Nausea is the side effect people expect from a GLP-1 medicine, and the labels confirm it: it is the most reported adverse reaction on every one of them. What the labels also allow, and what most pages about it skip, is a closer reading of when it happens, how long a single episode lasts, and how rarely it ends treatment.
This page reads the current US prescribing information for nine GLP-1 products, retrieved from FDA's label service and DailyMed on 26 September 2026, alongside the one published analysis that timed these episodes and the one head-to-head weight trial that counted them in both drugs at once. It is one symptom across the class. The class-wide side-effect picture is on the GLP-1 side effects hub, and how the side effects move with dose is on the dose and side effects page.
The rates, label by label
"Excess per 100" is our arithmetic: the drug rate at the highest dose shown minus the placebo rate in the same table.
| Label (population) | Placebo | Nausea on drug, by dose | Excess per 100 | Vomiting, top dose vs placebo |
|---|---|---|---|---|
| Wegovy 2.4 mg (weight, three trials, 2,116 on drug) | 16% | 44% | 28 | 24% vs 6% |
| Wegovy 2.4 vs 7.2 mg (weight, two 72-week trials) | 13% | 35% · 39% | 26 | 22% vs 6% |
| Saxenda 3 mg (weight, 3,384 on drug) | 13.8% | 39.3% | 25.5 | 15.7% vs 3.9% |
| Foundayo (weight, Trials 1 and 2) | 10% | 26% · 34% · 35% (5.5, 9, 17.2 mg) | 25 | 24% vs 4% |
| Zepbound (weight, Studies 1 and 2) | 8% | 25% · 29% · 28% (5, 10, 15 mg) | 20 | 13% vs 2% |
| Trulicity (type 2 diabetes) | 5.3% | 12.4% · 21.1% (0.75, 1.5 mg) | 15.8 | 12.7% vs 2.3% |
| Victoza (type 2 diabetes) | 5% | 18% · 20% (1.2, 1.8 mg) | 15 | 9% vs 2% |
| Ozempic injection (type 2 diabetes) | 6.1% | 15.8% · 20.3% (0.5, 1 mg) | 14.2 | 9.2% vs 2.3% |
| Semaglutide tablets (type 2 diabetes) | 6% | 11% · 20% (7, 14 mg) | 14 | 8% vs 3% |
| Mounjaro (type 2 diabetes) | 4% | 12% · 15% · 18% (5, 10, 15 mg) | 14 | 9% vs 2% |
Two patterns hold across the table. The weight-management labels sit higher than the diabetes labels — roughly 20 to 28 extra cases per 100 against 14 to 16 — which fits their higher doses but also reflects different people and trial designs. And where a label prints doses, nausea generally climbs with them, although Zepbound's three doses land within four points of each other. The Wegovy label adds that findings were similar in its trial of the Wegovy tablets; its adverse-reaction tables are for the injection.
Each label also carries the standard statement that rates from different trial programmes cannot be directly compared. The placebo column shows why: nausea on a dummy treatment ranged from 4% in the Mounjaro diabetes pool to 16% in the Wegovy weight trials.
When it happens
Four labels say it outright. Zepbound's: "The majority of nausea, vomiting, and/or diarrhea events occurred during dose escalation and decreased over time." Mounjaro's uses almost the same words. Foundayo's says the incidence "was higher during the FOUNDAYO dosage escalation period and decreased over time". The Rybelsus label says the majority of reports occurred during dose escalation.
The fullest timing data come from a pooled analysis of the STEP 1 to 3 trials of semaglutide 2.4 mg — 2,117 people on the drug and 1,262 on placebo, followed for 68 weeks. The share of people with nausea, diarrhoea or vomiting peaked at about week 20 and fell afterwards, with the steepest fall for nausea. Week 20 is not arbitrary: the Wegovy injection schedule steps up every four weeks and reaches its maintenance dose at week 17. The cumulative number of people having a first gastrointestinal event levelled off after the same point.
The escalation schedules themselves are on our titration schedules page, reported from each label.
How long a single episode lasts
This is the figure almost nobody quotes. In the same pooled analysis, the median duration of a single episode was 8 days for nausea, 3 days for diarrhoea and 2 days for vomiting — and similar to the durations on placebo. Constipation was the exception: a median of 47 days on semaglutide against 35 on placebo.
So the rates in the label table are rates of people who reported the symptom at least once over the trial, not of people who felt sick throughout it. The analysis states that rates of nausea, diarrhoea and vomiting stayed above placebo for the whole treatment period, but that participants typically recovered from each event within a few days.
The STEP 4 trial adds the view from the other side of escalation. Of 902 people who entered its 20-week run-in on semaglutide, 71.4% reported a gastrointestinal event. Among the 803 who were then randomised, new gastrointestinal events after week 20 occurred in 41.9% of those who continued the drug and 26.1% of those switched to placebo — and only two people stopped treatment for gastrointestinal reasons after week 20.
How often it ends or changes treatment
In the pooled STEP analysis, gastrointestinal events led to a dose reduction or a temporary pause in 12.5% of people on semaglutide against 1.7% on placebo, and 4.3% stopped permanently because of them.
