Can you lose fat and build muscle at the same time? The evidence says yes.

The old rule says you have to choose: eat more to build muscle, eat less to lose fat, never both at once. Controlled trials say otherwise. Losing fat while gaining muscle, known as body recomposition, has been measured in beginners, in people returning to training, in people carrying extra fat, and in men treated for genuine testosterone deficiency. It is harder for the lean and the highly trained, and the conditions matter.

The rule has a logic to it. Building tissue costs energy, and a calorie deficit means the body is short of energy. So, the reasoning goes, a deficit must mean losing muscle along with fat.

The flaw is that the body is not short of energy when it is carrying stored fat. It draws on that store to cover the gap, while muscle responds to two signals that work regardless of the deficit: hard training and enough protein. Give muscle both signals and it can grow while fat shrinks underneath it.

This matters more after 45 than at any other age. Muscle is the tissue people lose first and rebuild most slowly, and it carries strength, balance, blood sugar control and the capacity to recover from illness. A plan that takes off 20 pounds without asking what those pounds were made of has done half the job. A plan that takes off fat and keeps or adds muscle is the one worth paying for.

What the trials show

The cleanest demonstration is also the most extreme. In a trial at McMaster University, 40 young men ate about 40% fewer calories than they burned for four weeks while lifting and doing interval training six days a week. The group eating 2.4 g of protein per kg of body weight a day gained 1.2 kg of lean mass and lost 4.8 kg of fat. The group eating half that protein held their lean mass steady and lost 3.5 kg of fat (Longland, AJCN 2016). Same deficit, same training: the protein decided whether muscle grew.

Nobody over 45 should copy that protocol. Its value is the principle: a deficit does not, by itself, rule out building muscle. The rest of the literature shows how far the principle reaches.

• Beginners and people returning to training. Muscle that has never been trained, or has not been trained in years, responds fastest. These are the people for whom recomposition is most reliable, and a 2020 review found it documented in trained lifters as well, provided training is progressive and protein is high (Barakat, Strength Cond J 2020).
• Older adults losing weight. Across six randomized trials in older adults with obesity, adding resistance training to a calorie-restricted diet prevented about 94% of the lean mass the diet alone would have cost, with the same fat loss (Sardeli, Nutrients 2018).
• Women after menopause. Across 101 trials and 5,697 postmenopausal women, resistance training added about 0.9 kg (2 lb) of fat-free mass and lowered body-fat percentage by about 1.2 points (Khalafi, Front Endocrinol 2023).
• Men treated for low testosterone. Covered below: in a year-long trial, testosterone changed what the weight loss was made of.

The pattern runs one way. The more fat someone carries and the less trained they are, the more room there is to lose fat and gain muscle at once. The leaner and more experienced someone is, the slower it goes, and at some point a dedicated phase of eating more makes more sense. For most people reading this, that point is a long way off.

The four levers

Recomposition is not a special diet. It is four ordinary levers set at the same time, and it fails when any one of them is left out.

1. Training that progresses. Muscle grows when it is asked to do more than it can comfortably do now. For adults over 45 that usually means two or three full-body sessions a week built on multi-joint lifts, with loads that rise over weeks (Fragala, NSCA 2019). Walking and cardio help the heart and the fat loss, but they do not send the signal that builds muscle. Our piece on strength training after 50 covers the dose in detail.

2. Protein, more than most people eat. In trials of resistance training, extra protein added to the gains up to about 1.6 g per kg of body weight a day, with little benefit above that on average (Morton, Br J Sports Med 2018). In a deficit, and in older muscle, the need sits at the higher end. Men around 71 needed about 0.4 g per kg at a meal to get the full muscle-building response, compared with 0.24 g per kg in men around 22 (Moore, J Gerontol 2015). In practice that is 30 to 40 g at each meal, not a large dinner and a coffee for breakfast. Kidney disease changes these numbers, so that conversation comes first.

3. A modest deficit, not a crash. A review of training studies found that eating less blunted lean-mass gains while strength still improved, and that a deficit of about 500 calories a day was, on average, the point at which lean mass stopped rising (Murphy and Koehler, Scand J Med Sci Sports 2022). The more aggressive the diet, the harder the muscle has to fight. People carrying more fat can tolerate a larger gap; lean people need a smaller one.

4. Sleep. In a small study of ten adults on the same diet for two weeks, cutting sleep from 8.5 to 5.5 hours reduced the fat lost by 55% and increased the lean mass lost by 60% (Nedeltcheva, Ann Intern Med 2010). Same calories, different body. Short sleep shifts the loss from fat toward muscle, which is why our piece on why you feel heavier, tired and sore starts with sleep.

Creatine monohydrate is the one supplement with consistent evidence here. Taken alongside training, it adds a small amount of lean mass in older adults, about 1.2 to 1.4 kg in pooled trials (Chilibeck, Open Access J Sports Med 2017). It works only with the training, not instead of it.

Where hormones fit

Hormones are the lever people ask about first and the one that matters least for most of them. For a specific group of men, though, the effect is large.

In a year-long Australian trial, 100 men with obesity and low or borderline testosterone, median age 53, went on a very-low-calorie diet for ten weeks and then a maintenance phase, with injections of testosterone or placebo throughout (Ng Tang Fui, BMC Med 2016). Both groups lost about the same weight. What the weight was made of differed. By the end, the testosterone group had lost 2.9 kg more fat, including more of the visceral fat around the organs, and kept 3.4 kg more lean mass. The placebo group lost about 4 kg of lean mass over the year; the testosterone group lost about half a kilogram.

