Abstract A008: Counterfactual analysis of the impact of insurance type on ovarian cancer mortality according to race/ethnicity; a population-based study of the California Cancer Registry, 2010-2019

Abstract Background: Ovarian cancer is the 5th leading cause of cancer death among U.S. females. Black and Hispanic females have greater mortality after an ovarian cancer diagnosis even when stage at diagnosis is considered. Health insurance coverage, which relies upon employment status and income, is strongly associated with extent of cancer treatment and survivability; underinsurance disproportionately affects minoritized racial and ethnic groups. Thus, we sought to quantify the potential improvement to race/ethnicity-specific population level mortality with improved insurance coverage. Methods: We identified adult females diagnosed with epithelial ovarian cancer (EOC) between 2010 and 2019 in the California Cancer Registry and extracted data on insurance, age at diagnosis, race/ethnicity, marital status, date of diagnosis, tumor stage, histology, vital status (through Dec 31, 2021), and date and cause of death. Insurance was classified as public (Medicaid, VA, Medicare among 65 years of age, and others), private (including Medicare among 65+ years of age), or uninsured. Race/ethnicity was classified as non-Hispanic (NH) Asian American, NH Black, Hispanic, or NH White. Mean survival time in days was estimated using inverse probability of censoring weighting (IPCW) negative binomial regression with stage at diagnosis, sociodemographic, and tumor factors as covariates. We used g-computation to estimate two counterfactual scenarios: the effect on mortality among the entire study population if everyone were privately insured (ATE), and the effect among the publicly insured if they were privately insured (ATU). Analyses were stratified by racial/ethnic group. Results: Among 14,938 people with EOC, there were 8,434 deaths. The study population comprised 3,519 with public insurance, 11,111 with private insurance, and 308 uninsured. In fully adjusted models, mean survival was 1,561 days (CI=1,539, 1,583) for the privately insured and 1,405 days (CI=1,368, 1,442) for the publicly insured—a difference of nearly 4 months. If everyone were privately insured the change in total mortality would be an additional 32.5 days (CI=21.6, 43.4) survived. Counterfactual increases in survival were observed for each racial/ethnic group and were greatest for Hispanic females (survival difference=90.8 days; CI=60.4, 123.2). Among those who were publicly insured the change in mortality would be an additional 126.7 days (CI=84.6, 168.9). Counterfactual increases in survival among publicly insured were again observed for every group and were greatest for Hispanic females, who had an estimated increase in survival or more than seven additional months (survival difference=214.3 days; CI=142.5, 289.8). Conclusions: Even after accounting for stage at diagnosis, insurance type was associated with substantial inequities in ovarian cancer mortality that varied across racial/ethnic groups and were strongest for Hispanic females. Improving cancer care and navigation for those who are publicly insured could substantially reduce racial/ethnic inequities in ovarian cancer mortality. Citation Format: Serge Atherwood, Sylvia Sudat, Alison J. Canchola, Aly Cortella, Kathryn E. Kemper, Su-Ying Liang, Mark Segal, Judith M. E. Walsh, Mindy C. DeRouen. Counterfactual analysis of the impact of insurance type on ovarian cancer mortality according to race/ethnicity; a population-based study of the California Cancer Registry, 2010-2019 abstract. In: Proceedings of the 18th AACR Conference on the Science of Cancer Health Disparities; 2025 Sep 18-21; Baltimore, MD. Philadelphia (PA): AACR; Cancer Epidemiol Biomarkers Prev 2025;34(9 Suppl):Abstract nr A008.