Skip to main content

Page last checked

Symptom question · incretin class

Does a GLP-1 cause diarrhea?

Yes. 6 of the 36 compounds here list diarrhea as a reported side effect. The rate differs by drug and study. The table below gives each source’s figure.

Source: checked against 36 compound guides; rates are shown exactly as reportedFDA label

The label gives a frequency measured in a trial. FDA data shows the share of later reports that mentioned diarrhea. The two sources use different methods, but here they point in the same direction.

Neither of them tells you whether it will happen to you, and neither is a probability for any one person. What they establish is that this is an expected, documented reaction rather than a surprise. That is a smaller claim than the one most pages on this question make.

This question is usually typed with a brand in it — Ozempic, Wegovy, Mounjaro, Zepbound. The reaction belongs to the molecule under the brand: semaglutide under the first two, tirzepatide under the second two. Both are in the table below, and both name it.

Medically reviewed by Jennifer Montecillo, MD · non-practicing medical reviewer

Information checked 2026-08-20

At a glance

Compounds that list it

7 of 36

Only in a combined category

1

No published human safety data

11

Best available source

FDA label

Which compounds list diarrhea

We checked every covered compound for this one reaction. Three sources name it directly, and one includes it in a combined stomach-and-gut figure. The others either do not list it or have no human safety data. We keep those answers separate so a missing result is never mistaken for proof that a reaction cannot happen.

◆ Diarrhea across 36 compoundssource shown for every row
Every compound covered on this site, showing whether it lists diarrhea and at what frequency its own cited source states.
CompoundHow it appearsFrequency reportedSource type
TirzepatideZepbound, MounjaroDiarrheaVery common (≥10%)FDA label
SemaglutideWegovy, Ozempic, RybelsusDiarrheaVery common (≥10%)FDA label
LiraglutideVictoza, SaxendaDiarrheaReported in Victoza trials (≥5%)FDA label
RetatrutideDiarrheaReported in trials (dose-dependent)Human trial
CagrilintideStomach and gut events overall (nausea, constipation, diarrhea)counted inside a grouped figure, not on its ownReported in 41–63% across trial doses (vs 32% placebo)Human trial
SurvodutideNot listed in the source checkedThe label or study checked for this compound does not list this reaction.—Human trial
BPC-157No published human safety dataNo human trial has published enough safety data to list reactions for this compound.—no human dose shown to work
TesamorelinEgrifta SV, Egrifta WRNot listed in the source checkedThe label or study checked for this compound does not list this reaction.—FDA label
PT-141VyleesiNot listed in the source checkedThe label or study checked for this compound does not list this reaction.—FDA label
CJC-1295Not listed in the source checkedThe label or study checked for this compound does not list this reaction.—limited human data
IpamorelinNot listed in the source checkedThe label or study checked for this compound does not list this reaction.—Human trial
TB-500No published human safety dataNo human trial has published enough safety data to list reactions for this compound.—no human dose shown to work
GHK-CuNo published human safety dataNo human trial has published enough safety data to list reactions for this compound.—no human dose shown to work
SermorelinNot listed in the source checkedThe label or study checked for this compound does not list this reaction.—limited human data
MOTS-cNo published human safety dataNo human trial has published enough safety data to list reactions for this compound.—no human dose shown to work
AOD-9604Not listed in the source checkedThe label or study checked for this compound does not list this reaction.—limited human data
NAD+Not listed in the source checkedThe label or study checked for this compound does not list this reaction.—limited human data
EpitalonNo published human safety dataNo human trial has published enough safety data to list reactions for this compound.—no human dose shown to work
Kisspeptin-10No published human safety dataNo human trial has published enough safety data to list reactions for this compound.—no human dose shown to work
DSIPNot listed in the source checkedThe label or study checked for this compound does not list this reaction.—limited human data
HexarelinNot listed in the source checkedThe label or study checked for this compound does not list this reaction.—limited human data
IGF-1 LR3No published human safety dataNo human trial has published enough safety data to list reactions for this compound.—no human dose shown to work
PEG-MGFNo published human safety dataNo human trial has published enough safety data to list reactions for this compound.—no human dose shown to work
LL-37Not listed in the source checkedThe label or study checked for this compound does not list this reaction.—limited human data
KPVNo published human safety dataNo human trial has published enough safety data to list reactions for this compound.—no human dose shown to work
Thymosin alpha-1Not listed in the source checkedThe label or study checked for this compound does not list this reaction.—limited human data
SS-31ForzinityNot listed in the source checkedThe label or study checked for this compound does not list this reaction.—FDA label
SemaxNot listed in the source checkedThe label or study checked for this compound does not list this reaction.—limited human data
SelankNo published human safety dataNo human trial has published enough safety data to list reactions for this compound.—no human dose shown to work
Melanotan IINot listed in the source checkedThe label or study checked for this compound does not list this reaction.—limited human data
MazdutideDiarrheaReported by 39.4% on 9 mg weekly (vs 6.5% on placebo)Human trial
PinealonNot listed in the source checkedThe label or study checked for this compound does not list this reaction.—limited human data
AHK-CuNo published human safety dataNo human trial has published enough safety data to list reactions for this compound.—no human dose shown to work
VK2735Not listed in the source checkedThe label or study checked for this compound does not list this reaction.—Human trial
ZenagamtideNot listed in the source checkedThe label or study checked for this compound does not list this reaction.—Human trial
PemvidutideDiarrheaIn RECLAIM, the sponsor reported 20% with pemvidutide versus 10% with placeboHuman trial

