Background: A survival difference between white and black females has been reported after diagnosis and treatment for breast carcinoma. Although multivariate analyses demonstrate that the survival difference is less or nonexistent when additional prognostic factors are considered, study results have been inconsistent. The objective of this study was to examine further the role of ethnicity as an independent prognostic factor for breast carcinoma survival among white and black females treated over a 30-year period.
Methods: Among 3382 eligible white and black females who registered at M. D. Anderson Cancer Center (MDACC) from 1958 to 1987, ethnicity, age, socioeconomic status (SES), stage of disease, and treatment were examined. Data were obtained from the hospital tumor registry and survival was compared using actuarial life tables and multivariate Cox regression analyses.
Results: Univariate analyses demonstrated that the survival difference between white and black females was significant, with black females having a 1.63 times higher risk of mortality at 5 years compared with white females (confidence interval (CI), 1.47-1.82), with ethnicity a significant predictor of survival (P < 0.001). After controlling for SES, stage, and treatment, the relative risk was 1.12 (CI, 1.00-1.25) and ethnicity was no longer a strong predictor of survival (P = 0.048).
Conclusions: In this hospital population, the white and black female survival difference, which was highly significant when only univariate analyses were considered, became marginally significant after controlling for other prognostic factors.