GLP-1 Medications and Mental Health: What the Latest Research Actually Shows

If you take a GLP-1 medication like semaglutide (Ozempic, Wegovy) or tirzepatide (Mounjaro, Zepbound), you have probably seen conflicting headlines about what these drugs do to mood. In 2023 and 2024, reports of suicidal thoughts triggered regulatory reviews on two continents. By 2026, the picture had changed considerably, and in a direction most people did not expect.

Here is where the evidence actually stands, and what it means if you are managing depression, anxiety, bipolar disorder, or another psychiatric condition alongside a GLP-1.

Where the concern started

The original signal came from spontaneous adverse event reports, not from clinical trials. Patients and clinicians submitted reports of suicidal ideation to the FDA's adverse event reporting system, and European regulators received similar reports. Because these systems collect voluntary reports without a comparison group, they are good at flagging possible problems and poor at confirming them.

The FDA reviewed those reports alongside clinical trial data, observational studies, a cross-product meta-analysis, and its own postmarketing surveillance data. In January 2024, the agency reported that its preliminary evaluation "has not found evidence that use of these medicines causes suicidal thoughts or actions," while noting that the small number of events meant a small risk could not be entirely ruled out.

The European Medicines Agency reached the same conclusion three months later. In April 2024, its safety committee closed a formal review of dulaglutide, exenatide, liraglutide, lixisenatide, and semaglutide, stating that the available evidence "does not support a causal association" between GLP-1 receptor agonists and suicidal or self-injurious thoughts and actions.

What changed in 2026

In January 2026, the FDA went further and requested removal of the suicidal behavior and ideation warning from the labels of Saxenda (liraglutide), Wegovy (semaglutide), and Zepbound (tirzepatide).

The evidence behind that request was substantial. It included a meta-analysis of 91 placebo-controlled trials covering 107,910 patients, and a retrospective cohort study of more than 2.2 million patients comparing GLP-1 initiators against patients starting a different class of diabetes medication. The FDA concluded that it "did not find an increased risk" for suicidal ideation or behavior, or for anxiety, depression, irritability, or psychosis, and that "the totality of these studies does not support a causal relationship."

The study that surprised psychiatrists

In 2026, The Lancet Psychiatry published a Swedish national register study of 95,490 people who already had depression or anxiety and were taking medication for diabetes. Researchers used a within-individual design, which compares each person to themselves during periods on and off the medication, reducing the influence of differences between patients.

They did not find harm. They found the opposite direction of association. Compared with periods of non-use, semaglutide use was associated with a lower risk of psychiatric deterioration (adjusted hazard ratio 0.58, 95% CI 0.51 to 0.65), as was liraglutide to a smaller degree (0.82, 0.76 to 0.89). Exenatide and dulaglutide showed no association either way. Across the class, hospitalization for self-harm was also lower (0.56, 0.34 to 0.92).

The researchers were careful about what this means. Their stated conclusion was that these medications "might be useful dually effective therapeutic options" for anxiety and depression co-occurring with diabetes and obesity, and that randomized trials are warranted. Independent commentators noted the study is observational, drawn from a single country, and cannot rule out residual confounding.

What this does not mean

No GLP-1 medication is approved to treat depression or anxiety, and no randomized controlled trial has tested one for that purpose. A separate 2026 meta-analysis of randomized trials found no statistically significant association between GLP-1 receptor agonists and any psychiatric outcome in either direction, including depression, suicidal ideation, anxiety, and sleep disorders.

The reasonable reading is that these medications appear psychiatrically safe for most people, and that the association with better mental health outcomes in observational data is promising but unproven. A GLP-1 is not a substitute for psychiatric treatment.

Practical considerations if you take psychiatric medication

Lithium requires closer monitoring

This one is worth flagging clearly. Since 2025, several published case reports have described elevated lithium levels or lithium toxicity after a patient started a GLP-1 or changed the dose. In June 2026, the European Medicines Agency opened a formal safety signal investigation into a potential interaction between lithium and GLP-1 agonists leading to increased lithium levels, and requested supplementary information from manufacturers.

