Scientific deep-dive

Which Weight-Loss Drug Works Best With Antipsychotics? What the Rankings Show

Five network meta-analyses rank drugs for antipsychotic weight gain, and they name three different leaders. Here is how semaglutide, liraglutide, metformin and topiramate compare, and how far to trust the order.

By Eli Marsden · Founding Editor
Editorially reviewed & fact-checked against primary sources · How we verify contentLast reviewed
8 min read·19 citations

Semaglutide has the largest estimated effect in the most complete analysis so far: −10.98 kg against placebo across 95 studies in people with schizophrenia-spectrum disorders taking antipsychotics, on three semaglutide trials and moderate certainty evidence.[1] Metformin has a smaller estimate (−3.86 kg) but rests on 16 trials.[1] Earlier network analyses put liraglutide or metformin with lifestyle changes at the top,[4][5] and the authors of one 2026 analysis say the comparative ranking of the GLP-1 drugs “is uncertain.”[2] These comparisons are mostly indirect, so the order is a guide to questions for a prescriber and not a settled league table.

About this article

This page compares the rankings and the non-GLP-1 options. For trial detail on semaglutide with clozapine or olanzapine, what happens after stopping, and the clozapine constipation warning, see our semaglutide and antipsychotic weight gain review. Nobody should stop or change an antipsychotic to lose weight without their psychiatrist.

Five network analyses, three different winners

A network meta-analysis estimates how treatments compare even when trials never tested them against each other, usually by linking them through placebo. That is why the same drugs can land in different orders as trials are added or grouped differently. The five analyses below were published between 2024 and 2026.[1][2][3][4][5]

What each network meta-analysis reported as its leading result (abstract figures)
AnalysisEvidence baseLeading resultNote
Hegde, Gen Hosp Psychiatry 202468 trials, BayesianMetformin 750 mg + lifestyle −7.5 kg named most effectiveSibutramine 10 mg had the largest estimate, −8.0 kg
Goetz-Kundera, Compr Psychiatry 202527 studies, 17 drugsLiraglutide −5.3 kg beyond 12 weeks; first by SUCRAAuthors favor metformin and liraglutide
Mansour, J Psychopharmacol 202655 trials, 2,977 people, search to July 2025Semaglutide −13.5 kg vs usual careMetformin + NutriEx −6.34 kg second
Stogios, JAMA Psychiatry 202695 studies, 39 drugs, 5,898 people, search to Dec 2025Semaglutide −10.98 kg vs placeboModerate certainty for the top five
Kishi, Transl Psychiatry 20269 trials, 595 people, GLP-1 drugs onlySemaglutide SMD −2.101 vs control83.9% schizophrenia-spectrum; liraglutide SMD −0.945; exenatide not reported as significant

Three things explain most of the movement. The first is timing. The semaglutide trials were published from mid-2025 onward,[6][7][8] and the two earlier analyses do not list semaglutide among their significant results, likely because few or none of those trials were available to them.[4][5]

The second is how treatments are split. Hegde analyzed each dose separately, which produced metformin 750 mg at −5.7 kg and metformin 1,000 mg at −2.3 kg.[5] A lower dose beating a higher one may reflect which small trials sit behind each estimate. The same paper put sibutramine at the top on point estimate, with an interval running from −16 to −0.21 kg.[5] Sibutramine is not an option in the US: the FDA says it was withdrawn from the market in December 2010 after data showed increased risk of heart attack and stroke.[18]

The third is thin evidence at the top. Kishi’s authors say semaglutide “may be the preferred treatment option” but that its ranking is uncertain “due to sparse networks, substantial heterogeneity, and predominant indirect evidence.”[2] Note that Kishi’s figures are standardized mean differences, a unitless measure, and cannot be read as kilograms.

The top five, and how much is behind each

Stogios gives the fullest picture, so its numbers anchor the table below, alongside other estimates for the same drugs. All five were rated moderate certainty in that analysis. Only semaglutide and metformin reached the 5% weight change the authors treated as clinically meaningful.[1] Read differences between rows as numerically larger or smaller, not as proven gaps: these abstracts report each drug against placebo or usual care, not against the others.

