A cohort perspective of U.S. adult mortality

This dissertation advances a cohort perspective to analyze trends in racial and educational disparities in U.S. adult mortality. The project is organized around three themes. First, I emphasize that recent temporal changes in U.S. adult mortality risk are rooted in cohort forces. Unfortunately, much of the mortality literature has failed to account for the fact that the sociohistorical conditions of U.S. cohorts have changed dramatically, and these changes have tremendous implications for population health and mortality trends. My work clearly shows the pitfalls of omitting these cohort effects from analyses of U.S. adult mortality risk. Second, I illustrate that because exposure to social and health conditions have changed over time, resources in adulthood are growing increasingly important in shaping U.S. adult mortality risk. In this regard, my findings also highlight growing disparities in U.S. mortality across race/ethnic gender groups. Third, I advance a cohort theory of U.S. mortality, drawing from both “fundamental cause” theory and a life course perspective of mortality but couching them in a cohort framework to highlight the importance of historical changes in U.S. social and health contexts in both childhood and adulthood. This cohort perspective is then used to analyze three central topics in the U.S. mortality literature: the black-white crossover in older-adult mortality, the growing educational gap in U.S. adult mortality, and the origins and persistence of black-white inequalities in U.S. adult mortality. I estimate hierarchical age-period-cohort cross-classified random effects models using National Health Interview Survey-Linked Mortality Files between 1986 and 2006 to simultaneously analyze age, period, and cohort patterns of U.S. adult mortality rates. I find (1) the black-white crossover is a cohort-specific phenomenon, (2) educational disparities in U.S. adult mortality rates are growing across birth cohorts, not time periods, and (3) racial disparities in U.S. adult mortality rates stem from cumulative racial stratification across both cohorts and the life course. Such findings have direct consequences for both mortality theories and policy recommendations. Only by considering the disparate sociohistorical conditions that U.S. cohorts have endured across their life courses can we fully understand and address current and future health disparities in the United States.