Association for Public Policy Analysis & Management

Previous studies have estimated the population-level effect of health insurance coverage gains on ED utilization, but none has examined the longer-run trajectory of effects or focused on potentially avoidable ED utilization after the Marketplace and Medicaid expansion. We use two well-known algorithms to identify preventable and non-emergent ED visits in seven states using data from the Healthcare Cost and Utilization Project for the years 2008 through 2016. We use one-, two-, and three-year lags of insurance coverage in a county-level fixed effects model to identify whether prior changes in population-level coverage affect current ED use. We also estimate spatial regression models that incorporate geographic lags, allowing for the outcomes and health care supply and demand in bordering counties to factor into the estimation of effects of population-level insurance coverage on ED visit rates for non-elderly adults in a given county. We do not find compelling evidence to suggest delayed effects of coverage gains on reducing ED use generally or visits for non-emergent or preventable conditions. However, Medicaid expansion was associated with reductions in total ED visits per 1,000 non-elderly adults (b=-4.84, p<0.05), ED visits for ambulatory care sensitive conditions (b=-0.56, p<0.05), and ED visits for non-emergent or emergent but primary care treatable conditions (b=-2.99, p<0.05). Alternative payment models and focusing on social determinants of health, both through and outside of the health care system, are potential avenues for building upon coverage expansion to achieve the goals of more appropriate use of care and continued low growth in health care expenditures.