Alonso-Pe ez, En ique; Ding, Xuejie; Rich e , Da id; Gelle , Paul; O’Sulli an, Julie
Lo aine
A icle — Published Ve sion
In e sec ional sleep dispa i ies: associa ion be ween mul iple social
in e sec ions, pe cei ed neighbo hood dep i a ion, and sleep
dis u bance in Eu ope
Eu opean Jou nal o Public Heal h
P o ided in Coope a ion wi h:
WZB Be lin Social Science Cen e
Sugges ed Ci a ion: Alonso-Pe ez, En ique; Ding, Xuejie; Rich e , Da id; Gelle , Paul; O’Sulli an, Julie
Lo aine (2025) : In e sec ional sleep dispa i ies: associa ion be ween mul iple social in e sec ions,
pe cei ed neighbo hood dep i a ion, and sleep dis u bance in Eu ope, Eu opean Jou nal o Public
Heal h, ISSN 1464-360X, Ox o d Uni e si y P ess, Ox o d, Vol. 35, Iss. 5, pp. 821-827,
h ps://doi.o g/10.1093/eu pub/cka 130
This Ve sion is a ailable a :
h ps://hdl.handle.ne /10419/329649.2
S anda d-Nu zungsbedingungen:
Die Dokumen e au EconS o dü en zu eigenen wissenscha lichen
Zwecken und zum P i a geb auch gespeiche und kopie we den.
Sie dü en die Dokumen e nich ü ö en liche ode komme zielle
Zwecke e iel äl igen, ö en lich auss ellen, ö en lich zugänglich
machen, e eiben ode ande wei ig nu zen.
So e n die Ve asse die Dokumen e un e Open-Con en -Lizenzen
(insbesonde e CC-Lizenzen) zu Ve ügung ges ell haben soll en,
gel en abweichend on diesen Nu zungsbedingungen die in de do
genann en Lizenz gewäh en Nu zungs ech e.
Te ms o use:
Documen s in EconS o may be sa ed and copied o you pe sonal
and schola ly pu poses.
You a e no o copy documen s o public o comme cial pu poses, o
exhibi he documen s publicly, o make hem publicly a ailable on he
in e ne , o o dis ibu e o o he wise use he documen s in public.
I he documen s ha e been made a ailable unde an Open Con en
Licence (especially C ea i e Commons Licences), you may exe cise
u he usage igh s as speci ied in he indica ed licence.
h ps://c ea i ecommons.o g/licenses/by-nc/4.0/
Eu opean Jou nal o Public Heal h, Vol. 35, No. 5, 821–827
© The Au ho (s) 2025. Published by Ox o d Uni e si y P ess on behal o he Eu opean Public Heal h Associa ion.
This is an Open Access a icle dis ibu ed unde he e ms o he C ea i e Commons A ibu ion-NonComme cial License (h ps://c ea i ecommons.
o g/licenses/by-nc/4.0/), which pe mi s non-comme cial e-use, dis ibu ion, and ep oduc ion in any medium, p o ided he o iginal wo k is p ope ly
ci ed. Fo comme cial e-use, please con ac [email p o ec ed] o ep in s and ansla ion igh s o ep in s. All o he pe missions can be ob ained
h ough ou Righ sLink se ice ia he Pe missions link on he a icle page on ou si e— o u he in o ma ion please con ac jou nals.pe mis-
[email p o ec ed].
h ps://doi.o g/10.1093/eu pub/cka 130 Ad ance Access published on 18 Augus 2025
...............................................................................................................
In e sec ional sleep dispa i ies: associa ion be ween
mul iple social in e sec ions, pe cei ed neighbo hood
dep i a ion, and sleep dis u bance in Eu ope
En ique Alonso-Pe ez
1,2,
�, Xuejie Ding
2,3,4,5
, Da id Rich e
6,7
, Paul Gelle
1,2,8
,
Julie Lo aine O’Sulli an
1,2,8
1
Ins i u e o Medical Sociology and Rehabili a ion Science, Cha i �
e—Uni e si €
a smedizin Be lin, Co po a e Membe o
F eie Uni e si €
a Be lin, Humbold -Uni e si €
a zu Be lin, Be lin, Ge many
2
Eins ein Cen e Popula ion Di e si y (ECPD), Be lin, Ge many
3
Ox o d in Be lin, Be lin, Ge many
4
WZB Be lin Social Science Cen e , Be lin, Ge many
5
Depa men o Social Wo k and Social Adminis a ion, Uni e si y o Hong Kong, Hong Kong, China
6
Depa men o Educa ion and Psychology, F eie Uni e si €
a Be lin, Be lin, Ge many
7
SHARE Be lin Ins i u e, Be lin, Ge many
8
Ge man Cen e o Men al Heal h (DZPG), Pa ne si e Be lin-Po sdam, Be lin, Ge many
�Co esponding au ho . Cha i �
e—Uni e si €
a smedizin Be lin (Campus Mi e), Ins i u e o Medical Sociology and
Rehabili a ion Science, Vi chowweg 22/23, 10117 Be lin, Ge many. E-mail: [email p o ec ed].
Abs ac
The p e alence o sleep dis u bance, ela ed wi h social s a us and p i ilege, is une enly dis ibu ed wi hin
socie ies. Indi idual social de e minan s ha a e embedded wi hin b oade neighbo hood con ex s in e sec
and join ly shape sleep dispa i ies. This s udy inco po a es a quan i a i e in e sec ional amewo k o be e
unde s and he s uc u al inequali ies in sleep dis u bance o olde adul s, ocusing on he social–ecological
model o sleep and how indi idual and social con ex ac o s in e ac . Ou sample consis ed o 17 035 indi iduals
aged 50þ om wa es 4 and 5 o he Su ey o Heal h, Aging and Re i emen in Eu ope (SHARE). We c ea ed 72
unique in e sec ional s a a by in e ac ing indi idual axes o social inequali y (sex/gende , amily ca egi ing,
educa ion, occupa ion) wi h pe cei ed neighbo hood dep i a ion. To in es iga e he a ia ions in sleep dis u b-
ance ac oss in e sec ional s a a, we employed in e sec ional Mul ile el Analysis o Indi idual He e ogenei y and
Disc imina o y Accu acy (MAIHDA). In e sec ional s a a explained a ai magni ude o he a iance in sleep
dis u bance (6.3%). The mos disad an aged g oups, pa icula ly women wi h low educa ion, low-skill occupa-
ions who we e ca egi e s in pe cei ed highly-dep i ed neighbo hoods, exhibi ed he la ges numbe o sleep
dis u bance. Sex/gende and pe cei ed neighbo hood dep i a ion we e he main p edic o s o such di e ences.
