GLP-1 side effects: which ones mean you should call your prescriber

TL;DR

• Seek urgent assessment - not an appointment - for sudden vision loss or rapidly worsening eyesight, severe abdominal pain radiating to the back, severe pain with fever, jaundice, vomiting that will not stop, facial or throat swelling, or distension with inability to pass stool or wind.
• Call your prescriber today for persistent abdominal pain, pain under the right ribs, persistent vomiting or diarrhoea, or dehydration signs such as dark urine, passing very little urine or dizziness on standing - and nothing here is a reason to change how you take a prescribed medicine, which is the prescriber's decision.
• "Common" is a frequency, not a reassurance - fluid loss through vomiting and diarrhoea is the mechanism by which acute kidney injury has been reported, which is why regulators strengthened renal warnings across the class.
• The FDA added ileus to the Ozempic label in September 2023 on postmarketing reports; the tirzepatide and Wegovy labels already carried it.
• Pancreatitis sits in Warnings and Precautions, yet pooled data does not show a clear increase - a 43-trial meta-analysis and a pooling of seven cardiovascular outcome trials (~56,000 participants) both found no significant difference. Severe persistent pain radiating to the back is still urgent every time.
• The EMA concluded in June 2025 that NAION is a very rare side effect of semaglutide - specific to semaglutide medicines, not established as a class effect.

Educational content, not medical advice. This article contains no doses, schedules or protocols.

---

The short answer - the symptoms that warrant a call today

Most people reading this are somewhere in the first few months of GLP-1 therapy, have a symptom, and want to know whether it is the ordinary cost of the drug working or something that has gone wrong. That is a reasonable question. It is also, in almost every specific case, a question that cannot be answered without an examination - and often without bloods or imaging.

So this article does something narrower and more useful than triage. It describes what the regulatory labelling actually flags, what the pooled trial evidence supports, and what the pattern of a symptom tends to look like when clinicians take it seriously. The purpose is to lower the threshold at which you pick up the phone, not to give you a way to decide you do not need to.

That third row is deliberate. There is no category on this page for wait and see, because this article cannot tell which symptoms belong in it for you.

Why "common" and "harmless" are not the same word

The most frequent mistake people make with GLP-1 side effects is treating common as a reassurance. It is not one. It is a frequency, and frequency says nothing about consequence.

Nausea is common on these drugs. So is vomiting, and so is diarrhoea. That is well established across the randomised trial programmes for semaglutide and tirzepatide, and gastrointestinal adverse events are the single largest category of reported effects in those trials. But the labelling for tirzepatide products explicitly describes gastrointestinal reactions as "sometimes severe," and regulators have moved to strengthen renal warnings across the GLP-1 class precisely because the common symptom - fluid loss through vomiting and diarrhoea - is the mechanism by which the serious outcome, acute kidney injury, has been reported to happen. Common and serious are not opposite ends of a scale. They are the same event at different intensities.

This is the point where a page like this can do real harm if it is written carelessly. Telling a reader that nausea is normal invites them to file a symptom under normal and stop thinking about it. So, plainly: the absence of a red flag on this page is not permission. A symptom that is not listed here still warrants a call. A symptom that is listed here but feels mild to you still warrants a call. Severity as experienced by a patient and severity as assessed by a clinician are different measurements, and the second one is the one that determines what happens next.

There is a companion article to this one - the GLP-1 side-effect timeline - which answers a genuinely different question. That page answers when will this happen: what tends to appear, in what order, and how the arc typically runs. This page answers when does this stop being normal. They are not substitutes. Knowing that nausea often clusters early does not tell you that your nausea today is that nausea.

Gastrointestinal symptoms: the normal arc, and the deviations from it that matter

The mechanistic account of GLP-1 gastrointestinal effects is uncontroversial: these drugs slow gastric emptying and act on central appetite and nausea pathways. Slowed emptying is not a side effect in the incidental sense - it is part of how the drug produces the effect it is prescribed for. Nausea, fullness, reflux, constipation and diarrhoea all follow plausibly from that.

