3106 - Neighborhood Trajectory as a Predictor for Proximity to Radiation Therapy

Purpose/Objective(s): Radiation therapy (RT) is a requisite of multidisciplinary cancer care and often requires frequent travel. Longer travel times are associated with delayed and inadequate treatment. While prior research has evaluated the influence of distance traveled, less is known about location of RT centers relative to marginalized communities, of which members are less likely to receive guideline concordant care. Neighborhood Trajectories (NT) describe such communities by depicting historic residential segregation (via Home Owners’ Loan Corporation (HOLC) maps) and current socioeconomic characteristics into four categories: Advantage Stable, Advantage Reduced, Disadvantage Stable, and Disadvantage Reduced. The primary objective of this study was to evaluate the accessibility of RT centers according to NT.

Materials/Methods: RT centers across the United States were identified and geocoded from state health departments, the Imaging and Radiation Oncology Core, and Radiation Dosimetry Services. NT’s exist for the 32,153 census block groups that overlap a HOLC graded neighborhood by >50% in >200 metropolitan areas in the US. Travel time from the population-weighted centroid of each NT to the closest RT center was calculated with ArcGIS Pro. Mean drive time to the closest RT center were compared by region & NT type. Statistical significance of differences in mean drive times and variance was performed using ANOVA testing.

Results: 2,513 RT centers were identified. Mean travel times from each NT to the nearest RT center ranged from 6.2 to 10.1 minutes (Table 1). Evaluation of averages by region and NT demonstrated statistically significant variation (p<0.001).

Conclusion: Average drive time to the nearest RT center varied significantly by region and NT, however absolute differences were <3 minutes and unlikely clinically meaningful. Our results are likely due to use of census blocks for which HOLC data are available, namely in urban areas. While the difference in drive time by NT varied by short intervals, a significant difference was seen and warrants further investigation. Future analyses will explore potential variation in access across broader geographic areas and additional barriers to timely care that may vary by residential neighborhood, such as referral & insurance networks.