Is euglycaemic ketoacidosis a real concern with these drugs?

Both agents showed statistically significant disproportionality signals in the analysis described here, and the reported cases were frequently serious. That is a reason for the question to be studied properly, which is what a signal is for. It is not, by itself, a measurement of how often it happens.

Why not just divide the reports by prescriptions?

People do try, and the result is called a reporting rate. It is better than nothing but still unreliable, because the fraction of true events that get reported is unknown, varies by drug, by country, by seriousness and over time. Dividing an unknown fraction by a known denominator does not produce a rate.

Does the Weber effect always hold?

It is a described pattern, not a law, and studies since the original description have found it holds for some drugs and periods and not others - particularly where a drug's use expands sharply years after launch, as has happened across this class. It is a reason for caution about time trends rather than a correction that can be applied.

How long before real risk estimates arrive?

For a drug launched in 2022, cohort studies with adequate follow-up begin appearing around three to five years later. That gap is precisely the period in which spontaneous report counts are least interpretable and most quoted.

GLP-1 & Incretin Science

Why a Safety Report Count Climbs

JMWritten & reviewed by Jack Muncaster · Founder, UK PeptidesLast reviewed 2026-08-233 cited sources

A rising count of adverse event reports for a newly marketed drug is expected and largely uninformative about risk. Exposure grows, awareness grows, and reporting itself follows a well-documented curve that peaks a few years after launch and then declines.

Key facts

Observation: Tirzepatide ketoacidosis reports in non-diabetics
2022 rate: 1-2 cases per quarter
2025 rate: 28-34 cases per quarter
Tirzepatide US approval: May 2022 (type 2 diabetes)
Semaglutide US approvals: 2017 (diabetes), 2021 (weight management)
What is missing: Number of people exposed in each quarter
Named effect: Weber effect / notoriety bias
Correct interpretation: A prompt for a denominator-based study

The observation, and why it looks alarming

Makhmutov and Qureshi analysed FAERS reports from January 2021 to December 2025 and found that ketoacidosis reports for tirzepatide in non-diabetic patients rose from one or two per quarter in 2022 to 28 to 34 per quarter by 2025 - a fifteen- to thirty-fold increase over three years. Written as a headline that reads as an escalating safety problem. Written with its denominator it reads as nothing much at all, and the denominator is exactly what a spontaneous reporting system does not have.

Exposure is the first and largest explanation

Tirzepatide was first approved in the United States in May 2022. In the first quarter of that curve there were almost no people taking it. By 2025 there were millions. If the per-person risk were exactly constant, the raw count would still have risen by something close to the same factor as the prescription volume. A count that grows in step with uptake carries no information about whether the drug became more dangerous, and separating the two requires a denominator that FAERS cannot supply.

Reporting behaviour has its own curve

Beyond exposure, reporting rates for a given drug are not stable over time. The pattern usually called the Weber effect describes reporting for a newly marketed drug rising for roughly the first two years and then declining, independent of any change in the underlying risk. Layered on top is notoriety bias: once an association is discussed in the literature or the press, clinicians who see the event start attributing it to the drug and reporting it, which increases counts for a reason that is entirely about attention. A drug approved in 2022 and a drug approved in 2017 sit at different points on both curves at any given moment, which is one of the strongest reasons not to compare their raw counts.

What the same paper's controlled comparison shows

The disproportionality analysis in that paper is the part that carries information, precisely because it normalises against the rest of the database. Semaglutide showed a ketoacidosis reporting odds ratio of 3.15 (95% CI 2.78 to 3.56); tirzepatide showed 1.22 (95% CI 1.06 to 1.39). Both are above one and both are statistically distinguishable from it, so both drugs have a signal. But the drug whose raw count was climbing fastest has the smaller ratio - which is what you would expect if the climbing count were mostly uptake. The trend line and the ratio point in opposite directions, and only one of them is adjusted for anything.

Where the counts do carry information

Two places. The proportion of reported cases that were serious is not affected by the missing denominator in the same way, because it is a ratio within the reported cases: hospitalisation was required in 74.3% of semaglutide cases and 71.3% of tirzepatide cases in that analysis, which says something real about the severity of the cases that do get reported, whatever their frequency. And an abrupt discontinuity - counts jumping in a period when exposure did not - is a genuine signal, because the usual explanations do not fit. A smooth rise tracking a product launch is not that.

The general rule

Any raw adverse event count for a recently launched drug should be read as a measure of how much the drug is being used and discussed, with a risk signal somewhere inside it that the count alone cannot isolate. The way to isolate it is a study with people in the denominator: a cohort, a case-control design, a self-controlled case series, or a health-record analysis where exposure time is measured. Those studies are slower, which is why the fast, biased number is the one that circulates first.

