Post on 11-Sep-2021
transcript
Schatz et al. 2015 Older person’s living arrangements and health
1
Older person’s living arrangements, quality of life and disability in rural South Africa:
Confirming social positioning?
Enid Schatz, University of Missouri
Margaret Ralston, Mississippi State
Sangeetha Madhavan, University of Maryland
Don Willis, University of Missouri
F. Xavier Gomez-Olive, University of the Witwatersrand
Mark Collinson, University of the Witwatersrand
Prepared for Submission to:
Population Association of America Meeting
Spring 2016
**Please DO NOT cite without author’s permission**
Schatz et al. 2015 Older person’s living arrangements and health
2
Short abstract:
While older persons are usually regarded as dependent household members, we believe that
living arrangements are the result of and/or result in older people being dependent on those with
whom they live in some cases, and taking on productive roles in others. In this paper, we extend
a typology established in previous work of older persons’ living arrangements based on social
positioning to examine associations between older person’s social positioning and health. Using
2010 survey and census data from Agincourt, South Africa, we provide evidence that older
persons in “productive” arrangements on average report worse quality of life and higher levels of
disability than older persons in “dependent” arrangements (two generation, linear linked).
Further, within each category women report worse outcomes than men. However, when
controlling for a number of individual and household characteristics, living arrangements are no
longer significantly associated with differences in quality of life or disability.
Schatz et al. 2015 Older person’s living arrangements and health
3
Long Abstract:
In earlier work, we have argued that the social positioning of older persons in their households is
not homogenous (Schatz et al. 2014). Despite the common assumption that older persons are
dependent members of households, our work begins to explore whether there are ways that some
older South Africans instead play productive roles in their households. We have established a
typology of older persons’ households that outlines how older persons’ social position differs
due to the living arrangements, related to household membership. We believe that different
living arrangements are the result of and/or result in older people being dependent on those with
whom they live in some arrangements, and older person taking on active and productive roles in
other arrangements.
In places like South Africa, where there is a fairly generous non-contributory government
sponsored old-age pension, older person’s often use this pension to support not only themselves,
but also their family’s needs. There is substantial evidence from South Africa that older persons
pool their pensions with their households, and that this sharing results in better health of all
household members (Ardington et al. 2010; Burns, Keswell, and Leibbrandt 2005; Case and
Deaton 1998; Duflo 2003; May 2003). Thus, older persons are playing financially productive
roles in their households. Further caregiving roles for those sick with or orphaned by HIV/AIDS,
taken on mainly by older women also can be read as active and productive household
contributions (Bohman, van Wyk, and Ekman 2011; Boon et al. 2010; Schatz 2007; Schatz and
Seeley 2015). However, whether push or pull reasons dominate older persons’ taking on
productive or dependent roles is not always clear, and thus uncertainty remains as to what the
relationship between health and living arrangements might be.
Schatz et al. 2015 Older person’s living arrangements and health
4
Being a productive member may mean providing carework and other types of work
(gardening, cleaning) to the household, or contributing financial resources to the household
whether through wages or private or government-funded pensions. Older persons may end up in
productive roles because (a) they are in good health, (b) they are needed as substitutes for
mothers when women migrate (Madhavan et al 2012), (c) they have pension income to pool in
the household, or (d) some combination of all of these. Older persons who are in these
productive roles may or may not desire to be in these roles, and may or may not find them
rewarding; the engagement and feel of being needed could lead to better physical and mental
health. Being a dependent member also has a number of possible reasons and implications.
Dependency may be a result of being in poor health and needing care. Dependency also may be a
result of having children who have resources and allow the older person to ‘enjoy the leisure’ of
old-age. Thus, the quality of life of a dependent older person could be poor due to poor health, or
could be excellent due to feelings of being cared for, physically or emotionally.
In order to investigate further our earlier typology of rural South African households, we
use a cross-section of census and survey data from the Agincourt Health and socio-Demographic
Surveillance site to assess the association in 2010 between older persons living arrangements and
health by making use of two World Health Organization composite measures—one focused on
quality of life and the other on disability. Each of these measures provides insight into how older
persons report their health and wellbeing in this setting.
Households as a Social Environment
Health and wellbeing are often considered to be the result of interactions between
individuals and their environment. Living arrangements and kin play an important role in
Schatz et al. 2015 Older person’s living arrangements and health
5
creating one’s social environment and support systems through social roles, norms, histories, and
emotions, as well as the household economy (Hughes and Waite 2002). The impact can be
positive, but it is just as possible that excess claims on kinship obligations can be burdensome
(Portes 1998). The roles and expectations placed on household members differ in terms of the
care and resources they provide or are provided with. Some household members provide more
resources to their household than they receive in return, and others receive more than they give.
