The Impact of Hospital Safety-Net Burden on Outcomes for Primary Supratentorial Brain Tumor Surgery

Abstract

Abstract INTRODUCTION Safety-net hospitals treat a disproportionate share of vulnerable patient populations. While outcomes at these institutions for neurosurgical procedures, such as cerebral aneurysm surgery, have been researched, the impact of safety-net burden on brain tumor surgery is poorly characterized. METHODS Using International Classification of Diseases-9 diagnosis codes, we identified all adult admissions in the National Inpatient Sample from 2002 to 2011 undergoing craniotomy for a primary supratentorial brain tumor (191.0-5, 191.8-9, 225.0 and 237.5), excluding patients with brain metastasis (198.3). For each hospital, we quantified safety-net burden as the percentage of patients on Medicaid or without insurance. Hospitals in the top quartile of safety-net burden were categorized as high-burden hospitals (HBHs) while the remainder were low-burden hospitals (LBHs). Survey-weighted multivariate regression was used to make national estimates and adjust for 12 confounding variables: age, sex, insurance, household income, severity of illness and risk of mortality scores, Charlson Comorbidity Index, malignant status, hospital ownership, teaching status, region, and volume. RESULTS We analyzed 162 828 total admissions admitted to 1135 hospitals for brain tumor craniotomy in 2002 to 2011. A total of 212 hospitals (19%) were classified as HBHs, treating 16 914 admissions. HBHs were more likely to be low-volume and public hospitals (both P < .001). Moreover, patients at HBHs were less likely to be white and had higher severity of illness scores (both P < .001). When evaluating outcomes, hospital safety-net status was not associated with mortality (P = .260), favorable discharge disposition (P = .765), or perioperative complications (P = .757). However, admission to HBHs was associated with higher length of stay (+ 0.74 d, P = .007) and inpatient costs (+ $2 448, P = .002). CONCLUSION Although safety-net burden was not associated with mortality, disposition, or complications, patients at safety-net hospitals exhibited increased length of stay and costs, potentially due to factors like resource limitations or more advanced disease presentation among safety-net patients. Amidst potential reforms like “pay-for-performance” reimbursement models, it is critical to further study neurosurgical outcomes at safety-net hospitals.