Elsevier

Social Science & Medicine

Volume 67, Issue 12, December 2008, Pages 2017-2026
Social Science & Medicine

Social participation and healthy ageing: An international comparison using SHARE data

https://doi.org/10.1016/j.socscimed.2008.09.056Get rights and content

Abstract

Using the Survey of Health, Ageing & retirement in Europe (SHARE) data for respondents aged 50 years and over in 2004, this study evaluates the potential contribution of increased social participation to self-reported health (SRH) in 11 European countries. The probability to report good or very good health is calculated for the whole sample (after controlling for age, education, income and household composition) using regression coefficients estimated for individuals who do and for those who do not take part in social activities (with correction for selection bias in these two cases). Counterfactual national levels of SRH are derived from integral computation of cumulative distribution functions of the predicted probability thus obtained. The analysis reveals that social participation contributes by three percentage points to the increase in the share of individuals reporting good or very good health on average. Higher rates of social participation could improve health status within the whole sample and within most countries. Context and institutional arrangements (such as income inequality) may explain differences in social participation health efficiency.

Introduction

Europe is the world's oldest continent in demographic terms. It has the highest median age of all continents (38 years) and 20.6% of its population is aged 60 years and over. By 2050, this figure will reach 34.5% and the number of ‘oldest old’ (people aged 80 years and over) is expected to grow by 180% (source United Nations, 2007). Europe is getting older. Country members of the European Union at the Hampton Court informal Summit in October 2005 stressed that demographic ageing is one of the main challenges that the European Union will have to face in the years to come. The threat is that with ageing comes poor health and in consequence, reduced economic performance and increased health care public expenditures (European Commission, 2006). Health promotion of the ageing population is not only a public health policy target, it also is an intermediate objective for economic sustainability. Following Jamieson's (1994) prior work, policy interventions in Europe already dealing with this issue often draw inspiration from theoretical frameworks promoting “active ageing” (WHO, 2002) or “healthy ageing” (WHO, 2006) as a process of increasing opportunities for health to enable older people to take part in society. Amongst the various directions that may help achieving this goal, individuals' involvement in social activities (or social capital) may be decisive (see Agren & Berensson, 2006).

During the last decade, a wide range of social capital measures were found to be associated with various health outcomes (cf. Hawe and Shiell, 2000, Szreter and Woolcock, 2004, Islam et al., 2006), giving substance to Putnam's (2000: 326) famous assertion that “in none is the importance of social connectedness so well established as in the case of health and well-being.” However, significant evidence suggests that the health–social capital nexus should not be taken for granted since correlations between some proxies are frequently unobserved. In particular, social participation is a form of social capital which relationships with health are not unambiguous. Some studies providing statistical evidence that these two concepts are linked (e.g. Lindström, 2004, Petrou and Kupek, in press) is being challenged by a growing body of the literature reporting the absence of correlation between participation in social activities and self-reported health (D'Hombres et al., 2007, Greiner et al., 2004, Veenstra et al., 2005, Ziersch and Baum, 2004) or other health outcomes (Ellaway & Macintyre, 2007).

A close look at the literature advocates that the positive effects of social participation on health could be significant for the sub-population of older people (Veenstra, 2000, Kondo et al., 2007). One reason could be that older people have more time to take part in social activities due to retirement (Christoforou, 2005) or fewer familial constraints (Bolin, Lindgren, Lindstrom, & Nysetdt, 2003). This investment in social capital could help maintaining them in good health. At least two arguments may help in justifying this assumption. First, the number of cohort acquaintances an individual has throughout his life may decrease after a certain age (Glaeser, Laibson, & Sacerdote, 2002). Involvement in associations and other social groups may help maintaining (if not increase) the size of social networks. Second, retirement has been found to be associated with a decrease of individuals' cognitive capacities (Adam, Bay, Bonsang, Germain, & Perelman, 2006). Social participation may slow down this process as it often requires cerebral efforts from the individuals and thus help preserve their mental health (cf. Almedom, 2005). If these two statements are true, then taking part in social activities could help improve older adults' health status.

In order to test these assumptions, we use cross-sectional self-reported data (2004) from non-institutionalized individuals aged 50 years and over participating in the Survey of Health, Ageing, and Retirement in Europe (SHARE). The probability to report good or very good health is calculated for the whole sample (after controlling for age, education, income and household composition) using regression coefficients estimated for individuals who do and for those who do not take part in social activities (with correction for selection bias in these two cases). Counterfactual national levels of self-reported health (SRH) are derived from integral computation of cumulative distribution functions of the predicted probability thus obtained, and compared with the current probability to report good/very good health status. This counterfactual conditional process allows us to measure the potential effect a change in the rates of social participation could have on SRH. Although counterfactual (conditional) analysis is often the basis of experimental methods for establishing causality in medicine and social sciences (cf. Morgan and Winship, 2007, Pearl, 2000), our aim is not to test the direction of causation. Based on recent development of the literature (Folland, 2007, D'Hombres et al., 2007), we postulate here that involvement in voluntary associations influences respondents' health status. Notice that this study provides—in passim—a secondary analysis of the determinants of social participation since correction for selection bias requires as a preliminary to estimate the probability that a person is involved in one or more social activities.

