Marriage and Health in Malawi

PROJECT SUMMARY To date, the vast majority of research on marriage and health has almost exclusively focused on Europe or North America. Despite the centrality of marriage to social life in sub-Saharan Africa (SSA), research on its relationship with health is still in its nascent stages in this context. While

these early studies have documented that marriage and health are indeed interrelated, there has been little consideration of the implications of other key features of the African context, such as different marriage patterns and customs (e.g., early marriage, polygyny), persistence of traditional gender norms, and the HIV/AIDS epidemic, for the link between marriage and health.

The limited body of research in sub-Saharan Africa suggests that it is a unique and important setting to study the relationship between marriage and health, with social norms, infectious diseases, and marital practices that differ from contexts where most research on marriage and health have been set. In this proposed study, we intend to use a rich, longitudinal dataset to

examine the link between marriage and health in Malawi. In doing so, we address several methodological challenges that have affected research on this relationship. We use longitudinal data and statistical approaches to reduce common biases in examining the connection between marriage and mental and physical health, such as unobserved characteristics associated with

both marriage and heath, and the selection of individuals with different health status into marital change (mainly remarriage and dissolution). We take advantage of detailed retrospective marriage histories to identify union start and end dates, status of marriage (still married, divorced/separated, and widowed), reasons for divorce/separation, and polygynous unions. We

also combine data from husbands and wives to produce a couple-level dataset that is used to conduct dyadic analyses. We use a validated instrument (the SF-12) that measures multiple dimensions of mental and physical health, as well as specific subscales of each (e.g., depression, anxiety, physical functioning). We analyze data from individuals aged 15 to over 80

years, which allows us to account for differences in the marriage-health relationship across the life course. Finally, we take advantage of data collected from migrants who have left the study area to minimize attrition bias and account for the connection between marriage and migration.