Tirzepatide vs Semaglutide Side Effects

A study published 21 September 2026 in Diabetes, Obesity and Metabolism compared how tirzepatide and semaglutide show up in spontaneous adverse-event reports, the kind of database where a patient or a clinician can file a report describing a suspected drug reaction^[1]. Liu and Wang pulled 151,654 reports naming tirzepatide and 64,987 naming semaglutide, then split each set by who filed it. Healthcare professionals filed only 5.04% of the tirzepatide reports but 23.42% of the semaglutide reports, a gap that turns out to matter more than the raw totals.

What the reports show

Among reports filed by the general public, semaglutide was the only one of the two drugs to meet the study's four-method signal threshold for gastrointestinal disorders and metabolic disorders, with reporting odds ratios of 3.64 and 4.25. Pancreatitis signalled for both agents, and that signal evened out between the two drugs once the researchers looked only at people with type-2 diabetes.

The healthcare-professional reports told a different story. In that cohort, tirzepatide showed higher hypoglycaemia reporting than semaglutide, at a ratio of reporting odds ratios of 1.91, diabetic ketoacidosis was flagged only for semaglutide (reporting odds ratio 3.52), the tirzepatide gastrointestinal signal strengthened to a reporting odds ratio of 2.53, and a new eye-disorder signal appeared for semaglutide (reporting odds ratio 2.61)^[1]. The authors describe the overall picture as reporting patterns that differ by drug and by who is doing the reporting, and they call for confirmation in other datasets before anyone treats these as settled findings.

A separate FAERS analysis of tirzepatide alone, published in the same journal earlier in 2026, found fewer reports of vomiting, constipation and appetite suppression for tirzepatide than for semaglutide in its comparison group, while flagging its own new signals: eructation, hunger, food craving and starvation ketoacidosis. Medication errors and injection-site reactions were the strongest signals overall in that dataset, and the median time from starting the drug to a reported event was 13 days, with two-thirds of reports arriving within the first month^[2]. A third pharmacovigilance study covering multiple GLP-1 drugs, examining more than 260,000 reports, found semaglutide carried the strongest gastroesophageal reflux signal of the class, a result that has now turned up in more than one independent analysis^[4].

Taken together, the four datasets agree on the broad shape of the picture even when the exact numbers differ: gastrointestinal complaints dominate the general-public reports for both drugs, semaglutide's signal for gastrointestinal and reflux-related events tends to run stronger than tirzepatide's in the general population, and the clinically distinct events, hypoglycaemia, ketoacidosis, and eye disorders, only separate the two drugs once the analysis narrows to reports filed by clinicians.

What this data can and can't tell us

Spontaneous-reporting databases like this one do not record how many people actually took each drug, so there is no denominator to divide the report counts into. That means a report count, or even a signal that clears a statistical threshold, is not the same thing as an incidence rate, and it cannot establish that a drug caused the reaction described. A team writing about GLP-1 pharmacovigilance data in a different comparison put it plainly: findings from this kind of surveillance should be interpreted cautiously and confirmed with larger real-world studies before they change clinical practice^[3].

The reporter split adds another layer. Tirzepatide's overall report count is more than double semaglutide's, but healthcare professionals filed a much smaller share of them. That imbalance can shift which signals show up and how strong they look, independent of anything happening in patients' bodies. Semaglutide has also been on the market longer under more brand names and across more indications, and has accumulated a different mix of clinical experience, which affects who reports what and when. None of that is captured by a headline report count.

None of this means tirzepatide is the riskier drug for hypoglycaemia or that semaglutide is the riskier drug for ketoacidosis. It means the two drugs generate different kinds of reports from different kinds of reporters, and that difference needs a study designed to measure actual risk, not a count of spontaneous reports, before it becomes a safety conclusion either way.

Why the healthcare-professional split matters

General-public reports tend to describe symptoms a patient notices directly, like nausea or stomach upset, which is consistent with the gastrointestinal signals both drugs picked up in that cohort. Healthcare-professional reports more often describe events confirmed with a lab test or a clinical diagnosis, like documented hypoglycaemia or ketoacidosis, which is where the tirzepatide and semaglutide pictures diverged most sharply in this study. That distinction is part of why the researchers stratified by reporter type instead of pooling everyone together.

What this means for readers

If you are on either drug, gastrointestinal symptoms remain the most commonly reported issue across every dataset reviewed here, and any new or worsening symptom, especially confusion, sweating, or signs of low blood sugar, or nausea with abdominal pain that will not settle, is worth discussing with a clinician promptly rather than waiting it out. See the tirzepatide regulation pages for how each drug is classified where you live. This kind of reporting-pattern study is a starting point for further research, not a verdict on which drug is safer.