While some mul iplica i e e ec s we e ound, addi i e e ec s p edomina ed. Gi en he impo ance o sleep o
heal h, coupled wi h inc easing social inequali ies, ou indings sugges ha in e sec ionali y is a aluable ame-
wo k o mapping and add essing sleep dispa i ies. Tailo ed in e en ions should go beyond indi idual ac o s o
include communi y-le el measu es, a ge ing socially ulne able g oups, especially women expe iencing neigh-
bo hood dep i a ion.
...............................................................................................................
In oduc ion
Sleep dis u bance, including dis up ions in quali y, iming, and
du a ion, is associa ed wi h psychia ic symp oma ology, ca dio-
ascula disease, me abolic synd ome, and inc eased mo ali y [1–4].
Al hough dis u bed sleep is inc easingly p e alen among olde
Eu opean adul s [5], i emains unde - esea ched and unde -
p io i ized in public heal h in e en ions [6]. Mos sleep esea ch
ocuses on biological de e minan s, ye social and en i onmen al
ac o s a e equally c i ical [7]. Social epidemiology o sleep has
un a eled how social de e minan s like sex/gende , educa ion, o
occupa ion p edic sleep dis u bance [8], while he social–ecological
model o sleep heal h ecognizes how b oade social–en i onmen al
se ings like neighbo hood dep i a ion in e ac o a ec sleep dis-
u bance [9].
The social–ecological model o sleep heal h p o ides a comp ehen-
si e app oach o unde s anding how indi idual, social, and socie al
ac o s collec i ely shape sleep [9]. Indi idual ac o s (e.g. SES,
beha io s) exis wi hin social con ex s (e.g. amily, neighbo hood),
and hese wi hin b oade socie al con ex s (e.g. disc imina ion, econ-
omy) (Supplemen a y Fig. S1). These in e connec ed le els in e ac
wi h en i onmen al ac o s (e.g. pe cei ed neighbo hood sa e y)
in luencing indi idual beha io s and ice e sa, which unde sco es
he impo ance o examining sleep social de e minan s beyond he
indi idual le el [1]. The social and physical en i onmen , and pa -
icula ly neighbo hoods, gene a e and pe pe ua e sleep dispa i ies
h ough social (e.g. no ms), psychological (e.g. pe cei ed sa e y, dis-
c imina ion), and physical ac o s (e.g. pollu ion, access o se ices).
In e sec ionali y heo y posi s how s uc u al inequali ies and sys ems
o powe /opp ession uniquely a ec indi iduals a he in e sec ions o
hei social cha ac e is ics [10]. In his ega d, in e sec ionali y heo y
sugges s ha s uc u al o ces shape neighbo hood pe cep ions,
u he a ec ing ulne abili y o sleep isks [11, 12]. Likewise,
esea ch e ealed ha pe cei ed neighbo hood cha ac e is ics o en
ha e s onge associa ions wi h sleep han objec i e measu es [13,
14]. While adi ional in e en ions ocus on indi idual ac o s like
medica ion, he apy, o sleep hygiene, add essing con ex ual and
s uc u al de e minan s is c ucial o de eloping public heal h s a -
egies ha p omo e sleep equi y [1].
Wi hin and ac oss neighbo hoods, indi iduals a he in e sec ion
o mul iple disad an aged social posi ions a e mo e likely o expe i-
ence sleep dis u bance due o con inuous exposu e o disc imin-
a ion, inancial insecu i y, o unsa e social en i onmen s [15].
Following he PROGRESS-Plus amewo k [16, 17], we iden i ied
pa icula socio-demog aphic cha ac e is ics ha s a i y sleep
heal h oppo uni ies and ou comes: sex/gende , amily ca egi ing,
educa ion, occupa ion, and place o esidence. No ably, women ex-
pe ience mo e sleep dis u bance han men, and besides physiologic-
al di e ences, his is due o adi ional gende no ms ha p io i ize
ca egi ing and household du ies o e sleep [18, 19]. Likewise, ac o s
such as amily du ies can impac sleep iming and quali y, wi h mo e
sleep dis u bance epo ed by amily ca egi e s [20]. Fu he mo e,
lowe SES (i.e. educa ion, occupa ion) is linked o mo e sleep dis-
u bance due o less sleep heal h li e acy, limi ed access o es ul
wo king condi ions, and p essu es o low-skill jobs like nigh shi s
[2, 21]. Beyond indi idual cha ac e is ics, pe cei ed neighbo hood
dep i a ion is associa ed wi h mo e sleep dis u bance, ein o cing
geog aphical inequali ies in sleep quali y [11, 13]. Pe cei ed neigh-
bo hood con ex s a ec sleep h ough s ess, physiological
esponses, and social engagemen [11, 14]. Acco dingly, u banici y
is associa ed wi h ac o s (noise, dep i a ion) ha may esul in sleep
dis u bance [13]. O e all, he in e connec edness o social de e mi-
nan s o sleep highligh s he need o an in e sec ional app oach [22,
23]. Ye , mos s udies ha e examined hese ac o s sepa a ely, limi -
ing he unde s anding o how in e sec ional sleep dispa i ies eme ge.
While schola s ad oca e he use o in e sec ionali y o ad ance he
unde s anding o sleep dispa i ies ac oss social subg oups [22, 23],
quan i a i e sleep esea ch a ely inco po a ed in e sec ional
app oaches. In e sec ional mul ile el analysis o indi idual he e o-
genei y and disc imina o y accu acy (MAIHDA) is a no el,
in e sec ionali y-based quan i a i e me hod o s udying heal h dis-
pa i ies [24]. MAIHDA models heal h ou comes by nes ing indi id-
uals wi hin social s a a, de ined by unique in e sec ions o socio-
demog aphic cha ac e is ics. Compa ed o adi ional mul ile el
models wi h in e ac ions, MAIHDA o e s ad an ages in scalabili y,
pa simony, and handling small subg oup samples [25]. Unlike o he
models, MAIHDA ames in e sec ing social de e minan s (e.g. gen-
de , SES) as indica o s o sys emic opp ession (e.g. sexism, classism),
add essing he s uc u al oo s o sleep dispa i ies [22, 23]. Howe e ,
sleep ou comes emain unexplo ed using his amewo k.
The cu en s udy add esses he limi ed e idence on in e sec ional
sleep dispa i ies wi hin neighbo hood-le el con ex s by in eg a ing
an in e sec ional lens in o he social–ecological model o sleep, o
s udy social de e minan s o sleep dis u bance. We aim o map
dispa i ies in sleep dis u bance ac oss in e sec ional s a a, assess
he ela i e impac o pe cei ed neighbo hood dep i a ion on hese
dispa i ies, and iden i y s a a expe iencing in e sec ional in e -
ac ion e ec s.
