The Medicare Effect on Head and Neck Cancer Diagnosis and Survival

Abstract

<h3>Purpose/Objective(s)</h3> Medicare provides health insurance benefits to over 63 million Americans. Up to 25% of near elderly adults are uninsured or have a gap in insurance coverage prior to Medicare eligibility. A "Medicare effect" has been proposed to account for increased healthcare utilization at age 65 and has been associated with more cancer diagnoses and lower cancer-specific mortality in lung, breast, colon, and prostate cancers. The existence of such an effect in head and neck squamous cell carcinoma (HNSCC) has not been studied. We sought to investigate whether an increase in cancer diagnoses occurs with Medicare eligibility and whether this affects cancer stage at diagnosis and survival. <h3>Materials/Methods</h3> Patients were queried from the SEER database. Incidence rates of HNSCC among patients diagnosed in 2000-2016 were calculated for patients ages 60-64, 65, and 66-70. Multivariable Poisson regression compared incidence among age 65 to 60-64 and 66-70. For patients with first primary HNSCC diagnosed from 2007-2016, multivariable logistic regression compared odds of late-stage diagnosis (stages III/IV, and a multivariable competing risks proportional hazards model compared hazard of cancer-specific mortality of age 65 to 60-64 by insurance. <h3>Results</h3> There were 57,953 HNSCCs diagnosed from 2000-2016, with 48.6% among age 60-64 (incidence rate (IR) 41.2 per 100,000 persons), 9.6% age 65 (IR 46.9 per 100,000 persons), and 41.9% age 66-70 (IR 48.1 per 100,000 persons). Controlling for covariates, incidence among age 65 was 14% higher than for 60-64 (incidence rate ratio (IRR) = 1.14, 95% confidence interval (CI) 1.10 – 1.19), but incidence among age 66-70 compared to 65 was not significantly different (IRR = 1.03, 95% CI 0.99 – 1.07). Compared to patients age 65, ages 60-64 with Medicaid (odds ratio (OR) = 1.85, 95% CI 1.61 – 2.13) or who were uninsured (OR = 2.02, 95% CI 1.63 – 2.49) had increased odds of being diagnosed with late-stage HNSCC. Compared to patients age 65, ages 60-64 with Medicaid (sub-distribution hazard ratio (sdHR) = 1.26, 95% CI 1.14 – 1.40) or who were uninsured (sdHR = 1.17, 95% CI 1.02 – 1.35) had increased hazard of cancer-specific mortality, while ages 60-64 with private insurance had decreased hazard of cancer-specific mortality (sdHR = 0.84, 95% CI 0.77 – 0.91). <h3>Conclusion</h3> Medicare eligibility is associated with a significant increase in incidence of HNSCC. In the population studied, patients ages 60-64 had were more likely to be diagnosed with late-stage HNSCC and have increased cancer-specific mortality compared to their Medicare-insured counterparts. Collectively, this supports the notion of a "Medicare effect" in HNSCC wherein patients defer initiating cancer-related treatment until they are Medicare eligible.