Everyone knows someone for whom these drugs did very little and someone for whom they did a great deal. An analysis across a federated health platform tried to characterize the two ends, sorting 135,349 treated people into super responders above 15% weight loss, moderate responders between 5% and 15%, a minimal group below 5%, and regainers [1]. The question of how early you can tell which group you are in is handled in slow responders at week eight.
The stable finding is the middle. Moderate response ran at 40% to 42% regardless of brand, which is a narrow band across four products and 135,349 people. The ends are where the spread appeared: super responders were 34% on Zepbound, 26% on Wegovy, 24% on Mounjaro and 10% on Ozempic.
The demographic picture is easier to read and harder to act on. Super responders were younger by about four years on average, were 80% female against 58% to 65% in the other groups, and were 90% white against 80%. The authors attribute this to some mixture of biological, behavioral and social factors, which is the honest answer and not a useful one — a distribution like that reflects who reaches treatment and stays on it as much as who responds to it, a pattern already visible in what else weight-drug users were already carrying.
One part of the paper does not parse, and it is worth flagging rather than smoothing over. The conditions it describes as enriched among super responders include fibromyalgia at a rate ratio of 0.2 and osteoarthritis at 0.5, both for Zepbound. A ratio below 1 describes a condition that is less common in that group, not more, so either the label or the direction is wrong in the summary as published. Psoriasis at a rate ratio of 2.5 for Wegovy is the only one of the three whose direction matches the description.
There is no randomization here, no adherence data and no record of what dose anyone actually took, and the response groups are defined by the outcome being explained. What the study is good for is scale: it establishes that the spread in real-world results is enormous, that about four in ten land in the middle whatever they take, and that a published average is a poor guide to a personal result — which is the same reason the headline figure on a seller's page deserves the scrutiny given it in the headline against the bill and two thirds stop within a year.