A blank result means the label or study checked here does not list the reaction. It does not prove the reaction cannot happen, and it does not replace the full prescribing information. A pharmacist or clinician can review the complete label alongside the rest of a person’s medicines.

What people reported to the FDA

The first table summarizes labels and published trials. This table shows reports filed with the FDA after the medicines went on sale. Each percentage is the share of reports for that product that mention this symptom—not the share of patients who experienced it.

Each GLP-1 brand in the FDA snapshot, and what share of its reports names this symptom.
ProductShare of its own reportsAll reports
Ozempicsemaglutide9.0% of Ozempic reports66,161 reports
Rybelsussemaglutide7.9% of Rybelsus reports7,379 reports
Victozaliraglutide7.1% of Victoza reports44,086 reports
Trulicitydulaglutide6.9% of Trulicity reports88,332 reports
Mounjarotirzepatide6.3% of Mounjaro reports93,975 reports
Zepboundtirzepatide6.2% of Zepbound reports70,141 reports
Saxendaliraglutide5.5% of Saxenda reports10,040 reports
Wegovysemaglutide5.5% of Wegovy reports21,044 reports

A row reading “below” is not a zero. The snapshot keeps each product’s 25 most-reported terms, so a term missing from that list was either never reported or reported and ranked 26th. All that can be proved is that it sits under the 25th. Read from FDA on 2026-08-17.

Reports are not rates and a report is not proof the drug did it. How to read this data, and the questions it answers on its own.

3Read deeperWhy it happens, when it appears, and what the page proves

Why the gut is where this class shows up

These compounds act on receptors that are not confined to the pancreas. The same signaling that slows how fast the stomach empties also acts along the intestine. The gut is dense with the nerves and muscle that signaling reaches. So a change in bowel habit is close to the drug’s known action rather than off to one side of it.

That is also why the same class produces both constipation and diarrhea, which sounds contradictory and is not. Motility can move in either direction, and which way it moves differs between people and between doses. Both reactions appear on the same labels, at the same frequency band, from the same trials.

This paragraph is general pharmacology rather than a value read off a record. It is here because the alternative invites the reader to supply their own account. The ones in circulation are worse.

The opposite reaction is on the same records

Constipation, Stomach and gut events overall (nausea, constipation, diarrhea) and Stomach and gut events overall (nausea, vomiting, constipation) is also listed for Tirzepatide (very common (≥10%)), Semaglutide (very common (≥10%)), Liraglutide (reported in Saxenda trials (≥5%)), Retatrutide (reported in trials), Cagrilintide (reported in 41–63% across trial doses (vs 32% placebo)), Survodutide (reported in 75% of survodutide recipients (vs 42% placebo)), VK2735 (treatment-related event in 37 of 175 participants in the phase 2 safety population) and Pemvidutide (in RECLAIM, the sponsor reported 26% with pemvidutide versus 6% with placebo). Each result links to its original source.

Constipation sits on these labels at the same frequency band as diarrhea, from the same trials. A page that reports one and not the other implies a direction the sources do not state. This section exists so that neither reaction is read as the characteristic one.

What neither document says about how long it lasts

A label frequency band answers how many people in a trial reported something. It does not answer when it started, how long it went on, or whether it settled. A share of FDA side-effect reports does not answer those either — a report records that something was filed, with no follow-up attached.

Neither source reports how long it lasted. That means there is no reliable duration to quote from them. A precise timeline from elsewhere would be based on a different source or on an estimate.

What this page does and does not establish

Diarrhea is the reaction where the evidence is strongest in this section, which makes it the easiest place to overclaim. Each row separates what the two documents support from what gets built on top of them.