Mean weight difference compared with placebo in the Stogios network, with other published estimates
DrugStogios 2026Trials behind itAnother estimate
Semaglutide−10.98 kg (−13.33 to −8.62)3−13.5 kg vs usual care (Mansour)
Liraglutide−5.43 kg (−8.54 to −2.33)2−5.3 kg beyond 12 weeks (Goetz-Kundera)
Topiramate−3.95 kg (−5.89 to −2.02)5−4.82 kg for prevention, interval crosses zero (Cochrane)
Metformin−3.86 kg (−5.02 to −2.70)16−3.32 kg, 20 studies (Peng)
Exenatide−2.97 kg (−5.83 to −0.11)3−4.17 kg beyond 12 weeks (Goetz-Kundera)

The comparison figures come from Mansour,[3] Goetz-Kundera,[4] the 2022 Cochrane review of prevention[12] and Peng.[11] Exenatide shows why a ranked list can mislead. It reaches significance in two network analyses,[1][4] yet Kishi lists only liraglutide and semaglutide as significant[2] and a pairwise review described its reduction as smaller and non-significant.[19]

What sits behind the semaglutide estimate

The Stogios abstract does not name its three semaglutide trials. Three placebo-controlled trials we read show how varied this small evidence base is.

  • HISTORI: 154 Danish adults with schizophrenia, prediabetes and BMI of at least 27 took up to 1.0 mg weekly for 30 weeks. Weight fell 9.21 kg more than placebo. The primary outcome was HbA1c, and weight was a secondary endpoint.[7]
  • Sass: 73 Danish adults who started clozapine or olanzapine within the past 5 years took 1 mg weekly for 26 weeks. Weight fell 9.2 kg more than placebo. The primary analysis was intention-to-treat on HbA1c, the authors call secondary outcomes exploratory, and 78% completed.[8]
  • COaST: Australian adults on clozapine with BMI of at least 26 took semaglutide titrated to 2.0 mg for 36 weeks. The between-group difference was −13.46% of body weight, the primary outcome, analyzed with a mixed model for repeated measures in all randomized participants. Only 31 people were randomized against a target of 80, because the study drug became unavailable. It ran independently of industry support.[6]

Doses, durations and populations differ across these trials, which is part of what the word heterogeneity is pointing to. Pooled pairwise estimates for GLP-1 drugs as a group also show it: one 2026 review reports −6.54 kg with an I² of 92%,[9] and another reports −6.10 kg with an I² of 99% and low to moderate certainty.[10]

Options beyond the top five

Metformin

Metformin has the deepest evidence by trial count. A meta-analysis of 20 double-blind studies and 1,070 patients found −3.32 kg (95% CI −4.57 to −2.07) and reported that the effect held from 12 to 24 weeks.[11] Cochrane, looking at prevention of weight gain, rated its −4.03 kg estimate low certainty.[12] A broader view of the drug is in our metformin review.

Topiramate

Topiramate looks different depending on the question. It sits third in the Stogios treatment network,[1] while Cochrane’s prevention analysis of three studies found that it “did not appear effective,” with very low certainty.[12] For how the drug performs outside psychiatry, see our topiramate review.

Olanzapine with samidorphan

This combination pill is a different strategy: the data measure less gain on olanzapine, not loss of weight already gained. An individual patient data analysis of 1,063 people found a 3.68% gain at week 12 on the combination compared with 5.43% on olanzapine alone, a difference of −1.75%. Its authors include Alkermes employees.[14] A separate meta-analysis of four trials (1,416 people), whose authors declared no conflicts, found −1.18 kg and noted that naltrexone, another opioid-receptor blocker, did not separate from placebo.[13]

