The 30-second summary
- A new analysis of the SELECT trial, published on 24 September 2026 in Diabetes Care, looked at 7,594 people who took semaglutide, the drug in Wegovy, for at least a year.
- The median peak weight loss was 12.5%, reached after a median of 19 months.
- Everything doctors routinely know about you at the start explained less than 10% of the difference between people.
- Women lost the most, a median of 16.6%, against 12.4% for men.
What did the study look at?
SELECT is the large trial in which semaglutide 2.4 mg cut major cardiovascular events by 20% in 17,604 adults with overweight or obesity and existing heart disease, without diabetes. It ran for years, which makes it one of the few places to see how weight loss on semaglutide plays out over the long run.
This prespecified analysis took everyone on semaglutide who stayed on treatment for at least a year, 7,594 people or 86.3% of those assigned to the drug. For each person it found their peak weight loss, the most they lost at any point, and asked whether anything known at the start could have predicted it.
The participants had an average age of 62, and 72% were men. Everyone had established heart disease.
How much did people lose, and when?
| Median peak weight loss | |
|---|---|
| Everyone on semaglutide for a year or more | 12.5% |
| Women | 16.6% |
| Men | 12.4% |
| Highest dose reached was 2.4 mg | 13.3% |
| Highest dose reached was 0.24 mg | 6.0% |
The middle half of people lost between 7.9% and 17.9% at their peak. So a quarter lost less than 7.9%, and a quarter more than 17.9%.
The peak came after a median of 19 months. For the middle half, it came between 10 and 30 months. That is far later than the 68-week mark most trial headlines use. In STEP 1, the average loss at 68 weeks was 14.9% (Wilding et al., 2021), and many people in SELECT were still losing after that point.
Over the whole trial, the earlier SELECT weight report found the average loss held for up to four years, at 10.2% against 1.5% on placebo at 208 weeks (Ryan et al., 2024).
Why could nobody predict it?
The researchers fed every routinely available starting fact into a model: age, sex, weight, medical history, lab results and clinical measurements. Together, they explained less than 10% of the variation in peak weight loss. Sex alone made up 85.3% of what the model could explain, about 8% of the total.
In plain terms, two people who look identical on paper can lose very different amounts. The authors conclude that the difference likely comes from things the trial did not measure, biological, behavioural or genetic.
The link with dose is real but needs care. People whose highest dose was 2.4 mg lost far more than those who stayed at 0.24 mg. This study cannot say why some people stayed low, and side effects, choices and medical reasons all play a part. It is not evidence that anyone should push their own dose higher.
What are the limits?
- SELECT's population is specific. Everyone had established heart disease, the average age was 62, and most were men. Younger people, or groups that are mostly women, may see different numbers.
- Only people who stayed on treatment for a year were included, so this describes those who continued, not everyone who started.
- The analysis was partly industry-run. Several authors work for Novo Nordisk, which makes semaglutide.
- Diet and activity were not part of the model, so the study cannot say how much they shaped the result.
- The full paper is behind a paywall. This article uses the published abstract.
What does it mean if you take semaglutide?
Three useful things.
Your number is your own. If a friend lost 20% and you are at 9%, this study suggests that is not a sign you are doing something wrong. Most of the difference between people could not be explained by anything measured at the start.
The peak often comes late. A median of 19 months means that for many people, a slowdown at month six or nine is not the end of the loss. Our guide to the plateau on a GLP-1 covers what tends to help, and why a stall is often temporary.
Judge the loss by what it is made of. Since the amount is hard to control, the quality is worth attention: how much is fat and how much muscle. A protein floor and regular strength work are the two levers with the best evidence, and what to eat on Wegovy turns them into meals.
If your weight loss has stopped well short of what you hoped, talk to your prescriber about it. Do not change your dose on your own.
Sources
- Lingvay I, Kushner RF, Kaul U, et al. Predictors of Peak Body Weight Loss With Semaglutide in the SELECT Trial. Diabetes Care 2026. PubMed
- Ryan DH, Lingvay I, Deanfield J, et al. Long-term weight loss effects of semaglutide in obesity without diabetes in the SELECT trial. Nat Med 2024. PubMed
- Wilding JPH, Batterham RL, Calanna S, et al. Once-Weekly Semaglutide in Adults with Overweight or Obesity. N Engl J Med 2021. PubMed