Rising educational gradients in mortality among U.S. whites: What are the roles of marital status and educational homogamy?

The educational gradient in U.S. mortality has been rising among non-Hispanic whites. A common intuition sees the growing educational divide in marital status and increases in educational homogamy as potential explanations. To empirically assess this possibility, we analyze mortality from 1986 to 2015 using the National Health Interview Survey Linked Mortality Files (562,584 persons; 204,756 deaths). At the individual level, being unmarried and spouse's lower educational attainment are associated with higher mortality. Counterfactual analyses, however, reveal that the growing educational divide in marital status and increases in educational homogamy contribute little (8%-15%) to the widening educational gradient in mortality. Our simulation analyses further show that extreme educational divide in marital status could substantially drive up mortality inequality, whereas educational homogamy, even when pushed to the maximum level, would play a limited role in increasing mortality inequality. Combined, the results suggest that changes in the educational divide in marital status have the potential to affect mortality inequality, but the actual changes over the past decades were not strong enough to have a major impact; increases in education homogamy, in contrast, barely affect mortality inequality by education. Along with other research showing small effects of educational homogamy on income inequality, our research suggests that the consequences of changing marriage patterns for social inequalities in general and health inequalities in particular may be more limited than commonly assumed.