Social determinants of health – A question of social or economic capital? Interaction effects of socioeconomic factors on health outcomes

Abstract

Social structures and socioeconomic patterns are the major determinants of population health. However, very few previous studies have simultaneously analysed the “social” and the “economic” indicators when addressing social determinants of health. We focus on the relevance of economic and social capital as health determinants by analysing various indicators. The aim of this paper was to analyse independent associations, and interactions, of lack of economic capital (economic hardships) and social capital (social participation, interpersonal and political/institutional trust) on various health outcomes. Data was derived from the 2009 Swedish National Survey of Public Health, based on a randomly selected representative sample of 23,153 men and 28,261 women aged 16–84 year, with a participation rate of 53.8%. Economic hardships were measured by a combined economic hardships measure including low household income, inability to meet expenses and lacking cash reserves. Social capital was measured by social participation, interpersonal (horizontal) trust and political (vertical/institutional trust) trust in parliament. Health outcomes included; (i) self-rated health, (i) psychological distress (GHQ-12) and (iii) musculoskeletal disorders. Results from multivariate logistic regression show that both measures of economic capital and low social capital were significantly associated with poor health status, with only a few exceptions. Significant interactive effects measured as synergy index were observed between economic hardships and all various types of social capital. The synergy indices ranged from 1.4 to 2.3. The present study adds to the evidence that both economic hardships and social capital contribute to a range of different health outcomes. Furthermore, when combined they potentiate the risk of poor health.

Introduction

Today, there is ample evidence of associations between levels of economic capital (e.g, low income and self-reported financial stress) and various health outcomes (Laaksonen et al., 2007; Lorant, Croux et al., 2007; Lynch, Kaplan, & Shema, 1997; Weich & Lewis, 1998; Wildman, 2003; Zimmerman & Katon, 2005), and also of the association between individual level social capital (e.g., civic engagement, interpersonal and institutional trust) and diverse health behaviours and health outcomes (Hyyppa, Maki, Impivaara, & Aromaa, 2007; Kawachi, Kennedy, & Glass, 1999; Subramanian, Kim, & Kawachi, 2002; Veenstra, 2002). However, very few previous studies have simultaneously analysed associations of both social and economic capital in relation to health at the individual level. While most previous studies including both social capital and economic capital indicators have investigated one of the variables as predictor variable including the other as control variable, few others have investigated the importance of both in relation to health. Carlson (2004) has found both economic (economic satisfaction) and social factors (social capital; organisational activity and trust) to be important in explaining health differences (self-rated health) between countries in central and Eastern Europe, the former Soviet Union and Western countries. Carlson concluded that economic factors seemed to be most important in relation to area differences in self-rated health.

Rose (2000) has analysed the importance of household income and social capital (networks, friends, life control, and trust) in Russia, and found both aspects to be equally important and independent of each other with respect to physical and emotional health. Stuckler and colleagues have also linked rapid privatisation, social capital and health outcomes in Eastern Europe (e.g., Stuckler, King, & McKee, 2009).

Stafford, De Silva, Stansfeld, and Marmot (2008) have found associations between neighbourhood social capital (trust, reciprocity and associational membership) and common mental disorders (GHQ-12) in the subgroup living in deprived circumstances in England and Scotland. Finally, a recent study by Sun, Rehnberg, and Meng (2009) has found an association between poverty (minimum living allowance) and low individual level social capital (a five dimensional measure) in rural China. The researchers also found a synergy effect between lack of social capital (neighbourhood social cohesion) and poverty, indicating an aggravating effect on health (self-rated health) of the two.

Hence, among the scarce evidence so far, most studies have found independent associations between both the ‘social’ and the ‘economic’ capital variables and health. However, it is also possible, as previously investigated by Sun et al. (2009) in China, that there might be interactive effects between lack of social capital and economic hardships on health outcomes. It is thus most possible that social capital generates material outcomes and vice versa, and that a combination contributes to a double burden on health.

