GLP-1 and Exercise Statistics (2026)

Q: What percentage of weight lost on a GLP-1 is muscle?

It depends on the drug and lifestyle, but the trial substudies put it at roughly 25 to 45 percent of total weight lost. The STEP 1 body composition substudy found about 45 percent of the weight lost on semaglutide came from lean mass, while the SURMOUNT-1 substudy found closer to 25 percent for tirzepatide. The 2024 Neeland review summarizes the overall range across studies as roughly 20 to 40 percent, in line with what any rapid weight loss produces without resistance training and adequate protein.

Q: Do people exercise more after starting a GLP-1?

No. A 2026 study presented at ENDO, the Endocrine Society's annual meeting, found the opposite. Among 753 adults with obesity tracked with Fitbit data, daily steps fell from about 5,047 to 4,487 after starting a GLP-1, a decline of roughly 560 steps per day, and moderate-to-vigorous activity dropped from about 28 minutes to 22 minutes per day. The researchers concluded that exercise cannot be treated as optional for people on these medications.

Q: How much protein should someone on a GLP-1 eat to protect muscle?

Clinical guidance points to 1.2 to 1.6 grams of protein per kilogram of body weight per day during active weight loss on a GLP-1. The Morton et al. 2018 meta-analysis of 49 studies found total protein intake was the strongest dietary driver of resistance-training results, with benefits plateauing near 1.62 grams per kilogram per day. Appetite suppression from the medication makes reaching this target the main practical challenge.

Q: How much muscle loss does resistance training prevent during weight loss?

Most of it. A 2018 meta-analysis by Sardeli et al. in Nutrients pooled six trials of older adults in a calorie deficit and found resistance training preserved nearly all lean mass while still allowing significant fat loss. The 2024 Neeland review applies the same evidence to GLP-1 users and recommends resistance training at least twice weekly plus adequate protein to close most of the lean-mass gap.

Q: Is tirzepatide better than semaglutide for preserving muscle?

The trial substudies suggest a small edge, but the drugs are closer than the headlines imply. SURMOUNT-1 reported about 75 percent of weight lost on tirzepatide was fat and 25 percent lean, versus roughly 55 percent fat and 45 percent lean for semaglutide in STEP 1. The absolute kilograms of muscle lost are similar because tirzepatide produces more total weight loss, and resistance training plus protein is a much larger lever than the choice of drug.

Summary The picture from the data is consistent. GLP-1 medications produce large weight loss, and roughly 25 to 45 percent of that weight comes from lean mass depending on the drug (Wilding et al., 2021 for semaglutide; Look et al., 2025 for tirzepatide). People also tend to move less after starting, not more: a 2026 study presented at ENDO found daily steps dropped by about 560 and moderate-to-vigorous activity fell by nearly six minutes a day (Maharjan et al., 2026). The countermeasures are well established. Resistance training preserves nearly all lean mass during a deficit (Sardeli et al., 2018), and protein intake of 1.2 to 1.6 g per kg per day supports it, with benefits plateauing near 1.62 g per kg (Morton et al., 2018).

GLP-1 receptor agonists (semaglutide as Ozempic and Wegovy, tirzepatide as Mounjaro and Zepbound) reshaped weight loss in a few short years. The conversation around them has been dominated by the weight numbers, which are genuinely impressive. This page is about the quieter set of numbers: what happens to muscle and movement, and what the evidence says you can do about it. It's a reference, organized by question, with every statistic sourced.

A quick framing note before the numbers. Muscle loss during weight loss is not unique to GLP-1s. It happens with any large, fast reduction in body weight. What's specific here is the combination of very large weight loss and a documented tendency to become more sedentary, which together raise the stakes on the muscle question. Let's go through it.

How much of the weight lost on a GLP-1 is muscle?

This is the headline question, and the trial substudies answer it directly using DXA scans, the gold standard for measuring body composition. The short version: a lot, unless you intervene.