What the mechanism does not tell you is where the boundary sits between an expected consequence of slowed transit and a clinical problem. That boundary has been the subject of active regulatory attention. In September 2023 the FDA added ileus - a failure of normal bowel motility that can amount to an intestinal blockage - to the Ozempic label, on the basis of postmarketing reports through its adverse-event reporting system; the tirzepatide and Wegovy labels already carried it. It is worth being precise about the evidence tier here: postmarketing reporting establishes that events were reported, not that the drug caused them. Spontaneous reports have no denominator and are subject to reporting bias, which is exactly why they generate labelling language rather than incidence figures. Pharmacovigilance analyses published since have found increased reporting odds for bowel obstruction and gastroparesis relative to comparator drugs, which is a signal worth acting on and is not the same as an established rate.

The features that tend to change a clinician's reading of a gastrointestinal symptom are about pattern rather than presence.

Read that table for the fourth column, not the second. The left-hand descriptions exist to explain what clinicians mean by "the expected arc" - they are not a checklist for self-clearance. Symptoms in the middle column are not diagnoses; they are the features that make an assessment more urgent. Nothing in the second column entitles anyone to conclude they are fine, and no combination of columns replaces someone examining you.

Upper-right abdominal pain and the gallbladder question

Pain under the right ribs deserves its own paragraph because it is common enough to be worth naming and specific enough to be worth acting on.

The association between GLP-1 receptor agonist use and gallbladder or biliary disease is not speculative. It has been demonstrated in pooled randomised trial data - the strongest evidence tier available for a drug-safety question - and the tirzepatide labelling reports acute gallbladder disease as an observed event in placebo-controlled trials. Two mechanisms appear to contribute: a direct effect on gallbladder motility, and the rapid weight loss that the therapy is prescribed to produce, which raises gallstone risk regardless of how the weight is lost. The full treatment is in GLP-1s and gallbladder risk, which also explains why a relative-risk figure cannot be turned into a personal one.

What matters for someone with the symptom now is the pattern clinicians look for: pain in the right upper abdomen or upper-middle abdomen, often after a fatty meal, sometimes radiating to the right shoulder blade, persistent rather than fleeting. Add fever, or jaundice, and it becomes urgent rather than same-day.

The complication is that early GLP-1 nausea and biliary pain can overlap in how they feel, and right-upper-quadrant pain has a differential diagnosis that includes several things unrelated to the drug entirely. Separating them requires examination and usually imaging. This is the clearest example on the page of a judgement that is not available to a reader, however carefully they read.

Signs of dehydration and why they compound quietly

Dehydration is the most under-reported item in this whole category, for a structural reason: it does not arrive as an event. It accumulates. Someone vomiting intermittently, eating less because appetite has fallen, and drinking less because drinking makes them feel full can lose more fluid than they replace for several days without any single moment that feels like an emergency.

The regulatory position here is direct. Health regulators have required strengthened warnings across GLP-1 labels about serious kidney injury linked to the drugs' gastrointestinal effects, and the proposed mechanism named in that literature is volume depletion from vomiting, diarrhoea or reduced fluid intake rather than any direct toxic effect on the kidney. That is an important distinction: it means the kidney risk described is largely downstream of a symptom you can notice, which is the reason to notice it.

Signs that warrant contacting your prescriber today include dark urine, passing much less urine than usual, dizziness or light-headedness on standing, unusual weakness or fatigue, a dry mouth that does not resolve, and confusion. Anyone with pre-existing kidney disease, or taking medicines that affect kidney function or blood pressure, is having a different conversation with their clinician about all of this than someone without - see what GLP-1s should not be mixed with.

One caution about the framing you will encounter elsewhere: some patient-facing summaries pair dehydration guidance with instructions about the medicine itself. This article does not, and will not. Whether a prescribed medicine continues, and on what terms, is a decision for the prescriber who has your history, your bloods and your reason for being on it. Your job is to report the symptom promptly. Their job is everything after that.

