Abstract 1188: The effect of insurance status on diagnosis stage and overall survival among children and adolescents with cancer: A propensity score weighting analysis

Abstract

Abstract Background. Differences in access, delivery or utilization of health care may impact childhood and adolescent cancer survival. In this study, we evaluated the impact of insurance coverage on diagnosis stage, treatment received, and overall survival among children and adolescents diagnosed with cancer based on a counterfactual framework. Methods. We identified 58,421 individuals diagnosed with any cancer type at ≤ 19 years old from 2004-2010 from the National Cancer Database. We applied the inverse probability of treatment weighting (IPTW) estimator to correct for observed selection biases generated by insurance enrollment at diagnosis or initial treatment and examined associations between insurance status and diagnosis stage, treatment received (any vs. none), and all-cause mortality using weighted logistic and weighted Cox proportional hazard regressions. Results. Compared to privately insured individuals, there was a higher percentage of younger individuals in the Medicaid category for all cancers, whereas the uninsured category had a higher percentage of older individuals. Children and adolescents with Medicaid and no insurance were more likely to be non-White and from lower income counties. Applying the IPTW estimator, all the imbalances on the above covariates were removed. Results from weighted analysis showed a 15% (95 confidence interval (CI): 0.78-0.93) lower odds of being diagnosed at metastatic stage had children and adolescents with Medicaid been enrolled on private insurance but comparable odds for uninsured individuals (odds ratio: 0.98, 95%CI: 0.79-1.22), vs. those with private insurance. The odds of receiving any treatment were 17% (95%CI: 1.08-1.28) and 68% (95%CI: 1.39-2.04) higher had those with Medicaid and no insurance been enrolled on private insurance at diagnosis or initial treatment. Medicaid and uninsured individuals would have a 24% (95%CI: 0.72-0.80) and 24% (95%CI: 0.66-0.87) lower hazard of death during 5 years of follow-up if they were enrolled on private insurance at diagnosis or initial treatment. Conclusion. Children and adolescents without private insurance would have a lower risk of being diagnosed at metastatic stage, higher odds of receiving any treatment, and lower hazard of death within five years following cancer diagnosis had they been enrolled on private insurance at diagnosis or initial treatment. Additional research is needed to further understand underlying mechanisms. Citation Format: Xiaoyan Wang, Kimberly Johnson. The effect of insurance status on diagnosis stage and overall survival among children and adolescents with cancer: A propensity score weighting analysis [abstract]. In: Proceedings of the Annual Meeting of the American Association for Cancer Research 2020; 2020 Apr 27-28 and Jun 22-24. Philadelphia (PA): AACR; Cancer Res 2020;80(16 Suppl):Abstract nr 1188.