سرمایه اجتماعی، وضعیت اجتماعی و اقتصادی و پریشانی روانی در میان بزرگسالان استرالیا

High levels of social capital may be associated with positive mental health in adults. However, quantifying the various dimensions of social capital has presented a challenge due in part to the diverse definitions and measures used. Data from a representative, population-wide survey of Australian adults aged 16 years and older were used to investigate the links between dimensions of social capital and mental health morbidity. Social capital comprised three constructs and was measured at the individual level: feelings of trust and safety, community participation and neighbourhood connections and reciprocity. Mental health was measured by the 10-item Kessler (K10) instrument and assessed symptoms of psychological distress (i.e., depression and anxiety) over the previous month. Community participation showed a weak, and neighbourhood connections and reciprocity a moderate association with distress. Having higher levels of trust and feeling safe were consistently associated with low levels of psychological distress, after adjusting for socio-demographic characteristics and health conditions. The results clearly demonstrate that having trust in people, feeling safe in the community and having social reciprocity are associated with lower risk of mental health distress. The implications for conceptualising and measuring the individual and collective (contextual) dimensions of social capital are discussed. The findings also suggest the importance of examining the interrelationships between socio-economic status, social capital and mental health for community-dwelling adults.

In recent years, emerging epidemiological data have underlined the magnitude of mental health problems worldwide (WHO, 2003). Depression in particular has attracted international attention since that disorder alone makes a large contribution to the global burden of disease (Ustun, Ayuso-Mateos, Chatterji, Mathers, & Murray, 2004). Depending on the instruments used, the annual prevalence of one or more depressive episodes ranges from 0.7% to 19.9% across diverse community samples, and the lifetime prevalence is estimated to be as high as 14% (Chiu, 2004; Henderson, Andrews, & Hall, 2000; The ESEMeD/MHEDEA 2000 Investigators, 2004). Recurrent and untreated mental health symptoms have been linked to risk of major psychiatric disorders, suicide, chronic physical diseases, general disability and poor work performance (Waghorn, Chant, While, & Whitehead, 2005). In Australia, mental disorders contribute substantially to disability from non-fatal diseases, accounting for 8% of the total disease burden (Mathers, Vos, & Stevenson, 1999). Yet, it remains difficult to explain the variations in the prevalence of mental distress across time and settings, although there are suggestions that economic and social conditions may be implicated. In relation to economic factors, there is substantial evidence from population-based studies that those who are unemployed (Bartley, 1994), who have relatively low incomes (Weich & Lewis, 1998) and/or low standards of living (Lewis et al., 1998) are at higher risk of depression. Communities who enjoy advantageous social conditions may be relatively protected from adverse mental health outcomes (Greiner, Li, Kawachi, Hunt, & Ahluwalia, 2004; Hawe & Shiell, 2000; Ross, 2000; Whitley & McKenzie, 2005). Such communities are characterised by strong social reciprocity, civic and social participation, and high levels of trust and bonding. Conversely, there is a link between loneliness or infrequent social contacts and the onset of depression (Prince, Harwood, Thomas & Mann, 1998). The relevant social elements identified are increasingly incorporated within the notion of social capital making it important to examine the role of that construct in determining mental health status at both the individual and community levels. Several investigators have explored various individual- and group-level mechanisms that might explain how social capital relates to mental health (McKenzie, Whitley, & Weich, 2002; Whitley & McKenzie, 2005). Kawachi and Berkman (2001) have argued that social networks and social supports, as distinct from socio-economic status, can buffer the negative effects of life events on mental health for individuals already under stress. These researchers also hypothesised that regardless of mental health status, engaging in social relationships which result in exposure to positive emotional support will enhance individual psychological well-being. Hence, there may be two key elements, the cognitive (what people ‘feel’, e.g., perception of trust) and structural (what people ‘do’ e.g., degree of social participation) that comprise social capital that are protective of mental health (Harpham, Grant, & Thomas, 2002). For example, the perception of trust or availability of social support generates stress-buffering effects by increasing feelings of security, self-esteem and confidence in one's coping abilities. The structural mechanism impacting on mental health, in contrast, is determined by access to networks and the effectiveness of social interactions within them. Kawachi, Kim, Couts, and Subramanian (2004) go further to postulate that the two elements are linked in that communities rich in social capital promote perceptions of friendly, cohesive and safe societies, characteristics that in turn may encourage interactions among socially isolated individuals who are vulnerable to mental distress. Harpham et al. (2002) hypothesise that social capital could play a key role in reducing risks to factors that cause mental distress. For instance, high social capital could mitigate the distress caused by a job loss or a marriage breakdown—both are life events with potentially detrimental effects on mental health. At present, the empirical evidence supporting the role of social capital in determining mental health is limited and somewhat inconclusive (Almedom, 2005; De Silva, McKenzie, Harpham, & Huttly, 2005). A key challenge is that social capital is a complex construct, making it important to examine more fully which elements are most relevant to promoting mental health. The impact of each element might be influenced by whether social capital is measured at the individual or group level and by related factors, such as socio-economic status. A study assessing social integration according to social networks, church attendance, trust and control found that social capital was strongly correlated with emotional health in the previous year (Rose, 2000). In a population survey of adults, McCulloch (2001) found a significant association between GHQ-12 and social capital measured by perceptions of neighbourhood