Proposed mechanisms include delayed gastric emptying changing absorption, transient changes in sodium and fluid handling, and reduced clearance after significant weight loss. None of these fully explains the reports on its own, and causation is not established. If you take lithium and are starting or adjusting a GLP-1, tell both prescribers and expect more frequent level checks.

Oral medication absorption may shift

GLP-1 medications slow gastric emptying, which is part of how they work. Pharmacokinetic modeling has predicted meaningfully higher drug exposure and delayed peak concentrations for several oral medications. That modeling tested cardiometabolic drugs rather than psychiatric ones, so applying it to antidepressants or antipsychotics is inference rather than direct evidence. Still, if you notice new side effects or a change in how a medication feels after starting a GLP-1, that is worth reporting rather than waiting out.

There is real benefit for antipsychotic-associated weight gain

For patients on antipsychotics, weight gain is one of the most common reasons treatment gets abandoned. A 2026 meta-analysis of three randomized trials in schizophrenia spectrum disorders found semaglutide produced a mean weight reduction of 11.32 kg and a BMI reduction of 3.58, with improvements in HbA1c and fasting glucose, and no consistent evidence of worsening psychosis, cognition, or functioning. Gastrointestinal side effects were more common, as expected. A separate real-world inpatient cohort found roughly 12 kg of weight loss at 12 months.

Eating disorder screening matters

Clinicians who work in eating disorders have raised a reasonable concern: appetite suppression in someone with a history of restriction, bingeing, or purging can reinforce disordered patterns. The recommendation in the literature is ongoing screening and monitoring rather than a single question at the first visit.

The alcohol use disorder research

A related line of research has moved quickly. A 2025 phase 2 randomized trial in JAMA Psychiatry found that low-dose semaglutide reduced laboratory alcohol self-administration, drinks per drinking day, and weekly craving in 48 adults with alcohol use disorder. A larger 26-week randomized trial published in The Lancet in 2026 enrolled 108 adults with alcohol use disorder and obesity, and found heavy drinking days fell 41.1 percentage points with semaglutide plus cognitive behavioral therapy versus 26.4 points with placebo plus therapy.

This is not an approved use, and both trials are relatively small. But it is one of the more interesting findings in addiction medicine in years.

What to bring up with your psychiatric provider

If you are on a GLP-1 or considering one, make sure your psychiatric provider knows. Specific things worth discussing: your full medication list, including anything prescribed by a weight management clinic or telehealth service; any history of an eating disorder; whether you take lithium or any medication with a narrow therapeutic window; and any change in mood, sleep, or energy after starting or increasing the dose.

Coordination between prescribers matters more here than with most medications, largely because GLP-1s are frequently prescribed outside a patient's usual care team.

The bottom line

The strongest available evidence does not support the idea that GLP-1 medications cause suicidal thoughts, depression, or anxiety, and regulators on both sides of the Atlantic have now said so explicitly. The observational finding that these medications may be associated with better psychiatric outcomes is real but not yet confirmed by randomized trials. The interaction concerns that deserve attention are narrower and more specific, particularly lithium monitoring and eating disorder history.

If you have questions about how a GLP-1 fits with your psychiatric care, that is a conversation worth having rather than a decision to make alone.

This article is for general education and is not medical advice. Do not start, stop, or change any medication without talking to your prescriber.

If you are managing a psychiatric condition alongside a GLP-1 and want a provider who will look at the whole medication picture, Birmingham Psychiatry and Behavioral Health is accepting new patients. Reach us at appointments@birminghampsych.com or (205) 201-0658.

BPBH

Empowering adults with compassionate mental health care.

Connect with us!

appointments@birminghampsych.com

(205) 201-0658

© 2025. All rights reserved.