Published Sep 27, 2026 | 7:00 AM ⚊ Updated Sep 27, 2026 | 7:00 AM
One possibility is that semaglutide's psychiatric association is indirect.
Synopsis: A Swedish nationwide study found semaglutide was associated with a 21% lower risk of psychiatric hospitalisation among people with bipolar disorder and diabetes and/or obesity. Psychiatrists say the finding may reflect metabolic or direct brain effects, but caution that it remains observational. Randomised trials are needed before semaglutide can be considered a bipolar treatment.
A Swedish study found a 21% lower risk of psychiatric hospitalisation among semaglutide users with bipolar disorder, raising questions about whether GLP-1 drugs could have effects beyond treating diabetes and obesity.
The study, published in Acta Psychiatrica Scandinavica, included 14,694 people with bipolar disorder who were prescribed antidiabetic medication between 2009 and 2024. Of them, 5,200 used a glucagon-like peptide-1 receptor agonist (GLP-1RA), including 3,682 who used semaglutide.
Semaglutide use was associated with a 21% lower risk of psychiatric hospitalisation compared with periods when the same individuals were not taking a GLP-1RA. The drug was also associated with a 17% lower risk of hospitalisation following a bipolar relapse.
The finding does not establish that semaglutide treats bipolar disorder. The researchers used observational health-register data and said the results need to be tested in randomised controlled trials before the drug can be considered a treatment for the condition.
Also Read: Karnataka launches salt reduction plan, targets 30% cut in salt intake by 2030
Dr Girishchandra B G, Medical Director and Senior Consultant, Adult Psychiatrist at Maarga Mind Care, said the study design makes the finding more notable because researchers compared people with themselves rather than comparing separate groups.
“Instead of comparing different groups of people, researchers compared the same people during periods on the drug versus off it. That rules out a lot of things that usually complicate this kind of research like genetics, how severe the illness was to start, other health problems, social and economic circumstances because those factors stay constant when you’re comparing someone to themselves,” he told South First.
He described the 21% reduction as “a real finding” but cautioned against interpreting it as proof that semaglutide treats bipolar disorder.
“A 21% lower risk is a real finding, but a moderate one, not a dramatic one. The statistics leave some room for uncertainty,” he said.
The study’s within-person design compared periods of GLP-1RA use with periods without GLP-1RA use in the same individuals and adjusted for several time-varying factors, including other psychiatric and diabetes medications.
Dr Girishchandra said the fact that the association was seen with semaglutide, rather than consistently across all drugs in the class, also warrants further investigation.
“So this leans toward a possible direct effect, but it isn’t proof. It’s enough to justify a trial, not enough to call it a treatment,” he said.
Also Read: Final 5 minutes before smoke killed the newborn: The fiery night Adilabad RIMS lost three babies
People with bipolar disorder have higher rates of obesity and type 2 diabetes, creating an overlap between metabolic and psychiatric health. The study notes that type 2 diabetes and obesity occur more frequently among people with bipolar disorder.
One possibility is that semaglutide’s psychiatric association is indirect. Improvements in blood sugar, weight and overall physical health could potentially reduce factors that make bipolar disorder more difficult to manage.
Dr Girishchandra said better metabolic health could contribute through several pathways, including improved sleep, reduced inflammation and fewer complications from diabetes.
“But the fact that it’s specific to semaglutide suggests something more is going on,” he said. “That could be a direct effect on the brain, a difference in how strongly the drug acts, or differences in how well each drug reaches the brain.”
The study itself raises the possibility of effects beyond weight loss and metabolic control. GLP-1 receptor stimulation has been linked in previous research to changes in neuroinflammation, oxidative stress and neuroprotective processes, although the researchers stress that these mechanisms have not been established in bipolar disorder.
Also Read: Syphilis is curable, but why does Telangana record one of India’s highest death counts?
GLP-1 receptors are found not only in tissues involved in glucose regulation but also in the brain.
According to Dr Girishchandra, this provides one possible biological explanation for the findings.
“GLP-1 receptors aren’t just in the gut and pancreas, they’re also in brain areas that regulate mood, including regions tied to stress and reward,” he said.
Early research suggests that GLP-1 drugs could have neuroprotective and anti-inflammatory effects, although there is currently no definitive evidence showing that these effects prevent bipolar episodes.
