Semaglutide & liraglutide lower risk of worsening mental illness




GLP-1 RAs are licensed to treat diabetes and obesity, but data on whether these medications alleviate or exacerbate anxiety, depression, and self-harm are mixed. A cross-institute team of researchers conducted a national cohort study using Swedish electronic health registers in 95,490 patients with depression or anxiety disorders (mean age, 50.6 years; female, 59.4 percent) who used any antidiabetic medications between the years 2009 and 2022. [Lancet Psychiatry 2026;13:327-335]
Semaglutide: The most effective GLP-1 RA
The primary composite outcome was worsening mental illness, which was defined as psychiatric hospital admissions, sick leave from work (either full-time or part-time work; sickness absence for >14 days due to psychiatric reasons), hospitalization due to self-harm, and death due to suicide.
“Semaglutide was better than any other GLP-1 RAs regarding our main outcome of worsening mental illness,” commented the researchers. “[After a mean follow-up of 5.2 years,] we found a 42 percent lower risk of worsening mental illness in patients using semaglutide [adjusted hazard ratio (HR), 0.58; 95 percent confidence interval(CI), 0.51–0.65] and an 18 percent lower risk in those using liraglutide [HR, 0.82; 95 percent CI, 0.76–0.89] vs non-use of GLP-1 RAs.” However, exenatide (HR, 1.01; 95 percent CI, 0.69–1.46) and dulaglutide (HR, 1.01; 95 percent CI, 0.85–1.20) showed no significant effect.
With regard to secondary outcomes, semaglutide was the only GLP-1 RA to be significantly associated with decreased risks of worsening depression (HR, 0.56; 95 percent CI, 0.44–0.71), worsening anxiety (HR, 0.62; 95 percent CI, 0.52–0.73), and worsening substance use disorder (HR, 0.53; 95 percent CI, 0.35–0.80). Liraglutide was associated only with a lower risk of worsening depression (HR, 0.74; 95 percent CI, 0.64–0.87). In contrast, neither exenatide nor dulaglutide reduced risks of depression or anxiety disorders.
Not a class effect
“The finding implies that this is not a class effect,” pointed out the researchers. Although semaglutide is superior to liraglutide, dulaglutide, and exenatide in terms of glycaemic control and weight loss, the study does not provide evidence that weight loss directly caused improvement in mental health,” clarified the researchers. “The relationship between GLP-1 RAs, weight loss, and mental health is likely to be complex.”
GLP-1 RAs reduce self-harm
Although a lack of statistical power precluded comparison between individual GLP-1 RA, GLP-1 RAs as a group were associated with a reduced risk of self-harm (HR, 0.56; 95 percent CI, 0.34–0.92) vs non-use of GLP-1 RAs.
The diminished risk of self-harm with GLP-1 RA use demonstrated in the study contradicts earlier concerns that GLP-1 RAs might increase the risk of suicidal behaviour. [Lancet Psychiatry 2026;13:327-335]