358 (2024) 70–78
77
Ba , P.B., 2018. Ea ly neighbo hood condi ions and ajec o ies o dep essi e symp oms
ac oss adolescence and in o adul hood. Ad . Li e Cou se Res. 35, 57–68. h ps://doi.
o g/10.1016/j.alc .2018.01.005.
B azil, N., Cla k, W.A.V., 2017. Indi idual men al heal h, li e cou se e en s and dynamic
neighbou hood change du ing he ansi ion o adul hood. Heal h Place 45, 99–109.
h ps://doi.o g/10.1016/j.heal hplace.2017.03.007.
Buu, A., DiPiazza, C., Wang, J., Pu le , L.I., Fi zge ald, H.E., Zucke , R.A., 2009. Pa en ,
amily, and neighbo hood e ec s on he de elopmen o child subs ance use and
o he psychopa hology om p eschool o he s a o adul hood. J. S ud. Alcohol
D ugs 70, 489–498. h ps://doi.o g/10.15288/jsad.2009.70.489.
Choi, K.W., S ein, M.B., Nishimi, K.M., Ge, T., Coleman, J.R.I., Chen, C.-Y.,
Ra ana ha a ho n, A., Zheu lin, A.B., Dunn, E.C., B een, G., Koenen, K.C., Smolle , J.
W., 2020. An exposu e-wide and Mendelian andomiza ion app oach o iden i ying
modi iable ac o s o he p e en ion o dep ession. Am. J. Psychia y 177, 944–954.
h ps://doi.o g/10.1176/appi.ajp.2020.19111158.
COVID-19 Men al Diso de s Collabo a o s, 2021. Global p e alence and bu den o
dep essi e and anxie y diso de s in 204 coun ies and e i o ies in 2020 due o he
COVID-19 pandemic. Lance 398, 1700–1712. h ps://doi.o g/10.1016/S0140-6736
(21)02143-7.
Flou i, E., Midouhas, E., Ruddy, A., 2016. Socio–economic s a us and amily s uc u e
di e ences in ea ly ajec o ies o child adjus men : indi idual and neighbou hood
e ec s. Heal h Place 37, 8–15. h ps://doi.o g/10.1016/j.heal hplace.2015.11.005.
Glymou , M.M., A endano, M., Kawachi, I., 2014. Social Epidemiology, 2nd ed. Ox o d
Uni e si y P ess. h ps://doi.o g/10.1093/med/9780195377903.003.0002.
Gomm, S., Be naue , T., 2023. A e ac ual and pe cei ed en i onmen al condi ions
associa ed wi h a ia ion in men al heal h? En i on. Res. 223, 115398 h ps://doi.
o g/10.1016/j.en es.2023.115398.
Goodwin, R.D., Die ke , L.C., Wu, M., Galea, S., Ho en, C.W., Weinbe ge , A.H., 2022.
T ends in U.S. dep ession p e alence om 2015 o 2020: he widening ea men
gap. Am. J. P e . Med. 63, 726–733. h ps://doi.o g/10.1016/j.
amep e.2022.05.014.
Gudi-Minde mann, H., Whi e, M., Roczen, J., Riedel, N., D ege , S., Bol e, G., 2023.
In eg a ing he social en i onmen wi h an equi y pe spec i e in o he exposome
pa adigm: a new concep ual amewo k o he social exposome. En i on. Res. 233,
116485 h ps://doi.o g/10.1016/j.en es.2023.116485.
He nandez-Fe e , C., Wellenius, G.A., Tamayo, I., Basaga˜
na, X., Sunye , J., V ijheid, M.,
Gonzalez, J.R., 2019. Comp ehensi e s udy o he exposome and omic da a using
exposome Bioconduc o Packages. Bioin o ma ics 35. h ps://doi.o g/10.1093/
bioin o ma ics/b z526.
Hu, H., Liu, X., Zheng, Y., He, X., Ha , J., James, P., Laden, F., Chen, Y., Bian, J., 2023.
Me hodological challenges in spa ial and con ex ual exposome-heal h s udies. C i .
Re . En i on. Sci. Technol. 53, 827–846. h ps://doi.o g/10.1080/
10643389.2022.2093595.
Huang, Y., Lu, J., ˇ
Si ůˇ
cek, J., 2023. The associa ions be ween social en i onmen and
adolescen s’ psychosoma ic heal h: an ecological pe spec i e. F on . Psychol. 14,
1141206 h ps://doi.o g/10.3389/ psyg.2023.1141206.
