A recent research letter published in JAMA by Chetty et al describes the experience of a secret shopper in obtaining GLP-1 receptor agonists (GLP-1 RA) from online sources. The purported patient was a 28 year old male with a BMI of 35 and a history of hyperlipidemia, hypertension, prediabetes, and reported sleep apnea. A lengthy and highly detailed questionnaire published as a supplement to the research article provided specific questions that recorded information from the online provider’s prescribing practice. For example, the criterion for the online provider asking about any prior medical condition was fulfilled if there was at least one question about any prior medical condition. Responses were sought about whether the simulated patient received personalization of care that might prompt modifications of the dosage, such as “Is it important to you to minimize possible nausea and GI side effects?” or “During your weight loss journey would you like your doctor to be able to tailor your dosage to your personalized needs and goals?.” The validity of the patient’s obesity was assessed by a photograph that included either the face, upper body, or full body or a photograph of the patient on a scale. Payment information either before or after the prescription was assessed. The median time of the clinician video or call, when it occurred, was recorded.
Of 49 websites that prescribed GLP-1 RAs, 17 sold compounded medication only, 5 sold branded only, and 27 sold both. 45 websites prescribed and 34 mailed the medication. Of the 4 non-prescribing websites, 2 required blood work, one denied medication based on the finding of an existing prescription, and 1 prescribed but later withdrew a prescription based on a disparity between the patient’s photograph and reported weight.
Almost all online prescribers assessed weight and height, medical conditions, medications and allergies, contraindications, and weight loss goals. Approximately half of the prescribers asked about eating disorders and the patient’s last medical evaluation, but only a third requested blood pressure, glucose, cholesterol or triglyceride values. One-quarter of the websites required a photo of the face, and 20% required a full body photo. About one-third required a video with a clinician, but only half of these were MDs or DOs; 20% were with an advance practice registered nurse. The mean time to prescriptions was one day or less, and one was issued in less than 5 minutes. The median supply of medication was one month and mean cost was $217. Almost 90% required payment information before the prescription and two thirds required payment before the prescription was dispensed.
Because an estimated 20% of US adults are currently receiving a GLP-1 RA obtained online, Chetty et al’s research letter provides enormously important insights and raises concerns relevant to patient care. The overriding questions about the ability to pay, the limited time for consultation with a provider (if available), and the observation that when consultation occurred, the interaction appeared cursory raises concern about the quality of care. One consultation lasted only 5 minutes, and the median for all consultations was 9 minutes. These findings suggest that many of these online prescribers are running a medication mill with money rather than quality patient care as the object. Only half of the prescribers asked about eating disorders, and less than half required a visible confirmation of obesity. These findings do not exclude the possibility that prescriptions are being sought for cosmetic reasons or to potentiate an eating disorder, like anorexia. Potential adverse effects of the medications, like nausea or vomiting and where and when to seek support are not routinely addressed. The importance of follow-up or engagement of the patient’s primary care provider is not routinely counseled. Finally, the 503A compounding facilities, which are licensed by states, are intended to provide drugs for specific patients, and may lack the capacity, oversight, and quality of the production of drugs for population use, as required by 503B facilities. Their use heightens concerns about the composition of online compounded GLP-1 RAs.
As Chetty and his co-authors conclude, “… while direct-to-consumer platforms have increased patient access to GLP1 RAs, these findings suggest that some do so through impersonal care that prioritizes quick prescriptions over comprehensive care.” To understand the consequences of online prescribing that Chetty et al describe, we will need a system to document the potential adverse health effects of online care and the composition of the product that is being prescribed.