Three essays on socioeconomic status, social support, and the Health and Retirement Study

The associations between both socioeconomic status and mortality and social support and mortality are widely recognized, though the mechanisms that underlie these associations are less well understood. The Health and Retirement Study allows for a rich exploration of these associations. In the first essay, I investigate mortality differentials by both education and household wealth, first for all causes combined and then over seven different causes of death. Both education and wealth are strongly associated with mortality. I find that the SES gradient is strongest for lung cancer and respiratory diseases, and the gradient is weakest for cancers other than lung and cerebrovascular diseases. In the second essay, I explore how mortality risk increases upon becoming widowed, and what factors increase or mitigate that change in risk. I find that becoming widowed is associated with a nearly 50% increase in mortality risk, net of demographic controls. Some evidence indicates that this risk increase may be due to consequences of adjusting to a new living environment, although some is certainly due to selection into widowhood. Factors that mitigated widowhood mortality risk included education and wealth; children did not have an association with widowhood mortality risk. A spouse may be an irreplaceable source of support. In the third essay, I explore mortality differentials by religion. I find that substantial differences exist, with Mainline Protestants and Jews having the most favorable mortality. Differences were partially explained by socioeconomic status. While Catholics, evangelical Protestants, and black Protestants benefit from favorable attendance patterns, attendance (or lack of) at services explains much of the higher mortality of those with no religious preference. Health behaviors do not mediate the relationship between mortality and religion, except among evangelical Protestants.