Of 79 websites selling compounded GLP-1 drugs, 29 said or implied that the products were FDA approved. They are not. That finding, from a study of compounded semaglutide advertising and its tirzepatide and liraglutide equivalents, sits beside another: 39 of the sites did not report the full set of adverse effects, warnings and precautions, and contraindications.

The paper is "Online Advertising of Compounded Glucagon-Like Peptide-1 Receptor Agonists," published in JAMA Health Forum on January 17, 2025. Its authors are Ashwin K. Chetty, Reshma Ramachandran, Joseph S. Ross and Alissa S. Chen of Yale, and Mahima Chillakanti of the David Geffen School of Medicine at UCLA. We read the full open-access text on PubMed Central.

Two of the authors ran searches on Google Shopping between July and September 2024, using brand and generic drug names, and took every website that appeared in Google's sponsored results. They found 98 unique sites selling any GLP-1 drug. Seventy-nine sold a compounded version, or a prescription for one, and those went into the analysis.

All 79 sold compounded semaglutide. Fifty-seven (72.2%) sold compounded tirzepatide and three sold liraglutide.

What the websites said, and didn't

Start with the word "compounded." Sixty-eight sites (86.1%) used it at least once, which leaves 11 (13.9%) that never disclosed the drugs were compounded at all. Only 13 (16.5%) said so consistently, by the authors' definition of always calling the product compounded within the same sentence or phrase. Seven sites called the drugs generic.

On approval status, 34 sites (43.0%) said the compounded drugs were not FDA approved. The 29 that said or implied the opposite come to 36.7%.

Risk information was patchy rather than absent, and the difference is worth getting right. Sixty-eight sites mentioned adverse effects of some kind. Fifty mentioned contraindications and 49 mentioned warnings and precautions. The 49.4% figure is the share that failed to give all three.

Then efficacy. Sixty-five sites (82.3%) cited clinical trial evidence for their efficacy claims, and 32 (40.5%) made an efficacy claim that does not appear in the label of the FDA-approved branded drug.

Brand names were everywhere. Sixty-two sites (78.5%) used one in their advertising, and 50 (63.3%) used a phrase such as "same active ingredients as" the branded product.

The median first-month price, discounts included, was $231 for compounded semaglutide and $330 for tirzepatide. A median is the middle value; the middle half of semaglutide prices ran from $189 to $294. Fifty-two sites displayed a certification mark, and for 50 of them it was LegitScript's.

Limits, as the authors give them

Two limitations are named. Search results varied by location, so another searcher might have seen a different set of sites. And because the reviewers read everything, subpages and blogs included, the paper "likely overestimated the amount of information that typical consumers review." Our reading: a real shopper probably saw less of everything the reviewers counted, disclosures included.

Several authors disclose grants or fees outside the submitted work, including FDA grants to Ramachandran and Ross.

The authors conclude that the sites "often partially informed and sometimes misinformed potential consumers," and say FDA could require sellers to disclose and define compounding, including the lack of FDA approval. They also propose unique naming conventions for compounded drugs and greater FDA authority over misleading advertising of them.

What it means for a practice

That is where the study stops. What follows is our analysis, not the authors'.

Three cautions first. These were websites surfaced by Google Shopping ads, not med spas as such: 56 of the 79 sold online only, 17 sold online and in person, and six sold in person only. The paper never identifies a business type. The data are also old. The searches ran in mid-2024, when the authors describe GLP-1 shortages as ongoing; FDA wrote in an update dated April 1, 2026 that tirzepatide and semaglutide "do not currently appear on the 503B bulks list or on FDA's drug shortage list," and Guidepoint Qsight's January 2026 industry deck lists "GLP-1 shortages end" among its 2025 highlights. The prices are history. And nothing here is guidance on prescribing.

What has held up is the list of things the reviewers looked for, because it reads like a regulator's checklist. FDA's consumer page on unapproved GLP-1 drugs, marked current as of October 1, 2026, says compounded drugs "are not FDA approved" and tells consumers to treat it as a red flag when a company "makes claims such as the compounded drug is the same as an FDA-approved drug." Most of the sites in this study used a same-ingredients line.

An owner with a weight-loss page can run the authors' form against it in an afternoon. Does every mention say compounded? Is the lack of FDA approval said outright? Is "generic" anywhere on the page? Does any efficacy claim go beyond the branded label? For a sense of how closely the agency reads promotional copy, see our report on its September 2026 untitled letter to Evolus over an Instagram reel.

The page is advertising. Have it reviewed like advertising.

Sources