Guide

Muscle loss on a GLP-1: why it happens, how to stop it

Some lean tissue leaves in any large deficit, medication or not. The share is not fixed, almost nobody in the trials was eating enough protein or lifting anything, and that is the opening. Here is what the research shows and what to do about it.

The short answer

Pooled across 20 randomised trials, 35.2% of the weight lost on semaglutide was lean mass and 25.4% on tirzepatide. Diet alone cost 26.2%, which is the point: this is what large weight loss does, not what the injection does. Adding resistance training took it to 17.5%, and that is the number you can move.

Take the muscle or fat quiz

The mechanism, in one paragraph

Your body needs a steady supply of amino acids every day, whether or not you are eating. When food stops providing them, it takes them from the largest available store, which is skeletal muscle. In an ordinary diet that process is slow, because hunger pushes you to eat and you keep some protein coming in. On a GLP-1, appetite drops faster than habits change, so protein disappears from the plate first while the deficit gets larger than anything you have run before. Add no strength signal and muscle becomes the cheapest fuel your body has.

Both halves of that are fixable. That is the entire point of the work we do, and it is why we measure strength weekly rather than trusting the scale.

What the research actually found

You will still see "40% of the weight lost is muscle" quoted everywhere. That came from one body composition sub-study of one medication, and by 2026 there is something better to quote. Across 20 randomised trials, an average of 35.2% of the weight lost on semaglutide (Ozempic, Wegovy) was lean mass. Losing the same weight through diet alone cost 26.2% as lean mass, which was not significantly different from the medications (p = 0.42). The muscle comes off because the weight comes off, not because of the injection. Adding resistance training brought it down to 17.5%, the most favourable result of any approach studied. Eisa N et al, Diabetes, Obesity and Metabolism 2026

The same analysis put the other two molecules lower. For tirzepatide (Mounjaro, Zepbound) the figure was 25.4%. Eisa N et al, Diabetes, Obesity and Metabolism 2026 For liraglutide (Saxenda, Victoza) the figure was 26.8%. Eisa N et al, Diabetes, Obesity and Metabolism 2026

A second review, of 35 separate studies, came out at a similar place with wider spread. Across 35 studies the median share of weight loss coming from muscle-related measures was 28.3%, with wide variation between studies. Batsis JA et al, Annals of Internal Medicine 2026

Two pieces of context matter. First, lean mass on a scan is not all muscle: it includes water, connective tissue and the extra tissue that supported a larger body, some of which you no longer need at a lower weight. Second, and this is the part the headlines leave out, the comparison group tells you what is really going on. Losing the same weight without a medication cost about the same share of lean tissue. The injection is not doing something strange to your muscle. It is producing a very large deficit, very reliably, in people who are mostly not eating protein or lifting anything.

Which is the good news, because nobody in those trials was asked to train or to hit a protein target. That is the variable nobody controlled, and it is the one you can.

Why the percentage on your scan can look fine while you get weaker

Lean mass as a percentage of body weight was unchanged in pooled trials, which is why people can lose real muscle and still look like the numbers are fine. Karakasis P et al, Metabolism 2025

It is worth sitting with that. If you lose 20 kg (44 lb) and a third of it was lean tissue, you have less muscle than you had, and yet the percentage of you that is lean can be identical to the day you started, because the fat left too. A body composition report can read like success while the stairs get harder. The weekly strength test exists for exactly this gap.

Why it matters after 45

Muscle does three jobs that become more valuable with age. It is the tissue that lets you stand up from a low chair, carry shopping and catch yourself when you trip. It is metabolically active, so it accounts for a meaningful share of what you burn at rest. And it is the buffer that protects you when you are ill or immobile for a fortnight.

Lose it in your fifties and you do not get it back easily. Muscle already declines gradually from midlife, and older muscle responds less strongly to a given meal of protein. The lean mass lost on the strongest of these medications, around 6 kg (13 lb), has been described as comparable to a decade or more of ageing. Locatelli JC et al, Diabetes Care 2024

Then there is the regain problem. Weight regained after stopping appears to come back disproportionately as fat, so repeated cycles can leave someone heavier in fat and lighter in muscle than when they started. Bosomworth NJ, Canadian Family Physician 2025 That is how somebody ends a year lighter, weaker, and with a worse body composition than they started with.

How to tell whether it is happening to you

You cannot feel muscle loss, and the mirror is a poor judge because fat is leaving at the same time. There is also a hole in the research here that nobody advertises. In a 2026 review of 35 studies in the Annals of Internal Medicine, not one reported whether people could actually do more or less afterwards. Scans were measured. Getting out of a chair was not. Batsis JA et al, Annals of Internal Medicine 2026

So the published work cannot tell you what any of this did to the things you would actually notice, and no home test diagnoses anything. What a weekly test does is compare you with you. Three numbers, checked every Sunday, do the job.

