GLP-1 Side Effects by Label Frequency: What Section 6.1 of Each FDA Label Prints

GLP-1 side effects by label frequency are published in one place for each product: Section 6.1, Clinical Trials Experience, of the FDA-approved prescribing information. That section contains a table of adverse reactions that occurred more often on the drug than on placebo, above a stated threshold, in a named pool of trials. This page reproduces the symptom rows it covers (nausea, vomiting, diarrhea, constipation, abdominal pain, dyspepsia, eructation, fatigue, hair loss and injection-site reactions) from those tables for six labels, not the full tables; most of those tables also print rows this page does not carry, headache, dizziness and decreased appetite among them. All six were read on DailyMed, the National Library of Medicine's mirror of current FDA labeling, accessed 14 September 2026: Ozempic (semaglutide injection, Novo Nordisk, label revised May 2026), Wegovy (semaglutide injection and tablets, Novo Nordisk, revised June 2026), Rybelsus (oral semaglutide, Novo Nordisk, on a label shared with Ozempic tablets and revised January 2026), Mounjaro (tirzepatide, Eli Lilly, revised August 2026), Zepbound (tirzepatide, Eli Lilly, revised August 2026) and Saxenda (liraglutide, Novo Nordisk, revised June 2026).

Read across the six tables, the order of frequency is the same everywhere. Nausea comes first on every label. Diarrhea, vomiting and constipation follow in some order. Abdominal pain, dyspepsia and eructation sit lower. Fatigue, hair loss and injection-site reactions appear on some tables and not others, and where they are missing the label usually says why. The absolute percentages differ a great deal between labels, and the largest reason is not the molecule but the table's rules: the type 2 diabetes labels (Ozempic, Mounjaro, Rybelsus) list reactions reported by at least 5% of treated patients, while the weight-management labels (Wegovy, Zepbound, Saxenda) list reactions reported by at least 2%, in trials where patients were treated for up to 56, 68 or 72 weeks (Saxenda mean 46 weeks) versus mean exposures of 32.9, 36.6 and 41.8 weeks (Ozempic, Mounjaro, Rybelsus).

Two sentences from the labels frame everything below. Each one opens Section 6.1 with the statement that adverse reaction rates observed in the clinical trials of a drug cannot be directly compared to rates in the clinical trials of another drug and may not reflect the rates observed in practice. And five of the six, in their gastrointestinal subsections, state that nausea, vomiting and diarrhea were reported mostly during dose escalation; the sixth, Saxenda, states that the percentage of patients reporting nausea declined as treatment continued. This page is organised by symptom so that it answers the question a reader actually asks, how common is this, but the answer is always a label figure with its population attached, never a ranking. Nothing here is medical advice; a symptom that worries you belongs with your prescriber or pharmacist, not with a table.

Nausea is the first row of every Section 6.1 table. The Ozempic label's Table 1 (pool of two placebo-controlled trials in type 2 diabetes, 521 patients exposed, mean exposure 32.9 weeks) prints nausea at 6.1% on placebo, 15.8% on the trial's 0.5 mg arm and 20.3% on its 1 mg arm. The Wegovy label's Table 3 (three placebo-controlled trials in 2,116 adults with obesity or overweight, up to 68 weeks) prints 16% on placebo and 44% on Wegovy. The Mounjaro label's Table 1 (SURPASS-1 and SURPASS-5, 718 patients with type 2 diabetes, mean exposure 36.6 weeks) prints 4% on placebo and 12%, 15% and 18% across its 5 mg, 10 mg and 15 mg arms. The Zepbound label's Table 1 (Study 1 and Study 2, 2,519 adults with obesity or overweight, up to 72 weeks) prints 8% on placebo and 25%, 29% and 28% across its 5 mg, 10 mg and 15 mg arms. The Rybelsus label's Table 2 (two placebo-controlled trials in type 2 diabetes, 1,071 patients, mean exposure 41.8 weeks) prints 6% on placebo and 11% and 20% on its 7 mg and 14 mg arms. The Saxenda label's Table 2 (five placebo-controlled trials, 3,384 adults with overweight or obesity) prints 13.8% on placebo and 39.3% on Saxenda. All via DailyMed, accessed 14 September 2026.

