Social capital dynamics and health in mid to later life: findings from Australia
Social capital dynamics and health in mid to later life: findings from Australia
Vasoontara Yiengprugsawan 0 1 2
Jennifer Welsh 0 1 2
Hal Kendig 0 1 2
0 National Centre for Epidemiology and Population Health, Research School of Population Health (NCEPH), The Australian National University , Canberra , Australia
1 The Australian Research Council Centre of Excellence in Population Ageing Research (CEPAR) , Canberra , Australia
2 & Vasoontara Yiengprugsawan
Purpose The influence of social capital has been shown to improve health and wellbeing. This study investigates the relationship between changes in social capital and health outcomes during a 6-year follow-up in mid to later life in Australia. Methods Nationally representative data from the Household, Income and Labour Dynamics in Australia (HILDA) survey included participants aged 45 years and over who responded in 2006, 2010 and 2012 (N = 3606). Each of the three components of social capital (connectedness, trust and participation) was measured in Waves 2006 and 2010 and categorised as: 'never low', 'transitioned to low', 'transitioned out of low' and 'consistently low'. Health outcomes in 2012 included self-rated overall health, physical functioning, and mental health based on the Short Form 36-item health survey (SF-36). Multivariable logistic regression assessed changes in social capital (measured in 2006 and 2010) predicted poor health (measured in 2012), adjusting for covariates. Results Consistently low trust was significantly associated with higher odds of transitions into poor physical functioning (AOR 1.54; 95% Confidence Interval 1.06-1.22), poor mental health (AOR 1.59; 95% CI 1.08-2.36) and poor self-rated health (AOR 1.86; 95% CI 1.27-2.72). Transition into low trust was also a predictor of poor selfrated health after adjusting for covariates (AOR 1.74; 95% CI 1.11-2.73). Changes in social connectedness in both directions (transitioned out of and into low) were statistically associated with poor self-rated health (AORs 1.40; 95% CI 1.00-1.97 and 1.61; 95% CI 1.11-2.34, respectively) after adjusting for confounders as well as other social capital components. Conclusions Our longitudinal findings reveal social capital dynamics and effects on health in mid to later life. Social trust and connectedness could be important enablers for older persons to be more active in the community and potentially benefit their health and wellbeing over time.
Social capital; Social participation; Trust; Self-rated health; Ageing; Middle and older adults
Centre for Research on Ageing, Health and Wellbeing,
Research School of Population Health (CRAHW), The
Australian National University, Canberra, Australia
Putnam’s seminal work on social capital, building from
concepts of social democracy [
], has been applied in a
range of empirical work linking social bonding to
beneficial health outcomes [
] and overall life satisfaction [
Much of this work has focused on psycho-social
resources—notably trust, social support, social networks and
reciprocity while community dimensions have also been
addressed primarily in terms of social inequalities and
spatial segregation [
In the past few decades, emerging research in Western
countries has focused on social capital and its role in later
]. A cross-national study in Europe has reported
that, regardless of the levels of social trust and social
networks, there were similar associations between social
capital and self-assessed health among older adults in
Finland, Poland and Spain . Another comparative study
among the elderly reported that low trust was associated
with adverse self-rated health in both the US and Germany;
in addition, lack of social participation was also associated
with poor self-rated health and depression in Germany [
International reviews of public policy have argued for
improving social capital as an important strategy for
reducing social exclusion and inequality among
disadvantaged older people [
In Australia, there have been calls to consider social
capital as part of the public health agenda [
monitoring population health [
]. A cross-sectional study
in two suburbs of Adelaide found that those who were
better off materially had better access to social capital;
further, perceived material advantage as well as social
capital was associated with mental and physical health
]. An early national study found that measures of social
capital and perceived material wellbeing predicted mental,
but not physical health [
]. Another national
cross-sectional Australian study has shown that structural
(community participation) and cognitive (social cohesion)
components of social capital related to general health,
mental health and physical functioning [
empirical longitudinal data are limited especially for older
The aim of this research is to provide longitudinal
evidence on the changes in social capital and effects on health
outcomes in mid to later life. In particular, we set out to
investigate the relationship between three components of
social capital (connectedness, trust, participation) and
effects on vulnerability in terms of health during a 6-year
follow-up among participants aged 45 years and over in
Data and sample
This study used nationally representative data from the
Household, Income and Labour Dynamics in Australia
(HILDA) survey. HILDA data are primarily collected
using face to face or telephone interviews but information
on more sensitive topics, including social attitudes and
health is collected using a mail back self-completed
questionnaire. This study is based on Waves 6, 10 and 12
(collected in 2006, 2010 and 2012) because of special topic
modules relating to social capital in these waves.
