Impact of Primary Payer on Survival in Patients

Abstract

Prior studies have demonstrated better outcomes for cancer patients with health insurance coverage compared to those with none, but less has been reported regarding the effect of insurance type. The U.S. health insurance landscape is highly variable in patients <65 years old, given they do not have access to uniform coverage such as Medicare. We sought to determine whether the type of primary payer (private versus government-sponsored insurances [GSI]) impacted survival in insured patients younger than 65 with localized head and neck cancer undergoing definitive radiotherapy (RT). The National Cancer Data Base was queried for patients age <65 with stage I-IVB squamous cell carcinoma of the oropharynx, nasopharynx, hypopharynx, or larynx diagnosed from 2004 to 2014 and treated with definitive RT (total dose: 6600 – 8160 cGy) with or without chemotherapy. Patients with no insurance, unknown insurance status, or oropharynx cancer (OPC) with unknown HPV status were excluded. Multivariate Cox regression and propensity score matching were used to adjust for imbalances in covariates. A Bonferroni correction was applied to p-values from subgroup analyses to account for multiple hypothesis testing. The analysis included 27,292 HNSCC patients with median follow-up of 52.1 months. In propensity score-matched cohorts, 3-year overall survival (OS) was 75.2% versus 63.5% for privately insured versus government-insured patients, respectively (p<0.001). In multivariate analysis, private health insurance was associated with significantly improved OS (HR 0.61, 95% CI 0.57-0.65, p<0.001) in comparison to GSI after accounting for other covariates. In subgroup analysis, private insurance was associated with improved OS across all subgroups. However, the greatest impact of private insurance was observed among patients with HPV-positive OPC (interaction p<0.001), stage T1-2 tumors (interaction p<0.001), age ≤50 (interaction p=0.001), no comorbidity (interaction p=0.001), and those receiving chemotherapy (interaction p=0.001) when performing tests of interaction in multivariable models corrected for multiple hypothesis testing. For example, private insurance was associated with a 54% (95% CI 46%-62%) relative decrease in the risk of death for HPV-positive OPC, in comparison to a 37% (95% CI 34%-40%) relative decrease in mortality for HPV-negative cancers. Patients under 65 with HNSCC with private health insurance have improved OS relative to patients with GSI. Although this effect was observed in all subgroups, the relative survival difference was largest in young, healthy patients with HPV-positive OPC and those receiving chemotherapy. This illustrates a significant disparity in our complex healthcare-delivery system which may be mitigated with future policy changes.