Search results for who cannot take these drugs return a long, undifferentiated list, and most of what is on it is not a contraindication at all. The distinction is worth getting right, because flattening it misleads in both directions: it frightens people who are eligible, and it reassures people who genuinely need a conversation first.
The formal list is short
On both the semaglutide and tirzepatide labels, the contraindications section names two things. The first is a personal or family history of medullary thyroid carcinoma, or a diagnosis of multiple endocrine neoplasia syndrome type 2. The second is a prior serious hypersensitivity reaction to the drug or to any ingredient in it.
That is the whole list. Everything else circulating as a rule is drawn from the warnings and precautions section, where the instruction is to weigh, monitor or discuss rather than to refuse.
Two consequences follow. Most people reading a scare list online are eligible, and price is the variable that will actually decide where they buy — see the cost board and the oral options. The small number of people the first column applies to should treat it as settled rather than negotiable.
Why the thyroid rule exists, and what the human data show
Both products carry a boxed warning about thyroid C-cell tumors, which originates in rodent studies: in rats, these drugs produced dose-dependent and duration-dependent C-cell tumors at clinically relevant exposures. Whether that translates to humans has not been determined, and the labels state as much in the warning itself.
The largest observational test of it is a Scandinavian cohort covering Denmark, Norway and Sweden. Thyroid cancer occurred in 76 of 145,410 people treated with a GLP-1 receptor agonist, against 184 of 291,667 on a comparator drug — a hazard ratio of 0.93 (95% CI 0.66 to 1.31) over a mean 3.9 years. For medullary thyroid cancer specifically the hazard ratio was 1.19, with a confidence interval from 0.37 to 3.86.[1]
Read that carefully in both directions. The study found no substantially increased risk across the population, and its own authors note the upper confidence bound still allows for as much as a 31% relative increase. A wide interval on a rare cancer is not the same as an all-clear, which is why the family-history rule stays in place regardless.
The gut conditions that change the calculation
These drugs work partly by slowing gastric emptying, so an existing motility problem is a genuinely different situation. The semaglutide labeling states that the drug is not recommended in severe gastroparesis, and severe gastrointestinal adverse reactions have their own warnings section on both products.
A multicenter cohort across thirteen hospitals found semaglutide initiation for weight management associated with gastrointestinal dysmotility or obstruction at a hazard ratio of 3.91 (95% CI 1.42 to 10.82).[2] Wide interval, small numbers, but the direction is consistent with the mechanism rather than a surprise.
On the biliary side, a meta-analysis of 55 randomized trials covering 106,395 participants found an increased risk of gallstones, risk ratio 1.46 (95% CI 1.09 to 1.97), and of reflux, risk ratio 2.19 (95% CI 1.48 to 3.25). It found little or no effect on pancreatitis, bowel obstruction, gastrointestinal bleeding or gastroparesis in the trial record.[3]
That last sentence is the useful one, and it cuts against the internet consensus. Pancreatitis holds a warning on the label and is a reason to disclose a prior episode, but the randomized evidence has not shown these drugs causing it at a measurable rate. A prior attack still belongs on an intake form, because the label instructs stopping the drug if pancreatitis is suspected and prior history changes how a symptom is read.
The rest of the conversation
A 2026 review of safety across this drug class collects the remaining items: worsening of existing diabetic retinopathy, acute kidney injury when vomiting causes dehydration, and injection-site reactions. It also catalogs rarer reports drawn from pharmacovigilance sources that trials have not confirmed.[4] Pregnancy and planned conception belong on the same list — that timing has its own set of rules.
Three more are easy to miss. Stacking a second GLP-1 on top of the first is not recommended on the tirzepatide label, which matters when a person orders from two sellers at once. Anyone using insulin needs a hypoglycemia plan built with a clinician. And anesthesia teams now ask about these drugs before elective procedures, because delayed stomach emptying raises an aspiration concern under sedation.
What this means when you are buying online
Every item above depends on an intake that asks. A form that collects a weight and a card number but never asks about thyroid family history is skipping the only question the label treats as absolute. How each seller handles that is part of what a provider review is for. The criteria behind those assessments are written up in the methodology, and commercial relationships are listed in the disclosures.
One more caveat specific to this market. Most sellers here supply compounded product, which does not arrive with the FDA-reviewed labeling quoted above, so the warnings that shape this page are the branded products’ and not a document in the box. Tolerability once you have started is a different subject, covered in the side-effect data.
None of this is medical advice and none of it is a diagnosis. The purpose of a page like this is to tell you which facts about yourself are worth volunteering before a prescription is written, not to decide anything on your behalf.