Medicaid unwinding linked to disruptions in opioid addiction treatment

The massive national effort to return the Medicaid enrollment process to its pre-pandemic rules starting in April 2023 may have disrupted the care of people receiving treatment for opioid addiction, a new University of Michigan study suggests. 

The researchers call this finding concerning, because the disruption may increase the risk of overdose or other negative impacts from opioid use disorder in states that were most aggressive in removing people from Medicaid coverage during the “unwinding” process in 2023. 

The study has implications for the current debate over the future of Medicaid funding, which may result in further changes in eligibility in some or all states. Although opioid overdose deaths have declined somewhat, they are still as high as they were early in the pandemic, with around 80,000 Americans a year dying from this cause. 

The new analysis, published in JAMA Network Open, looks at prescriptions for buprenorphine, a medicine that can help someone with opioid use disorder achieve and maintain recovery. It includes data from more than 569,000 adults across the country who had Medicaid coverage and received buprenorphine during the pre-unwinding period.

The researchers, led by U-M postdoctoral fellow Joanne Constantin, Ph.D., compared what happened with buprenorphine prescriptions, and how patients paid to fill them, in two groups of states.

The first group included the states where the percentage of people with Medicaid dropped the most from the month before unwinding began to the end of 2023: Arkansas, Colorado, Idaho, Kansas, Montana, New Hampshire, North Dakota, Oklahoma, Texas, Utah, West Virginia and Wyoming. 

The second group was states with the smallest percentage drops: California, Connecticut, Delaware, Hawaii, Illinois, Maine, Massachusetts, Minnesota, Nebraska, Nevada, Virginia and Wisconsin. 

A few states that expanded their Medicaid programs in other ways during 2023, or delayed the unwinding process, were not analyzed. Neither were data from people age 65, who also have access to Medicare drug coverage.

Key findings

The researchers found that patients were less likely to continue to fill buprenorphine prescriptions and more likely to pay for buprenorphine with cash or private insurance if they lived in states with the biggest versus smallest Medicaid enrollment drops.

Constantin and the paper’s senior author, Kao-Ping Chua, M.D., Ph.D., note that the majority of disenrollments during the unwinding were due to administrative issues, such as individuals not providing information about their income to their state by a deadline. 

They also remark that their analysis includes buprenorphine prescription information from all individuals in each state who received the medication through Medicaid during the years leading up to the pandemic and during its early years -- not just those who lost Medicaid coverage during the unwinding. 

That means the actual drops in prescriptions may be much larger among those who were disenrolled from coverage, which has been seen with other studies of care and prescription access among people who lost Medicaid coverage. 

“It’s important to study the effects of unwinding because it offers a natural experiment to look at how changes in Medicaid coverage affect use of potentially life-saving medications such as buprenorphine, which patients may need to take for months or years to achieve and maintain recovery,” said Constantin. 

Adds Chua, “This study is highly relevant to the ongoing debates about whether to cut federal Medicaid funding. If this funding is cut, many people will be disenrolled from Medicaid – just as they were during the unwinding process. Some of those people will be patients using buprenorphine and will be in jeopardy of having their access to buprenorphine blocked.”

The unwinding process focused on returning to the rules for determining eligibility for Medicaid in each state that were in effect before the pandemic – such as annual verification that a person’s income was near, at or below the federal poverty level or that they met other eligibility requirements. 

The future of Medicaid may change those eligibility criteria further if states receive less from the federal government or experience changes in their ability to fund the state’s share of the program from tax revenues. 

Nearly 72 million Americans are currently enrolled in Medicaid after the end of the unwinding, and an additional 7.2 million children are enrolled in the Children’s Health Insurance Plan or CHIP which also relies on Medicaid funding. 

Ten states have not expanded Medicaid under the Affordable Care Act, but those that have done so offer coverage to all individuals up to 138% of the federal poverty level, or about $21,000 for an individual and $36,000 for a family of three. 

In an invited commentary accompanying the new paper, Tulane University psychiatrist Ramin Mojtabai, M.D., Ph.D., writes, "If passed, the proposed legislations could lead to further Medicaid disenrollments, with potential consequences for the health and well-being of vulnerable adults with behavioral health conditions, such as opioid use disorders. By losing Medicaid coverage, these individuals are at increased risk of discontinuation of lifesaving medication treatments and of experiencing drug overdose. Safeguarding access to needed care for this vulnerable population requires concerted effort on the part of states and local health authorities and advocacy groups in the coming years." 

Chua is the director of, and Constantin a fellow at, the Susan B. Meister Child Health Evaluation and Research Center, or CHEAR. Both are members of the Department of Pediatrics at the U-M Medical School. Chua is also a member of the U-M Institute for Healthcare Policy and Innovation and serves on the leadership of the U-M Opioid Research Institute.

In addition to Constantin and Chua, the study’s authors are Genevieve M. Kenney, Ph.D., of the Urban Institute and Kosali Simon, Ph.D. of the University of Indiana, Bloomington. The study was funded by the National Institute on Drug Abuse of the National Institutes of Health (R01DA056438).

Reference: Medicaid Unwinding and Changes in Buprenorphine Dispensing, JAMA Network Open. 2025;8(5):e258469. doi:10.1001/jamanetworkopen.2025.8469