Humph ey, J.L., Roo , E.D., 2017. Spa io- empo al neighbo hood impac s on
in e nalizing and ex e nalizing beha io s in U.S. elemen a y school child en: e ec
modi ica ion by child and amily socio-demog aphics. Soc. Sci. Med. 180, 52–61.
h ps://doi.o g/10.1016/j.socscimed.2017.03.014.
Huppe z, C., Ba els, M., de Geus, E.J.C., an Beijs e eld , C.E.M., Rose, R.J., Kap io, J.,
Sil en oinen, K., 2017. The e ec s o pa en al educa ion on exe cise beha io in
childhood and you h: a s udy in Du ch and Finnish wins. Scand. J. Med. Sci. Spo s
27, 1143–1156. h ps://doi.o g/10.1111/sms.12727.
Hyypp¨
a, M.T., M¨
aki, J., 2001. Why do Swedish-speaking Finns ha e longe ac i e li e?
An a ea o social capi al esea ch. Heal h P omo . In . 16, 55–64. h ps://doi.o g/
10.1093/heap o/16.1.55.
Ji aj, S., Mu ay, E.T., No man, P., Nicholas, O., 2020. The impac o li e cou se
exposu es o neighbou hood dep i a ion on heal h and well-being: a e iew o he
long- e m neighbou hood e ec s li e a u e. Eu . J. Pub. Heal h 30, 922–928.
h ps://doi.o g/10.1093/eu pub/ckz153.
Kajan ie, M., Mande backa, K., McCallum, A., No kola, I.-L., A man, M., Fo ssas, E.,
Ka onen, S., Ko einen, M., Leyland, A., Keskim¨
aki, I., 2006. How o Ca y ou
Regis e -Based Heal h Se ices Resea ch in Finland?: Compiling Complex S udy Da a
in he REDD P ojec . S akes, Helsinki.
Keil, A.P., Buckley, J.P., O’B ien, K.M., Fe guson, K.K., Zhao, S., Whi e, A.J., 2020.
A quan ile-based g-compu a ion app oach o add essing he e ec s o exposu e
mix u es. En i on. Heal h Pe spec . 128, 47004.
Keles, B., McC ae, N., G ealish, A., 2020. A sys ema ic e iew: he in luence o social
media on dep ession, anxie y and psychological dis ess in adolescen s. In . J.
Adolesc. You h 25, 79–93. h ps://doi.o g/10.1080/02673843.2019.1590851.
Kemppainen, L., Kemppainen, T., Fokkema, T., W ede, S., Kou onen, A., 2023.
Neighbou hood e hnic densi y, local language skills, and loneliness among olde
mig an s—a popula ion-based s udy on Russian speake s in Finland. In . J.
En i on. Res. Public Heal h. h ps://doi.o g/10.3390/ije ph20021117.
Koops, J.C., Lie b oe , A.C., Gau hie , A.H., 2021. Socio-economic di e ences in he
p e alence o single mo he hood in No h Ame ica and Eu ope. Eu . J. Popul. 37,
825–849. h ps://doi.o g/10.1007/s10680-021-09591-3.
K oks ad, S., Weiss, D.A., K oks ad, M.A., Rangul, V., K aløy, K., Ingul, J.M.,
Bje kese , O., Twenge, J., Sund, E.R., 2022. Di e gen decennial ends in men al
heal h acco ding o age e eal poo e men al heal h o young people: epea ed
c oss-sec ional popula ion-based su eys om he HUNT S udy, No way. BMJ Open
12, e057654. h ps://doi.o g/10.1136/bmjopen-2021-057654.
Liddell, B.J., Nicke son, A., Sa o , L., I ancic, L., B yan , R.A., 2016. The gene a ional
gap: men al diso de p e alence and disabili y amongs i s and second gene a ion
immig an s in Aus alia. J. Psychia . Res. 83, 103–111. h ps://doi.o g/10.1016/j.
jpsychi es.2016.08.011.
Lin, B.D., P ies, L.-K., Sa ac, H.S., an Os, J., Ru en, B.P.F., Luykx, J., Guloksuz, S., 2022.
Nongene ic ac o s associa ed wi h psycho ic expe iences among UK biobank
pa icipan s: exposome-wide analysis and Mendelian andomiza ion analysis. JAMA
Psychia y 79, 857–868. h ps://doi.o g/10.1001/jamapsychia y.2022.1655.
Mo issey, K., Kinde man, P., 2020. The impac o childhood socioeconomic s a us on
dep ession and anxie y in adul li e: es ing he accumula ion, c i ical pe iod and
social mobili y hypo heses. SSM - Popul. Heal. 11, 100576 h ps://doi.o g/10.1016/
j.ssmph.2020.100576.
