Outcomes of Lymphoma Among American Adolescent and Young Adult Patients Varied by Health Insurance-A SEER-based Study

Abstract

Introduction: Impacts of health insurance status on survival outcomes among adolescent and young adult (AYA, 15 to 39 years of age) patients with lymphoma in the United States are insufficiently known. This study aimed to clarify associations between health insurance status and overall survival (OS) estimates in this population.

Materials and methods: We examined 18 Surveillance, Epidemiology, and End Results registries in the United States and analyzed American AYA patients with lymphoma diagnosed during January 2007 and December 2016. Health insurance status was categorized, and Kaplan-Meier and multifactor Cox regressions were adopted using hazard ratio and 95% confidence interval. Probable baseline confounding was modulated by multiple propensity score.

Results: A total of 21,149 patients were considered; ~28% were 18 to 25 years old, and 63.5% and 7.5% had private and no insurance, respectively. Private insurance rates increased in the 18 to 25 age group (60.1% to 6.1%, P<0.001) following the 2010 Patient Protection and Affordable Care Act (ACA), and lymphoma survival rates improved slightly 1 to 5 years postdiagnosis. Five-year OS rates decreased with age (93.9%, 90.4%, and 87.0% at 15 to 17, 18 to 25, and 26 to 39, respectively) and differed among insurance conditions (81.7%, 79.2%, 89.2%, and 92.0% for uninsured, Medicaid, insured, and insured/no specifics, respectively). Risk of death was significantly higher for those with Medicaid or no insurance than for those with private insurance in multiple propensity score-adjusted models (hazard ratio [95% confidence interval]=1.07 [1.03-1.12]), independent of stage at diagnosis.

Conclusions: No or insufficient insurance was linked to poor OS in our sample in exposure-outcome association analysis. Insurance coverage and health care availability may enhance disparate outcomes of AYAs with cancer. The ACA has improved insurance coverage and survival rates for out sample. Nevertheless, strategies are needed to identify causality and eliminate disparities.