Content uploaded by Rania Tohme
Author content
All content in this area was uploaded by Rania Tohme on Apr 18, 2018
Content may be subject to copyright.
EMHJ • Vol. 19 No. 12 • 2013 Eastern Mediterranean Health Journal
La Revue de Santé de la Méditerranée orientale
1026
Living arrangements of older adults in Lebanon:
correlates of living with married children
O. Shideed,1 A. Sibai 1 and R. Tohme
1
ABSTRACT Rapid increases in the proportion of older adults in the population present major challenges to
policy-makers worldwide. Using a nationally representative sample from the PAPFAM survey in Lebanon, this
study examined the living arrangements of older adults (aged ≥ 65 years), and their correlates, with a focus on
co-residence with married children. Of 1774 older adults 17.1% co-resided with their married children: 28.1%
of the 559 unmarried (widowed/divorced/single) and 11.3% of the 1071 married older adults. Among both the
married and unmarried, the likelihood of co-residence was significantly lower in regions outside the capital and
decreased with increasing socioeconomic status. Among the unmarried elderly, co-residence with a married
child was also significantly associated with increasing age and availability of sons, as well as presence of a vascular
disorder and speech problems. While solitary living has traditionally been the focus for policy-makers, older
people living with a married child may also be a vulnerable group.
1Department of Epidemiology and Population Health, Faculty of Health Sciences, American University of Beirut, Beirut, Lebanon (Correspondence
to O. Shideed: shideedo@who.int).
Received: 28/06/12; accepted: 20/02/13
65
%11.3%28.1%17.11774
Modalités de vie des personnes âgées au Liban : corrélats de la cohabitation avec des enfants mariés
RÉSUMÉ L’augmentation rapide de la proportion de personnes âgées dans la population pose des difficultés
majeures aux responsables de l’élaboration des politiques dans le monde. À l’aide d’un échantillon représentatif
au niveau national issu de l’enquête du Projet panarabe pour la santé de la famille (PAPFAM) au Liban, la présente
étude a examiné les modalités de vie des personnes agées (65 ans ou plus) et leurs corrélats, en étudiant plus
particulièrement la cohabitation avec des enfants mariés. Sur 1774 personnes âgées, 17,1 % cohabitaient avec
leurs enfants mariés, tout comme 28,1 % des 559 personnes âgées sans conjoint (veufs/divorcés/célibataires)
et 11,3 % des 1071 personnes âgées mariées. Chez les personnes âgées mariées comme chez les autres, la
probabilité de cohabiter était très inférieure dans les régions par rapport à la capitale. En outre, plus le statut
socioéconomique augmentait, plus cette probabilité diminuait. Parmi les personnes âgées sans conjoint, la
cohabitation avec un enfant marié était fortement associée à un âge plus avancé et au fait d’avoir des fils, ainsi qu’à
la présence de troubles vasculaires et de la parole. Si les personnes âgées vivant seules ont traditionnellement fait
l’objet d’attention de la part des responsables politiques, celles partageant un toit avec un enfant marié peuvent
aussi représenter un groupe vulnérable.
1027
Introduction
Recent studies in Lebanon have pro-
jected an increase in the proportion of
older adults in the population, similar
to contemporary Europe [1]. Yet, amid
a number of other urgent problems,
ageing and issues related to the health
and care of older adults have not been
adequately addressed by policy-makers
as an area of priority concern. In most
Middle Eastern countries, societies
value family ties, and older adults are
viewed as a resource to the family. While
cultural and ethical values still protect
the majority of older people in Leba-
non, it is likely that recent changes in
family structure, with a trend towards
smaller families, youth migration and
increasing participation of women in
the workforce have weakened the family
as a social institution [1,2].
Studies have underscored the im-
portance of living arrangements to the
well-being of older adults [3,4]. is
issue is of great importance especially in
regions characterized by the scarcity of
social security safety nets, public institu-
tions and pension plans. As mentioned
earlier, Lebanon is undergoing a demo-
graphic transformation characterized
by lower mortality and fertility rates
and longer longevity. But rather than
occurring in isolation, the transforma-
tion is embedded in cultural, social and
economic circumstances that are also
changing and thereby leading to an im-
balance in the traditional relationships
and exchange of resources between
generations [5]. e traditional con-
cept of intergenerational support and
provision of care and assistance to older
adults is “on the brink” [6].
