GLP-1 Drugs and Suicidal Thoughts: FDA Findings
GLP-1 Psychology
107,910 participants reviewed
The FDA's meta-analysis of 91 clinical trials found no increased risk of suicidal ideation or behavior with GLP-1 receptor agonists. In January 2026, the FDA requested removal of the suicidality warning from drug labels.
FDA Drug Safety Communication, January 2026
FDA Drug Safety Communication, January 2026
View as imageIn July 2023, the FDA announced it was investigating reports of suicidal thoughts and self-harm behavior in patients taking GLP-1 receptor agonists. The investigation lasted over two years, examined 91 clinical trials with 107,910 participants, and reached a clear conclusion: no evidence of increased risk. On January 13, 2026, the FDA took the unusual step of requesting that manufacturers remove the suicidal behavior and ideation warning from the labels of liraglutide (Saxenda), semaglutide (Wegovy), and tirzepatide (Zepbound). For broader context on the psychological dimensions of GLP-1 weight loss, see our body image after GLP-1 weight loss pillar article.
Key Takeaways
- The FDA meta-analyzed 91 clinical trials with 107,910 participants (60,338 on GLP-1 drugs, 47,752 on placebo) and found no increased risk of suicidal ideation or behavior
- In January 2026, the FDA requested removal of the suicidal ideation warning from GLP-1 drug labels
- A real-world cohort study of 240,000+ patients found semaglutide was associated with 49-73% lower risk of suicidal ideation compared to other obesity and diabetes drugs
- FAERS pharmacovigilance data did show disproportionate reporting of suicidal events, but reporting databases cannot establish causation or incidence
- A Danish nationwide case-time-control study found no temporal association between GLP-1 RA use and suicide or suicide attempts
- The investigation began in July 2023 after post-marketing reports and took over two years to reach its conclusion
The Timeline: From Signal to Resolution
July 2023: The European Medicines Agency (EMA) announced a review of GLP-1 RA drugs following reports of suicidal thoughts and self-harm, primarily associated with liraglutide. The FDA simultaneously disclosed it was evaluating the same signal.
January 2024: The FDA's preliminary evaluation found no evidence of a causal link. The agency stated it had not found evidence that use of these medicines causes suicidal thoughts or actions, but continued its investigation.
April 2024: The EMA concluded its review, finding that available evidence does not support a causal link between GLP-1 RAs and suicidal or self-harm thoughts and behaviors.
January 13, 2026: The FDA issued a final drug safety communication requesting removal of the suicidal behavior and ideation risk information from the labeling of liraglutide (Saxenda), semaglutide (Wegovy), and tirzepatide (Zepbound). The FDA based this decision on a comprehensive meta-analysis of 91 clinical trials across GLP-1 drug development programs.
This timeline illustrates how pharmacovigilance is supposed to work: a signal is detected, investigated thoroughly, and resolved with data rather than speculation.
The FDA's Meta-Analysis: 91 Trials
The FDA's analysis was the most comprehensive examination of GLP-1 drugs and suicidality ever conducted. It pooled data from 91 clinical trials across multiple GLP-1 receptor agonist development programs, covering 107,910 participants. Of these, 60,338 received a GLP-1 drug and 47,752 received placebo.
The meta-analysis found that GLP-1 users did not have an increased risk for suicidal ideation or behavior, or other relevant psychiatric adverse events, compared with placebo. The analysis covered semaglutide, liraglutide, tirzepatide, dulaglutide, and exenatide across both diabetes and obesity indications.
The size of this analysis matters. With nearly 108,000 participants, it had statistical power to detect even small increases in rare events. That it found no signal across this volume of data is strong evidence against a class-wide effect.
Real-World Evidence: The Surprising Protective Signal
While the FDA was conducting its investigation, independent researchers published findings that told an unexpected story. A large retrospective cohort study using electronic health records of over 240,000 patients examined the association between semaglutide and suicidal ideation in real-world clinical practice.
