After two decades of escalation, drug overdose deaths declined for three years straight from 2023 to 2025. From 2023 to 2024 alone, the decline was nearly 27 percent.
That progress is real but fragile. While disruptions in the illicit fentanyl supply were a major contributor to the decline in overdose deaths, the evidence suggests multiple factors worked together to make it possible. The historic expansion of Medicaid-funded treatment was also a potentially important enabling factor.
But Medicaid is facing significant cuts. The One Big Beautiful Bill Act is expected to strip Medicaid coverage from approximately 1.6 million people with substance use disorders. Because Medicaid covers nearly half of nonelderly adults with opioid use disorder (OUD), coverage losses of this scale are a major threat to the primary payer for addiction treatment.
Recent Urban Institute research, funded by the Foundation for Opioid Response Efforts (FORE), underscores what’s at stake. We analyzed data on Medicaid-covered prescription fills for three medications used to treat OUD and reverse opioid overdose—buprenorphine, naltrexone, and naloxone—from 2010 through 2025. Our analysis, which focuses on buprenorphine specifically, can help policymakers at every level make more informed decisions about upholding and enhancing access to effective medications for substance use disorders in Medicaid.
Why we focus on buprenorphine
Buprenorphine treatment is one of the two opioid-agonist medications associated with an 80 percent lower risk of opioid overdose death. It is, by far, the most common medication treatment for OUD in Medicaid, making up 57 percent of such prescriptions in 2022. (Methadone made up 29 percent, and naltrexone made up 6 percent.)
Buprenorphine also accounts for the majority of the prescription fills we examined. Nationally, Medicaid-covered buprenorphine fills rose roughly 10-fold—from 936,000 to 9.4 million—between 2010 and their peak in 2021, before easing to 8.2 million by 2025.
Source: Urban Institute analysis of Medicaid State Drug Utilization Data, 2010–25, accessed via Medicaid.gov, June 22, 2026.
Notes: OUD = opioid use disorder. States are categorized as having either a higher or a lower OUD rate based on whether their average estimated OUD rate over 2010–25 was above or below the median across states in its Medicaid expansion group. For complete methods, see Clemans-Cope and Payton (2026).
States that have expanded Medicaid and have higher opioid use disorder rates have higher numbers of Medicaid-covered buprenorphine fills
To understand differences in Medicaid-covered buprenorphine fills across states, we divided states into groups by whether and when they expanded Medicaid under the Affordable Care Act (ACA) and whether they have high or low rates of OUD. We categorized each state as having a high or low OUD rate by comparing the state’s average estimated OUD rate over 2010–25 with the median OUD rate for all the states in its Medicaid expansion group over the period. If the state OUD rate is below the group median, we categorize it as having a lower OUD rate. If the state OUD rate is higher than the group median, the state has a higher OUD rate.
The increase in prescription fills was overwhelmingly concentrated in states that expanded Medicaid under the ACA, particularly those with higher OUD rates. States that expanded Medicaid early, before 2019, accounted for about 80 percent of all Medicaid buprenorphine fills throughout the 16-year study period. Among these states, those with the largest increases had the greatest OUD burdens.
These patterns demonstrate Medicaid’s important role in financing access to addiction treatment over the past decade. In 2021, an estimated 61 percent of Medicaid enrollees with OUD had gained their coverage through ACA expansions, and by 2023, Medicaid was financing more than half of all OUD medications nationally.
Buprenorphine fills per Medicaid enrollee estimated to have OUD vary significantly across states but have likely increased over time
To understand the extent to which Medicaid enrollees with OUD are accessing medication treatment, we examine buprenorphine fills relative to the number of Medicaid enrollees estimated to have OUD.
The differences between the state groups we examined are wide: In 2025, buprenorphine fill rates per 1,000 Medicaid enrollees estimated to have OUD varied roughly 11-fold across the state groups. By 2025, states that had expanded Medicaid later (2019–23) and had lower OUD rates had the highest rate of any subgroup (10,974 per 1,000), followed by states that expanded early and have higher OUD rates (9,347). The states with the lowest rate of Medicaid-covered buprenorphine fills per 1,000 Medicaid enrollees were those that had not expanded by 2024 and had lower OUD rates, reaching only 985 fills per 1,000 enrollees in 2025.
