Every current US label for the GLP-1 weight and diabetes drugs records reflux and stomach pain more often on the drug than on placebo. In the Wegovy weight trials, gastroesophageal reflux disease (GERD) was reported by 5% of people on the drug against 3% on placebo, and abdominal pain by 20% against 10%. On Zepbound the reflux figures were 4% to 5% against 2%. Two large cohort studies have since found a raised risk of a reflux diagnosis in people starting these drugs. What no study has done is measure reflux directly in people taking semaglutide or tirzepatide.
This page sets out the label rates, what the observational studies found, and the one randomised trial that measured reflux with a pH probe. The labels were read on 17 October 2026 from DailyMed, using the manufacturers' current versions. It reports what the documents say; it does not advise on treatment.
What the labels call it
The labels' adverse-reaction tables use four terms that between them cover what patients call acid reflux, heartburn and stomach pain:
| Label term | What it covers |
|---|---|
| Gastroesophageal reflux disease (GERD) | Stomach contents or acid coming back up into the oesophagus |
| Dyspepsia | Indigestion: upper-stomach discomfort, burning or fullness after eating |
| Abdominal pain | On Wegovy, a pooled term including upper and lower abdominal pain, gastrointestinal pain, tenderness, discomfort and epigastric discomfort |
| Abdominal distension | Bloating |
The patient Medication Guides for Wegovy, Zepbound and Foundayo use the everyday word: each lists heartburn among the common side effects. Belching and "sulfur burps", which the labels record as eructation, have their own page: sulfur burps on GLP-1 drugs.
The rates, label by label
| Label (trials) | Placebo | Drug arms | GERD | Dyspepsia | Abdominal pain |
|---|---|---|---|---|---|
| Wegovy (adult weight trials, n=2,116 on drug) | 3 / 3 / 10 | 2.4 mg | 5 | 9 | 20 |
| Wegovy (adolescents, n=133 on drug) | 2 / — / 6 | 2.4 mg | 4 | — | 15 |
| Zepbound (weight trials) | 2 / 4 / 5 | 5 / 10 / 15 mg | 4 / 4 / 5 | 9 / 9 / 10 | 9 / 9 / 10 |
| Foundayo (weight trials) | 2 / 4 / 7 | 5.5 / 9 / 17.2 mg | 6 / 6 / 7 | 12 / 16 / 13 | 13 / 14 / 14 |
| Saxenda (adult weight trials, n=3,384 on drug) | 1.7 / 2.7 / 3.1 | 3.0 mg | 4.7 | 9.6 | 5.4 (upper: 5.1) |
| Mounjaro (type 2 diabetes trials) | 0.4 / 3 / 4 | 5 / 10 / 15 mg | 1.7 / 2.5 / 1.7 | 8 / 8 / 5 | 6 / 5 / 5 |
| Ozempic (type 2 diabetes trials) | 0 / 1.9 / 4.6 | 0.5 / 1 mg | 1.9 / 1.5 | 3.5 / 2.7 | 7.3 / 5.7 |
| Trulicity (type 2 diabetes trials) | 0.5 / 2.3 / 4.9 | 0.75 / 1.5 mg | 1.7 / 2.0 | 4.1 / 5.8 | 6.5 / 9.4 |
All figures are percentages of patients. The placebo column gives the placebo rate for GERD / dyspepsia / abdominal pain in that order. Sources: each manufacturer's prescribing information on DailyMed, adverse-reactions section 6.1 (Wegovy effective 2026-06-18; Zepbound 2026-08-28; Foundayo 2026-07-29; Saxenda 2026-02-25; Mounjaro 2026-08-27; Ozempic 2026-06-01; Trulicity 2026-06-16). For Mounjaro, Ozempic and Trulicity, the GERD figures come from the label's paragraph of less common gastrointestinal reactions rather than its main table.
What the table does and does not show. On every label, reflux was recorded more often on the drug than on placebo. The gap is a few percentage points: 2 on Wegovy, 2 to 3 on Zepbound, 3 on Saxenda, 4 to 5 on Foundayo, and 1 to 2 on the three diabetes labels (our subtraction from the rates above). The gaps sit in different trials, with different populations, durations and ways of collecting side effects, so they do not rank the drugs — no trial in any of these labels compares two of them for reflux. The diabetes labels' lower figures also come from lower doses than the weight-management labels.
Stomach pain is the larger signal: on Wegovy, abdominal pain was twice as common on the drug as on placebo in the adult weight trials, and the label's 7.2 mg trial recorded it in 12% on 7.2 mg, 9% on 2.4 mg and 7% on placebo. The Wegovy label says the gastrointestinal reactions in this group, reflux and indigestion among them, "were most frequently reported during dosage escalation". None of the labels gives a separate timeline for reflux.
What the cohort studies found
Two large studies have compared reflux diagnoses in people starting a GLP-1 drug with people starting another diabetes medicine. Both used health records, not trials.
| Study | Who | Comparison | Finding |
|---|---|---|---|
| Noh et al., Annals of Internal Medicine 2025 | 24,708 new GLP-1 users and 89,096 new SGLT-2 inhibitor users with type 2 diabetes, UK records, 2013–2021 | GLP-1 drugs against SGLT-2 inhibitors over a median 3 years | GERD risk ratio 1.27 (95% CI 1.14 to 1.42), absolute difference 0.7 per 100 people; complications risk ratio 1.55, 0.8 per 1,000 |
| Liu et al., Gut 2024 | 177,666 matched pairs with type 2 diabetes, TriNetX database | GLP-1 drugs against other second-line diabetes drugs, split by how long the drug acts | Erosive reflux disease hazard ratio 1.15 overall — but 1.22 for short-acting drugs and 0.99 for long-acting ones |
The second study's split is the detail most summaries miss. Its raised risk came entirely from the short-acting drugs — liraglutide, lixisenatide and exenatide, defined there as acting for a day or less. The long-acting drugs, the group that includes the weekly injections, showed no increase in erosive reflux. The first study did not report results by individual drug in its abstract. Both studies say what an observational design can say: an association after statistical adjustment. Noh and colleagues name the limitation themselves — no information on diet or lifestyle — and the comparison groups are diabetes patients on other medicines, not people taking these drugs for weight loss.