Respondents were included in this study if they were aged
45 years or older in Wave 6 (2006) and returned their
questionnaire in all the three waves (N = 3606). Appendix
1 includes information on sample and inclusion criteria in
the supplementary data.
We measured three components of social capital: ‘‘low
connectedness’’—infrequent contact with friends or
relatives or perceptions that neighbours are unwilling to help;
‘‘low trust’’—low generalised trust; and ‘‘low
participation’’’—no club membership and only infrequent
attendance at community events (more information in Appendix
2 in the Supplementary Material). Each was measured in
Waves 6 and 10, allowing us to further categorise
components according to transitions between waves: ‘never
low’, ‘transitioned to low’, ‘transitioned out of low’ and
We focused on three measures: self-rated overall health,
physical functioning and mental health in Wave 12 based
on the international standardised medical outcomes study
Short Form 36-item health survey [
]. Respondents were
considered to have poor self-rated health if they reported
their overall health as ‘poor’ or ‘fair’ or poor physical
functioning or mental health if their score was in the
bottom 20% of scores for their age group (Appendix 2 in the
In order to assess the main effects of social capital, the
following potential confounding variables from Wave 6
were grouped into categories: sex, age groups, marital
status, employment status, household equivalised annual
income, region of residence, number of people in the
household and whether the respondent had a long-term
Multivariable logistic regression assessed the extent to
which transitions in connectedness, trust and participation
(measured in Waves 6 and 10) predicted poor health
(measured in Wave 12) and taking into account covariates
including baseline health from Wave 6. Analyses were run
separately for each health outcome. Respondents reporting
poor health (assessed with the cut points noted above) at
the baseline of the study in 2006 were excluded from the
analysis. Models were first adjusted for confounders
(Model 1) and then additionally for other components of
social capital (Model 2). Data were weighted to the
Characteristics of the sample are presented in Table 1:
approximately 75% aged between 45 and 65 years, 17%
were 65–74 years, and 8% were 75? years. Across the
three components of social capital: low social
connectedness 34%; low trust 29%, and low participation 23% were
reported in 2006. The number and weighted percent of
respondents in each of social capital dynamics for social
connectedness, trust, and participation between 2006 and
2012 and the multivariable associations with health
outcomes are shown in Table 2.
Transition into low connectedness between 2006 and
2010 was associated with poor mental health (Adjusted
Odds Ratio, AOR 1.54; 95% Confidence Interval
1.02–2.33). However, once adjusted for trust and
participation dynamics, the effect size was still high but no longer
statistically significant (AOR 1.38; 95% CI 0.90–2.10).
Transitions out of and into low connectedness were
significant predictors in reporting poor self-rated health
(AORs 1.53; 95% CI 1.10–2.14 and 1.76; 95% CI
1.24–2.52, respectively) after adjusting for confounders as
well as other social capital components (AORs 1.40; 95%
CI 1.00–1.97 and 1.61; 95% CI 1.11–2.34, respectively).
