Income Inequality, Social Capital, and Risk for Cardiovascular Disease and Depression in South Africa
MetadataShow full item record
CitationAdjaye-Gbewonyo, Kafui. 2016. Income Inequality, Social Capital, and Risk for Cardiovascular Disease and Depression in South Africa. Doctoral dissertation, Harvard T.H. Chan School of Public Health.
AbstractChronic stress associated with living in societies with high income inequality may increase risk for cardiovascular disease (CVD) and mental illness. Research has also linked social capital, including trust of other people, to mental health outcomes. South Africa has one of the highest levels of income inequality recorded worldwide and is experiencing a growing burden of non-communicable diseases. Data additionally suggest that levels of trust in South Africa are low. Using longitudinal data from adults interviewed in Waves 1 to 3 (2008-2012) of the National Income Dynamics Study, this dissertation examined whether district-level income inequality calculated from census and survey data was associated with several physical and behavioral risk factors for CVD (Paper 1) and with depressive symptoms (Paper 2), as well as whether trust at the individual and district levels was associated with depressive symptoms (Paper 3).
Income inequality, measured using Gini coefficients, and prevalence of most CVD risk factors increased during the study time period, while depressive symptoms, measured using the Center for Epidemiological Studies of Depression Short Form (CES-D-10), decreased. In pooled cross-sectional regression models, higher district inequality was associated with lower BMI and waist circumference. In longitudinal fixed-effects models, Gini coefficients were not significantly associated with any CVD risk factors. District-level Gini coefficients were not significantly associated with CES-D-10 scores or with high depressive symptoms (scores of 10 or higher) in either cross-sectional or longitudinal fixed-effects models. Low generalized and personalized trust at the individual level were associated with lower CES-D-10 scores and reduced risk of high depressive symptoms in pooled cross-sectional and longitudinal fixed-effects models. Neither district-level generalized nor personalized trust was associated with depressive symptoms. However, a cross-level interaction was observed suggesting that lower levels of trust are associated with reduced depressive symptoms when districts have low percentages of highly trusting individuals but not when district-level distrust is low.
In conclusion, the results of this dissertation do not support the income inequality hypothesis in relation to CVD risk factors or depressive symptoms in South Africa. Moreover, findings suggest that trust may not be beneficial for depressive symptoms for all individuals or in all contexts.
Citable link to this pagehttp://nrs.harvard.edu/urn-3:HUL.InstRepos:27201741