Pancreatitis: what the labelling says, what the evidence says, what it feels like

Pancreatitis is the item on this list where the labelling and the pooled evidence point in usefully different directions, and it is worth being honest about that rather than picking whichever is more alarming.

What the labelling says. The FDA-approved labelling for semaglutide and tirzepatide products carries acute pancreatitis in Warnings and Precautions. The wording is specific: patients should be observed for signs and symptoms of acute pancreatitis, described as persistent severe abdominal pain, sometimes radiating to the back, which may or may not be accompanied by nausea or vomiting; and if pancreatitis is suspected, the drug should be discontinued and not restarted if confirmed. Note carefully who that last instruction is addressed to. It is written for the prescriber, as an instruction about clinical management. It is not an instruction to a patient reading a label at home, and it should not be read as one.

What the evidence says. Pooled randomised trial data have generally not supported a clear class-wide increase in acute pancreatitis. An analysis pooling seven cardiovascular outcome trials in type 2 diabetes (approximately 56,000 participants) reported no significant difference versus placebo for acute pancreatitis, and a 2023 meta-analysis across 43 trials likewise found no clear association, with confidence intervals crossing one in both cases. Living systematic reviews continue to accumulate placebo-controlled trial data on this question. This is meaningfully reassuring at the level of populations - and it is not a statement about you. Trial populations are selected, trials of this size are underpowered for genuinely rare events, and an average effect across tens of thousands of people has no bearing on what is causing pain in one person tonight.

What it feels like. Persistent, severe abdominal pain, often in the upper abdomen, frequently boring through to the back, often worse lying flat, sometimes with vomiting. That description comes directly from the labelling and it is the reason the labelling exists. It cannot be distinguished from several other causes of severe upper abdominal pain without examination and blood tests, and no reader should attempt to.

The practical synthesis: the population-level evidence does not suggest pancreatitis is a likely outcome of GLP-1 therapy, and severe persistent abdominal pain radiating to the back is nonetheless an urgent-assessment symptom every time it occurs. Both of those statements are true simultaneously, and holding both is the correct posture.

Vision changes, severe persistent vomiting, and the small set of genuine emergencies

Vision. In June 2025 the EMA's Pharmacovigilance Risk Assessment Committee concluded its review of semaglutide medicines and determined that non-arteritic anterior ischaemic optic neuropathy (NAION) - a form of sudden vision loss caused by interrupted blood supply to the optic nerve - is a very rare side effect of semaglutide, and recommended the product information for Ozempic, Rybelsus and Wegovy be updated accordingly. "Very rare" is a defined regulatory frequency category, not a description of any individual's risk. The underlying signal came from large epidemiological studies in adults with type 2 diabetes that suggested roughly a two-fold increase in NAION among people exposed to semaglutide compared with unexposed people; those are observational studies, which is a weaker tier than randomised evidence and cannot fully exclude confounding by the conditions that themselves raise NAION risk. The WHO issued a corresponding alert in June 2025, and the MHRA issued a Drug Safety Update on semaglutide and NAION in February 2026. The regulatory guidance is unambiguous on what a patient should do: sudden vision loss or rapidly worsening eyesight requires immediate medical attention. Note that this finding is specific to semaglutide; it has not been established as a class effect, and that is a real limitation rather than a reassurance about other agents.

Severe persistent vomiting. Treated as urgent for two independent reasons - it drives the volume depletion described above, and it accompanies several of the serious presentations on this page, including biliary obstruction, pancreatitis and ileus. Vomiting that will not stop is not a symptom to wait out.

The rest of the genuine emergency set. Severe abdominal pain with fever. Jaundice. Abdominal distension with inability to pass stool or wind. Signs of a severe allergic reaction - swelling of the face, lips or throat, or difficulty breathing. Symptoms of low blood sugar in anyone also taking insulin or a sulfonylurea, a combination the tirzepatide labelling specifically flags as raising hypoglycaemia risk. Each of these resolves to urgent assessment, not to a phone call you make tomorrow.