Smaller studies

Nizatidine shows up in several analyses, at −6.82 kg in short-term trials in one[4] and −5.46 kg in another,[3] but Stogios rates its evidence very low certainty.[1] In a 12-week trial of 84 adults on clozapine or olanzapine, empagliflozin and metformin did not differ significantly on weight; with no placebo arm, the trial cannot say how much either drug did on its own.[16] An 8-week orlistat trial of 60 people found a significant difference in BMI change and in some cholesterol measures, which is too short and small to place on a ranking.[17]

Tirzepatide: the gap

None of the network abstracts above reports a tirzepatide estimate. The closest data we found are observational. In a health-records study of 1,809 matched pairs with schizophrenia-spectrum disorders, tirzepatide use was associated with a lower composite of death, hospitalization and suicidal behavior (hazard ratio 0.66), with a hazard ratio of 0.18 for death alone.[15] The authors say control-outcome checks supported validity, but an association that large over one year could still partly reflect who gets prescribed a new drug. The abstract reports no weight outcome, so it cannot rank tirzepatide for weight. Any higher efficacy in people on antipsychotics would be an inference from other populations.

Side effects that travel with the ranking

Across the Stogios network there were “no major concerns” with gastrointestinal effects or dropouts between interventions.[1] Reviews limited to GLP-1 drugs show the class pattern more clearly. Pooling eight trials, one found higher risks of nausea (RR 1.61), vomiting (RR 2.10) and constipation (RR 2.34), no significant rise in serious adverse events (RR 0.72, 95% CI 0.49 to 1.06), and concluded that psychiatric safety “could not be established owing to sparse data.”[9] COaST saw no change in clozapine levels or PANSS symptom scores and low rates of constipation, in a trial of only 31 people.[6]

What to raise with a prescriber

  • Which antipsychotic you take. Some trials enrolled only people on clozapine or olanzapine,[6][8] so results may not carry over evenly to other medicines.
  • Whether the goal is preventing gain or losing weight. The Cochrane and samidorphan data address prevention,[12][14] while the Stogios network examines weight reduction.[1]
  • Your blood sugar. The HISTORI and Sass trials selected for prediabetes or early glucose changes.[7][8]
  • How constipation and other medicines will be monitored, particularly with clozapine, and who on your care team is in charge of that plan.
  • How long treatment would last. The semaglutide trials above ran 26 to 36 weeks,[6][7][8] and one review calls for longer-term trials with standardized safety reporting.[9]

Frequently Asked Questions

No single answer is settled. In a 2026 JAMA Psychiatry network meta-analysis of 95 studies, semaglutide had the largest estimate at −10.98 kg compared with placebo, based on three trials. Metformin had a smaller estimate, −3.86 kg, based on 16 trials. Other analyses order drugs differently, so this is a decision to make with a prescriber.
Semaglutide's estimate is numerically larger, but that comparison is indirect and runs mostly through placebo. Metformin has far more trials behind it. The semaglutide trials also differ in dose and population, and one of them enrolled only 31 people, so the size of the gap is uncertain.
None of the five network analyses reviewed here reports a tirzepatide estimate in their abstracts. One observational records study linked tirzepatide use with lower rates of death, hospitalization and suicidal behavior in people with schizophrenia-spectrum disorders, but it shows an association, not cause, and its abstract reports no weight outcome.
They include different trials from different dates, group doses differently, and sometimes use different comparators such as placebo or usual care. Several top-ranked drugs rest on only two or three small trials, so a new trial can move a drug up or down. One 2026 analysis calls the ranking of GLP-1 drugs uncertain.