In many cases the expectation is for downward flows of resources to support children when they
are young (Caldwell and Caldwell 1993; Goody 1982), but with an upward flow in old age.
Political, social, economic, and cultural factors may shift these flows of resources. It is important
to also take into account household composition as a possible proxy for economic resources.
Households with multiple income earners have an economic advantage over single-earner
households that often translates into a health advantage (Hughes and Waite 2002). Thus, the
influence of living arrangements on health is closely tied to the way those arrangements pool or
drain resources from the household itself. Further, the uneven distribution of demands and
resources across household members may result in different health and well-being outcomes for
each household member depending on one’s expected role (Hughes and Waite 2002).
Living Arrangements and Health
Results from research examining the impact of living arrangements on health and well-
being remain mixed depending on place, group, and which particular measure of health and well-
being is focused on (Hays 2002). While certain living arrangements have a protective effect, the
type of living arrangement that is protective varies quite based on the specifics of each study.
In high-income settings, extensive research has shown the positive health effects of
marriage (living with a spouse) for men, with less positive outcomes for women (Koball et al.
Schatz et al. 2015 Older person’s living arrangements and health
6
2010; Pienta, Hayward, and Jenkins 2000). Among older adults, Michael et al. (2001) found that
older women who live alone in the United States had lower risk of decline in mental health and
vitality. Yet, other work has found that women in late adulthood (age 51-61) report better self-
rated health when living with only their husband, or with husband and children, than when living
in any other arrangements (Hughes and Waite 2002). Moreover, instrumental support with daily
activities from children, can have an entirely different associations with health of older
individuals depending on living arrangement and marital status. While instrumental support of
the general elderly in Spain is associated with poor self-rated health and high levels of
depression, which is likely due to the poor underlying functional abilities of those who receive
such support, the opposite is true for elderly widow(er)s who live alone (Zunzunegui, Béland,
and Otero 2001). In other words, living arrangements and marital status appear to moderate the
relationship between some forms of support and the self-rated health and levels of depression
among Spanish elderly.
Norms related to living with adult children differ greatly across low and middle-income
countries (LMIC); living with adult children is less common in African countries than in Asia
(Bongaarts and Zimmer 2002). These norms may influence the way that living arrangements are
associated with older people’s health. Among South Korean elderly (age 65 or older) with
physical disabilities, those living with a spouse reported better life satisfaction than those living
with others or living alone (Kim, Hong, and Kim 2014). Additional evidence from Korea also
shows significantly better physical health status, self-esteem, and family support among those
who live with family compared to those who live alone (Sok and Yun 2011). With a sample
spanning fifteen countries across sub-Saharan Africa, McKinnon, Harper, and Moore (2013) find
that living with children, regardless of whether they are of working age, offers protection against
Schatz et al. 2015 Older person’s living arrangements and health
7
depressive symptoms for individuals age 50 or older. However, a study of Demographic and
Health Surveys from 22 African countries showed that HIV is impacting household living
arrangements, with greater numbers of older people living alone in high-prevalence countries,
and potentially decreasing their familial support and increasing the care they must provide to
others (Kautz et al. 2010).
The South African context, however, differs significantly from many other places, even
within sub-Saharan Africa, due to the rising morbidity and or mortality among migrant workers
related to HIV/AIDS, the impact this has on elders who become caregivers, and the influence of
a non-contributory pension program for those in old age (Bohman et al. 2011; Case and Deaton
1998; Schatz and Ogunmefun 2007; Ssengonzi 2009). Qualitative work of elderly households
suggests that within the context of a population deeply affected by HIV/AIDS a great deal of
resources are directed to the younger generations with HIV/AIDS or their vulnerable children
affected by the disease who are living with the elderly, placing greater demands on the aging
population and elderly women in particular (Schatz 2007; Schatz and Ogunmefun 2007;
Ssengonzi 2009). Thus, the major beneficiaries of social programs such as the old-age pension
program in South Africa may be those for whom the elderly are providing care and support,
namely persons living with HIV/AIDS and/or orphans and vulnerable children affected by
HIV/AIDS (Case and Menendez 2007; Duflo 2003; Schatz and Williams 2012) Moreover, HIV-
related care giving appears to result in a perceived cost to the emotional, physical, and
psychological health of elderly caregivers (Schatz and Seeley 2015b; Ssengonzi 2009).