The semaglutide number comes from the STEP 1 body composition substudy (Wilding et al., 2021), published in Diabetes, Obesity and Metabolism. Participants lost an average of 15.3 kg, of which about 6.9 kg was lean tissue. That's roughly 45 percent of total weight lost coming off as lean mass. The tirzepatide number comes from the SURMOUNT-1 substudy (Look et al., 2025), which reported about 75 percent of weight lost was fat and 25 percent lean. The broader synthesis by Neeland et al. (2024) puts the overall range across studies at roughly 20 to 40 percent.

~45%

of weight lost on semaglutide came from lean mass (STEP 1 substudy)

Wilding et al., 2021 · Diabetes Obes Metab · PMC8089287

~25%

of weight lost on tirzepatide came from lean mass (SURMOUNT-1 substudy)

Look et al., 2025 · Diabetes Obes Metab · DOI

20 to 40%

lean-mass fraction range across GLP-1 studies

Neeland et al., 2024 · Diabetes Obes Metab · DOI

The nuance that gets lost in the headlines: this ratio is normal for rapid weight loss. What makes the absolute muscle loss large is that the total weight loss is large. Losing 15 percent of your body weight at a 45 percent lean fraction takes off more muscle than losing 5 percent at the same ratio. The lever that changes the ratio is lifestyle, which is where the rest of these numbers come in. We cover the body-composition science in full on our page about GLP-1 drugs and muscle loss.

Do people move more or less after starting a GLP-1?

Less. This is the finding that reframes the whole muscle question, and it's new. You'd reasonably expect that losing weight makes movement easier and people become more active. The largest wearable-based look at this found the reverse.

The study, led by Sajana Maharjan and presented at ENDO 2026 (the Endocrine Society's annual meeting), used the NIH All of Us Research Program, which links electronic health records to participants' Fitbit data. From 1,950 adults with obesity who started a GLP-1, the team analyzed the 753 who had enough wearable data. The cohort was 78.6 percent female with a mean age of 52.7 years.

Daily steps fell from an average of 5,047 before starting the medication to 4,487 after, a drop of about 560 steps per day. Moderate-to-vigorous physical activity declined from about 27.9 minutes per day to 22.2, a loss of roughly 5.7 minutes. Both changes were highly statistically significant. The largest declines showed up in men and in people with joint or muscle pain, while age, heart failure, and prior stroke did not change the pattern. The authors' takeaway was blunt: there was no evidence that weight loss led to more activity, so exercise cannot be treated as optional for people on these drugs.

Activity measure	Before GLP-1	After GLP-1	Change
Daily steps	5,047	4,487	-560
Moderate-to-vigorous activity (min/day)	27.9	22.2	-5.7

Maharjan et al., presented at ENDO 2026. Retrospective pre-post cohort, n=753 adults with obesity, NIH All of Us Research Program (electronic health records linked to Fitbit). Both changes P < .001.

Put this next to the muscle-loss numbers and the concern sharpens. A body in a large calorie deficit that's also moving less has two reasons to shed muscle instead of just fat. This is the mechanistic case for treating exercise as part of the prescription, not an afterthought. Our blog answers the direct version of this question in do I need to lift weights on Ozempic.

Get an evidence-based plan built for you

FitCraft, our mobile fitness app, pairs you with an AI coach who builds you a personalized plan around your goals, schedule, and fitness level. Every FitCraft program is designed by Domenic Angelino, MPH (Brown University) and NSCA-CSCS, with research published in the Journal of Strength and Conditioning Research and Medicine & Science in Sports & Exercise.

Take the Free Assessment Free • 2 minutes • No credit card

How much muscle can resistance training save?

Most of it. This is the most encouraging set of numbers on the page, and it's not new science. Resistance training during a calorie deficit has a long, consistent evidence base.

The cleanest number comes from a 2018 meta-analysis by Sardeli et al. in Nutrients, which pooled six randomized trials of older adults in caloric restriction. Resistance training preserved nearly all lean mass while still allowing significant fat loss. The Cava, Yeat, and Mittendorfer (2017) review in Advances in Nutrition reached the same conclusion for weight loss generally: higher protein plus resistance training protects muscle and strength far better than dieting alone. The 2024 Neeland review took this established evidence and made it the formal recommendation for GLP-1 users: resistance training at least twice a week, plus adequate protein.

