Objective: To examine racial/ethnic disparities in mortality and survival in a large nationwide and population-based cohort of women with breast cancer after simultaneously controlling for differences in comorbidity, treatment, and socioeconomic status.

Methods: A cohort of 35,029 women with stage I-IIIA breast cancer at age > or = 65 from 1992 to 1999 was identified from the surveillance, epidemiology, and end results-medicare linked databases with up to 11 years of follow-up. Cox proportional hazard regression analysis was performed to determine the risk of all-cause and breast cancer-specific mortality.

Results: African-American women with breast cancer were more likely to live in the poorest quartiles of socioeconomic status at the census tract level than whites (73.7% versus 20.7%, P < 0.001). Those living in communities with the lowest socioeconomic status were 11% more likely to die than those in the highest (hazard ratio, 1.10; 95% confidence interval, 1.04-1.16). The risk of dying changed slightly after controlling for race/ethnicity (1.11; 1.05-1.18). Compared with white women with breast cancer, crude hazard ratios of all-cause and breast cancer-specific mortality were 1.35 (1.27-1.45) and 1.83 (1.56-2.16) for African-Americans. After adjusting for treatment and socioeconomic status, hazard ratio of all-cause mortality was no longer significant in African-Americans (1.02; 0.84-1.10), whereas the risk of breast cancer-specific mortality was marginally higher in African-Americans (1.21; 1.01-1.46).

Conclusions: Racial disparities in overall survival between African-American and white women with breast cancer were not present after controlling for treatment and socioeconomic status. Efforts to eliminate these barriers have important public health implications for reducing disparities in health outcomes.