Living arrangements can be a double-edged sword for certain households—helping some
members and burdening others. Particularly for a region impacted by HIV/AIDS, the flow of
resources between generations seems to be a reversal of what is seen in places where the disease
Schatz et al. 2015 Older person’s living arrangements and health
8
is absent. Moreover, this demand for care often initiates a drastic disruption in the living
arrangements of the elderly, resulting in widespread consequences ranging from crowded
sleeping arrangement, abandoned gardens (sources of supplemental nutrition), selling off
personal property, and a negative impact on social engagement and relationships—particularly
marital relations for which prolonged absences related to care giving were straining, infusing
them with misunderstanding and distrust (Ssengonzi 2009). This is much different from work
focused on the U.S. family that highlights these multigenerational households wherein both
children and parents are adults resulting in relationships that have, in later life, become more like
“friendships” (Blieszner and Mancini 1987), or instances when children report a role-reversal
wherein they have become the primary caretakers of their parents (Fischer 1985).
This paper further extends existing research by exploring associations between particular
types of living arrangements and older persons’ health and wellbeing through measures of
quality of life and disability.
DATA & METHODS
We use data from the Agincourt Health and socio-Demographic Surveillance System
(Agincourt HDSS) census along with the 2010 World Health Organization Study of Global
Aging and Adult Health survey (WHO-SAGE). The census, run by the MRC/University of the
Witwatersrand Rural Public Health and Health Transitions Unit (Tollman, Director), has
collected data annually from all households in the Agincourt sub-district since 1992. As of 2010,
the site covered 27 villages—approximately 15,600 households and 89,000 individuals. In 2010,
the Agincourt HDSS collected health and wellbeing data on persons over the age of 50 through
an abbreviated WHO-SAGE survey. The instrument contained two modules adapted from the
Schatz et al. 2015 Older person’s living arrangements and health
9
full WHO-SAGE questionnaire: Health Status and Activities of Daily Living (following the
WHO Disability Assessment Scale version II (WHODAS-II) model), and Subjective Wellbeing.
Approximately 60 per cent of the target population completed the questionnaire with only 0.4 per
cent refusing. Others were either not found (35%), ineligible (4%) or dead (1.6%). The resulting
sample contains 5,980 individuals age 50 and above, about 25% male and 75% female.
Variables
Table 1 describes the living arrangement typology and the health and wellbeing variables.
In previous work we created a typology of living arrangements that includes the four categories
described in Table 1 (Schatz et al. 2014). In this paper we reduce these categories into two
groups: productive older person households and dependent older person households. Older
persons living in single generation households and those in complex linked multigeneration
households fall into the category of “productive arrangements.” Older persons in two generation
and linear linked multigenerational households are in “dependent arrangements.” Older persons
living in single generation households are considered productive because they are not able to
depend on younger household members. Complex linked multigenerational are households in
which older persons, particularly pensioners, may need to take on more of a productive role. The
productive role may include financial contributions, whether from pensions or from income-
generating activities. In addition, productive roles may be in the form of physical and in-kind
support, such as providing care for the sick or young. Complex linked multigenerational
households have additional individuals who may be seeking assistance from other productive
household members. In two-generation households, we expect that for the most part the older
person is the parent(s) of the other generation in the household, and thus can depend on them for
financial and physical support. In linear linked multigenerational households older persons are
Schatz et al. 2015 Older person’s living arrangements and health
10
also more likely to be dependents of the household because the head of household is likely to be
their son (or daughter) who would assume primary responsibility for caregiving and financial
provision. Age is an important factor in household living arrangements, making it important to
control for this in regressions below—a larger proportion of those living in single generation
households are in the older age categories (70+), and in two generation households, a larger
proportion are in the youngest age category (50-59).
In order to explore health and wellbeing, we look at quality of life and disability. These
variables are WHO-constructed composite measures; each measure is based on multiple
questions in the WHO-SAGE survey and converted to a 0-100 scale. (1) The WHOQoL (World
Health Organization Quality of Life measure) is based on questions on self-rated general health
and questions on satisfaction. The World Health Organization defines quality of life as “the
individual’s perception of their position in life in the context of the culture and value systems in
which they live and in relation to their goals, expectations, standards and concerns.” (2) The
WHODAS II (World Health Organization Disability Assessment Schedule II) scale assesses day-
to-day functioning in six activity domains. Ten questions assess individuals’ difficulty
performing certain activities during the past 30 days.
