The part of the number nobody celebrates

The scale gives you one number and lets you assume it's all fat. It isn't. Weight lost is always a mixture, and on a GLP-1 the mixture is worth paying attention to, because the losses are fast and the appetite suppression makes it easy to eat in a way that tilts the mixture the wrong direction.

Reviews of medically induced weight loss put lean mass at roughly 25 to 40% of the total lost when nothing is done to protect it (Prado et al., 2024). So on a 20 kg loss, somewhere between 5 and 8 kg of it may not be fat.

The good news, and it is genuinely good news, is that this ratio is not fixed. It responds to two things, both of them boring, both of them well studied, and neither of them requiring a gym membership or a supplement cupboard. This piece is the numbers behind those two things and how to actually run them while your appetite is suppressed.

A note on names: tirzepatide is Mounjaro in the UK, Ireland and Australia, and Zepbound in the US. Semaglutide injection is Wegovy in all four. The muscle question is identical either way.

The honest counterpoint

Before the protocol, the caveat, because this topic attracts a lot of alarmism and I'd rather you got the balanced version.

Some reduction in lean mass during weight loss is expected and appropriate. A smaller body needs less muscle to carry it around, less connective tissue and less blood volume, and "lean mass" on a DXA scan includes all of that, not just muscle. Three researchers who've spent careers on this made exactly that argument in a 2024 JAMA viewpoint, questioning how much of the weight-loss-induced muscle loss reported in GLP-1 trials is actually clinically meaningful (Conte, Hall & Klein, 2024).

The broader review evidence points the same way on function. Compared with people of normal weight, people with obesity carry more muscle but of poorer quality. Diet-induced weight loss reduces muscle mass without reliably reducing muscle strength, and overall physical function tends to improve after weight loss, mostly because there's less fat to move around (Cava et al., 2017).

So the honest framing isn't "GLP-1s are melting your muscle." It's this: you have a measurable amount of influence over the composition of what you lose, the levers are cheap, and the case for pulling them is strongest if you're older, if you're losing quickly, or if you were not carrying much muscle to begin with.

Why a GLP-1 makes this harder than ordinary dieting

Three things stack up, and it's worth naming them because each has a different answer.

Total intake falls fast, and protein falls fastest. When appetite drops sharply, people don't reduce their food evenly. Protein-rich foods are heavy, filling and the least appealing thing to force down when you're not hungry or slightly nauseated. Toast is easy. Chicken is not. So protein as a share of intake tends to fall exactly when it matters most.

The rate of loss is quicker than most diets. Faster loss generally means a larger deficit, and larger deficits are associated with a higher share of lean mass in what comes off.

You're training into a headwind. A meta-analysis of resistance-training studies found that being in an energy deficit impairs the lean-mass gains you'd otherwise get from training, though it doesn't impair the strength gains (Murphy & Koehler, 2022). That's worth knowing so you set the right expectation: on a deficit, the realistic win is keeping the muscle and getting noticeably stronger, not adding size.

Lever one: the protein number

The range with the most support behind it is 1.2 to 1.6 g of protein per kg of body weight a day while you're losing weight. That's a range rather than a single figure on purpose, because a range is what the evidence supports and anyone quoting you a number to the gram is adding certainty that isn't there.

What that looks like in practice:

• 70 kg: 84 to 112 g a day, or roughly 28 to 37 g at each of three meals
• 80 kg: 96 to 128 g a day, or roughly 32 to 43 g per meal
• 90 kg: 108 to 144 g a day, or roughly 36 to 48 g per meal
• 100 kg: 120 to 160 g a day, or roughly 40 to 53 g per meal

The protein target calculator runs this for your weight and adjusts for age and whether you're training, and it shows the working rather than hiding a formula.

Two findings anchor the range. The first is a trial that pushed protein hard: young men in a 40% energy deficit, training six days a week, randomised to either 1.2 or 2.4 g per kg a day. The higher-protein group gained 1.2 kg of lean mass while the lower-protein group essentially held steady at 0.1 kg, and the higher-protein group also lost more fat, 4.8 kg against 3.5 kg (Longland et al., 2016). That's a striking result and it deserves its caveats: 40 young men, four weeks, an extreme training volume and a supervised diet. It shows the direction of the effect clearly. It isn't a promise about a 55-year-old on a GLP-1.

The second anchors the top end. A meta-analysis of 49 resistance-training studies covering 1,863 people found that protein supplementation did increase strength and fat-free mass, but that intakes above roughly 1.62 g per kg a day produced no further gain in fat-free mass (Morton et al., 2018). That's where the 1.6 ceiling in the range comes from. Beyond it you're mostly buying expensive certainty.