The results, published in Nature Medicine, found that semaglutide was associated with a 49% to 73% lower risk of first-time or recurring suicidal ideation compared to other medications prescribed for obesity and type 2 diabetes. Among patients using semaglutide for weight loss, 0.11% experienced first-time suicidal ideation, compared to 0.43% with other weight loss medications. For diabetes patients, the rates were 0.13% vs. 0.36%.
The NIH highlighted these findings, noting that people taking semaglutide had lower risk of suicidal thoughts than those on other treatments for the same conditions.
This does not mean semaglutide prevents suicidal ideation. The lower risk likely reflects differences in the patient populations, confounding by indication, or the metabolic improvements (weight loss, improved glucose control) that accompany GLP-1 therapy. Obesity and diabetes are themselves associated with elevated rates of depression and suicidal ideation, and treating these conditions may indirectly improve mental health.
The Pharmacovigilance Data That Started the Concern
The FAERS (FDA Adverse Event Reporting System) database showed disproportionate reporting of suicidal events with semaglutide and liraglutide.[1] McIntyre and colleagues analyzed FAERS reports through 2023 and found elevated reporting odds ratios for suicidality with GLP-1 receptor agonists.
It is essential to understand what FAERS data can and cannot tell us. FAERS is a passive surveillance system where healthcare providers and patients voluntarily report adverse events. It captures raw reports, not incidence rates. A disproportionate signal in FAERS means that suicidal events are reported more often than expected relative to the total reports for that drug. But it cannot determine whether the drug caused the event, the rate at which it occurs, or how it compares to the background rate in the population taking the drug.
Several factors inflate FAERS signals for GLP-1 drugs specifically:
Massive prescription volume. Semaglutide and liraglutide are among the most prescribed medications in the world. More prescriptions mean more reports of every adverse event, including coincidental ones.
Media amplification. After initial reports, media coverage prompted additional reporting. Patients and clinicians who read about the concern were more likely to report any psychiatric symptom, creating a reporting cascade.
Population risk factors. Patients taking GLP-1 drugs for obesity often have higher baseline rates of depression, anxiety, and suicidal ideation. These are comorbidities of obesity itself, not drug effects.
Stimulated reporting. After the FDA announced its investigation in July 2023, reporting of psychiatric events with GLP-1 drugs increased, a well-documented phenomenon in pharmacovigilance.
The WHO VigiBase analysis (World Health Organization's global adverse event database) found similar disproportionality signals, but with the same limitations as FAERS data.
The Lancet Danish Study
A nationwide case-time-control study published in eClinicalMedicine (The Lancet) examined suicide and suicide attempts among GLP-1 receptor agonist users in Denmark using complete national registry data. This design compares each patient to themselves at different time points, eliminating confounding by patient characteristics.
The study found no temporal association between GLP-1 RA initiation and suicide or suicide attempt. Patients were not at higher risk during the period after starting GLP-1 drugs compared to before. This within-person design is particularly powerful because it controls for all time-invariant confounders, including baseline psychiatric risk.
Why the Signal Appeared: Confounding by Indication
The most likely explanation for the initial pharmacovigilance signal is confounding by indication. Obesity and type 2 diabetes are associated with elevated rates of depression (approximately 2-3 times the general population), anxiety, and suicidal ideation. Patients seeking treatment for these conditions already carry higher psychiatric risk.
When a drug is prescribed to millions of people with elevated baseline psychiatric risk, some will experience suicidal ideation during treatment. This will be reported to pharmacovigilance databases, creating a signal that reflects the population's risk rather than the drug's effect.
The FDA's controlled clinical trial data, where patients on GLP-1 drugs are compared to similar patients on placebo, is the appropriate tool for separating drug effect from background risk. The 91-trial meta-analysis did exactly this and found no difference.
