Zepbound side effects: what the 2026 label says, by dose
Nausea hit 25 to 29 percent of adults taking Zepbound 5, 10, or 15 mg in the two placebo-controlled weight-loss trials, against 8 percent on placebo. Both figures come from the FDA prescribing information for Zepbound, revised August 2026. Zepbound side effects are common and mostly digestive. Most people who get them keep going, since 4.8 to 6.7 percent stopped the drug for an adverse reaction, against 3.4 percent on placebo.
The tables below put the rate for each dose next to placebo. One section covers how long episodes last, and one compares Zepbound with semaglutide using SURMOUNT-5, the first head-to-head trial in obesity. The sources are the label, the SURMOUNT-1 paper, a pooled SURMOUNT-1 to 4 analysis of digestive tolerability, the SURMOUNT-5 registry results, and an FDA safety notice.
Zepbound contains tirzepatide. The week-by-week arc for semaglutide is in our guide to how long semaglutide side effects last. The semaglutide version of this guide is Wegovy side effects.
Quick reference: Zepbound side effects
- Nausea: 25% (5 mg), 29% (10 mg), 28% (15 mg) vs 8% on placebo
- Any digestive reaction: 56% at every dose vs 30% on placebo
- Most nausea, vomiting, and diarrhea came during dose escalation. No source gives a median episode length
- Stopped for any side effect: 4.8% (5 mg), 6.3% (10 mg), 6.7% (15 mg) vs 3.4% on placebo
- Hair loss: 7.1% of women vs 0.5% of men on Zepbound
- Oral birth control: the label says to add a barrier method or switch for 4 weeks after starting and after each dose increase
Educational information, not medical advice. Talk to your prescriber before changing how you take any medication.
The most common Zepbound side effects, by dose
The most common Zepbound side effects are nausea, diarrhea, vomiting, constipation, and stomach pain, and nausea, diarrhea, and vomiting run higher on the 10 and 15 mg doses than on 5 mg. The label's Table 1 pools two placebo-controlled trials, Studies 1 and 2, which are SURMOUNT-1 and SURMOUNT-2 according to Lilly's medical information page. Together they treated 2,519 adults with obesity or overweight for up to 72 weeks. The average age was 47, 37 percent were men, and 25 percent had type 2 diabetes. The table lists every reaction that hit at least 2 percent of Zepbound users and ran higher than placebo.
| Side effect | 5 mg (N=630) | 10 mg (N=948) | 15 mg (N=941) | Placebo (N=958) |
|---|---|---|---|---|
| Nausea | 25% | 29% | 28% | 8% |
| Diarrhea | 19% | 21% | 23% | 8% |
| Vomiting | 8% | 11% | 13% | 2% |
| Constipation | 17% | 14% | 11% | 5% |
| Abdominal pain | 9% | 9% | 10% | 5% |
| Indigestion (dyspepsia) | 9% | 9% | 10% | 4% |
| Injection site reactions | 6% | 8% | 8% | 2% |
| Fatigue | 5% | 6% | 7% | 3% |
| Hypersensitivity reactions | 5% | 5% | 5% | 3% |
| Burping (eructation) | 4% | 5% | 5% | 1% |
| Hair loss | 5% | 4% | 5% | 1% |
| Reflux (GERD) | 4% | 4% | 5% | 2% |
| Gas (flatulence) | 3% | 3% | 4% | 2% |
| Bloating (abdominal distension) | 3% | 3% | 4% | 2% |
| Dizziness | 4% | 5% | 4% | 2% |
| Low blood pressure (hypotension) | 1% | 1% | 2% | 0% |
Eight percent of people on placebo also reported nausea, and 5 percent reported constipation or stomach pain. The drug effect is the difference between columns, which for nausea is 17 to 21 extra people per 100. Vomiting grows from 6 extra per 100 at 5 mg to 11 extra at 15 mg, and diarrhea from 11 to 15 extra.
Constipation runs the other way, falling from 17 percent at 5 mg to 11 percent at 15 mg. Hair loss sits at 4 to 5 percent on every dose against 1 percent on placebo, and the label ties it to weight reduction.