Me hods
Da a and sample
We used da a om wa es 4 and 5 o he Su ey o Heal h, Aging and
Re i emen in Eu ope (SHARE), he la ges Eu opean panel s udy
wi h in o ma ion on demog aphic, socioeconomic, and heal h ou -
comes o people aged 50þ[26]. SHARE da a a e collec ed wi h
compu e -assis ed pe sonal in e iews (CAPIs), and he su ey has
been ex ensi ely desc ibed elsewhe e [26]. Wa e 4 (2011) is he
la es wa e including mul iple sleep a iables, while wa e 5 (2013)
collec ed in o ma ion on pe cei ed neighbo hood con ex . O he
38 296 esponden s aged 50þpa icipa ing in bo h wa es, we
excluded: (a) 707 esponden s (1.85%) wi h missing amily ca egi -
ing in o ma ion; (b) 486 esponden s (1.27%) wi h missing
educa ion in o ma ion; (c) 5308 esponden s (13.87%) wi h missing
occupa ion in o ma ion; (d) 13 296 esponden s (34.74%) wi h miss-
ing pe cei ed neighbo hood con ex in o ma ion; (e) 1075 espond-
en s (2.81%) who changed place o esidence be ween wa e 4 and
wa e 5; and ( ) 389 esponden s (1.02%) missing a sleep a iable.
The inal sample consis ed o N¼17 035 (Supplemen a y Table S1).
Measu es
Ou come a iable
Ou ou come a iable was a sleep dis u bance index (SDI), a p e i-
ously used composi e sco e based on wo sel - epo ed sleep i ems in
wa e 4 [4]. The i ems cap u ed sleep quali y “ o he pas six mon hs
a leas ” h ough he ques ions “Ha e you been bo he ed by sleep
p oblems?” and “Ha e you had ouble sleeping o a change in sleep
pa e n?”. Bo h we e coded as Yes/No. Since in e nal consis ency as
a measu e o eliabili y was accep able (C onbach’s α¼.76)
(Supplemen a y Table S2), we calcula ed he index by adding he
o al numbe o sleep dis u bance (SDI ange 0–2).
Main exposu e
We de ined in e sec ional s a a wi h indi idual-le el and
neighbo hood-le el ac o s associa ed o sleep dis u bance: sex/gen-
de (G), amily ca egi ing (F), educa ion (E), occupa ion (O), and
pe cei ed neighbo hood dep i a ion (N). The a iable selec ion was
in o med by he PROGRESS-Plus amewo k [17]. The unique com-
bina ions o all possible ca ego ies esul ed in 72 in e sec ional s a a
([2G] ×[2F] ×[3E] ×[2O] ×[3N] ¼72) [24]. We adop ed he ap-
p oach p oposed by E ans27, whe e con ex ual-le el a iables a e
in e ac ed wi h indi idual-le el a iables o c ea e in e sec ional
s a a. Hence, we allowed he e ec o pe cei ed neighbo hood de-
p i a ion o be unique o each combina ion o indi idual de e mi-
nan s, modeling he con ex ual social p ocess o in e sec ional sleep
dispa i ies. Mo e han 73% o he 72 s a a consis ed o a leas 30
obse a ions, indica ing a su icien sample size (Supplemen a y
Table S3) [25].
Sex/gende was coded as women o men, e lec ing he bina y
op ions p o ided o sex in SHARE. Acknowledging he limi a ions
o a bina y ca ego iza ion, we op ed o he e m sex/gende o ec-
ognize he con la ion o sex and gende in a single su ey i em.
Family ca egi ing was coded as No/Yes a e he ques ion “Du ing
he las 12 mon hs, is he e someone li ing in you household whom
you ha e helped egula ly wi h pe sonal ca e?”. Educa ion was
assessed ollowing he ISCED-1997, ca ego izing i in o high (5–6),
medium (3–4), o low (0–2). Occupa ion was ob ained om he
p esen o la es -held wo k posi ions, and coded in o wo majo
g oups acco ding o ISCO-88: high-skill (HS) o low-skill (LS)
occupa ions.
Pe cei ed neighbo hood dep i a ion was de i ed om ques ions
on he local a ea (e e ywhe e wi hin a 20-min walk o a kilome e
om home), whe e pa icipan s we e asked on hei ag eemen wi h
he ollowing: “ andalism o c ime is no a big p oblem in his a ea”,
“ his a ea is kep e y clean”, “I eally eel pa o his a ea”, and “I I
we e in ouble, he e a e people in his a ea who would help me”.
Answe s we e dicho omized as ag ee (s ongly ag ee o ag ee) o
disag ee (disag ee o s ongly disag ee), e e se coded and agg e-
ga ed in a single a iable ( alues 0–4, wi h highe sco es indica ing
mo e dep i a ion). Ou inal a iable was coded as low dep i a ion
(0–1), medium dep i a ion (2–3), o high dep i a ion (4), aligned
wi h p e ious s udies [27]. Since neighbo hood dep i a ion was only
a ailable in wa e 5, con a y o all o he a iables measu ed in wa e
4, we only included pa icipan s who did no change place o esi-
dence be ween wa e 4 and wa e 5.
S a is ical analysis
Se e al s udies in eg a ed social–con ex ual aspec s o in e sec ion-
ali y in ou key ways: (i) in e ac ing con ex ual-le el a iables wi h
822 Alonso-Pe ez e al.
axes o social posi ion [28]; (ii) eco-in e sec ional mul ile el (EIM)
modeling wi h a eas o esidence nes ed wi hin in e sec ional s a a
[29]; (iii) mul ile el models wi h indi iduals nes ed wi hin a ea o
esidence [30]; and (i ) c oss-classi ying in e sec ional social s a a
wi h con ex ual a iables [31]. App oaches (ii), (iii), and (i ) use
objec i e con ex ual measu es; howe e , since we ocus on pe cei ed
neighbo hood dep i a ion and i s impo ance o sleep dispa i ies
[14], we adop ed app oach (i).
We used he in e sec ional MAIHDA amewo k, which is based
on mul ile el models whe e indi iduals a e classi ied wi hin in e -
sec ional s a a: indi iduals we e placed a le el 1, nes ed wi hin
in e sec ional s a a a le el 2 [24]. Recen publica ions ha e
desc ibed he MAIHDA me hod ho oughly, while p o iding an
o e iew o i s ad an ages compa ed o adi ional mul ile el mod-
els wi h in e ac ions [24, 32]. We applied es ic ed maximum like-
lihood (REML) es ima ion o i linea mul ile el models.
