The GLP-1 Problem No One Talks About: Why Ozempic Users Are Losing Muscle, Not Just Fat

The scale is the wrong instrument for this job.

A patient loses 45 pounds on semaglutide, feels great about it, and comes to me for a baseline. We run a DEXA scan. Twelve of those 45 pounds were lean tissue. She traded a quarter of her weight loss for muscle she had spent her whole life building, and nobody at the online clinic that wrote the prescription ever measured anything except her weight.

I see a version of this every month. GLP-1 drugs work. The problem is that "losing weight" and "losing fat" are not the same thing, and most GLP-1 prescribing treats them as if they were.

This post is about that gap: how much muscle these drugs cost you, why it matters more than the number on the scale, and what a physician-led program does differently to protect it. If you want the broader case for medical supervision on these drugs (labs, GI complications, thyroid screening, the exit strategy), read our companion piece on GLP-1 medications with a full medical team. This one stays on muscle.

What the Trials Actually Measured

The headline numbers are real. In STEP 1 (Wilding and colleagues, New England Journal of Medicine, 2021), adults on weekly semaglutide 2.4 mg lost 14.9% of body weight over 68 weeks against 2.4% on placebo. In SURMOUNT-4 (Aronne and colleagues, JAMA, 2024), participants lost an average of 20.9% during 36 weeks of tirzepatide.

What those headline numbers do not tell you is what the lost weight was made of.

The best answer we have comes from a 2026 systematic review in the Annals of Internal Medicine (Batsis and colleagues). They pooled 35 randomized trials of semaglutide, tirzepatide, liraglutide and dulaglutide that measured body composition with DEXA, bioimpedance, CT or MRI. The median share of total weight loss that came from muscle-related measures was 28.3%, with a range across studies of roughly 16% to 40%. Two thirds of the trials exceeded the benchmark the reviewers set for expected muscle loss. And not one of the 35 trials measured whether participants could function better, worse or the same afterward.

Read that last sentence again. Thirty-five trials. Zero measurements of strength or physical function.

A narrative review in Diabetes Care (Locatelli and colleagues, 2024) put the loss in plainer terms: incretin drugs cause a rapid lean mass loss of roughly 10%, about 6 kg, which the authors compared to a decade or more of normal aging.

A caveat, because honesty matters here. "Lean mass" on a DEXA scan is not all skeletal muscle. It includes organ tissue and water, and some of what these scans register as lean loss is liver shrinkage, which is not a bad thing in someone with fatty liver. That is a fair point and it does not rescue the picture. The CT and MRI studies in the Batsis review, which measure muscle directly, still showed muscle losses above benchmark in two thirds of cases.

Why Losing Muscle Is Worse Than It Sounds

Muscle is not decoration. It is the largest metabolically active tissue you own, and three things happen when you lose a chunk of it during pharmacological weight loss.

Your resting metabolism drops. Resting energy expenditure tracks the mass of your organs and lean tissue closely; Wang and colleagues (American Journal of Clinical Nutrition, 2005) built a model that predicted resting energy expenditure from organ and tissue mass with a correlation of 0.92. Less lean tissue, fewer calories burned at rest, and a body primed to regain fat the day the drug stops.

Your strength predicts your survival. In the PURE study (Leong and colleagues, Lancet, 2015), which followed nearly 140,000 adults in 17 countries, grip strength was a stronger predictor of all-cause and cardiovascular death than systolic blood pressure. Every 5 kg drop in grip strength carried a 16% higher risk of dying during follow-up. Weak is not a cosmetic problem.

The regain comes back as fat. In the STEP 1 extension (Wilding and colleagues, Diabetes, Obesity and Metabolism, 2022), participants regained two thirds of their lost weight within a year of stopping semaglutide. In SURMOUNT-4 (Aronne and colleagues, JAMA, 2024), people switched from tirzepatide to placebo regained 14% of body weight over 52 weeks while those who stayed on the drug lost another 5.5%. If you lose 40 pounds with a quarter of it muscle and regain 30 of them, mostly as fat, you end up heavier in body fat and lighter in muscle than the day you started. That is the worst possible trade.

We have written more about what happens to the body as muscle goes in our post on the effects of muscle loss.

This Is Not Unique to GLP-1s. The Speed Is.