Digestive reactions and dropout rates by dose
Digestive reactions affected 56 percent of Zepbound users at every dose against 30 percent on placebo, and 1.9 to 4.3 percent stopped the drug because of them. Severe digestive reactions and dropouts both climb with the dose. The pattern is easier to see in one place.
| Measure | 5 mg | 10 mg | 15 mg | Placebo |
|---|---|---|---|---|
| Any digestive reaction | 56% | 56% | 56% | 30% |
| Severe digestive reaction | 1.7% | 2.5% | 3.1% | 1% |
| Stopped drug for digestive reasons | 1.9% | 3.3% | 4.3% | 0.5% |
| Stopped drug for any adverse reaction | 4.8% | 6.3% | 6.7% | 3.4% |
The label adds that most people who stopped for an adverse reaction did so in the first few months, mainly over digestive problems. The SURMOUNT-1 paper, which covers only Study 1, reports 4.3, 7.1, and 6.2 percent stopping on 5, 10, and 15 mg vs 2.6 percent on placebo. Across SURMOUNT-1 to 4, the pooled analysis found 27.8 to 72.8 percent of tirzepatide users with at least one digestive event vs 12.2 to 32.5 percent on placebo, and 1.0 to 10.5 percent stopping for it. The range is wide because the four trials differ. SURMOUNT-2 enrolled people with type 2 diabetes, and SURMOUNT-4 opened with a 36-week tirzepatide lead-in.
Nausea does not explain the weight loss. In that analysis, weight loss was similar in people with no nausea, nausea alone, or any nausea, vomiting, or diarrhea. Modeling tied nausea, vomiting, diarrhea, and indigestion to at most 3.1 percent of total weight reduction.
How long Zepbound side effects last
Most nausea, vomiting, and diarrhea on Zepbound happened during dose escalation and decreased over time, but no source gives a median length for a single episode. The label says so in these words: "The majority of nausea, vomiting, and/or diarrhea events occurred during dose escalation and decreased over time." Neither the label, the SURMOUNT-1 paper, the pooled SURMOUNT-1 to 4 analysis, nor Lilly's medical information page says how many days one bout of nausea lasts. The closest figure measures when side effects start, not how long they last. A 2026 analysis of FDA adverse event reports, which are spontaneous reports and not trial data, found that most tirzepatide digestive events occurred within 3 months, with a median onset of 16 days.
The semaglutide trials went further. The pooled STEP analysis behind our Wegovy side effects guide reported median episode lengths in days. No equivalent has been published for tirzepatide.
What the trials do document is timing. All four SURMOUNT weight-loss trials used a 20-week dose-escalation period, and the pooled analysis says digestive events were "primarily reported during dose escalations." First use of anti-nausea and anti-diarrhea medicine was most frequent in the first 24 weeks. The label's dropout data point the same way, with most stopping in the first few months. The Medication Guide sets no number of days. It says to tell your provider about stomach problems that are severe or will not go away.
Serious side effects of Zepbound and warning signs
The serious side effects of Zepbound are uncommon, and the label's Warnings and Precautions section lists ten entries. These are the ones with a symptom you can recognize, with the trial rate where the label gives one.
Severe digestive reactions. Severe reactions were reported in 1.7 percent (5 mg), 2.5 percent (10 mg), and 3.1 percent (15 mg) of users vs 1 percent on placebo. The label does not recommend Zepbound in people with severe gastroparesis, a slow-emptying stomach. Postmarketing reports also include ileus, intestinal obstruction, and severe constipation with fecal impaction.
Kidney injury from dehydration. Postmarketing reports include acute kidney injury, some cases needing hemodialysis, mostly in people who lost fluid to nausea, vomiting, or diarrhea. In the trials it was reported in 0.5 percent of Zepbound users vs 0.2 percent on placebo. The label tells prescribers to monitor kidney function in people reporting such reactions, especially during dose initiation and escalation.
Gallbladder disease. Cholecystitis, inflammation of the gallbladder, occurred in 0.7 percent of Zepbound users vs 0.2 percent on placebo. Gallstones were near even, 1.1 percent vs 1 percent. The label says these events were associated with weight reduction. The Medication Guide lists upper stomach pain, fever, yellow skin or eyes, and clay-colored stools as reasons to call right away.
Pancreatitis. Adjudicators confirmed acute pancreatitis in 0.2 percent of Zepbound users and 0.2 percent on placebo in the weight-loss trials. In the two sleep apnea trials, 467 people in all, the rate was 0.84 per 100 years on Zepbound and 0 on placebo. The warning sign is persistent or severe belly pain, sometimes reaching the back, with or without nausea or vomiting. The label says to discontinue Zepbound if pancreatitis is suspected.