Vomiting runs lower but follows the same shape. Ozempic: 2.3% placebo, 5% and 9.2%. Wegovy: 6% placebo, 24%. Mounjaro: 2% placebo, then 5%, 5% and 9%. Zepbound: 2% placebo, then 8%, 11% and 13%. Rybelsus: 3% placebo, then 6% and 8%. Saxenda: 3.9% placebo, 15.7%. The Wegovy and Saxenda labels also name vomiting among the most common reasons for permanent discontinuation, at 1.2% versus 0% and 1.7% versus under 0.1% respectively.

Diarrhea is the row where placebo rates are highest, which matters for reading the gap. Ozempic: 1.9% placebo, 8.5% and 8.8%. Wegovy: 16% placebo, 30%. Mounjaro: 9% placebo, then 12%, 13% and 17%. Zepbound: 8% placebo, then 19%, 21% and 23%, with a footnote that the row includes frequent bowel movements. Rybelsus: 4% placebo, then 9% and 10%. Saxenda: 9.9% placebo, 20.9%. The overall gastrointestinal figures put the three rows in context: the Wegovy label reports gastrointestinal adverse reactions in 73% of treated adults versus 47% on placebo, the Saxenda label about 68% versus 39%, the Zepbound label 56% at each arm versus 30%, the Mounjaro label 37.1% to 43.6% versus 20.4%, the Ozempic label 32.7% and 36.4% versus 15.3%, and the Rybelsus label 32% and 41% versus 21%. Timing, and what the labels say about reactions fading, is a separate question covered in GLP-1 side effects by week.

Constipation is the row that most separates the weight-management labels from the diabetes labels. The Wegovy label prints 24% versus 11% on placebo; the Saxenda label 19.4% versus 8.5%; the Zepbound label 17%, 14% and 11% across its three arms versus 5%, with a footnote that the row includes hard feces. The Mounjaro label prints 6%, 6% and 7% versus 1%; the Ozempic label 5% and 3.1% versus 1.5%; the Rybelsus label 6% and 5% versus 2% (each label's Section 6.1, via DailyMed, accessed 14 September 2026). Separately, every one of the six labels lists severe constipation including fecal impaction in Section 6.2 as a postmarketing report without a frequency. What to write down when constipation lasts, and what the labels say about it, is the subject of GLP-1 constipation.

Abdominal pain is printed on all six tables, often as a grouped term. The Wegovy label prints 20% versus 10%, and its footnote explains the row combines abdominal pain, upper and lower abdominal pain, gastrointestinal pain, abdominal tenderness, abdominal discomfort and epigastric discomfort. Zepbound prints 9%, 9% and 10% versus 5% with a similar grouped footnote. Rybelsus prints 10% and 11% versus 4%. Ozempic prints 7.3% and 5.7% versus 4.6%. Mounjaro prints 6%, 5% and 5% versus 4%. Saxenda prints abdominal pain at 5.4% versus 3.1% and, as a separate row, upper abdominal pain at 5.1% versus 2.7%.

Dyspepsia and eructation are where the diabetes labels drop below their 5% threshold and move the figure into the running text instead of the table. The Ozempic label states dyspepsia at 1.9% on placebo, 3.5% and 2.7% on its two arms, and eructation at 0%, 2.7% and 1.1%. The Rybelsus label lists dyspepsia at 0.6% on placebo, 3% on its 7 mg arm and 0.6% on its 14 mg arm, and eructation at 0%, 0.6% and 2%. The Mounjaro label prints dyspepsia in its table at 3% placebo, 8%, 8% and 5%, and eructation in the text at 0.4%, 3.0%, 2.5% and 3.3%. On the weight-management tables both rows appear in full: Wegovy dyspepsia 9% versus 3% and eructation 7% versus under 1%; Zepbound dyspepsia 9%, 9% and 10% versus 4% and eructation 4%, 5% and 5% versus 1%; Saxenda dyspepsia 9.6% versus 2.7% and eructation 4.5% versus 0.2%. Eructation is the label's word for belching; the sulfur-smelling version people describe online has no row of its own on any label, which is explained in sulfur burps on a GLP-1.

