(S003) Disparities in Stage at Diagnosis and Survival in Adult Cancer Patients According to Insurance Status

Gary V. Walker, MD, MPH, Stephen R. Grant, BS, Ashleigh Guadagnolo, MD, MPH, Matthew Koshy, MD, Usama Mahmood, MD; UT MD Anderson Cancer Center

Background: The Patient Protection and Affordable Care Act seeks to increase the rate of participation in insurance plans for the 48 million individuals without coverage. The purpose of this study is to determine the influence of insurance status on the stage of presentation and survival among non–Medicare-age adult patients with the 10 most common causes of cancer death using the Surveillance, Epidemiology, and End Results (SEER) public use database. 

Methods: A total of 473,722 patients aged 18–64 years who were diagnosed with 1 of the top 10 causes of cancer death (breast, prostate, lung, colorectal, head and neck, non-Hodgkin lymphoma, liver, pancreatic, ovarian, esophageal) in the SEER database from 2007–2010 were analyzed. Demographic information was obtained, including age, gender, race, year of diagnosis, and marital status. Insurance status was defined as insured, Medicaid, or uninsured. Extent of disease was categorized as local (no nodal or metastatic disease), regional (nodal disease), or metastatic (any distant disease). Definitive treatment was defined as surgery (for breast, colorectal, liver, ovarian, and pancreatic cancers) or surgery and/or radiation therapy (for esophageal, head and neck, lung, and prostate cancers). A Cox proportional hazards model was used for multivariate analyses to assess the effect of patient and tumor characteristics on cause-specific death, stratified by insurance status.

Results: The median follow-up was 17 months (range: 0–47 mo). A total of 371,628 (78.4%) had insurance, 55,135 (11.6%) had Medicaid, and 22,442 (4.7%) did not have insurance. On univariate analysis, the following demographic characteristics were associated with the lack of insurance: younger age, male gender, nonwhite race, being unmarried, rural residence, and residing in a county with a higher poverty level. Overall, patients with insurance were less likely to present with metastatic disease (16.9%) than those with Medicaid (29.1%) or without insurance coverage (34.7%) and more likely to present with local disease (60.8%) compared with those with Medicaid (42.2%) or without insurance coverage (40.3%) (P < .001). The unadjusted 3-year cause-specific survival among those with insurance was 83.4%, compared with 64.1% among those with Medicaid and 61.8% among those without insurance coverage (P < .0001). In a Cox regression that adjusted for age, race, gender, marital status, residence (urban vs rural), percentage of county below federal poverty level, stage (local, regional, distant), and receipt of definitive treatment, patients were more likely to die of their disease if they had Medicaid (hazard ratio [HR] = 1.53; 95% confidence interval [CI], 1.49–1.56) or did not have insurance coverage (HR = 1.52; 95% CI, 1.48–1.57) compared with those with insurance (P < .001).

Conclusions: Among patients with the top 10 causes of cancer death, those with Medicaid or without insurance were more likely to present with advanced-stage disease. This group also experienced worse survival when controlling for demographic information, stage at diagnosis, and receipt of definitive treatment. The expansion of insurance coverage would be expected to substantially impact the presentation and outcome of cancer in the US.

Proceedings of the 96th Annual Meeting of the American Radium Society - americanradiumsociety.org