Any large calorie deficit costs lean tissue. Chaston and colleagues (International Journal of Obesity, 2007) reviewed 55 cohorts losing more than 10 kg by diet or surgery and found the proportion of weight lost as fat-free mass rose with the degree of caloric restriction and the speed of loss, and fell when exercise was added.

GLP-1 drugs create a very large, very fast deficit with almost no effort from the patient. That is the point of them. It is also why the muscle problem is worse on these drugs than on a sensible diet: the appetite suppression can push intake below 1,000 calories a day without the person noticing, protein-rich foods often become unappealing, and fatigue during dose escalation means people move less. Large deficit, low protein, no training. Every lever pointed the wrong way.

Who Loses the Most

Not everyone loses muscle at the same rate. In my practice, the patients I watch most closely on a GLP-1 are:

• Adults over 50. You are already losing muscle to age. A drug-driven loss stacks on top of it and can push someone from "a little weaker" into frailty territory within a year.
• Women. Lower starting muscle mass means each pound of lean tissue lost is a larger fraction of the total.
• Anyone who is not lifting. Without a resistance stimulus, the body has no reason to keep expensive tissue during a deficit.
• Rapid responders. Losing 20% of body weight in under a year is the highest-risk scenario for muscle, whatever the cause.
• Low-protein eaters. Which, on these drugs, is almost everyone unless someone is counting.

If two or more of those describe you, the plan below is not optional.

The Four Things That Protect Muscle on a GLP-1

1. Lift weights. This is the biggest lever by far.

A 2026 meta-analysis in Diabetes, Obesity and Metabolism (Deller and colleagues) pooled 34 randomized trials with 1,455 overweight or obese participants on calorie restriction, with or without exercise. Exercise prevented nearly half (45.7%) of the fat-free mass loss that calorie restriction alone produced, and programs that included strength training had the largest effect.

The evidence on GLP-1 drugs specifically points the same direction. In a randomized trial (Lundgren and colleagues, New England Journal of Medicine, 2021), adults who had lost 13 kg on a low-calorie diet were assigned for a year to exercise, liraglutide, both, or placebo. The combination group lost the most weight, dropped body fat percentage about twice as much as either treatment alone, and was the only group that improved insulin sensitivity, HbA1c and cardiorespiratory fitness. Then the researchers stopped everything and watched for another year (Jensen and colleagues, EClinicalMedicine, 2024). Weight regain was 6 kg larger after stopping liraglutide alone than after stopping supervised exercise, and the people who had exercised held onto their body composition. The drug alone did not.

What this looks like in practice: three to four resistance sessions a week, full-body or upper/lower split, compound movements, sets taken close to failure, loads progressing over time. Beginners can start with two sessions of goblet squats, rows, push-ups and lunges. The specifics matter less than showing up and adding weight. On Rebel Peak a certified strength coach programs this for you around your energy, your GI symptoms during titration, and your DEXA numbers.

2. Eat protein like it is a prescription.

The RDA of 0.8 g/kg is a floor for avoiding deficiency, not a target for someone in a large deficit. Phillips and colleagues (Applied Physiology, Nutrition, and Metabolism, 2016) put the range for optimal health at 1.2 to 1.6 g/kg per day, and Morton and colleagues (British Journal of Sports Medicine, 2018), pooling 49 trials of resistance training, found no further lean mass gains above about 1.6 g/kg. For a 180-pound person that is roughly 100 to 130 grams a day.

Distribution matters too. Mamerow and colleagues (Journal of Nutrition, 2014) found that spreading protein evenly across three meals, about 30 g each, produced 25% higher 24-hour muscle protein synthesis than eating the same total skewed toward dinner. On a GLP-1, that means protein first at every meal, and a shake when solid food will not go down. A registered dietitian who understands how appetite behaves on these drugs makes this achievable. Willpower alone rarely does.

3. Do not let the deficit run away from you.

Murphy and colleagues (Scandinavian Journal of Medicine and Science in Sports, 2022) pooled resistance-training trials run in an energy deficit and found that a deficit of roughly 500 calories a day was enough to prevent lean mass gains, and advised anyone trying to preserve muscle during weight loss to stay under that. A patient eating 800 calories a day on semaglutide is in a 1,500-calorie deficit. No amount of lifting will hold muscle through that. Sometimes the right move is a slower titration or a lower dose, which is a decision your physician makes with your body composition data in hand, not a corporate dosing calendar.