The paper is structured as follows: the next section presents data from the SHARE project and some descriptive statistics. The method section deals with econometric issues and the different tests applied here. Regression results and interpretations are given in the results section; while comments, limitations and further research issues are displayed in the discussion. Conclusions are drawn in the final section.

Section snippets

The Survey of Health, Ageing, and Retirement in Europe (SHARE)

This study used cross-section of individual-level data from Release 2.0.1 of the first wave of the Survey of Health, Ageing and Retirement in Europe (SHARE) collected in 2004. SHARE has been developed on the basis of prior successful experiments which are the Health and Retirement Survey (HRS) in the United States, and the English Longitudinal Survey of Ageing (ELSA). SHARE is a bi-annual longitudinal survey with the aim to carry out international comparisons and analysis of economic and social

Counterfactual analysis

The initial intuition, with the aim to measure the contribution of social participation to self-reported health, was to run a regression with the former variable as one of the determinants of the latter. Although individual logistic models make it possible to estimate the influence of social participation on SRH (by providing the predicted value of y at x-bar) for each of the 11 European countries in the sample, we believe that this method is not desirable here for several reasons. First,

Estimation results

Table A1 in annex reports Probit estimates of equation 1. Statistical inference points out that the initial model can be interpreted as far as correctly predicted outcomes are high enough (68.5%) and the chi-squared value of the LR test indicates that all coefficients are simultaneously and significantly different from zero (p < 0.001). In addition, the usual predictors of health status are significant and associated with the expected signs for the overall sample. Unsurprisingly, age is a very

Discussion

Our results raise the key question of why is relative health efficiency of social participation different between countries? Or alternatively, why the ranking of countries by SRH and rates of associational membership (Table 1) differs from the one based on the relative influence of social capital on health (Table 3)? At the outset, we have to acknowledge that the second ranking is sensitive to model specification (rho parameter) so that one may say that the north–south gradient in SRH and

References (48)

  • M. Lindström

    Social capital, the miniaturisation of community and self-reported global and psychological health

    Social Science & Medicine

    (2004)
  • J.M. Oakes

    The (mis)estimation of neighborhood effects: causal inference for a practicable social epidemiology

    Social Science & Medicine

    (2004)
  • G. Veenstra

    Social capital, SES and health: an individual-level analysis

    Social Science & Medicine

    (2000)
  • G. Veenstra et al.

    Who you know, where you live: social capital, neighborhood and health

    Social Science & Medicine

    (2005)
  • S. Adam et al.

    Occupational activities and cognitive reserve: a frontier approach applied to the survey of health, ageing and retirement in Europe (SHARE), CREPP WP 2006/05, HEC School of Management

    (2006)
  • G. Agren et al.
    (2006)
  • G.S. Becker et al.

    Social economics: Market behaviour in a social environment

    (2000)
  • Christoforou, A. (2005). On the determinants of social capital in Greece compared to countries of the European Union....
  • N. Craig

    Exploring the generalisability of the association between income inequality and self-assessed health

    Social Science & Medicine

    (2004)
  • B. D'Hombres et al.

    Does social capital determine health? Evidence from eight transition countries

    (2007)
  • A.S. Deaton et al.

    Ageing and inequality in income health

    American Economic Review, Papers and Proceedings

    (1998)
  • S. Durlauf

    On the empirics of social capital

    Economic Journal

    (2002)
  • M. Erlinghagen et al.

    The participation of older Europeans in volunteer work

    Ageing & Society

    (2006)
  • Cited by (159)

    View all citing articles on Scopus

    The SHARE data collection has been primarily funded by the European Commission through the 5th framework programme (project QLK6-CT-2001-00360 in the thematic programme Quality of Life). Additional funding came from the US National Institute on Ageing (U01 AG09740-13S2, P01 AG005842, P01 AG08291, P30 AG12815, Y1-AG-4553-01 and OGHA 04-064). Further support by the European Commission through the 6th framework program (projects SHARE-I3, RII-CT-2006-062193, and COMPARE, CIT5-CT-2005-028857) is gratefully acknowledged. For methodological details see Börsch-Supan and Jüerges (2005). The authors would like to thank the two anonymous referees, Jean-Pierre Lachaud, Michel Grignon, IRDES members, and participants to the ESPE 2008 conference for useful comments on a previous version of this paper. The authors thank Linda Marie Messenger for text and language improvement of this article.

    View full text