Fatigue is printed as a table row on three labels. The Wegovy label's Table 3 lists fatigue at 11% versus 5%, with a footnote that the row includes asthenia; its Table 4, from the two 72-week trials of the higher-strength injection, lists fatigue at 9% and 11% across its two Wegovy arms versus 5% on placebo. The Zepbound label's Table 1 lists fatigue at 5%, 6% and 7% versus 3%, footnoted as including asthenia, lethargy and malaise. The Saxenda label's Table 2 lists fatigue at 7.5% versus 4.6% and, separately, asthenia at 2.1% versus 0.8%. The other three labels are less specific. The Ozempic label says only that fatigue, dysgeusia and dizziness were associated with Ozempic at a frequency above 0.4%, with no placebo comparison. The Mounjaro label does not list fatigue in its placebo-controlled table; it reports 10% fatigue in a separate cardiovascular outcomes trial of 6,647 adults, again without a placebo figure in that sentence. The Rybelsus label's Section 6.1 does not list fatigue at all (all via DailyMed, accessed 14 September 2026). Why a symptom this common has so few numbers, and how to log it usefully, is covered in GLP-1 fatigue.

Hair loss follows the same split. The Wegovy label prints hair loss at 3% versus 1% in adults, 4% versus 0% in its pediatric table, and in the text gives 3.3% of patients on Wegovy (4% of women, 0.9% of men) versus 1% on placebo, rising to 5.8% in the trials of the higher-strength injection. The Zepbound label prints 5%, 4% and 5% versus 1%, and states in the text that hair loss was reported by 7.1% of women versus 0.5% of men on Zepbound, against 1.3% and 0% on placebo. Both labels state that hair loss adverse reactions were associated with weight reduction. The Ozempic, Mounjaro, Rybelsus and Saxenda labels list alopecia only under Skin and Subcutaneous Tissue in Section 6.2, Postmarketing Experience, where the label states that because reports are voluntary from a population of uncertain size it is not always possible to estimate their frequency. That is the correct citation when someone asks whether hair loss is on the Ozempic label: it is, without a number. The detail sits in GLP-1 hair loss.

Injection-site reactions span the widest range of any row, from 0.2% to 13.9%, and the range is mostly a difference in how each label counts. The Ozempic label reports injection-site reactions, such as discomfort and erythema, in 0.2% of treated patients in placebo-controlled trials. The Wegovy label reports 1.4% versus 1% on placebo. The Mounjaro label reports 3.2% versus 0.4%, and adds that in its seven-trial pool the rate was 4.6% among patients who developed anti-tirzepatide antibodies versus 0.7% among those who did not. The Zepbound label prints 6%, 8% and 8% versus 2% as a grouped table row covering bruising, erythema, pruritus, pain and rash, and states 11.3% among antibody-positive patients versus 1% among antibody-negative. The Saxenda label prints 13.9% versus 10.5%, the highest placebo rate on any row of this page, with erythema, pruritus and rash each reported by 1% to 2.5% of treated patients. Rybelsus is an oral tablet and has no such row.

Ozempic (semaglutide injection, Novo Nordisk). Table 1 of the label, via DailyMed, accessed 14 September 2026, is drawn from two placebo-controlled trials in adults with type 2 diabetes, one monotherapy trial and one in combination with basal insulin, reflecting 521 patients exposed to Ozempic for a mean of 32.9 weeks. Rows are included at 5% or more. The label states the arms as placebo (N=262), 0.5 mg (N=260) and 1 mg (N=261); those are trial randomisation arms as the label prints them, not a schedule. Selected rows (gastrointestinal, fatigue, hair loss, injection site), which in Ozempic's case is every row the table prints: nausea 6.1, 15.8, 20.3; vomiting 2.3, 5, 9.2; diarrhea 1.9, 8.5, 8.8; abdominal pain 4.6, 7.3, 5.7; constipation 1.5, 5, 3.1.

Wegovy (semaglutide injection, Novo Nordisk). Table 3 comes from three randomised, double-blind, placebo-controlled trials of the once-weekly injection in 2,116 adults with obesity or overweight, treated for up to 68 weeks, against 1,261 on placebo, with rows at 2% or more. Selected rows (gastrointestinal, fatigue, hair loss, injection site): nausea 16, 44; diarrhea 16, 30; vomiting 6, 24; constipation 11, 24; abdominal pain 10, 20; fatigue 5, 11; dyspepsia 3, 9; eructation under 1, 7; hair loss 1, 3. The label states the tablet's trial produced a similar profile and prints no separate table for it.