Fi s , we i ed an unadjus ed null model (Model 1) o map sleep
heal h dispa i ies ac oss in e sec ional s a a. This model allowed us
o calcula e he a iance pa i ion coe icien (VPC)—equi alen o
he in aclass co ela ion coe icien (ICC)—a measu e o disc im-
ina o y accu acy e lec ing he be ween-s a a a iance o sleep dis-
u bance [32]. MAIHDA li e a u e sugges s he ollowing VPC
classi ica ion o disc imina o y accu acy: nonexis en (0–1), poo
(>1 o ≤5), ai (>5 o ≤10), good (>10 o ≤20), e y good (>20 o
≤30), and excellen (>30) [32]. Second, we i ed pa ially adjus ed
models (Models 2a–2e) by including one s a a-de ining a iable as
ixed e ec in each sequen ial model a a ime. In addi ion o he
VPC, we calcula ed he p opo ional change in a iance (PCV) in
each model, which indica es he p opo ion o be ween-s a a a i-
ance explained by he added main e ec s [32]. E e y PCV e lec s
he ex en in which each s a a-de ining a iable (i.e. in e sec ional
dimensions) con ibu e o he a iance o sleep dis u bance. Finally,
we i ed a ully adjus ed model (Model 3), which displays he con-
ibu ion o each s a a-de ining a iable o he join e ec on a e -
age. The VPC illus a es he emaining be ween-s a a a iance a e
adding all main e ec s. The PCV e lec s he a iance explained by
main e ec s only (addi i e e ec s), hence 1-PCV ep esen s he un-
explained a iance due o in e sec ional in e ac ions (mul iplica i e
e ec s). We included coun y dummies o con ol o c oss-coun y
a ia ion, al hough p io s udies wi h he same da a showed ha
coun y di e ences explained li le a iance in sleep dis u bance
(Supplemen a y Table S5 and Fig. S2) [33].
Based on Model 3, we calcula ed he p edic ed sleep dis u bance
and s a a-le el esiduals o each s a um, he la e indica ing po-
en ial in e ac ion e ec s o each o he in e sec ional s a a (highe
o lowe sleep dis u bance han expec ed om he main e ec s only)
[32]. Finally, gi en he associa ion o u banici y and sleep dis u b-
ance o g oups aged 66þ[34], we pe o med a sensi i i y analysis
by con olling o u banici y ( u al/u ban li ing a ea) in Model 4.
All analyses we e conduc ed wi h S a a/BE
®
18.0, whe e s a is ical
signi icance was assessed by 95% con idence in e als no including
ze o.
Resul s
The a e age age o he sample was 64.4 yea s old, wi h jus o e hal
o pa icipan s being emale (Table 1). A mino i y o he sample
unde ook egula amily ca egi ing on adul s (8.0%), whe eas
mos esponden s had medium o low educa ion. Almos wo-
hi ds o pa icipan s had a high-skill occupa ion. Mos indi iduals
pe cei ed medium o low neighbo hood dep i a ion, whe eas 20.5%
pe cei ed high neighbo hood dep i a ion. Abou wo- hi ds o
esponden s li ed in u ban a eas. One- hi d o he sample epo ed
ha ing ouble sleeping, while 22.4% epo ed ecen sleep com-
plain s. A la ge sha e (37.0%) o s udy pa icipan s epo ed a leas
one ype o sleep dis u bance occu ing ecen ly.
Highly ulne able in e sec ional s a a, such as women wi h
ca egi ing du ies and low SES, had he highes sleep dis u bance
on a e age (Fig. 1). Pa icula ly, indi iduals wi h he same social
cha ac e is ics exhibi ed be ween-neighbo hood sleep dispa i ies,
wi h high-dep i ed neighbo hoods en ailing highe sleep dis u b-
ance. The s a um comp ising women unde aking amily ca e-
gi ing, wi h medium educa ion, low-skill occupa ion, and li ing
in high-dep i ed neighbo hoods had he highes sleep dis u b-
ance (SDI ¼1.0). This implies a h ee- old di e ence compa ed
o he s a um wi h he leas sleep dis u bance (SDI ¼0.3), com-
p ised by men no unde aking amily ca egi ing, wi h high edu-
ca ion, high-skill occupa ion, and li ing in low-dep i ed
neighbo hoods.
The VPC o he null model (Model 1) indica ed ha 6.3% o he
a iance in sleep dis u bance was a ibu able o he in e sec ional
s a a (Table 2). This implies a good le el o disc imina o y accu acy
[32]. In he pa ially adjus ed models (Models 2a–2e), he PCVs
e ealed ha sex/gende (77.2%) and pe cei ed neighbo hood de-
p i a ion (40.3%) we e he la ges con ibu o s o di e ences in
sleep dis u bance ac oss s a a. In con as , amily ca egi ing
(11.4%), educa ion (6.1%), and occupa ion (1.8%) had a lowe con-
ibu ion. The ully adjus ed model (Model 3) showed ha , on a e -
age, women expe ienced mo e sleep dis u bance han men, as did
people unde aking amily ca egi ing compa ed o hose who did
no . Indi iduals wi h lowe educa ion epo ed highe sleep dis u b-
ance han hei medium and high educa ion coun e pa s, while
hose in low-skill occupa ions had mo e sleep dis u bance han hei
high-skill coun e pa s. High pe cei ed neighbo hood dep i a ion
was signi ican ly associa ed wi h mo e sleep dispa i ies, compa ed
o medium and low pe cei ed dep i a ion.
The VPC in Model 3 indica ed ha a e adjus ing o main e ec s,
only 0.6% o he be ween-s a a a iance emained. A PCV o 90.9%
e ealed ha he majo i y o in e sec ional a iance in sleep dis u b-
ance was explained by addi i e e ec s, whe eas he emaining a i-
ance (9.1%) was due o mul iplica i e e ec s (i.e. in e sec ional
in e ac ions). Likewise, he esidual analysis e ealed ha only se en
s a a exhibi ed signi ican mul iplica i e e ec s (Fig. 2 and
Supplemen a y Table S4). Fou o hem had highe sleep dis u bance
han expec ed om he addi i e e ec s only—CIs abo e 0 indica ing
Table 1. Desc ip i e s a is ics o he s udy sample
a
Va iable N%
To al 17 035
Age (mean ± SD) 64.4 (9.3)
Sex/gende
Male 7859 46.1
Female 9176 53.9
Family ca egi ing
No 15 670 92.0
Yes 1365 8.0
Educa ion
High 4119 24.2
Medium 6933 40.7
Low 5983 35.1
Occupa ion
High-skill occupa ion 10 580 62.1
Low-skill occupa ion 6455 37.9
Pe cei ed neighbo hood dep i a ion
Low dep i a ion 4760 27.9
Medium dep i a ion 8791 51.6
High dep i a ion 3484 20.5
Li ing a ea
U ban 10 791 63.7
Ru al 6147 36.3
Sleep ou comes
Sleep complain s pas 6 mon hs 3816 22.4
T ouble sleeping/change in pa e n 5860 34.4
Sleep dis u bance index (mean ± SD) 0.6 (0.8)
a: The sleep dis u bance index (SDI) akes alues be ween 0 and 2;
SD: s anda d de ia ion.
