Losing fat without losing muscle comes down to three things: lifting heavy enough to give your body a reason to keep the muscle, eating enough protein to work with, and measuring well enough to know which way things are actually going. The third is where most people come unstuck, because bodyweight cannot tell you what you lost, and several of the tools that claim to are less precise than their marketing suggests.

This guide covers what happens to your body in a deficit, which measurements are worth trusting, how much change is real and how much is noise, and what the research genuinely supports. It also covers what it does not support, because this subject attracts more confident nonsense than almost anything else in fitness.

Expertise reviewed by Mariyan Voykov, Manager and Senior Coach at Body Transformation London. MSc Sports and Exercise Science, BSc Biomedical Science, Level 3 Personal Trainer.

What actually gets lost when you lose weight

You never lose pure fat. Every kilogramme that leaves is a mixture, and the mixture matters.

The long-standing rule of thumb is that roughly a quarter of the weight you lose is fat-free mass and three quarters is fat. That is a useful anchor, but a critical review in Obesity Reviews points out it is not a constant. It shifts with age, activity and how long you have been dieting, and losses of lean tissue during dieting are often smaller than people assume.

What reliably changes the ratio is what you do alongside the deficit. A network meta-analysis of 62 randomised trials covering 4,429 people compared caloric restriction on its own against restriction paired with training. Caloric restriction alone was the only approach that significantly lost lean body mass, at −1.66 kg (95% CI −3.12 to −0.19). Every arm that added training protected lean tissue better, although those individual confidence intervals crossed zero. The honest reading is that dieting alone loses lean mass, rather than that training reliably adds it.

“Lean mass” is not the same thing as muscle

This trips up almost everyone reading a scan for the first time. Fat-free mass on a body composition report includes water, stored glycogen, organs and connective tissue, not just skeletal muscle.

It matters most in the first fortnight. Rapid early weight loss carries a large amount of glycogen and the water bound to it, so a scan taken three weeks into a hard deficit often shows a dramatic drop in “lean mass” that is mostly fluid. People see that number, panic, and change course at exactly the wrong moment. Give it two to three months of scans before drawing conclusions.

Why the scale cannot answer the question

The scale gives you one number combining fat, muscle, glycogen, gut contents and water. Two people can lose the same four kilogrammes over eight weeks, one mostly fat and the other a worrying amount of muscle, and the scale reports them identically.

It also moves for reasons unrelated to progress. A salty dinner, a hard leg session, poor sleep or where you are in your menstrual cycle can shift it more than a week of genuine fat loss.

That is not an argument for ignoring bodyweight. It is an argument for not using it alone. We go through the practical signals in am I losing muscle or fat.

The measurement stack

No single tool answers the question. Each covers a blind spot in the others, and the cheap ones are better than people expect.

1. The tape measure

The most useful measurement here costs almost nothing, and it has the strongest backing from national guidance.

NICE guideline NG246 recommends that adults with a BMI below 35 use waist-to-height ratio alongside BMI as a practical estimate of central fat. NICE’s own phrasing is the clearest instruction in the guideline: keep your waist to less than half your height.

Two things to know. The BMI-below-35 limit is real, and most articles on this get it wrong. And taping is less reliable than it looks, with error between two different measurers running as wide as 15 cm. The full method, and how much change counts as real, is in the waist-to-height ratio guide.

2. The body composition scan

A bioimpedance scan is the tool most gyms lead with, including ours. Ours is an InBody 270. It is useful, and this kind of device is routinely oversold, so here is the straight version.

Against DEXA, the reference method, it carries a real bias. In the one study comparing that model with DXA, in 88 adults, body fat read 24.6% against DXA’s 27.7%, fat mass 1.9 kg low and fat-free mass 2.6 kg high, all at p<0.0001 (Garcia JR et al., Journal of Exercise and Nutrition, 2020). In short, about 3.1 percentage points low on body fat, and the direction is the same across every model tested: underestimates fat, overestimates fat-free mass. NICE is direct about it too: do not use bioimpedance as a substitute for BMI.

So why use one? Because absolute accuracy is not the job. Repeatability is, and on that these devices do well. Bioimpedance test-retest reliability under controlled conditions is 0.998 or better (Looney DP et al., Frontiers in Nutrition, 2024), measured on other devices, since no test-retest study of the 270 itself has been published. Because the bias is systematic it largely cancels out when you subtract one scan from another. The number may be a few points off, but the direction and size of the change between two properly standardised scans is meaningful.

