GLP-1s and Existing Depression or Anxiety: What the Swedish Study Actually Found

Matt Cole, host, Gila Podcast — edited by Sezen Soykut

TL;DR: A 2026 Lancet Psychiatry study followed more than 95,000 people in Sweden who already had a depression or anxiety diagnosis and used an antidiabetic medication. It found that during the periods those same people were using semaglutide, their existing mental illness was less likely to get worse — not that GLP-1s prevent depression or anxiety from developing in the first place. It's an observational study, so it can't establish cause. Here's what it actually measured, and what it found.

This article covers research on GLP-1s and mental health, including self-harm and suicide. It is not medical advice. For questions about your own medication, talk to your prescriber. If you're thinking about harming yourself, call or text 988 in the US, or contact your local crisis line.

The study, precisely

Researchers at the Karolinska Institutet, the University of Eastern Finland, and Griffith University published "Association between GLP-1 receptor agonist use and worsening mental illness in people with depression and anxiety in Sweden: a national cohort study" in The Lancet Psychiatry in 2026. A person with related lived experience was involved in the study's design and write-up.

Who was in it. The cohort included 95,490 people identified from Swedish national health registers covering 2009 to 2022. Every person in the study already had a diagnosis of depression or an anxiety disorder, and used an antidiabetic medication during that period. Of those, 22,480 used a GLP-1 medication during the follow-up period. This is not a study of the general population, and it is not a study of everyone who takes a GLP-1 — it's specifically people living with an existing mental illness who also used an antidiabetic medication.

What was measured. The main outcome was worsening of the person's existing mental illness — psychiatric hospitalization, more than 14 days of sick leave for psychiatric reasons, hospitalization for self-harm, or death by suicide. It was not new-onset depression or anxiety in people who didn't have it before, and it was not a measure of symptom severity day to day. To reduce confounding, the researchers compared periods when the same person was using a GLP-1 with periods when that same person wasn't. Reducing confounding isn't the same as removing it: for example, people may be more likely to start or stay on a medication during stretches when they're already doing better.

What it found

According to the study's published summary, during periods of semaglutide use, compared with periods when the same people used no GLP-1 medication, the risk of worsening mental illness was 42% lower (adjusted hazard ratio, or aHR, 0.58; 95% CI 0.51–0.65). When depression and anxiety were analyzed separately, semaglutide was associated with a lower risk of worsening depression (aHR 0.56; 95% CI 0.44–0.71) and of worsening anxiety (aHR 0.62; 95% CI 0.52–0.73).

Liraglutide showed a smaller association: an 18% lower risk of worsening mental illness (aHR 0.82; 95% CI 0.76–0.89). Analyzed separately, it was associated only with a lower risk of worsening depression (aHR 0.74; 95% CI 0.64–0.87), not worsening anxiety.

Exenatide (aHR 1.01; 95% CI 0.69–1.46) and dulaglutide (aHR 1.01; 95% CI 0.85–1.20) were not associated with a lower risk of worsening mental illness — a reminder that "GLP-1" is not one drug with one effect.

For self-harm, the study's summary reports a result for GLP-1 medications as a group, not for semaglutide alone: use was associated with a lower risk of self-harm (aHR 0.56; 95% CI 0.34–0.92). That confidence interval is wide, so the true size of the difference is uncertain.

What it didn't find, and can't tell you

This is the part the previous version of this page got wrong, so it's worth being direct about it.

• It did not measure whether GLP-1s prevent depression or anxiety. Every person in the cohort already had the diagnosis. The study cannot speak to whether starting a GLP-1 lowers your risk of developing depression or anxiety if you don't currently have it.
• It is observational, not a randomized trial. A within-person design like this one reduces some kinds of bias, but it cannot establish that semaglutide caused the lower rate of worsening. The authors write that randomized controlled trials evaluating these findings are warranted.
• It doesn't apply evenly across GLP-1s. The lower risk of worsening was seen with semaglutide and, to a lesser extent, liraglutide. The study did not find that association for exenatide or dulaglutide.
• Ethnicity data were not available, so the study can't show whether the findings hold equally across ethnic groups.

None of this makes the finding meaningless. It comes from a large national cohort, using a design that reduces some kinds of bias. It's just a narrower and more conditional finding than "GLP-1s reduce depression and anxiety."

How this fits with the broader safety picture

This finding sits alongside, not instead of, the regulatory record on GLP-1s and mental health, which asks a related but different question: whether GLP-1 medicines raise the risk of suicidal thoughts or behavior. In April 2024, the European Medicines Agency's safety committee concluded that the available evidence does not support a causal association between GLP-1 medicines and suicidal and self-injurious thoughts and actions. In January 2026, the FDA concluded that the studies it reviewed do not support a causal relationship, and asked the makers of Saxenda, Wegovy, and Zepbound to remove the suicidal behavior and ideation warning from those labels. Neither of those reviews is about this Swedish cohort, and neither changes what this study did and didn't measure. Our fuller guide to GLP-1s and mood covers the more mixed picture, including adverse-event reporting signals this study doesn't address.

If you're struggling right now

If you already have depression or anxiety and you're on a GLP-1, this study offers context, not a prediction: it describes patterns across many people and can't tell you how your own mood will respond. If you're having thoughts of harming yourself, please reach out now, without waiting for your next appointment: call or text 988 in the US, or contact your local crisis line. If you're in immediate danger, call your local emergency number. Tell your prescriber what you're feeling, even if you're not sure it's "serious enough" to mention.

This article is not medical advice. An observational study like this one isn't a basis for starting, stopping, or changing a medication. Those decisions belong in a conversation with your prescriber, who knows your diagnosis and treatment plan.

Matt Cole is a brand host of the Gila podcast and is not a clinician. Articles bylined to Matt are written by the Gila editorial team and edited by Sezen Soykut, Gila's founder and editor-in-chief.

Sources

1. Taipale H, Taylor M, Lähteenvuo M, Mittendorfer-Rutz E, Tanskanen A, Tiihonen J. "Association between GLP-1 receptor agonist use and worsening mental illness in people with depression and anxiety in Sweden: a national cohort study." Lancet Psychiatry. 2026;13(4):327–335. doi:10.1016/S2215-0366(26)00014-3. https://doi.org/10.1016/S2215-0366(26)00014-3 (PubMed: https://pubmed.ncbi.nlm.nih.gov/41862258/)
2. European Medicines Agency, Pharmacovigilance Risk Assessment Committee (PRAC). "Meeting highlights from the Pharmacovigilance Risk Assessment Committee (PRAC), 8-11 April 2024." https://www.ema.europa.eu/en/news/meeting-highlights-pharmacovigilance-risk-assessment-committee-prac-8-11-april-2024
3. U.S. Food and Drug Administration. "FDA Requests Removal of Suicidal Behavior and Ideation Warning from Glucagon-Like Peptide-1 Receptor Agonist (GLP-1 RA) Medications." Drug Safety Communication, January 13, 2026. https://www.fda.gov/drugs/drug-safety-communications/fda-requests-removal-suicidal-behavior-and-ideation-warning-glucagon-peptide-1-receptor-agonist-glp