In e sec ional sleep dispa i ies 823
a haza dous e ec , whe eas h ee s a a had lowe sleep dis u bance
han expec ed—CIs below 0 indica ing p o ec i e e ec . Acco ding o
hese esiduals, he combina ion o high SES (high educa ion and
high-skill occupa ion) wi h mid/high pe cei ed neighbo hood dep i -
a ion ampli ied he isk o sleep dis u bance h ough in e sec ional
in e ac ions, as hese s a a showed mul iplica i e haza dous e ec s.
Con e sely, in e sec ional p o ec i e e ec s o sleep dis u bance
we e e iden o s a a wi h mid/low neighbo hood dep i a ion, com-
bined wi h a mixed pa e n o highe and lowe SES. Finally, he
sensi i i y analysis e ealed ha u banici y was no signi ican ly
associa ed wi h sleep dis u bance, wi h addi i e e ec s explaining
a la ge pa o he be ween-s a a a iance (Supplemen a y Table S5).
Discussion
In his s udy, we explo ed inequali ies in sleep dis u bance ac oss
in e sec ional s a a, ocusing on di e ences in pe cei ed neighbo -
hood dep i a ion. Using a Eu opean sample o olde adul s and
in e sec ional MAIHDA, we ound signi ican di e ences in sleep
dis u bance ac oss s a a, p ima ily explained by addi i e a he han
Figu e 1. P edic ed sleep dis u bance o each in e sec ional social s a um, wi h poin es ima es and 95% con idence in e als ob ained
om MAIHDA Model 3 (N¼17 035). The g ey colou -scheme indica es pe cei ed neighbo hood con ex . Highe SDI means mo e sleep
dis u bance. Abb e ia ions: HS ¼high-skill occupa ion; LS ¼lLow-skill occupa ion; Hi Ed ¼high educa ion; Me Ed ¼medium educa ion.
Table 2. In e sec ional MAIHDA models on he sleep dis u bance index (n¼17 035)
a
Coe icien (95% CI) Model 1 Model 2a Model 2b Model 2c Model 2d Model 2e Model 3
Cons an 0.6 (0.5, 0.7) 0.4 (0.4, 0.5) 0.5 (0.5, 0.6) 0.5 (0.4, 0.6) 0.6 (0.5, 0.7) 0.4 (0.4, 0.6) 0.3 (0.2, 0.3)
Sex/gende
Male Re . Re .
Female 0.3 (0.3, 0.4) 0.3 (0.3, 0.3)
Family ca egi ing
No amily ca egi ing Re . Re .
Family ca egi ing 0.1 (0, 0.2) 0.1 (0.1, 0.2)
Educa ion
High educa ion Re . Re .
Mid educa ion 0.1 (−0.1, 0.2) 0.1 (0.0, 0.1)
Low educa ion 0.2 (0.0, 0.3) 0.2 (0.1, 0.3)
Occupa ion
High-skill Re . Re .
Low-skill 0.0 (−0.1, 0.1) 0.1 (0.0, 0.1)
Pe cei ed neighbo hood
Low dep i a ion Re . Re .
Medium dep i a ion 0.1 (−0.1, 0.2) 0.1 (0.0, 0.1)
High dep i a ion 0.3 (0.0, 0.3) 0.2 (0.1, 0.3)
Random e ec s
Be ween-s a a a iance (95% CI) 0.04 (0.03, 0.06) 0.01 (0.01, 0.02) 0.04 (0.02, 0.06) 0.04 (0.02, 0.06) 0.04 (0.03, 0.06) 0.02 (0.01, 0.04) 0.01 (0.00, 0.01)
VPC (%) 6.3% 1.5% 5.6% 5.9% 6.3% 2.6% 0.6%
PCV (%) – 77.2% 11.4% 6.1% 1.8% 40.3% 90.9%
a: Model 3 con ols o coun y dummies; CI ¼con idence in e al; VPC ¼ a iance pa i ion coe icien ; PCV ¼p opo ional change in
a iance. Coe icien s wi h 95% CI no including ze o we e conside ed s a is ically signi ican .
824 Alonso-Pe ez e al.
mul iplica i e e ec s. Sex/gende and pe cei ed neighbo hood de-
p i a ion con ibu ed he mos o hese di e ences, ollowed by
amily ca egi ing du ies. No ably, indi iduals wi h high SES and
mid/high neighbo hood dep i a ion expe ienced he in e sec ional
haza dous e ec s. Ou indings highligh how he in e sec ion o
social de e minan s in luences sleep dispa i ies, which e lec
b oade in e sec ing inequali ies beyond indi idual ac o s.
Ou esul s e ealed ha 37.0% o he sample epo ed a leas one
ecen sleep dis u bance, highligh ing a signi ican public heal h
issue among Eu opean olde adul s [5]. Consis en wi h p e ious
esea ch, we ound ha ac o s such as being a woman, amily ca e-
gi ing, low educa ion, low-skill occupa ions, and pe cei ing he
neighbo hood as dep i ed we e associa ed wi h highe sleep dis u b-
ance [6, 8, 35]. These ac o s con ibu e o sleep dis u bance h ough
s esso s like inancial insecu i y, daily disc imina ion, less sleep
heal h li e acy, and con lic i e wo k–li e balances [36]. No iceably,
he combina ion o hese de e minan s led o highe sleep dis u b-
ance, wi h mul iple social in e sec ions and pe cei ed neighbo hood
dep i a ion join ly c ea ing in e sec ional sleep dispa i ies. While
be ween-s a a di e ences we e mos ly addi i e, ou indings e lec
ha b oade social o ces, including sys ems o powe and opp es-
sion, con ibu e o sleep dispa i ies [23].
Pe cei ed neighbo hood dep i a ion la gely con ibu ed o he
isk o sleep dis u bance compa ed o indi idual de e minan s.
Addi ionally, s a a wi h high SES bu high pe cei ed neighbo hood
dep i a ion had haza dous in e sec ional e ec s, whe eas some wi h
low neighbo hood dep i a ion had p o ec i e in e sec ional e ec s.
Ou esul s indica e ha a sa e and cohesi e en i onmen migh
bu e he e ec s o disad an aged social iden i ies. This highligh s
he impo ance o placing in e sec ional sleep inequali ies [37], since
he same social posi ion may in luence sleep di e en ly depending
on he con ex [28]. These indings a e in line wi h p io wo k e-
ealing ha cohesi e and sa e en i onmen s con ibu e o less sleep
dis u bance [13], whe eas non-cohesi e and dep i ed a eas a e asso-
cia ed wi h wo se sleep ou comes [35]. Neighbo hood con ex s can
in luence sleep h ough psychosocial, physiological, and social en-
gagemen mechanisms [11]. Pe cei ed low social cohesion, di iness,
and c ime may c ea e s ess and anxie y, while inc easing allos a ic
load and in lamma o y bioma ke s, which con ibu e o sleep dis-
u bance [31, 38]. Con e sely, a ea belonging o communi y suppo
imp o es sleep ou comes h ough sha ed esou ces, ca e access, o
ein o cemen o social no ms and beha io s. These si ua ions c ea e
oppo uni ies o es ul sleep, wi h a g ea e pe cep ion o sa e y
allowing o elaxa ion, whe eas high pe cei ed dep i a ion jeopa d-
izes such beha io s and p e en s be e sleep ou comes. Social en i-
onmen s shape daily ou ines, in e ac ing wi h s esso s and
he eby inc easing he isk o sleep dis u bance. Ou applica ion o
he social–ecological amewo k shows ha in e ac ing indi idual
and social ac o s c i ically in luence sleep dis u bance, ampli ying
s uc u al dispa i ies in dep i ed neighbo hoods.
