Beyond the Scale: How to Track Body Composition on a GLP-1 Microdose With DEXA, Waist, and Strength
A bathroom scale tells you how much you weigh. It cannot tell you whether the kilo you lost was fat, muscle, or last night's pizza water. Here is how accurate each tracking method really is, and a schedule that fits a slow, low-dose protocol.
39%
SHARE OF WEIGHT LOST AS LEAN MASS ON SEMAGLUTIDE 2.4 MG IN THE STEP 1 DEXA SUBSTUDY (WILDING, NEJM 2021)
3-5+
TYPICAL INDIVIDUAL ERROR, IN BODY-FAT PERCENTAGE POINTS, OF CONSUMER SMART SCALES VERSUS DEXA
0.5
WAIST-TO-HEIGHT RATIO BELOW WHICH CARDIOMETABOLIC RISK IS LOWER (ASHWELL, OBESITY REVIEWS 2012)
A 51-year-old architect started a low-dose semaglutide protocol in March and weighed himself every morning. By week five he had lost 2.8 kg, then gained 1.1 kg over a single weekend, and decided the protocol had stopped working. His belt had moved one notch in the same period. His DEXA scan at week 16 showed 4.1 kg of fat gone and 0.6 kg of lean mass lost, a split most people on full-dose trials would envy. The scale had been telling the truth about his weight and very little about his progress.
Body composition matters more on GLP-1 drugs than on most diets because the drugs reduce appetite so effectively that protein intake and training can fall without anyone noticing. The large trials measured this directly with DEXA, and the numbers show that a meaningful share of weight lost is lean tissue. Most people on a protocol have no DEXA scanner, but they do have a scale, a tape measure, and their own strength, and used properly those three cover most of what matters.
This article covers what the trials found about fat versus lean loss, how accurate each measurement method is, why daily weight is so noisy, and a tracking schedule that suits the slower pace of a microdose protocol.
Key takeaways
- •In the STEP 1 DEXA substudy, about 39 percent of weight lost on semaglutide 2.4 mg over 68 weeks came from lean mass. In the SURMOUNT-1 DEXA substudy, lean mass was about 25 percent of weight lost on tirzepatide.
- •DEXA has a repeat-scan precision error of roughly 1 to 2 percent, which makes it the most practical accurate tool for tracking fat and lean mass over months, as long as scan conditions are kept identical.
- •Consumer bioimpedance smart scales commonly misjudge an individual's body fat by 3 to 5 or more percentage points against DEXA, and a litre of water or a run can move the reading by more than a month of real change.
- •Daily scale weight varies by 0.5 to 1 kg from water, glycogen, and gut contents. A 7-day rolling average of morning weigh-ins shows the real trend.
- •Waist circumference, with a waist-to-height ratio below 0.5 as the target, and a monthly strength test give most of the useful information a DEXA scan would, at no cost.
How much of GLP-1 weight loss is fat and how much is muscle?
Between a quarter and two-fifths of the weight lost in the major GLP-1 trials came from lean mass, with the rest from fat. In the STEP 1 trial of semaglutide 2.4 mg, 140 participants had DEXA scans. Over 68 weeks they lost about 19 percent of their fat mass and about 10 percent of their lean mass, so lean tissue made up roughly 39 percent of the total drop. In the SURMOUNT-1 DEXA substudy of 160 tirzepatide users, fat mass fell by about 34 percent and lean mass by about 11 percent, so lean tissue was closer to 25 percent of the loss.
Two points soften those numbers. First, "lean mass" on DEXA is everything that is not fat or bone, including water, organs, and the connective tissue that supported the extra weight, so true skeletal muscle loss is smaller than the headline figure. Second, the proportion of lean mass in the body actually rose in both trials, because fat fell faster. Body composition improved, even as absolute muscle fell. Our article on muscle preservation on a microdose covers the protein and training side in detail.
Where the weight came from: fat versus lean mass, % of total loss
DEXA substudies. Wilding JPH et al., NEJM 2021 (STEP 1, 68 weeks); Jastreboff AM et al., NEJM 2022 (SURMOUNT-1, 72 weeks). Rounded.
- Fat mass
- Lean mass
Both trials used full therapeutic doses. No trial has yet published DEXA data for a low-dose protocol, where weight loss is slower.
Why does the scale go up and down every day?