  1. One strength test, the same way every week

    Goblet squat repetitions at a fixed weight, push-ups, or the 30-second sit-to-stand if you are starting out. Fat loss does not make you weaker, so a number that falls three or four weeks running is a real signal. It measures what you can do, not how much muscle you have.

  2. Your protein average across the whole week

    Not your best day. Under 1.2 g per kg of goal weight explains most of the muscle loss we see in coaching.

  3. The tape around your upper arm

    Same arm, same spot, relaxed. Your waist will shrink whatever you do. If your arm is shrinking at the same rate, that is lean tissue leaving.

A DEXA scan every three to six months is the precise version if you want it, and it is a useful baseline. It is not necessary. The three numbers above cost nothing and catch the problem in the month it starts rather than the month after.

The warning signs people ignore

  • Feeling cold all the time Partly less food and less insulating fat, partly a smaller furnace. See feeling cold on a GLP-1.
  • Stairs feeling harder at a lower weight The clearest signal there is. You are carrying less and it feels heavier.
  • Hair shedding two to four months in Usually rapid loss and low protein together. See hair shedding.
  • A plateau with falling strength The combination that says the deficit is now coming out of muscle. See the six-month plateau.

The two fixes, and nothing else

Protein at 1.2 to 1.6 g per kilogram of goal body weight (0.55 to 0.7 g per pound), three or four times a day, at 0.3 to 0.4 g per kg a sitting. For a 70 kg (154 lb) goal that is 84 to 112 g a day and 21 to 28 g a meal. The detail, including how to eat it on a few bites a day, is in protein on a GLP-1, and the calculator gives you your own number in ten seconds.

Then two 30-minute full-body strength sessions a week: squat pattern, hinge, push, pull, carry, core brace, with enough load that the last two repetitions are hard. The programme is in strength training on a GLP-1. Two sessions is the minimum effective dose, and it is enough.

Creatine, collagen, amino acid drinks and the rest sit a long way behind those two, and none of them substitutes for either. Ask your prescriber before adding any supplement, particularly if you have kidney problems or take other medication.

Who is most at risk

Four groups show up repeatedly in coaching. People losing very fast, because a bigger deficit takes more from everywhere. People over 60, whose muscle is already declining and responds less strongly to a meal. Women in or past menopause, where falling oestrogen pulls in the same direction on muscle and bone. And anyone who started out with relatively little muscle to begin with, which includes most people who have dieted repeatedly for decades.

If you are in more than one of those groups, treat protein and training as part of the prescription rather than an optional extra. The detail is in over 50 on a GLP-1 and GLP-1 and menopause.

What muscle loss looks like after you stop

This is where the cost is collected. Appetite returns when the medication stops, and for most people weight follows. Someone who loses 20 kg (44 lb), a third of it lean tissue, and then regains 12 kg (26 lb) of mostly fat is worse off on body composition than before they started, at a similar weight on the scale. It is the outcome we see most often in people who did everything right except protect their muscle, and it is covered in coming off a GLP-1.

If it is already happening

Nothing here is lost permanently as long as you act. Get protein to the floor this week, start the two sessions this week, and keep your dose exactly as prescribed while you do, because changing or skipping medication is your prescriber's decision and never a response to a strength test. Then give it four weeks and read the trend rather than the days.

Most people we coach see the strength test stop falling within a fortnight and start climbing inside six weeks, while the scale keeps moving. If your numbers are still falling after a month of proper protein and training, that is worth taking to your prescriber along with your food log, because thyroid, iron and vitamin D all produce the same picture.

Free, one page

Forty foods, ranked by protein per bite

The hard part is not knowing you need protein, it is hitting the number on a day when food is the last thing you want. One page, grams and ounces, what a real serve looks like, and the first-bite rule for when appetite is gone. Print it and put it on the fridge.

Then five short emails over five days on keeping muscle while the weight comes off, and a short note every fortnight after that if you want it. One click to stop, any time.

Sent. Check your inbox in the next minute or two, and look in junk if it is not there.

Questions people ask

How much muscle do you lose on a GLP-1?

Across 20 randomised trials, an average of 35.2% of the weight lost on semaglutide (Ozempic, Wegovy) was lean mass. For tirzepatide (Mounjaro, Zepbound) the figure was 25.4%. Losing the same weight through diet alone cost 26.2% as lean mass, which was not significantly different from the medications (p = 0.42). The muscle comes off because the weight comes off, not because of the injection. Adding resistance training brought it down to 17.5%, the most favourable result of any approach studied.