The p esen s udy ein o ces he impo ance o social de e mi-
nan s o sleep [7], since indi iduals in mo e ad an aged social posi-
ions (men, no amily ca egi ing, high SES) epo ed less sleep
dis u bance. Likewise, disad an ageous social posi ions a e linked
o ins i u ionalized opp ession and s uc u al disc imina ion, ep o-
ducing in e sec ional social inequali ies on o sleep dispa i ies. I is
impe a i e no o ake o g an ed he p i ilege o access o com o ,
sa e y, and p i acy, which a e essen ial o good sleep, as sugges ed
Figu e 2. S a a-le el esiduals ob ained om MAIHDA Model 3 o each in e sec ional s a um, and hei 95% con idence in e als (CIs).
In e sec ional s a a a e anked om lowes o highes esidual. A esidual o ze o (ho izon al line) equals o he expec ed alue based on
main e ec s only, hus s a a wi h 95% CI abo e (below) ze o displayed mul iplica i e haza dous (p o ec i e) e ec s, colo ed in ed
(g een). Abb e ia ions: M ¼men; W ¼women; Hi edu ¼high educa ion; Me edu ¼medium educa ion; Lo Edu ¼low educa ion; HS ¼high-
skill occupa ion; LS ¼low-skill occupa ion; Hi dep i . ¼high neighbo hood dep i a ion; Me dep i ¼medium neighbo hood dep i a ion;
Lo dep i . ¼low neighbo hood dep i a ion.
In e sec ional sleep dispa i ies 825
by he ew s udies applying in e sec ional MAIHDA wi h
neighbo hood-le el con ex s [28, 31]. MAIHDA p o es as a use ul
me hodology o un a el how b oade social con ex s ma e o di -
e en in e sec ional subg oups in c ea ing sleep and heal h dispa -
i ies. In socie ies wi h ising inequali y, his unde sco es he need o
in e sec ionali y-in o med s a egies o iden i y and suppo sub-
g oups a highe isk o sleep dis u bance [39].
Pe spec i es and implica ions
The p omo ion o good sleep is an o e looked public heal h oppo -
uni y especially in olde adul s, conside ing i s ela ion wi h nume -
ous heal h ou comes and he modi iable na u e o ce ain social
de e minan s [1]. We ound a complex pa e n linking mul iple so-
cial inequali ies and hei in e sec ions o sleep dispa i ies. No ably,
people in he same social posi ion epo ed a ying sleep dis u b-
ance depending on hei pe cei ed neighbo hood dep i a ion. This
emphasizes he need o in eg a e social con ex s in quan i a i e
in e sec ional analyses o heal h inequali ies, as place shapes he
complexi y o li ed expe iences wi hin popula ion subg oups [37].
Fu u e esea ch on sleep dispa i ies should inco po a e an in e sec-
ional lens and conside b oade social ac o s ha a e beyond he
scope o indi iduals, while s ill a ec ing hei sleep.
Ou esul s sugges ha indi idual sleep in e en ions, pa icula ly
o women wi h ca egi ing du ies and low SES, could be less e ec i e
unless en i onmen al ac o s we e included. I is c ucial o shi he
ocus om indi iduals as he only agency ac o o, ins ead, in es iga -
ing he implica ions o neighbo hood-le el sleep p e en ion app oaches
[23, 40]. In combina ion wi h indi idual-le el sleep beha io al ecom-
menda ions o he apy, a ge ed sleep p omo ion in e en ions should
be placed a mul iple con ex ual le els, o ins ance well-main ained
neighbo hoods, a ea-speci ic equi able access o esou ces, and sleep-
iendly media and wo kplaces. Echoing ecen calls o ac ion in dis-
ad an aged neighbo hoods [40], MAIHDA could se e as a ool o
de elop a ge ed in e en ions in “sleep dese s”, whe e s uc u al dis-
c imina ion equi es cul u ally ailo ed in e en ions a he communi y
le el. Based on he social–ecological model o sleep, we encou age he
p io i iza ion o con ex and connec ion o e indi idualism in public
heal h p og ams wi h he aim o p omo e sleep heal h equi y.
S eng hs and limi a ions
We used a la ge, ep esen a i e sample wi h alida ed scales, ena-
bling esul s gene aliza ion. The social–ecological model o sleep
heal h guided ou design, p o iding insigh s in o mul ile el socio-
en i onmen al de e minan s o sleep. We also applied in e sec ion-
ali y heo y h ough MAIHDA modeling o add ess how in e sec-
ions o social de e minan s join ly con ibu e o sleep dispa i ies. A
limi a ion was sel - epo ed sleep dis u bances, as objec i e meas-
u es like polysomnog aphy o accele ome e s we e una ailable in
SHARE. Fu u e esea ch should combine subjec i e and objec i e
sleep e alua ions o mo e comp ehensi e insigh s. Neighbo hood
con ex was also sel - epo ed, lacking geog aphical da a o objec -
i e dep i a ion measu es. While subjec i e neighbo hood assess-
men s a e alid p edic o s o sleep, u u e s udies should in eg a e
objec i e a ea dep i a ion da a wi h in e sec ionali y. Addi ionally,
SHARE lacks ace/e hnici y and sexual iden i y indica o s, c i ical
ac o s in US s udies on sleep dispa i ies among ulne able g oups.
We excluded mig a ion backg ound gi en i s limi ed ope a ionali-
za ion, which could no adequa ely cap u e acial/e hnic dimen-
sions. Hence, Eu opean s udies should inco po a e ace/e hnici y
and sexual iden i y da a o be e unde s and he in e sec ionali y
o sleep dispa i ies. Las ly, ou sample included only olde adul s
aged 50þ, a g oup p one o sleep dis u bances and cumula i e social
inequali ies. Fu u e esea ch should examine in e sec ional sleep
dispa i ies in younge coho s.
In conclusion, using an in e sec ional lens is c ucial o iden i y
sleep and heal h dispa i ies and in o m p ecision public heal h
measu es. Gi en he impo ance o neighbo hood ac o s, u u e
esea ch should in eg a e b oade con ex s and ocus on mul ile el
sleep p omo ion in e en ions add essing ac o s beyond
he indi idual.
