Skip to main content

Sources last read

The study

Wrong-dose reports pile up in the drugs you measure yourself

For Mounjaro and Zepbound, the thing people report most is not nausea. It is the wrong amount going in. That one term covers 18% of everything filed about Mounjaro and 18% of everything filed about Zepbound. For Ozempic, Wegovy, Saxenda, Victoza and Rybelsus it does not reach the top 25 things people report at all.

Source: FDA side-effect reports, read 2026-08-17

What matters first

Products compared
9
Mounjaro wrong-dose share
18%
Zepbound wrong-dose share
18%
Measured rows
3

What the reports show

Below are nine drugs, grouped by what a person has to do to get a dose out of the box. The bar is how much of that drug’s own report pile names a wrong dose. Every group label comes from the product’s own FDA label, not from us.

The three at the top all share one thing. Each is sold in a form where a person draws the dose out of a vial. One of them starts as a dry powder that has to be mixed first.

◆ Wrong dose given — share of each drug’s own reportsShare of that drug’s reports
9Exact values — 9 rows in the data table
Wrong dose given — share of each drug’s own reports
DrugWhat you do to get a doseShare of that drug’s reports
Zepboundsingle-dose pen, single-dose vial, and a multi-dose vial holding four 0.6 mL dosesYou measure the dose yourself18% of Zepbound reports
Mounjarosingle-dose pen, single-dose vial, and a multi-dose vial holding four 0.6 mL dosesYou measure the dose yourself18% of Mounjaro reports
Egrifta SV2 mg of dry powder in a single-dose vial, mixed with 10 mL of Sterile Water before useYou measure the dose yourself11% of Egrifta SV reports
Ozempica 3 mL pen set to one of several doses, holding four dosesYou pick the dose on a dialbelow 2.3% of Ozempic reports
Saxendaa 3 mL pen set to 0.6, 1.2, 1.8, 2.4 or 3 mgYou pick the dose on a dialbelow 1.9% of Saxenda reports
Victozaa pen set to 0.6, 1.2 or 1.8 mgYou pick the dose on a dialbelow 1.7% of Victoza reports
Trulicitysingle-dose pen at 0.75, 1.5, 3 or 4.5 mg — one strength per penIt comes ready-measured6.3% of Trulicity reports
Wegovysingle-dose pen or syringe, one strength each; a four-dose pen at 2.4 mgIt comes ready-measuredbelow 2.0% of Wegovy reports
Rybelsusa 3 mg, 7 mg or 14 mg tabletYou swallow a tabletbelow 2.1% of Rybelsus reports

Hatched bars are ceilings, not measurements. Five drugs do not carry this term in their 25 most-reported, so all that can be said is that it sits below the 25th — shown as the open bar. Read from FAERS on 2026-08-17.

  • Zepbound, Mounjaro, Egrifta SV sit at the top. Each is sold in a form a person draws from.
  • Trulicity comes ready-measured and sits in the middle, at 6.3%.
  • None of the five pen and tablet products at the bottom reaches 2.5%.

The check that makes this a finding

A gap that big could just mean the two piles of reports are different piles. Maybe one group of patients files more, or files differently. So here is the same chart for nausea, across the same drugs, from the same data.

Nausea sits between 10% and 15% for every one of the 8 GLP-1 brands. No gap. The same ruler reads the same on the control, and only splits apart on the dose term.

◆ Nausea — share of each drug’s own reportsShare of that drug’s reports
9Exact values — 9 rows in the data table
Nausea — share of each drug’s own reports
DrugWhat you do to get a doseShare of that drug’s reports
ZepboundYou measure the dose yourself11% of Zepbound reports
MounjaroYou measure the dose yourself10% of Mounjaro reports
Egrifta SVYou measure the dose yourselfbelow 2.2% of Egrifta SV reports
OzempicYou pick the dose on a dial15% of Ozempic reports
SaxendaYou pick the dose on a dial14% of Saxenda reports
VictozaYou pick the dose on a dial15% of Victoza reports
TrulicityIt comes ready-measured12% of Trulicity reports
WegovyIt comes ready-measured14% of Wegovy reports
RybelsusYou swallow a tablet14% of Rybelsus reports

Egrifta SV is not a GLP-1 and nausea is not one of its common effects, so it is not a control for it. Its bar is shown as a ceiling for the same reason as the others: the term is outside its 25 most-reported.

The drug that is not a GLP-1

Egrifta SV treats a different problem, in different patients, for a different company. It is not a weight-loss drug and it is not in the same family. It comes as 2 mg of dry powder in a single-dose vial, mixed with 10 mL of Sterile Water before use.

