Residential racial segregation and mortality among black, white, and Hispanic urban breast cancer patients in Texas, 1995 to 2009

Abstract

BACKGROUND The authors investigated whether residential segregation (the degree to which racial/ethnic groups live separately from one another in a geographic area) 1) was associated with mortality among urban women with breast cancer, 2) explained racial/ethnic disparities in mortality, and 3) whether its association with mortality varied by race/ethnicity. METHODS Using Texas Cancer Registry data, all-cause mortality and breast-cancer mortality were examined among 109,749 urban black, Hispanic, and white women aged ≥50 years who were diagnosed with breast cancer from 1995 to 2009. Racial (black) segregation and ethnic (Hispanic) segregation of patient's neighborhoods were measured and were compared with the larger metropolitan statistical area using the location quotient measure. Shared frailty Cox proportional hazard models were used to nest patients within residential neighborhoods (census tract) and were controlled for race/ethnicity, age, diagnosis year, tumor stage, grade, histology, neighborhood poverty, and county-level mammography availability. RESULTS Greater black segregation and Hispanic segregation were adversely associated with cause-specific mortality and all-cause mortality. For example, in adjusted models, Hispanic segregation was associated with cause-specific mortality (adjusted hazard ratio, 1.24; 95% confidence interval, 1.05-1.46). Compared with whites, blacks had higher mortality for both outcomes, whereas Hispanics demonstrated equivalent (cause-specific) or lower (all-cause) mortality. Segregation did not explain racial/ethnic disparities in mortality. Within each race/ethnicity strata, segregation was either adversely associated with mortality or was not significant. CONCLUSIONS Among urban women with breast cancer in Texas, segregation has an independent, adverse association with mortality, and the effect of segregation varies by patient race/ethnicity. The novel application of a small-area measure of relative racial segregation should be examined in other cancer types with documented racial/ethnic disparities across varied geographic areas.