Background. African American women are seen with more advanced breast cance
rs, are less likely to be treated with breast-conserving surgery and genera
lly have poorer prognoses than white women. There are a myriad of potential
causes for these phenomena. The purpose of this study was to measure racia
l differences in the surgical treatment of breast cancer among women with c
omparable health care access and delivery.
Methods. The Breast Cancer Registry of the Department of Surgery at Henry F
ord Hospital was accessed for all patients between January 1, 1990, and Dec
ember 31, 1997 for whom data on race, tumor characteristics, stage, and tre
atment specifics were available. Socioeconomic information was collected wi
th use of 1990 census block data. Proportions of women who received each tr
eatment weve compared for African Americans and whites with use of the rela
tive risk (RR) and 95 % confidence intervals (CI). We used multiple logisti
c regression to obtain estimates of the relative risk, controlling for pote
ntial confounding factors.
Results. Of the 1699 patients in the database, 1250 had sufficient informat
ion for analysis. A total of 8.7% of African American women were diagnosed
with late-stage disease (ie, stage III or IV) compared with 7.9% of whites.
Nevertheless, African American women had a lower frequency of stage I dise
ase (30.5% vs 36.2%) and a higher frequency of stage II disease (36.8% vs 3
1.4%). Overall and adjusted risk estimates for age, tumor stage, marital st
atus, median income, and type of insurance revealed no substantive or stati
stically significant differences between African American and white patient
s. The adjusted RR for local excision was 1.39 (95% CI 0.78 to 2.49), for l
umpectomy and axillary dissection RR 0.92 (95% CI 0.66 to 1.29), for simple
mastectomy RR 0.84 (95% CT 0.41 to 1.72), and for modified radical mastect
omy RR 1.00 (95 % CI 0.73 to 1.36).
Conclusions. In this setting of equal access to health care, African Americ
an women still have higher frequencies of stage II disease, although the fr
equencies for late-stage disease are similar: Nevertheless, no surgical dif
ferences were found in this population, even after the effects of socioecon
omic indicators and stage at diagnosis were controlled for. Survival differ
ences between African American and white women are unlikely to be explained
by differences in treatment.