Investigational compounds. If you are taking an investigational compound such as retatrutide, which is investigational and not approved by any major regulator for any indication, everything above applies with less certainty rather than more. There is no approved label to define the warning set, and the safety characterisation rests on trial data with the limitations described in retatrutide side effects. Investigational status is a reason to lower your threshold for calling, not to raise it, and it is a strong argument for having a clinician who knows what you are taking - see whether you need a doctor to start.

What to bring to the call - a short, non-diagnostic checklist

This is not a triage tool and it produces no conclusion. It exists because clinical assessment is faster and better when the history is clear, and because people under stress forget the details that turn out to matter.

Before you call, note down:

• When it started. Date if you can, and whether it began suddenly or built up.
• What it feels like. Location, and whether it stays in one place or moves. Constant or in waves. Sharp, dull, cramping, burning. If it radiates, where to.
• What changes it. Food, particularly fatty food. Lying flat. Movement. Vomiting. Time of day.
• What else changed. Any new medicine or supplement, including anything over the counter. Alcohol. A significant change in what or how much you have been eating or drinking. Any illness in the past fortnight.
• What you have measured. Temperature if you have taken it. Urine colour and roughly how often you are passing urine. Whether you have kept fluids down in the last 12 hours.
• The medicine itself. What you are taking and how long you have been on it. Bring the pen, box or vial to the appointment if you have one.

Deliberately absent from that list: any instruction to change, delay or withhold anything. Do not make that decision on the basis of a webpage, including this one. Bring the information; let the clinician make the call.

Frequently asked questions

How do I tell the difference between normal side effects and something serious? You often cannot, and this is the single most important thing on the page. The symptoms overlap substantially - early GLP-1 nausea, biliary pain, pancreatitis and simple dehydration can all present as some combination of upper abdominal discomfort and vomiting, and the labelling itself describes the serious presentations in terms that sound like the common ones. What separates them is examination, blood tests and sometimes imaging, none of which you have at home. The practical implication is not that you should worry more; it is that the threshold for a call should be low, because the cost of an unnecessary phone call is small and the cost of a missed presentation is not. Report the symptom and let someone with the tools do the distinguishing.

Should I stop taking it until I can speak to someone? That is not a question this article will answer, and you should be sceptical of any consumer page that does. Decisions about a prescribed medicine belong to the clinician who prescribed it, who knows why you are on it, what else you take, what your bloods look like and what stopping would cost you. Labelling language about discontinuation is written as guidance to prescribers, not as instructions to patients. What you can do without any ambiguity is contact that clinician promptly, or seek urgent assessment if the symptom is on the urgent list and you cannot reach them.

My symptom is not on this list. Does that mean it is fine? No. This page covers the presentations that regulatory labelling and pooled evidence have specifically flagged, which is a narrow slice of what can happen to a person taking a medicine. It is not a complete list of GLP-1 adverse effects, it does not cover the effects of anything else you might be taking, and it says nothing about the many conditions that arise independently of the drug and would need attention regardless. A new or unexpected symptom is a reason to contact your prescriber whether or not it appears here. The absence of your symptom from this page carries no information at all.

Does this differ between semaglutide, tirzepatide and the newer compounds? Partly. The gastrointestinal profile and the gallbladder signal are broadly class-level findings. The NAION conclusion, by contrast, is specific to semaglutide medicines and has not been established across the class - though that reflects where the evidence has been gathered rather than a demonstration that other agents are exempt. The tirzepatide labelling adds specific language about hypoglycaemia risk when combined with insulin or a sulfonylurea. For investigational compounds such as retatrutide, which remains investigational and unapproved, no approved warning set exists at all, and the safety picture rests on trial data with the usual gap between trial conditions and real-world use. In every case the symptom-level guidance on this page is the same, because the action is the same: contact the person supervising your therapy.