One living arrangement of particu-
lar interest in the Lebanese context is liv-
ing with married children. In Lebanon
as in other Arab countries, the norm
is for adult children to co-reside with
their parents until marriage, when they
establish individual households [7,8].
e literature has indicated that living
with children provides mutual benet
for the older adult who receives social,
nancial and health assistance from the
adult child and in turn helps in caring
for the grandchildren or caring for the
house [9–13]. Married children are of-
ten considered as important resources
of economic security for their older
parents [14]. Older women are more
likely to live with their children in Egypt,
because women outlive men, and are
more disabled at older ages [8]. A higher
percentage of older adults co-reside
with their married male child than with
their married female child [8,9].
Published studies on the living ar-
rangements of older adults in the East-
ern Mediterranean Region as well as
in Lebanon remain limited in scope,
and a majority lacked health-related
characteristics that are regarded in the
literature as signicant confounders
in the older person’s choice of living
arrangement [9]. Using data from a
national study conducted in Lebanon,
the current study aimed to examine
the living arrangements of older adults
in Lebanon, and their correlates, with
a focus on those living with children.
Earlier studies have shown that co-
residence with children is likely to bring
material advantages, especially for the
older adults. Nevertheless, it is unclear
whether living with married children
presents comparable advantages as
living with unmarried children [15].
e distinction is particularly important
because in the context of Lebanon, chil-
dren are expected to live with parents
until marriage.
Methods
Study design
is study of older Lebanese compared
the living arrangements of married
and currently unmarried (widowed,
divorced or separated) individuals. We
also assessed the demographic, socio-
economic and health correlates of living
with married children compared to
other living arrangements.
Study population and sample
e study population consisted of
persons aged ≥ 65 years who were in-
cluded in the Household Health Survey
conducted in 2004 by the Lebanese
Ministry of Social Aairs in collabora-
tion with the League of Arab States as
part of the Pan Arab Project for Family
Health (PAPFAM) [16]. e sample
for the study was selected using a 3-stage
stratied cluster sample design. In the
rst stage, 15 geographical areas were
randomly selected from the 6 governo-
rates of Lebanon, and in the second, 449
population survey units were randomly
sampled. e third stage included the
sampling of 7098 main dwellings. A to-
tal of 6505 households were visited and
interviews were completed for 5532,
yielding a response rate of 85%. A total
of 1812 older adults were identied, of
whom 1774 completed the interviews,
yielding a 97.9% response rate.
e study sample included ever-
married subjects with at least 1 child
alive. Older adults who were never mar-
ried or did not have any children still
alive were excluded because they were
not at-risk for the outcome (living with
married children). e sample included
1684 (95.7%) ever-married individuals
and 52 (3.0%) with no children alive.
e nal analysis was based on data
from 1630 older adults, 1071 married
and 559 unmarried.
Variables
e outcome variable of interest was
the living arrangements of older adults.
In this study living arrangements
were initially grouped into 5 mutually
exclusive categories according to the
relation of the older adult to other
household members: living alone, living
with spouse (no children), living with
unmarried children (with or without
spouse), living with at least 1 married
child and/or ospring (with or without
spouse) and living with non-spouse/
non-children others. Except for those
living alone, the remaining categories in-
cluded those who may or may not have
EMHJ • Vol. 19 No. 12 • 2013 Eastern Mediterranean Health Journal
La Revue de Santé de la Méditerranée orientale
1028
been living with a sibling, other relatives
or non-relative others. For the objective
of the study, living arrangements were
dichotomized into a binary variable:
living with married children versus all
other arrangements.
Exposure variables which assessed
the demographic, socioeconomic
and health status of the sample were
included. ese variables, according to
existing literature, were expected to have
an eect on the living arrangements of
older adults.
Demographic variables included
age (65–69, 70–74, 75–79, 80–84 or ≥
85 years), sex male or female, region Bei-
rut, Mount Lebanon, North Lebanon,
South Lebanon, Nabatieh or Beka’a,
number of sons and daughters alive
(0, 1–2, 3–4 or ≥ 5) and marital status
(single, married, widowed, divorced
or separated). Because few respond-
ents reported being single, divorced or
separated, they were merged for analysis
with those who were widowed into one
category: unmarried.
Socioeconomic variables included
educational level (no formal schooling,
aended preparatory school or higher
education), work status (never worked,
worked in past or currently working),
nancial status in household (nan-
cially independent, having dependants
or nancially dependent on others),
income sucient for main expenses
(used to assess adequacy of monthly in-
come, dichotomized yes/no), monthly
per capita income.