What This Means for Patients and Prescribers
The FDA's January 2026 decision to request warning removal is consequential. It means:
- The strongest available evidence (91 RCTs, 107,910 participants) found no link
- Real-world evidence suggests semaglutide may be associated with lower, not higher, psychiatric risk
- Prescribers no longer need to screen specifically for suicidal ideation as a GLP-1-related risk
- Patients who experienced anxiety about taking these drugs because of the warning can be reassured
This does not mean psychiatric symptoms in patients on GLP-1 drugs should be ignored. Depression, anxiety, and suicidal ideation are common in the populations receiving these medications. Routine mental health screening is appropriate regardless of which medications patients take. The evidence simply says GLP-1 drugs do not add to that risk.
For the broader question of how GLP-1 drugs affect mood and mental health, see our articles on semaglutide and depression, GLP-1 and alcohol, and eating disorders. For the related concept of how these drugs change the psychological experience of food, see what is food noise.
The Broader Psychiatric Safety Profile
The suicidal ideation investigation is part of a broader examination of GLP-1 drugs and mental health. Several other psychiatric outcomes have been studied:
Depression. Multiple pharmacovigilance studies reported signals for depression with semaglutide. However, a 2026 Lancet Psychiatry study of Swedish national cohort data found that GLP-1 RA use in patients with pre-existing depression and anxiety was not associated with worsening mental illness. The weight loss and metabolic improvements associated with GLP-1 drugs may actually improve depressive symptoms in many patients.
Anxiety. Anxiety is commonly reported during GLP-1 dose escalation, but controlled data do not show elevated rates compared to placebo. Rapid dietary changes, body image adjustment, and GI side effects may contribute to transient anxiety that resolves with dose stabilization.
Substance use changes. Intriguingly, some research suggests GLP-1 drugs may reduce alcohol consumption and addictive behaviors by modulating reward circuitry. GLP-1 receptors are expressed in brain regions involved in reward processing, and preclinical data shows reduced alcohol self-administration with GLP-1 RA treatment. This is explored in our article on GLP-1 and alcohol.
The overall picture that emerges is that GLP-1 receptor agonists have a neutral to possibly beneficial psychiatric safety profile, though this does not eliminate the need for standard mental health monitoring in populations with high baseline psychiatric comorbidity.
Remaining Questions
Individual vulnerability. The meta-analysis examines average risk across large populations. Whether specific subgroups (patients with pre-existing psychiatric conditions, adolescents, patients on concurrent psychotropic medications) might have differential risk has not been fully characterized.
Mechanism. The biological mechanism by which GLP-1 receptor agonists could affect mood is plausible (GLP-1 receptors are expressed in brain regions involved in mood regulation), but the clinical data does not support a harmful effect. Some researchers speculate the brain effects may actually be beneficial, given the real-world data showing lower psychiatric risk.
Long-term effects. The mean clinical trial follow-up was typically 6-18 months. Whether prolonged GLP-1 receptor activation over years or decades has different psychiatric effects is unknown.
Dose effects. The higher doses used for obesity (semaglutide 2.4 mg, tirzepatide 15 mg) have not been studied as extensively for psychiatric outcomes as the lower diabetes doses. Whether higher brain GLP-1 receptor activation from obesity doses produces qualitatively different psychiatric effects is unknown but biologically plausible.
Withdrawal effects. Some patients report mood changes when discontinuing GLP-1 drugs. Whether this reflects a pharmacological withdrawal, the psychological impact of weight regain, or coincidence has not been studied systematically.
The Bottom Line
The FDA's investigation into GLP-1 drugs and suicidality is resolved. A meta-analysis of 91 clinical trials with 107,910 participants found no evidence of increased suicidal ideation or behavior. Real-world evidence from 240,000+ patients suggested semaglutide was associated with lower, not higher, psychiatric risk compared to alternative treatments. In January 2026, the FDA requested removal of the suicidality warning from GLP-1 drug labels. The initial pharmacovigilance signal was likely driven by confounding, media-amplified reporting, and the elevated baseline psychiatric risk of the population taking these drugs.