Race and ethnicity and breast cancer outcomes in an underinsured population

Background The disparity in breast cancer mortality between African American women and non-Hispanic white women has been the subject of increased scrutiny. Few studies have addressed these differences in the setting of equal access to health care. We compared the breast cancer outcomes of underinsured African American and non-Hispanic white patients who were treated at a single institution. Methods We conducted a retrospective review of medical records for breast cancer patients who were treated at Wishard Memorial Hospital from January 1, 1997, to February 28, 2006. A total of 574 patients (259 non-Hispanic whites and 315 African Americans) were evaluated. A Cox proportional hazards regression analysis for competing risks was performed. All statistical tests were two-sided. Results Sociodemographic characteristics were similar in the two groups, and both racial groups were equally unlikely to have undergone screening mammography during the 2 years before diagnosis. Most (84%) of the patients were underinsured. The median time from diagnosis to operation, receipt of adequate surgery, and use of all types of adjuvant therapy were similar in the two groups. Median follow-up was 80.3 months for non-Hispanic whites and 77.9 months for African Americans. After accounting for the effect of comorbidities, African American race was statistically significantly associated with breast cancer-specific mortality (African Americans vs non-Hispanic whites: 26.0% vs 17.5%, P =. 028; hazard ratio [HR] of death = 1.64, 95% confidence interval [CI] = 1.06 to 2.55). Adjustment for age at diagnosis, clinical stage, and hormone receptor status attenuated the effect, and the effect of race on breast cancer-specific survival was no longer statistically significant (HR of death from breast cancer = 1.43, 95% CI = 0.89 to 2.30). After adjustment for sociodemographic factors, the hazard ratio for race was further attenuated (HR = 1.26; 95% CI = 0.79 to 2.00). Conclusions In this underinsured population, African American patients had poorer breast cancer-specific survival than non-Hispanic white patients. After adjustment for clinical and sociodemographic factors, the effect of race on survival was no longer statistically significant.