The Effect of Health Insurance on Childhood Cancer Survival in Pediatric Patients Treated With Radiation in Multi-Center Health Systems

• Published in: International journal of radiation oncology, biology, physics Vol. 111; no. 3; pp. e321 - e322
• Main Authors: Fine, J., Ransdell, J., Kwon, D., Zhao, W., Isrow, D.
• Format: Journal Article
• Language: English
• Published: Elsevier Inc 01.11.2021
• DOI: 10.1016/j.ijrobp.2021.07.991
• ISSN: 0360-3016

The impact of health insurance on cancer identification and survival remains understudied in the pediatric cancer population. This study was designed to assess the association between health insurance status and childhood cancer survival. Data from 464 pediatric oncology patients age < 19 who were diagnosed between Jan. 1990 and Aug. 2019 and treated for cancer with radiation therapy at a large public hospital or a multi-center private hospital in a diverse metropolitan city were reviewed retrospectively. Variables collected included patient demographics, treatment, toxicities, recurrences, and death. Continuous variables were compared using ANOVA tests; categorical variables were compared with Chi-square test and Fisher's exact test excluding unknown. Cox regression modeling was used for univariable (UVA) and multivariable (MVA) analyses to assess associations between study variables and clinical outcomes: overall (OS) and recurrence free survival (RFS). Event-free individuals (alive for OS; alive without recurrence for RFS) were censored at date of last follow-up. Hazard Ratios (HR) and corresponding 95% confidence intervals (CI) were calculated with P-values of < 0.05 considered significant. Mean survival time was 1.52 years shorter in pediatric oncology patients with Medicaid/Medicare than in pediatric oncology patients with private insurance (P = 0.015). In UVA, Medicaid/Medicare patients had significantly worse OS than privately insured patients (HR 2.0, 95% CI 1.27, 3.70). RFS was also significantly worse in Medicaid/Medicare patients compared to privately insured patients (HR 1.79, 95% CI 1.27, 2.5). Inferior OS and RFS in Medicaid/Medicare patients remained significant in MVA when adjusting for diagnosis (OS: HR 2.0, 95% CI 1.04, 3.13; RFS: HR 1.52, 95% CI 1.08, 2.13). No significant survival differences were identified by race, ethnicity, sex, socioeconomic status, or treatment site. Across insurance groups, no significant differences between the times from diagnosis to chemotherapy initiation (P = 0.276), diagnosis to radiation initiation (P = .592), or initial radiation consult to radiation start (P = 0.391) were observed. The analysis reveals significantly worse survival outcomes in pediatric cancer patients with Medicaid/Medicare. The disparity in survival between private vs. Medicaid/Medicare patients exists despite no differences in treatment initiation times or differences in survival between treatment sites. These findings suggest that factors outside of treatment in the hospital systems are contributing to the inferior outcomes in Medicaid/Medicare-insured pediatric cancer patients. These factors may include later stage of cancer at time of diagnosis due to inferior healthcare accessibility. Further research is necessary to discern what factors are driving the disparities observed in the Medicaid/Medicare pediatric cancer population in order to improve their outcomes.