Health-related variables included
self-rated health assessed by asking the
older adult “Do you think your health in
general is good, fair or bad?” (coded in
3 categories). Participants were asked if
they suered from any disabling chron-
ic diseases that required continuous
monitoring: hypertension, vascular dis-
order, diabetes mellitus or pulmonary
disorder (coded yes/no). ey were
asked if they had any communication
problems for 3 senses: speech, hearing
and sight (coded yes/no). Diculty
with activities of daily living (ADL)
[17] were assessed for 5 activities: using
toilet, bathing, dressing, moving in/out
of chair and eating (coded no help or
partial/total help), as well as the total
ADL score.
Statistical analysis
Frequency distributions of living
arrangements as well as baseline
characteristics (demographic, socioec-
onomic and health-related conditions )
were evaluated for the total sample and
for married and unmarried individu-
als separately. Dierences by marital
status were assessed using Pearson
chi-squared test of association. Age-
adjusted odds ratios (OR) and their
respective 95% condence intervals
(CI) comparing each exposure with its
reference category were estimated.
A multivariate logistic regression
model was estimated for married and
unmarried individuals separately. Vari-
ables included in the multivariate model
were age, sex, region, number of sons
alive and number of daughters alive,
literacy, monthly per capita income,
chronic diseases, ADL score and com-
munication problems. OR and 95%
CI were obtained, and signicance was
assessed via the Wald test statistic.
Results
Overview of the study
population
e total sample included 1630 older
adults: 1071 (65.7%) married and 559
(34.2%) unmarried. Overall, 41.3% of
the individuals had 1–2 sons alive and
43.2% had 1–2 daughters alive. As for
the socioeconomic factors, they were
similar for the 2 groups. With regard to
literacy, 59.0% of the older adults had
no formal schooling and almost 40%
reported a monthly per capita income
< US$ 100. Table 1 shows the distri-
bution of demographic and socioeco-
nomic characteristics among the study
population, stratied by marital status.
Compared with married individuals,
those who were unmarried were signi-
cantly more likely to be aged ≥ 85 years
(9.5% versus 2.8%; P < 0.001), to be
female (78.9% versus 32.5%; P < 0.001)
and to be living in Beirut (19.1% versus
10.9%; P < 0.001).
e distribution of health-related
conditions among married and unmar-
ried individuals is presented in Table 2.
Currently unmarried older adults were
slightly more likely to suer from ill-
nesses and co-morbidities than married
older adults. Currently unmarried older
adults had signicantly higher ADL
scores, indicating need greater need
for help. Diculties in hearing and
sight was signicantly more common
for currently unmarried than married
respondents: (19.1% versus 11.9% P
< 0.001) and (22.9% versus 13.9% P <
0.001) respectively.
Living arrangements
e ndings revealed that the high-
est percentage of older adults (47.3%)
lived in a nuclear family with their un-
married children, with or without their
spouse (Table 3). Living with at least 1
married child accounted for 17.1% of
the total sample, with the currently un-
married being more than twice as likely
to be living with their married children
as their married counterparts (28.1%
versus 11.3%). Around 10% were living
alone and an additional 24.2% were
living with their spouse. e majority
of the older adults (91.0%) reported
being satised with their current living
arrangement and there was no dier-
ence between married and unmarried
individuals.