4. Measure what you are losing.

If you take one thing from this post: get a DEXA scan before you start, and repeat it every three to six months while you are on the drug. It is a cash-pay add-on, it takes about ten minutes, and it is the only way to know whether the weight coming off is fat, muscle or both. Bioimpedance scales are less precise but fine for tracking trends. Your lifts are the free daily signal: if the scale drops and your strength holds, you are doing this right. If the scale drops and your strength drops with it, something needs to change this week, not next quarter.

What We Do at Rebel Health Alliance

Whether a GLP-1 belongs in your protocol is a decision your physician makes with you, based on your history and labs. When it does, the program is built around body composition, not body weight.

Before the first dose: the initial lab panel (about 30 tests at wholesale cash-pay rates, including fasting insulin, HbA1c, a full thyroid panel, B12, vitamin D and iron; we lay out the markers in our guide to the blood panel your doctor isn't running), plus a baseline DEXA as a cash-pay add-on. If fasting insulin comes back high, that shapes the whole approach; see how to lower fasting insulin. Then a protein target set to your lean mass and a training program set to your experience.

During treatment: your physician titrates against your tolerance and your DEXA trend, not a schedule. If lean loss is running high, protein goes up, training changes, or the dose comes down. You message your team anytime and replies usually land the same day. On Rebel Peak, the dietitian and the strength coach work from the same bloodwork and the same scan as your physician. On Rebel Health, the physician-only membership, your doctor sets the same targets and you can add coaching a la carte.

Our members' repeat DEXA scans are one of the outcomes we publish. Visceral fat on repeat DEXA fell 12% (n=28) across members with a baseline and a follow-up scan, with labs through March 2026. That is a practice audit, not a trial, and individual results vary. The full set is on our results page, and you can read how the program is structured on how it works.

Your Move

The drug handles appetite. It does nothing for muscle, and left alone it will take a good share of yours. Protein, progressive resistance training, a deficit you can actually train through, and a DEXA scan every few months are what turn "weight loss" into fat loss.

A prescription is not a plan. Let us build you one that keeps your muscle.

Book a 20-minute call with our physician-led team to talk through GLP-1 therapy with body composition tracking built in.

Frequently asked questions

Does Ozempic directly cause muscle loss?
Not through its mechanism. Semaglutide suppresses appetite, which creates a large calorie deficit, and any large deficit costs lean tissue. What makes GLP-1 drugs different is the size and speed of the deficit and how little protein people end up eating on them. The 2026 Annals of Internal Medicine review by Batsis and colleagues found a median 28% of weight lost in incretin trials came from muscle-related measures.

How much protein should I eat on a GLP-1?
Aim for 1.2 to 1.6 grams per kilogram of body weight per day, split across three or four meals of roughly 30 grams each. For most adults that is 100 to 140 grams a day. Above about 1.6 g/kg there is no additional muscle gain in people who lift. If you cannot hit that with food, a protein shake counts.

Can I actually gain muscle while on semaglutide or tirzepatide?
Some people do, with a structured program: three or more resistance sessions a week, protein at the top of the range, and a deficit that is not extreme. Most people on a well-run program lose a small amount of lean mass rather than gaining, and that is a good outcome. Losing a quarter or more of your weight as lean tissue is the outcome to avoid.

How often should I get a DEXA scan on a GLP-1?
Before you start, then every three to six months while on the drug, and once more a few months after you stop. It is a cash-pay add-on and the single most useful test for this problem. Between scans, your strength numbers in the gym are the best day-to-day signal.

Will the muscle come back after I stop the drug?
Muscle you lost can be rebuilt with training and protein, but it is far harder than keeping it in the first place, and weight regain after stopping tends to come back as fat. In the STEP 1 extension, participants regained two thirds of their lost weight within a year. The people who fare best after stopping are the ones who were lifting the whole way through.

This article is for educational purposes only and does not constitute medical advice. Consult a qualified physician before starting, stopping or modifying any medication. Individual results vary, and GLP-1 therapy should be tailored to your specific medical history.

Dr. Alec Weir is the Chief Medical Officer at Rebel Health Alliance (https://rebelhealthalliance.io), a 100% virtual, physician-led longevity practice in all 50 states. Learn more about how the program works or read the FAQ.