Abstract
INTRODUCTION: Safety net hospitals (SNHs) care for a substantial population of vulnerable patients and are often resource-limited. These limitations may impact treatment decisions for high-risk prostate cancer (hPCa). We performed the first population-based analysis examining SNH status and treatment decisions for localized hPCa. METHODS: National cancer database (NCDB) was queried from 2010-2016 for patients with non-metastatic hPCa. SNH status was defined as hospitals with the 95(th) percentile of Medicaid and uninsured caseload. Non-curative-intent treatment was defined as androgen deprivation monotherapy (ADT) or no treatment. Outcomes assessed were treatment choice and overall survival (OS) by SNH status. RESULTS: A total of 95 747 patients with hPCa were included; 112 hospitals were identified as SNHs, with mean Medicaid/uninsured caseload of 24.4% compared to 3.2% at non-SNHs (p<0.01). Patients at SNHs were independently associated with greater odds of non-curative-intent treatment (odds ratio [OR] 2.2, p<0.01). Results were consistent across subgroups: private insurance (OR 2.2, p<0.01), age <65 (OR 2.3, p<0.01), and at academic centers (OR 1.9, p<0.01). There was no difference in OS among SNHs and non-SNHs when patients received curative treatment. Among patients who did not receive curative treatment, OS was greater at SNHs (hazard ratio 0.82, p=0.02). CONCLUSIONS: Patients at SNHs were more likely to receive non-curative treatment independent of known socioeconomic risk factors. Private insurance or treatment at academic centers did not mitigate these disparities. Increased resources may be needed at SNHs, especially in the context of healthcare expansion, which may further strain these facilities.