Associations between co-
variables and living with
married children
Binary logistic regression was carried
out to assess associations with our main
outcome variable, living with married
children compared with other living
arrangements. Tables 4 and 5 present
the age-adjusted OR for demographic,
1029
socioeconomic and health-related
factors. e likelihood of living with
married children increased signi-
cantly with increasing age. In general,
older adults in Beirut were more likely
to co-reside with their married children
compared with those in all other gov-
ernorates. Currently unmarried older
adults with at least 1 son were signi-
cantly more likely to be co-residing with
a married child (OR = 4.08; 95% CI:
1.77–9.41), and married individuals
Table 1 Distribution of sociodemographic factors of adults aged ≥ 65 years stratified by marital status
Sociodemographic variables Total Married UnmarriedaP-value
(n = 1630) (n = 1071) (n = 559)
No. %No. %No. %
Age (years)
65–69 504 30.9 375 35.0 129 23.1 < 0.001
70–74 540 33.1 378 35.3 162 29.0
75–79 326 20.0 198 18.5 128 22.9
80–84 178 10.9 91 8.5 87 15.6
≥ 85 83 5.1 30 2.8 53 9.5
Mean (SD) 73.2 (6.1) 72.3 (5.5) 74.9 (6.8) < 0.001
Sex
Male 841 51.6 723 67.5 118 21.1 < 0.001
Female 788 48.4 348 32.5 440 78.9
Region
Beirut 224 13.7 117 10.9 107 19.1 < 0.001
Beka’a 177 10.9 126 11.8 51 9.1
Mount Lebanon 632 38.7 394 36.8 238 42.5
Nabatieh 125 7. 7 92 8.6 33 5.9
North 310 19.0 220 20.5 90 16.1
South 163 10.0 122 11.4 41 7. 3
No. of sons alive
0 111 6.8 86 6.3 43 7. 7 0.006
1–2 673 41.3 422 39.4 251 44.9
3–4 578 35.5 382 35.7 196 35.1
≥ 5 268 16.4 199 18.6 69 12.3
No. of daughters alive
0 107 6.6 68 6.4 39 7. 0 0.510
1–2 703 43.2 451 42.1 252 45.1
3–4 540 33.1 359 33.6 181 32.4
≥ 5 279 1 7. 1 192 17.9 87 15.6
Literacy
No formal schooling 948 59.0 588 55.6 360 65.3 < 0.001
≤ Preparatory 525 32.6 360 34.1 165 29.9
Higher 135 8.4 109 10.3 26 4.7
Per capita income (US$/month)
≤ 50 270 16.6 199 18.6 71 12.7 < 0.001
50–≤ 100 375 23.0 247 23.1 128 22.9
100–≤ 150 273 16.7 167 15.6 106 19.0
150–≤ 200 177 10.9 138 12.9 39 7. 0
≥ 200 375 23.0 224 20.9 151 27.0
No response 160 9.8 96 9 64 11.4
aSingle, divorced, separated and widowed.
SD = standard deviation.
EMHJ • Vol. 19 No. 12 • 2013 Eastern Mediterranean Health Journal
La Revue de Santé de la Méditerranée orientale
1030
with at least 1 daughter were more likely
to live with a married child, although
this was not signicant (OR = 2.23; 95%
CI: 0.92–5.37). With regard to socio-
economic factors, the overall the results
showed that married and unmarried
individuals of higher socioeconomic
status were less likely to be living with
married children. is was evident for
monthly per capita; as income increased,
the likelihood of living with married
children signicantly decreased.
e data also showed that both
married and unmarried older adults
living with married children were more
likely to report chronic diseases. For
unmarried older adults, hypertension
and vascular disease were signicantly
associated with living with married chil-
dren (OR = 1.67; 95% CI: 1.15–2.43
and OR = 2.02; 95% CI: 1.37–2.98
respectively). Diculty in performing
ADL was also signicantly associated
with living with married children among
unmarried individuals (OR = 1.16; 95%
CI: 1.02–1.32). Similarly, older adults
living with married children were sig-
nicantly more likely to report com-
munication problems; specically, there
were higher odds of having hearing and
Table 2 Distribution of healt h-related factors of adults aged ≥ 65 years stratified by marital status
Health-related variables Total Married UnmarriedaP-value
(n = 1630) (n = 1071) (n = 559)
No. %No. %No. %
Chronic diseases
Hypertension
Yes 596 36.6 345 32.2 251 44.9 < 0.001
No 1034 63.4 726 67.8 308 55.1
Vascular disorder
Yes 434 26.6 262 24.5 172 30.8 0.006
No 1196 73.4 809 75.5 387 69.2
Diabetes mellitus
Yes 358 21.9 218 20.3 140 25.0 0.029
No 1273 78.1 854 79.7 419 75.0
Pulmonary disorder
Yes 101 6.2 61 5.7 40 7. 2 0.244
No 1530 93.8 1011 94.3 519 92.8
Activit ies of daily living
ADL score
0 1387 85.0 946 88.2 441 78.8 < 0.001
1 52 3.2 26 2.4 26 4.6
2 46 2.8 19 1.8 27 4.8
3 34 2.1 15 1.4 19 3.4
4 30 1.8 13 1.2 17 3.0
5 83 5.1 53 4.9 30 5.4
Mean (SD) 0.48 (1.29) 0.40 (1.22) 0.63 (1.40) < 0.001
Communication problems
Speech problems
Yes 98 6.0 56 5.2 42 7. 5 0.065
No 1532 94.0 1015 94.8 517 92.5
Hearing problems
Yes 234 14.4 127 11.9 107 19.1 < 0.001
No 1396 85.6 944 88.1 452 80.9
Sight problems
Yes 277 17.0 149 13.9 128 22.9 < 0.001
No 1353 83.0 922 86.1 431 7 7. 1
aSingle, divorced, separated and widowed.
