U.S. racial/ethnic mortality gap adjusted for population structure

Background: Racial mortality disparities in the U.S. are well-documented and central to the debate on social inequalities in health. We argue that standard measures that are used to describe the disparities, such as life expectancy or years of life lost, underestimate those disparities. Methods: We analyze contemporary U.S. mortality disparities comparing Blacks and Hispanics to Whites using CDC and NCHS data. We estimate mortality disparities using standard metrics and a novel approach that weights mortality inequalities by the population fraction that is exposed to the inequalities. We then express the magnitude of these inequalities by comparing them to the loss of life due to leading causes of death. Results: Based on the exposure-adjusted measure, the Black mortality disadvantage is as deadly or deadlier than circulatory diseases, the top cause of death in the U.S; and 43% (men) and 87% (women) larger than the disadvantage as measured by life expectancy. For Hispanics, the exposure-adjusted mortality advantage over Whites is over two times larger, for both men and women, than what life expectancy disparities would imply, and 21% (men) and 11% (women) larger than when measured using standard years of life lost. Conclusions: Mortality inequalities experienced by real populations can differ markedly from the inequalities that are calculated for synthetic populations that are used in standard calculations. We show that racial/ethnic disparities in the U.S. are underestimated if not adjusted for the populations experiencing the inequalities. For health policy the exposure-adjusted inequalities are likely to provide a more reasonable signal on where to allocate scarce resources.