How to read your InBody results comes down to one habit: attend to the few numbers the machine measures well, and know what is not on the sheet at all. Skeletal muscle mass, lean body mass and total body water are the trustworthy end of the printout. Body fat percentage is useful as a trend but carries a real offset against DEXA. The Segmental Lean Analysis is more useful than most people realise, and the Body Composition History strip along the bottom is the single most useful thing on the page. Our device is an InBody 270, and it prints no visceral fat, no segmental fat and no hydration ratio, so a good deal of the advice floating around about “which InBody numbers to ignore” describes a machine you were never standing on.

Expertise reviewed by Giuseppe Marchesani, Senior Coach at Body Transformation London. Level 3 Personal Trainer.

The machine passes a current through you at two frequencies, 20 and 100 kHz, and measures resistance. Everything on the printout comes from that reading plus your height, age and sex. Only your weight was actually weighed. The rest was inferred, and some inferences are far sounder than others. For the wider picture, start with our pillar guide to losing fat without losing muscle.

How to read your InBody results, line by line

These are the sections that appear on the sheet, in the order they appear on it.

Body Composition Analysis

The top block, and the foundation of everything below it: total body water, dry lean mass, body fat mass and weight, stacked so you can see how they add up. Total body water is the raw material of the whole estimate, so a large swing there explains a large swing in everything else. Dry lean mass is protein and mineral, the part that is not water and not fat.

ESPEN puts typical prediction error for fat-free mass at 3.5 to 6% (Kyle et al., 2004), so read this block as an estimate with a margin, not a set of weighed quantities.

Muscle-Fat Analysis

Three bars: weight, skeletal muscle mass and body fat mass, each plotted against a normative range referenced to your age and sex.

Skeletal muscle mass is the best number on the sheet. Against DEXA in 84 older adults, appendicular skeletal muscle index agreed almost perfectly, intraclass correlation 0.965 (Buch et al., 2022). Anchor your tracking here. It is still estimated largely from body water, so it moves with hydration and glycogen, not only with contractile tissue.

The bars themselves are a comparison with other people. The change in your own bar between scans is the part that concerns you.

Obesity Analysis

BMI and percent body fat. Body fat percentage is the number everyone looks at first and the one that deserves the least emotional weight. In the one study comparing this model with DXA, 88 adults read 24.6% body fat on the InBody against 27.7% by DXA, a gap of about 3.1 percentage points, with fat mass under-read by about 1.9 kg and fat-free mass over-read by about 2.6 kg, all at p<0.0001 (Garcia et al., 2020).

So treat the percentage as a point on your own private scale, not a number to set beside a friend’s DEXA result. Read it as a trend across scans on the same device, and expect it to sit below what a DEXA would say.

BMI is on the sheet because it belongs there. NICE says explicitly not to use bioimpedance as a substitute for BMI in adults (NICE NG246, 1.9.9).

What is not here, and this matters: there is no visceral fat figure on this model. It does not measure or report visceral fat area or visceral fat level, so nobody should be quoting you one from it. Total adiposity has to be read from percent body fat, BMI and waist-to-height ratio instead. For that last one NICE recommends keeping your waist under half your height in adults with a BMI below 35 (NG246, 1.9.15). A tape measure answers the question this sheet cannot, and costs nothing.

Segmental Lean Analysis

Five segments, left arm, right arm, trunk, left leg and right leg, each given in kilogrammes and as a percentage of what would be expected, referenced to your age and sex. This is a genuine strength of the device, and the part clients most often skim.

Used properly it answers two questions. Is there a left-to-right asymmetry worth programming around, often after an old injury or a one-sided sport. And is your upper body keeping pace with your lower body, which for most desk-based clients it is not.

One caution. Segmental figures drift across a single day as fluid moves out of the legs into the trunk and arms: hand-to-hand impedance rose 27.2 Ω and foot-to-foot fell 31.8 Ω between morning and evening, while whole-body impedance barely changed (Oshima and Shiga, 2006). Compare these numbers morning to morning, or not at all.