Acknowledgemen s
This pape uses da a om SHARE Wa es 4 and 5 (DOIs: 10.6103/
SHARE.w4.800, 10.6103/SHARE.w5.800). We would like o hank
all membe s o he Eins ein Cen e Popula ion Di e si y (ECPD). E.
A.P. is a membe o he Eu opean Public Heal h Associa ion (mem-
be ship ID: 62951).
Au ho con ibu ions
E.A.P., P.G., and J.L.O.S. concep ualized he esea ch idea and E.A.
P. w o e he o iginal a icle. E.A.P. ca ied ou he ini ial da a ana-
lysis, wi h inpu om P.G. and J.L.O.S. X.D. and J.L.O.S. c i ically
e ised he a icle and analyses. P.G. acqui ed he unding o ob ain
inancial suppo o his publica ion. All au ho s app o ed he
inal a icle.
Supplemen a y da a
Supplemen a y da a a e a ailable a EURPUB online.
Con lic o in e es : The au ho s ha e no con lic s o in e es
o disclose.
Funding
The SHARE da a collec ion has been unded by he Eu opean
Commission h ough FP5 (QLK6-CT-2001-00360), FP6 (SHARE-
I3: RII-CT-2006-062193, COMPARE: CIT5-CT-2005-028857,
SHARELIFE: CIT4-CT-2006-028812), FP7 (SHARE-PREP: GA N�
211909, SHARE-LEAP: GA N�227822, SHARE M4: GA N�
261982), and Ho izon 2020 (SHARE-DEV3: GA N�676536,
SERISS: GA N�654221) and by DG Employmen , Social A ai s &
Inclusion. Addi ional unding om he Ge man Minis y o
Educa ion and Resea ch, he Max Planck Socie y o he
Ad ancemen o Science, he US Na ional Ins i u e on Aging
(U01_AG09740-13S2, P01_AG005842, P01_AG08291,
P30_AG12815, R21_ AG025169, Y1-AG-4553-01, IAG_BSR06-11,
OGHA_04-064, HHSN271201300071C), and om a ious na ional
unding sou ces is g a e ully acknowledged (see www.sha e-e ic.eu).
This wo k was unded by he Eins ein Founda ion Be lin (g an
numbe EZ-2019-555-2), suppo ing he Eins ein Cen e
Popula ion Di e si y (ECPD) in Be lin. The sala ies om E.A.-P.
( o ally), X.D. (pa ially) and J.L.O.S. (pa ially) a e unded by he
g an o he Eins ein Cen e Popula ion Di e si y (ECPD). Open
Access unding was pa ially enabled and o ganized by he Open
Access Publica ion Fund o he Medical Lib a y a Cha i �
e–
Uni e si €
a smedizin Be lin and he Ge man Resea ch Founda ion
(DFG). The es o Open Access unding wwas pa ially enabled
by he Open Access unds om he Eins ein Cen e Popula ion
Di e si y (ECPD).
Da a a ailabili y
This s udy uses da a o he Su ey o Heal h, Ageing, and Re i emen
in Eu ope (SHARE). SHARE da a a e publicly a ailable upon ee
egis a ion (h ps://sha e-e ic.eu/da a/, accessed on 21
Sep embe 2024).
826 Alonso-Pe ez e al.
Re e ences
10Hale L, T oxel W, Buysse DJ. Sleep heal h: an oppo uni y o public heal h o
add ess heal h equi y. Annu Re Public Heal h 2020;41:81–99.
20Jackson CL, Hu FB, Redline S e al. Racial/e hnic dispa i ies in sho sleep du a ion
by occupa ion: he con ibu ion o immig an s a us. Soc Sci Med 2014;118:71–9.
30Lube kin EI, Jia H. Bu den o disease due o sleep du a ion and sleep p oblems in
he elde ly. Sleep Heal h 2018;4:182–7.
40S e niczuk R, Theou O, Rusak B e al. Sleep dis u bance is associa ed wi h inciden
demen ia and mo ali y. Cu Alzheime Res 2013;10:767–75.
50Calem M, Bisla J, Begum A e al. Inc eased p e alence o insomnia and changes in
hypno ics use in England o e 15 yea s: analysis o he 1993, 2000, and 2007
Na ional Psychia ic Mo bidi y Su eys. Sleep 2012;35:377–84.
60Duncan DT, Kawachi I, Redline S. The Social Epidemiology o Sleep. Ox o d, Uni ed
Kingdom: Ox o d Uni e si y P ess, 2019.
70G andne MA. Sleep, heal h, and socie y. Sleep Med Clin 2022;17:117–39.
80Hale L, Emanuele E, James S. Recen upda es in he social and en i onmen al
de e minan s o sleep heal h. Cu Sleep Med Rep 2015;1:212–7.
90G andne MA. Social-ecological model o sleep heal h. Sleep and Heal h.
Camb idge, Massachuse s, Uni ed S a es o Ame ica: Else ie Academic P ess,
2019, 45–53.
10 C enshaw K. Mapping he ma gins: in e sec ionali y, iden i y poli ics, and iolence
agains women o colo . S an L Re 1991;43:1241.
11 Alhasan DM, Gas on SA, Jackson WB e al. Neighbo hood social cohesion and sleep
heal h by age, sex/gende , and ace/e hnici y in he Uni ed S a es. In J En i on Res
Public Heal h 2020;17:9475.
12 G andne MA, Williams NJ, Knu son KL e al. Sleep dispa i y, ace/e hnici y, and
socioeconomic posi ion. Sleep Med 2016;18:7–18.
13 Adjaye-Gbewonyo D, Ng AE, Jackson CL e al. The pe cei ed neighbo hood
walking en i onmen and sel - epo ed sleep heal h in a na ionally ep esen a i e
sample o he Uni ed S a es. Heal h Place 2023;83:103066.
14 Johnson DA, B own DL, Mo gens e n LB e al. The associa ion o neighbo hood
cha ac e is ics wi h sleep du a ion and day ime sleepiness. Sleep Heal h 2015;
1:148–55.
15 Johnson DA, Jackson CL, Williams NJ e al. A e sleep pa e ns in luenced by ace/
e hnici y—a ma ke o ela i e ad an age o disad an age? E idence o da e. Na Sci
Sleep 2019;11:79–95.
16 E ans T, B own H. Road a ic c ashes: ope a ionalizing equi y in he con ex o
heal h sec o e o m. Inj Con ol Sa P omo 2003;10:11–2.
17 O’Neill J, Tabish H, Welch V e al. Applying an equi y lens o in e en ions: using
PROGRESS ensu es conside a ion o socially s a i ying ac o s o illumina e
inequi ies in heal h. J Clin Epidemiol 2014;67:56–64.
18 Ba anowski M, Jabkowski P. Gende and socioeconomic pa e ning o sel - epo ed
sleep p oblems ac oss Eu opean coun ies. Eu J Public Heal h 2023;33:242–8.