The labels give the stopping figures for nausea itself where they name it. Saxenda's: nausea 2.9% against 0.2% on placebo, vomiting 1.7% against under 0.1%. Wegovy's figures, and the gap between how many report nausea and how many stop, are set out on the side effects hub, which we do not repeat here. Zepbound's label adds that most people who stopped because of adverse reactions did so "during the first few months of treatment" because of gastrointestinal reactions.
The one head-to-head count
Label tables for different drugs come from different trials. SURMOUNT-5 is the exception: 751 adults with obesity randomised to tirzepatide (up to 15 mg) or semaglutide (up to 2.4 mg) for 72 weeks. Its posted results on ClinicalTrials.gov count the gastrointestinal events in both arms:
| Event (SURMOUNT-5, posted results) | Tirzepatide (374) | Semaglutide (376) |
|---|---|---|
| Nausea | 163 (43.6%) | 167 (44.4%) |
| Vomiting | 56 (15.0%) | 80 (21.3%) |
| Diarrhoea | 88 (23.5%) | 88 (23.4%) |
| Constipation | 101 (27.0%) | 107 (28.5%) |
| Eructation | 37 (9.9%) | 29 (7.7%) |
The percentages are our arithmetic from the posted counts. Nausea was almost identical in the two arms, which is not what a reader comparing the Zepbound and Wegovy labels (28% against 44%) would expect. Vomiting was less common on tirzepatide and belching slightly more common. The trial was open-label and sponsored by Lilly, which makes tirzepatide. How the two drugs compare on other outcomes is on our tirzepatide vs semaglutide page, and how the class compares on side effects overall is on which GLP-1 has the fewest side effects.
Sulfur burps: the counted version
"Sulfur burps" is how people describe it; the labels call it eructation, and several count it:
| Label | Placebo | Eructation on drug |
|---|---|---|
| Wegovy 2.4 mg (weight) | under 1% | 7% |
| Zepbound (weight) | 1% | 4% · 5% · 5% |
| Foundayo (weight) | 1% | 6% · 8% · 8% |
| Mounjaro (diabetes) | 0.4% | 3.0% · 2.5% · 3.3% |
None of the labels describes the smell, and the medical term does not separate sulfur-smelling belches from ordinary ones.
Does nausea mean it is working?
The semaglutide analysis tested this. People with gastrointestinal side effects lost about the same share of their weight as people without them (11.4% to 17.7% against 9.6% to 17.1% across the three trials), and a mediation analysis attributed less than one percentage point of the extra weight loss over placebo to those side effects. Several of the authors were Novo Nordisk employees, and Novo Nordisk makes semaglutide; the conclusion is also consistent with the drug's intended action being on appetite rather than through feeling ill.
What this page cannot tell you
- Whether one person's nausea is the drug. Up to 16% of people on placebo reported nausea too.
- Which drug will suit one person. The label rates are not comparisons, and the one head-to-head trial found near-identical nausea.
- What to do about it. The labels warn that fluid lost through vomiting or diarrhoea can injure the kidneys, and the patient information on the Mounjaro and Trulicity labels tells patients to talk to their healthcare provider about any side effect that bothers them or does not go away. Dose changes belong to the prescriber.
Sources
- WEGOVY (semaglutide) prescribing information, Novo Nordisk, SPL effective 2026-06-18 — DailyMed, Tables 1, 3 and 4.
- ZEPBOUND (tirzepatide) prescribing information, Eli Lilly, effective 2026-08-28 — DailyMed, section 6.1 and Table 1.
- MOUNJARO (tirzepatide) prescribing information, Eli Lilly, effective 2026-08-27 — DailyMed, section 6.1.
- FOUNDAYO (orforglipron) prescribing information, Eli Lilly, effective 2026-07-29 — DailyMed, section 6.1 and Table 1.
- SAXENDA (liraglutide) prescribing information, Novo Nordisk, effective 2026-02-25 — DailyMed, Table 2.
- TRULICITY (dulaglutide) prescribing information, Eli Lilly, effective 2026-06-16 — DailyMed, Table 1.
- VICTOZA (liraglutide) prescribing information, Novo Nordisk, effective 2025-10-14 — DailyMed, section 6.1.
- OZEMPIC (semaglutide) injection prescribing information, Novo Nordisk, SPL effective 2026-06-01 — DailyMed, Table 1; OZEMPIC and RYBELSUS tablets, effective 2026-01-30 — DailyMed, Table 2.
- Wharton S, et al. Gastrointestinal tolerability of once-weekly semaglutide 2.4 mg in adults with overweight or obesity, and the relationship between gastrointestinal adverse events and weight loss. Diabetes Obes Metab 2022;24:94-105 — PubMed 34514682, full text PMC9293236.
- SURMOUNT-5 (tirzepatide vs semaglutide), posted results — ClinicalTrials.gov NCT05822830, adverse events module, read 2026-09-26; main report Aronne LJ, et al. N Engl J Med 2025 — PubMed 40353578.