GLP-1 receptor agonists for antipsychotic weight gain: about 6.5 kg lost

Design

Systematic review and meta-analysis of 8 RCTs

Population

535 adults on antipsychotics with overweight or metabolic dysregulation

Primary outcome

Change in weight, BMI, HbA1c and total cholesterol

Effect

Weight MD −6.54 kg (−9.80 to −3.29); HbA1c MD −0.32%

This meta-analysis pooled eight randomised trials with 535 adults on antipsychotics who were overweight or had metabolic dysregulation, comparing GLP-1 receptor agonists with control.

GLP-1 receptor agonists reduced body weight by 6.54 kg (95% CI −9.80 to −3.29), BMI by 2.34 kg/m², waist circumference by 5.2 cm and HbA1c by 0.32%. Semaglutide produced more weight loss than liraglutide, and longer treatment produced more. Nausea (RR 1.61), vomiting (RR 2.10) and constipation (RR 2.34) were more common, but serious adverse events and overall discontinuation were not. Heterogeneity for weight was very high (I² 92%), and psychiatric adverse events were too few to pool.

For people whose weight gain persists despite lifestyle measures and, where feasible, a switch to a lower-liability antipsychotic, this supports adding a GLP-1 receptor agonist, ideally with the diabetes or obesity team. Constipation deserves particular attention in people on clozapine, whose gut motility is already slowed.

• Consider a GLP-1 receptor agonist when antipsychotic weight gain persists despite lifestyle support and review of the antipsychotic.
• Semaglutide produced more weight loss than liraglutide in these trials.
• Warn about nausea, vomiting and constipation; check bowel habit closely in people on clozapine.
• Monitor mood and psychotic symptoms after starting — psychiatric safety data were too sparse to pool.
• Cost and access limit use in India; metformin remains a cheaper first option many services already use.

Why it matters

It gives psychiatrists an evidence-based option beyond advice and switching for one of the commonest reasons patients stop antipsychotics.

Don't overread it

Psychiatric safety in this population was not established; the trials were small and short.

The statistics, in plain English

A mean difference of −6.54 kg means the treated groups lost about 6.5 kg more than controls on average. I² of 92% means the trials varied a great deal, so the effect in any one patient could be much larger or smaller.

Read the rest in the app

You have read your two free briefings this month. The app carries all 27 specialties, every morning, free — and this finding is waiting in it.

Get Daily Doctor — free

Scan to keep reading on your phone. No account needed to start.