Extended research context

The GLP-1 & Incretin Science deep dive

Deep dive: the two routes to a bigger effect

Every compound trying to beat GLP-1 alone has taken one of two routes. The first adds more receptors from the same hormone family — GIP in tirzepatide, GIP and glucagon in retatrutide. The second adds a non-incretin satiety hormone, which in practice means amylin: CagriSema combines cagrilintide with semaglutide, and amycretin engages both receptors from one molecule. Both routes work, because they recruit signalling pathways that do not fully overlap. Neither has escaped the constraint that binds all of them, which is that gastrointestinal tolerability worsens as effect size grows.

Deep dive: why a percentage is not a result

The most-quoted numbers in this field are the least comparable. REDEFINE 1 reported 22.7% and 20.4% for the same compound in the same trial — the first among participants who adhered to treatment, the second across everyone randomised. TRIUMPH-1 reported 28.3% in an uncomplicated obesity population while TRIUMPH-3 reported up to 22.6% in adults with established cardiovascular disease, using the same compound. Before any two figures can be compared they have to match on estimand, population, duration, comparator and whether the number is placebo-adjusted. Most published comparisons match on none of them.

Deep dive: what happens after the trial stops

Every headline figure describes weight while treatment continues. The STEP-1 extension found that a year after semaglutide was stopped, participants had given back roughly two-thirds of what they lost, moving from 17.3% mean reduction to a net 5.6% — though average weight remained below baseline and nearly half stayed at least 5% down. Meta-analysis puts regain at around 0.8 kg per month. This is why maintenance studies such as TRIUMPH-6 matter more to the field's future than another two points of peak reduction.

Research applications

▸Comparing incretin and amylin compounds on a like-for-like basis
▸Interpreting estimands, thresholds and placebo-adjusted figures in trial reports
▸Tracking the obesity pipeline across sponsors and jurisdictions
▸Understanding receptor pharmacology behind GLP-1, GIP, glucagon and amylin
▸Distinguishing licensed medicines from investigational compounds

Handling checklist

✓Identify which estimand a quoted percentage comes from before citing it
✓Check the trial population and baseline BMI against the comparison you are making
✓Confirm the duration and whether the reduction curve had plateaued
✓Read discontinuation rates alongside efficacy figures
✓Verify every NCT identifier against ClinicalTrials.gov rather than secondary reporting

Common research-handling mistakes

Learnt from thousands of researcher orders across our UK labs.

✗ Comparing headline percentages across different trials

Fix: Population, duration, estimand and comparator all differ; the numbers are not interchangeable.

✗ Quoting the larger of two figures from the same trial

Fix: Name the estimand. Efficacy and treatment-policy answer different questions.

✗ Treating peak reduction as a durable outcome

Fix: Substantial regain follows cessation across the class; peak figures describe a maintained state.

✗ Assuming an oral route means a weaker mechanism

Fix: Route and receptor count are independent. Orforglipron is weaker because it hits one receptor, not because it is a tablet.

✗ Reading investigational compounds as available treatments

Fix: Most of this pipeline holds no authorisation anywhere; mazdutide is approved only in China.

Continue researching

Peer-reviewed guides, comparators and matched reference materials.

Frequently asked questions

Is euglycaemic ketoacidosis a real concern with these drugs?: Both agents showed statistically significant disproportionality signals in the analysis described here, and the reported cases were frequently serious. That is a reason for the question to be studied properly, which is what a signal is for. It is not, by itself, a measurement of how often it happens.
Why not just divide the reports by prescriptions?: People do try, and the result is called a reporting rate. It is better than nothing but still unreliable, because the fraction of true events that get reported is unknown, varies by drug, by country, by seriousness and over time. Dividing an unknown fraction by a known denominator does not produce a rate.
Does the Weber effect always hold?: It is a described pattern, not a law, and studies since the original description have found it holds for some drugs and periods and not others - particularly where a drug's use expands sharply years after launch, as has happened across this class. It is a reason for caution about time trends rather than a correction that can be applied.
How long before real risk estimates arrive?: For a drug launched in 2022, cohort studies with adequate follow-up begin appearing around three to five years later. That gap is precisely the period in which spontaneous report counts are least interpretable and most quoted.

Primary sources & clinical trials

Peer-reviewed research and registered trials from PubMed, ClinicalTrials.gov, PubChem, FDA and NIH. All links open in a new tab and point to the primary source, so every claim can be verified at origin.

Written and reviewed by

Jack Muncaster · Founder, UK Peptides

Jack founded UK Peptides in Manchester after repeatedly receiving research compounds with missing or recycled paperwork. He is responsible for supplier selection, batch release decisions and the content published in this research library. Every article here is sourced to primary literature and every product page to a signed third-party certificate.

How we write and check this content Report a factual error in this article

More GLP-1 & Incretin Science articles

Popular across the research hub

One flagship guide from every other research category — keep exploring.

← Back to Forum