Scale weight moves by 0.5 to 1 kg from one day to the next because of water, not fat. Each gram of glycogen stored in muscle and liver binds roughly 3 grams of water, so a carbohydrate-heavy dinner can add close to a kilo overnight. Sodium does the same thing through fluid retention. Gut contents, especially on a GLP-1 drug that slows stomach emptying and bowel transit, add several hundred grams more. None of this is body fat, and all of it reverses within a few days.
The pattern is predictable enough to have been studied. A 2014 analysis in Obesity Facts by Orsama and colleagues tracked daily weigh-ins in 80 adults and found weight rose over the weekend and fell through the working week. The fix is a 7-day rolling average of morning weigh-ins, taken after the toilet and before food or drink. Most scale apps calculate this automatically, and it strips out most of the noise that made the architect in the introduction want to quit.
Daily weigh-ins versus the 7-day average over six weeks, kg
Illustrative data for one person losing about 0.4 kg per week, with normal daily fluctuation.
- Daily weigh-in
- 7-day average
Day 15 reads heavier than day 1 on the daily line. The average shows 0.7 kg lost by then.
How accurate are DEXA, smart scales, and calipers?
DEXA is the most accurate widely available method, while consumer smart scales are the least accurate for any single reading. A DEXA scan uses two low-energy X-ray beams to separate fat, lean tissue, and bone, with a radiation dose lower than a day of natural background exposure. Repeat-scan precision is roughly 1 to 2 percent for total fat and lean mass, which means a change of about 1 kg or more between scans on the same machine is likely real.
Bioelectrical impedance analysis, the technology in smart scales, sends a weak current through the body and estimates water content from the resistance, then infers fat from water using population equations. A 2004 review in Clinical Nutrition by Kyle and colleagues laid out why this works reasonably for groups and poorly for individuals. In a 2018 comparison of 3,655 measurements, Achamrah and colleagues found BIA and DEXA agreed on average but diverged widely at high and low BMI. Eight-electrode clinic devices perform better than foot-to-foot scales because current passes through the arms and trunk as well as the legs. Skinfold calipers can be reasonable in trained hands, and much less so when measuring yourself.
Typical individual error in body-fat estimate, percentage points
Approximate values from validation literature (Kyle 2004; Achamrah 2018). Actual error varies by device, body size, and hydration.
A 5-point error on a person at 30 percent body fat means the true figure could plausibly sit anywhere from 25 to 35 percent.
Why does my smart scale show different body fat at different times of day?
A smart scale's body-fat number changes through the day because it measures water, and body water shifts constantly. Drinking a litre of water adds conductive fluid and often raises the fat reading, because the scale also adds a kilogram of weight it attributes partly to fat. A run lowers the reading, since warm muscle and increased blood flow reduce resistance in the legs. Neither changes actual fat by a measurable amount.
This matters more on a GLP-1 protocol than usual, because fluid intake often drops in the first weeks when appetite and thirst signals quieten. Mild dehydration raises resistance and can make body fat look higher, or make lean mass appear to fall. Anyone who uses a smart scale should step on it under identical conditions every time, first thing in the morning, and read only multi-week trends. Our article on hydration and micronutrients covers fluid targets on a protocol.
One person, one day, one smart scale: body-fat reading, %
Illustrative readings for a person whose true body composition did not change during the day.
A 2.8-point swing in one day is larger than the real change most people see in two months.
Is waist circumference a better measure than weight?
Waist circumference is a better guide to metabolic progress than weight alone, because it tracks the abdominal and visceral fat most linked to risk and it does not rise when muscle or water increases. A 2012 meta-analysis in Obesity Reviews by Ashwell, Gunn, and Gibson pooled data from more than 300,000 adults and found waist-to-height ratio predicted diabetes, high blood pressure, and cardiovascular disease better than BMI. The UK's NICE guidance adopted a waist-to-height ratio below 0.5 as the healthy threshold in 2022.
Measuring well takes a minute. Stand relaxed, breathe out gently, and place a flexible tape horizontally at the navel without compressing the skin. Take two readings and use the average. Measuring every two weeks at the same time of day is enough, since a real change of 1 to 2 cm takes a few weeks on a low-dose protocol. Waist tends to keep falling even during scale plateaus, which is one of the most useful things a microdoser can know. The biomarker tracking guide covers the blood tests that pair with it.
