Purpose: Women awaiting breast biopsy after abnormal screening mammography can experience psychological distress, and long delays to biopsy have been associated with worse breast cancer outcomes. We evaluated the risk of not receiving a timely biopsy following abnormal screening across racial/ethnic groups adjusting for woman- and healthcare-level factors.
Materials and Methods: We performed a cohort study using data collected prospectively in 2009-2019 from 113 facilities across seven Breast Cancer Surveillance Consortium breast imaging registries. We included asymptomatic women 40-79 years old without a history of breast cancer recalled from screening and with a final recommendation for breast biopsy. Risk factors included in the analysis were woman-level (self-reported race/ethnicity, age, family history of breast cancer, breast density, history of prior breast biopsy, time since prior mammogram, geocoded educational attainment, geocoded household income, and residential rurality) and healthcare-level factors (exam modality, academic affiliation, and availability of on-site biopsy services). Racial/ethnic groups were White, Black, Asian, Hispanic, and other/mixed. We used sequential log-binomial regression models to evaluate women’s risk of not receiving a biopsy within 60 days of abnormal screening, adjusting for multilevel factors.
Results: The cohort included 45,471 abnormal screening mammograms, of which 16.3% were not resolved with biopsy within 60 days. Black, Asian and Hispanic women were at increased risk of not having biopsy within 60 days compared to White women (unadjusted RR[95%CI], 1.38[1.21, 1.58], 1.57[1.14,2.16], and 1.36[1.10,1.69], respectively). There was no significant risk attenuation after adjusting for nonmodifiable woman-level risk factors. Additional adjustment for modifiable woman-level factors slightly attenuated the risk of no biopsy for each racial/ethnic minority group (RR[95% CI],1.33[1.18,1.50], 1.50[1.16,1.94], and 1.29[1.06,1.58], for Black, Asian and Hispanic women, respectively). Adjustment for most healthcare-level factors did not significantly attenuate risk of no biopsy except adjusting for specific screening facility resulted in the most risk attenuation (RR[95%CI], 1.25[1.14, 1.37], 1.20[1.11,1.30], 1.12[1.01,1.23] for Black, Asian and Hispanic women, respectively).
Conclusion: Among women recommended for breast biopsy, there were racial/ethnic differences in receipt of timely biopsy. Adjusting for modifiable woman-level factors resulted in slight risk attenuation, while adjusting for screening facility resulted in the largest risk attenuation of no timely biopsy for women in minority racial/ethnic groups. Persistent increased risk of not receiving biopsy after accounting for multilevel factors suggests other unmeasured factors contribute to differences in time to biopsy.
Clinical Relevance Statement: Multilevel measured and unmeasured factors contribute to delays in completing breast biopsy in a timely manner among women in minority racial/ethnic groups.
.jpg "Marissa B. Lawson, MD photo")