It is also the only product in this set with a mixing step. Its most-reported term is a missed dose. Its second is a wrong dose, at 11%. Put the four dose terms together and they cover between 29% and 34% of Egrifta SV reports.

That matters because it lands outside the GLP-1 story entirely. Whatever is going on is not about one drug family or one company.

What this cannot tell you

Start with the big one. Nobody counted the people who took these drugs and filed nothing. So none of these numbers is a rate, and none of them says how likely anything is. They are shares of a pile of reports, and that is all.

Then there is a second problem, and it is the one that could undo the whole finding.

The report piles are not alike

FDA marks some reports serious, meaning somebody died, went to hospital, or came to lasting harm. The share marked serious swings wildly across these nine drugs.

◆ How much of each drug’s report pile FDA marked seriousShare of that drug’s reports
9Exact values — 9 rows in the data table
How much of each drug’s report pile FDA marked serious
DrugShare of that drug’s reports
Zepbound8.9% of Zepbound reports
Egrifta SV21% of Egrifta SV reports
Mounjaro26% of Mounjaro reports
Trulicity26% of Trulicity reports
Wegovy45% of Wegovy reports
Victoza47% of Victoza reports
Ozempic49% of Ozempic reports
Rybelsus56% of Rybelsus reports
Saxenda57% of Saxenda reports

Serious describes the outcome somebody wrote down, not a finding that the drug caused it. Low is not good and high is not bad; a pile with more everyday reports in it simply has a smaller share marked serious.

Zepbound sits at 8.9%. Ozempic sits at 49%. That is a huge difference in what kind of report reaches FDA in the first place.

A wrong dose is usually an everyday event, not a serious one. So a drug whose pile holds more everyday reports will hold more wrong-dose reports, whatever the device does. Part of the gap on this page could be that, and this data cannot split the two apart.

The companies line up almost as neatly

There is a second pattern in the same nine rows. Every Eli Lilly drug here — Zepbound, Mounjaro and Trulicity — sits above every Novo Nordisk one. Companies run their own reporting, so that split alone could produce a gap this size.

Two things cut against it. Trulicity is a Lilly drug that comes ready-measured, and it sits at 6.3% — well under the 18% of its own stablemate. And Egrifta SV belongs to neither company, yet sits near the top.

Neither story works on its own. Trulicity and Wegovy both come ready-measured, yet they are about three times apart, so the box cannot be the whole answer. Trulicity and Mounjaro share a maker, yet they are about three times apart the other way, so the maker cannot be either.

The piles are also new

A total is a curve about uptake and attention before it is anything else. Reports naming Zepbound went from a few dozen to tens of thousands in three years.

◆ Reports naming Zepbound, by yearReports received
4Exact values — 4 rows in the data table
Reports naming Zepbound, by year
YearReports received
2023118 reports
202411,200 reports
202532,097 reports
2026part-year — the snapshot stops mid-year26,667 reports

Counts of reports FDA received in each calendar year. The last bar covers a part year only: the snapshot was taken on 2026-08-17.

Where this data agrees with the trials

None of the above means FAERS is junk. Take Vyleesi, a different drug again. Nausea covers 43% of its reports.

Its own label puts nausea at 40.0% in the trials, against 1.3% on placebo. Two very different ways of counting, landing in the same place. That is what this data looks like when nothing odd is happening.

The Vyleesi record, with its trial figures →

Take the data

Both files carry every figure on this page, the query behind each drug, and the warnings above in their header.

Quote it, check it, or argue with it. If a figure here is wrong, the query that produced it is printed below and the raw source is a public endpoint.

How to cite this table

HealthspanLedger. Dose-handling reports by product form, from FDA FAERS. Edition 2026.08 (version 4ccde919bfa3). Derived from FDA Adverse Event Reporting System (FAERS), openFDA drug event API, retrieved 2026-08-17. Licensed CC BY 4.0. https://healthspanledger.com/dosing-errors

Analysis by HealthspanLedger · edition 2026.08 · version 4ccde919bfa3 · retrieved 2026-08-17

The underlying records are us government work, public domain and carry no copyright: the publisher’s own license. This table — how we select, join, label and bound the facts — is ours, and it is published under Creative Commons Attribution 4.0 International (CC BY 4.0). Reuse it, including commercially, and credit HealthspanLedger with a link to this page.

If you draw your own doses

This page is about arithmetic, not about whether a drug is safe. Drawing a dose out of a vial asks a person to do a sum. Picking one on a dial does not, and a ready-made pen asks for nothing at all.