The dose matters and it's lower than people fear. During a deficit, the goal is to preserve muscle, which takes far less work than building it. Around 10 hard sets per major muscle group per week is enough, comfortably within two or three short full-body sessions. The full week-by-week layout lives in our GLP-1 muscle preservation workout plan.

What protein intake does the research point to?

The protein target is one of the best-established numbers in this whole area. For active weight loss, clinical guidance lands at 1.2 to 1.6 grams of protein per kilogram of body weight per day, with the top of the range for older adults and anyone starting with low muscle mass.

The anchor for that range is the Morton et al. (2018) meta-analysis in the British Journal of Sports Medicine, which pooled 49 studies with 1,863 participants. It found total daily protein was the strongest dietary driver of resistance-training gains, and that benefits plateaued near 1.62 g per kg per day. Above that, more protein didn't add measurable gains in the pooled data. So there's both a floor and a ceiling to aim for.

1.2 to 1.6

grams of protein per kg per day recommended during GLP-1 weight loss

Neeland et al., 2024 · Cava et al., 2017 · PMC5421125

1.62 g/kg

the point where added protein stops improving training gains

Morton et al., 2018 · Br J Sports Med · PMID 28698222

49 studies

pooled in the Morton protein meta-analysis (n=1,863)

Morton et al., 2018 · Br J Sports Med · PMID 28698222

The practical catch is unique to GLP-1s: appetite suppression makes hitting even the low end genuinely hard. When you're eating far less overall, protein is the macro most people fall short on, because high-protein foods are filling and the hunger that would normally push you to eat is gone. That's why the standard advice for GLP-1 users is to eat protein first at every meal and lean on protein-dense, low-volume options.

The weight-loss numbers, for context

To keep the muscle numbers in perspective, here are the weight-loss figures they sit inside. The pivotal semaglutide trial (Wilding et al., 2021, STEP 1) randomized 1,961 adults and produced a mean weight loss of 14.9 percent versus 2.4 percent on placebo over 68 weeks. Tirzepatide in SURMOUNT-1 pushed average weight loss on the highest dose to roughly 22 percent. These are the numbers that make the absolute muscle loss meaningful: the bigger the total, the more lean tissue rides along without countermeasures.

Drug (trial)	Mean weight loss	Fat fraction of loss	Lean fraction of loss	Source
Semaglutide (STEP 1)	~14.9%	~55%	~45%	PMC8089287
Tirzepatide (SURMOUNT-1)	~22%	~75%	~25%	DOI

Weight-loss figures from the pivotal STEP 1 (Wilding et al., 2021) and SURMOUNT-1 trials; body-composition fractions from their DXA substudies. Fractions are approximate and vary with population and lifestyle.

What the numbers add up to

Line up the statistics and a clear, actionable story emerges. GLP-1 medications produce large weight loss, of which a meaningful fraction is muscle. People tend to move less after starting, which pushes that fraction higher. And the two countermeasures, resistance training two to three times a week and protein at 1.2 to 1.6 g per kg per day, are cheap, well-evidenced, and close most of the gap. None of these numbers argue against the medications. They argue for pairing them with a floor of resistance training and protein from day one.

Preserving muscle isn't just about looking a certain way. Muscle drives strength, metabolic health, and long-term function, which is why the loss matters most for older adults and anyone at risk of sarcopenia. For the why-it-matters-beyond-weight side, see our research page on sarcopenia and resistance training.

Frequently Asked Questions

What percentage of weight lost on a GLP-1 is muscle?

It depends on the drug and lifestyle, but the trial substudies put it at roughly 25 to 45 percent of total weight lost. The STEP 1 body composition substudy found about 45 percent of the weight lost on semaglutide came from lean mass, while the SURMOUNT-1 substudy found closer to 25 percent for tirzepatide. The 2024 Neeland review summarizes the overall range across studies as roughly 20 to 40 percent, in line with what any rapid weight loss produces without resistance training and adequate protein.

Do people exercise more after starting a GLP-1?