“Bipolar disorder tends to come with low-grade inflammation in the body and brain, especially during mood episodes,” he said. “GLP-1 drugs have shown anti-inflammatory effects in other conditions like heart disease so it’s a reasonable guess that they’re doing something similar here, we just don’t know for sure yet.”
The researchers similarly suggest that GLP-1 drugs could potentially influence brain function directly as well as through weight loss, but say these possibilities need to be investigated in controlled trials.
The findings could be relevant in India, where diabetes and obesity are major health concerns and GLP-1 drugs are increasingly being used for weight management.
Dr Girishchandra said the overlap between metabolic disease and bipolar disorder makes the research particularly relevant to Indian clinical practice.
“India has one of the largest and fastest-growing burdens of type 2 diabetes globally and obesity is rising in urban cities, so the population overlap with bipolar disorder is substantial and clinically relevant,” he said.
He added that people with bipolar disorder can also face metabolic risks associated with some commonly used psychiatric medicines.
“If semaglutide’s psychiatric benefit is replicated in future research, it could be a genuinely useful option for Indian patients who have both conditions since many people with bipolar disorder already carry significant metabolic risk, partly from the illness itself and partly from weight-gain-prone mood stabilisers and antipsychotics,” he said.
However, the Swedish findings cannot be directly transferred to India.
The Swedish study was conducted using national health, prescription and social insurance registers in a healthcare system where treatment is broadly publicly funded. The researchers themselves caution that the findings may not be generalisable to countries with different healthcare systems.
Dr Girishchandra said affordability and access could become important confounders in an Indian study.
“In India, GLP-1 drugs are expensive and mostly paid out of pocket. So, people who get prescribed semaglutide are likely to be wealthier, with better access to healthcare, food, stable housing, and more consistent treatment, all things that independently lead to better mental health outcomes anyway,” he said.
“This means if similar data were collected in India, it would be hard to tell whether any mental health benefit came from the drug itself or just from these underlying advantages.”
Also Read: While India watched its borders for virus, mpox left a quiet trail in Kerala and Telangana
The prospect of using semaglutide in people taking antipsychotics or mood stabilisers also raises practical questions.
Dr Girishchandra said there are currently no major known interactions with lithium or antipsychotics, but clinicians should pay attention to tolerability and medication management when the drugs are combined.
“The real risk is semaglutide’s nausea and appetite loss stacking onto existing side effects from mood stabilisers, which can hurt adherence and that’s a practical issue, not a safety one,” he said.
The Swedish study did not establish semaglutide as a psychiatric treatment and did not test it specifically as an intervention for bipolar disorder. Its findings were based on periods of medication exposure in people who already had bipolar disorder and diabetes and/or obesity.
As semaglutide use expands beyond diabetes treatment into weight management, psychiatrists may increasingly encounter patients taking the drug for metabolic reasons.
Dr Girishchandra said this could make psychiatric monitoring more relevant, particularly among people with an existing or previously unrecognised history of mental illness.
“If there really is a brain effect here, that means people could see unexpected mood changes without anyone screening for it,” he said.
He cautioned that people using GLP-1 drugs without medical supervision, particularly those with a personal or family history of mental illness, should be monitored.
The authors of the Swedish study conclude that semaglutide’s potential psychiatric effects should be tested in a randomised controlled trial.
Dr Girishchandra said such a trial would need to go beyond simply measuring hospital admissions.
“We’d need a real trial, one where some patients get semaglutide and others get a placebo and neither they nor the doctors know who’s getting what,” he said.
The trial should measure mood symptoms directly, include patients with bipolar disorder both with and without diabetes or obesity, and follow participants long enough to capture relapses, he said.
“It’d also help to test different doses and to actually look at what’s happening in the body including inflammation levels, brain scans,” he said.
“And safety needs close watching especially anything to do with mood or suicidal thoughts since this is a group that can be sensitive to any kind of destabilisation.”
For now, the Swedish study adds a new question to the expanding research around GLP-1 drugs and the brain: whether the psychiatric benefits observed during semaglutide use reflect better metabolic health, a direct neurological effect, or a combination of both. The answer will require prospective clinical trials rather than observational associations alone.
(Edited by Fayisa CA)
