Ozempic Side Effects: What Happens, How Common, and How Long

Nausea is the one you should expect. It showed up in 15.8 percent of people on 0.5 mg and 20.3 percent on 1 mg, against 6.1 percent on placebo, and most of it happened in the weeks right after a dose increase rather than at a fixed point in treatment.

Every number on this page comes from Novo Nordisk’s prescribing information for Ozempic: a pool of two placebo-controlled trials in adults with type 2 diabetes, 521 people exposed to Ozempic for a mean of 32.9 weeks, plus the warnings and postmarketing sections. Not from an average of other articles.

Quick answer

Nausea leads at 15.8 to 20.3 percent depending on dose, then diarrhea at 8.5 to 8.8 percent, vomiting at 5 to 9.2 percent, abdominal pain at 5.7 to 7.3 percent and constipation at 3.1 to 5 percent. Most of it clusters around dose escalation and settles once you hold a dose. Between 3.1 and 3.8 percent of people stop because of it, against 0.4 percent on placebo. Three things the label warns about that most articles skip: your eyes, your gallbladder, and telling your anaesthetist.

Ozempic side effects by dose, from the label

Placebo n=262, 0.5 mg n=260, 1 mg n=261, pooled from two placebo-controlled trials: one of Ozempic on its own, one of Ozempic added to basal insulin.

A note on where these numbers come from, because it matters: this was a type 2 diabetes population at 0.5 and 1 mg. If you are on Ozempic off-label for weight, or you are at 2 mg, you are outside the population that produced these percentages. Wegovy, which is the same molecule at higher doses, reports higher rates.

The pattern in that table nobody writes about

Read the constipation row again. 5 percent, then 3.1. It gets less common as the dose goes up.

Abdominal pain does the same thing: 7.3 percent, then 5.7.

Now read the vomiting row. 5 percent, then 9.2. That one nearly doubles.

So “everything gets worse at higher doses” is not what the table shows. The profile changes shape as you climb: the low end leans more towards constipation and stomach ache, the top end more towards vomiting.

Two honest caveats. Novo publishes no statistical test on those gaps, and at roughly 260 people per arm the difference between 7.3 and 5.7 percent is about four patients, small enough to be chance. And the constipation half is the part that holds up elsewhere: Lilly’s tirzepatide label shows the same fall across three doses and much larger numbers, 17 then 14 then 11 percent, with vomiting rising 8 to 11 to 13. Abdominal pain does not fall there (9, 9, 10), so the stomach-ache half looks specific to this Ozempic table rather than to the drug class.

Either way, the practical point stands: write down which dose you were on when something started. “I get nauseous on Ozempic” is not a useful sentence at your next appointment. “Nausea started the week I went to 1 mg and it is easing” is.

When they start, and how long they actually last

Here is the honest version, because this question attracts more invented answers than any other.

The label gives no duration in days. What it says is that the majority of reports of nausea, vomiting and diarrhea occurred during dose escalation. That is the whole of it. Anyone telling you “two to four weeks” is quoting a number that does not exist in the source.

The shape is still clear: worst after a step-up, easier once you hold.

On timing inside the week, the pharmacology helps. Semaglutide peaks one to three days after your shot, its half-life is about a week, and steady state takes four to five weeks of dosing. So the drug is still fully on board when the next dose is due, nothing clears between shots, and your first month is not what maintenance will feel like.

Does Ozempic cause nausea, and what actually helps

15.8 percent at 0.5 mg, 20.3 percent at 1 mg, against 6.1 percent on placebo. It is the most common side effect and the most common reason people stop.

What the label supports: it concentrates around dose increases, and the majority of it happens there.

One thing worth knowing rather than assuming: 0.5 mg, 1 mg and 2 mg are all approved maintenance doses. Moving up is something your prescriber decides with you based on how you are responding, not a schedule you are behind on. That is a fact about the label rather than a suggestion about your dose, and where you settle is a conversation with the person who prescribed it.

Does Ozempic cause diarrhea?

8.5 percent at 0.5 mg and 8.8 percent at 1 mg, against 1.9 percent on placebo. Notably flat between doses, unlike nausea.

It matters more than the number suggests because of what sits downstream of it. The label warns about acute kidney injury caused by dehydration from vomiting and diarrhea, and acute kidney injury appears in the postmarketing section too. The mechanism is preventable and the prevention is fluid.

