Guideline-concordance along the cancer care continuum and breast cancer mortality by race and ethnicity: a SEER-Medicare study

Abstract Purpose To examine the relationship between guideline-concordant breast cancer care and hazard of cancer death by patient race and ethnicity. Methods We used SEER-Medicare data to identify 212,555 older women diagnosed with invasive breast cancer between 2000 and 2017. Guideline-concordant diagnostic workup, locoregional treatment, and initiation of systemic therapy were defined using NCCN guidelines. Hazards of breast cancer death 2 and 5 years from diagnosis by each guideline-concordance outcome overall and stratified by race and ethnicity were estimated using Cox proportional hazards models. Results Non-concordant diagnostic workup, locoregional treatment, and systemic therapy initiation were each associated with increased hazards of 2- and 5-year breast cancer mortality (diagnostics HR 2-year (95% CI) 1.33 (1.25–1.41), HR 5-year 1.29 (1.23–1.35); locoregional HR 2-year 2.10 (1.98–2.23), HR 5-year 1.83 (1.76–1.90); systemics HR 2-year 1.67 (1.51–1.84), HR 5-year 1.56 (1.45–1.68)). Non-concordant diagnostic workup and systemic therapy initiation were associated with greater hazard of 2- and 5-year breast cancer death among Black, Asian/Pacific Islander, Hispanic White, and non-Hispanic White patients; there was no consistent association among American Indian/Alaska Native patients for either outcome. Locoregional treatment was strongly associated with hazards of cancer death for all groups. Conclusion Equitable delivery of guideline-recommended breast cancer care from diagnosis through treatment across racial and ethnic groups may mitigate survival disparities. Efforts to improve access to high-quality care must be informed by and responsive to the social and structural root causes of health inequities.