Both genetic analysis and a meta-analysis of clinical studies found no causal connection between GLP-1 receptor agonists and anxiety, depression, suicidal behavior, bipolar disorder, or schizophrenia — while eating disorder symptoms actually improved.
No causal link to 7 mental disordersBoth Mendelian randomization (genetic) and meta-analysis (clinical) independently confirmed that GLP-1 receptor activation is not causally linked to anxiety, depression, suicide, bipolar disorder, schizophrenia, eating disorders, or chronic depression
What the researchers found
Using both genetic analysis (Mendelian randomization) and a meta-analysis of clinical studies, researchers found no causal link between GLP-1 receptor activation and seven mental disorders: anxiety, bipolar disorder, chronic depression, depression, eating disorders, suicide, and schizophrenia. All genetic odds ratios were non-significant after correction.
The meta-analysis of real-world clinical data confirmed these findings, showing no significant differences between GLP-1RA users and controls for suicidal ideation/behavior (OR=0.92, p=0.67), anxiety (OR=1.03, p=0.93), or depressive symptoms (SMD=-0.25, p=0.39). However, one notable exception emerged: GLP-1RA users showed significantly reduced eating disorder symptoms (SMD=-0.71, p=0.006).
Why it matters
As millions of people take GLP-1 receptor agonists like semaglutide and liraglutide for diabetes and weight loss, concerns about potential psychiatric side effects — particularly suicidal thoughts and depression — have generated significant public attention and regulatory scrutiny. This study provides robust reassurance from two independent lines of evidence (genetic and clinical) that GLP-1 receptor activation does not cause mental health disorders. The finding that eating disorder symptoms actually decreased in GLP-1RA users is a positive bonus.
The numbers in context
7 mental disorders tested · anxiety OR=0.870 · depression OR=0.985 · suicide OR=1.040 · all PFDR>0.25 (non-significant) · suicidal ideation meta-analysis OR=0.92, p=0.67 · eating disorders SMD=-0.71, p=0.006 (significant improvement)
How the study worked
The researchers employed a multi-method approach: (1) Two-sample Mendelian randomization using genetic variants as instruments to assess causal relationships between GLP-1R and mental disorders, (2) Linkage Disequilibrium Score Regression (LDSC) to evaluate genetic correlations, (3) Bayesian co-localization to validate shared genetic architecture, and (4) a meta-analysis of published clinical studies comparing GLP-1RA users to controls on mental health outcomes. Results were validated using independent datasets (eQTLGen Consortium and Psychiatric Genomics Consortium).
Who was studied
Genetic data from large-scale GWAS databases; clinical data from meta-analysis of GLP-1RA clinical studies
What this study cannot tell us
Mendelian randomization tests the effect of lifelong genetic variation in GLP-1R, which may differ from the effect of short-term drug exposure. The meta-analysis depends on the quality and reporting of included clinical studies. Some mental health outcomes may be underreported in clinical trials focused on metabolic endpoints. The eating disorders finding, while significant, had only a small number of studies and warrants further investigation. The study examined GLP-1RAs as a class rather than individual drugs.
How to read the evidence
This study earns a strong evidence grade because it combines Mendelian randomization (which approximates a natural experiment) with a meta-analysis of clinical data, validated across multiple independent genetic databases. This multi-method approach provides more robust causal inference than observational studies alone.
When this study was published
Published in 2026, this is a brand-new study directly addressing current public and regulatory concerns about GLP-1 drug safety.
The bigger picture
This study arrives at a critical moment in the GLP-1 drug story. Regulatory agencies including the FDA and EMA have been actively investigating reports of psychiatric side effects from semaglutide and liraglutide. Providing genetic-level evidence that GLP-1R activation does not cause mental disorders significantly strengthens the safety case for this drug class. The unexpected finding that eating disorder symptoms improved adds an intriguing therapeutic angle — some researchers are already exploring whether GLP-1 drugs could specifically benefit people with binge eating disorder.
Questions still open
- Could GLP-1 receptor agonists be developed as treatments for eating disorders, given the significant symptom reduction observed?
- Do individual GLP-1 drugs (semaglutide vs. liraglutide vs. tirzepatide) differ in their mental health profiles despite the class-level safety signal?
- How should regulatory agencies weigh this genetic evidence alongside individual case reports of psychiatric events in GLP-1 drug users?
Common questions
Do GLP-1 drugs like Ozempic cause depression or suicidal thoughts?
What about the reports of psychiatric side effects from GLP-1 drugs?
Read the original research
Causal association between glucagon-like peptide-1 receptor agonists and mental disorders: insight from genetic and real-world evidence.
Journal of affective disorders, 403, 121452
Citation
Ouyang, Chenhao; Zhang, Xiaoyue; Zhang, Minghai; Miao, Yan; Zhu, Zicheng; Zeng, Yunting; Ban, Hong; Wu, Yuting; Peng, Nanqin; Ling, Jitao; Li, Chen; Zhang, Deju; Yu, Peng; Zhang, Jing; Liu, Xiao; Huang, Tongsheng. (2026). Causal association between glucagon-like peptide-1 receptor agonists and mental disorders: insight from genetic and real-world evidence.. Journal of affective disorders, 403, 121452. https://doi.org/10.1016/j.jad.2026.121452