Impacts of the Affordable Care Act on Utilization of Emergency and Non-Emergency Medical Services

The Affordable Care Act (ACA) aimed to achieve nearly universal health insurance coverage in the United States through a combination of regulations, mandates, subsidies, exchanges, and Medicaid expansions. We use data from the Medical Expenditure Panel Survey (MEPS) to investigate the impacts of the ACA on the health care utilization and expenditures of non-elderly adults. A difference-in-difference-in-differences strategy separately identifies the effects of the ACA’s expansions of private and Medicaid coverage by leveraging variation in states’ Medicaid expansion decisions and individuals’ pre-ACA insurance status. Intuitively, impacts of the ACA’s insurance expansions should be concentrated among those who lacked insurance prior to the law, and such individuals are more likely to be affected in states that participated in the Medicaid expansion. Similar methods have been used to study the ACA’s effects on outcomes such as health insurance coverage, access to care, risky health behaviors, and self-assessed health. However, they have not been previously used to investigate impacts on health care spending. Theoretically, the net effect on spending is ambiguous. On one hand, insurance lowers the effective price of care, which should increase utilization across-the-board. On the other hand, insurance improves access to primary and preventive care, which could potentially reduce use of expensive emergency services. The MEPS data allow us to examine a wide range of outcomes, including frequency of and expenditures on physician and non-physician office visits; inpatient, outpatient, and emergency room hospital visits; and prescription drugs. Of particular interest are combined effects on non-emergency services, emergency services, and all services. For the expenditure outcomes, we further subdivide by payer, including out-of-pocket, private insurance, and Medicaid. We also conduct subsample analyses by age, race/ethnicity, education, income, gender, marital status, presence of children in the home, and area rurality. Using these results, we compute the extent to which the ACA reduced disparities on the bases of race/ethnicity and socio-economic status in both Medicaid expansion and non-expansion states. Finally, we estimate instrumental variables models to determine the impacts of private coverage and Medicaid coverage on the various utilization outcomes.