The claimWhat the evidence shows
Diarrhea is a documented reaction of this class.Established. Three sources name it and one includes it in a combined figure. FDA reports include it for all eight brands.
Roughly one in ten people taking it get diarrhea.Not what the label says. A frequency band is the rate observed in a specific trial population on a specific schedule. Reading it as a personal probability drops the population, the dose and the comparison group.
Ozempic causes more diarrhea than Wegovy.Not established by FDA reports. A larger share of reports does not mean a higher rate among everyone who used the product. The data does not include people who filed nothing, and report shares also change with market age and prescribing patterns.
It means the dose is too high.Not established here, and not a judgment this page makes. Nothing in either document ties a reaction to a dose decision, and a dose decision belongs to the prescribing clinician.

When this stops being a nuisance

Most of what is described above is uncomfortable rather than dangerous. Two things on these sources turn it into something to raise promptly, and both are on the labels.

Acute pancreatitis

Urgent

Signs: Severe pain high in the abdomen that will not let up and may bore through to the back, often with nausea or vomiting that does not settle. People describe it as different from ordinary post-dose nausea, in that it does not come and go with meals.

The prescribing information lists this as a reaction that needs medical assessment. Call a clinician without waiting for the next appointment, or emergency services if the pain is severe.

Acute kidney injury from dehydration

Prompt

Signs: Vomiting or diarrhea bad enough that you cannot keep fluids down, passing much less urine than usual, dizziness when you stand up, or unusual weakness.

The label links the risk to dehydration, so ongoing fluid loss is worth reporting early. A clinician can check kidney function.

Recognition is the only job of this section. Assessment belongs to a clinician, who can weigh a symptom against the rest of the picture in a way no page can.

EmergencySome symptoms should not wait for an appointment. Call emergency services (911 in the US) if they are severe, getting worse fast, or involve trouble breathing, chest pain, fainting or confusion.

Common questions

Is diarrhea listed on the Ozempic and Wegovy labels?

The semaglutide label used for both brand names lists it in its highest frequency range.

That record is a transcribed extract of the reactions its cited label reports most prominently, not a complete copy of the document. The full label, read by a pharmacist or clinician alongside the rest of a medicine list, is where a complete list belongs.

How long does it last?

No label or trial reports a set duration.

That is a real limit rather than an evasion. A frequency band counts people, not days, and an FDA side-effect report has no follow-up attached to it. A page giving you an exact week count is quoting something other than these two documents.

Why does the same drug cause both diarrhea and constipation?

Both are on the same labels, at the same frequency band, from the same trials.

Motility can shift in either direction, and these sources do not say what decides which. What they do establish is that neither reaction is the characteristic one, and a page reporting only whichever fits its argument is choosing.

Does it mean the medicine is working?

Nothing in these sources connects a reaction to whether the treatment is achieving anything.

No study shows that a side effect means the drug is working. Treating discomfort as a progress signal can encourage people to ignore symptoms they should report.

Why does the report-stream table have no dashed rows?

Because every one of the eight brands carries this term among its twenty-five most-reported, so every row is a measurement rather than a ceiling.

Most terms in this section are not like that. Where a term is missing from a product’s stored list, the honest statement is a bound. It sits at or below the twenty-fifth term’s count. That is what a dashed row means. This table has none, which is itself worth noticing.

Want the full picture for one compound? Open its guide for uses, dosage information, side effects and serious warnings.

Every tirzepatide reaction, not only diarrhea

9 common and 6 serious tirzepatide reactions, in full past diarrhea.

The full semaglutide profile behind its diarrhea entry

Frequencies as the semaglutide source states them, read directly rather than via diarrhea.

Which pages list constipation, beyond the diarrhea question

Named constipation entries, grouped figures and absences kept distinct, apart from diarrhea.

The nausea question, alongside diarrhea

One reaction, nausea, read down every page instead of out of diarrhea alone.

The pages each diarrhea row is read from

See the strongest source available for every compound, with diarrhea as context

How long diarrhea lasts, and why nobody says

No source cited on this site states how long any of these reactions lasts, with diarrhea as one example

What else mazdutide lists besides diarrhea

No boxed warning applies to mazdutide, on the mazdutide page rather than on diarrhea.

The full pemvidutide profile behind its diarrhea entry

Frequencies as the pemvidutide source states them, read directly rather than via diarrhea.

Every cagrilintide reaction, not only diarrhea

4 common and 0 serious cagrilintide reactions, in full past diarrhea.

Every symptom in this section is listed on the symptom questions index.

38primary sources, each with the day we read it — open the documents used on this page