Medicaid expansion status and its impact on racial and ethnic disparities in access to insurance coverage

Background: Despite widespread evidence that Medicaid expansion states showed greater reductions in uninsurance and better health outcomes than non-expansion states across all demographics, there is limited research that examines the success of these states in addressing racial/ethnic disparities in insurance coverage. Research Question: This study seeks to examine how measures of racial/ethnic disparities within the 50 States and the District of Columbia were differentially impacted across the US, identify the impact of Medicaid expansion status on these trends, and to examine whether these trends are homogeneous across expansion states. Methods: Rates of uninsurance for adults aged 18-64, stratified by self-reported race and state residency, from the years 2011-2016 were collected from the American Community Survey and used to examine trends in uninsurance across racial/ethnic groups and quantify measures of disparities in coverage using relative and absolute measures over time. Using these measures, we examined differences between expansion and non-expansion states, in aggregate, and examined the distributions of these values to assess homogeneity of these measures within the expansion states and within non-expansion states. Multiple difference-in-difference (DD) models were developed to assess the causal impact of expansion status on any differences in the trends of these values over time, between expansion and non-expansion states. Results and Discussion: We found that there were statistically significant reductions in uninsurance across all racial/ethnic groups (p < 0.001 in all cases), with larger insurance gains in expansion states. Black and White populations showed comparable absolute reductions in uninsurance, with Hispanic populations showing smaller decreases. Disparities in coverage for both Black and Hispanic populations were observed using both relative and absolute measures (p < 0.001 in all cases). The relative measures showed no change in disparities for Black populations as a result of the ACA, regardless of expansion status, while Hispanic populations saw a worsening of disparities in expansion states (p < 0.001), but no change in non-expansion states. Absolute measures showed statistically significant reductions disparities in Black expansion populations (p < 0.01) and in Hispanic populations, regardless of expansion status (p < 0.05); but no change in Black disparities in non-expansion states. The DD models showed that there was no statistically significant impact of expansion status on the differences in trends between expansion and non-expansion states. Finally, there was widespread heterogeneity in these measures within expansion states and within non-expansion states. Conclusions: These results suggest that racial/ethnic disparities in insurance coverage continue to persist nationally, and were not significantly different in states that adopted the Medicaid expansion. Future research should examine states with the smallest disparities and those with the greatest improvement in these measures, to develop logic models for successful implementation of ACA programs and guide future initiatives that address these disparities and improve coverage overall.Bachelor of Science in Public Healt