Health Insurance and Disparities: More than Access

PROJECT SUMMARY Cancer health disparities in the US persist, particularly between racial/ethnic groups and urban/rural residency, despite nearly 92% of Americans having health insurance coverage. In 2021, Medicaid covered ~19% of people; 54.2% were covered by employer-based insurance; Medicare covered another 18.4%; 10.2% were covered by direct purchase plans, and the remaining 3.5% were covered by military plans. Previous research has focused on patients without insurance, but factors associated with coverage, such as cost sharing, out-of-pocket limits, provider networks, and drug formularies can vary widely, potentially contributing to substantial disparities in access to and costs of care. Progress toward ameliorating disparities is hindered by the absence of multiple-payer data that could inform policies. Population-based registries do not adequately capture health insurance coverage or the full cancer care experience. Moreover, few data sources capture OOP costs that can both influence treatment decisions and lead to financial hardship. We address this gap by building on an established linkage between the Colorado Central Cancer Registry (CCCR) and the Colorado All-Payer Claims Data (APCD). APCDs are an emerging source for comprehensive data on healthcare utilization within a state. We supplement these data with plan-level information and census and health services files to better understand the social determinants of health context in which patients reside. The resulting data will contain person, treatment, provider/facility, plan, and geographic data for the years 2012-22 that allows longer-term survivorship assessments. Our specific aims are to: Aim 1: Compare differences in outcomes by race/ethnicity and urban/rural residency, controlling for plan type, patient demographics, market and provider characteristics, and social determinants of health. Aim 2. Decompose the relative contribution of plan type, patient demographics, market and provider characteristics, and social determinants of health in explaining observed outcome differences. Aim 3: Estimate the burden of OOP expenditures after a cancer diagnosis by plan type and by race/ethnicity and urban-rural residency.