[Table 1 about here]
We explore demographic, individual characteristics and other household characteristics
of the population. They include household size, percent of household under 15, percent with
orphan in household, percent with foster child in household, socio-economic status (SES),
education, employment status, nationality of origin, and self-reported health. SES is determined
Schatz et al. 2015 Older person’s living arrangements and health
11
from a household asset score derived from 34 variables collected in 2009 (including information
about the type and size of dwelling, access to water and electricity appliances and livestock
owned and transport available). The score was derived through principal component factor
analysis and then divided into quintiles (Gómez-Olivé et al. 2010). Education is categorized as
no formal education or some education. Employment status, collected in 2008 is coded as
currently working or not. The majority of those not working were not looking for work but had
retired, having concluded their working career. Employment status focused on those with
permanent formal work, so may not capture those doing informal income-generating activities.
“South African” captures self-identification as South African or Mozambican. Self-rated health
is categorized as “bad” or not.
Analysis
We first present descriptive statistics by living arrangement to explore the nature and
strength of the relationship between key household and individual characteristics and living
arrangements. We then take a look descriptively at potential differences in WHOQOL and
WHODAS by living arrangements. Because of important differences in percent female in living
arrangement categories, and gender differences in reporting on health and wellbeing, we examine
these relationships separately for men and women. Finally we examine whether a relationship
between WHOQOL and living arrangements, and WHODAS and living arrangements remain in
OLS regression models with and without individual and household control variables, clustering
by household. We have limited our sample to respondents who are in productive or dependent
households, leaving a sample N 4703 individuals in 4487 households. Earlier work included an
ambiguous “other” household category, we have dropped those households in this paper to be
able to directly compare productive to dependent households.
Schatz et al. 2015 Older person’s living arrangements and health
12
PRELIMINARY RESULTS
Table 2. shows background characteristics and health and wellbeing indictors by living
arrangements for persons aged 50 plus. Over half (56%) of the sample lived in complex linked
multigenerational, signaling that the majority of older persons in Agincourt live in household
arrangements where they are likely to be productive members. Another 12% of older persons
live in single generation households. While the latter do no contain any children, complex linked
multigenerational have the highest percent of foster and orphan children. Single generation
households have the highest percentage in the lowest socioeconomic status quintile and the
lowest representation in to be in the highest quintile, and the highest percentage of older people
reporting bad self-rated health. Two-generation households have the highest percentage of older
people reporting to be currently working, and lowest percentage reporting bad self-rated health.
Linear linked multigenerational households, where older adults are theorized to be dependent
members of the household, have the highest percentage in the highest socioeconomic status
quintile. Thus, from these descriptive statistics, it appears that those in ‘productive’ arrangements
are at a disadvantage on in a number of realms: older, poorer self-reported health, worse socio-
economic status, and worse off in terms of both quality of life and reported disability.
[Table 2 about here]
Table 3 shows the relationship between living arrangements and three health and
wellbeing variables by sex for persons 50 years or older. We include percent reporting bad self-
rated health as well as the quality of life and disability variables. As in Table 2, there is
consistency among these measures, with higher percentages of poor self-rated health being
reported in groups also reporting worse quality of life and higher levels of disability. For women,
Schatz et al. 2015 Older person’s living arrangements and health
13
being in a productive living arrangement is associated with reporting worse quality of life and
higher levels of disability, as well as a higher percentage of women reporting bad self-rated
health. The difference is statistically significance for all three measures for women. Similarly,
difference between productive and dependent households for men is statistically significance for
two of the measures. Among men being in a “productive” arrangement was associated with
worse health, quality of life and disability outcomes.
[Table 3 about here]
Tables 4a and 4b show results for OLS regression of living arrangements (and individual
and household controls) on WHOQOL and on WHODAS, with separate regressions for men
and women, clustering by household. The tables display unstandardized and standardized
regression coefficient and robust standard errors. In Table 4a [WHOQOL], Model 1, the
univariate model, being in a productive arrangement is associated with significantly worse
reports of quality of life for women (increasing on the 0-100 scale is actually ‘worse’
WHOQOL). While the coefficient goes in the same direction for men, the relationship is not
significant. When we add control variables in Model 2, productive arrangements no longer
are significant. For women, smaller household size, younger age, having some education
and not currently working are associated with significantly better quality of life. Reporting
bad self-rated health is related to worse quality of life; this is has the strongest effect in the
model (beta=.367). For men, fewer of the controls have a meaningful relationship with
WHOQOL. Having some education and not currently working are associated with
significantly better quality of life, while bad self-rated health again has a strong association
(beta=.387) with reports of worse quality of life.
