2051 - Effect of Distance and Sociodemographic Variables on Radiotherapy Recommendation, Initiation and Completion

J. S. Hogan¹, E. J. Orav², and M. Lam³; ¹Washington University School of Medicine, St. Louis, MO, ²Department of Medicine, Brigham and Womens Hospital, Boston, MA, ³Department of Health Policy and Management, Harvard T.H. Chan School of Public Health, Boston, MA

Purpose/Objective(s): Access to cancer care in the United States is affected by many variables, with outcomes varying significantly based on timely access to care. Prior single-cancer type studies have shown that distance decreases the likelihood of receiving radiotherapy (RT). However, it is unclear where in the cancer care process this occurs, and if this relationship is consistent across common cancer types. This study aims to address the effects of sociodemographic variables and increasing distance between the patient’s residence and the hospital on the likelihood of 1) being recommended RT, 2) initiating RT once recommended, and 3) completing RT once started. Materials/

Methods: Using the National Cancer Database, we identified patients diagnosed with the ten most common solid cancers (bladder, breast, colorectal, uterine, kidney, lung, melanoma, oropharyngeal, pancreas, or prostate) between 2018-2021. The three primary outcomes were: 1) being recommended RT, 2) initiating recommended RT, and 3) completing started RT. The distance variable (defined as the crowfly distance between the patient’s residence and the hospital) was categorized as 0-10, 10-50, and >50 miles (similar to prior studies). For each outcome, all patients with the outcome of interest and available distance data were included in the analysis. Other covariates included patient characteristics (age, sex, diagnosis year, Charlson comorbidity, metastatic stage, and cancer type), sociodemographic characteristics (race, ethnicity, median income quartile, and insurance status), geographic region, and facility type. We performed univariate and multivariable analyses to study the relationship between the covariates and each outcome.

Results: A total of 3,068,919 patients were included in the RT recommendation analysis, 1,068,749 in the RT start analysis, and 959,064 patients in the RT completion analysis. In adjusted models, patients living >50 miles away, non-White patients, and Hispanic patients were less likely to be recommended RT. Of patients who were recommended RT, Black patients were less likely to initiate RT, and uninsured patients, those on Medicaid, or those in a lower median income quartile were less likely to initiate or complete RT (p< 0.05 for all results).

Conclusion: This study demonstrates how sociodemographic factors and distance impact access to RT along the cancer care continuum, highlighting the necessity for support before and after RT has been recommended to ensure equitable access to care. To increase the likelihood of completing treatment, additional support may be beneficial for patients with recommended RT who are uninsured, on Medicaid, and with low household incomes.