Neu cou , L., Cas agn´
e, R., Mabile, L., Khala ba i-Sol ani, S., Delpie e, C., Kelly-
I ing, M., 2023. Assessing how social exposu es a e in eg a ed in exposome
esea ch: a scoping e iew. En i on. Heal h Pe spec . 130, 116001 h ps://doi.o g/
10.1289/EHP11015.
Ni, M.Y., Yao, X.I., Cheung, F., Wu, J.T., Schooling, C.M., Pang, H., Leung, G.M., 2020.
De e minan s o physical, men al and social well-being: a longi udinal en i onmen -
wide associa ion s udy. In . J. Epidemiol. 49, 380–389. h ps://doi.o g/10.1093/ije/
dyz238.
Ohanyan, H., Po engen, L., Huss, A., T aini, E., Beulens, J.W.J., Hoek, G., Lake eld, J.,
Ve meulen, R., 2022. Machine lea ning app oaches o cha ac e ize he obesogenic
u ban exposome. En i on. In . 158, 107015 h ps://doi.o g/10.1016/j.
en in .2021.107015.
Pe son Waye, K., L¨
o e, J., Le che , P., Dzhambo , A.M., Kla e, M., Sch eckenbe g, D.,
Belke, C., Leis , L., Ris o ska, G., Je am, S., Kanninen, K.M., Selande , J., A a , A.,
Lachmann, T., Cla k, C., Bo eldoo en, D., Whi e, K., Jul ez, J., Fo as e , M.,
Kap io, J., Bol e, G., Psyllidis, A., Gulli e , J., Boshuizen, H., Bozzon, A., Fels, J.,
Ho nikx, M., an den Hazel, P., Webe , M., B ambilla, M., B aa -Eggen, E., Van
Kamp, I., Vincens, N., 2023. Adop ing a child pe spec i e o exposome esea ch on
men al heal h and cogni i e de elopmen - concep ualisa ion and oppo uni ies.
En i on. Res. 117279 h ps://doi.o g/10.1016/j.en es.2023.117279.
P eache , K.J., Wichman, A.L., MacCallum, R.C., B iggs, N.E., 2008. La en G ow h Cu e
Modeling. SAGE Publica ions, Inc., Thousand Oaks h ps://doi.o g/10.4135/
9781412984737.
P ies, L.-K., Moo e, T.M., Visoki, E., So elo, I., Ba zilay, R., Guloksuz, S., 2022.
Es ima ing he associa ion be ween exposome and psychosis as well as gene al
psychopa hology: esul s om he ABCD s udy. Biol. Psychia y Glob. Open Sci. 2,
283–291. h ps://doi.o g/10.1016/j.bpsgos.2022.05.005.
Pulkkinen, L., 2017. Human de elopmen om middle childhood o middle adul hood:
g owing up o be middle-aged. In: Human De elopmen om Middle Childhood o
Middle Adul hood: G owing Up o be Middle-aged, 1s ed. Rou ledge, London.
h ps://doi.o g/10.4324/9781315732947.
Pulkkinen, L., Kap io, J., Rose, R.J., 1999. Pee s, eache s and pa en s as assesso s o he
beha iou al and emo ional p oblems o wins and hei adjus men : he
Mul idimensional Pee Nomina ion In en o y. Twin Res. Hum. Gene . 2, 274–285.
h ps://doi.o g/10.1375/ win.2.4.274.
Rose, R.J., Sal a o e, J.E., Aal onen, S., Ba , P.B., Bogl, L.H., Bye s, H.A., Heikkil¨
a, K.,
Ko honen, T., La ala, A., Pal iainen, T., Ranji , A., Whipp, A.M., Pulkkinen, L.,
Dick, D.M., Kap io, J., 2019. FinnTwin12 coho : an upda ed e iew. Twin Res.
Hum. Gene . 22, 302–311. h ps://doi.o g/10.1017/ hg.2019.83.
Rosseel, Y., 2012. La aan: an R package o s uc u al equa ion modeling. J. S a . So w.
48, 1–36. h ps://doi.o g/10.18637/jss. 048.i02.
Schinasi, L.H., Benma hnia, T., De Roos, A.J., 2018. Modi ica ion o he associa ion
be ween high ambien empe a u e and heal h by u ban mic oclima e indica o s: a
sys ema ic e iew and me a-analysis. En i on. Res. 161, 168–180. h ps://doi.o g/
10.1016/j.en es.2017.11.004.