Does Ozempic cause constipation?

5 percent at 0.5 mg, falling to 3.1 percent at 1 mg, against 1.5 percent on placebo.

Worth knowing that the postmarketing section is blunter than the trial table here: it lists severe constipation including faecal impaction, along with ileus and intestinal obstruction. Those are rare enough to have no frequency attached, but constipation that is not resolving is worth raising rather than waiting out.

Does Ozempic cause “Ozempic face”?

Not as a side effect of the drug, no. Facial volume loss appears nowhere in the prescribing information.

What is happening is simpler and it is worth saying plainly: the fat pads that give cheeks and temples their shape are among the first to go when you lose weight quickly, and skin does not tighten as fast as fat disappears. The same thing happens after rapid weight loss from any cause. The nickname stuck to Ozempic because Ozempic made rapid weight loss common, not because semaglutide does something to your face.

The only lever the medication gives you is pace. Slower loss gives skin more time.

Does Ozempic cause hair loss?

Alopecia is listed in the postmarketing section of the label. Postmarketing means it has been reported since approval but without a reliable frequency, so there is no percentage to quote and anyone giving you one is guessing.

The likeliest mechanism is the same as the face: rapid weight loss and reduced intake, which is a well-documented trigger for temporary shedding. Protein intake matters here more than most people expect, and appetite suppression is exactly what makes it hard.

The eye one, which most articles skip

This one deserves more attention than it gets. In Novo’s cardiovascular outcomes trial, diabetic retinopathy complications occurred in 3.0 percent of people on Ozempic against 1.8 percent on placebo.

The label goes further, and this is the number that should decide what you do about it. The increase was concentrated almost entirely in people who already had retinopathy: 8.2 percent against 5.2 percent on placebo among those with a history of it, versus 0.7 percent against 0.4 percent among those without. For most people the absolute risk is very small. For people already being treated for retinopathy it is not.

Novo does not say what caused it. What the label does say, separately, is that rapid improvement in glucose control has been associated with temporary worsening of diabetic retinopathy, and that the effect of long-term glycaemic control has not been established. Connecting those two is inference, not something Novo states.

If you have diabetic retinopathy, the label says you should be monitored. Worth raising before you start rather than after.

Before any surgery, tell them you take Ozempic

This is the newest warning on the label and most content has not caught up with it.

Ozempic slows how fast your stomach empties. There have been rare reports, across GLP-1 drugs as a class rather than Ozempic specifically, of people undergoing general anaesthesia or deep sedation who still had food in their stomach despite following the fasting instructions, which creates a risk of aspiration.

Novo is honest that the data are not yet good enough to say whether pausing the drug or fasting longer would help. So the instruction is simple and it is on you: tell your healthcare providers before any planned surgery or procedure that you take Ozempic. Include dental work under sedation and anything involving an anaesthetist.

Gallstones, and a number that looks backwards

Cholelithiasis was reported in 1.5 percent of people on 0.5 mg and 0.4 percent on 1 mg.

That is the higher rate at the lower dose, which is not what you would expect, and it is worth stating rather than smoothing over. With numbers this small in a trial of a few hundred people, the gap may be noise rather than signal. What is established is that gallbladder problems track with weight loss generally, and cholecystitis and gallbladder removal both show up in the postmarketing section.

Pain in the upper right of your abdomen, particularly after a fatty meal, is the one to mention.

What happens if you miss a dose

Take it as soon as possible within 5 days. Past that, skip it and take your next dose on your usual day.

Because the half-life is about a week, one late shot does not undo anything. What it does do is blur your read on the following week, which is the argument for writing down that it happened.

The serious ones, in proportion

The boxed warning is for thyroid C-cell tumours, based on rodent studies. Ozempic is contraindicated if you or your family have medullary thyroid carcinoma or MEN 2.

