OBJECTIVE: High rates of screening mammography have been reported among Hispanic women in the U.S. However, a potential harm of screening is a false positive result with recall breast imaging or biopsy. Our objective was to identify factors associated with false positive results on screening mammography among a predominantly Hispanic population in New York City.
METHODS: We enrolled women receiving mammography at Columbia University Medical Center in New York, NY. They completed a questionnaire on breast cancer risk factors and gave consent to access their medical records for breast imaging and biopsy reports for the past 15 years. Breast cancer risk was assessed using the Gail model and eligibility for BRCA genetic testing was determined using a family history screener. High breast density was defined qualitatively as heterogeneously or extremely dense. Recall breast imaging was based upon a BIRADS score of 0, 3, 4, 5, or 6 on the screening mammogram. False positive breast biopsies were any biopsies that did not yield breast cancer.
Results: From November 2014 to May 2015, 1325 women were enrolled: median age 58 years (range, 29-89); white/black/Hispanic/other (%): 10/10/76/4; 25% met high-risk criteria for breast cancer; 31% had high breast density; 71% were undergoing annual mammography; 53% had at least one recall breast imaging and 6% had at least one false positive breast biopsy. In multivariable analysis, high breast cancer risk, high breast density, and more frequent screening mammograms were associated with recall breast imaging and biopsy.
CONCLUSION: Based upon our results, a potential strategy to reduce the false positive rates on screening mammography is to target women at high risk for breast cancer or with high breast density for screening with breast tomosynthesis, which has less false positives than digital mammography. Additionally, we may adopt less frequent breast cancer screening in average risk women to further reduce the harms of screening.
