A Yale secret shopper study in JAMA found two thirds of online GLP 1 prescriptions (Ozempic, Wegovy) skip clinician interaction, exposing a ready made telehealth middleman layer.
A Yale researcher spent the last several months posing as a patient at nearly 50 telehealth sites selling GLP-1 weight-loss drugs, the class behind Ozempic, Wegovy, and Mounjaro. The result, published Monday in JAMA (the Journal of the American Medical Association), is the first peer-reviewed benchmark for a problem clinicians have been describing anecdotally: it has become "extraordinarily fast and easy" to walk away with a prescription, and two-thirds of the time the prescription was issued without any clinician-patient interaction at all.
The reason is structural, not accidental. A turnkey telehealth middleman layer, documented by Wired and confirmed by the Yale study, has lowered the clinical bar for almost any operator to sell semaglutide or tirzepatide. The same infrastructure that delivers a prescription in minutes also determines whether anyone actually screens for the conditions that make these drugs risky.
GLP-1 agonists work by mimicking a gut hormone to slow digestion and blunt appetite. They are approved for type 2 diabetes and, at higher doses, for weight management. They are also associated with documented safety risks: pancreatitis, severe gastrointestinal events, thyroid concerns, and dangerous interactions with other medications. The prescribing standard matters because the patient can be harmed even when the prescription itself is legitimate.
The Yale secret-shopper methodology was blunt by design. A researcher filled out intake questionnaires at 49 telehealth sites without revealing a relevant history, including past pancreatitis, current use of contraindicated medications, and an eating-disorder diagnosis, then recorded what happened. Two-thirds of sites issued a prescription without any clinician-patient interaction, and the sites that did schedule a video or phone visit rarely probed the red flags the shopper had built into the intake form.
Researchers cannot say from this study how many patients were actually harmed. But they can say what the front door of online GLP-1 care looks like for a consumer who shows up with the exact risk factors a clinician should catch.
Compounded GLP-1s complicate the picture further. When the FDA declared GLP-1 shortages starting in 2022, compounders were allowed to produce semaglutide and tirzepatide under specific regulatory pathways (503A for traditional compounders and 503B for outsourcing facilities). KFF Health News reports that this market has since boomed alongside telehealth prescribing. Compounded versions are legal in that framework; they are not FDA-approved, and the agency has warned repeatedly that dosing errors in compounded products can produce concentrations far outside the labeled range. Yale's separate advertising study found compounded-drug marketing can mislead consumers about what they are buying.
The telehealth industry's growth is documented; what is missing is regulatory parity with in-person prescribing. STAT+ reported that the broader boom in telehealth GLP-1 prescribing is outpacing safety and oversight guardrails, and the new Yale study is the first peer-reviewed quantification of how thin those guardrails have become.
What a clinically defensible online evaluation should include, given what these drugs can do:
The Yale study makes the gap concrete: what these drugs require, and what online care currently delivers. Telehealth has real value for patients with limited access to specialists; the standard for these specific drugs has simply slipped.
The next milestone to watch is whether the FDA or state medical boards act on the JAMA publication. Until then, the burden of clinical guardrails is being carried almost entirely by the consumer standing at the intake form.