# Semaglutide and Suicidal Ideation: The Large Matched Cohort Studies | This GLP What the large matched cohorts found on GLP-1 drugs and suicidal ideation, why every one of them is observational, and where the systematic review still disagrees. Publisher: This GLP Editorial Team Canonical: https://thisglp.com/research/glp1-and-suicidal-ideation Updated: 2026-09-18 Published: 2026-09-16 [← Research](https://thisglp.com/research) Safety # Semaglutide and Suicidal Ideation: The Large Matched Cohort Studies Regulators opened investigations after reports of suicidal thoughts. Four matched cohorts published since — covering millions of patients — have all pointed the other way, and none of them is a randomized trial. [This GLP Editorial Team](https://thisglp.com/about)September 20267 min readUpdated September 2026 Hazard ratios for suicidal ideation, four cohorts · 1.0 · no difference · 0.27 · 0.44 · 0.49 · 0.75 · All four are observational. None is a randomized trial. **Research, not medical advice.** This GLP is independent and educational. Nothing here is a substitute for individualized care from a licensed clinician, and reading it creates no doctor–patient relationship. Talk to your own provider before starting, stopping, or changing any medication, peptide, or therapy. See our [methodology](https://thisglp.com/methodology) and [terms](https://thisglp.com/terms). ## Key takeaways - European regulators opened an investigation after reports of suicidal thoughts on these drugs. The large matched cohorts published since have pointed the other way. - In 240,618 patients with overweight or obesity, semaglutide was associated with a hazard ratio of 0.27 for incident suicidal ideation and 0.44 for recurrent ideation against other anti-obesity drugs. - Against earlier GLP-1 drugs, semaglutide showed a hazard ratio of 0.49 for suicidal ideation and 0.81 for depression. - Tirzepatide and semaglutide looked the same as each other over two years. - Every one of these is observational. A systematic review of 37 studies still calls the psychiatric picture preliminary and mixed. This is the safety question with the widest gap between what a reader has heard and what has been measured. The reports came first, the regulatory investigations followed, and the large studies arrived afterwards and mostly disagreed with the premise. That sequence is worth understanding before reading any single number below — the same three stages ran for [sudden vision loss](https://thisglp.com/research/does-semaglutide-cause-vision-loss), and knowing which stage a question has reached tells you how much weight its headline deserves. ## The cohort that answered the regulators A study of electronic health records assembled 240,618 patients with overweight or obesity who were prescribed either semaglutide or a non-GLP-1 anti-obesity medication, matched them on propensity score, and followed them for six months [\[1\]](https://thisglp.com/research/glp1-and-suicidal-ideation#ref-1). Incident suicidal ideation carried a hazard ratio of 0.27 on semaglutide, and recurrent ideation 0.44 (95% CI 0.32 to 0.60). The finding held across sex, age and ethnicity, and replicated in a separate population of 1,589,855 patients with type 2 diabetes. A hazard ratio below 1.0 means the event happened less often in the semaglutide group. The authors’ own conclusion is phrased negatively — the findings do not support a higher risk — which is the correct strength of claim for this design. ## Semaglutide against the rest of its own class A 2026 analysis of a global records network covering more than 192 million patients ran two matched comparisons [\[2\]](https://thisglp.com/research/glp1-and-suicidal-ideation#ref-2). Semaglutide against earlier GLP-1 receptor agonists, across 80,115 matched pairs, showed lower first-year rates of depression (HR 0.811, 95% CI 0.770 to 0.855), anxiety (HR 0.915, 95% CI 0.871 to 0.961) and suicidal ideation (HR 0.488, 95% CI 0.339 to 0.702). Tirzepatide against semaglutide, across 85,546 matched pairs, showed no meaningful difference on the composite psychiatric outcome in either year one (HR 0.984, 95% CI 0.950 to 1.019) or year two (HR 1.002, 95% CI 0.960 to 1.046). A nominally higher anxiety hazard appeared for tirzepatide in year two (HR 1.052, 95% CI 1.001 to 1.106), and the authors themselves flag it as fragile given the number of comparisons made. ## People who were already at risk A 2026 target-trial emulation in 606,434 US veterans with type 2 diabetes looked specifically at patients who already had a substance use disorder [\[3\]](https://thisglp.com/research/glp1-and-suicidal-ideation#ref-3). Among them, starting a GLP-1 receptor agonist rather than an SGLT-2 inhibitor was associated with a lower rate of suicidal ideation or attempt (HR 0.75, 95% CI 0.67 to 0.83), a difference