RESUMO
BACKGROUND: Racial disparities in the use of breast-conserving surgery (BCS) have been reported and may be due to advanced stage at diagnosis. Our hypothesis was that low-income and ethnic minority patients have an increased tumor size at diagnosis and decreased likelihood of BCS. PATIENTS AND METHODS: A retrospective review was conducted of early stage breast cancer patients from 10 hospitals in Harris County, Texas, between 2004 and 2011. Clinical stage was calculated on the basis of data from the institutional tumor registries and electronic medical records. Zip code-based socioeconomic factors were downloaded from the US Census Bureau (http://www.census.gov/). Linear regression was used to identify predictors of tumor size, and logistic regression was used to identify predictors of BCS. RESULTS: The cohort included 3937 patients, comprising 2546 (65%) whites, 535 (14%) African Americans, 482 (11%) Hispanics, and 374 (10%) Asian/others. Multivariate linear regression demonstrated socioeconomic status (SES), younger age, African American, Hispanic race, and hormone receptor-negative tumors to be associated with increased tumor size at diagnosis (P < .05). Hispanic and Asian/other race, larger tumor size, combined estrogen receptor-negative/progesterone receptor-negative tumors were associated with not receiving BCS. CONCLUSION: Race and SES were both associated with larger tumor size at diagnosis. Larger tumor size, negative hormone receptor status, and Hispanic and Asian race were associated with lack of receipt of BCS. Breast cancer screening programs should target both minority and low SES groups. Rates of BCS should be interpreted cautiously when used as a quality metric because of the multiple factors, including tumor size and biology, contributing to its use.
