Reviews/GLP-1 Receptor Agonists/No increase in suicidality
Safety · GLP1-SAF-03 · v2
Across the strongest studies available, GLP-1 drugs do not appear to raise the risk of suicidal thoughts or behaviour. A pooled analysis of 91 placebo-controlled trials covering 107,910 people found the same event rate on drug and placebo [24], and large national-register studies tracking real patients found no rise, or a fall, in suicide, self-harm and suicidal thinking [6][22][14][11][19]. Trials looking specifically at semaglutide and tirzepatide also found no increase versus placebo [9][25]. The main reasons for caution are a large real-world study that reported a doubling of suicidal behaviour [3], and several reporting databases that flag suicidal thoughts specifically for semaglutide [5][12][16][18]. Because suicide is rare, even the reassuring studies cannot rule out a small increase, and the strongest trials left out people with serious existing mental illness [9][25].
The most direct test, a subject-level pooled analysis of 91 placebo-controlled trials (N=107,910), found a rate ratio of 1.0 (0.7, 1.5) for suicidal ideation/behaviour [24]. Independent active-comparator cohorts across Swedish, Danish, Spanish, French and US data were null or protective: suicide death HR 1.25 (0.83-1.88) with the suicide-and-self-harm composite at 0.83 (0.70-0.97) [6], self-harm HR 0.93 and 0.94 [22], SIS HR 1.04 (0.35, 3.14) [11], suicide or attempt OR 0.62 (0.51-0.75) [14], and reduced self-harm 0.56 (0.34-0.92) [19]. Dedicated post-hoc RCT analyses of semaglutide [9] and tirzepatide [25] reported suicidal ideation/behaviour in 1% or fewer of participants with no difference from placebo. Against this, one propensity-matched cohort of 162,253 patients reported a 106% elevated risk of suicidal behaviour [3], and pharmacovigilance databases repeatedly flag a semaglutide-specific suicidal-ideation signal [5][12][16][18][30]. The contested point: The evidence divides on whether a semaglutide-specific suicidal-ideation signal in spontaneous-reporting databases [5][12][16][18][30] reflects a real drug-level effect that a class-wide null averages away, and on how to weigh one large cohort reporting a 106% increase in suicidal behaviour [3] against the many null cohorts and the pooled trial data [24][6][22]. Whether the reporting signals are causal or a product of notoriety bias after the 2023 regulatory reviews is left unresolved by the corpus [2][10].
All 30 papers behind this claim with their DOIs and what each does to it, the panel’s arbitration, the evidence gap map and the open research questions. Free, and sent to your inbox.
We’ll also send you other claims from GLP-1 Receptor Agonists, one at a time, and tell you when a release or a pivotal trial changes a grade. Unsubscribe by replying.
One reviewer's only job is to attack the finding. The challenges below were weighed before the grade was set.
A large propensity-matched cohort [3] (N=162,253) reports a 106% increased risk of suicidal behaviour — primary cohort data, not a reporting artefact — directly contradicting a class-wide 'do not increase' statement and cannot be dismissed as notoriety bias.
Six independent pharmacovigilance analyses [5][12][16][18][30] detect a semaglutide-specific suicidal-ideation signal; the claim explicitly covers 'suicidal ideation,' so a class-wide null may be averaging over a real drug-specific effect.
Where it remains uncertain. One propensity-matched cohort of 162,253 patients reported a 106% elevated risk of suicidal behaviour [3], and several pharmacovigilance analyses flag a semaglutide-specific suicidal-ideation signal [5][12][16][18].
Aleatory
Suicide and completed suicide are rare, so estimates rest on few events: 77 versus 71 suicide deaths [6] and a liraglutide completed-suicide signal on only 14 cases [21], leaving wide confidence intervals around any true effect.
Epistemic
Whether the suicidal-ideation signal is a drug-specific property of semaglutide [5][12][16][18][30] or absent at class level [24][7] is not resolved; a class-wide null may mask a compound-level effect.
Inferential
Cohort estimates depend on propensity matching and active comparators to remove confounding by indication and psychiatric history; residual confounding could explain either the protective [1][19] or the discordant [3] results.
The load-bearing papers from a corpus of 30, each classified by what it does to the claim. Every DOI is taken from the corpus the claim was graded on.
Represents the semaglutide-specific disproportionality signal that qualifies a class-wide null.
“Significant disproportionality was detected only for semaglutide-associated suicidal ideation (ROR, 1.45; 95% CI, 1.18-1.77; IC, 0.53; 95% CI, 0.19-0.78)”
Panel auditInternal validity: low · scope of inference: Reporting signal for semaglutide-associated suicidal ideation; liraglutide showed no signal.Key limitations: disproportionality not risk · reporting bias · signal limited to ideation · antidepressant co-reporting confound
Large active-comparator national-register cohort covering the hard outcome (suicide death).
“does not show an association between use of GLP-1 receptor agonists and an increased risk of suicide death, self-harm, or incident depression and anxiety-related disorders”
Panel auditInternal validity: high · scope of inference: Mostly T2D adults; suicide death and self-harm over mean 2.5-year follow-up.Key limitations: suicide death rare · wide CI on primary outcome · mostly T2D population
Randomised placebo-controlled (post-hoc) confirmation for semaglutide in obesity/T2D.
“treatment with semaglutide, 2.4 mg, did not increase the risk of developing symptoms of depression or suicidal ideation/behavior vs placebo”
Panel auditInternal validity: high · scope of inference: Semaglutide 2.4 mg in people with overweight/obesity WITHOUT known major psychopathology.Key limitations: excludes major psychopathology · post hoc analysis · very low event rate · semaglutide only
Discordant large cohort; included to represent genuine corpus disagreement, lowering consistency.
“patients on GLP-1 RAs exhibited a 195% higher risk of major depression, a 108% increased risk for anxiety, and a 106% elevated risk for suicidal behavior”
Panel auditInternal validity: low · scope of inference: Liraglutide/semaglutide in obesity, risk of developing suicidal behaviour post-marketing.Key limitations: effect sizes implausibly large · no CI reported · confounding by indication · percent-increase framing only
4 of 30 papers shown. The full review lists all of them, with the panel’s arbitration →
Claims are graded at each quarterly release, and re-graded between releases only when a pivotal trial lands. Every change is a new version of the claim.
Gold-standard RCT and active-comparator cohort evidence consistently supports no class-wide increase, but a discordant cohort, drug-specific reporting signals and residual imprecision in high-risk subpopulations preclude the top tier.
Release 2026-q4 re-graded every claim from its corpus under protocol v0.19. An earlier release's grade is shown for the record; it came from an earlier protocol and is not a movement.
Assumptions that carry weight but are not themselves established.
The panel records the gap that would most move the claim, and the study that would close it.
High-psychopathology population untested in trials. The reassuring RCTs excluded people with known major psychopathology [9][25], yet the claim implicitly covers these patients. The panel rates this the highest-importance open gap for the claim, and names the study design that would close it: An RCT enrolling patients with established psychiatric illness, with C-SSRS as a primary safety outcome.
Open question: Does semaglutide specifically increase suicidal ideation even if the GLP-1 class does not? Why it is still open: Multiple databases detect a semaglutide-only signal [5][12][16][18][30] while the pooled trial analysis is null for the class [24].
Priors is a sceptical first reader, not peer review, and not medical advice.
Release 2026-q4 · graded 21 Oct 2026 · protocol v0.19