There are several plausible pathways by which a combination of economic and social capital exposures may result in a higher risk of poor health. Lack of economic or financial resources may restrain the individual's choice of social activities, membership in organizations and contact surfaces, i.e. lack of economic resources may result in lower social capital. Unemployed people for instance report lower social capital than people employed or self-employed (Lindström, 2000). Second, low social capital with poorer social networks and lower trust may contribute to lower economic resources as a result of minimised job opportunities, thus low social capital may lead to economic hardships (Granovetter, 1974). Third, a lack of economic and social capital might also affect health directly through psychobiological pathways. These may be caused by high stress levels due to financial difficulties and feelings of mistrust caused by social isolation. Fourth, lack of economic and social capital might also affect health indirectly via social support mechanisms. For example, lack of economic and social capital might contribute to less inclination to participate in society and to benefit from support mechanisms provided by society. A lack of both social and economic capital can thus be considered as a lack of capabilities in several areas or dimensions, which severely constrains life (Sen, 1992, 1999) and contributes to poor health outcomes.

Lack of economic capital, is in this study considered as a multidimensional phenomenon. According to Sen's (Sen, 1992, 1999) capability approach, poverty is a complex, multifaceted concept dependent on each person's own personal characteristics and other social circumstances. Economic capital is also conceptualised in the light of Ringen (1988), who noticed that income and consumption are not always related and therefore suggested that a household should only be classified as deprived if it has both a low income and other symptoms of poverty. In our study economic hardships are measured by three indicators, capturing different dimensions of economic vulnerability (both income and non-income related), combined into one single economic hardships measure.

Social capital is examined at the individual level “comprising social resources that evolve in accessible social networks and social structures characterized by mutual trust” (Rostila, 2010, p.14). These social resources can facilitate access to various instrumental and expressive returns, which might benefit both the individual and the collective (Rostila, 2010). These resources do not reside within the individual (i.e., intrapersonal resources) but in the structure of his/her social networks, and to gain access to social capital an individual must be related to others. Thus, social capital in this view is always relational and inherent in the social structure (Rostila, 2010). Social capital is also conceptualised as the cognitive and structural dimensions that constitute the preconditions for social capital (e.g. Harpham, Grant, & Thomas, 2002; Rostila, 2010).

In our study, the structural dimension highlights the “social” side of the concept and the behavioural manifestation of social networks, i.e. social participation. The cognitive dimension highlights the social trust that emerges in social relations and is measured by horizontal (interpersonal) and vertical (political/institutional) aspects of trust.

A wide range of health outcomes is selected in order to get a deeper understanding of the effects of socioeconomic determinants on broad measures of health. Self-rated health measures a combination of different aspects of health and has proved to be a robust and reliable measure of a person's overall health status and a strong predictor of mortality (Burstrom & Fredlund, 2001; Ferraro, Farmer, & Wybraniec, 1997; Idler & Benyamini, 1997; Lundberg & Manderbacka, 1996).

A vast literature has shown an association between psychological distress and adverse health outcomes such as CHD (cardiovascular diseases) (Eaker, Sullivan, Kelly-Hayes, D'Agostino, & Benjamin, 2005; Kawachi, Sparrow, Vokonas, & Weiss, 1994; Kubzansky, Kawachi, Weiss, & Sparrow, 1998), and also subsequent risks of suicide attempt, psychiatric disease, hospital care and all-cause mortality (Eaton, Badawi, & Melton, 1995; Fogel, Eaton, & Ford, 2006; Ringback Weitoft & Rosen, 2005).

In Sweden, musculoskeletal disorders are the most reported causes of poor health, and the leading causes of work absence, long-term work disability and early retirement, especially among women (Persson, 2001). In addition, some musculoskeletal disorders have been associated with inflammatory process that increases the risk for CHD (Maradit-Kremers, Nicola, Crowson, Ballman, & Gabriel, 2005; Solomon et al., 2003) and contribute substantially to the burden of disease (Moradi, Allebeck, Jacobsson, & Mathers, 2006).

We expect positive associations between economic hardships and aspects of social capital, respectively, and poor health on all three outcomes. We also hypothesise positive interactions between economic hardships and social capital in relation to each of the three health outcomes. In this study we therefore aim to investigate independent associations of lack of economic capital (economic hardships) and social capital (social participation, interpersonal and political/institutional trust) on various health outcomes (self-rated health, psychological distress and musculoskeletal disorders). Furthermore, we aim to investigate whether our social capital variables interact with economic hardships to increase the odds of health problems. The paradox that men have higher mortality but lower prevalence of both mental and somatic chronic health problems is the reason for stratification by sex in this study. Thus, we also attempt to investigate if the health effects differ by gender.