People with a history of depression or anxiety who used certain GLP-1 receptor agonists had fewer signs of worsening mental health than during periods when they were not taking those medicines, according to a large Swedish register study.
The findings are encouraging but preliminary. They show an association, not proof that GLP-1 medicines improve mental health. The study also could not establish whether the results differed according to whether a medicine was prescribed for diabetes or weight loss.
What the study examined
Taipale and colleagues used Swedish national health registers to study 95,490 people with a pre-existing diagnosis of depression or anxiety who had been prescribed non-insulin antidiabetic medication. The records covered the period from 2009 to 2022.
The researchers used a within-individual design. Rather than comparing one group of people with another, they compared periods when the same person used a GLP-1 receptor agonist with periods when that person did not. This approach removes the influence of characteristics that do not change over time, such as personality and other comorbidities.
Participants were 59.7% female. Overall, 81.5% had anxiety, 54.9% had depression and 36.4% had both conditions. About 23.5% were prescribed a GLP-1 receptor agonist during follow-up.
What counted as worsening mental health
The study defined worsening mental health as a combined outcome involving psychiatric hospitalisation, psychiatric sick leave lasting more than 14 days, hospitalisation because of self-harm or death by suicide.
Use of semaglutide and liraglutide was associated with a lower risk of this combined outcome than periods without GLP-1 receptor agonist use. Semaglutide was also associated with a lower risk when compared with other second-line antidiabetic medicines.
The results varied by medicine and outcome:
- Semaglutide was associated with reduced risks of worsening depression, anxiety and substance use.
- Liraglutide was associated with a reduced risk of worsening depression.
- Liraglutide was linked with lower risk in women, but not in men, for the combined worsening mental health outcome.
- Any GLP-1 receptor agonist use was associated with a lower risk of self-harm, although the study could not reliably test individual medicines for this outcome.
- Semaglutide was associated with fewer psychiatric or self-harm hospitalisations among inpatient events.
- Semaglutide and liraglutide were associated with a lower risk of sick leave for any psychiatric reason.
Exenatide and dulaglutide were not associated with a significant reduction in the combined outcome. However, those analyses involved substantially smaller samples, so the findings may have been affected by limited statistical power.
Why the results do not prove a mental health benefit
A within-individual comparison can account for stable personal characteristics, but it cannot eliminate factors that change over time. For example, the reasons someone starts or stops a medicine may also be related to their mental health. The authors acknowledged that residual confounding could remain.
The study did not include detailed measures of clinical severity. It therefore could not show whether the results reflected changes in weight, changes in symptoms or another difference during medication use.
It was also unclear whether GLP-1 receptor agonists prescribed for diabetes had different mental health associations from those prescribed for weight loss. The study’s findings have not yet been replicated in a randomised controlled trial.
What this means for people taking GLP-1 medicines
The findings do not establish GLP-1 receptor agonists as treatments for depression, anxiety, substance use or self-harm. They also do not mean that a person should start, stop or change a prescribed medicine based on this study alone.
The source recommends that mental health symptoms be assessed and monitored during treatment, with treatment adjusted when appropriate. Anyone experiencing worsening symptoms or thoughts of self-harm should seek urgent support from a healthcare professional or local emergency service.
AI tools were used to assist with the preparation of this article.
Leave a Reply