There is no segmental fat analysis on this model. The five segments are lean tissue only, so this sheet cannot tell you where you carry fat. Any provider reading a fat distribution off this sheet is inventing it.

Body Composition History

The strip showing your last 10 tests. If you take one thing from this page, take this: the history is the most useful section on the sheet, because every honest use of the device is a comparison rather than a snapshot.

A single scan tells you where you are, with a margin of error you cannot fully quantify. Ten scans under the same protocol tell you the direction and rough rate of travel, which is the question you actually came in with. Read the line, not the last dot on it.

That is also the argument for scanning often. We rescan every three to four weeks, so a full history strip is roughly thirty weeks of trend printed on one page. Ten dots make a line you can read. Any two adjacent dots do not.

Body Fat-Lean Body Mass Control, and Lean Body Mass

Lean body mass is everything that is not fat: muscle, water, glycogen, bone, organs and connective tissue. It is not a synonym for muscle, and treating it as one is the commonest misreading of the sheet. On the 270 it is over-read against DXA by about 2.6 kg (Garcia 2020).

The Control figures are the machine’s own suggested adjustment to your fat mass and lean mass, generated by comparing you with a population norm. They are arithmetic, not a plan, and they take no account of your training age, your schedule or what you are willing to do for the next year. Treat them as a conversation starter.

Basal Metabolic Rate

Not measured. Calculated. The official InBody 270 training manual states that the device uses the Cunningham equation, a regression based on how much lean body mass you have. So this box is a population equation applied to an estimated input, which is worth knowing before you build a calorie target on it.

Prediction equations in general struggle at the individual level. The widely used Mifflin-St Jeor equation put only 71% of women in a 125-participant study within ±10% of their measured resting metabolic rate (Thom et al., 2020), and accuracy falls further in those carrying more body fat (Frankenfield, 2013). Roughly three people in ten sit outside 10%, enough to break a calorie target. No trial has shown that measuring metabolic rate improves results either.

The segmental impedance table

The raw resistance values at each frequency for each segment, printed at the foot of the sheet. It is there for quality control rather than for you. If a coach ever wants to check whether a strange result came from poor electrode contact, this is where they look.

One thing you will not find, and why it matters

There is no extracellular-to-total-body-water ratio on this sheet. The lowest frequency it uses, 20 kHz, is too high to isolate the fluid sitting outside your cells, so the machine cannot flag fluid retention, swelling or an inflammatory response.

On a higher-specification InBody, a rising ratio is the standard explanation when someone’s lean mass appears to jump two kilogrammes in a week. Here that check is unavailable, so a hydration-driven swing looks exactly like a real change in lean mass, with nothing on the page to contradict it. That is precisely why the standardisation protocol below is not optional busywork on this device. It is doing a job the hardware cannot do for you.

Which numbers deserve your attention?

Two rules govern the whole table. ESPEN puts changes under 1.5 to 2 kg inside the noise of bioimpedance (Kyle 2004), so a difference in mass smaller than that is not yours to celebrate. And for body fat percentage, a shift of less than roughly 2 to 3 percentage points cannot be confidently told apart from measurement noise on this class of device: minimum detectable change came out at 2.12 to 2.73 points (McLester 2020). That study tested the InBody 230, 720 and 770 rather than the 270, but the 230 runs the identical measurement engine to ours, two frequencies at 20 and 100 kHz through eight-point tetrapolar electrodes, so we treat it as a reasoned inference rather than a direct finding on this model. Worth adding that the same study found “the research grade models, InBody720 and InBody770, added minimal benefit over the portable InBody230.” How the scanner stacks up against DEXA for measuring change, and when a DEXA scan is worth paying for, is set out in DEXA vs InBody scan.

Does a workout change your InBody reading?

