This site has published that incretin therapy takes roughly six kilograms of lean mass with the fat. It is about to publish that the lean-mass effect is 0.78 kilograms. Both are accurate, and anybody who reads them side by side deserves to know why.
What this analysis did
A search of five databases to the end of 2024 turned up 2,776 articles, of which three randomized trials covering 171 participants met the criteria: adults with obesity, without type 2 diabetes, on a GLP-1, with body composition reported. [1] Against control, lean mass fell by 0.78 kg more, 95% CI −1.37 to −0.16, and fat mass by 3.43 kg more, 95% CI −5.94 to −0.93. Heterogeneity was high in both, at 63.2% and 78.2%.
What the split looks like
Taking the two point estimates at face value, the extra weight lost on the drug was about 18.5% lean and the rest fat. That division is our arithmetic rather than the paper’s — it uses only the central estimates, ignores both confidence intervals, and ignores the substantial heterogeneity. Treat it as a shape, not a measurement.
The shape is worth having anyway, because it is more favorable than the proportions this site has published from the larger literature. Around a third of total weight lost being lean is a real and often-quoted figure; the incremental picture attributable to the drug looks better than that. Both belong in view — the proportions across twenty trials are the larger evidence base, and this is the cleaner comparison.
Why the evidence base is this small
Two thousand seven hundred and seventy-six articles screened, three trials included. Measuring body composition properly requires DEXA or similar imaging at two timepoints, and most weight-loss trials simply weigh people. The question everybody asks is the one fewest trials were built to answer.
Restricting to people without type 2 diabetes narrows it further, and it is the right restriction for most readers here — but it means three trials and 171 people are carrying the whole estimate, with heterogeneity high enough that the trials disagreed considerably among themselves.
What it changes practically
Not the advice, which is unchanged and boring: resistance training and adequate protein during weight loss, whatever is causing the weight loss. The authors end on exactly that, and what it costs to do is the part no seller includes — which is its own line in the real total.
What it should change is how you read a frightening lean-mass number. The first question is always compared to what — against a control group, or against where the person started. If the answer is the second, part of what is being described is the cost of losing weight at all, which applies to every method. The same distinction sits under any figure that looks too clean and under every expected-loss estimate anybody shows you.