Semaglutide and Panic Attack Reports: What a New Study Found

What happened / Why it matters

What happened. A study in Pharmacoepidemiology and Drug Safety, listed in PubMed on Oct. 28, 2026, looked at reports of panic attacks with semaglutide in FDA’s adverse event reporting system. Reports were more common with semaglutide than with two comparison drugs, even after adjustment. But the authors say the pattern appears to track publicity and regulatory attention, and that a link between the drug and panic attacks “cannot be excluded.”

Why it matters. Headlines about this kind of study tend to land at one of two extremes: “semaglutide causes panic attacks” or “study debunks the panic attack scare.” The study supports neither. This article explains what the numbers show, how a regulatory notice can raise the number of reports without changing what a drug does, and what remains open. It is not medical advice.

What did the researchers do?

According to the abstract:

• FDA reports, 2018 to 2024. They compared reports in which semaglutide, dulaglutide (another GLP-1 drug) or empagliflozin (a diabetes drug from a different class) was the primary suspected drug. Panic attacks were recorded in 87 of 33,267 semaglutide reports (0.26%), 28 of 54,282 dulaglutide reports (0.05%) and 13 of 25,380 empagliflozin reports (0.05%).
• A look at timing. They checked what happened before and after July 1, 2023, when a regulatory communication about psychiatric reports with these drugs was issued. The full text ties the date to a European Medicines Agency safety review that began in July 2023, after an alert from Iceland’s medicines agency, with a parallel evaluation by FDA.
• A review of 12 other studies. They searched for randomized trials, cohort studies, other drug-safety analyses and case reports that looked at panic attacks or acute anxiety with semaglutide.

What did the reports show?

The study uses the reporting odds ratio (ROR), a measure of how much more often an event appears in reports for one drug than in reports for another. It is not a risk, and it cannot show how often an event happens.

Two things stand out. First, the signal does not disappear after adjustment over the whole period: both adjusted figures stay above 1 with intervals that exclude 1. Second, before July 2023, the adjusted difference was not statistically significant. Panic attack reports per 10,000 semaglutide reports rose from 6.7 in 2018 to 2022 to 42.0 in 2023 to 2024, and the authors report a drug-by-period interaction (p = 0.007), meaning the rise differed between drugs. According to the full text, dulaglutide reports also rose, from 3.7 to 10.9 per 10,000. When semaglutide was compared with all other drugs in the database, the adjusted figure was 1.48 (1.18 to 1.87) overall and 0.76 (0.48 to 1.18) before July 2023.

How can a regulatory notice raise the number of reports?

The authors call it “notoriety bias.” In plain terms: reports to FDA’s system come voluntarily from patients and health professionals, and from manufacturers, which must report what they receive; they are not routinely checked by a clinician. When a regulator announces it is reviewing a possible side effect, or the news covers it, patients and clinicians are more likely to notice that symptom, connect it to the drug, and file a report. The number of reports rises even if nothing about the drug’s effect has changed. Reports do not say how many people took the drug, and a rise in reports cannot be turned into a rate.

The authors describe the timing pattern as “suggestive rather than definitive evidence of notoriety bias,” according to the full text. They also note other possible reasons for the rise: more people using the drug, different kinds of patients using it and greater recognition of symptoms.

What did the trials show?

The systematic review found 12 studies, but only two measured panic attacks specifically. The others measured anxiety more broadly, which is not the same thing, according to the full text.

For randomized trials, the review found only a pooled analysis of the STEP trials in weight management. Panic attacks were reported in 3 of 2,116 people on semaglutide and 2 of 1,261 on placebo (relative risk 0.89, 95% CI 0.15 to 5.34). With five events in total, that comparison is too imprecise to show either an increase or no increase. Cohort studies that compared semaglutide with other active drugs did not find a consistent increase in anxiety outcomes.

So does the study rule out a link?

No. The authors’ interpretation, according to the full text, is that the pattern is “more readily explained by confounding by indication and notoriety-driven reporting than by a pharmacological effect, though an association with panic attacks cannot be excluded.” The abstract says the trials and active-comparator cohorts “did not show a consistent increase,” and that reports-based analyses “are hypothesis-generating and cannot estimate incidence or establish causality.”

“Confounding by indication” means the people who take a drug can differ from those who take another in ways that affect the outcome. Here that matters because none of the comparison drugs is approved for weight management: obesity was the recorded indication in 12.3% of semaglutide reports versus 0.4% and 0.1% for the comparators, per the full text. That leaves the adjustment for indication resting on few reports.

What are the limits?

• Reports, not rates. The study cannot say how often panic attacks occur in people taking semaglutide.
• Incomplete records. According to the full text, the indication was recorded in fewer than half of reports, and age was missing in about half of the semaglutide reports.
• Unverified reports. The reports were not checked clinically.
• Few trial events. The trial data include five events.
• Not a tested explanation. The publicity explanation is the authors’ interpretation of a timing pattern. It was not tested directly.

The authors declare no conflicts of interest; the funding statement reads, according to the full text, “The authors have nothing to report.”

What does this mean for a reader?

• Semaglutide is FDA-approved in specific products, for example Wegovy for weight management and Ozempic for type 2 diabetes. Each label has its own safety information.
• Anyone taking semaglutide who notices anxiety, panic or other changes in mood should tell their prescriber. Do not stop or change a medicine because of a news article.
• If you or someone you know is in crisis or having thoughts of self-harm, seek help right away: in the U.S., call or text 988 or call emergency services.
• This article gives no dosing guidance and does not link to or recommend sellers.

U.S. status of semaglutide checked Nov. 2, 2026. Based on the study abstract and the full text in PubMed Central; we did not read its supplements.