Three limits worth holding onto, and they are specific to this model. Muscle readings are its strongest output and fat percentage is weaker. It reports no visceral fat at all, so nobody should be quoting you one from it: total adiposity comes from percent body fat, BMI and waist-to-height ratio instead. And it reports no segmental fat, so it cannot tell you where you carry fat. What it does give you is a five-segment breakdown of lean mass, which is a genuine strength, and a good way to spot a left-to-right or upper-to-lower imbalance worth training around.

There is a fourth limit that matters more than any of them. With only two measurement frequencies, the 270 cannot report the ratio of extracellular to total body water, which is the standard check on a higher-specification machine for whether a sudden lean-mass jump was fluid rather than tissue. Without it, a hydration swing looks exactly like real change. That makes standardising the conditions more important on this device, not less.

The question nobody in the industry wants to answer is how much change is real. Bioimpedance cannot reliably resolve changes under roughly 1.5 to 2 kg (Kyle UG et al., Clinical Nutrition, 2004), and for body fat percentage a shift under roughly 2 to 3 percentage points cannot be confidently separated from noise: minimum detectable change came out at 2.12 to 2.73 points (McLester CN et al., Journal of Clinical Densitometry, 2020). That study tested the InBody 230, 720 and 770 rather than the 270, but the 230 shares the 270’s exact measurement engine, two frequencies at 20 and 100 kHz through eight-point tetrapolar electrodes, so read it as a reasoned inference rather than a finding on this model. So if anyone tells you a scan shows you gained 0.4 kg of muscle this month, they are reading noise and selling it as progress. The precision figures are in DEXA vs InBody.

Standardisation is what makes repeat scans mean anything: same time of day, same hydration state, same conditions every time. Roughly 830 ml of a glucose solution shifted a reading by about 1.2 percentage points of body fat within 20 minutes, which is the same size as a month of real progress. The full protocol is in is an InBody scan accurate, and the line-by-line read of the printout is in how to read your InBody results. Our own service page is here.

3. The gym logbook

The cheapest and most sensitive early-warning system you own is your training log.

Muscle you are losing shows up as strength you cannot hold, usually before any scan detects it. If your working weights drift down across three or four weeks in a deficit, that is information, and it arrives faster and with less measurement error than anything a machine will tell you. If your loads hold steady or creep up while your waist shrinks, that is the outcome you wanted, whatever the scale says.

4. Function

Body composition is a proxy. What matters is whether you can do things.

In a study of 139,691 adults across 17 countries, every 5 kg reduction in grip strength was associated with an all-cause mortality hazard ratio of 1.16 (95% CI 1.13 to 1.20), making grip a stronger predictor of mortality than systolic blood pressure. That is an association rather than a cause, and squeezing a gripper is not an intervention. Grip strength is a marker of how robust you are overall.

Simple functional tests are cheap, standardised and far less noisy than a scan printout. More in the grip strength guide.

5. Resting metabolic rate

Most coaches estimate resting metabolic rate from an equation, and for many people that is fine. Not for everyone. Worth knowing that the BMR box on a body composition printout is an equation too: the manufacturer’s training manual for our model states the device uses the Cunningham equation, a regression from lean body mass, so that figure is an estimate rather than a measured calorie target. The Mifflin-St Jeor equation is unbiased on average, yet in a study of 125 women it predicted only 71% of them within 10% of their measured value, and accuracy degrades further in people with obesity, which is precisely the group most likely to be receiving the advice.

The honest limit: no trial has shown that measuring resting metabolic rate rather than estimating it improves your results. What it does is remove a guess for the roughly three people in ten the equation gets wrong. See RMR and metabolic testing.

6. Blood markers

This is where a personal trainer’s job stops, and we would rather say so plainly than blur it.

Blood testing should be arranged and interpreted by an appropriately qualified clinician, with results going to your GP. Coaches can note what a flagged result means for how you train, and refer onward. We do not diagnose, we do not order tests, and we do not interpret them. “Dietitian” is a legally protected title in the UK and our coaches are coaches, not dietitians.