conditions. Nevertheless, because social capital was measured using global categories (low, medium or high), it was not possible to determine which specific dimensions were related to poor mental health. Greiner et al. (2004) found some discrimination amongst the dimensions of social capital, with perceived community safety and social cohesion rather than community involvement being associated with positive mental health. Focusing on deficits in social capital, Steptoe and Feldman (2001) found that neighbourhood problems (e.g., vandalism and disturbance, street litter, air and noise pollution) were associated with psychological distress as measured by GHQ-12. Moffitt and the E-Risk Study Team (2002) reported similar findings among young mothers. In a recent study, Ziersch and colleagues (2005) reported an association between perceived neighbourhood environments (connections, trust, reciprocity, safety, civic action) and the mental health subscale of the SF-12. Similar findings were reported in another study of Korean adults, but again using different measures of social capital and mental health (Cho, Park, & Echevarria-Cruz, 2005). Negative studies have also been published. Harpham, Grant, and Rodriguez (2004) found that trust in others or in institutions and civic participation were not independently related to anxiety and measuring social cohesion as a combination of perceived trust and neighbourhood relations (by aggregating individual responses to the area level), a study of African–American women found no association between social capital and anxiety and depression measured by the 18-item mini-mood and anxiety symptom questionnaire (Cutrona, Russell, Hessling, Brown, & Murry, 2000). A recent qualitative study conducted in a neighbourhood with high rates of depression found that residents were generally satisfied with the levels of trust and reciprocity in their neighbourhood, regardless of their mental health status (Whitley & Prince, 2005a). The researchers concluded that poor mental health does not link to deficits in social capital. This study, however, relied on in-depth interview data collected from only 32 residents. Measurement challenges may partially explain the mixed findings to date. For example, different measures of mental health have been used, ranging from general concept of ‘emotional health’ (Rose, 2000) to validated psychological screening tools such as the GHQ, SF-12 and CES-D (Ross, Reynolds, & Geis, 2000; Steptoe & Feldman, 2001; Ziersch, Baum, MacDougall, & Putland, 2005). Most studies have relied on previously validated instruments of social capital (Kawachi, Kennedy, Lochner & Prothrow-Sith, 1997; Sampson, Raudenbush, & Earls, 1997), but in most instances, the psychometric properties of these instruments have not been systematically reevaluated when applied to different populations. Further, approaches to measuring social capital have diverged in that some studies have assessed the construct as a property of individuals, while others have considered it to be a property of groups. Studies adopting the latter perspective commonly aggregate individual responses to the community level in order to gain an ecological perspective (e.g., Cutrona et al., 2000), but this approach has been criticised because it assumes that social capital can be defined by locality when the characteristic may be geographically dispersed (McKenzie et al., 2002; Subramanian, Kim, & Kawachi, 2002). Conceptualising and measuring social capital at an individual level, however, has also been criticised (Henderson & Whiteford, 2003), in that psychological distress may influence the way individuals perceive and describe their social environments, creating potential circularity in drawing causes or inferences. For example, reports of low social capital may reflect underlying depression rather than representing an independent cause of these difficulties. These potential biases need to be considered in interpreting the results of any studies in the field. This study reports on a secondary analysis of data collected from a large state-wide survey of Australian adults (New South Wales (NSW) Population Health Survey) conducted by the center for Epidemiology and Research of the NSW Department of Health (Centre for Epidemiology and Research, 2004). The current analysis examined the cognitive and structural components of social capital (Harpham et al., 2002), and their relationship to mental health status using a widely applied measure of population psychological distress. In the analysis, we included possible confounding with socio-demographic and socio-economic factors. Specifically, we aimed to investigate: (1) whether high cognitive and structural social capital are independent correlates of positive mental health after taking into account socio-demographic, individual- and geographical-level socio-economic status and health-related behaviours and (2) which particular elements of social capital exert the greatest impact on mental health even when moderated by indices of socio-economic status.

By examining distinct dimensions of social capital within the context of community environments and demonstrating consistent patterns of findings across them we can be confident that the associations between social capital and psychological distress are robust. In this study, trusting people, feeling safe in one's community and engaging in social reciprocity emerged as the most relevant dimensions of social capital in explaining variations in psychological distress morbidity. However, we did not find a statistically significant and independent inverse association between community participation and mental distress, suggesting that focusing on the social and contextual features of the community rather than its structural elements in the first instance might have substantial practical relevance. Building trust between individuals, building a socially cohesive environment that will promote social reciprocity and establishing new—while enhancing existing—social networks are amenable to change. Examining their effects on mental health outcomes in controlled trials would further provide a strong scientific base for effective mental health promotion. Moving from (opportunistic) correlational analyses to longitudinal studies and to intervention research is also a necessary next step for reaffirming the causal nature and mechanisms underpinning these associations. Finally, to ensure that the public health utility of the concept of social capital in explaining mental health morbidity is not overstated, the empirical evidence indicated in this study underscores the need to better understand how economic conditions moderate the relationship between psychological distress and social capital.