Two cautions keep that result honest. These were men whose testosterone was already low, so it says nothing about adding testosterone to normal levels. And weight loss raises testosterone by itself in men with obesity, in proportion to the weight lost (Corona, Eur J Endocrinol 2013). A man with borderline levels at 260 pounds may not have them at 220. Diagnosis comes first, with symptoms and two early-morning tests, and a recheck after meaningful weight loss. Our piece on raising testosterone naturally covers that sequence.

For women the picture is different. Menopause shifts body composition toward fat and away from lean tissue, which makes training and protein more important, not less. Menopause hormone therapy is prescribed for symptoms and to protect bone; in pooled studies it was linked to about 5% greater strength, which helps, but it is not a muscle-building treatment (Greising, J Gerontol 2009). Testosterone for women is supported for one use, low sexual desire after menopause, at doses that keep levels within the female range, and at those levels it does not build muscle (Global Consensus, JCEM 2019). For women, the recomposition levers are the four above.

On a GLP-1, the question changes

Medications like tirzepatide (Zepbound) have made large weight losses common, which turns recomposition from a fitness goal into a medical one. In the body-scan substudy of SURMOUNT-1, about a quarter of the weight lost on tirzepatide was lean mass, and the share was almost identical in adults under 50, 50 to 64 and 65 or older (Look, Diabetes Obes Metab 2025). The semaglutide trial STEP 1 reported closer to 40%, though the two trials differed in design and are not a head-to-head comparison.

Lean mass on a scan is not all muscle; it includes water, organs and other tissue, and some loss is expected with any large weight loss. The goal is not zero. It is to make the share as small as possible, with the same four levers: resistance training from the first dose, protein at every meal even when appetite is low, a pace of loss the muscle can keep up with, and sleep. The Athlos program goal is that no more than about a fifth of the weight lost is lean tissue. It is the target each plan is measured against on every scan, not a promise of any one person’s result.

That is the subject of its own piece, coming next in this series. The short version: on a GLP-1, the medication handles the appetite, and the plan has to handle the muscle. Our medical weight loss page describes how that is monitored.

How to tell it’s working

Recomposition hides from the bathroom scale. A person who loses 4 kg of fat and gains 1.5 kg of muscle has lost 2.5 kg, a number that looks like a disappointing month. This is the most common reason people quit a plan that is working.

The scale measures everything at once: fat, muscle, water, bone and last night’s dinner. To see recomposition you have to measure the parts separately.

• DXA is the reference standard. A low-dose X-ray scan separates fat, lean tissue and bone, region by region, and shows the fat around the organs.
• Bioimpedance scales and handheld devices estimate composition from how an electrical signal moves through the body. They are convenient but swing with hydration, and are best used for trends on the same device under the same conditions.
• Tape and strength cost nothing and move in the right direction when recomposition is happening: a smaller waist at the same weight, and loads rising in the gym.

The numbers that matter are fat mass, lean mass and the share of any weight lost that came from each. Strength is the check that the lean mass is muscle that works. A plan built on those numbers can tell success from stalling in weeks rather than months, and can change course before the muscle is gone.

What the research does not show

• The headline trial was young men for four weeks. Longland’s result came from men in their twenties under close supervision, on a deficit and a training load few people over 45 should attempt. It proves the principle; it does not set the dose for an older reader.
• Many of the changes are small next to the measurement. A gain of a kilogram of lean mass sits close to the error of even a good body scan, and lean mass includes water and stored carbohydrate as well as muscle. That is why strength is measured alongside it.
• The evidence in trained lifters is thinner. The 2020 review that found recomposition in trained people drew on fewer and smaller studies than the work in beginners.
• The 500-calorie figure is an average. It came mostly from studies of people who were not already trained, and the right deficit for any one person depends on how much fat they carry.
• The testosterone result applies to men whose levels were low. It was one trial, with injections, over a year. It says nothing about men with normal levels, and long-term safety is a separate question covered in our men’s health pieces.
• The sleep study was ten people for two weeks. The direction fits everything else known about sleep and muscle; the size of the effect is uncertain.
• Few trials have tested strength programs on top of the newer weight-loss drugs. The case for training and protein on tirzepatide rests largely on older trials of diet plus exercise and on what the drug trials measured without a structured program.

None of this weakens the central finding, which comes from randomized trials across ages and both sexes: with progressive training and enough protein, people can lose fat while keeping, and often adding, muscle.

Where that leaves you

You do not have to choose between losing fat and building muscle, and after 45 you should not. The plan that does both is unglamorous: lift two or three times a week with loads that rise, eat protein at every meal, keep the deficit modest, and protect your sleep. Then measure what the weight is made of, not just what it is.

If you are a man carrying weight around the middle with symptoms of low testosterone, test properly before you treat, and test again after the weight starts to come off. If you are on a GLP-1 or considering one, the levers matter more, not less, because the drug makes it easy to lose muscle without noticing.

The Athlos Transformation Program is built on exactly this: a measured baseline, coached strength training three times a week at our partner facility, protein targets from the first week, medication only when a physician finds it indicated, and fat and lean mass tracked separately every step of the way. The free Transformation Assessment is the place to start.

Written by Athlos Health, Torrance, California, and medically reviewed by Christopher Traughber, M.D. This article is general education, not medical advice. Testing and treatment are recommended only after an individual evaluation, and the risks of any treatment are discussed in consultation and in writing. Every figure above is linked to its source below.