References

  1. 1.Stogios N, Agarwal SM, Maksyutynska K, et al. Pharmacological Interventions for Weight Reduction in Patients With Schizophrenia Treated With Antipsychotics: A Systematic Review and Network Meta-Analysis. JAMA Psychiatry. 2026. PMID: 42418179.
  2. 2.Kishi T, Ikuta T, Sakuma K, et al. Glucagon-like peptide-1 receptor agonists for weight management in mental illness: a network meta-analysis. Transl Psychiatry. 2026. PMID: 42420255.
  3. 3.Mansour MEM, Alsaadany KR, Mustafa MMM, et al. Efficacy and safety of pharmacological and non-pharmacological interventions for antipsychotic-induced weight gain in individuals with schizophrenia: A systematic review and network meta-analysis of 55 clinical trials. J Psychopharmacol. 2026. PMID: 41439550.
  4. 4.Goetz-Kundera Z, Przekopińska B, Walichniewicz P, et al. Comparative effects of pharmacological treatment for weight gain in patients treated with antipsychotics. A systematic review and network meta-analysis. Compr Psychiatry. 2025. PMID: 40934730.
  5. 5.Hegde NC, Mishra A, Maiti R, et al. Pharmacological interventions for antipsychotic-induced weight gain in schizophrenia: A network meta-analysis. Gen Hosp Psychiatry. 2024. PMID: 38878592.
  6. 6.Siskind D, Baker A, Arnautovska U, et al. Efficacy and safety of semaglutide versus placebo for people with schizophrenia on clozapine with obesity (COaST): a phase 2, multi-centre, participant and investigator- blinded, randomised controlled trial in Australia. Lancet Psychiatry. 2025. PMID: 40506208.
  7. 7.Ganeshalingam AA, Uhrenholt N, Arnfred S, et al. Semaglutide Treatment of Antipsychotic-Treated Patients With Schizophrenia, Prediabetes, and Obesity: The HISTORI Randomized Clinical Trial. JAMA Psychiatry. 2025. PMID: 40900607.
  8. 8.Sass MR, Klausen MK, Schwarz CR, et al. Semaglutide and Early-Stage Metabolic Abnormalities in Individuals With Schizophrenia Spectrum Disorders: A Randomized Clinical Trial. JAMA Psychiatry. 2026. PMID: 41335431.
  9. 9.Zhou L, Zhang Y, Shen C, Wu F. Efficacy and safety of GLP-1 receptor agonists for antipsychotic-related overweight and metabolic dysregulation: a systematic review and meta-analysis. Eur J Clin Pharmacol. 2026. PMID: 42920001.
  10. 10.Abbas N, Memon AA, Zulfiqar R, et al. Semaglutide and Other GLP-1 Agonists for Antipsychotic-Associated Metabolic Dysfunction in Schizophrenia: A Systematic Review and Meta-Analysis. J Clin Psychopharmacol. 2026. PMID: 42503999.
  11. 11.Peng TR, Chen JA, Lee JA, et al. The Optimal Dosage and Duration of Metformin for Prevention and Treatment of Antipsychotic-Induced Weight Gain: An Updated Systematic Review and Meta-Analysis. Schizophr Bull. 2025. PMID: 39509416.
  12. 12.Agarwal SM, Stogios N, Ahsan ZA, et al. Pharmacological interventions for prevention of weight gain in people with schizophrenia. Cochrane Database Syst Rev. 2022. PMID: 36190739.
  13. 13.Lee K, Twohig M, Idoko NP, Williams BD. Centrally-acting opioid receptor antagonists as a treatment for antipsychotic-induced weight gain: A systematic review and meta-analysis of clinical trial data. J Psychopharmacol. 2025. PMID: 40406979.
  14. 14.Correll CU, Doane MJ, McDonnell D, et al. Olanzapine/Samidorphan Effects on Weight Gain: An Individual Patient Data Meta-Analysis of Phase 2 and 3 Randomized Double-Blind Studies. J Clin Psychiatry. 2025. PMID: 39832309.
  15. 15.Liu TH, Hsu CH, Wu JY, et al. Association between tirzepatide use and risk of mortality, hospitalization, and suicidal behavior in patients with schizophrenia spectrum disorders: A one-year retrospective cohort study of 3618 patients. Eur Neuropsychopharmacol. 2026. PMID: 41389474.
  16. 16.Dolati A, Ahmadimoghaddam D, Zamanirafe M, et al. Empagliflozin as an Alternative to Metformin for Antipsychotic-Related Weight Gain: Double-Blind Randomized Controlled Trial. Hum Psychopharmacol. 2026. PMID: 42135901.
  17. 17.Xie P, Shao T, Long Y, et al. Orlistat for the treatment of antipsychotic-induced weight gain: an eight-week multicenter, randomized, placebo-controlled, double-blind trial. Lipids Health Dis. 2024. PMID: 39049073.
  18. 18.US Food and Drug Administration. Questions and Answers about FDA's Initiative Against Contaminated Weight Loss Products. FDA. 2021. https://www.fda.gov/drugs/frequently-asked-questions-popular-topics/questions-and-answers-about-fdas-initiative-against-contaminated-weight-loss-products
  19. 19.Sobral MVS, Menechini RB, Marçal JBC, et al. Effects of glucagon-like peptide-1 receptor agonists on metabolic outcomes in antipsychotic-treated patients with schizophrenia: A systematic review and meta-analysis. Schizophr Res. 2026. PMID: 42497473.