[Table 4a about here]
Schatz et al. 2015 Older person’s living arrangements and health
14
In Table 4b [WHODAS], Model 1, being in a productive arrangement has a significant
relationship with reporting higher levels of disability for both women and men; this
relationship appears to be even stronger than the one with WHOQOL. Among older women,
having some education, being younger and being Mozambican (as opposed to South
African) are associated with reports of lower levels of disability. Among older men, the
relationship between productive arrangements and disability is muted by the controls in
Model 2. Having some education, being younger and currently working make it less likely
that men report disability; it is likely that men who report higher levels of disability are less
able to work. It is unclear why being in the second lowest SES category is associated with
significantly higher likelihood of reporting higher levels of disability, but not for other low
SES categories. We would like to explore this further.
[Table 4b about here]
DISCUSSION & NEXT STEPS
In this paper, we provide evidence that older persons in “productive” arrangements on average
report worse quality of life and higher levels of disability than older persons in “dependent”
arrangements (two generation, linear linked). Further, within each category women report worse
outcomes than men. However, these relationships are muted for both men and women when
adding individual and household level controls to regression models. Research from other
settings has also suggested there is no difference in health and wellbeing of older persons
depending on living arrangements themselves, but that the meaningful difference lies between
those that are in living arrangements in concordance with their preference and those that are not
(Sereny 2011). This perspective emphasizes the need for a good fit between the individual and
Schatz et al. 2015 Older person’s living arrangements and health
15
their home environment/living arrangement, rather than emphasizing one particular arrangement
as universally ideal. We hope to explore this further with qualitative data (both existing and new
data collection) that will help to understand how and why older persons end up in particular
living arrangements, their assessment of their social role, and how their social role impacts their
views on their health, quality of life and level of disability, and how their health impacts their
social role in the household.
Where we do find significant relationships (even if only in bivariate models), we still
cannot say which direction the causality is occurring as the data we are analyzing are cross-
sectional. Thus, we cannot say if it is living ‘productive’ arrangements that lead to worse health
and wellbeing outcomes, or if there is selection into these households such that those who are
already worse off end up in productive arrangements. It is interesting that older persons in
‘dependent’ arrangements on average report better health, better quality of life, and less
disability. This might say something about their selection into these households, their being taken
care of in their old age, or the connection between care-networks and perceptions of health and
wellbeing. These are all issues we hope to explore further in future iterations of this paper and
with other data from the site.
While it is clear that there is some association between living arrangements and health
outcomes, we would like to do additional model testing to explore which set of predictors are
contributing to the muting of this relationship, as well as interaction terms to see if there are
differences when we look at older/younger or richer/poorer individuals in productive versus
dependent arrangements. Perhaps there is some combination of these that are particularly salient
while others are not. Further, we plan to make use of longitudinal data on a limited sample for
which we have health data from 2006 to see if controlling for poor health at a previous time
Schatz et al. 2015 Older person’s living arrangements and health
16
period affects the relationship between the 2010 quality of life and disability variables and living
arrangements.
Schatz et al. 2015 Older person’s living arrangements and health
17
References
Ardington, Cally et al. 2010. “The Impact of AIDS on Intergenerational Support in South Africa:
Evidence From the Cape Area Panel Study.” Research on Aging 32(1):97–121.
Blieszner, Rosemary and Jay A. Mancini. 1987. “Enduring Ties: Older Adults’ Parental Role and
Responsibilities.” Family Relations 36(2):176–80.
Bohman, D., Neltjie C. van Wyk, and Sirkka-Liisa Ekman. 2011. “South Africans’ Experiences
of Being Old and of Care and Caring in a Transitional Period.” International Journal of
Older People Nursing 6(3):187–95.
Bongaarts, John and Zachary Zimmer. 2002. “Living Arrangements of Older Adults in the
Developing World: An Analysis of Demographic and Health Survey Household
Surveys.” Journal of Gerontology: Social Sciences 57B(3):S145–57.
Boon, Hermien et al. 2010. “Correlates of Grief Among Older Adults Caring for Children and
Grandchildren as a Consequence of HIV and AIDS in South Africa.” Journal of Aging
and Health 22(1):48–67.
Burns, Justine, Malcolm Keswell, and Murray Leibbrandt. 2005. “Social Assistance, Gender, and
the Aged in South Africa.” Feminist Economics 11(2):103–15.