Solan aus, T., Leinonen, J., Punam¨
aki, R.-L., 2004. Child en’s men al heal h in imes o
economic ecession: eplica ion and ex ension o he amily economic s ess model in
Finland. De . Psychol. h ps://doi.o g/10.1037/0012-1649.40.3.412.
S inson, S., 1985. Sex di e ences in en i onmen al sensi i i y du ing g ow h and
de elopmen . Am. J. Phys. An h opol. 28, 123–147. h ps://doi.o g/10.1002/
ajpa.1330280507.
T aini, E., Huss, A., Po engen, L., Rookus, M., Ve schu en, W.M.M., Ve meulen, R.C.H.,
Bella ia, A., 2022. A mul ipollu an app oach o es ima ing causal e ec s o ai
pollu ion mix u es on o e all mo ali y in a la ge, p ospec i e coho . Epidemiology
33.
an de Weije , M.P., Baselmans, B.M.L., Ho enga, J.-J., Dolan, C.V., Willemsen, G.,
Ba els, M., 2022. Expanding he en i onmen al scope: an en i onmen -wide
associa ion s udy o men al well-being. J. Expo. Sci. En i on. Epidemiol. 32,
195–204. h ps://doi.o g/10.1038/s41370-021-00346-0.
Visse , K., Bol , G., Finkenaue , C., Jonke , M., Weinbe g, D., S e ens, G.W.J.M., 2021.
Neighbou hood dep i a ion e ec s on young people’s men al heal h and well-being:
a sys ema ic e iew o he li e a u e. Soc. Sci. Med. 270, 113542 h ps://doi.o g/
10.1016/j.socscimed.2020.113542.
Wang, Z., Zelle s, S., Whipp, A.M., Heinonen-Guzeje , M., Fo as e , M., Júl ez, J., an
Kamp, I., Kap io, J., 2023. The e ec o en i onmen on dep essi e symp oms in la e
adolescence and ea ly adul hood: an exposome-wide associa ion s udy and win
modeling. Na . Men . Heal. h ps://doi.o g/10.1038/s44220-023-00124-x.
Weck o h, M., Ala-Man ila, S., Ballas, D., Ziogas, T., Ikonen, J., 2022. U bani y,
neighbou hood cha ac e is ics and pe cei ed quali y o li e (QoL): analysis o
indi idual and con ex ual de e minan s o pe cei ed QoL in 3300 pos al code a eas
in Finland. Soc. Indic. Res. 164, 139–164. h ps://doi.o g/10.1007/s11205-021-
02835-z.
Whea on, B., Cla ke, P., 2003. Space mee s ime: in eg a ing empo al and con ex ual
in luences on men al heal h in ea ly adul hood. Am. Sociol. Re . 68, 680–706.
h ps://doi.o g/10.2307/1519758.
Z. Wang e al.
Jou nal o A ec i e Diso de s 358 (2024) 70–78
78
Whipp, A.M., Ko honen, T., Rae uo i, A., Heikkil¨
a, K., Pulkkinen, L., Rose, R.J.,
Kap io, J., Vuoksimaa, E., 2019. Ea ly adolescen agg ession p edic s an isocial
pe sonali y diso de in young adul s: a popula ion-based s udy. Eu . Child Adolesc.
Psychia y 28, 341–350. h ps://doi.o g/10.1007/s00787-018-1198-9.
Whi e, J., G eene, G., Fa ewell, D., Duns an, F., Rodge s, S., Lyons, R.A., Humph eys, I.,
John, A., Webs e , C., Phillips, C.J., Fone, D., 2017. Imp o ing men al heal h
h ough he egene a ion o dep i ed neighbo hoods: a na u al expe imen . Am. J.
Epidemiol. 186, 473–480. h ps://doi.o g/10.1093/aje/kwx086.
Xia, Y., Yang, Y., 2019. RMSEA, CFI, and TLI in s uc u al equa ion modeling wi h
o de ed ca ego ical da a: he s o y hey ell depends on he es ima ion me hods.
Beha . Res. Me hods 51, 409–428. h ps://doi.o g/10.3758/s13428-018-1055-2.
Xue, Y., Le en hal, T., B ooks-Gunn, J., Ea ls, F.J., 2005. Neighbo hood esidence and
men al heal h p oblems o 5- o 11-yea -olds. A ch. Gen. Psychia y 62, 554–563.
h ps://doi.o g/10.1001/a chpsyc.62.5.554.
Z. Wang e al.
