Can GLP-1 Drugs Prevent Worsening Mental Illness? Perhaps, Study Suggests

People with pre-existing depression or anxiety were less likely to have mental health worsening while taking a GLP-1 medication, observational Swedish data suggested.

Compared with periods of non-use in the same individual, periods during which people were on semaglutide (Ozempic, Wegovy; adjusted HR [aHR] 0.58, 95% CI 0.51-0.65) or liraglutide (Victoza, Saxenda; aHR 0.82, 95% CI 0.76-0.89) were associated with a lower risk of a decline in mental health over an average follow-up of 5.2-years.

Worsening mental illness was defined as a composite of hospitalization due to mental disorder or self-harm, sick leave for psychiatric reasons, or suicide, Mark Taylor, MD, of the Edinburgh Practice in Scotland, and colleagues reported in Lancet Psychiatry.

This was the first study to link GLP-1 drug use with a lower risk of worsening mental illness in the same individuals, the researchers noted.

“We were surprised by the magnitude of the associations,” Taylor told MedPage Today. “The reduction in sick leave — or health-related absences from work — was the most unexpected. As mental health is now the most common reason people are on sick leave or absent from work due to ill-health, this has major implications for public health.”

Taylor explained that beyond weight loss and improved self-esteem, better glycemic control can improve mood regulation. “It also appears that some GLP-1 receptor agonists have a central ‘brain’ effect, perhaps via dopamine reward pathways, and may be anti-inflammatory or stimulate brain recovery,” he said.

The risk of worsening depression was lower when people were taking semaglutide (aHR 0.56, 95% CI 0.44-0.71) or liraglutide (aHR 0.74, 95% CI 0.64-0.87). Semaglutide use was also tied to lower risks of worsening anxiety (aHR 0.62, 95% CI 0.52-0.73) and substance use disorder (aHR 0.53, 95% CI 0.35-0.80).

Two other GLP-1 agents — exenatide (Byetta, Bydureon) and dulaglutide (Trulicity) — did not show these associations, however. This suggests the outcomes may not be a class-wide effect, the researchers noted.

The results “cannot simply be attributable to better medication adherence through weekly injectables, unless the reduced risk is entirely due to differential weight loss between GLP-1 receptor agonist agents,” they wrote.

Of the 171 suicide deaths reported during the study, only one occurred while a patient was taking a GLP-1 drug. The findings allay prior concerns about suicidality, Taylor said, noting “there are no safety concerns from a psychiatric perspective” in the study.

The FDA investigated suicidality after its adverse event reporting database received reports of suicidal ideation in 2023, though no safety signals appeared in clinical trials. By 2024, a preliminary report from the agency cleared these drugs of causing suicidal thoughts or actions. In January 2026, the FDA requested that manufacturers remove suicidality warnings from labels of GLP-1 agents that carried it; this included products indicated for weight management, including tirzepatide (Zepbound), semaglutide, and liraglutide.

The current study used a within-individual design and included 22,480 GLP-1 drug users from national Swedish electronic health registers. All participants had diagnoses of depression or anxiety disorder and used non-insulin antidiabetic medications from 2009 to 2022.

Average age of GLP-1 drug users was 49.9 and 63.2% were female. Overall, 80% had anxiety, 49.1% had depression, and 29.1% had both. Most GLP-1 drug users (82.2%) were prescribed an antidiabetic drug for type 2 diabetes.

While semaglutide was associated with protective mental health benefits across both genders, benefits of liraglutide were limited to women. In people who tried multiple GLP-1 drugs, the risk of worsening mental illness was significantly lower with semaglutide compared with other agents (aHR 0.67, 95% CI 0.59-0.77).

The researchers acknowledged limitations inherent to the study’s observational design, including a lack of individual data on psychiatric symptom severity, weight changes, or HbA1c levels. They called for a randomized controlled trial specifically targeting patients with comorbid diabetes and mood disorders.