19 Rich e D, K €ame MD, Tang NK e al. Long- e m e ec s o p egnancy and
childbi h on sleep sa is ac ion and du a ion o i s - ime and expe ienced mo he s
and a he s. Sleep 2019;42:zsz015.
20 Ha is EC, D’Angelo S, Syddall HE e al. Rela ionships be ween in o mal ca egi ing,
heal h and wo k in he heal h and employmen a e i y s udy, England. Eu J
Public Heal h 2020;30:799–806.
21 Akimo a ET, Taiji R, Ding X e al. Gene-x-en i onmen analysis suppo s p o-
ec i e e ec s o e eningness ch ono ype on sel - epo ed and ac ig aphy-de i ed
sleep du a ion among hose who always wo k nigh shi s in he UK Biobank. Sleep
2023;46:zsad023.
22 Cace es BA, Hickey KT, Hei kempe EM e al. An in e sec ional app oach o
examine sleep du a ion in sexual mino i y adul s in he Uni ed S a es: indings om
he beha io al isk ac o su eillance sys em. Sleep Heal h 2019;5:621–9.
23 Duba RT. #NoJus iceNoSleep: c i ical in e sec ions o ace-e hnici y, income, edu-
ca ion, and social de e minan s in sleep heal h dispa i ies. Sleep Heal h 2022;8:7–10.
24 E ans CR, Williams DR, Onnela J-P e al. A mul ile el app oach o modeling heal h
inequali ies a he in e sec ion o mul iple social iden i ies. Soc Sci Med 2018;
203:64–73.
25 E ans CR, Bo ell LN, Bell A e al. Cla i ica ions on he in e sec ional MAIHDA
app oach: a concep ual guide and esponse o Wilkes and Ka imi (2024). Soc Sci
Med 2024;350:116898.
26 B€o sch-Supan A, B and M, Hunkle C e al.; SHARE Cen al Coo dina ion Team.
Da a esou ce p o ile: he Su ey o Heal h, Ageing and Re i emen in Eu ope
(SHARE). In J Epidemiol 2013;42:992–1001.
27 Oglia i G, Ryg J, Ande sen-Ranbe g K e al. Pe cei ed neighbou hood en i onmen
and alls among communi y-dwelling adul s: c oss-sec ional and p ospec i e ind-
ings om he Su ey o Heal h, Ageing and Re i emen in Eu ope (SHARE). Eu J
Ageing 2022;19:1121–34.
28 E ans CR. Rein eg a ing con ex s in o quan i a i e in e sec ional analyses o heal h
inequali ies. Heal h Place 2019;60:102214.
29 Al a ez CH, E ans CR. In e sec ional en i onmen al jus ice and popula ion heal h
inequali ies: a no el app oach. Soc Sci Med 2021;269:113559.
30 Me lo J, €
Obe g J, Khala K e al. Geog aphical and sociodemog aphic di e ences in
s a in dispensa ion a e acu e myoca dial in a c ion in Sweden: a egis e -based
p ospec i e coho s udy applying analysis o indi idual he e ogenei y and dis-
c imina o y accu acy (AIHDA) o basic compa isons o heal hca e quali y. BMJ
Open 2023;13:e063117.
31 Holman D, Bell A, G een M e al. Neighbou hood dep i a ion and in e sec ional
inequali ies in bioma ke s o heal hy ageing in England. Heal h Place 2022;77:102871.
32 Axelsson Fisk S, Mulina i S, Wem ell M e al. Ch onic obs uc i e pulmona y
disease in Sweden: an in e sec ional mul ile el analysis o indi idual he e ogenei y
and disc imina o y accu acy. SSM Popul Heal h 2018;4:334–46.
33 Van De S aa V, B acke P. How well does Eu ope sleep? A c oss-na ional s udy o
sleep p oblems in Eu opean olde adul s. In J Public Heal h 2015;60:643–50.
34 Ande s MP, B eckenkamp J, Ble ne M e al. Associa ion be ween socioeconomic
ac o s and sleep quali y in an u ban popula ion-based sample in Ge many. Eu J
Public Heal h 2014;24:968–73.
35 Johnson DA, Billings ME, Hale L. En i onmen al de e minan s o insu icien sleep
and sleep diso de s: implica ions o popula ion heal h. Cu Epidemiol Rep 2018;
5:61–9.
36 Sco elle AJ, Hewi B, Lallukka T e al. Time use, ime p essu e and sleep: is gende
an e ec modi ie ? Eu J Public Heal h 2023;33:411–7.
37 Bamb a C. Placing in e sec ional inequali ies in heal h. Heal h Place 2022;
75:102761.
38 Chen X, Redline S, Shields AE e al. Associa ions o allos a ic load wi h sleep apnea,
insomnia, sho sleep du a ion, and o he sleep dis u bances: indings om he
Na ional Heal h and Nu i ion Examina ion Su ey 2005 o 2008. Ann Epidemiol
2014;24:612–9.
39 Cha u VK, Manza MD, Kuma y S e al. The global p oblem o insu icien sleep
and i s se ious public heal h implica ions. Heal hca e 2018;7:1.
40 A a ian H, Dunie z GL, Ga idia-Rome o R e al. Add essing sleep dese s: a p o-
posed call o ac ion. Sleep Heal h 2024;10:S15–8.
Key poin s
•The e we e subs an ial sleep dispa i ies ac oss in e sec ional
g oups.
•Sex/gende and pe cei ed neighbo hood dep i a ion we e key
d i e s o sleep dispa i ies.
•Pe cei ed neighbo hood dep i a ion was linked o sleep
dis u bance ega dless o indi idual social de e minan s.
•The use o an in e sec ional lens h ough MAIHDA is c ucial
o shape public heal h policies.
•Mul ile el a ge ed sleep p omo ion in e en ions should
conside b oade social con ex s.
In e sec ional sleep dispa i ies 827
© The Au ho (s) 2025. Published by Ox o d Uni e si y P ess on behal o he Eu opean Public Heal h Associa ion.
This is an Open Access a icle dis ibu ed unde he e ms o he C ea i e Commons A ibu ion-NonComme cial License (h ps://c ea i ecommons.o g/licenses/by-nc/4.0/), which
pe mi s non-comme cial e-use, dis ibu ion, and ep oduc ion in any medium, p o ided he o iginal wo k is p ope ly ci ed. Fo comme cial e-use, please con ac ep in s@oup.
com o ep in s and ansla ion igh s o ep in s. All o he pe missions can be ob ained h ough ou Righ sLink se ice ia he Pe missions link on he a icle page on ou si e— o
u he in o ma ion please con ac [email p o ec ed].
Eu opean Jou nal o Public Heal h, 2025, 35, 821–827
h ps://doi.o g/10.1093/eu pub/cka 130
O iginal Manusc ip
