Drawing upon the same dataset as the previous study, this analysis emphasised the focus upon social capital by considering migration and trusting relationships alongside leisure participation (the latter of which was arranged over the same seven categories and frequency scale as before). Individual-level social capital was captured through questions about migration (from one municipality to another), residential stability, trust in family relations, close friends and trust in them and leisure and social participation (as before). This time, deaths during the first five years of follow-up were excluded, to allow for undiagnosed diseases at the time of the baseline study. In the reduced cohort, all-cause and cardiovascular mortality (including strokes) up to November 2004 was established, with reference to the Finnish National Registry for Cause of Death as before. Four sets of Cox proportional hazard models were constructed around individual-level social capital, with all non-significant confounders omitted.
This study found that ‘In men, leisure social participation only just predicted all-cause mortality, but none of the measures of individual-level social capital predicted cardiovascular mortality. Economic status slightly modified the effect of leisure participation in men, thus emerging as a tentative mediator between social capital and health in men’ (p. 594). The measures taken to be indicative of social capital were less robust. While interpersonal trust proved to be a predictor of both all-cause and cardiovascular mortality in women, residential stability (enabling longevity of trust networks) was rejected as a measure of social capital.
The researchers responsible for this study assessed its strengths and limitations. As with the Swedish studies, it was based on a robust population sample and accurate mortality data. While reverse causation during the first five years of follow-up was avoided within this revised research design, the problem of the long intervening period persisted.
Markku T. Hyyppä, Juhani Mäki, Olli Impivaara and Arpo Aromaa, ‘Individual-Level Measures of Social Capital as Predictors of All-Cause and Cardiovascular Mortality: A Population-Based Prospective Study of Men and Women in Finland’, European Journal of Epidemiology, 22, 2007, pp. 589–97.