Mortality Differences in Windowhood

Being widowed elevates mortality risk, relative to married men or women of the same age. I investigate how risk varies by subpopulation of widows and widowers. Specifically, using data from the Health and Retirement Study, I test for differences in widowhood mortality by education, by number of children, and by how sudden or expected the death of the pre-decedent spouse was. Consistent with other studies, I find an increased hazard of mortality upon widowhood. In contrast with other studies but consistent with the larger literature on SES and mortality, education is protective in widowhood. Number of children has a u-shaped association with mortality in widowhood, with those having 3-4 children having the lowest levels of mortality after death of a spouse. Lingering deaths (deaths after a chronic condition) of the predecedent spouse are much for the surviving spouse than sudden deaths or other types of death. Important gender differences occur with each of these effects. These findings illuminate mechanisms through which mortality is affected by widowhood, and provide evidence on the power of SES and social support in vulnerable populations.

Widowhood/Mortality/spousal death/spousal death

62851