Mounjaro (tirzepatide, Eli Lilly). Table 1 pools SURPASS-1 and SURPASS-5 in adults with type 2 diabetes, 718 patients exposed for a mean of 36.6 weeks, rows at 5% or more, arms printed as placebo (N=235), 5 mg (N=237), 10 mg (N=240) and 15 mg (N=241). Selected rows (gastrointestinal, fatigue, hair loss, injection site): nausea 4, 12, 15, 18; diarrhea 9, 12, 13, 17; vomiting 2, 5, 5, 9; constipation 1, 6, 6, 7; dyspepsia 3, 8, 8, 5; abdominal pain 4, 6, 5, 5.

Zepbound (tirzepatide, Eli Lilly). Table 1 pools the label's Study 1 and Study 2, 2,519 adults with obesity or overweight treated for up to 72 weeks, rows at 2% or more, arms printed as placebo (N=958), 5 mg (N=630), 10 mg (N=948) and 15 mg (N=941). Selected rows (gastrointestinal, fatigue, hair loss, injection site): nausea 8, 25, 29, 28; diarrhea 8, 19, 21, 23; vomiting 2, 8, 11, 13; constipation 5, 17, 14, 11; abdominal pain 5, 9, 9, 10; dyspepsia 4, 9, 9, 10; injection-site reactions 2, 6, 8, 8; fatigue 3, 5, 6, 7; eructation 1, 4, 5, 5; hair loss 1, 5, 4, 5.

Rybelsus (oral semaglutide, Novo Nordisk). Table 2 of the combined Ozempic tablets and Rybelsus label pools two placebo-controlled trials in adults with type 2 diabetes, 1,071 patients exposed for a mean of 41.8 weeks, rows at 5% or more, arms printed as placebo (N=362), 7 mg (N=356) and 14 mg (N=356). Selected rows (gastrointestinal, fatigue, hair loss, injection site): nausea 6, 11, 20; abdominal pain 4, 10, 11; diarrhea 4, 9, 10; vomiting 3, 6, 8; constipation 2, 6, 5.

Saxenda (liraglutide, Novo Nordisk). Table 2 pools five double-blind, placebo-controlled trials in 3,384 adults with overweight or obesity, treated for up to 56 weeks (mean treatment duration 46 weeks, median 56), against 1,941 on placebo, rows at 2% or more. Selected rows (gastrointestinal, fatigue, hair loss, injection site): nausea 13.8, 39.3; diarrhea 9.9, 20.9; constipation 8.5, 19.4; vomiting 3.9, 15.7; injection-site reaction 10.5, 13.9; dyspepsia 2.7, 9.6; fatigue 4.6, 7.5; abdominal pain 3.1, 5.4; eructation 0.2, 4.5. The serious risks that sit above these tables, in the boxed warning and Section 5, are collected separately in GLP-1 warnings and precautions.

A Section 6.1 percentage is the share of trial participants who reported a reaction at least once during the trial, on a drug arm, next to the share who reported it on placebo. It is not a probability for you, it is not a severity, and it is not a duration. The placebo column is the most under-read part of the table: diarrhea at 16% on placebo in the Wegovy trials, or injection-site reactions at 10.5% on placebo in the Saxenda trials, tells you how much of a symptom is simply the background rate of a long trial in a large group. The gap between the two columns is the label's evidence, and the labels themselves warn that the gap cannot be lined up against another product's gap because the trials differ in population, duration, threshold and design.

The trial arms are printed here exactly as the labels print them because that is what a reference page owes its reader. They are not an escalation ladder, and this pillar does not publish one. Which product a person uses, at what strength, and how that changes over time is set by their prescriber; the pharmacist is the right person to ask about a label figure that seems to describe what they are feeling.

What a table cannot record is the thing a prescriber will ask about: which day after a dose the nausea arrived, whether constipation eased after a change in fibre or fluids, whether the fatigue tracked the injection day or the week. That is a dated note, not a percentage. GLP-1 Tracker, a free iPhone app that stores its data on the device, keeps each logged dose with its date and site next to weight, meals and a note about how the week went, so the timing is accurate when someone qualified asks; the procedure for a useful entry is in how to keep a GLP-1 symptom journal. The app organises what you enter and interprets nothing. Any symptom that is severe, persistent or frightening is a call to your prescriber or pharmacist, not a search for a percentage; every one of these labels lists ileus, intestinal obstruction and severe constipation including fecal impaction among its postmarketing reports, which is reason enough to treat a symptom that will not settle as a clinical question rather than a logging one.