The Medicaid GLP-1 coverage study found a sharp fall in prescriptions after four states stopped paying for weight-loss treatment. Medicaid-paid prescriptions fell by about 84% relative to states that kept coverage, within one quarter. That measures paid prescriptions, rather than how many people stopped treatment.
Medicaid is the joint federal-state health insurance program. GLP-1 medicines act on a gut-hormone signal involved in appetite and blood sugar.
What the Medicaid GLP-1 coverage study found
As of November 20, 2026, the day this article was written, a JAMA research letter published November 16 reports the change. Bryant Shuey and colleagues compared four states that ended weight-loss coverage on January 1 with ten that maintained it. The four were California, New Hampshire, Pennsylvania and South Carolina.
The researchers used state-reported prescription counts, including payments made directly by Medicaid and through plans run by insurers. The weight-loss group included Wegovy, containing semaglutide; Zepbound, containing tirzepatide; and Saxenda, containing liraglutide.
These are quarterly prescriptions per 100,000 Medicaid members, averaged across states:
| State group | October–December 2025 | January–March 2026 |
|---|---|---|
| Four states ending coverage | 2,914 | 421 |
| Ten states maintaining coverage | 3,006 | 2,758 |
The researchers used difference-in-differences: comparing the change in states that cut coverage with the change in states that kept it. That comparison produced the 84% relative decline. The estimated difference was 2,245 fewer quarterly prescriptions per 100,000 members.
This is an average across states. A large state didn't get more weight in that calculation simply because it had more members.
What this means when Medicaid stops paying
A coverage decision can interrupt access even when the prescription itself hasn't changed. Medical suitability and payment are separate questions. The Medicaid coverage history and state overview explains why the answer varies by state.
The letter notes that prescriptions didn't fall to zero. It raises other medical uses, such as obstructive sleep apnea, repeated airway blockage during sleep, as one possible explanation. It cannot identify why each remaining prescription was paid.
As of November 20, 2026, the day this article was written, California's April 1, 2026 notice describes case-by-case review. It lists Wegovy for heart and blood-vessel disease and Zepbound for obstructive sleep apnea. This requires prior authorization, the program's review before it pays. Members younger than 21 have a separate medical-need review for weight-loss treatment under federal child-health protections.
Those examples aren't automatic approvals. The prescriber documents the actual condition, and the state program decides whether its coverage requirements are met. Having the same active ingredient as another brand doesn't establish the same coverage.
What the study cannot tell you
The records don't follow individual people. They cannot show prescriptions bought with private payment, changes in weight or health, or what happened beyond the first quarter after coverage ended.
The University of Pittsburgh's announcement explains the estimate of 175,500 to 526,600 fewer people potentially filling these prescriptions. It assumes one to three prescriptions per person per quarter. It is an estimate, not a count of people known to have stopped treatment.
Changes in prescriptions for brands used for diabetes, a condition with high blood sugar, did not differ clearly between the state groups. That doesn't establish what happened to each patient's care. Other policy changes, including different authorization requirements, could also have contributed to the findings.
As of November 20, 2026, the day this article was written, a Cureus study published November 6 also reports a large decline. Fabrice Nyambod and colleagues found an 84.7% relative drop after five state withdrawals, compared with twelve continuing states. It used the same public Medicaid data, so it isn't an independent set of patients confirming the result.
The guide to stopping semaglutide covers separate evidence about appetite and weight after treatment ends. Those outcomes weren't measured here.
What the authors say about compounded medication
The JAMA authors write that “compounded formulations have considerable safety concerns.” The line appears where they discuss options for patients who lose coverage; in the same sentence, they write that direct-to-consumer pharmacies for GLP-1 access “may not be financially feasible.”
Promise is a telehealth company through which compounded GLP-1 medication is prescribed, so it has a commercial interest in this subject.
Semaglutide prescribed through Promise is dispensed by a licensed U.S. compounding pharmacy as a compounded medication, which is different from an FDA-approved product: the formulation offered here is not FDA-approved. Compounding means a pharmacy prepares a medication for an individual prescription. FDA explains that it does not verify compounded drugs' safety, effectiveness or quality before marketing.
A licensed provider may still prescribe — that decision is between you and your doctor.
What to watch before the next refill
The useful documents are the written denial, the current state coverage rules and the prescriber's treatment plan. Three questions can make that conversation more concrete:
- Does a still-covered medical use apply? The diagnosis must reflect the person's actual health needs. The prescriber supplies the records; the state program reviews coverage.
- Is there an exception or appeal route? California's member notice describes a State Hearing after a coverage denial. It generally allows 90 days from the notice to request one. The denial's deadlines and any rules for continuing treatment during review matter.
- Is a manufacturer self-pay program an option? Such programs exist for branded prescriptions; the TrumpRx guide identifies the channels. Eligibility and affordability still need checking. A cash purchase isn't Medicaid coverage.
Rationing or stretching a prescribed dose to make it last is not a substitute for a plan with the prescriber. Dose changes belong in that conversation.
At Promise, a licensed provider reviews every request, and not everyone qualifies. Later Medicaid data and studies following individual patients are needed to show whether the prescription losses persisted and what happened to health.





