Protocol logistics

The most-reported term for tirzepatide products is not a symptom

By Emma Johari, Editor

8 min

Articles are summaries, researched and drafted from the sources listed on each page. How this site is written.

Regulatory sources · 10 sources

On this page
  1. The database holds more than adverse events
  2. What tops the tirzepatide lists
  3. The generic string tells a different story again
  4. What the regulatory record says alongside it
  5. Why the error terms climb the way they do
  6. Common questions

FDA's adverse event database holds three things together: adverse event reports, medication error reports and product quality complaints.

That is why a ranked list of reported terms mixes symptoms with terms about how a product was handled or used.

For the two marketed tirzepatide products, the single most-reported term describes a dosing error rather than a symptom.

None of these counts is a rate. There is no denominator, nothing is verified, and the counts rise with how many people are using a product.

This page describes what a ranked list of reported terms actually contains. It reports no amounts, no schedules and no administration instructions for any compound, and it gives no advice about anyone's own use — that requires someone who can examine you.

The database holds more than adverse events

FDA describes its adverse event system as a database containing adverse event reports, medication error reports and product quality complaints resulting in adverse events. Three different kinds of thing, coded in the same MedDRA vocabulary and returned by the same query.59

That design decision is why a ranked list of reported terms is not a list of side effects. It is a list of everything anyone coded about a product, and MedDRA has terms for circumstances of use, product handling and product quality alongside its terms for symptoms and diagnoses.59

What tops the tirzepatide lists

Ranking reported terms for MOUNJARO in the extract dated 28 April 2026 gives INCORRECT DOSE ADMINISTERED 16,692, INJECTION SITE PAIN 8,880, NAUSEA 8,656, OFF LABEL USE 7,394, DIARRHOEA 5,180 and EXTRA DOSE ADMINISTERED 4,991 — every one a count of reports mentioning the term, with no denominator and no causal assessment behind it.1

For ZEPBOUND the same ranking gives INCORRECT DOSE ADMINISTERED 11,413, NAUSEA 5,407, INJECTION SITE PAIN 4,299, EXTRA DOSE ADMINISTERED 3,993, DRUG INEFFECTIVE 3,230 and ACCIDENTAL UNDERDOSE 2,604 — again counts of reports, not counts of people or events.2

  • The top term in both lists is a medication-error term.
  • Three of the top six in each list describe product handling or administration rather than a bodily effect.
  • Drug ineffective and therapeutic response changed describe an absence of expected effect, which is not an adverse event in the ordinary sense either.
  • One report can carry many terms, so these entries overlap and do not sum to anything.12

The generic string tells a different story again

Query the generic string TIRZEPATIDE rather than either brand and the top term becomes PRODUCT TAMPERING at 408 reports, ahead of nausea at 327. Those tampering reports are concentrated in two years: 4 in 2023, 175 in 2024 and 229 in 2025.34

A term that appears in one query and not in another, for the same molecule, is a property of which reports used which product name. Reports filed under a generic name for a molecule with two marketed brands are an unusual subset by definition, and the terms in that subset reflect whatever drove those particular filings.36

What the regulatory record says alongside it

The medication-error material is not confined to the database. FDA has published an alert stating it received reports of adverse events, some requiring hospitalisation, that may relate to overdoses from dosing errors with compounded injectable semaglutide products, and that errors arose both from people measuring and administering the product themselves and from health care providers miscalculating.7

FDA has separately published on unapproved GLP-1 products used for weight loss, describing products the agency has not evaluated for safety, effectiveness or quality. Those two documents are regulatory statements, produced by review; they are a different kind of evidence from a term count, and they are the stronger kind.8

Why the error terms climb the way they do

The Mounjaro dosing-error term over time runs 1,206 (2022), 5,692 (2023), 4,481 (2024) and 4,605 (2025) — a curve that begins in the product's launch year and then flattens, matching the shape of its overall report series rather than tracking anything about the product changing.10

Reporting is also structurally sensitive to who is filing. Manufacturers forward what reaches them, and a manufacturer running patient-support and pen-training programmes hears about device and administration problems through channels that do not exist for symptoms managed at home. The counts reflect that plumbing, and the plumbing is not visible in the number.6

Tirzepatide is the clearest example on the site of a ranked FAERS list being topped by a handling term rather than a symptom, and it holds for both marketed products.

Common questions

Does incorrect dose administered at the top of the list mean the product is hard to use correctly?

