Insurance Status is an Independent Predictor of Overall Survival in Patients With Stage III Non–small-cell Lung Cancer Treated With Curative Intent

Abstract

IntroductionPopulation studies suggest an impact of insurance status on oncologic outcomes. We sought to explore this in a large single-institution cohort of patients with non–small-cell lung cancer (NSCLC). Materials and MethodsWe retrospectively analyzed 342 consecutive patients (January 2000 to December 2013) curatively treated for stage III NSCLC. Patients were categorized by insurance status as uninsured (U), Medicare/Medicaid + Veterans Affairs (M/M + VA), or Private (P). The χ2 test was utilized to compare categorical variables. The Kaplan-Meier approach and the Cox proportional hazard models were used to analyze overall survival (OS) and freedom from recurrence (FFR). ResultsCompared with M/M + VA patients, P insurance patients were more likely to be younger (P < .001), married (P < .001), Caucasian (P = .001), reside in higher median income zip codes (P < .001), have higher performance status (P < .001), and undergo consolidation chemotherapy (P < .001) and trimodality therapy (P < .001). Diagnosis to treatment was delayed > 30 days in U (67.3%), M/M + VA (68.1%), and P (52.6%) patients (P = .017). Compared with the M/M + VA and U cohorts, P insurance patients had improved OS (median/5-year: 30.7 months/34.2%, 19 months/17%, and 16.9 months/3.8%; P < .001) and FFR (median/5-year: 18.4 months/27.3%, 15.2 months/23.2%, and 11.4 months/4.8%; P = .012), respectively. On multivariate analysis, insurance status was an independent predictor for OS (P = .017) but not FFR. ConclusionCompared with U or M/M + VA patients, P insurance patients with stage III NSCLC were more likely to be optimally diagnosed and treated, resulting in a doubling of median OS for P versus U patients. Improved access to affordable health insurance is critical to combat inequities in access to care and has potential for improvements in cancer outcomes.