What GLP-1 Adverse-Event Reports Can—and Cannot—Establish

A report that someone was hospitalized after using a GLP-1 medication deserves attention. But it does not, by itself, answer three different questions: Did the medication cause the event? How often does that event happen among users? Does it tell us anything about a particular telehealth provider? Treating those as interchangeable questions can turn a legitimate safety concern into an unsupported statistic—or encourage someone to dismiss a warning because the evidence is incomplete.

FDA guidance provides a useful starting point: adverse-event reports lack the denominator needed to calculate a rate and do not automatically establish causation. This guide explains how to read those limits alongside concrete FDA concerns and public documentation from CoreAge Rx, Ro, Enhance.MD, Sesame, Walgreens and Found. The pages were consulted October 19, 2026; that is a research date, not a claim that every page was updated then. The aim is practical evidence literacy, not a safety ranking or an individual treatment recommendation.

At a glance

Reports are not rates: Adverse-event reports lack a denominator for calculating an event rate and do not automatically establish causation.
Sources: [1]
Measurement errors matter: FDA has received reports of errors measuring or administering compounded medication, some associated with hospitalization.
Sources: [1]
Warm injectable deliveries require action: FDA recommends not using an injectable GLP-1 drug that arrives warm or with insufficient refrigeration, and contacting the provider or dispensing pharmacy.
Sources: [1]
Compounding has a specific evidence boundary: Compounded preparations are not FDA approved. FDA recommends them when an individual's medical needs cannot be met by an approved drug, with a prescription and a state-licensed pharmacy.
Sources: [1]
Access claims are not response-time guarantees: Ro advertises unlimited provider messaging and side-effect management; its published 24/7 care-team access language does not specify a response deadline.
Sources: [24]

Start with the missing denominator

A rate needs both an event count and a defined population in which those events occurred. A collection of adverse-event reports does not provide that denominator. Consequently, a report count cannot tell a reader what percentage of users experienced the problem. It also cannot establish that one company has a higher event rate than another. FDA's warning about this limitation is not a reason to ignore reports; it is a boundary on the arithmetic they support.

The same restraint applies when a provider is named in a story or discussion. A general FDA concern is not a measured event rate for CoreAge Rx, Found or any other business covered here. The consulted documentation does not establish comparable provider-specific safety rates. Claims that a platform is safer because fewer reports are mentioned—or more dangerous because more reports are discussed—therefore go beyond what this evidence can establish.

Match the evidence to the question it can answer
Evidence type	Useful for	Does not establish
Adverse-event report	Identifying a concern for attention	An event rate or automatic causation
FDA public guidance	Understanding documented regulatory concerns	A safety ranking of these providers
Randomized clinical trial	Studying a defined product and population	Outcomes for a telehealth company's customers
Provider service page	Documenting advertised access and responsibilities	Verified clinical performance
Pharmacy or testing claim	Identifying questions about preparation and quality	A provider-specific adverse-event rate

Sources: [1]

Separate sequence from causation

An event occurring after medication use establishes a sequence, not necessarily its cause. FDA explicitly cautions that adverse-event reports do not automatically establish causation. This leaves room for investigation without either declaring the medicine responsible or declaring the event unrelated. For readers, the disciplined wording is that an event was reported—not that the report proves the drug caused it, or that everyone using a similarly named product faces a quantified risk.

This distinction also protects against the opposite mistake: treating uncertainty as reassurance. FDA has received reports of measurement or administration errors involving compounded medication, including some associated with hospitalization. That is a concrete concern even though the reports do not yield a population rate. The practical response is to clarify instructions with the prescriber or pharmacist, rather than use an article, advertisement or online discussion to calculate a dose.

Sources: [1]

Identify the exact preparation before borrowing evidence

The medication name alone may leave important differences unresolved. CoreAge Rx says its compounded tirzepatide is typically formulated with B-12, while discussing a version without B-12 requires the provider's involvement. It also markets separate semaglutide and tirzepatide microdosing offerings. Enhance.MD markets an injectable Elite formulation combining semaglutide and tirzepatide, alongside Core and Advanced programs. These are documented offerings, not evidence that trials of other preparations establish their benefits or adverse-event frequencies.

