Location as Destiny: Identifying Geospatial Disparities in Radiation Treatment Interruption by Neighborhood, Race, and Insurance

Elsevier,   International Journal of Radiation Oncology*Biology*Physics, Volume 107, Issue 4, 2020, Pages 815-826, ISSN 0360-3016, https://doi.org/10.1016/j.ijrobp.2020.03.016.
Authors: 
Daniel V. Wakefield MD, Matthew Carnell BS, Austin P.H. Dove MD, Drucilla Y. Edmonston MD, Wesley B. Garner MD, MPH, Adam Hubler BS, Lydia Makepeace BS, Ryan Hanson MS, Esra Ozdenerol Ph, Stephen G. Chun MD, Sharon Spencer MD, Maria Pisu PhD, Michelle Martin PhD, Bo Jiang MPH, Rinaa S. Punglia MD, MPH, David L. Schwartz MD, FACR

Purpose

Radiation therapy interruption (RTI) worsens cancer outcomes. Our purpose was to benchmark and map RTI across a region in the United States with known cancer outcome disparities.
Methods and Materials

All radiation therapy (RT) treatments at our academic center were cataloged. Major RTI was defined as ≥5 unplanned RT appointment cancellations. Univariate and multivariable logistic and linear regression analyses identified associated factors. Major RTI was mapped by patient residence. A 2-sided P value <.0001 was considered statistically significant.
Results

Between 2015 and 2017, a total of 3754 patients received RT, of whom 3744 were eligible for analysis: 962 patients (25.8%) had ≥2 RT interruptions and 337 patients (9%) had major RTI. Disparities in major RTI were seen across Medicaid versus commercial/Medicare insurance (22.5% vs 7.2%; P < .0001), low versus high predicted income (13.0% vs 5.9%; P < .0001), Black versus White race (12.0% vs 6.6%; P < .0001), and urban versus suburban treatment location (12.0% vs 6.3%; P < .0001). On multivariable analysis, increased odds of major RTI were seen for Medicaid patients (odds ratio [OR], 3.35; 95% confidence interval [CI], 2.25-5.00; P < .0001) versus those with commercial/Medicare insurance and for head and neck (OR, 3.74; 95% CI, 2.56-5.46; P < .0001), gynecologic (OR, 3.28; 95% CI, 2.09-5.15; P < .0001), and lung cancers (OR, 3.12; 95% CI, 1.96-4.97; P < .0001) compared with breast cancer. Major RTI was mapped to urban, majority Black, low-income neighborhoods and to rural, majority White, low-income regions.
Conclusions

Radiation treatment interruption disproportionately affects financially and socially vulnerable patient populations and maps to high-poverty neighborhoods. Geospatial mapping affords an opportunity to correlate RT access on a neighborhood level to inform potential intervention strategies.