The Effect of Public Insurance Expansions on Substance use Disorder Treatment Utilization: Evidence from the Affordable Care Act

This study explores the effect of Medicaid expansions under the ACA on substance use disorder (SUD) treatment utilization (including Rx drug abuse treatment) among low-income adults. This population has historically had little access to insurance but has elevated prevalence of SUDs. We leverage variation in public insurance eligibility generated by U.S. states’ decisions to expand Medicaid to these adults between 2010-2015.

Problems related to substance use are a public health concern in the U.S. In 2015, over 20 million individuals in the U.S. met diagnostic criteria for an SUD. SUDs are characterized by clinically significant impairment related to the use of alcohol or psychoactive drugs. The harms related to substance use are believed to be a leading contributor to the poor health of Americans relative to residents of other high income countries. Substance use contributes to a range of social problems including healthcare costs, crime, and reduced labor market productivity. These costs are estimated to be $519B per year in the U.S. While the effectiveness of SUD treatment is well-established, only one-tenth of individuals who meet the diagnostic criteria for SUDs receive treatment in any year. Although there are myriad reasons for failure to receive treatment, lack of insurance coverage and inability to pay are key barriers.

The ACA provides an opportunity to increase treatment utilization among individuals with SUDs and to alter the financing of such treatment. Medicaid expansion provides millions of previously uninsured adults with insurance coverage that includes SUD treatment services. Due in large part to the substantial increases in the number of covered individuals and services, some scholars argue that ‘no illness will be more affected than substance use disorders’ by the ACA.

We study the effects of Medicaid expansion under the ACA on treatment utilization and use of Medicaid as source of payment for such treatment. We leverage administrative data drawn from the Treatment Episodes Data Set (TEDS) between 2010-2015, and the Medicaid State Drug Utilization Data (SDUD) between 2011 and 2015. TEDS includes nearly nine million admissions to specialty SUD treatment facilities while SDUD captures all prescriptions for medications purchased at retail and online pharmacies used to treat SUDs in outpatient settings financed, at least partially, by Medicaid. We couple these administrative data sets with differences-in-differences regression models.

Our results suggest that states expanding Medicaid experienced no increase in admissions to specialty SUD treatment post-expansion relative to non-expanding states. We find that post-expansion prescriptions for medications used to treat SUDs paid for by Medicaid increased by 43% in expanding states relative to non-expanding states. Among patients receiving specialty treatment in expanding states, we find that Medicaid coverage and use of Medicaid to pay for treatment increased substantially. For example, we find that the probability of using Medicaid as a source of payment increased by 75% among patients with a decline in payments from state and local governments. In a supplementary analysis, we examine changes in fatal alcohol poisonings and psychoactive drug overdoses. We do not find any evidence of changes in such deaths.