The Role of Facility Variation on Racial Disparities in Use of Hypofractionated Whole Breast Radiation Therapy

Abstract

Hypofractionated radiation therapy is a less burdensome and less costly approach that is efficacious for most patients with early-stage breast cancer. Concerns about racial disparities in adoption of medical advances motivate investigation of the use of hypofractionated radiation in diverse populations. The goal of our study was to determine whether hypofractionated whole breast radiation therapy after breast-conserving surgery was being similarly used across racial groups in the state of Michigan. A prospectively collected statewide quality consortium database from 25 institutions was queried for patients with breast cancer who completed hypofractionated (HF) or conventionally fractionated whole breast radiation therapy from January 2012 to December 2018. We used patient-level multivariable modeling to evaluate associations between HF use and race, controlling for patient and facility factors, and multilevel modeling to account for patient clustering within facilities. Of 9634 patients analyzed, 81% self-reported race as white, 17% as black, and 2% as Asian, similar to statewide and national distributions. In addition, 31.7% of whites were treated at teaching centers compared with 66.7% of blacks and 64.8% of Asians. In 2018, HF was used in 72.7% of whites versus 56.7% of blacks and 67.6% of Asians (P = .0411). On patient-level multivariable analysis, black and Asian races were significantly associated with a lower likelihood of HF receipt (P < .001), despite accounting for treatment year, age, laterality, body mass index, breast volume, comorbidities, stage, triple-negative status, intensity modulated radiation therapy use, teaching center treatment, and 2011 American Society for Radiation Oncology Hypofractionation Guideline eligibility. On multilevel analysis, race was no longer significantly associated with HF receipt. We observed that black and Asian patients receive hypofractionated whole breast radiation therapy less often than whites, despite more frequent treatment at teaching centers. Multilevel modeling eliminated this disparity, suggesting that differences in facility-specific HF use appear to have contributed. Further inquiry is needed to determine whether reduction of facility-level variation may reduce disparities in accessing HF treatment.