Yes, and more importantly it makes the reading less comparable to your last one. How far that pushes the number, and what it means for accuracy in general, is covered in is an InBody scan accurate. Fluid is what the machine is most sensitive to: about 832 mL of a glucose solution raised measured body fat by roughly 1.2 percentage points within 20 minutes, and a control group drifted up 0.7 points over two hours with no intervention at all (Schierbauer et al., 2023). A study that deliberately broke every rule found smaller differences, −1.9 to +0.4 points and none significant, but used consumer devices and was underpowered (Randhawa et al., 2021).

In practice a broken protocol moves a reading by roughly one to two percentage points of body fat. That is minor beside the offset against DEXA, but it is about the size of a month of honest work, so a rescan taken under different conditions can invent a result or bury one.

If you want rescans to mean anything, run every one of them to the same standardisation protocol, which we set out in full in our guide to InBody accuracy.

What is a good skeletal muscle mass percentage?

There is no defensible single answer, and anyone quoting a clean target range is quoting a population comparison, not a clinical standard. Bioimpedance equations are validated to about 34 kg/m², and ESPEN concluded the method is “more suitable for population groups than for individuals” (Kyle 2004).

Function does have agreed thresholds. The European consensus on sarcopenia flags grip strength below 27 kg in men and 16 kg in women, five chair stands over 15 seconds, or gait speed of 0.8 m/s or slower (Cruz-Jentoft et al., 2019). So ask instead whether your skeletal muscle mass is holding or rising while you get stronger. That one you can answer.

How often should you do an InBody scan?

Scan often, and read the line rather than the last point.

Every three to four weeks is a good interval, and it is the one we work to. Not because three weeks resolves a real change, because it does not, but because it keeps you accountable, keeps the habit intact, and adds another point to the trend line. What a short gap cannot do is settle an argument on its own. A single three-week gap sits inside the measurement noise: on this class of device a body fat change under roughly 2 to 3 percentage points, or a mass change under 1.5 to 2 kg, cannot be confidently separated from error.

So the value of frequent scanning is the shape of the trend across two to three months, not the difference between consecutive scans. That is precisely what the Body Composition History is for. The sheet stores and prints your last 10 tests, and ten three-weekly scans is about thirty weeks of trend on one page. The line is the evidence. The gap between scan nine and scan ten is not.

If you are in a fast loss phase, or your weight loss is overseen by a clinician, the changes are larger, so more of each gap clears the noise floor. The rule does not change: judge the line.

One thing a scan is not: a replacement for BMI. NICE NG246 (1.9.9) says not to use bioimpedance as a substitute for BMI, which we unpack in is an InBody scan accurate. Your scan sits alongside BMI and waist-to-height ratio, not above them.

What to actually do with this

1. Write down your skeletal muscle mass and body fat percentage. Those two carry most of the information.
2. Look at the body composition history before you look at today’s numbers. The line matters more than the dot.
3. Check the segmental lean columns for a left-to-right gap or an upper-to-lower gap, and take them as a programming prompt rather than a verdict.
4. Measure your waist the same day, on skin, at the end of a normal breath out, and divide by your height. That is where central adiposity comes from, because this sheet has no visceral fat figure.
5. Act on nothing that sits inside the noise floor. Wait for the next scan.
6. Track what the machine cannot fudge: what you lift, for how many reps, and how you move.

If muscle mass holds or rises while fat mass falls and your lifts go up, the programme is working. If it falls across two or three standardised scans while strength drops too, that is a real signal.

Getting it measured properly

A scan is only as good as the protocol around it and the reading after it. If you want yours run consistently and programmed around, that is what our InBody body composition testing is for, at the studio on Finchley Road. Otherwise, book a call first. We will tell you if a scan is not what you need.

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. BTX coaches are coaches, and “dietitian” is a legally protected title in the UK: anything needing clinical interpretation belongs with your GP or a registered dietitian.

Frequently asked questions

Should I do my InBody scan fasted?
Ideally yes, or consistently three to four hours after food. Consistency matters more than fasting itself.