The count does not support that conclusion or its opposite. It records that reports mentioning the term were filed, in a system with no denominator, at a time when the product's user base was growing quickly.

Why do symptoms and handling problems appear in the same ranked list?

Because FDA's system holds adverse event reports, medication error reports and product quality complaints together, all coded in the same dictionary. The query cannot separate them.

What is product tampering doing in an adverse-event database?

It is a coded term like any other, and product quality complaints are part of what the system collects. Its presence records what reporters described; it establishes nothing about any particular product.

Sources

  1. Regulatory
    openFDA drug adverse event endpoint — query: search=patient.drug.medicinalproduct:"MOUNJARO"&count=patient.reaction.reactionmeddrapt.exactU.S. Food and Drug Administration, openFDA, 2026api.fda.gov/drug/event.json?search=patient.drug.medicinalproduct ↗↩ Back to text
  2. Regulatory
    openFDA drug adverse event endpoint — query: search=patient.drug.medicinalproduct:"ZEPBOUND"&count=patient.reaction.reactionmeddrapt.exactU.S. Food and Drug Administration, openFDA, 2026api.fda.gov/drug/event.json?search=patient.drug.medicinalproduct ↗↩ Back to text
  3. Regulatory
    openFDA drug adverse event endpoint — query: search=patient.drug.medicinalproduct:"TIRZEPATIDE"&count=patient.reaction.reactionmeddrapt.exactU.S. Food and Drug Administration, openFDA, 2026api.fda.gov/drug/event.json?search=patient.drug.medicinalproduct ↗↩ Back to text
  4. Regulatory
    openFDA drug adverse event endpoint — query: search=patient.drug.medicinalproduct:"TIRZEPATIDE" AND patient.reaction.reactionmeddrapt.exact:"PRODUCT TAMPERING"&count=receivedateU.S. Food and Drug Administration, openFDA, 2026api.fda.gov/drug/event.json?search=patient.drug.medicinalproduct ↗↩ Back to text
  5. Regulatory
    FDA Adverse Event Reporting System (FAERS) — what the database containsU.S. Food and Drug Administration, 2026www.fda.gov/drugs/surveillance/fda-adverse-event-reporting-syste ↗↩ Back to text
  6. Regulatory
    FDA Adverse Event Monitoring System (AEMS) Public Dashboard [formerly FAERS] — limitations to the dataU.S. Food and Drug Administration, 2026www.fda.gov/drugs/questions-and-answers-fdas-adverse-event-repor ↗↩ Back to text
  7. Regulatory
    FDA alerts health care providers, compounders and patients of dosing errors associated with compounded injectable semaglutide productsU.S. Food and Drug Administration, 2026www.fda.gov/drugs/human-drug-compounding/fda-alerts-health-care- ↗↩ Back to text
  8. Regulatory
    FDA's Concerns with Unapproved GLP-1 Drugs Used for Weight LossU.S. Food and Drug Administration, 2026www.fda.gov/drugs/drug-alerts-and-statements/fdas-concerns-unapp ↗↩ Back to text
  9. Secondary source
    MedDRA: the basics — hierarchy and preferred termsMedical Dictionary for Regulatory Activities (MedDRA) MSSO, 2026www.meddra.org/how-to-use/basics ↗↩ Back to text
  10. Regulatory
    openFDA drug adverse event endpoint — query: search=patient.drug.medicinalproduct:"MOUNJARO" AND patient.reaction.reactionmeddrapt.exact:"INCORRECT DOSE ADMINISTERED"&count=receivedateU.S. Food and Drug Administration, openFDA, 2026api.fda.gov/drug/event.json?search=patient.drug.medicinalproduct ↗↩ Back to text

Keep reading

Peptide basics

How to read an adverse-event database without drawing the wrong conclusion

The same molecule returns 6,618 reports under one product name and 57,104 under another. That single fact rules out most of what people do with these numbers.Reviewed 9 minRegulatory sources · 12 sourcesSemaglutideTirzepatidePeptide basics

Tirzepatide adverse-event reports by year, and what the curve measures

Zepbound went from 37 reports in 2023 to 10,547 in 2024. It also launched in November 2023. Those two facts are the same fact.Reviewed 8 minRegulatory sources · 10 sourcesTirzepatidePeptide basics

"Off label use" is one of the most-reported FAERS terms. It is not an injury.

It ranks fourth for semaglutide and fourth for Mounjaro. It describes how a product was used, not something that happened to a body.Reviewed 7 minRegulatory sources · 10 sourcesSemaglutideTirzepatide