Racial/Ethnic and Economic Segregation and Survival in the United States

Life expectancy differences across racial/ethnic and economic groups persist in the U.S., but little is known about the combined effects of racial and income segregation in explaining old age survival across neighborhoods. We operationalized neighborhoods using census tracts, with 65,655 of them nested within 3,015 counties, nested within 49 states. We linked census track life table data for ages 45-74 (n=196,965) from the National Vital Statistics System with 5-year population estimates (2011-2015) stratified by race/ethnicity and income from the American Community Survey. We measured racial/ethnic and income segregation using the Index of Concentration of the Extremes at the census tract level, and the indexes of dissimilarity and exposure at the county level. Using three-level random intercept models, we assessed the direct and contextual relationship between survival at ages 45-74 with racial/ethnic and income segregation. Regardless of racial/ethnic stratification, a high concentration of neighborhood poverty was associated with a lower probability of survival relative to affluent neighborhoods (-4.83%; 95% CI -4.86, -4.79), although the relationship was larger in neighborhoods with high concentration of Blacks (-5.61%; -5.67, -5.54). Black-white county-level unevenness also had the largest negative association in those neighborhoods (-0.26%; -0.32, -0.20). Furthermore, Black isolation was negatively associated with a lower survival probability (-0.21%; -0.29, -0.13), but Hispanic isolation was positively associated (0.23%; 0.16, 0.30). Opposite relationships resulted from Black-White (0.06; -0.01, 0.14) and Hispanic-White (-0.13; -0.21, -0.05) interactions. Finally, high exposure to neighborhood poverty/affluence was associated with lower/higher probability of neighborhood survival, but the associations were the strongest for Blacks. SignificanceRacial/ethnic and income residential segregation have been associated with negative health outcomes and increased risk of mortality in the U.S. We tested the association between residential segregation at different geographic levels with the survival probability at ages 45-74 across rural/urban neighborhoods. Results indicate that the racial/ethnic and income geographic composition of the population are important factors for explaining this relationship. That is, poverty concentration, minority-white unevenness, and exposure to poverty, had detrimental outcomes for all communities, irrespective of their racial/ethnic composition. However, while minority-isolation/-exposure to whites is detrimental/protective for Black communities, the opposite happened in Hispanic communities. Our research provides important insights in understanding the extent to which residential segregation explains survival disparities across neighborhoods based on their race/ethnicity composition.