of roughly 10 fewer events per 1,000 people over three years. What the systematic review says, which is less tidy A 2026 systematic review screened 342 papers and included 37 [\[4\]](https://thisglp.com/research/glp1-and-suicidal-ideation#ref-4). Its summary is that semaglutide is effective and well tolerated across psychiatric populations, with possible benefit in binge eating disorder and alcohol use disorder — and that it is also linked to depressive symptoms and suicidal ideation, alongside possible antidepressant effects, with the evidence described as preliminary and its use in depression still controversial. Four cohorts pointing one way and a review calling the picture unsettled are not a contradiction. The cohorts measure average rates across very large groups. The review is describing what happens to individuals, which averages are poor at showing. ## What to do with this None of these studies randomized anybody to semaglutide to find out. People who get prescribed one drug differ from people who get prescribed another, and propensity matching reduces that problem without eliminating it. The honest summary is that the signal which prompted the investigations has not reproduced in large matched populations, and that this is not the same as proof it does not exist for any individual. Practically, this is a reason to want a prescriber you can reach. Most of the sellers tracked here publish nothing about who reviews a case or how often, which is the silence counted in [what sellers will not tell you](https://thisglp.com/research/what-glp1-sellers-do-not-publish) and enumerated in [the questions they leave open](https://thisglp.com/tools/questions-they-leave-open). Anyone with a psychiatric history should be raising it with a clinician who can act on it rather than an intake form, and anyone deciding whether to stop should read [what happens after stopping](https://thisglp.com/research/stopping-glp1-and-weight-regain) first. The related question of whether these drugs affect drinking is covered in [semaglutide and alcohol use](https://thisglp.com/research/does-semaglutide-reduce-alcohol-use). ## Frequently asked Do GLP-1 drugs cause suicidal thoughts? The large matched cohorts published since the regulatory investigations found lower rates on semaglutide, not higher — a hazard ratio of 0.27 for incident ideation in one study of 240,618 patients. All of them are observational rather than randomized. Why did regulators investigate then? Reports of suicidal thoughts during treatment prompted European agencies to look. A reporting signal is a reason to investigate rather than a finding, and the studies run since have not reproduced it at population scale. Is tirzepatide different from semaglutide here? Across 85,546 matched pairs the two looked the same on the composite psychiatric outcome over two years. A slightly higher anxiety hazard appeared for tirzepatide in year two, which the authors describe as fragile. Does any of this settle the question? No. A 2026 systematic review of 37 studies still describes the psychiatric evidence as preliminary and mixed, and reports links in both directions. Anyone with a psychiatric history should be discussing it with a clinician who can act on it. ## Sources 1. \[1\] Wang W, et al. (2024). Association of semaglutide with risk of suicidal ideation in a real-world cohort Nature Medicine. [PMID 38182782](https://pubmed.ncbi.nlm.nih.gov/38182782/) 2. \[2\] Chen SC, et al. (2026). Neuropsychiatric Outcomes With Tirzepatide, Semaglutide, and Other GLP-1 Receptor Agonists Diabetes, Obesity and Metabolism. [PMID 42420795](https://pubmed.ncbi.nlm.nih.gov/42420795/) 3. \[3\] Cai M, et al. (2026). Glucagon-like peptide-1 receptor agonists and risk of substance use disorders among US veterans with type 2 diabetes: cohort study BMJ. [PMID 41781010](https://pubmed.ncbi.nlm.nih.gov/41781010/) 4. \[4\] Carminati M, et al. (2026). Glucagon-like peptide-1 receptor agonist semaglutide through the lens of psychiatry: a systematic review of potential benefits and risks International Clinical Psychopharmacology. [PMID 40577093](https://pubmed.ncbi.nlm.nih.gov/40577093/) ## More in Safety - [ Medication list · Check combinations · Fluid loss · Changes kidney risk · Pharmacist · Choose a safe option ### Can You Take Ibuprofen With Ozempic? Check Kidney and Stomach Risks ](https://thisglp.com/research/can-you-take-ibuprofen-with-ozempic) - [ 1.9% · Placebo · 8.5% · Ozempic 0.5 mg · 8.8% · Ozempic 1 mg ### Does Ozempic Cause Diarrhea? Rates, Timing, and When to Call ](https://thisglp.com/research/does-ozempic-cause-diarrhea) - [ Sit down · Prevent a fall · Check context · Glucose and fluids · Get help · Severe symptoms ### Does Ozempic Cause Dizziness? What to Check First ](https://thisglp.com/research/does-ozempic-cause-dizziness)