Ozempic muscle loss: what the studies have actually measured

Of the 13.6kg lost on average by the STEP 1 participants whose body composition was measured, 5.3kg was not fat. That ratio, 38%, fed the headlines about Ozempic muscle loss, and then the worry of many people on weight loss jabs who feel weaker going up a flight of stairs.

The figure is accurate. The conclusion drawn from it is less so, for two reasons: lean mass is not muscle, and any weight loss takes some of it, with or without a medicine. That is no excuse for doing nothing, because strength can be defended very well, provided you look after it.

What the trials measured

Two sub-studies are the reference. Both used dual-energy X-ray absorptiometry, or DXA, a scan that distinguishes fat, bone and everything else, the remainder being called lean mass.

The first covers 140 participants in STEP 1, the trial of semaglutide at 2.4mg, the Wegovy dose (Ozempic, prescribed for type 2 diabetes, contains the same molecule at lower doses). Its results were presented in the Journal of the Endocrine Society, 2021. The second, published in Diabetes, Obesity and Metabolism, 2025, concerns 160 participants in SURMOUNT-1, the trial of tirzepatide (Mounjaro).

The two trials were not run in the same people or with the same machines. This table therefore cannot be used to say that one medicine spares muscle better than the other. It shows that fat falls far more than the rest. In STEP 1 the proportion of lean mass in the body actually rose by 3 percentage points by the end of the trial: participants had less lean mass in absolute terms, but it made up a larger share of them.

One detail of SURMOUNT-1 deserves attention. In the placebo group, which had lost about 5% of body weight on lifestyle advice alone, the split was identical: three quarters fat, one quarter lean mass. The medicine amplifies the weight loss; it does not change its nature.

Lean mass is not muscle

DXA files under lean mass everything that is neither fat nor bone mineral: muscle, but also the organs, the body's water and the water held in fat tissue itself.

The authors of the joint advisory from four American learned societies (Obesity, 2025) estimate that muscle makes up about half of lean mass. On that basis the 38% in STEP 1 would correspond to a muscle loss of the order of 20% of the weight lost. It is an estimate, not a direct measurement.

A review devoted to the subject in Diabetes, Obesity and Metabolism, 2024 underlines how much the results vary: depending on the trial, lean mass accounts for 40 to 60% of the weight lost, or barely 15%. Its authors take from MRI studies that the reduction in muscle volume appears proportionate to the weight loss, and that the muscle left behind contains less infiltrated fat. They speak of an adaptation more than a wasting.

Is Ozempic muscle loss a serious problem?

For most people, nothing suggests that it is. A lighter body needs less muscle to move itself around, and a review in Advances in Nutrition in 2017 had already concluded, of conventional diets, that weight loss reduces muscle mass without impairing strength and improves physical function overall.

Caution is called for in certain groups. The 2024 review names older people and those with severe illness, who are more exposed to sarcopenia, that is, a loss of muscle and strength that ends up getting in the way of daily life. There is a more basic limitation as well: the trials weighed tissue, and very few measured strength or walking. Little is known about the effect of these treatments on what really matters.

Strength training, the best-supported lever

Resistance exercise is what best protects muscle during weight loss. The clearest demonstration comes from a trial published in the New England Journal of Medicine, 2017, run in 160 older adults with obesity who were dieting, with no medicine involved. All lost about 9% of their weight in six months. Those who did aerobic exercise alone lost 5% of their lean mass, those who did resistance training, 2%. In that last group strength increased by 19%.

On GLP-1 medicines the data are scarcer. A Danish trial in the New England Journal of Medicine, 2021 compared, over a year, Saxenda (liraglutide), an exercise programme, and the two together. The combination lowered body fat percentage by 3.9 points, about twice as much as either strategy alone, and it alone improved cardiorespiratory fitness.

The 2025 advisory turns that into a recommendation: pair treatment with a structured programme, with strength training at least three times a week and at least 150 minutes of moderate aerobic activity. The NHS guideline for every adult asks for the same 150 minutes and for strengthening activities that work all the major muscle groups on at least two days a week, so the expert advice builds on a familiar base.

In concrete terms it means working the large muscle groups against a load, using resistance bands, dumbbells, gym machines or your own body weight. A physiotherapist or a qualified exercise professional can help you start without injury, especially after years away from sport, and the NHS suggests speaking to your GP first if you have not exercised for some time.

Protein, the other half of the answer

Protein contributes to the maintenance of muscle mass, and intake drops easily when appetite disappears. The experts suggest aiming for 1.2 to 1.6g per kilo a day during weight loss, against the UK reference intake of 0.75g for a healthy adult. Our GLP-1 diet guide explains how to get there with ordinary food.

One useful clarification: according to the 2017 review, eating more protein helps to preserve lean mass but does not increase strength. Protein and exercise do not stand in for one another.

When appetite is low, small protein-rich helpings spread through the day are easier than three proper meals: a pot of yoghurt, an egg, a glass of milk, a tin of fish. The 7-day GLP-1 diet plan puts numbers on a week of eating that way.

What you can keep an eye on yourself

The scales say nothing about muscle. A few everyday tests say more: getting up from a low chair, carrying the shopping, climbing two flights of stairs without stopping. If those become harder over the months while the weight is falling, talk to your prescriber, who can measure grip strength or arrange a body composition scan. The 2025 advisory in fact recommends assessing muscle strength and function from the start of treatment.

Medicines designed to preserve muscle alongside these treatments are being studied. None is available today. In the meantime the changes people notice first tend to be the visible ones, in the mirror, a subject we take up in Ozempic face, while the less visible ones are listed in our guide to GLP-1 side effects.

This article is for information only and is no substitute for advice from a healthcare professional. Never change your treatment without talking to your doctor.