It is also worth knowing that the “comprehensive metabolic panel” sold widely online is not what current UK guidance recommends. NICE advises against using routine liver blood tests to rule out fatty liver disease, and against testing thyroid function without clinical suspicion. UK biochemistry guidance says vitamin D testing is not indicated for tiredness or non-specific aches when bone biochemistry is normal. What a UK adult aged 40 to 74 is actually offered is the NHS Health Check every five years, covering weight, waist, blood pressure, cholesterol and possibly blood sugar.

How to train for losing fat without losing muscle

The evidence here is stronger and simpler than the measurement evidence.

A systematic review of 114 trials covering 4,184 people found resistance training combined with caloric restriction was the most effective approach for reducing body fat percentage (effect size −3.8%, 95% CI −4.7 to −2.9), and that resistance training was the most effective single method for increasing lean mass compared with untrained controls (0.8 kg, 95% CI 0.6 to 1.0; Lopez P et al., Obesity Reviews, 2022). The finding held across ages and both sexes.

Do not drop cardio to do it. In a trial of 160 older adults in the New England Journal of Medicine, all exercising groups lost around 9% of bodyweight, but combined aerobic and resistance training improved physical function by 21% against 14% for either alone, and strength rose 18 to 19% in the groups that lifted against 4% in the aerobic-only group.

The principles we programme around:

• Chase load, not fatigue. In a deficit, maintaining or adding weight on the bar is the signal you are keeping muscle. Grinding yourself into the floor to burn calories is not.
• Trim volume before intensity. Recovery is impaired when energy is low. Fewer hard sets, kept heavy, beats many mediocre ones.
• Prioritise compound movements. Time and recovery are limited. Spend them where they cover the most tissue.
• Autoregulate. Some days in a deficit are poor. Adjust the session rather than abandon the week.

Full detail in strength training during weight loss and our strength training programmes.

How to eat, without the exaggeration

Higher protein during a deficit helps preserve lean tissue. It helps less than the internet claims.

The clearest evidence is a meta-analysis of 24 trials covering 1,063 people, comparing higher-protein diets at around 30% of energy against standard diets at 12 to 18%. Higher protein produced 0.87 kg more fat loss and 0.43 kg more fat-free mass retained. That lean tissue benefit is about four hundred grammes. It is real, statistically solid, and nothing like the transformation protein marketing implies. Over a long programme it is worth having. It is not magic.

The ranges the literature uses sit between roughly 1.0 and 1.6 g per kg of bodyweight per day. The European Association for the Study of Obesity recommends a minimum of 1.0 to 1.2 g/kg for older adults, spread across three meals at 25 g or more each, alongside moderate-to-high-intensity resistance training. The British Nutrition Foundation puts the principle simply: adequate protein and strength-based activity are essential to help maintain muscle mass.

More in protein to maintain muscle while losing weight and our nutrition coaching.

What happens when a fast weight-loss phase ends

Whatever route you took to the deficit, the end of a rapid loss phase is where most progress is lost, and it is the least planned-for part of the process.

A systematic review and meta-regression of 48 studies covering more than 3,200 people found roughly 60% of lost weight was regained within 52 weeks of stopping, with regain plateauing around 60 weeks and a model projecting about 25% of the original loss retained long term. Carry the caveats with that figure: the plateau is extrapolated beyond the data, only six trials informed the modelling, and most included studies carried a moderate risk of bias. Note also what it does not say. “You put it all back on” is not what the research shows.

Two popular claims deserve correcting, because both are wrong in ways that change what people do.

“You regain fat, not muscle.” There is no evidence for this. The composition of regain has not been established, and the researchers say so directly. The best available proxy, from 43 studies covering 2,379 people losing weight through diet and exercise, found the fat-free-mass fraction of regain at 21.6% was almost identical to the fat-free-mass fraction of the original loss at 19.6%.

“Your metabolism is permanently broken.” The famous figure of roughly 500 kcal per day of metabolic adaptation comes from 14 people from a televised competition, and it is an explicit outlier. A systematic review of 33 studies covering 2,528 people put typical adaptation at around 30 to 100 kcal per day, and found it may be attenuated or absent once weight has stabilised. Adaptation is real. It is generally far smaller than the headline, and whether it drives regain is genuinely disputed.

Our guide to keeping weight off after rapid weight loss covers what to do about it week to week.

What the evidence does not show

Any page selling you measurement and coaching should be straight about the limits of its own argument. Ours are these.