SD = standard deviation.
1031
sight problems among married (OR
= 1.99; 95% CI: 1.21–3.27 and OR =
1.80; 95% CI: 1.12–2.89 respectively )
than currently unmarried individuals
(OR = 1.69; 95% CI: 1.09–2.64 and OR
= 1.41; 95% CI: 0.92–2.16 respectively).
Among currently unmarried individuals
living with married children, diculty
in speech was 3.5 times more common
than among the married (95% CI:
1.85–6.72).
Multivariate analysis
Tables 4 and 5 also show the multivari-
ate analyses for the selected variables.
Age was found to be a signicant factor
for living with married children only in
unmarried individuals (OR = 1.79; 95%
CI: 1.08–2.93), whereas region was a
signicant predictor for living with mar-
ried children among both married and
unmarried individuals (OR = 0.36; 95%
CI: 0.20–0.63 and OR = 0.40; 95% CI:
0.24–0.67 respectively), with a negative
association in other regions compared
with Beirut. Among the currently
unmarried, having at least 1 son alive
signicantly increased the likelihood of
living with a married child (OR = 5.40;
95% CI: 2.12–13.74). A signicant as-
sociation was observed with monthly
per capita; as monthly income increased,
older adults were less likely to live with
their married children (Table 4).
For chronic diseases, only vascular
disease among the currently unmar-
ried retained its signicant association
with living with a married child (OR
= 1.69; 95% CI: 1.08–2.64). For
communication problems, only speech
among the currently unmarried was
signicantly associated with co-residing
with a married child (OR = 4.19; 95%
CI: 1.71–10.23) (Table 5).
Discussion
Living with others, in particular children,
has been considered as one arrange-
ment in which older persons are closer
to their social networks and are more
likely to receive support when needed.
It is expected that co-residence would
be associated with mutual benet for
both the older adult and their children
by increasing material and non-material
resource exchanges between the genera-
tions, as well as expanding their support
networks [18]. However, it is not clear
whether the advantages conferred by
living with married children are compa-
rable to those of living with unmarried
children [15].
e main purpose of this study was
to assess the living arrangements of old-
er adults and the correlates of living with
married children among older adults in
Lebanon. In the Lebanese context, the
norm is for children to live with their
parents until they are married, at which
time they are then expected to leave their
parent’s home and establish their own
households. e results of this study
show that overall 17.1% of older adults
in Lebanon live with at least 1 married
child. is percentage is similar to that
of a previous study conducted in Beirut
(1993–1994), where the percentage of
older adults living with at least 1 married
child was 19.1% [14]. e likelihood of
living with married children was around
2.5 times greater among the unmarried
when compared with the married older
adults (28.1% versus 11.3%.)
Consistent with other studies, this
study demonstrated that older adults
co-residing with married children was
positively associated with the avail-
ability of living children and with age
of the older adult [2,6,9,14,17]. In the
Lebanese context, due to strong fam-
ily relations, increases in age and in de-
pendency (both nancial and physical),
is oen accommodated by the family.
Living with married children was in-
versely associated with socioeconomic
status, including education of the older
person and monthly per capita income.
Also, those co-residing with married
children were more likely to be receiv-
ing aid from government sources than
their unmarried counterparts. ese
results were consistent for both married
and unmarried people, although the
strength of the association was stronger
among the married. Findings from
other published studies are inconsist-
ent in this regard. For example, studies
elsewhere in the Arab region showed
co-residence with children was associ-
ated with higher socioeconomic status.
