Race, socioeconomic status, and breast cancer treatment and survival

CJ Bradley, CW Given, C Roberts - Journal of the National …, 2002 - academic.oup.com
CJ Bradley, CW Given, C Roberts
Journal of the National Cancer Institute, 2002academic.oup.com
Background: Previous studies have found that African-American women are more likely than
white women to have late-stage breast cancer at diagnosis and shortened survival.
However, there is considerable controversy as to whether these differences in diagnosis and
survival are attributable to race or socioeconomic status. Our goal was to disentangle the
influence of race and socioeconomic status on breast cancer stage, treatment, and survival.
Methods: We linked data from the Metropolitan Detroit Surveillance, Epidemiology, and End …
Abstract
Background: Previous studies have found that African-American women are more likely than white women to have late-stage breast cancer at diagnosis and shortened survival. However, there is considerable controversy as to whether these differences in diagnosis and survival are attributable to race or socioeconomic status. Our goal was to disentangle the influence of race and socioeconomic status on breast cancer stage, treatment, and survival. Methods: We linked data from the Metropolitan Detroit Surveillance, Epidemiology, and End Results (SEER)1 registry to Michigan Medicaid enrollment files and identified 5719 women diagnosed with breast cancer, of whom 593 were insured by Medicaid. We first calculated the unadjusted odds ratios (ORs) associated with race, Medicaid insurance, and poverty for breast cancer stage at diagnosis, breast cancer treatment, and death. We then estimated the ORs of having late-stage breast cancer at diagnosis, breast-conserving surgery, no surgery, and death using logistic regression after controlling for clinical and nonclinical factors. All statistical tests were two-sided. Results: Before controlling for Medicaid enrollment and poverty, African-American women had a higher likelihood than white women of each unfavorable breast cancer outcome. However, after controlling for covariates, African-American women were not statistically significantly different from white women on most outcomes except for surgical choice. African-American women were more likely than white women to have no surgery (adjusted OR = 1.62; 95% confidence interval [CI] = 1.11 to 2.37). Among women who had surgery, African-American women were more likely to have breast-conserving surgery than were white women (adjusted OR = 1.63; 95% CI = 1.33 to 1.98). Conclusions: The linkage of Medicaid and SEER data provides more in-depth information on low-income women than has been available in past studies. In our Metropolitan Detroit study population, race was not statistically significantly associated with unfavorable breast cancer outcomes. However, low socioeconomic status was associated with late-stage breast cancer at diagnosis, type of treatment received, and death.
Oxford University Press