Allergic reactions. Severe hypersensitivity reactions occurred in 0.1 percent of Zepbound users and no one on placebo, and postmarketing reports include anaphylaxis and angioedema. The Medication Guide lists facial or throat swelling, trouble breathing or swallowing, severe rash, fainting, and a very rapid heartbeat.
Low blood sugar. In the type 2 diabetes trial, blood sugar below 54 mg/dL occurred in 4.2 percent of Zepbound users vs 1.3 percent on placebo. With a sulfonylurea the rate was 10.3 percent, vs 2.1 percent without one. The label tells prescribers to consider lowering insulin or sulfonylurea doses when Zepbound starts.
Eye, surgery, and pen safety. The label reports temporary worsening of diabetic retinopathy with rapid improvement in blood sugar control. It reports rare cases of food or liquid entering the lungs during anesthesia despite fasting, and tells patients to inform providers before any procedure. It says never to share a KwikPen between people, even with a new needle.
The thyroid tumor warning has its own section below. For these risks, the Medication Guide tells people to contact their provider, and to stop Zepbound and get medical help right away for signs of a serious allergic reaction.
Tirzepatide side effects and cancer risk
Zepbound carries a boxed warning for thyroid C-cell tumors because tirzepatide caused them in rats, and whether it causes them in people is unknown. A boxed warning is the label's most prominent warning. It rests on animal data, not on human cases.
In a 2-year study, rats received tirzepatide twice weekly at 0.15, 0.5, or 1.5 mg/kg, which is 0.1 to 1 times the exposure of the 15 mg human dose. C-cell adenomas rose significantly in male rats at 0.5 mg/kg and above and in female rats at 0.15 mg/kg and above. Adenomas and carcinomas combined rose in both sexes at every dose. In a 6-month study in transgenic mice, tirzepatide was not tumorigenic.
For people, the label's language is direct. It is unknown whether Zepbound causes thyroid C-cell tumors, including medullary thyroid carcinoma (MTC), because the human relevance of the rodent findings has not been determined. The adverse reactions section reports no thyroid cancer rate. Zepbound is contraindicated in anyone with a personal or family history of MTC or with Multiple Endocrine Neoplasia syndrome type 2 (MEN 2), an endocrine system condition.
The label lists symptoms to report: a neck mass, trouble swallowing, shortness of breath, or persistent hoarseness. It says routine calcitonin blood tests or thyroid ultrasound are of uncertain value for catching MTC early. The SURMOUNT-5 registry results list no thyroid tumor among serious adverse events in either arm. That is 72 weeks in about 750 people, far too short and small to settle a cancer question.
Zepbound side effects in females
Zepbound side effects in females differ most on hair loss, and the label adds instructions on birth control pills, pregnancy, and breastfeeding. In the pooled weight-loss trials, hair loss was reported by 7.1 percent of women on Zepbound and 0.5 percent of men. On placebo the split was 1.3 percent of women and 0 percent of men. No Zepbound user and one placebo user stopped treatment because of hair loss. The label says hair loss reactions were associated with weight reduction. Our guide to semaglutide hair loss explains why shedding follows fast weight loss on any GLP-1. Its timing data come from semaglutide.
Birth control pills. The label says Zepbound may reduce the effectiveness of oral hormonal contraceptives because it delays gastric emptying, and that the delay is largest after the first dose and shrinks over time. It advises switching to a non-oral method, or adding a barrier method, for 4 weeks after starting Zepbound and for 4 weeks after each dose increase. Hormonal contraceptives that are not taken by mouth should not be affected. Which method fits is a decision for your prescriber.
Pregnancy. The label says weight loss offers no benefit to a pregnant patient and may cause fetal harm, and that Zepbound should be discontinued when a pregnancy is recognized. Human data are insufficient to evaluate the risk of birth defects or miscarriage, and animal studies suggest there may be risks to the fetus. A pregnancy exposure registry exists. For the fertility side of the question, see GLP-1 fertility and PCOS.
Breastfeeding. In 11 healthy lactating women given a single 5 mg dose, tirzepatide was undetectable in 164 of 171 breast milk samples. The label says there are no data on effects on a breastfed infant or on milk production.
Other Zepbound side effects in the label
The label documents a second tier of side effects beyond nausea. Injection site reactions hit 6 to 8 percent of Zepbound users vs 2 percent on placebo, and the label ties them to antibodies, at 11.3 percent in people who developed anti-tirzepatide antibodies vs 1 percent in those who did not. Hypersensitivity reactions of any kind ran 5 percent vs 3 percent, mostly rash and itching. Fatigue ran 5 to 7 percent vs 3 percent, burping 4 to 5 percent vs 1 percent, reflux 4 to 5 percent vs 2 percent, and dizziness 4 to 5 percent vs 2 percent.