Studies conducted in Kuwait by Shah et
al. [19] and in Egypt by Yount [8] found
that high income and standards of living
encouraged co-residence due to the
Table 3 Distribution of living arrangements of adults aged ≥ 65 years stratified by marital status
Living arrangements Total Married Unmarried
aP-value
(n = 1630) (n = 1071) (n = 559)
No. %No. %No. %
Alone 171 10.5 0 0.0 171 30.6 <0.001
Spouse 395 24.2 395 36.9 0 0.0
Unmarried children 771 47.3 555 51.8 216 38.6
Married children 278 1 7. 1 121 11.3 157 28.1
Non-spouse/non-children
others 15 0.9 0 0.0 15 2.7
aSingle, divorced, separated and widowed.
EMHJ • Vol. 19 No. 12 • 2013 Eastern Mediterranean Health Journal
La Revue de Santé de la Méditerranée orientale
1032
nancial ability of the older adult to
keep his/her children close. In contrast,
studies conducted in the United States
[20] and in selected countries in the
Far East [21] have shown an inverse
relationship between co-residence and
socioeconomic status. e association
between lower socioeconomic status
and co-residence implies that pooling
of resources, in an aim to overcome
scarcity of nancial resources, is the
driving factor for such an arrangement.
In other words, higher per capita in-
come enables purchase of privacy. is
explanation is supported by our nding
that co-residence was more likely to
occur in Beirut compared with other
governorates. e high costs of living
associated with residing in the capital
[22] may explain these results. Our
ndings concur with those found in
Turkey, where living in metropolitan
areas was associated with co-residence
due to high costs of housing and living
[23,24]. e dynamics of living with
married children (i.e. who moved in
with whom, and why?) could not be
determined in the current study.
With regard to health-related factors,
this study showed that selected co-mor-
bid conditions, such as vascular disor-
ders, hypertension and communication
problems, were associated with living
with a married child, and the ndings
were only signicant among the unmar-
ried. An earlier study conducted in Beirut
showed that co-residence with a mar-
ried child was the most disadvantageous
among all living arrangements with the
highest all-cause and cardiovascular
disease mortality risk [14]. However, dif-
culty in performing ADL
in this study
Table 4 Age-adjusted and multivariate-adjusted odds ratios of sociodemographic characteristics of married and unmarried
adults aged ≥ 65 years
Sociodemographic
variable
Married Unmarried
a
Age-adjusted Mult ivariate adjusted Age-adjusted Multivariate adjusted
OR 95% CI OR 95% CI OR 95% CI OR 95% CI
Age (years)
65–79 Ref. Ref.
≥ 80 1.77b1.06–2.95 1.36 0.76–2.45 1.71b1.15–2.56 1.79b1.08–2.93
Sex
Male Ref. Ref.
Female 1.05 0.70–1.57 0.99 0.64–1.53 1.20 0.75 1.89 0.98 0.58–1.66
Region
Beirut Ref. Ref.
Other 0.50b0.30–0.82 0.36b0.20–0.63 0.44b0.28–0.68 0.40b0.24–0.67
No. of sons alive
0 Ref. Ref.
≥ 1 1.23 0.60–2.53 0.96 0.44–2.10 4.08b1.77–9.41 5.40b2.12–13.7
No. of daughters
alive
0 Ref. Ref.
≥ 1 2.23 0.92–5.37 2.07 0.82–5.26 1.80 0.93–3.48 0.85 0.40–1.83
Literacy
No formal
schooling Ref. Ref.
Other 0.76 0.51–1.13 0.91 0.59–1.42 0.76 0.51–1.13 1.16 0.72–1.87
Per capita income
(US$/month)
≤ 50 Ref. Ref.
50– ≤ 150 0.52b0.32–0.82 0.52b0.32–0.86 0.39b0.23–0.67 0.38b0.21–0.69
≥ 150 0.29b0.17–0.50 0.26b0.14–0.48 0.14b0.08–0.27 0.12b0.06–0.25
Government aid
No Ref. Ref.
Yes 2.02 0.78–5.23 2.19 0.81–5.90 1.94 0.77–4.93 2.31 0.80–6.65
aSingle, divorced, separated and widowed; bP < 0.05.
Ref. = reference category.
OR = odds ratio; CI = confidence interval.
1033
had no signicant relationship with
co-residence with married children.
e laer result is similar to ndings
in Lebanon from data using a crude
measure of disability among unmarried
older Lebanese females [25]. Follow-up
studies and in-depth interviews as well
as an objective assessment of the health
status of the older adult may assist in
interpreting such results.