Sources: Urban Institute analysis of Medicaid State Drug Utilization Data, 2010–25, accessed via Medicaid.gov, June 22, 2026; American Community Survey data, 2010–16, from the US Census Bureau; and Centers for Medicare & Medicaid Services Performance Indicator Medicaid Enrollment data, 2017–25; National Survey on Drug Use and Health small-area estimates, 2022–23, from the Substance Abuse and Mental Health Services Administration; State Health Access Data Assistance Center analysis of CDC WONDER mortality data, 2010–23; and Vital Statistics Rapid Release Provisional Drug Overdose Death Counts, 2024–25.
Notes: OUD = opioid use disorder. States are categorized as having either a higher or a lower OUD rate based on whether their average estimated OUD rate over 2010–25 was above or below the median across states in their Medicaid expansion group. For complete methods, including how enrollees with estimated OUD are computed, see Clemans-Cope and Payton (2026).
As a rough benchmark for estimating treatment rates, we divided the total number of Medicaid-covered buprenorphine prescription fills nationally in 2025 by nine prescription fills for medications to treat OUD per treated enrollee per year. This implied an estimated 59 percent of enrollees with OUD received buprenorphine last year—an increase from 2019, when the rate was about 36 percent, using the same rough calculations.
Though these are implied, approximate rates that are sensitive to the OUD prevalence and enrollment estimates behind them, the broader direction of rising treatment concentrated in states that have expanded Medicaid is consistent with other research. In 2019, Mojtabai and colleagues studied the impact of Medicaid expansion on specialty outpatient treatment and found rates of medication treatment for OUD rose from 39.1 percent in 2010–13 to 50.2 percent in 2014–16 in Medicaid expansion states but remained essentially flat in nonexpansion states. Similarly, Saloner and colleagues found that, among Medicaid enrollees diagnosed with OUD in West Virginia (an early expansion state), the buprenorphine treatment rate rose from roughly one-third to over three-quarters from 2014 to 2016.
Research from Siegal and colleagues this year suggests OUD treatment gains were greatest in states that expanded Medicaid in 2019 or later. But the picture is not uniform. In 2021, Olfson and colleagues found that treatment increases in expansion states were similar to those in nonexpansion states, after accounting for payer shifts. And in 2020, Gertner and colleagues found that prescriber supply often limited increases in treatment rates.
Policy choices could protect—or erode—access to life-saving opioid use disorder treatment
The current policy environment makes our findings especially urgent. Federal funding reductions under the One Big Beautiful Bill Act are expected to reduce federal Medicaid spending by more than $900 billion over the next decade. And Urban researchers have estimated that the law’s Medicaid expansion work requirements and six-month eligibility redeterminations could reduce Medicaid enrollment by 4.9 million to 10.1 million people in 2028.
On top of this, in April 2026, SAMHSA issued two Dear Colleague letters that pull back from evidence-based opioid treatment and overdose prevention, directing grantees to discuss discontinuing medication among patients receiving medication treatment for OUD and barring federal funds for fentanyl test strips.
Though the One Big Beautiful Bill Act includes some narrow exemptions from Medicaid cuts for people with substance use disorders, research shows that many people with OUD who remain eligible will likely struggle with the complex documentation required to prove eligibility. If enrollees lose coverage as a result—even briefly—it could interrupt their access to buprenorphine and immediately increase the risk of overdose death.
To uphold the historic OUD treatment gains in Medicaid and sustain the decrease in overdose deaths in recent years, policymakers should preserve Medicaid coverage for people with opioid use disorder (including protecting populations from work requirement disenrollments) and maintain federal funding for SAMHSA addiction treatment and harm-reduction programs. In addition, states that have not yet expanded Medicaid should consider expansion as a tool for closing treatment gaps.
Ultimately, people’s lives depend on Medicaid coverage, access to medication for OUD, and the broader behavioral health services, care coordination, and social supports required for lasting recovery. The provision of effective, Medicaid- and SAMHSA-funded health care is the baseline needed to protect—and continue—a decade of hard-won progress in treating addiction.
