Women on semaglutide reported a wider range of menstrual problems to federal drug-safety monitors than women on other GLP-1 drugs, and one drug in the class produced no menstrual signal at all.
That split is the finding of a new analysis of the FDA Adverse Event Reporting System, or FAERS, published July 9 in Obstetrics & Gynecology ↗. Connor Frey and Mahyar Etminan restricted the database to female patients aged 12 to 55, the reproductive years, and looked at every report filed through March 2026. Their question was narrow: when one of these drugs appears in a report alongside a menstrual complaint, does that pairing show up more often than you would expect by chance?
How the database works
FAERS is a passive collection point. Doctors, pharmacists, drugmakers, and patients file reports when someone on a drug has a bad experience, and the reports pile up without any check on whether the drug caused the problem. It cannot give a rate, because nobody counts the far larger number of people who took the drug and had nothing to report. What it can do is flag disproportion: a drug-and-symptom pair that turns up more often than the rest of the database would predict. The authors ran that disproportionality test and a Bayesian version of it, a second method that guards against small-number flukes.
The drugs diverged
Semaglutide, the molecule in Ozempic and Wegovy, carried the broadest signal. Reports tied it to heavy menstrual bleeding, spotting between periods, menstrual clots, infrequent periods (oligomenorrhea), and cycles that skipped ovulation (anovulatory cycles). That is five distinct kinds of menstrual disruption.
Tirzepatide, sold as Mounjaro and Zepbound, flagged for two: spotting between periods and menstrual clots. Liraglutide, the older daily injection sold as Victoza and Saxenda, produced no significant signal at all.
Three drugs that hit the same receptor, GLP-1R, and three different menstrual fingerprints. The authors read that as evidence of pleiotropic effects, meaning the drugs touch menstrual physiology through more than one pathway rather than a single shared mechanism.
Read it carefully
A FAERS signal is a lead, not a verdict. The database is shaped by how much a drug is prescribed and how much it is talked about, and semaglutide is both the most prescribed GLP-1 drug and the one at the center of a running social-media conversation about "Ozempic periods." Notoriety pulls in reports. Part of semaglutide's wider signal could be that more patients, and more news stories, primed more reporting. Disproportionality analysis corrects for raw volume to a degree, but it cannot fully untangle a drug that is genuinely different from a drug that is merely famous.
Rapid weight loss on its own can shift menstrual cycles, and all three drugs drive weight loss, so a cleaner study would still have to separate the drug's direct effect from the effect of the pounds coming off. None of that is possible inside a spontaneous-report database. What the analysis does establish is that the question is worth asking in a real study, and that the answer may not be the same for every drug in the class.
The practical takeaway the authors land on is modest and sensible: menstrual health belongs in the conversation when a reproductive-aged patient starts one of these drugs.
Peptidemodel hosts cards for semaglutide ↗, tirzepatide ↗, and liraglutide ↗, all built against the GLP-1R target ↗ they share. Same receptor, and on the evidence here, not the same effect on a period.