Overall, the results of the study
showed an advantage for currently mar-
ried older adults over their unmarried
counterparts. e unmarried (mostly
widowed and divorced) were more like-
ly to host within their own households
their married children, and this living
arrangement did not indicate a positive
outcome, whether in terms of economic
assets or health indicators. e driv-
ing forces behind co-residence among
married older people were mainly of
socioeconomic origin, while among the
unmarried, these included additionally
health status of the older adult. Widowed
older adults tend to be more depend-
ent and have fewer choices when com-
pared with their married counterparts
[8,9,20,21,26,27]. Studies conducted
in Egypt revealed that older adults may
resume living with a child when they re-
quire nancial assistance, when they are
widowed, or when they need assistance
in ADL due to declining health [8].
Table 5 Age-adjusted and multivariate-adjusted odds ratios of health-related characteristics of married and unmarried adults
aged ≥ 65 years
Health-related variable Married Unmarried
a
Age-adjusted Multivariate adjusted Age-adjusted Mult ivariate adjusted
OR 95% CI OR 95% CI OR 95% CI OR 95% CI
Co-morbidities
Hypertension
No Ref. Ref.
Yes 1.22 0.82–1.81 1.09 0.70–1.69 1.67
b1.15–2.43 1.36 0.88–2.10
Vascular disorder
No Ref. Ref.
Yes 1.16 0.76–1.78 0.96 0.59–1.55 2.02b1.37–2.98 1.69b1.08–2.64
Diabetes mellitus
No Ref. Ref.
Yes 1.48 0.96–2.30 1.33 0.81–2.17 1.12 0.72–1.72 0.89 0.54–1.48
Pulmonary disorder
No Ref. Ref.
Yes 1.84 0.93–3.63 1.49 0.69–3.20 1.83 0.94–3.57 1.21 0.56–2.59
Activities of daily living
ADL score 1.10 0.96–1.26 0.98 0.82–1.16 1.16b1.02–1.32 1.02 0.86–1.22
Communication problems
Speech problems
No Ref. Ref.
Yes 1.45 0.68–3.09 1.02 0.41–2.57 3.52b1.85–6.72 4.19b1.71–10.2
Hearing problems
No Ref. Ref.
Yes 1.99b1.21–3.27 1.31 0.62–2.78 1.69b1.09–2.64 1.07 0.54–2.12
Sight problems
No Ref. Ref.
Yes 1.80b1.12–2.89 1.00 0.49–2.06 1.41 0.92–2.16 0.75 0.41–1.38
aSingle, divorced, separated and widowed; bP < 0.05.
Ref. = reference category; OR = odds ratio; CI = confidence interval.
Conclusion
e aim of this study was to dene
the correlates of living with married
children among older adults (aged ≥
65 years) in Lebanon. e ndings
revealed that a combination of demo-
graphic, socioeconomic and health-
related factors were correlated with
co-residence. Marital status of the older
adult appeared to aect co-residence
with a married child; being widowed
or divorced signicantly increased the
likelihood of co-residence. Among mar-
ried older adults, co-residence with a
married child was associated with lower
socioeconomic status, whereas among
EMHJ • Vol. 19 No. 12 • 2013 Eastern Mediterranean Health Journal
La Revue de Santé de la Méditerranée orientale
1034
the unmarried, availability of sons and
additionally declining health were as-
sociated with co-residence with a mar-
ried child. While solitary living among
older people has been traditionally the
focus of aention by policy-makers
and service providers in Lebanon [28],
the study ndings showed that those
living with a married child may also be
a vulnerable group. Such information
would help policy-makers in the coun-
try to design intervention measures
towards achieving the Millennium
Development Goals, especially in the
area of poverty reduction and equal
opportunities of development for older
adults.
Competing interests: None declared
References
1. Sibai AM et al. Population ageing in Lebanon: current status,
future prospects and implications for policy. Bulletin of the
World Health Organization, 2004, 82:219–225.
2. Sibai A, Yamout R. Family-based old-age care in Arab coun-
tries: between tradition and modernity (63–76). In: Groth H,
Sousa-Poza A, eds. Population dynamics in Muslim countries:
assembling the jigsaw. Heidelberg, Springer Berlin, 2012.
3. Chaudhuri A, Roy K. Gender differences in living arrangements
among older persons in India. Journal of Asian and African Stud-
ies, 2007, 44(3):259–277.
4. Shah NM et al. Decline in co-residence of parents and chil-
dren among older Kuwaiti men and women: what are the
significant correlates? Journal of Cross-Cultural Gerontology,
2011, 26:157–174.