The sum itself is small and unforgiving, and getting it wrong is the single most-reported problem with two of the biggest drugs in the country.

2Method and quotation limitsThe reproducible query, denominator rules, ceilings, and FDA cautions

How this was worked out

Every figure comes from one saved copy of FDA’s side-effect database, pulled on 2026-08-17. FDA itself last refreshed that data on 2026-07-30. The biggest pile here holds 93,975 reports; the smallest holds 1,500 reports.

Each drug was asked for separately, by both the name FDA files it under and the name whoever filed the report typed in. Asking only one way misses whole drugs: Wegovy has no entry under the first, and Ozempic files under the name of a tablet. The exact query for Zepbound was:

patient.drug.openfda.brand_name:"ZEPBOUND" OR patient.drug.medicinalproduct:"ZEPBOUND"

Paste that into FDA’s own endpoint and the same total comes back. Every drug’s query ships in the download.

What the shares are of

Every share on this page has the same bottom half: that drug’s own total reports. So 18% means 18% of the 70,141 reports naming Zepbound. It never means 18% of the people who took it. That number is not in this data and cannot be got from it.

The four terms, and the ones shown as ceilings

The headline uses a single term. The wider family adds three more, all of them about the amount rather than the drug:

  • INCORRECT DOSE ADMINISTERED — the headline term
  • EXTRA DOSE ADMINISTERED
  • ACCIDENTAL UNDERDOSE
  • PRODUCT DOSE OMISSION ISSUE

The saved copy keeps each drug’s top 25 terms, not all of them. So a term can be missing for two reasons: nobody reported it, or it came 26th. We cannot tell which.

Printing a zero there would be a made-up number, and it would flatter this page. So a missing term is published as a ceiling instead. A term outside the top 25 has to be at or below the 25th, because the top 25 are the 25 biggest. Those rows read “below” and their bars are hatched and open-ended.

The real figures for those five drugs are somewhere under the ceilings shown. The gap on this page is therefore the smallest gap the data allows, not the largest.

Read this before quoting anything above

These are counts of reports filed with the FDA, not rates. Nobody knows how many people took the drug and did not file a report, and a report is not a finding that the drug caused what was reported.

  • There is no denominator. FAERS counts reports that were filed. It does not count the people who took the drug, so there is no bottom half to the fraction. Any figure on this page that looks like a percentage is a share of this drug’s reports — never a share of the people who took it.
  • A report is not proof the drug did it. FDA does not require a reporter to establish cause. A report says somebody noticed something and thought the drug was worth naming. Some of these events were caused by the drug, some were caused by the illness it was prescribed for, and some were coincidence — FAERS does not separate them.
  • A bigger number can just mean a more-prescribed drug. More reports can simply mean that more people use the drug, that it is newer, or that it receives more news and legal attention. Raw report totals do not tell you which drug is safer.
  • The same event can be counted twice. One event can reach FAERS through more than one channel — the patient, the prescriber and the manufacturer can all file. FDA does not guarantee that the public files are de-duplicated.
  • FDA’s own words. Do not rely on openFDA to make decisions regarding medical care. While we make every effort to ensure that data is accurate, you should assume all results are unvalidated. We may limit or otherwise restrict your access to the API in line with our Terms of Service.
  • Retrieved 2026-08-17 from https://api.fda.gov/drug/event.json · FDA data last updated 2026-07-30 · license https://open.fda.gov/license/

How everything on this site is sourced →

When this moves

These counts are a snapshot

FDA refreshes FAERS about every three months. When we re-pull it, the shares above change — and which way they move is the whole point of the study.

Give an address and you get a short note on the next pull: what changed, by how much, and the query that produced it.

We mail a confirmation link and store nothing against your address until you open it. Sourcing updates only — what changed in the data, never what to take. One click unsubscribes, from any message. What the updates cover.

12primary sources, each with the day we read it — open the documents used on this page

The pages behind the study

What the Trulicity label sets as a starting dose

The Trulicity label names no maintenance dose at all, and the page keeps that gap open.

Victoza doses, and who the label covers

The Victoza label reaches down to age 10, on its own ceiling.

How high Zepbound is allowed to go

Zepbound runs on NDA 217806, and the schedule below is that document’s.

How high Mounjaro is allowed to go

Mounjaro runs on NDA 215866, and the schedule below is that document’s.

Where the Ozempic schedule stops

Ozempic starts at 0.25 mg and stops at 2 mg, 4 steps apart.

How PT-141 differs from the rest of its class

No published human dosing schedule was found for PT-141.