How can you tell if you are losing muscle without a scan?
Strength is the best free signal of muscle loss, because muscle function tends to fall before a scan would confirm a change. Grip strength with an inexpensive hand dynamometer, the number of chair stands in 30 seconds, or the weights and repetitions used in regular training all work, provided the same test is repeated the same way every four weeks. The European sarcopenia consensus (EWGSOP2, Cruz-Jentoft 2019) uses grip strength below 27 kg in men and 16 kg in women as a threshold for low strength.
A drop of 10 percent or more in any of these tests over a couple of months, while weight is falling, is a reason to review two things: protein and resistance training. Protein around 1.2 to 1.6 g per kg of body weight daily, spread over meals, and two to three resistance sessions a week have the strongest evidence for protecting lean mass during weight loss. Strength that holds steady or rises while the waist shrinks is about as good a sign as tracking can offer.
Illustrative 16-week tracking record on a low-dose protocol
Weekly average weight and waist (left axis) with grip strength (right axis). Illustration only.
- Weight, 7-day avg (kg)
- Waist (cm)
- Grip strength (kg)
5.5 kg and 7 cm down, grip unchanged: the pattern that suggests most of the loss is fat.
What should a tracking schedule on a GLP-1 microdose look like?
A sensible schedule combines a daily scale averaged weekly, a fortnightly waist measurement, a monthly strength test, and an occasional DEXA scan. Low-dose protocols typically produce slower weight loss than the 15 to 20 percent seen in full-dose trials, which makes noise a bigger share of each reading. Measuring less often, under tighter conditions, gives cleaner answers than measuring more.
DEXA is worth the cost at baseline and every three to six months for anyone over 50, anyone with low starting muscle mass, and anyone whose strength is falling. Scanning more often than every three months mostly captures machine precision error rather than real change. Book repeat scans on the same machine, at the same time of day, fasted, and without training beforehand.
| Measure | How often | Change worth acting on |
|---|---|---|
| Morning scale weight (7-day average) | Daily weigh-in, review the weekly average | A change in the average of about 0.5 kg or more over two weeks |
| Waist circumference at the navel | Every 2 weeks, same time of day | 2 cm or more; waist-to-height ratio below 0.5 is the usual target |
| Grip strength or 30-second sit-to-stand | Every 4 weeks | A fall of 10 percent or more is a flag to review protein and training |
| DEXA scan | Baseline, then every 3 to 6 months | Roughly 1 kg or more in fat or lean mass, on the same machine |
| Consumer smart-scale body fat | Optional; trend only | Only a sustained multi-week trend, never a single reading |
What is the takeaway?
The takeaway is that weight is one number among several, and on a slow protocol it is the noisiest. Averaging the scale, measuring the waist, and checking strength once a month shows whether the weight coming off is mostly fat, and that is the question GLP-1 trials made people worry about. A DEXA scan confirms the picture when the stakes are higher. The architect from the opening now weighs himself daily, looks only at the Sunday average, and has stopped worrying about Mondays.
"The scale answers how much. The tape and your grip answer what."
Frequently asked questions
How much of the weight lost on a GLP-1 medication is muscle?
In the DEXA substudy of the STEP 1 trial, about 39 percent of the weight lost on semaglutide 2.4 mg came from lean mass and 61 percent from fat. In the SURMOUNT-1 DEXA substudy of tirzepatide, lean mass made up about 25 percent of the loss. Lean mass on DEXA includes water, organs, and connective tissue as well as muscle, so these figures overstate true muscle loss somewhat.
Is DEXA the best way to measure body composition on a GLP-1 protocol?
DEXA is the most practical accurate method for tracking fat and lean mass over time, with a repeat-scan precision error of around 1 to 2 percent. A single scan costs roughly $50 to $150 in many US cities. Its accuracy depends on consistency: the same machine, the same time of day, and the same hydration and meal status for every scan.
Are smart scales accurate for body fat percentage?
Consumer smart scales that use bioelectrical impedance are not accurate enough to judge individual readings. Validation studies commonly find individual errors of 3 to 5 or more body-fat percentage points compared with DEXA, and the reading shifts with hydration, recent exercise, and meals. A smart scale can show a long-term trend if it is used under identical conditions, but a single number should not drive decisions.
Why does the scale jump up and down from day to day?