Two adjustments worth knowing. If you're 65 or over, sit toward the top of the range: older adults lose lean mass more readily and respond less to a given amount of protein, which is why the international position paper on protein for older people recommends more than the standard adult guideline, not less (Bauer et al., 2013). And if you're carrying a lot of body fat, using your full current weight can produce a very large number, since fat tissue doesn't need feeding with protein. The calculator handles that; if you're doing it on paper, it's a reasonable question for a dietitian.

One more practical point: spread it out. Protein is used more effectively divided across meals than concentrated into one, so a similar amount at each of three meals beats a small breakfast and an enormous dinner.

Lever two: the training signal

Protein is the raw material. Resistance training is the message telling your body the muscle is worth keeping. Neither works nearly as well alone.

The clearest number I know on this comes from a systematic review of 52 studies in adults over 50 with a BMI above 25. Among groups doing energy restriction alone, 81% lost 15% or more of their body weight as fat-free mass. Among groups doing energy restriction plus exercise, that fell to 39% (Weinheimer et al., 2010). Same weight loss, very different composition.

Resistance training specifically has an advantage over cardio here: both help preserve muscle mass during weight loss, but resistance training also improves muscle strength, which is what actually determines whether you can carry shopping and get off a low chair at 75 (Cava et al., 2017).

The dose that the preservation research is built on is modest: two or three sessions a week, twenty to thirty minutes, covering the major movement patterns. Not six days a week, not two hours. If you want an actual routine that needs no equipment, there's one in bodyweight training on a GLP-1, and I'd rather point you there than pad this piece out repeating it.

The protocol, on one page

Daily. Hit 1.2 to 1.6 g of protein per kg of body weight, spread across your meals. Eat the protein on the plate first, while you still have appetite to spend.

Two or three times a week. Twenty to thirty minutes of resistance work covering a push, a pull, a squat or hinge, and something for the core. Progress it by making it harder over time, whether that's more reps, more range, slower tempo or more load.

Every week. Weigh on a consistent schedule and read the trend, not the day.

Every month or two. Check something that isn't weight. Waist measurement, or a strength benchmark like how many press-ups or how long a wall sit. If your weight is falling and your strength benchmark is holding or improving, the composition is going the way you want, and that's more useful information than the scale gives you.

When you step up a dose. Expect appetite and possibly nausea to dip further for a week or two. Protect protein first and let the rest of the intake fall where it falls.

Hitting the target when you are not hungry

This is the part that decides whether any of the above happens, because a target you can't reach is just a source of guilt.

Front-load the day. Appetite on these medicines is usually best in the morning. A 30 to 40 g breakfast means you're not trying to find 100 g of protein at 8pm feeling full.

Go for density over volume. Greek yoghurt, cottage cheese, eggs, tinned fish, skyr and lean meat deliver protein without the plate looking intimidating. Volume is the enemy when your stomach is emptying slowly.

Drink some of it when eating is hard. A shake covers 20 to 25 g with almost no volume and no chewing, which on a bad nausea day may be the difference between hitting the target and not. What to look for on a label is in the protein powder guide.

Separate protein from liquid. Filling up on drinks right before a meal costs you the room you needed for the food. Drink between meals instead.

Miss the target on bad days and move on. This is a weekly average, not a daily exam. A run of days well under the range is worth looking at. One is not.

If nausea is severe enough that you can't eat properly for more than a few days, that's a conversation with your prescriber rather than something to push through, and it may be a reason to hold at your current dose rather than step up.

The questions people ask

How would I know if I'm losing muscle?

Without a DXA or a good bioimpedance scan, you can't measure it directly, and home scales that claim to are not precise enough to trust for small changes. The practical proxy is strength. If you can do more press-ups, hold a wall sit longer or carry the shopping further than three months ago while your weight has fallen, that's a reasonable sign. Losing strength quickly alongside the weight is worth raising with your prescriber.

Is 25 to 40% lean mass unique to GLP-1s?

No. That share is broadly what fast weight loss produces by any route, including very low calorie diets and bariatric surgery. What's different about GLP-1s is scale: a lot of people are now losing a lot of weight quickly, so a long-standing feature of rapid weight loss has become a mass-market question.

Do I need protein powder?

No. Food works fine and is generally cheaper per gram. Powder is a convenience tool for the days when volume or nausea is the obstacle, not a requirement.

Is creatine worth taking alongside this?

It has a reasonable evidence base for strength and lean mass alongside resistance training, and it's one of the better-studied supplements out there. There's a fuller look at the evidence and the safety questions in the creatine guide. Get the protein and the training in place first, though. Those are the levers doing most of the work.

This is information, not medical advice. If you have kidney disease or any condition affecting how you handle protein, or if low appetite is stopping you eating properly, talk to your prescriber or a dietitian before making changes.

Start with the number. Work out your protein range in about ten seconds, no sign-up, then track it free in Healthcount so you can see whether you're actually hitting it rather than assuming.