A large national study from Sweden suggests that GLP-1 receptor agonists — the class of drugs that includes semaglutide (sold as Ozempic and Wegovy) and liraglutide — may reduce the risk of worsening anxiety and depression in people taking diabetes medications. The findings, published in The Lancet Psychiatry, also link GLP-1 use to fewer psychiatric hospitalizations and less sick leave for mental health reasons, but researchers and clinicians caution that randomized trials are still needed before these drugs can be recommended as primary psychiatric treatments.
Study design and population
Researchers analyzed health records for 95,490 people in Sweden who had diagnoses of depression, anxiety, or both, and who were prescribed diabetes medications between 2009 and 2022. Using national electronic health registers that captured hospital admissions, sick leave, and deaths, investigators applied a within-individual design: each person served as their own control, comparing periods when they were taking GLP-1 receptor agonists with periods when they were not. This approach helps limit confounding by factors that don’t change over time, such as sex or baseline health.
Outcomes and comparisons
The primary composite outcome measured worsening mental illness, defined as psychiatric hospitalization, extended sick leave for psychiatric reasons, hospitalization for self-harm, or death by suicide. Secondary outcomes included worsening of depression or anxiety separately, substance use disorders, and self-harm events. The study also compared GLP-1s with other second-line diabetes drugs (for example, empagliflozin/Jardiance, dapagliflozin/Farxiga, and sitagliptin/Januvia) to see whether any benefits were independent of glucose control.
Key findings
– Average follow-up was 5.2 years; about 23.5% of participants used GLP-1 receptor agonists during the study. The most commonly used agents were semaglutide and liraglutide.
– Semaglutide use was associated with a 42% lower risk of the composite measure of worsening mental illness compared with the same individuals’ periods off GLP-1 treatment.
– Liraglutide was associated with an 18% reduced risk; exenatide and dulaglutide did not show significant associations.
– Semaglutide was linked to lower risks of worsening depression, anxiety, and substance use disorders; liraglutide was mainly associated with reduced risk of worsening depression. As a class, GLP-1 receptor agonists were associated with a lower risk of self-harm.
– Compared directly with other second-line antidiabetic medications, semaglutide still showed more favorable mental health outcomes, suggesting benefits beyond glucose lowering. Reduced risks were also reflected in fewer psychiatric hospitalizations and less psychiatric sick leave, with implications for functioning and quality of life.
Analyses and robustness checks
Statistical models adjusted for time-varying factors such as order and duration of medication use and concurrent psychiatric or antidiabetic treatments. The authors performed sensitivity analyses to address potential biases like medication sequence effects and carryover between treatment periods; results remained consistent when excluding initial days after starting or stopping medication and when focusing on use after official approval dates.
Possible mechanisms
Experts propose several mechanisms that could explain the observed associations. GLP-1 medications can cross the blood-brain barrier and interact with brain regions involved in reward and mood regulation. They may modify dopamine signaling in reward circuits, potentially reducing impulsivity and cravings that contribute to substance-related problems. GLP-1s also have anti-inflammatory effects that could lower central nervous system inflammation — a factor linked to depression and anxiety — and thereby help stabilize mood and emotional regulation.
Expert perspectives and caveats
Clinicians not involved in the research noted the potential for dual benefits in patients with overlapping metabolic and psychiatric disorders, including reduced hospitalizations and functional impairment. However, they emphasized that this was an observational study and cannot establish causality. GLP-1 receptor agonists should not yet be considered primary treatments for depression or anxiety without randomized controlled trials to confirm efficacy, optimal dosing, and safety for psychiatric indications. The findings do support the value of integrated care and point to an important area for future research at the intersection of psychiatry, addiction medicine, and metabolic health.