Can an InBody tell me if I gained muscle this month?
No. A few hundred grammes of “muscle gain” sits inside the noise of bioimpedance, so it is measurement error rather than progress. The same goes for body fat percentage, where anything under roughly 2 to 3 points is unreliable on this class of device. Scanning every three to four weeks is still worth doing, because it builds the history strip, but read the trend across two to three months rather than the gap between your last two scans.

Where is the visceral fat number on my sheet?
There is not one. This model does not measure or report visceral fat, so if you have been given a visceral fat figure, it did not come from this machine. Read total adiposity from percent body fat, BMI and waist-to-height ratio instead.

Why do my left and right arms show different muscle mass?
Often fluid distribution and time of day rather than a real imbalance, because segmental figures shift measurably between morning and evening (Oshima and Shiga, 2006). A gap that persists across several scans taken at the same time of day is worth programming around. A gap that appears once is not.

Can this scan tell me where I carry fat?
No. There is no segmental fat analysis on this model. The five-segment breakdown is lean tissue only.

Sources

• 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.
• Buch A et al. Nutrition & Diabetes, 2022;12:45. https://doi.org/10.1038/s41387-022-00223-1
• Kyle UG et al. Clinical Nutrition, 2004. https://www.espen.org/documents/BIA1.pdf
• Oshima Y, Shiga T. European Journal of Clinical Nutrition, 2006;60(8):938–941. https://doi.org/10.1038/sj.ejcn.1602402
• Schierbauer J et al. Metabolites, 2023;13(4):473. https://doi.org/10.3390/metabo13040473
• Randhawa AK et al. BMC Public Health, 2021;21:495. https://doi.org/10.1186/s12889-021-10552-y
• 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
• Cruz-Jentoft AJ et al. Age and Ageing, 2019;48:16–31. https://doi.org/10.1093/ageing/afy169
• NICE NG246, Overweight and obesity management. Published 14 January 2025, last updated 8 January 2026. https://www.nice.org.uk/guidance/ng246/resources/overweight-and-obesity-management-pdf-66143959958725

{“@context”: “https://schema.org”, “@type”: “FAQPage”, “mainEntity”: [{“@type”: “Question”, “name”: “Should I do my InBody scan fasted?”, “acceptedAnswer”: {“@type”: “Answer”, “text”: “Ideally yes, or consistently three to four hours after food. Consistency matters more than fasting itself.”}}, {“@type”: “Question”, “name”: “Can an InBody tell me if I gained muscle this month?”, “acceptedAnswer”: {“@type”: “Answer”, “text”: “No. A few hundred grammes of muscle gain sits inside the noise of bioimpedance, so it is measurement error rather than progress. The same goes for body fat percentage, where anything under roughly 2 to 3 points is unreliable on this class of device. Scanning every three to four weeks is still worth doing, because it builds the history strip, but read the trend across two to three months rather than the gap between your last two scans.”}}, {“@type”: “Question”, “name”: “Where is the visceral fat number on my sheet?”, “acceptedAnswer”: {“@type”: “Answer”, “text”: “There is not one. This model does not measure or report visceral fat, so if you have been given a visceral fat figure, it did not come from this machine. Read total adiposity from percent body fat, BMI and waist-to-height ratio instead.”}}, {“@type”: “Question”, “name”: “Why do my left and right arms show different muscle mass?”, “acceptedAnswer”: {“@type”: “Answer”, “text”: “Often fluid distribution and time of day rather than a real imbalance, because segmental figures shift measurably between morning and evening (Oshima and Shiga, 2006). A gap that persists across several scans taken at the same time of day is worth programming around. A gap that appears once is not.”}}, {“@type”: “Question”, “name”: “Can this scan tell me where I carry fat?”, “acceptedAnswer”: {“@type”: “Answer”, “text”: “No. There is no segmental fat analysis on this model. The five-segment breakdown is lean tissue only.”}}]}