5. Department of Economic and Social Affairs, Population Divi-
sion. Living arrangements of older persons around the world.
New York, United Nations, 2005.
6. Department of Economic and Social Affairs, Population Divi-
sion. World economic and social survey 2007: development in an
ageing world. New York, United Nations, 2007.
7. Khadr Z. Living arrangements and social support systems of the
older population in Egypt [doctoral thesis]. Ann Arbor, Michi-
gan, University of Michigan, 1997.
8. Yount K. The patriarchal bargain and intergenerational co-
residence in Egypt. Sociological Quarterly, 2005, 46:139–166.
9. Bongaarts J, Zimmer Z. Living arrangements of older adults in the
developing world: an analysis of DHS household surveys. New
York, Population Council, 2001 (Working Paper, No.148).
10. De Vos S, Holden K. Measures comparing living arrangements
of the elderly: an assessment. Population and Development
Review, 1998, 14:688–704.
11. Pal S. Effects of intergenerational transfers on elderly coresidence
with adult children: evidence from rural India. Bonn, Germany, In-
stitute for the Study of Labor, 2007 (Discussion Paper No. 2847).
12. VanWey L, Cebulko K. Intergenerational coresidence among
small farmers in Brazilian Amazonia. Journal of Marriage and
the Family, 2007, 69:1257–1270.
13. Yount K, Khadr Z. Gender, social change, and living arrange-
ments among older Egyptians during the 1990s. Population
Research and Policy Review, 2008, 27:201–225.
14. Sibai AM, Yount KM, Fletcher A. Marital status, intergenera-
tional co-residence and cardiovascular and all-cause mortality
among middle-aged and older men and women during war-
time in Beirut: gains and liabilities. Social Science and Medicine,
2007, 64:64–76.
15. Mehio-Sibai A, Beydoun MA, Tohme RA. Living arrangements
of ever-married older Lebanese women: is living with married
children advantageous? Journal of Cross-Cultural Gerontology,
2009, 24:5–17.
16. The Pan Arab Project for Family Health. Lebanon family health
survey 2004: Principal report. Ministry of Social Affairs, Lebanon
2006 (http://www.pdslebanon.org/UserFiles/File/Leba-
non%20Report%20English1.pdf, accessed 6 November 2013).
17. Katz S et al. Studies of illness in the aged. The index of ADL: A
standardized measure of biological and psychosocial function.
Journal of the American Medical Association, 1963, 185:914–919.
18. De Jong Gierveld J, Dykstra P, Schenk N. Living arrangements,
intergenerational support types and older adult loneliness in
Eastern and Western Europe. Demographic Research, 2012,
27:167–200.
19. Shah NM et al. Living arrangements of older women and
men in Kuwait. Journal of Cross-Cultural Gerontology, 2002,
17:337–355.
20. Elman C, Uhlenberg P. Coresidence in the early twentieth
century: Elderly women in the United States and their children.
Population Studies, 1995, 49:501–517.
21. DaVanzo J, Chan A. Living arrangements of older Malaysians:
who coresides with their adult children? Demography, 1994,
31:95–113.
22. Laithy H, Abu-Ismail K, Hamdan K. Poverty, growth and income
distribution in Lebanon. Brasilia, Brazil, International Poverty
Centre, 2008 (Country Study No.13).
23. Aykan H, Wolf D. Traditionality, modernity, and household
composition: Parent-child coresidence in contemporary Tur-
key. Research on Aging, 2000, 22:395–421.
24. Autaç IA. Intergenerational living arrangements in Turkey. Jour-
nal of Cross-Cultural Gerontology, 1998, 13:241–264.
25. Obeid M. Living arrangements of Lebanese unmarried elderly
women: evidence from the population and housing survey 1996.
[Master’s thesis]. Beirut, Lebanon American University of Bei-
rut, 2004.
26. Grau MA. Residence patterns of aged widows in three Medi-
terranean communities and the organization of the care. His-
tory of the Family, 2002, 7:157–173.
27. Chen F. Residential patterns of parents and their married chil-
dren in contemporary Chine: A life course approach. Popula-
tion Research and Policy Review, 2005, 24:125–148.
28. Tohme R et al. Socioeconomic resources and living arrange-
ments of older adults in Lebanon: who chooses to live alone?
Ageing and Society, 2011, 31:1–17.
