This study tested the hypothesis that differences between African American and white patients in stage of diagnosis of breast cancer and in treatment of breast cancer are caused by racial differences in access to care. The authors examined data from a managed care system with equivalent access to care across racial groups. If differences existed in stage of breast cancer presentation and treatment, then this hypothesis would be rejected.

The study population was identified as all African American and white women with newly diagnosed breast cancer during the period from 1990 to 1997 who were treated at a hospital in Michigan that serves an HMO but also other patient referrals. After excluding women whose stage of tumor and nodal status was unknown, the sample consisted of 1,250 women. About half of both black and white women were members of the HMO.

Almost 46% of African Americans were first seen with regional or distance disease compared with 39% of whites. This difference was largely due to stage II patients (36.8% of African Americans versus 31.4% of whites), rather than the later stages (8.7% of African Americans versus (7.9% of whites). Among stage II patients (for which staging depends both on tumor size and nodal status), African Americans were more likely to have tumors >2cm compared with whites (RR=1.48, 95% CI=1.08, 2.03) but less likely to have nodal involvement (RR=0.85, 95% CI=0.71, 1.00). After adjustment for age, tumor stage, marital status, estimated household income, and insurance, African Americans and whites received similar treatments.

The authors conclude that there is a higher rate of stage II breast cancers in African American women compared with white women, but similar rates of advanced disease in a managed care environment that provides equal access to health care. However, the equal distribution of treatments implies that "equal counseling to the treatment options will lead to an equal distribution of surgical treatments." As possible explanation for the stage at diagnosis patterns, the authors cite previous research suggesting differences in tumor biology, differences in mammography screening, and cultural differences that account for differences in perceived risk and knowledge of symptoms of breast cancer between African American and white women.

This study is limited by the lack data on breast cancer diagnosis and treatment for a comparison group of non-HMO patients in the same setting.