FDA notes that semaglutide sodium and acetate are different active ingredients from those in approved drugs; the evidence here does not justify attributing either salt to these providers. More broadly, compounded preparations do not undergo FDA premarket review for safety, effectiveness and quality. That absence alone proves neither clinical equivalence nor inequivalence. Our guide to formulation and route differences explains why a shared ingredient name is only the beginning of an evidence assessment.

Sources: [12] [13] [14] [8] [19] [27] [1]

Keep clinical trials in their own evidence category

Randomized trials and adverse-event reports answer different questions. STEP 1 studied a specified subcutaneous semaglutide regimen in a defined population; SURMOUNT-1 studied specified subcutaneous tirzepatide regimens. Neither was a trial of these telehealth companies or their compounded products. Their findings cannot simply be reassigned to microdosing programs, vitamin blends, combinations or different routes. Nor does comparing separate trials create a randomized head-to-head comparison between medications.

A provider may accurately cite a study while leaving the applicability of that study to its offering unresolved. That is why the question should be, “What exactly was studied, and how does it match this prescription?” rather than merely, “Is there a study?” Our explanation of why a clinical trial is not a company performance report develops this distinction. Provider outcomes require provider-specific evidence; a familiar trial citation does not supply it.

Sources: [3] [4]

Find the person responsible for clinical follow-up

Public service documents can help establish whom to contact, even when they cannot establish safety outcomes. CoreAge Rx assigns medical evaluation and prescribing to independent professional entities and describes its own role as technology and administrative coordination. Its help center directs medical questions to the portal's “Talk to the Doctor” function. Its “24/7 Provider Support” heading, however, is explained by text about customer support availability—not a documented promise of round-the-clock prescriber response.

Ro advertises regular check-ins, unlimited provider messaging and side-effect management. Found describes monthly follow-up for compounded tirzepatide and messaging between visits, while its 24/7 Aimee offering is an AI companion, not continuous human clinical availability. These differences are useful for planning contact routes, but none supplies a measured response time or an adverse-event rate. Before enrolling, ask who receives a clinical concern, how it is escalated and what response timing is actually promised.

Sources: [17] [26] [21] [24] [5] [15] [23]

Treat pharmacy and shipping questions as specific responsibilities

FDA's delivery advice is explicit: do not use an injectable GLP-1 drug that arrives warm or with insufficient refrigeration, and contact the provider or dispensing pharmacy. A shipping advertisement cannot override that advice. CoreAge Rx estimates arrival approximately two to three business days after pharmacy shipment, separately from earlier processing. Enhance.MD advertises medication delivery every four weeks. Neither a transit estimate nor a replenishment schedule, by itself, documents temperature control for a particular package.

Identifying the pharmacy is therefore a practical preparation step. CoreAge Rx says a partner is assigned by therapy and medication, but its FAQ and terms disagree about using a pharmacy outside the network. Enhance.MD names partners in its FAQ, while its terms give a partly different list; using a chosen pharmacy brings separate medication billing. These are unresolved documentary questions, not proof of a delivery failure. See our guide to reading cold-chain and delivery claims for the distinction between logistics promises and supporting evidence.

Sources: [1] [7] [12] [18] [8] [11] [20]

Read support and testing claims without turning them into outcomes

Found says its compounded semaglutide receives third-party potency, purity and sterility testing before shipment. That is an attributed quality-control claim; the consulted documentation does not identify the outside laboratory or provide certificates of analysis. Enhance.MD describes baseline laboratory testing and a private portal for reviewing results and communicating with the care team. Such documentation helps explain the advertised service process. It does not establish that adverse events have been eliminated or measured at a particular frequency.

Walgreens provides another useful distinction: its pharmacy-expert chat is advertised as available 24/7, but may involve a wait and is not for emergencies or a substitute for physician advice. Weight Management follow-up chat visits and additional video visits are separately advertised at $49 each; the pharmacy-chat page does not establish a zero fee. Sesame, meanwhile, advertises provider messaging and video care, but says the platform is not a pharmacy. A support label should not blur these different responsibilities.