Is it the medication that causes the muscle loss?

The pooled data says not really. Losing the same weight through diet alone cost 26.2% as lean mass, which was not significantly different from the medications (p = 0.42). The muscle comes off because the weight comes off, not because of the injection. That is why the answer is protein and load rather than a different prescription.

Is muscle loss on a GLP-1 permanent?

Not if you act while you are losing. Muscle regained after a large loss comes back slowly and only with protein and training, which is why protecting it now is far cheaper than rebuilding it later.

How do I know if I am losing muscle rather than fat?

Check three numbers every week: one strength test done the same way, your seven-day protein average, and a tape measure around your upper arm. A strength test falling for three or four weeks is the signal.

Does a body composition scan tell me more?

It is more precise and it makes a good baseline, and it has one blind spot. Lean mass as a percentage of body weight was unchanged in pooled trials, which is why people can lose real muscle and still look like the numbers are fine. The weekly strength test catches a problem in the month it begins, which a scan every six months cannot.

Will taking creatine stop muscle loss?

It sits a long way behind protein and training and replaces neither. Ask your prescriber before adding any supplement, particularly with kidney problems or other medication.

The research behind every figure on this page

Read the first three in order and they say something the headline version does not: this is what weight loss costs, and training is what changes the bill.

Across 20 randomised trials, an average of 35.2% of the weight lost on semaglutide (Ozempic, Wegovy) was lean mass. Eisa N et al, Diabetes, Obesity and Metabolism 2026

Losing the same weight through diet alone cost 26.2% as lean mass, which was not significantly different from the medications (p = 0.42). The muscle comes off because the weight comes off, not because of the injection. Eisa N et al, Diabetes, Obesity and Metabolism 2026

Adding resistance training brought the lean share of weight lost down to 17.5%, the most favourable result of any approach studied. Eisa N et al, Diabetes, Obesity and Metabolism 2026

Across 35 studies the median share of weight loss coming from muscle-related measures was 28.3%, with wide variation between studies. Batsis JA et al, Annals of Internal Medicine 2026

In a 2026 review of 35 studies in the Annals of Internal Medicine, not one reported whether people could actually do more or less afterwards. Scans were measured. Getting out of a chair was not. Batsis JA et al, Annals of Internal Medicine 2026

The papers these numbers come from

  • Eisa N, Barood O. Lean Mass Changes With Incretin Therapy Versus Lifestyle Intervention: A Systematic Review and Meta-Analysis of Randomised Controlled Trials. Diabetes, Obesity and Metabolism, 2026. 20 randomised controlled trials, 15,782 participants. doi:10.1111/dom.70666
  • Batsis JA, Gavras A, Gross DC, et al. Effect of Incretin-Based and Nonpharmacologic Weight Loss on Body Composition: A Systematic Review. Annals of Internal Medicine, 2026. 35 studies, median duration 26 weeks. doi:10.7326/ANNALS-25-00478
  • Karakasis P, Patoulias D, Fragakis N, Mantzoros CS. Effect of glucagon-like peptide-1 receptor agonists and co-agonists on body composition: Systematic review and network meta-analysis. Metabolism, 2025. 22 randomised controlled trials, 2,258 participants. doi:10.1016/j.metabol.2024.156113
  • Locatelli JC, Costa JG, Haynes A, et al. Incretin-Based Weight Loss Pharmacotherapy: Can Resistance Exercise Optimize Changes in Body Composition? Diabetes Care, 2024. Narrative review. doi:10.2337/dci23-0100
  • Bosomworth NJ. New drugs for weight loss: Why change in body composition matters and why nutrition and exercise remain paramount. Canadian Family Physician, 2025. GRADE-assessed review. doi:10.46747/cfp.711112705

Last reviewed 18 September 2026. We update these figures when a larger or better study replaces them, and we say what changed. How we handle evidence.

The free tools that go with this

The Keep Muscle Kit

The 12-week system for keeping what you have

The workbook, Sunday numbers tracker, 2-day strength plan, injection-day card, and meal plans at 100, 120 and 140 g protein. Instant download.

Get the Kit

A$47 / US$29 / £23, one payment.

Read next

Every guide, by medication and situation

EverStrong provides nutrition and strength coaching and education. It is not medical advice and does not replace your prescribing doctor. Consult your doctor before starting a new exercise or nutrition plan, especially with a heart, kidney, liver or metabolic condition, if pregnant, or with a history of disordered eating.