No trial has proved that preserving lean mass causes better long-term weight maintenance. Fat-free-mass loss does predict later regain, but the effect weakened after adjustment and fat mass loss was the stronger predictor. The relationship is observational.

Exercise for maintenance is contested. An overview of 12 systematic reviews and 149 studies found no significant effect of exercise on weight maintenance, although resistance training did preserve lean mass during loss. A more recent meta-analysis of 11 trials found 2.81 kg less regain with exercise during maintenance. Both are real findings and they disagree.

Nobody has run the decisive trial for people losing weight under medical supervision. The training and protein recommendations above are extrapolated from studies of diet-led weight loss. A UK trial is currently recruiting to test them directly. Until it reports, anyone claiming certainty here is going beyond the evidence.

Measuring your metabolic rate has not been shown to improve outcomes. It removes a guess. That is the only supportable claim.

If someone tells you this field is settled, they have not read it.

How we do it at BTX

Our position is straightforward: coached and measured, rather than guessed at. That means a baseline before anything is decided, a training block built around holding strength while fat comes off, a protein target you can hit in a working week, and a re-measure every three to four weeks under identical conditions so the numbers mean something. That interval is for accountability and for building the trend line. The judgement gets made on the shape of that line across two to three months, never on the gap between two consecutive scans, which sits inside the measurement error.

It also means telling you when a number is noise. A coach who reacts to every wobble on a scan is not reading data, they are guessing with extra steps.

If you want this measured properly and programmed around, book a call and we will start with a baseline.

Free InBody scan until 30 September

We are running complimentary body composition scans at our Finchley Road studio, Monday to Friday between 10am and 2pm. Thirty minutes: the scan itself, plus a coach taking you through what your numbers mean and which ones to ignore. No charge, no obligation, one per person.

Book your free scan or see what’s included.

This article is general information, not medical advice. If you have a medical condition, are pregnant, are taking any prescribed medication, or have a history of disordered eating, speak to your GP or a registered dietitian before changing how you train or eat.

Frequently asked questions

Can you lose fat and build muscle at the same time?
The evidence does not settle it. Resistance training plus caloric restriction was the most effective combination for reducing body fat percentage across 114 trials, and in the 62-trial network meta-analysis every training arm protected lean mass better than restriction alone, but those confidence intervals crossed zero. Holding onto the muscle you have is the defensible goal in a deficit.

How do I know if I am losing muscle?
Strength is the fastest signal. If your working weights fall steadily over three or four weeks in a deficit, that deserves attention. Scans help over longer periods, but they cannot resolve changes under roughly 1.5 to 2 kg, so a single scan will not answer the question.

How much protein do I need to keep muscle while losing weight?
The literature uses roughly 1.0 to 1.6 g per kg of bodyweight per day, spread across meals at 25 g or more each. The measured benefit of higher protein over standard intake during a deficit was about 0.43 kg of retained fat-free mass across 24 trials.

How often should I get a body composition scan?
Every three to four weeks is a good interval for accountability, but judge progress on the trend across two to three months rather than on any single gap between scans. What matters more than frequency is that every scan follows the same protocol, so the line is comparable.

Is BMI useless if I lift weights?
BMI is a population screening tool rather than an individual assessment, and it does not distinguish muscle from fat. NICE recommends using waist-to-height ratio alongside it for adults with a BMI below 35, and specifically advises against substituting bioimpedance for it.

