GLP-1 in Serious Mental Illness: 764,115 Matched Pairs
People with bipolar disorder, major depressive disorder or schizophrenia die earlier than the general population, and mostly of cardiovascular disease. A target trial emulation published in JAMA Psychiatry on 26 August 2026 asked whether starting a GLP-1 receptor agonist rather than an SGLT2 inhibitor is associated with a smaller gap.
The matched cohort ran to 1,528,230 adults, 764,115 pairs, of which 195,184 pairs had a serious mental illness.
A framing note. These are approved medicines and this is a study of patients. Peptra Labs supplies tirzepatide as reference material for laboratory research only, and nothing here is guidance about human use or about managing any psychiatric condition.
What the GLP-1 emulation actually compared
The design is worth stating precisely, because the comparator is the whole point.
The authors used the TriNetX Analytics Network with a new-user, active-comparator design and propensity score matching, splitting the population into a psychiatric cohort and a non-psychiatric one. The exposure was a first-recorded prescription for a GLP-1 receptor agonist. The comparator was a first-recorded prescription for an SGLT2 inhibitor.
An active comparator matters more here than in most database work. Comparing a new drug against no drug tells you mainly about who gets prescribed things. Comparing it against another modern agent, given for overlapping reasons to a similar clinical profile, removes a great deal of that. SGLT2 inhibitors are not a placebo; they have their own established mortality benefit, which makes this a demanding reference rather than an easy one.
Prespecified follow-up was 1, 4 and 10 years, with 4-year all-cause mortality as the primary outcome.
The GLP-1 numbers
Among the participants with serious mental illness, four-year mortality was 4.91 percent with a GLP-1 receptor agonist against 6.45 percent with an SGLT2 inhibitor. Hazard ratio 0.76, confidence interval 0.74 to 0.78, absolute risk difference 1.54 percentage points.
In a separately matched one-year cohort, mortality was 1.46 percent against 2.84 percent, a relative risk of 0.52.
| Subgroup, 10-year exploratory | Relative risk for mortality |
|---|---|
| Major depressive disorder | 0.55, CI 0.53 to 0.56 |
| Bipolar disorder | 0.57, CI 0.51 to 0.63 |
| Schizophrenia | 0.67, CI 0.59 to 0.75 |
Among participants with serious mental illness and type 2 diabetes, semaglutide against an SGLT2 inhibitor was associated with lower three-point and five-point major adverse cardiovascular events, myocardial infarction, stroke, heart failure and coronary artery bypass grafting.
The authors report that the primary mortality association persisted across the prespecified sensitivity analyses, which included stratification by diabetes, exclusion of baseline heart failure and chronic kidney disease, and censoring at crossover.
The line that names our catalogue
The conclusion contains a sentence that most coverage will skip: the benefits were driven by semaglutide and tirzepatide.
That is an agent-specific claim inside a GLP-1 class-level analysis, and it is the kind of detail that matters for anyone tracking which molecules the evidence actually attaches to. Tirzepatide is a dual GIP and GLP-1 agonist and it is in the Peptra Labs catalogue as reference material. Semaglutide is not in the catalogue at all.
The authors also close by saying prospective randomised trials are warranted, which is what an emulation should always say about itself.
Why the absolute GLP-1 numbers matter more than the ratios
The GLP-1 hazard ratio of 0.76 is a moderate effect. The absolute risk difference of 1.54 percentage points over four years is the number with practical meaning, and the authors are explicit that the absolute reductions were greater in the serious mental illness group than outside it.
That asymmetry is the finding. The same relative effect applied to a population with higher baseline mortality produces more avoided deaths, which is why interventions that look unremarkable in a healthy cohort can matter disproportionately in a high-risk one.
It is also the reason to be careful. A population with higher baseline risk is also a population where confounding by indication bites harder, because the decision to start a GLP-1 rather than something else in someone with schizophrenia is not a random one.
What this adds to a difficult month for this class
We have now covered three papers touching psychiatric and neurological endpoints, and they do not line up neatly.
In a cohort of 13,007 psychiatric patients and a cognitive score, semaglutide initiation was associated with fewer recorded cognitive signs, though not significantly against sitagliptin. In six years of European adverse event reports, the same compound accumulated a suicidal ideation reporting signal with an interval running from 1.57 to 26.81. Now a mortality benefit in the same broad population, in the best journal of the three.
Those are not contradictions. They are three different instruments pointed at three different GLP-1 questions, and the only honest summary is that the psychiatric population is where GLP-1 is being studied hardest right now, and where the evidence is least settled.
A fourth reading sits underneath. In a FAERS analysis of cardiac reporting published in July, semaglutide accumulated disproportionate cardiac signals against tirzepatide, and the authors attributed that to demographic and indication differences rather than toxicity. Here the same compound is associated with fewer cardiovascular events than a different comparator. Both can be true, because reporting and events are not the same measurement.
Where the catalogue sits
Neither compound in this analysis is supplied by us for human use, and the practical content for a laboratory is narrower than the headline.
Retatrutide, the triple agonist, appears in none of this work, because it has no market and therefore no real-world dataset. Every GLP-1 paper we have covered this month has depended on records generated by prescribing, which means the newest compounds are structurally invisible to this entire method until they are approved.
For laboratories, our European research buyer guide covers procurement and documentation, everything is supplied on a research use only basis, and our ranking of the most-studied research peptides counts published work rather than prescriptions.
References
- McIntyre RS, Zhang-James Y, Kwan ATH. Glucagon-Like Peptide 1 Receptor Agonists, Mortality, and Cardiovascular Outcomes in Serious Mental Illness. JAMA Psychiatry, 26 August 2026. doi 10.1001/jamapsychiatry.2026.2574
- De Giorgi R, Lipunova N, Mathias E, Shang F, Taquet M. Cognitive signs and symptoms in people with a psychiatric diagnosis on semaglutide. BMJ Mental Health, 20 August 2026
- Ammendolia I, Mondello C, Esposito E, et al. Psychiatric and eye disorders associated with use of glucagon-like peptide 1 receptor agonists (GLP-1 RAs). Naunyn-Schmiedeberg’s Archives of Pharmacology, 22 August 2026
- Singla A, Kaur G, Singh G, et al. Comparative Cardiovascular Pharmacovigilance of Semaglutide and Tirzepatide: A FAERS Database Analysis (2022-2026). Cureus, 25 July 2026
The products referenced on this site are supplied for laboratory research use only. They are not medicines and are not intended for human or veterinary use. This article summarises published research for informational purposes and is not medical advice. Statements about third-party studies belong to their authors.