Caldwell, John and Pat Caldwell. 1993. “The Nature and Limits of the Sub-Saharan African
AIDS Epidemic: Evidence from Geographic and Other Patterns.” Population and
Development Review 19(4):817–48.
Case, A. and A. Deaton. 1998. “Large Cash Transfers to the Elderly in South Africa.” The
Economic Journal 108(September):1330–61.
Case, Anne and Alicia Menendez. 2007. “Does Money Empower the Elderly? Evidence from the
Agincourt Demographic Surveillance Site, South Africa1.” Scandinavian Journal of
Public Health 35(69 suppl):157–64.
Duflo, E. 2003. “Grandmothers and Granddaughters: Old-Age Pension and Intra-Household
Allocation in South Africa.” World Bank Economic Review 17(1-25).
Fischer, Lucy R. 1985. "Elderly Parents and the Caregiving Role: An Asymmetrical Transition."
Pp. 105-114 in Warren A. Peterson and Jill Quadagno (eds.), Social Bonds in Later Life.
Beverly Hills, CA: Sage.
Goody, Esther N. 1982. Parenthood and Social Reproduction: Fostering and Occupational
Roles in West Africa. Cambridge [Cambridgeshire]: Cambridge University Press.
Hays, Judith C. 2002. “Living Arrangements and Health Status in Later Life: A Review of
Recent Literature.” Public Health Nursing (Boston, Mass.) 19(2):136–51.
Schatz et al. 2015 Older person’s living arrangements and health
18
Hughes, MARY ELIZABETH and LINDA J. Waite. 2002. “Health in Household Context:
Living Arrangements and Health in Late Middle Age.” Journal of health and social
behavior 43(1):1–21.
Kautz, T. et al., 2010. AIDS and declining support for dependent elderly people in Africa: retrospective analysis using demographic and health surveys. BMJ (Clinical Research Ed.), 340, p.c2841.
Kim, Hyun-Jun, Seunghye Hong, and Miok Kim. 2014. “Living Arrangement, Social
Connectedness, and Life Satisfaction among Korean Older Adults with Physical
Disabilities: The Results from the National Survey on Persons with Disabilities.” Journal
of Developmental and Physical Disabilities 27(3):307–21.
Koball, Heather L., Emily Moiduddin, Jamila Henderson, Brian Goesling, and Melanie
Besculides. 2010. “What Do We Know about the Link between Marriage and Health?”
Journal of Family Issues 31(8):1019–40.
Madhavan, S. et al., 2012. Child mobility, maternal status, and household composition in rural South Africa. Demography, 49(2), pp.699–718.
May, Julian. 2003. “Chronic Poverty and Older People in South Africa.” Chronic Poverty
Research Centre Working paper 25.
McKinnon, Brittany, Sam Harper, and Spencer Moore. 2013. “The Relationship of Living
Arrangements and Depressive Symptoms among Older Adults in Sub-Saharan Africa.”
BMC Public Health 13(1):682.
Michael, Yvonne L., Lisa F. Berkman, Graham A. Colditz, and Ichiro Kawachi. 2001. “Living
Arrangements, Social Integration, and Change in Functional Health Status.” American
Journal of Epidemiology 153(2):123–31.
Pienta, Amy Mehraban, Mark D. Hayward, and Kristi Rahrig Jenkins. 2000. “Health
Consequences of Marriage for the Retirement Years.” Journal of Family Issues
21(5):559–86.
Schatz, Enid. 2007. “‘Taking Care of My Own Blood’: Older Women’s Relationships to Their
Households in Rural South Africa.” Scandinavian Journal of Public Health 35(Suppl
69):147–54.
Schatz, Enid, Sangeetha Madhavan, Mark Collinson, F. Xavier Gómez-Olivé, and Margaret
Ralston. 2014. “Dependent or Productive? A New Approach to Understanding the Social
Positioning of Older South Africans through Living Arrangements.” Research on Aging
0164027514545976.
Schatz, Enid and Catherine Ogunmefun. 2007. “Caring and Contributing: The Role of Older
Women in Rural South African Multi-Generational Households in the HIV/AIDS Era.”
World Development 35(8):1390–1403.
Schatz et al. 2015 Older person’s living arrangements and health
19
Schatz, Enid and Janet Seeley. 2015, forthcoming. “Gender, Ageing and Carework in East and
Southern Africa: A Review.” Global Public Health.
Schatz, Enid and Jill Williams. 2012. “Measuring Gender and Reproductive Health in Africa
Using Demographic and Health Surveys: The Need for Mixed-Methods Research.”
Culture, health & sexuality 14(7):811–26.