Then the warnings with trial numbers behind them:

• Severe gastrointestinal reactions. 0.4 percent at 0.5 mg and 0.8 percent at 1 mg, against none on placebo.
• Pancreatitis. Seven adjudicated cases in Ozempic-treated patients, 0.3 per 100 patient-years, against three in comparator-treated patients. The postmarketing section adds necrotising pancreatitis, sometimes fatal, which is why persistent severe abdominal pain is not a wait-and-see symptom.
• Diabetic retinopathy complications. 3.0 percent against 1.8 percent, as above.
• Low blood sugar with a sulfonylurea. Severe hypoglycaemia in 0.8 percent at 0.5 mg and 1.2 percent at 1 mg. If you take a sulfonylurea or insulin, your prescriber may reduce that dose when you start.
• Heart rate. A mean increase of 2 to 3 beats per minute, against a mean decrease of 0.3 on placebo.
• Injection site reactions. 0.2 percent, which is dramatically lower than tirzepatide’s 6 to 8 percent.

Reported after approval, without reliable frequencies: anaphylaxis, angioedema, rash and urticaria; ileus and intestinal obstruction; cholecystitis and gallbladder removal; acute kidney injury; alopecia; headache and dysesthesia.

One more, easy to miss: Novo says to stop Ozempic at least 2 months before a planned pregnancy, which the label attributes to semaglutide’s long washout period. For reference, the label says semaglutide is still present for about 5 weeks after the last dose.

When to call your prescriber

• Persistent severe abdominal pain, especially radiating to your back, with or without vomiting. That is the pancreatitis presentation.
• Vomiting or diarrhea you cannot stay ahead of with fluids, because that is the route to kidney injury.
• Pain in the upper right abdomen after fatty meals.
• Any change in your vision, particularly if you have diabetic retinopathy.
• Any allergic reaction: swelling, trouble breathing, a spreading rash.
• Constipation that is not resolving, or a stomach that stops emptying.

The unglamorous truth about these appointments is that they go better with data. “Nausea, days one and two, every week since the 1 mg step on August 11, worst at 7 out of 10” gets you somewhere. “I’ve been feeling rough” gets you sympathy.

Ozempic versus Wegovy versus Zepbound

Ozempic is the gentlest of the three on the numbers that exist, but the trials are not directly comparable and it would be dishonest to pretend otherwise. Ozempic’s figures come from a type 2 diabetes population at 0.5 and 1 mg. Wegovy is the same molecule at up to 2.4 mg in an obesity population. Zepbound is a different molecule again.

Injection site reactions are the figures people reach for: Ozempic 0.2 percent, Wegovy 1.4 percent, Zepbound 6 to 8 percent. The gap is real and it does surprise people who switch, but it is not the clean comparison it looks like. Ozempic’s is a single narrative figure; Wegovy’s and Zepbound’s are composites of five and six separate terms (bruising, redness, itching, pain, rash, reaction) counted over much longer trials. Lilly also ties most of its rate to immunogenicity: 11.3 percent in people who developed anti-tirzepatide antibodies against 1 percent in those who did not.

So reactions are genuinely more common on tirzepatide, and the size of the gap is not something these three labels can settle.

For the full tirzepatide picture, Zepbound’s side effects by dose has the same treatment of Lilly’s label.

How common is it to quit?

3.1 percent on 0.5 mg and 3.8 percent on 1 mg stopped because of gastrointestinal side effects, against 0.4 percent on placebo.

Read that carefully, because it is narrower than it looks: it counts people who stopped because of gastrointestinal side effects, not everyone who stopped for any reason, which Novo does not report for these trials. So it is not a retention rate. What it does say is that fewer than 1 in 25 people left because of the symptoms this page is about, and that most of what makes people consider it happens during escalation, which is the part that passes.

Sources

All Ozempic percentages, the discontinuation rates, the retinopathy and gallstone figures, the heart rate change, the hypoglycaemia numbers, the pulmonary aspiration warning, the missed-dose window, the pregnancy washout and the postmarketing list come from Novo Nordisk’s prescribing information for Ozempic (semaglutide) injection, revised May 2026, sections 2.1, 2.2, 5.1 to 5.10, 6.1, 6.2, 8.3 and 12.3, as published on DailyMed (setid adec4fd2-6858-4c99-91d4-531f5f2a2d79). The adverse reaction table is pooled from two placebo-controlled trials, not one.

The Zepbound and Wegovy comparison figures come from Eli Lilly’s Zepbound prescribing information and Novo Nordisk’s Wegovy prescribing information.

Sources verified 9 September 2026.

This article is for tracking and education. It is not medical advice, and dosing questions belong with your prescriber.