GLP-1 Medications and Depression: What 100,000 Patients Tell Us

The data on GLP-1 drugs and mental health is contradictory — some studies show improvement, others signal risk. Here's every major finding, who's most affected, and what the latest Lancet study changes.

The relationship between GLP-1 medications and depression has been debated since the European Medicines Agency opened a safety review of suicidal ideation reports in 2023. Three years later, the picture is clearer — but more nuanced than either "GLP-1s cure depression" or "GLP-1s cause depression" would suggest.

In April 2026, The Lancet Psychiatry published the largest study to date — nearly 100,000 individuals tracked over 13 years — and the findings were striking. Lancet Psychiatry 2026

The Lancet Study: 42% Fewer Psychiatric Hospitalizations

Researchers from the University of Eastern Finland, Karolinska Institutet, and Griffith University analyzed data from Swedish national health registers spanning 2009 to 2022. The study included over 20,000 people who had used GLP-1 medications, compared against periods when the same individuals were not using them. This "within-individual" design — comparing a person to themselves — is one of the strongest methods for reducing confounding.

The key findings from this study, broken down by outcome:

Critically, semaglutide was the single most effective GLP-1 in the study, followed by liraglutide. The researchers noted this is not a uniform class effect — different GLP-1 receptor agonists had different psychiatric profiles.

The Contradictory Data: Why Some Studies Show Risk

Not all data points in the same direction. A 2024 cohort study published in Scientific Reports analyzed over 162,000 matched patients and found an association between GLP-1 use and increased risk of psychiatric disorders — including a 195% higher risk of major depression diagnoses among GLP-1 users compared to non-users. Nature Sci Rep 2024

How do you reconcile a study showing 44% fewer psychiatric hospitalizations with one showing 195% more depression diagnoses? The methodological differences matter enormously:

The Lancet study used a within-individual design — comparing the same person during on-treatment vs. off-treatment periods. This controls for baseline psychiatric vulnerability. The Scientific Reports study compared GLP-1 users to non-users, meaning the groups may have differed in underlying psychiatric risk from the start. People who seek GLP-1 treatment may already have higher rates of depression linked to obesity and metabolic disease.

Additionally, the FDA-commissioned post hoc analysis of the STEP 1, 2, 3, and 5 trials (the controlled clinical trials for Wegovy) found a statistically significant but clinically negligible reduction in depressive symptoms — an estimated treatment difference of −0.56 on the PHQ-9 scale, where the minimum clinically meaningful change is typically considered 5 points. Diabetes Obes Metab 2026

Pharmacovigilance Signals: What FAERS and EMA Data Show

The global pharmacovigilance database (VigiBase) analysis published in 2025 found no significant increase in overall psychiatric adverse drug reaction reporting for GLP-1 RAs as a class. However, semaglutide specifically showed significant signals for depressed mood disorders (adjusted reporting odds ratio 1.70) and suicidality (adjusted reporting odds ratio 1.45). Clin Nutr 2025

Importantly, sensitivity analyses that restricted data to the period before June 2021 — before the media attention and EMA safety review — found no signals at all. This raises the possibility that stimulated reporting (people becoming more aware of and more likely to report psychiatric side effects after media coverage) may account for some of the post-2021 signal.

The Most Likely Explanation: It's Both

The emerging consensus among researchers is that GLP-1 medications likely improve mental health for the majority of users — through weight loss, improved metabolic function, better sleep, and possibly direct neurological effects on reward circuitry. At the same time, a smaller subset of users may experience worsened mood, particularly during the adjustment period or at higher doses.

Who May Be at Higher Risk

Based on the available data, the demographic groups that showed higher risk for psychiatric adverse events in the VigiBase analysis were: women (higher risk of anxiety and suicidal ideation), younger adults aged 18–49 (higher risk of suicidal ideation), Black patients (higher risk of suicidal ideation), and patients with pre-existing psychiatric history. Individuals with a history of depression should be monitored closely, especially during dose titration.