GLP-1 users ate about 271 kcal less at test lunch

GLP-1 receptor agonists, and the dual GIP/GLP-1 agonist tirzepatide, produce substantial weight loss, and the assumed mechanism is that people simply eat less. Yet what actually happens to measured food intake on these drugs has rarely been pulled together quantitatively, with most attention going to weight on the scale. Objectively measured intake matters because it tells you whether the effect is appetite suppression at a meal, or something more diffuse. This review focused on one concrete, comparable measure: how much energy people consume at a standardised meal they can eat freely.

The authors searched six databases through March 2026 for studies reporting quantitative dietary outcomes in adults taking a GLP-1 or dual GIP/GLP-1 receptor agonist. For the primary analysis they pooled randomised or crossover placebo-controlled trials lasting at least four weeks that reported standardised ad libitum lunch intake, using a DerSimonian-Laird random-effects model. Sixteen studies entered the systematic review, and four arms from three trials, comprising 209 participants, contributed to the meta-analysis. Prespecified sensitivity analyses included alternative heterogeneity estimators, a prediction interval, subgroup and leave-one-out analyses, and the addition of an exploratory three-week phase 1 tirzepatide trial.

The pooled treatment difference in ad libitum lunch energy intake was 1132 kJ lower on drug than placebo (95% confidence interval 1449 to 815 kJ lower), roughly 271 kcal, with no heterogeneity at all between the contributing trials and a 95% prediction interval from 1828 to 436 kJ. Adding the exploratory phase 1 tirzepatide trial as a fifth comparison increased the estimated effect to 1421 kJ but introduced substantial heterogeneity of 70%, showing how sensitive the figure is to which studies are included. The comparison between semaglutide and tirzepatide was not statistically significant. The authors emphasise that habitual, real-world intake remains underreported across the literature.

This puts a number on the appetite mechanism widely assumed to explain GLP-1 weight loss, and the tight consistency across the three pooled trials is reassuring for the acute effect. What it does not do is establish how much people eat across a whole day, week or year, which is what actually determines weight change. The absence of a significant semaglutide-versus-tirzepatide difference should be read as insufficient evidence to separate them rather than proof they are equivalent. Better reporting of habitual dietary intake in longer trials is the clear gap.

This is a reason to think of these medicines as working substantially through reduced food intake at a meal, rather than through some mysterious metabolic route. It also frames the practical consequence: if intake falls by that order of magnitude, the composition of what remains on the plate becomes more important, and the authors highlight adequate protein alongside structured nutritional support. A single laboratory lunch is not your Tuesday, so treat the figure as an illustration of appetite effect size rather than a daily deficit. Prescribing and nutritional planning are matters for a clinician and dietitian.