Low trust was robustly associated with all three health
outcomes with an observed gradient of adverse health
outcomes from never low, transitioned into low, and
consistently low. In particular, consistently low trust were
significantly associated with higher odds of transitions
into poor physical functioning (AOR 1.54; 95% CI
1.06–1.22), poor mental health (AOR 1.59; 95% CI
1.08–2.36) and poor self-rated health (AOR 1.86; 95% CI
1.27–2.72). Transition into low trust was also a predictor
of poor self-rated health after adjusting for covariates
(AOR 1.74; 95% CI 1.11–2.73). Consistently, low social
participation was statistically associated with poor
selfrated health (AOR 1.53; 95% CI 1.02–2.31). However,
after further adjusting for trust and connectedness
dynamics, the associations attenuated and were no longer
We report findings on social capital dynamics and health
among nationally representative samples aged 45 years and
older in Australia. Across the three social capital
components, consistently low social trust dynamics were the
strongest predictors for all outcomes especially for poor
self-rated health. Notably, changes in social connectedness
in both directions (transitioned into and out of low) were
statistically associated with poor self-rated health. This
strong effect could reflect the relationship between social
connection and self-perceived health. Besides social trust,
other transitions were not statistically significant adjusting
for other social capital components.
Our findings on social trust predicting self-rated health
were in line with a longitudinal study in a sample of three
ageing cohorts in Finland which reported that stability and
change of high levels of trust over three years have
important effects on self-rated health [
]. However, a
comparative study has shown that Finland generally had
almost twice the higher proportion of trust as compared to
Spain and Poland [
]. Our older Australian samples
reported similar proportions of trust levels and have also
shown similar association with health outcomes to the
latter two countries (e.g. significant relationship between
trust and self-rated health in both Spain and Poland).
Our findings provide international evidence on the role
of social capital in later life [
]. In particular, having trust
could be an important enabler for older persons to be more
active in the community. Social capital through
participation could alleviate loneliness among older persons which
in turn could help to improve their health and wellbeing.
Promoting social capital and facilitating formal and
informal social networks can be an effective health promotion
strategy for older populations [
Some considerations for this study include firstly the
strength of representative national samples with an array of
sociodemographic and health covariates which could be
taken into account in the analyses. Secondly, there might
be bi-direction relationships between social capital and
] and consequently to minimise reverse
causality effects, our analyses were restricted to
participants who did not have poor health at the baseline. Thirdly,
we also investigated generalised trust, connectedness, and
participation as our social capital measures; however, these
measures may not capture all dimensions of social
relationships. In future studies with larger samples and stronger
cross-national comparative dimensions, such as the
longitudinal Australian survey of ageing populations now under
development, it would be possible to better understand the
influence of varying personal and social context—for
Bold values indicate statistically significance results (p \ 0.05)
Respondents reporting poor health (assessed with the cut points) at the baseline of the study in 2006 were excluded from the analysis. Estimates
were weighted to the population and were adjusted for the survey design. Model 1 is adjusted for: age groups, sex, marital status, employment
status, health condition, residence, number of people in the household, and income quintiles. Model 2 is further adjusted for all components of
social capital simultaneously
example gender, age, life history and social class variations
at different points across later life .
In Australia, there has been increasing research and
advocacy on behalf of ‘ageing well’, that is, the positive
dimensions of health and wellbeing, noting the attitudinal
and structural barriers facing people in mid to later life
]. This action can include psycho-social interventions
with vulnerable older people and extend to social actions
such as addressing age discrimination in the workplace as
well as related social policies [
]. New ways of
conceptualising challenges and opportunities over the life course
can greatly benefit Australia in the midst of rapid
Acknowledgements This research was supported by the Australian
Research Council Centre of Excellence in Population Ageing
Research (CE110001029). The authors wish to thank Peter Sbirakos
for editorial guidance throughout the process and Professor Catherine
D’Este for statistical advice in response to reviewers.
Open Access This article is distributed under the terms of the
Creative Commons Attribution 4.0 International License (http://crea
tivecommons.org/licenses/by/4.0/), which permits unrestricted use,
distribution, and reproduction in any medium, provided you give
appropriate credit to the original author(s) and the source, provide a
link to the Creative Commons license, and indicate if changes were
1. Putnam , R. D., Leonardi , R. , & Nanetti , R. ( 1993 ). Making democracy work: civic traditions in modern Italy . Princeton, NJ: Princeton University Press. xv, 258 .