Daily scale weight moves by 0.5 to 1 kg or more because of water, sodium, glycogen, and gut contents, not fat. Each gram of stored glycogen holds roughly 3 grams of water, and a salty meal can add several hundred grams of water overnight. A 7-day rolling average of morning weigh-ins removes most of this noise and shows the real direction.
Why is weight loss fast in the first two weeks and then slower?
The first 1 to 2 kg lost after eating less is mostly glycogen and the water bound to it, which is why early loss looks fast. Once glycogen settles at a new level, the scale reflects fat loss more closely, which typically runs at 0.25 to 1 percent of body weight per week. On a low-dose GLP-1 protocol the later pace is usually at the slower end of that range.
Is waist measurement better than weight for tracking progress?
Waist circumference tracks abdominal and visceral fat more closely than scale weight, and it does not rise when muscle or water increases. A 2012 meta-analysis by Ashwell and colleagues found waist-to-height ratio outperformed BMI for predicting cardiometabolic risk, and a ratio below 0.5 is the commonly used target. Measuring at the navel, after breathing out gently, every two weeks is enough.
How can I tell if I am losing muscle on a GLP-1 microdose?
Without a DEXA scan, the most useful signal is strength. A steady fall in grip strength, sit-to-stand repetitions, or the weights used in regular training over several weeks suggests muscle loss worth addressing. Protein intake of about 1.2 to 1.6 g per kg of body weight daily and resistance training two to three times a week are the two measures with the best evidence for protecting lean mass during weight loss.
How often should body composition be measured?
A practical schedule is daily morning weigh-ins reviewed as a weekly average, waist every two weeks, a strength check every four weeks, and a DEXA scan at baseline and every three to six months. Scanning more often than that mostly measures machine noise, because true changes in lean mass are small relative to the scan's precision error over a few weeks.
Clinical references
- Wilding JPH, Batterham RL, Calanna S, et al. Once-weekly semaglutide in adults with overweight or obesity (STEP 1). New England Journal of Medicine. 2021;384(11):989-1002.
- Jastreboff AM, Aronne LJ, Ahmad NN, et al. Tirzepatide once weekly for the treatment of obesity (SURMOUNT-1). New England Journal of Medicine. 2022;387(3):205-216.
- Kyle UG, Bosaeus I, De Lorenzo AD, et al. Bioelectrical impedance analysis, part I: review of principles and methods. Clinical Nutrition. 2004;23(5):1226-1243.
- Achamrah N, Colange G, Delay J, et al. Comparison of body composition assessment by DXA and BIA according to the body mass index: a retrospective study on 3655 measures. PLoS One. 2018;13(7):e0200465.
- Ashwell M, Gunn P, Gibson S. Waist-to-height ratio is a better screening tool than waist circumference and BMI for adult cardiometabolic risk factors: systematic review and meta-analysis. Obesity Reviews. 2012;13(3):275-286.
- Orsama AL, Mattila E, Ermes M, van Gils M, Wansink B, Korhonen I. Weight rhythms: weight increases during weekends and decreases during weekdays. Obesity Facts. 2014;7(1):36-47.
- Cruz-Jentoft AJ, Bahat G, Bauer J, et al. Sarcopenia: revised European consensus on definition and diagnosis (EWGSOP2). Age and Ageing. 2019;48(1):16-31.
- National Institute for Health and Care Excellence. Obesity: identification, assessment and management (CG189), waist-to-height ratio update. 2022.
- Neeland IJ, Linge J, Birkenfeld AL. Changes in lean body mass with glucagon-like peptide-1-based therapies and mitigation strategies. Diabetes, Obesity and Metabolism. 2024;26(Suppl 4):16-27.
Medical disclaimer. This article is for general educational and informational purposes only. It is not medical advice, not dosing instruction, and not a substitute for the prescriber or clinician responsible for your care. The body-composition trial data described come from full therapeutic doses of semaglutide and tirzepatide, not low-dose protocols. The daily weigh-in, smart-scale, and 16-week tracking charts are illustrations and do not describe any real individual or predict any outcome. GLP-1 microdosing is an off-label approach and should only be undertaken under a prescriber's supervision. Nothing here describes a treatment for obesity, sarcopenia, or any other condition. Unexplained weakness, rapid weight loss, dizziness, or signs of dehydration should be discussed with a clinician promptly.