Biomarkers · Updated September 2026
A DEXA scan measures fat, lean and bone by X-ray attenuation, region by region. A smart scale passes a current through your legs and runs the result through an equation. Both have a place; only one of them is a measurement.
Dual-energy X-ray absorptiometry passes two X-ray beams of different energies through the body and measures how much each is attenuated. Bone, lean tissue and fat attenuate differently, so the scan resolves the three compartments for every region — arms, legs, trunk, and the visceral fat estimate in the abdomen. The radiation dose is very small, a fraction of a chest X-ray. It is the reference method used in most clinical body composition research, and the one against which everything else is validated.
A bioelectrical impedance scale sends a small alternating current from one foot to the other and measures the resistance. Lean tissue is mostly water and conducts well; fat conducts poorly. The scale then estimates total body water from the impedance, converts that to lean mass with an assumed hydration constant, subtracts from body weight, and reports the remainder as fat. Every step of that chain uses a population equation with your age, sex and height as inputs. The scale is not measuring fat; it is measuring the electrical resistance of your lower body and inferring fat from a model.
Because the current takes the path of least resistance through the legs and lower trunk, a foot-to-foot scale barely samples the arms and upper body, and it is exquisitely sensitive to hydration. Drink a litre, sweat through a workout, eat a salty meal, or step on in the evening rather than the morning, and the reading moves by several percentage points of body fat with no change in actual fat. Skin temperature, recent exercise, the menstrual cycle, and whether your feet are dry all shift it. Validation studies against DEXA typically find that bioimpedance scales are roughly right on average across a group and wrong by 3–8 percentage points for a given individual, in either direction, and worse at the extremes of leanness or obesity.
Same time of day, same conditions: morning, after the bathroom, before food or drink, bare dry feet, every time.
Read the weekly average, not the daily number. Day-to-day noise is mostly water.
Treat the body fat figure as an index that goes up or down, not as a truth. A move from 24 to 22 over two months under identical conditions is meaningful; the 24 itself may be 20 or 28.
Ignore the segmental and visceral fat outputs on a foot-to-foot scale. They are extrapolations of an extrapolation.
| DEXA scan | Bioimpedance smart scale | |
|---|---|---|
| Method | Measures X-ray attenuation, three compartments | Estimates from electrical resistance and an equation |
| Individual accuracy | High; reference method | Low to moderate; ±3–8 points |
| Repeatability | Very good on the same machine | Poor day to day; fair for weekly averages |
| Regional data | Yes, arms, legs, trunk, visceral fat | Extrapolated on foot-to-foot; somewhat better with hand electrodes |
| Bone density | Yes | No |
| Radiation | Very small dose | None |
| Where | Clinic or scan centre | Bathroom |
| Frequency | Every 3–12 months | Daily |
| Cost | $50–250 per scan | $30–200 once |
| Best use | Baseline and periodic truth | Daily trend under fixed conditions |
The scale's weakness is the absolute number, and its strength is that you own it and use it daily. DEXA's strength is the absolute number, and its weakness is that it is a trip to a clinic and a fee. Used together they cover each other: a DEXA every six or twelve months calibrates what your scale is actually reading, and the scale, under fixed morning conditions, tracks the direction between scans. Someone losing weight who sees the scale's fat percentage drifting down and the DEXA confirming lean mass held has a genuinely useful picture. Someone reading a single scale number as a fact has a random number.
For the questions that matter most to longevity — whether lean mass is being preserved through a weight loss, whether resistance training is adding muscle in someone over fifty, whether bone density is holding — DEXA is the only one of the two that can answer. The scale cannot see bone at all, and its lean mass figure is the one most distorted by hydration.
Scales and standalone devices with hand grips as well as foot plates pass current through the whole body rather than the legs alone, and their segmental estimates are correspondingly less bad. They are still equation-based and still hydration-sensitive, but the upper-body sampling improves consistency. Clinical multi-frequency devices used in some gyms and clinics go further, using several current frequencies to separate intracellular from extracellular water, and their agreement with DEXA is better; they are not something to buy for a bathroom.