Sources

• Heymsfield SB et al. Obesity Reviews, 2014. https://onlinelibrary.wiley.com/doi/abs/10.1111/obr.12143
• Xie Y et al. Frontiers in Nutrition, 2025;12:1579024. https://www.frontiersin.org/journals/nutrition/articles/10.3389/fnut.2025.1579024/full
• Lopez P et al. Obesity Reviews, 2022. https://onlinelibrary.wiley.com/doi/10.1111/obr.13428
• Villareal DT et al. New England Journal of Medicine, 2017;376:1943-1955. https://www.nejm.org/doi/full/10.1056/NEJMoa1616338
• Wycherley TP et al. American Journal of Clinical Nutrition, 2012.
• Kim JE et al. Nutrition Reviews, 2016;74(3):210-224. https://academic.oup.com/nutritionreviews/article/74/3/210/1825906
• Di Vincenzo O et al. Obesity Facts, 2026;19(4):422-437. https://karger.com/ofa/article/doi/10.1159/000549751/940127/
• British Nutrition Foundation. Summary statement on weight loss and muscle mass, 2025.
• Garcia JR, Manimaleth R, Czartoryski P et al. Journal of Exercise and Nutrition, 2020;3(3):10. https://journalofexerciseandnutrition.com/index.php/JEN/article/view/65
• McLester CN, Nickerson BS, Kliszczewicz BM, McLester JR. Journal of Clinical Densitometry, 2020;23(3):443-450.
• Looney DP et al. Frontiers in Nutrition, 2024;11:1491931. https://doi.org/10.3389/fnut.2024.1491931
• Kyle UG et al. Clinical Nutrition, 2004. https://www.espen.org/documents/BIA1.pdf
• Schierbauer J et al. Metabolites, 2023;13(4):473. https://doi.org/10.3390/metabo13040473
• NICE. Overweight and obesity management, NG246, 2025 (updated 2026). https://www.nice.org.uk/guidance/ng246/resources/overweight-and-obesity-management-pdf-66143959958725
• NICE. Non-alcoholic fatty liver disease, NG49. https://www.ncbi.nlm.nih.gov/books/NBK384748/
• NICE. Thyroid disease: assessment and management, NG145. https://www.ncbi.nlm.nih.gov/books/NBK550859/
• NHS. NHS Health Check. https://www.nhs.uk/conditions/nhs-health-check/
• Thom G et al. Journal of Nutritional Science, 2020;9:e17. https://doi.org/10.1017/jns.2020.11
• Frankenfield DC. Clinical Nutrition, 2013. https://doi.org/10.1016/j.clnu.2013.03.022
• Leong DP et al. The Lancet, 2015;386(9990):266-273. https://doi.org/10.1016/S0140-6736%2814%2962000-6
• Budini B, Luo S. eClinicalMedicine, 4 March 2026.
• Turicchi J et al. Obesity Reviews, 2019. https://doi.org/10.1111/obr.12849
• Nunes CL et al. British Journal of Nutrition, 2022;127(3):451-469.
• Bellicha A et al. Obesity Reviews, 2021. https://doi.org/10.1111/obr.13256
• Wang J et al. Scientific Reports, 2026. https://www.nature.com/articles/s41598-026-57804-8

{“@context”:”https://schema.org”,”@type”:”FAQPage”,”mainEntity”:[{“@type”:”Question”,”name”:”Can you lose fat and build muscle at the same time?”,”acceptedAnswer”:{“@type”:”Answer”,”text”:”The evidence does not settle it. Resistance training plus caloric restriction was the most effective combination for reducing body fat percentage across 114 trials, and in the 62-trial network meta-analysis every training arm protected lean mass better than restriction alone, but those confidence intervals crossed zero. Holding onto the muscle you have is the defensible goal in a deficit.”}},{“@type”:”Question”,”name”:”How do I know if I am losing muscle?”,”acceptedAnswer”:{“@type”:”Answer”,”text”:”Strength is the fastest signal. If your working weights fall steadily over three or four weeks in a deficit, that deserves attention. Scans help over longer periods, but they cannot resolve changes under roughly 1.5 to 2 kg, so a single scan will not answer the question.”}},{“@type”:”Question”,”name”:”How much protein do I need to keep muscle while losing weight?”,”acceptedAnswer”:{“@type”:”Answer”,”text”:”The literature uses roughly 1.0 to 1.6 g per kg of bodyweight per day, spread across meals at 25 g or more each. The measured benefit of higher protein over standard intake during a deficit was about 0.43 kg of retained fat-free mass across 24 trials.”}},{“@type”:”Question”,”name”:”How often should I get a body composition scan?”,”acceptedAnswer”:{“@type”:”Answer”,”text”:”Every three to four weeks is a good interval for accountability, but judge progress on the trend across two to three months rather than on any single gap between scans. What matters more than frequency is that every scan follows the same protocol.”}},{“@type”:”Question”,”name”:”Is BMI useless if I lift weights?”,”acceptedAnswer”:{“@type”:”Answer”,”text”:”BMI is a population screening tool rather than an individual assessment, and it does not distinguish muscle from fat. NICE recommends using waist-to-height ratio alongside it for adults with a BMI below 35, and specifically advises against substituting bioimpedance for it.”}}]}