Semaglutide for Antipsychotic Weight Gain: What the Clozapine and Olanzapine Trials Show

Semaglutide is now the best-studied add-on for antipsychotic weight gain. In two randomized trials of people on clozapine, olanzapine and similar drugs, it took off about 9 kg more than placebo without worsening symptoms. The trials, metformin and clozapine safety.

10 min read

GLP-1 Drugs and Eating Disorders: The 2026 Expert Consensus Explained

A 45-member panel in World Psychiatry advises screening before GLP-1 treatment and generally avoiding it in active eating disorders, particularly restrictive ones. Here is what it recommends, how thin the evidence is, and what it leaves out.

7 min read

GLP-1 With Lithium, Lamictal, or Atypical Antipsychotics

Atypical antipsychotics cause 5-15 kg weight gain on average. GLP-1 receptor agonists reverse this without affecting psychiatric stability. We review the published cohort data, the lithium PK + dehydration interaction, and the practical co-management protocol.

11 min read

Ozempic Drug Interactions: Common Medications and What Actually Matters

Can you take ibuprofen, Tylenol, antibiotics, or metformin with Ozempic? GLP-1 drug interactions explained: the real cautions vs the reassuring pharmacology.

10 min read

Metabolic Longevity: Do Anti-Aging Drugs & Supplements Actually Help Weight and Metabolism?

An evidence-first ranking of the popular 'metabolic longevity' interventions — GLP-1 agonists, metformin, berberine, NAD precursors, rapamycin, and senolytics — by how much human weight and metabolic evidence actually supports them, with a link to each fully-cited deep-dive.

6 min read

Do Weight-Loss Drugs Improve Natural Fertility? What the Meta-Analyses Show

GLP-1 therapy was linked to more natural pregnancies than metformin in PCOS trials, but pooled data are small, overall pregnancy did not differ significantly, and live birth benefit remains unproven. Here is what the reviews, orlistat trials and real-world records show.

7 min read

Where to get semaglutide (Ozempic / Wegovy) safely: vetted online providers

Vetted telehealth providers that prescribe online. We compare pricing, form, and states served.

No insurance needed · vetted by our editors

WeightLossRankings.org is reader-supported. When you buy through links on our site, we may earn an affiliate commission. How we make money

7.2

HumeCare+

GLP-1 weight loss with real-time body-composition tracking to protect muscle

Read full review →

From $199/mo

Get started →
8.3

Collective

The lowest flat medication price if you can absorb an annual membership fee

Read full review →

From $59/mo

+ required membership

Get started →
8.3

CoreAge Rx

Lowest-price compounded semaglutide and tirzepatide with one flat price at every dose

Read full review →

From $79/mo

“starting at” price

Get started →

About this analysis

Weight Loss Rankings is an independent editorial publication. Our provider dataset is verified directly against each provider's website and updated continuously. Read our full methodology →