Low blood pressure was reported in 1.6 percent vs 0.1 percent, more often in people also taking blood pressure drugs (2.2 percent vs 1.2 percent). Heart rate rose by a mean of 1 to 3 beats per minute, with no increase on placebo. Dry mouth or throat was reported by 1 percent vs 0.1 percent, and altered taste by 0.4 percent vs none. Blood tests showed mean rises of 20 to 25 percent in pancreatic amylase and 28 to 35 percent in lipase, vs 2.1 and 5.8 percent on placebo. The label says their significance is unknown without other signs of pancreatitis.
Zepbound vs Wegovy side effects in a head-to-head trial
Zepbound and Wegovy caused nausea and diarrhea at nearly the same rates in SURMOUNT-5, and the two drugs split on vomiting, reflux, and injection site reactions. The SURMOUNT-5 trial randomized 751 adults with obesity and no diabetes to the maximum tolerated dose of tirzepatide (10 or 15 mg) or semaglutide (1.7 or 2.4 mg) for 72 weeks. It was open-label, and both drugs started low and climbed every 4 weeks. Average weight change was 20.2 percent with tirzepatide and 13.7 percent with semaglutide.
| Event | Tirzepatide | Semaglutide |
|---|---|---|
| Nausea | 43.6% (163) | 44.4% (167) |
| Constipation | 27.0% (101) | 28.5% (107) |
| Diarrhea | 23.5% (88) | 23.4% (88) |
| Vomiting | 15.0% (56) | 21.3% (80) |
| Reflux (GERD) | 6.1% (23) | 10.6% (40) |
| Fatigue | 10.4% (39) | 12.2% (46) |
| Burping | 9.9% (37) | 7.7% (29) |
| Hair loss | 8.3% (31) | 6.1% (23) |
| Injection site reaction | 8.6% (32) | 0.3% (1) |
| Any serious adverse event | 4.8% (18) | 3.5% (13) |
Semaglutide users had more vomiting, 21.3 percent vs 15.0 percent, and more reflux, 10.6 percent vs 6.1 percent. Tirzepatide users had more injection site reactions, 8.6 percent vs 0.3 percent, and slightly more hair loss, 8.3 percent vs 6.1 percent. Serious adverse events were 4.8 percent vs 3.5 percent. A summary of the NEJM paper from the American College of Cardiology reports that digestive events causing people to stop treatment were more common with semaglutide (5.6 percent) than tirzepatide (2.7 percent).
Treat the table as registry counts, not a significance test. The registry lists only events reaching 5 percent, and the percentages are our arithmetic. Still, this is a direct randomized comparison of the two drugs in obesity, and it beats lining up separate label tables, which the label itself warns against. For semaglutide rates by dose, see Wegovy side effects. The semaglutide schedule itself is in our Wegovy dosage chart. For the full drug-versus-drug page, see Wegovy vs Zepbound.
What changed on the Zepbound label in 2026
The Zepbound label was revised in August 2026, and its Recent Major Changes box lists five items dated January through August 2026. Older side effect pages will not reflect them. This section follows the DailyMed version that was live on September 30, 2026.
The suicidal thoughts warning is gone. On January 13, 2026, the FDA asked the makers of Saxenda, Wegovy, and Zepbound to remove the suicidal behavior and ideation warning after reviewing 91 placebo-controlled trials with 107,910 patients. The agency found no increased risk of suicidal behavior or ideation, or of anxiety, depression, irritability, or psychosis. The Zepbound label logs the removal in February 2026. Any page still listing suicidal thoughts as a Zepbound warning is out of date.
The other four entries are the severe digestive reactions warning (section 5.2, February 2026), the diabetic retinopathy warning (section 5.8, August 2026), the KwikPen sharing warning (section 5.10, January 2026), and a dosage and administration update (section 2.4, January 2026).
Why the Zepbound dose climbs slowly
The label says the escalation schedule exists "to reduce the risk of gastrointestinal adverse reactions." Zepbound starts at 2.5 mg once weekly for 4 weeks, a dose the label says is not approved for maintenance. The dose then rises in 2.5 mg steps after at least 4 weeks on the current dose, to a maintenance dose of 5, 10, or 15 mg. The SURMOUNT trials built in a 20-week escalation period. For the schedule step by step, see our guide to tirzepatide dosing for weight loss in units.
