GLP 1 access pathways study finds one in three users go outside insurance and regular pharmacies

About one in three Americans who have used a GLP 1 drug got it outside the usual doctor, pharmacy and insurer route, a new preprint on GLP 1 access pathways estimates. Researchers at the University of Southern California and RAND followed 9,150 adults from March 2024 to October 2025 and found that use rose from 8.2% to 12% of adults. The growth came mostly from people taking the drugs for weight loss, and many of them paid without insurance or filled through compounding, online or foreign pharmacies. The paper has not yet been peer reviewed.

What the study measured

The work comes from the Understanding America Study, a national internet panel run by the USC Center for Economic and Social Research. The team asked the same adults about GLP 1 and GIP drugs in three waves, in March 2024, December 2024 and October 2025, and weighted the answers to match the US adult population. Of the 9,150 people included, 1,274 reported current or past use and 945 more said they had never used the drugs but planned to. The preprint, titled GLP 1/GIP Uptake, Indication, and Access Pathways Among US Adults, was posted to medRxiv on September 2 by Ritika Chaturvedi and nine coauthors. It was funded by the National Institute on Aging, the Social Security Administration and the National Library of Medicine.

The team split access into two groups. A route counted as traditional when a person had a prescription from a regular clinician and filled it at an ordinary pharmacy. It counted as nontraditional when the prescription came from an online provider or a medical spa, when the fill came from a compounding, online or foreign pharmacy, or when there was no prescription at all. That definition matters for anyone who pays for or tracks these drugs. Insurers, employers and public health agencies mostly count use through pharmacy claims, and claims only capture the traditional route. Anything bought for cash through a telehealth site or a compounder can stay invisible to them, which is the gap this survey was built to measure.

One in three users outside the usual route

By October 2025, about 12% of US adults had used one of these drugs, which the authors put at roughly 32 million people. About one in three users, or around 9 million adults, reached treatment through a nontraditional route, according to the paper. Those users were concentrated among people taking the drugs for weight loss. Within that weight loss group, 33% had no conventional prescription and 41% filled through compounding, online or foreign pharmacies. Almost every person taking the drugs for type 2 diabetes used the traditional route instead. Weight loss use rose from 4.1% to 5.6% of adults between December 2024 and October 2025, while diabetes use stayed flat at about 5.3% to 5.4%.

About one in three users, or around 9 million adults, reached treatment through a nontraditional route, according to the paper.

People on the nontraditional route were about five times as likely to name a product that was not on the list of branded medicines, 19.8% against 4.1%, which the authors read as a sign of compounded versions. The survey cannot confirm what was in those vials, because the answers are self reported. Compounded copies are not FDA approved, and regulators have warned about online sellers and ads for them, as in the FDA warning letters over GLP 1 ads sent to online clinics in 2025. Among nontraditional users, 11.1% reported no diagnosis linked to the drugs, which the authors flag as possible cosmetic use, compared with 1.4% of traditional users. The researchers did not verify those reasons.

Coverage gaps behind GLP 1 access pathways

Money is the clearest thread in the data. Some 43% of people taking the drugs for weight loss said they had no drug coverage for them, while 97% of people taking them for diabetes said they did. That split follows the way plans pay. Many insurers and public programs cover GLP 1 drugs for type 2 diabetes but limit or exclude them for weight loss, and several have pulled back further this year, as with Massachusetts insurers cutting weight loss cover. When a plan says no, a patient faces a cash price for a branded pen, and lower priced compounded or online options become the alternative many choose. The study does not show whether those routes changed health outcomes.

The table below sets out the main figures the authors report. They describe a market where the newest users, the ones paying for themselves, are also the least visible to the systems that set coverage rules. Semaglutide held about half of all use, down from 54%, with Ozempic at 41% and Wegovy at 9.1%. Tirzepatide rose from 22% to 30%, split between Mounjaro at 20.1% and Zepbound at 10.3%, and new starts on tirzepatide now outnumber new starts on semaglutide. Of everyone who had used one of the drugs, 64.3% were still taking it at the last survey and 35.7% had stopped. The authors note that weight loss users stopped more often than diabetes users.

What it means for payers and patients

The authors argue that claims data alone now undercount who is taking these drugs, because a large share of weight loss use never reaches an insurer. That has practical effects. A payer that judges the cost of covering obesity drugs from its own claims may miss members who already take them and pay elsewhere. Safety tracking that relies on prescription records can miss side effects in people using compounded or foreign products. The team calls for population surveys that follow every route, and it offers its data, which it calls GLIMMER, as a public resource for payers and providers. Those are proposals from the authors, not policy, and no agency has said it will act on them.

The study has clear limits. Use, the reason for use and the access route were all reported by the people themselves, so some answers will be wrong, and people taking the drugs for both diabetes and weight may be classed in only one group. The panel ends in October 2025, before the cash price cuts and newer direct sales programs that came later, so today's split between routes may differ. For patients comparing GLP 1 access pathways, the message is practical rather than alarming: a lower price outside insurance often means a product that is not FDA approved. Our page on compounded copies and online sellers follows the rules on that side of the market.

Conclusion

The preprint on GLP 1 access pathways suggests that about 9 million US adults took these drugs through routes that insurance records do not see, mostly people paying for weight loss without coverage. The figures are self reported and not yet peer reviewed, but they give payers and regulators a reason to look past claims data. Patients weighing a cheaper source should ask a clinician or pharmacist whether the product is approved.