Peptide basics

Three in four people in the pivotal GLP-1 weight trial were women

This page does not tell you what any compound will do for you, and it contains no dose, schedule or route of administration. It sets out who was actually enrolled in the trials that GLP-1 weight-loss claims come from, so you can read those claims with the right population attached.

The number, and where it comes from

STEP 1 randomly assigned 1,961 adults with overweight or obesity to semaglutide or placebo over 68 weeks. The results posted to ClinicalTrials.gov record the sex of every one of them: 1,453 female and 508 male. That is 74.1% and 25.9%.¹²

The registry also records a mean age of 46 years for the whole randomised population. So the single most-cited body of GLP-1 weight-loss evidence describes, in the main, women in their mid-forties.¹

Nothing about that is hidden or contested. It is in the posted baseline characteristics table, one click from the trial's registry page, and anyone quoting a STEP 1 figure can check it in under a minute.¹

It is not one trial. It is the pattern

If STEP 1 were an outlier, the imbalance would be a curiosity. It is not an outlier. STEP 3 posted 495 female and 116 male participants out of 611, which is 81.0% female. STEP 4 posted baseline data for the 803 participants who reached randomisation: 634 female and 169 male, or 79.0% female.³⁴

Other compounds in the class recruited the same way. SURMOUNT-1 randomised 2,539 adults, 1,714 female and 825 male — 67.5% female. The SCALE obesity trial of liraglutide randomised 3,731 adults, 2,928 female and 803 male, or 78.5% female.⁵⁶

• STEP 1: 1,453 of 1,961 female (74.1%).
• STEP 3: 495 of 611 female (81.0%).
• STEP 4: 634 of 803 randomised female (79.0%).
• SURMOUNT-1: 1,714 of 2,539 female (67.5%).
• SCALE Obesity and Prediabetes: 2,928 of 3,731 female (78.5%).¹³⁴⁵⁶

A published review of the semaglutide programme reached the same conclusion from the other direction, reporting that men were 26.9% of those randomised to semaglutide in STEP 1, 22.6% in STEP 3 and 19.8% in STEP 4, and describing the male population as understudied in the phase-3a programme.⁸

What that changes when you read a claim

The first thing it changes is the sentence. A result from STEP 1 is not a result in adults; it is a result in a group that was three-quarters women with a mean age of 46. When a headline drops the second half, it is not summarising, it is generalising.¹

The second thing it changes is precision. A smaller group produces a wider interval around any estimate drawn from it. With 508 men in STEP 1 against 1,453 women, anything reported separately for men rests on roughly a third as many people, and should be read as correspondingly less settled.¹

The third thing it changes is what counts as a surprise. If a later study in a more male population reports a different average, that is not necessarily a contradiction. The two studies may simply be describing different people.⁷

The mirror image: the cardiovascular trial

The clearest proof that these rosters reflect recruitment rather than the drug is SELECT, which studied the same compound in people with established cardiovascular disease. It randomised 17,604 adults: 12,732 male and 4,872 female. That is 72.3% male, almost the exact inverse of STEP 1.⁷

Its mean age was 61.6 years against STEP 1's 46, and its minimum entry age was 45 rather than 18. Same molecule, two trials, two populations that barely overlap.¹⁷

A trial roster is not a market

It is tempting to read a 74% female trial as evidence that mostly women take these drugs. Survey data does not support that. KFF's May 2024 Health Tracking Poll of 1,479 US adults found 12% saying they had ever taken a GLP-1 agonist, and reported that similar shares of adults took them regardless of gender.⁹

So the two facts point in different directions: the evidence base is predominantly female, and self-reported use is not. Both can be true, and holding them together is the whole point of this page.¹⁹

Semaglutide is the compound with the largest and most-quoted weight-management trial record. STEP 1 is where most of the widely repeated numbers originate, and its 74.1% female roster travels with every one of them.

Common questions