DEXA is a reference, not a gold standard. Different manufacturers' machines and software versions give different absolute results for the same person, sometimes by a couple of percentage points, so scans should be compared only on the same machine. Hydration affects DEXA too, though far less than impedance. Very large individuals may not fit the scan field. And the visceral fat estimate is a model applied to the abdominal region, useful for trends rather than as an exact quantity.
Waist circumference, measured at the navel in the morning under fixed conditions, tracks visceral fat closely enough that it is used in clinical risk scoring, costs nothing, and has none of the hydration noise of a scale. For the single question of whether abdominal fat is rising or falling, a tape and a notebook are hard to beat.
DEXA results are comparable only on the same machine and software; scale results only under identical morning conditions. Comparing across either is comparing noise.
Scale manufacturers that publish a validation study against DEXA with the error range stated. Most publish nothing, and “clinically accurate” without a number is a slogan.
Whole-body current path, better consistency, less lower-body bias.
A scale or app that shows weekly averages rather than headlining the daily figure is designed by people who understand the measurement.
The reference method first, then the home tools in the order of how much they should be trusted.
The measurement everything else is validated against, with regional lean mass, a visceral fat estimate and bone density in the same fifteen-minute appointment. For anyone tracking muscle through a weight loss, training after fifty, or concerned about bone, it is the one that answers the question. Book it through a scan centre; the Amazon link is for the guide that explains the report.
View on AmazonPassing current through arms and trunk as well as legs removes the worst of the lower-body bias and makes the trend more consistent. Still an estimate, still hydration-sensitive, still to be read as a weekly average. The best home option short of a clinical device.
View on AmazonWeight to 100 g is a real measurement, and the app's weekly averaging is the useful feature. The body fat figure is an index for direction only. Buy it for the weight trend and the habit; treat everything else it displays as a rough estimate that improves with consistent conditions.
View on AmazonWaist circumference at the navel, morning, relaxed, tracks visceral fat well enough to feature in clinical risk scores and costs a few dollars. A self-tensioning tape removes the pull-tightness variable. Paired with weight, it answers most of what people want a scale's body fat number for.
View on AmazonA trained measurer with calipers and a multi-site equation matches DEXA about as well as any home method, and the same person measuring the same sites the same way tracks change reliably. The catch is the training; self-measurement at some sites is awkward. Better than a scale for anyone willing to learn.
View on AmazonAcross a group, roughly right on average; for an individual, typically within 3–8 percentage points of DEXA, in either direction, and worse at very low or very high body fat. Day-to-day readings swing by several points with hydration alone. Read them as a trend under fixed morning conditions, not as a number.
If the question is whether you are preserving muscle during weight loss, building it through training, or holding bone density, yes: it is the only home-accessible method that answers those directly, and one scan every six to twelve months is enough. If the question is simply whether weight is going the right way, a scale and a tape measure are sufficient.
Sweating reduced your body water, which raised electrical resistance, which the equation interprets as more fat. Nothing changed but hydration. Weigh in the morning, before food, drink and exercise, and the readings become comparable.
The dose from a body composition DEXA is very small, a fraction of a chest X-ray and comparable to a few hours of natural background exposure. It is not a reason to avoid an annual scan. Pregnancy is the standard exclusion.
Weight on any scale plus waist circumference with a tape, both in the morning, weekly averages. That pair tracks fat loss and visceral fat change reliably, costs almost nothing, and has far less noise than a bioimpedance body fat figure. Add a DEXA once a year if lean mass or bone is the concern.
Take the 7-question quiz to get your personalized supplement and lifestyle stack — tailored to your age, fitness level, and longevity concerns.
Take the QuizA 28-day fillable sleep log built around the seven levers that actually move sleep, so you test one per week and see what it did.