The one study that measured reflux directly
A diagnosis code records that a doctor wrote "GERD". A pH probe in the oesophagus records the reflux itself. A PubMed search on 17 October 2026 returned 36 records pairing GLP-1 drug terms with "gastroesophageal reflux" or "GERD" in the title or abstract; reading every title, one is a randomised study that used a pH probe. (Two 2026 papers from oesophageal-manometry labs reach opposite conclusions on whether these drugs change how the oesophagus contracts — Gala et al. reporting an association with motility abnormalities, and a second paper titled "Esophageal Motility Patterns Are Not Impacted" — but manometry measures muscle movement, not reflux.)
Quast et al., Diabetes Care 2020 randomised 57 people with type 2 diabetes to lixisenatide or liraglutide for 10 weeks and measured 24-hour oesophageal pH, sphincter pressure, gastric emptying and stomach acid before and after:
- Gastric emptying slowed on both drugs — by 52 minutes on lixisenatide and 25 minutes on liraglutide (half-emptying time).
- The number of reflux episodes and the overall reflux score did not change significantly on either drug.
- Stomach acidity fell by about 21%.
So the one direct measurement found slower emptying without more reflux, and slightly less acid. It is a small, short study with no placebo arm, run in people with diabetes, and it tested neither semaglutide nor tirzepatide. It cannot rule out reflux on the newer drugs; it does show that the usual explanation — slow emptying, therefore reflux — has not been confirmed where it was tested.
What is not established
- Reflux measured on semaglutide or tirzepatide. No pH-monitoring or endoscopy trial of either was found in the search above; their reflux figures come from trial reporting of side effects.
- How long reflux lasts. No label gives a duration for reflux or indigestion separately from the general statement about dose escalation. The general evidence on duration is on how long GLP-1 side effects last.
- Whether one drug causes less reflux than another. No head-to-head trial reports it. The cohort split between short- and long-acting drugs is one observational study's finding.
- Whether reflux treatment changes anything. No trial in these labels tested acid-suppressing medicines alongside a GLP-1 drug.
The labels also list severe stomach pain as a symptom of two serious warnings, pancreatitis and gallbladder disease, and the labels warn that slower emptying matters for surgery under anaesthesia, where it can raise the risk of inhaling stomach contents. Whether ongoing heartburn or stomach pain on one of these drugs needs a change of plan is a question for the prescriber. The wider picture of gastrointestinal effects is on the GLP-1 side effects hub.
Sources
- Current US labels, opened on DailyMed on 17 October 2026 in the manufacturers' own versions; the table or section named is where each label prints its reflux, GERD and dyspepsia rows. Novo Nordisk: Wegovy (set ID ee06186f-2aa3-4990-a760-757579d8f77b), Saxenda (3946d389-0926-4f77-a708-0acb8153b143, Table 2), Ozempic (adec4fd2-6858-4c99-91d4-531f5f2a2d79, section 6.1). Eli Lilly: Zepbound (487cd7e7-434c-4925-99fa-aa80b1cc776b, Table 1), Foundayo (8ac446c5-feba-474f-a103-23facb9b5c62, Table 1), Mounjaro (d2d7da5d-ad07-4228-955f-cf7e355c8cc0, section 6.1), Trulicity (463050bd-2b1c-40f5-b3c3-0a04bb433309, Table 1 and section 6.1). Search any of them at dailymed.nlm.nih.gov.
- Noh Y, Yin H, Yu OHY, Bitton A, Azoulay L. "Glucagon-Like Peptide-1 Receptor Agonists and Risk for Gastroesophageal Reflux Disease in Patients With Type 2 Diabetes." Ann Intern Med 2025;178(9):1268-1278, PubMed 40658955.
- Liu BD et al. "Shorter-acting glucagon-like peptide-1 receptor agonists are associated with increased development of gastro-oesophageal reflux disease and its complications in patients with type 2 diabetes mellitus." Gut 2024;73(2):246-254, PubMed 37739778.
- Quast DR et al. "Effects of Lixisenatide Versus Liraglutide (Short- and Long-Acting GLP-1 Receptor Agonists) on Esophageal and Gastric Function in Patients With Type 2 Diabetes." Diabetes Care 2020;43(9):2137-2145, PubMed 32647054.
- Gala K et al. Clin Gastroenterol Hepatol 2026;24(9):2620-2622, PubMed 41213390; "Esophageal Motility Patterns Are Not Impacted By Glucagon-Like Peptide-1 Receptor Agonist Use" (2026), PubMed 42133868.
- PubMed search, 2026-10-17: (glucagon-like peptide-1 receptor agonist OR GLP-1 RA OR semaglutide OR tirzepatide OR liraglutide) AND (gastroesophageal reflux OR GERD), title/abstract fields — 36 records.