Sereny, Melanie. 2011. “Living Arrangements of Older Adults in China: The Interplay Among
Preferences, Realities, and Health.” Research on Aging 33(2):172–204.
Sok, Sohyune R. and Eun K. Yun. 2011. “A Comparison of Physical Health Status, Self-Esteem,
Family Support and Health-Promoting Behaviours between Aged Living Alone and
Living with Family in Korea.” Journal of Clinical Nursing 20(11-12):1606–12.
Ssengonzi, R. 2009a. “The Impact of HIV/AIDS on the Living Arrangements and Well-Being of
Elderly Caregivers in Rural Uganda.” AIDS Care 21(3):309–14.
Ssengonzi, R. 2009b. “The Impact of HIV/AIDS on the Living Arrangements and Well-Being of
Elderly Caregivers in Rural Uganda.” AIDS Care 21(3):309–14.
Zunzunegui, M. V., F. Béland, and A. Otero. 2001. “Support from Children, Living
Arrangements, Self-Rated Health and Depressive Symptoms of Older People in Spain.”
International Journal of Epidemiology 30(5):1090–99.
Schatz et al. 2015 Older person’s living arrangements and health
20
Table 1. Descriptions of Household Typology and Dependent variables
Living arrangements
Productive Older Person Households
Single-generation Households include single persons, couples, and siblings living together.
Complex linked
multigeneration
Households include an older household head’s unmarried children or
fostered/orphaned grandchildren, and among younger heads, their siblings,
nieces/nephews, and/or aunts/ uncles, and/or (parents/daughters/sons)-in-
law. Skipped generation households (parental generation is missing) are also
included.
Dependent Older Person Households
Two generation Households include a head, his or her spouse, and children (or parents of the
head) and also includes single-parent households and those with step
children.
Linear linked
multigeneration
Households includes those in which (1) there is no break in generations and
(2) the middle generation is comprised of a married couple in the traditional
‘‘productive’’ age category (ages 15–49).
Health and wellbeing
WHOQOL World Health
Organization Quality of
Life (WHOQOL)
Enough energy for daily life
Enough money to meet needs
Satisfaction with:
Your health
Yourself
Ability to perform daily activities
Personal; relationships
Condition of your living place
Rate your overall quality of life
0 (high quality of life) to 100 (low quality of life)
World Health
Organization Disability
Assessment Schedule II
(WHODAS)
Interpersonal activities
Difficulties in daily living:
Standing
Walking
Household duties
Learning
Concentrating
Self-care
0 (high ability) to 100 (low ability)
Schatz et al. 2015 Older person’s living arrangements and health
21
Table 2. Background characteristics and health and wellbeing by living arrangements for persons aged 50 plus in 2010, Agincourt HDSS and SAGE
Single
generation
Two generation Linear linked
multigenerational
Complex linked
multigenerational
Dependent
Older HH
Productive
Older HH
Total
Household characteristics
Mean household size 1.33 4.51 10.16 7.58 8.02 6.47 6.97
Mean percent of
household under 15
0 10.42 30.65 29.86 22.98 24.58 24.06
Percent with orphan in
HH
0 1.57 12.53 16.21 8.37 13.33 11.74
Percent with foster
child in HH
0 0.87 25.05 27.92 15.89 22.98 20.69
Socioeconomic status
(quintiles)
First (lowest) 44.20 18.09 8.64 12.83 12.18 18.25 16.29
Second 23.57 21.45 17.70 19.90 19.11 20.53 20.07
Third 18.05 21.10 20.47 23.82 20.71 22.82 22.14
Fourth 8.10 17.73 22.17 19.90 20.51 17.86 18.72
Fifth (highest) 6.08 21.63 31.02 23.55 27.50 20.53 22.78
Individual characteristics
Percent Female 55.32 61.22 76.01 77.69 70.40 73.73 72.66
Five year age group
50-54 12.23 30.43 14.65 17.66 20.63 16.70 17.97
55-59 12.94 22.96 19.85 16.67 21.03 16.01 17.63
60-64 14.54 14.43 17.62 16.13 16.41 15.85 16.03
65-69 10.64 7.48 15.29 14.84 12.33 14.09 13.52
70-74 15.60 9.57 13.06 13.39 11.73 13.78 13.12
75plus 34.04 15.13 19.53 21.32 17.86 23.57 21.73
Percent no formal
education
69.46 48.74 65.16 58.46 59.00 60.37 59.93
Percent currently
working
17.55 28.00 19.85 18.96 22.94 18.71 20.07
Percent South African 71.01 68.52 65.07 75.21 66.38 74.48 71.87
Percent bad self-rated
health
21.81 15.01 15.97 18.54 15.61 19.12 17.99
Mean WHOQOL 48.31 46.23 45.97 47.24 46.07 47.43 46.99
Mean WHODAS 24.10 19.32 19.73 21.73 19.57 22.15 21.32
N (% of total) 564 (11.99) 575 (12.23) 942 (20.03) 2622 (55.75) 1517 (32.26) 3186 (67.74) 4703 (100)
Schatz et al. 2015 Older person’s living arrangements and health
22
Table 3. Health and Wellbeing by living arrangements for WOMEN and MEN separately age 50 plus in 2010, Agincourt HDSS and SAGE1
Dependent HH Productive HH P-value Total
Women
Percent bad self-rated health 16.81 20.53 .011 19.37
Mean WHOQOL 46.75 47.74 .015 47.43
Mean WHODAS 20.85 23.06 .001 22.37
N 1,068 2,349 3,417
Men
Percent bad self-rated health 12.75 15.27 .238 14.33
Mean WHOQOL 44.45 46.55 .003 45.81
Mean WHODAS 16.52 19.60 .003 18.53
N 449 837 1,286
1Displays results for mean significant difference tests between the two living arrangement categories.