2. Szreter , S. , & Woolcock , M. ( 2004 ). Health by association? social capital, social theory, and the political economy of public health . International Journal of Epidemiology , 33 ( 4 ), 650 - 667 .
3. Elgar , F. J. , et al. ( 2011 ). Social capital, health and life satisfaction in 50 countries . Health Place, 17 ( 5 ), 1044 - 1053 .
4. O 'Rand , A. M. ( 2006 ). Stratification and the life course: Life course capital, life course risks, and social inequality . In R. H. Binstock & L. K. George (Eds.), Handbook of aging and the social sciences . Cambridge: Elsevier, Academic Press.
5. Kawachi , I. ( 1999 ). Social capital and community effects on population and individual health . Annals of the New York Academy of Sciences , 896 , 120 - 130 .
6. Koutsogeorgou , E. , et al. ( 2015 ). Social capital and self-rated health among older adults: A comparative analysis of Finland Poland and Spain . Ageing Society , 35 , 653 - 667 .
7. Pollack , C. E. , & von dem Knesebeck , O. ( 2004 ). Social capital and health among the aged: Comparisons between the United States and Germany . Health Place, 10 ( 4 ), 383 - 391 .
8. Walker , A. ( 2010 ). Ageing and quality of life in Europe . In D. Dannefer & C. Phillipson (Eds.), The SAGE handbook of social gerontology . Thousand Oaks, CA: SAGE Publications.
9. Nyqvist , F. , & Forsman , A. ( 2015 ). Social capital as a health resource in later life: The relevance of context . Netherlands: Springer.
10. Baum , F. ( 1999 ). Social capital: Is it good for your health? issues for a public health agenda . Journal of Epidemiology and Community Health , 53 ( 4 ), 195 - 196 .
11. Siahpush , M. , & Singh , G. K. ( 1999 ). Social integration and mortality in Australia . Australian and New Zealand Journal of Public Health , 23 ( 6 ), 571 - 577 .
12. Ziersch , A. M. ( 2005 ). Health implications of access to social capital: Findings from an Australian study . Social Science and Medicine , 61 ( 10 ), 2119 - 2131 .
13. Phongsavan , P. , et al. ( 2006 ). Social capital, socio-economic status and psychological distress among Australian adults . Social Science and Medicine , 63 ( 10 ), 2546 - 2561 .
14. Ware , J. E. , Jr. , & Sherbourne , C. D. ( 1992 ). The MOS 36-item short-form health survey (SF-36). I. Conceptual framework and item selection . Medical Care , 30 ( 6 ), 473 - 483 .
15. Watson , N. ( 2012 ) Longitudinal and Cross-sectional Weighting Methodology for the HILDA Survey HILDA Project Technical Paper Series . No. 2 /12.
16. Nummela , O. , et al. ( 2009 ). The effect of trust and change in trust on self-rated health: A longitudinal study among aging people . Archives of Gerontology and Geriatrics , 49 ( 3 ), 339 - 342 .
17. Giordano , G. N. , & Lindstrom , M. ( 2016 ). Trust and health: Testing the reverse causality hypothesis . Journal of Epidemiology and Community Health , 70 ( 1 ), 10 - 16 .
18. Kendig , H. , McDonald , P. , & Piggott , J . (Eds.). ( 2016 ). Population ageing and Australia's future. Canberra: ANU Press for the Academy of Social Sciences in Australia.
19. Kendig , H. , & Browning , C. ( 2016 ). Social research and actions on ageing well . In H. Kendig, P. McDonald , & J. Piggott (Eds.), Population ageing and Australia's future. Canberra: ANU Press for the Academy of Social Sciences in Australia.
20. Windsor , T. D., Curtis , R. G. , & Luszcz , M. A. ( 2016 ). Social engagement in late life . In H. Kendig, P. McDonald , & J. Piggott (Eds.), Population ageing and Australia's future. Canberra: ANU Press for the Academy of Social Sciences in Australia.