Sources: [15] [6] [11] [22] [16] [25] [10]

Turn uncertainty into better questions

A useful consultation question connects the general concern to the actual prescription and service: Which preparation and route are proposed? Which pharmacy will dispense it? Who explains the administration instructions? Where should a question about tolerability go? What happens if refrigerated delivery is inadequate? These questions do not require treating every report as proven causation. They recognize FDA's documented concerns while asking the responsible professionals to explain how the relevant process works.

NIDDK recommends discussing benefits, side effects, current conditions, other medicines, family history and costs with a clinician, including supplements, allergies and pregnancy plans. Follow-up should address tolerance and response, not merely recurring payments. Our clinical follow-up guide offers a framework for that conversation. A provider's willingness to answer questions is useful service information, but it still should not be represented as a verified safety endorsement or a guarantee of an individual result.

Sources: [1] [2]

Questions & answers

Does FDA's compounded-drug warning establish a safety rate for CoreAge Rx?

No. FDA's general guidance is not a provider-specific event rate. CoreAge Rx documents compounded semaglutide and tirzepatide services, but the consulted material does not establish a denominator-based safety rate for its patients. Its preparation, dispensing pharmacy and clinical contact route are useful subjects for questions; a general report count cannot answer them.

Sources: [1] [26]

Does Ro's side-effect support prove fewer adverse events?

No. Ro advertises unlimited provider messaging, side-effect management and regular check-ins. Those are service claims, not comparative safety results. Its 24/7 care-team access language also does not specify a response deadline. Ask how clinical concerns are handled without assuming that messaging availability establishes either immediate care or a lower event rate.

Sources: [24] [1]

Can trial findings establish the safety of Enhance.MD's Elite combination?

Not from this evidence. Enhance.MD markets Elite as an injectable semaglutide-and-tirzepatide formulation. STEP 1 and SURMOUNT-1 were not studies of that offering or of Enhance.MD's patients. Their results cannot automatically establish outcomes for this combination. Request evidence that matches the proposed formulation rather than accepting separate ingredient studies as a complete answer.

Sources: [27] [3] [4]

Who handles medication questions when using Sesame?

Sesame describes clinical care through independently operated provider practices and says it is not a pharmacy. Its program advertises video care and provider messaging; its terms identify Truepill as a delivery partner. Clarify which clinician handles treatment concerns and which dispensing pharmacy handles preparation or delivery questions, rather than treating the platform as one undifferentiated clinical service.

Sources: [9] [10] [25]

Is Walgreens' pharmacy chat the same as a prescriber follow-up visit?

No. Walgreens advertises pharmacy-expert chat separately from its Weight Management consultations. The chat may require waiting and is not intended for emergencies. Weight Management follow-up chat visits and additional video visits are advertised at $49 each, excluding medication. Do not infer immediate prescriber access or free clinical visits from the pharmacy chat's 24/7 availability.

Sources: [22] [16]

Does Found's third-party testing claim replace FDA approval?

No. Found attributes potency, purity and sterility testing to an outside party, but the consulted documentation does not supply certificates or identify that laboratory. Compounded preparations remain non-FDA-approved. FDA recommends compounding when an individual's medical needs cannot be met by an approved drug. Neither the testing claim nor nonapproval alone establishes clinical equivalence or inequivalence.

Sources: [15] [1]

Our assessment

Adverse-event reports are reasons to pay attention, not ready-made probabilities. The balanced reading preserves both truths: FDA has documented meaningful concerns, and those reports do not automatically prove causation or establish provider-specific rates. Public provider pages can clarify advertised preparations, contact routes and pharmacy responsibilities, but they cannot substitute for clinical outcome evidence. Use each source for the question it can actually answer, and bring unresolved product and care-process questions to the treating clinician or dispensing pharmacist.

Sources & documentation

Public materials consulted October 19, 2026. Provider statements are attributed claims, not independently verified clinical outcomes. Terms can change; obtain a current written quote.

What GLP-1 Adverse-Event Reports Can—and Cannot—Establish

At a glance

01Start with the missing denominator

02Separate sequence from causation

03Identify the exact preparation before borrowing evidence

04Keep clinical trials in their own evidence category

05Find the person responsible for clinical follow-up

06Treat pharmacy and shipping questions as specific responsibilities

07Read support and testing claims without turning them into outcomes

08Turn uncertainty into better questions