Neuropsychiatric Safety of GLP-1 Receptor Agonists and Tirzepatide in Adults With Type 2 Diabetes Mellitus or Obesity
Het Tushar Jani, Helisse Levine, Abhiraj Patel, Richa Trivedi, and Xitij Dineshkumar Patel
AbstractObjective: Neuropsychiatric safety concerns about glucagon-like peptide-1 receptor agonists (GLP-1 RAs) and tirzepatide prompted regulatory review by the European Medicines Agency (EMA) and the US Food and Drug Administration (FDA). We synthesized the contemporary evidence in adults with type 2 diabetes mellitus or obesity, emphasizing study design, risk of bias, and the distinction between signal detection and causal inference.
Methods: The PubMed/MEDLINE, Embase, Web of Science, and Scopus databases were searched without a lower date limit through March 15, 2026 (English, adults ≥18 years). The review was reported with reference to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) 2020, and the PRISMA-S and PRISMA-Harms extensions. Evidence was grouped into prespecified categories: randomized syntheses, comparative observational studies and target-trial emulations, outlier non-active-comparator studies, pharmacovigilance analyses, and regulatory reviews. Risk of bias was appraised with the Cochrane Risk of Bias 2 tool and with Risk of Bias in Non-randomised Studies of Interventions; certainty was summarized using Grading of Recommendations Assessment, Development and Evaluation.
Results: Twenty studies and two regulatory reviews met the inclusion criteria. Randomized evidence, including two large 2025 meta-analyses and post hoc pooled analyses of semaglutide and tirzepatide, showed no increased risk of suicidality, depression, or anxiety, though psychiatric exclusion criteria limited generalizability. Active-comparator cohorts from the United States, Scandinavia, and the United Kingdom likewise showed no clear increase after confounding control. Passive-comparator and disproportionality analyses generated signals but could not separate drug effects from baseline psychiatric risk, obesity-related distress, or reporting bias. The EMA and FDA conclusions do not support a causal class-wide association between GLP-1 RAs and suicidal ideation or behavior.
Conclusions: Evidence does not support a clinically meaningful increase in neuropsychiatric harm with GLP-1 RAs or tirzepatide in the populations studied. Surveillance remains warranted for patients with substantial psychiatric morbidity, for obesity-only populations, and for long-term rare outcomes.
Rambam Maimonides Med J 2026;17(4):e0024