Purpose: The probability of a woman developing invasive breast cancer < age 40 is low (<1%). For in situ or invasive cancer, the incidence rate per 100,000 woman‐years for African American (AA) and white women, respectively, is 16.8 and 15.1 (Brinton et al, 2008). Yet, 34% of non‐Hispanic AA women, 30% of non‐Hispanic white women, and 22% of Hispanic women ages 30–39 reported ever having a mammogram from a national population‐based sample (Kapp et al, 2009). We describe first mammograms in younger women to investigate racial/ethnic differences (AA, Asian, Hispanic, white) in mammography outcomes.

Methods: Data are pooled from the National Cancer Institute's Breast Cancer Surveillance Consortium, a collaborative network of mammography registries created for the purpose of studying performance and outcomes in community practice. We included women ages 18–39, with no prior history of breast cancer, with a first mammogram (screening or diagnostic) between 1996 and 2005. We determined whether a cancer diagnosis (DCIS or invasive) was made within the 12 months following each mammogram and, using standard definitions, classified each mammogram as a true positive (TP), false positive (FP), true negative (TN) or false negative (FN).

Results: Our sample included 73,353 screening mammograms and 26,262 diagnostic mammograms. Our FP results for screening mammograms reflected modest variability across race/ethnicity (10.4–14.1%). AA woman had a TP to total screening mammogram ratio of 1 in 363, compared to 1 in 623 for white women, while the ratio of a FP to total screening mammograms was similar across all racial/ethnic groups (1 in 7–10). Among diagnostic mammograms, the FP variability was greater, ranging from 8.7% for white women to 18.2% for Asian women with an absolute risk difference of a TP of <1% among all racial/ethnic groups.

Conclusions: While the FP rates may vary moderately by race/ethnicity, their impact may vary substantially. Average risk AA women have greater odds than white women of reporting multiple mammograms < age 40 (Kapp et al, 2009); yet ≥40 are less likely to receive adequate mammography screening (Smith‐Bindman et al, 2006). Could early mammography testing adversely impact future mammography use? This deserves further study.

Citation Information: Cancer Prev Res 2010;3(1 Suppl):A92.

Copyright © January 7, 2010, American Association for Cancer Research
2010