The label also leaves room for tolerability. It tells prescribers to weigh response and tolerability when choosing a maintenance dose, and says a lower maintenance dose can be considered if one is not tolerated. On missed doses, it tells prescribers to instruct patients to take the dose within 4 days (96 hours) and to skip it after that.
Schedule changes belong to your prescriber. Our directory of GLP-1 prescribers lists providers by state. The Zepbound medication page covers the drug itself. What each dose costs is in our guide to Zepbound cost. For the same molecule sold for diabetes, see Mounjaro vs Zepbound.
Frequently asked questions
What is the most common side effect of Zepbound?
Nausea is the most common Zepbound side effect. In the two pooled weight-loss trials in the FDA label, nausea affected 25 percent of adults on 5 mg, 29 percent on 10 mg, and 28 percent on 15 mg, against 8 percent on placebo. Diarrhea came next at 19 to 23 percent vs 8 percent, then constipation at 11 to 17 percent vs 5 percent. Vomiting ran 8 to 13 percent vs 2 percent. Most of these events happened during dose escalation and decreased over time, according to the label.
How long do Zepbound side effects last?
No trial or label source gives a median length for one episode of Zepbound nausea, vomiting, or diarrhea. An analysis of FDA adverse event reports put the median onset at 16 days. The label says most of those events occurred during dose escalation and decreased over time, and the SURMOUNT trials used a 20-week escalation period. The pooled SURMOUNT-1 to 4 analysis says digestive events were primarily reported during dose escalations. The Medication Guide tells patients to contact their provider about stomach problems that are severe or will not go away.
What are the serious side effects of Zepbound?
The Zepbound label lists these serious risks: thyroid C-cell tumors (boxed warning), severe digestive reactions, acute kidney injury from dehydration, gallbladder disease, pancreatitis, serious allergic reactions, low blood sugar with insulin or sulfonylureas, worsening diabetic retinopathy, and aspiration during anesthesia. Serious digestive reactions occurred in 1.7 to 3.1 percent of users vs 1 percent on placebo. Severe allergic reactions occurred in 0.1 percent vs none. The Medication Guide says to call right away about severe belly pain that will not go away.
Does Zepbound cause thyroid cancer?
It is unknown whether Zepbound causes thyroid C-cell tumors, including medullary thyroid carcinoma, in people. The boxed warning exists because tirzepatide caused these tumors in rats. Zepbound is contraindicated in people with a personal or family history of medullary thyroid carcinoma or MEN 2. The label lists a neck mass, trouble swallowing, shortness of breath, and persistent hoarseness as symptoms to report. It reports no thyroid cancer rate from the weight-loss trials, and it says routine calcitonin tests or thyroid ultrasound are of uncertain value for early detection.
Does Zepbound cause hair loss in women?
Zepbound can cause hair loss in women. In the pooled weight-loss trials, 7.1 percent of women on Zepbound reported it, against 1.3 percent on placebo. Among men the rates were 0.5 percent on Zepbound and 0 percent on placebo. Across both sexes, 4 to 5 percent of Zepbound users had hair loss vs 1 percent on placebo. The label says hair loss was associated with weight reduction. No Zepbound user stopped treatment because of it, and one placebo user did.
Does Zepbound make birth control less effective?
Zepbound can make birth control pills less effective. The Zepbound label says the drug may reduce the effectiveness of oral hormonal contraceptives because it delays gastric emptying, and the delay is largest after the first dose. It advises switching to a non-oral method or adding a barrier method for 4 weeks after starting Zepbound and for 4 weeks after each dose increase. Hormonal contraceptives that are not taken by mouth should not be affected. A prescriber can advise which option fits.
Are Zepbound side effects worse than Wegovy side effects?
The head-to-head SURMOUNT-5 trial found no clear winner. Nausea was nearly identical, 43.6 percent on tirzepatide and 44.4 percent on semaglutide, and diarrhea matched at about 23 percent. Semaglutide users had more vomiting (21.3 vs 15.0 percent) and reflux (10.6 vs 6.1 percent). Tirzepatide users had more injection site reactions (8.6 vs 0.3 percent) and slightly more hair loss. Serious adverse events were 4.8 percent on tirzepatide and 3.5 percent on semaglutide. The trial was open-label.