Search volume for "GLP-1 overdose" is up sharply — the term has risen faster over the past year than anything else we track on this site. So we did the obvious thing and went looking for what an overdose is.
The labels do not say. Not one of them.
Every figure below carries its source and the date we read it. This page reports what regulatory documents and the FDA's adverse-event database record. It is not a guide to what to do about a dose, and it does not tell anyone what to take; if a dose has been taken in error, the Poison Help line on 1-800-222-1222 answers that question and every one of these labels points there.
Seven labels, no number
We read the OVERDOSAGE section of all seven approved US labels for this class via openFDA on 2026-09-16.
| Label | Effective | Does it state an overdose dose? | Effects described |
|---|---|---|---|
| Wegovy (semaglutide 2.4 mg) | 2024-04-23 | No | Severe nausea, severe vomiting, severe hypoglycemia — attributed to other GLP-1 receptor agonists |
| Ozempic (semaglutide) | 2026-01-30 | No | None |
| Rybelsus (oral semaglutide) | 2026-01-30 | No | None |
| Zepbound (tirzepatide) | 2026-08-28 | No | None |
| Mounjaro (tirzepatide) | 2026-07-29 | No | None |
| Saxenda (liraglutide) | 2026-02-25 | No | Severe nausea, severe vomiting, severe hypoglycemia |
| Victoza (liraglutide) | 2023-02-06 | No | Severe nausea, severe vomiting, severe hypoglycemia |
Four molecules, two manufacturers, labels revised across four years, and none of them defines the thing. That consistency is the finding. A label that omitted a number by accident would be one label; seven is a convention.
What they give instead is identical across the board: contact the Poison Help line on 1-800-222-1222 or a medical toxicologist, and initiate supportive treatment according to the patient's clinical signs and symptoms.
The one detail worth pulling out of the Wegovy label is its attribution. It says: "Overdoses have been reported with other GLP-1 receptor agonists." Not with semaglutide. The three effects most often quoted in coverage of semaglutide overdose are, on semaglutide's own weight-management label, borrowed from a different drug.
The number the labels do give is the clock
Where the labels are specific, it is about time, and it is the part most worth understanding.
Wegovy: "A prolonged period of observation and treatment for these symptoms may be necessary, taking into account the long half-life of WEGOVY of approximately 1 week."
Zepbound: "A period of observation and treatment for these symptoms may be necessary, taking into account the half-life of tirzepatide of approximately 5 days."
This is the practical difference between these medicines and most drugs people worry about overdosing on. A week is not an evening in an emergency department. Whatever an excess dose does, it does slowly and for a long time, and that is why the labels emphasise the duration of observation rather than an antidote — there is no reversal agent named on any of them.
What the adverse-event record actually holds
We queried FDA's adverse event reporting system on 2026-09-16. The result reframes the question.
The single most reported term for tirzepatide is not a symptom.
| Tirzepatide — most frequently reported terms | Reports |
|---|---|
| INCORRECT DOSE ADMINISTERED | 29,672 |
| Nausea | 17,092 |
| Injection site pain | 13,906 |
| Diarrhoea | 10,006 |
| EXTRA DOSE ADMINISTERED | 9,608 |
| Off label use | 8,941 |
| Vomiting | 8,267 |
| Drug ineffective | 6,965 |
| Constipation | 6,502 |
| ACCIDENTAL UNDERDOSE | 4,609 |
Out of 160,663 tirzepatide reports in total, a dosing error is reported more often than nausea. Three of the top ten terms are dosing errors, and one of them is an error in the opposite direction.
The term "Overdose" itself is much rarer, and the split inside it is the most telling number on this page:
| Tirzepatide | Semaglutide | |
|---|---|---|
| Total reports in the database | 160,663 | 73,001 |
| Reports carrying Overdose | 2,837 | 372 |
| Accidental overdose | 2,060 | 108 |
| Intentional overdose | 16 | 52 |
| Incorrect dose administered | 29,672 | 1,189 |
| Extra dose administered | 9,608 | 260 |
| Product dose omission issue | 4,474 | 1,358 |
| Underdose | 5,219 | 140 |
For tirzepatide, accidental overdose outnumbers intentional overdose by about 129 to 1.
So the question behind the search term is, in the record, a question about a mistake — a doubled dose, a misread pen, a vial measured wrong — and not about deliberate harm.
How serious are they
Of the 2,837 tirzepatide reports carrying Overdose:
| Classification | Reports | Share |
|---|---|---|
| Non-serious | 2,711 | 95.6% |
| Serious | 126 | 4.4% |
| — of which hospitalization | 59 | 2.1% |
| — of which life-threatening | 11 | 0.4% |
| — of which death | 4 | 0.1% |
One more caution about all of these counts, which we can demonstrate rather than assert. When this site queried the same database on 2026-09-10 for our tirzepatide review, semaglutide's incorrect-dose-administered total came back as 880. The same query on 2026-09-16 returned 1,189. Six days, a 35% move. These are live counts against a database that is added to continuously, so any figure here is a reading with a date attached, not a fixed quantity.
These figures need their limits stated, and the limits are severe. FAERS is a voluntary, spontaneous reporting system. Reports are unverified, can be duplicated, are submitted by manufacturers, clinicians and members of the public alike, and carry no denominator — nothing here says what share of people using the drug this represents, because the database does not know how many people use it. Above all, a report does not establish that the drug caused the event. A seriousness classification is an administrative field on a form, not a clinical adjudication.
What the table supports is narrow: among reports that reached the database with this term attached, the large majority were logged as non-serious. It does not support "GLP-1 overdose is safe", and nothing here should be read that way.
One label warning belongs beside it. Severe hypoglycemia is the effect the liraglutide labels name, and the risk of it is substantially higher for someone also taking insulin or a sulfonylurea. That is a person-specific fact, which is exactly why the labels route this to a toxicologist rather than to a table.
Why tirzepatide and not semaglutide
The gap is too large to pass over and we cannot close it honestly. Tirzepatide has roughly twice semaglutide's total reports but nearly eight times the overdose terms and twenty-five times the incorrect-dose reports.
Three explanations are available and we are naming them rather than choosing:
- Presentation. Dose-selection pens and multi-dose vials create error opportunities that a fixed single-dose autoinjector does not. This is plausible and we cannot test it from the database.
- Reporting behaviour. Manufacturer support programmes, patient-support call centres and pharmacovigilance practice differ between companies, and they shape how many minor events become reports.
- The compounded market. Compounded tirzepatide was widely sold in units and concentrations the branded pens never used, which is a well-documented source of real-world dosing error, and we cover what changed on compounded tirzepatide vs Zepbound.
The database can show the gap. It cannot explain it, and neither can we.
What this page connects to
That dosing errors dominate tirzepatide's adverse-event record is a finding this site first reported on our tirzepatide review. This page is where it is set against the labels.
For the counted adverse reactions from the labels' own placebo-controlled tables — the nausea and vomiting figures, and the discontinuation rates that tell a different story — see GLP-1 side effects. For what happens when someone comes off these medicines rather than takes too much, see stopping a GLP-1.
If a dose has been taken in error, the number on every one of these seven labels is 1-800-222-1222. If the question is about self-harm, in the US the 988 Suicide and Crisis Lifeline can be called or texted at any hour.