Schatz et al. 2015 Older person’s living arrangements and health
23
Table 4a. Regression models for WHOQOL run separated by sex
Women Men
Model 1 Beta Model 2 beta Model 1 beta Model 2 beta
Productive Arrangement .989* (.408) .042 .217 (.389) .009 1.865 (.705) .076 1.18 (.658) .048
Household Size -.147** (.050) -.052 -.054 (.085) -.020
Socioeconomic status (quintiles)
First (lowest) .443 (.661) .015 1.13 (1.088) .036
Second .363 (.549) .013 .882 (1.000) .029
Third .580 (.530) .022 1.519 (.856) .053
Fourth -.213 (.562) -.008 1.086 (.954) .035
Fifth (highest) REF REF REF REF
Five year age group
50-54 -1.865** (.632) -.068 .102 (1.147) .003
55-59 -.619 (.602) -.022 .204 (1.124) .006
60-64 -2.231*** (.593) -.073 -1.435 (1.037) -.047
65-69 -2.653*** (.624) -.082 -1.436 (1.025) -.043
70-74 -1.785** (.650) -.053 -2.417* (.990) -.077
75plus REF REF REF REF
No Education .855* (.424) .038 2.608*** (.667) .111
Currently Working -.425 (.480) -.015 -2.223** (.755) -.085
South African .665 (.461) .027 .351 (.824) .014
Bad Self-rated Health 10.214*** (.519) .367 12.957*** (1.012) .387
Constant 46.732 46.556 44.638 42.593
R2 .002 .163 .006 .199
p < .05*; p < .01**; p < .001***
Schatz et al. 2015 Older person’s living arrangements and health
24
Table 4b. Regression models for WHODAS separately by sex
Women Men
Model 1 beta Model 2 beta Model 1 beta Model 2 beta
Productive Arrangement 2.217** (.653) .057 .861 (.638) .022 3.090** (1.010) .084 1.659 (1.011) .045
Household Size -.126 (.085) -.027 -.269* (0.127) -.065
Socioeconomic status (quintiles)
First (lowest) -.320 (1.084) -.007 1.451 (1.668) .031
Second -.723 (.955) -.016 4.348** (1.613) .095
Third -1.056 (.918) -.024 1.769 (1.372) .042
Fourth -1.588 (.929) -.035 2.026 (1.447) .044
Fifth (highest) REF REF REF REF
Five year age group
50-54 -12.861*** (1.055) -.283 -5.332** (1.828) -.104
55-59 -12.424 *** (1.046) -.265 -5.452** (1.827) -.114
60-64 -12.370*** (1.090) -.245 -8.430*** (1.640) -.185
65-69 -9.900*** (1.092) -.186 -6.821*** (1.786) -.136
70-74 -6.403*** (1.206) -.115 -4.710** (1.717) -.100
75plus REF REF REF REF
No Education 1.926*** (.696) .051 2.809** (1.054) .080
Currently working -1.18 (.771) -.025 -4.271*** (1.050) -.110
South African 2.685** (.782) .066 1.877 (1.268) .049
Constant 20.756 29.069 16.451 20.803
R2 .003 .103 .007 .087
p < .05*; p < .01**; p < .001***