Related reading

• The GLP-1 side-effect timeline - the companion to this page: when effects tend to appear and how the arc typically runs, where this page covers when that arc stops being the explanation
• GLP-1s and gallbladder risk - the pooled trial evidence behind the right-upper-quadrant pain section, and why a relative risk cannot be made personal
• Vision changes on a GLP-1 - which eyesight changes settle and which are urgent, and why the two cannot be told apart without an eye examination
• Whether you need a doctor to start - what proper intake and follow-up actually cover, and why supervision is the mechanism that makes any of this useful
• Retatrutide side effects - how tolerability data read for an investigational compound with no approved label
• What GLP-1s should not be mixed with - relevant to the "what else changed" question and to anyone on medicines affecting kidney function or blood glucose

Part of The Complete Guide to GLP-1s.

Educational content, not medical advice. This article contains no doses, schedules or protocols, and nothing in it is a recommendation to start, stop, alter, pause or continue any medication. It is not a triage tool and cannot diagnose anything. If you have a symptom while taking a GLP-1 receptor agonist, contact the clinician who prescribed it; if the symptom is severe, sudden or accompanied by fever, jaundice, breathing difficulty or vision loss, seek urgent medical assessment immediately rather than reading further. Claims for the Malaysian market are subject to MAB requirements.

Explore it with verified supply. Every Vivre Labs lot is independently tested and the certificate of analysis is published, so you know exactly what is in the vial. Explore tirzepatide.

References

• FDA-approved prescribing information for semaglutide products (Ozempic, Wegovy). Warnings and Precautions: acute pancreatitis - observe for persistent severe abdominal pain, sometimes radiating to the back, with or without vomiting; discontinue if pancreatitis is suspected and do not restart if confirmed.
• FDA-approved prescribing information for tirzepatide products (Mounjaro, Zepbound). Warnings and Precautions: acute pancreatitis; acute gallbladder disease reported in placebo-controlled adult trials; gastrointestinal adverse reactions described as sometimes severe, with use not recommended in severe gastrointestinal disease including severe gastroparesis; increased hypoglycaemia risk when used with insulin or an insulin secretagogue.
• US FDA labelling action, September 2023: addition of ileus to the Ozempic label based on postmarketing reports received through the FDA Adverse Event Reporting System; ileus already present on the tirzepatide and Wegovy labels.
• European Medicines Agency, Pharmacovigilance Risk Assessment Committee, June 2025: NAION concluded to be a very rare side effect of semaglutide medicines (Ozempic, Rybelsus, Wegovy), with product information recommended for update; signal based on epidemiological studies in adults with type 2 diabetes suggesting approximately a two-fold increase in NAION with semaglutide exposure.
• World Health Organization, June 2025: alert to healthcare professionals regarding NAION risk associated with semaglutide medicines.
• MHRA Drug Safety Update, 5 February 2026: semaglutide and NAION.
• Regulatory labelling actions requiring strengthened warnings regarding serious kidney injury across GLP-1 receptor agonist products, with volume depletion from vomiting, diarrhoea or reduced fluid intake identified as the principal proposed mechanism.
• Pooled analysis of seven cardiovascular outcome trials of GLP-1 receptor agonists in type 2 diabetes (approx. 56,000 participants) reporting no significant difference versus placebo for acute pancreatitis (OR ≈1.05; 95% CI 0.78–1.40).
• 2023 systematic review and meta-analysis of 43 randomised trials reporting no clear association between GLP-1 receptor agonist use and acute pancreatitis (OR ≈1.24; 95% CI 0.94–1.64).
• Living systematic review and meta-analysis of placebo-controlled randomised trials of GLP-1 receptor agonists and acute pancreatitis (ongoing).
• Pharmacovigilance analyses of the FDA Adverse Event Reporting System reporting increased reporting odds for bowel obstruction and gastroparesis with GLP-1 receptor agonists relative to comparator drugs.
• Faillie J-L, et al. "Association of Glucagon-Like Peptide-1 Receptor Agonist Use With Risk of Gallbladder and Biliary Diseases: A Systematic Review and Meta-analysis of Randomized Clinical Trials." JAMA Internal Medicine, 2022.