Smart Scales and the Body Fat Number That Is a Guess

This looks at bioelectrical impedance body composition estimates from the practical end — what holds up once conditions stop being ideal.

What is actually measured

A small alternating current is passed between the electrodes and the device measures how much the body resists its passage. Lean tissue holds a great deal of water and salts and conducts well, while fat holds little water and conducts poorly. The measurement is therefore of resistance, from which everything else on the display is calculated rather than observed.

Muscle mass, bone mass, visceral fat and metabolic age are outputs of equations rather than separate measurements. Nothing is scanned, imaged or weighed separately, which is worth remembering when the app shows six decimal figures.

The equation and who it was built from

Manufacturers convert resistance into composition using formulas derived from reference populations measured by more accurate methods. Those formulas carry the characteristics of the group they came from, in age, sex, ethnicity and body type.

A person far from that reference group, including very lean athletes and people with obesity, gets a poorer estimate. Formulas are proprietary and differ between manufacturers, which is why two scales give two different answers for the same body. Amputation, implants, pregnancy and certain conditions affecting fluid all fall outside the assumptions these equations make.

Why the number moves overnight

Body water is the main determinant of conductance, so anything changing your hydration changes the estimate. Drinking, eating, sweating, alcohol, salt intake and menstrual cycle phase all shift it, sometimes by several percentage points.

Exercise beforehand changes blood distribution and skin condition, and both affect the reading. Skin temperature and how damp your feet are alter contact resistance directly at the electrodes. A body fat reading that fell two points overnight recorded a change in fluid, because fat does not move that quickly.

The foot-to-foot limitation

A bathroom scale has electrodes only under the feet, so the current travels up one leg and down the other. It largely bypasses the torso and arms, and the composition of those regions is estimated by extrapolation.

Handheld devices have the mirror-image problem, measuring the upper body and inferring the legs. Devices with both foot and hand contacts sample more of the body and are correspondingly better, though still estimates.

Even research-grade equipment carries meaningful error, so a consumer device promising precision is promising something unavailable.

Using it in the only way that works

The reading is more useful as a trend than as a value, because systematic error is at least consistent within one device. Measure at the same time of day, under the same conditions, on the same scale, and compare weekly averages rather than daily numbers. Do not compare your number against a friend on a different scale, since you are comparing two different equations.

Tape measurements at fixed sites, photographs and how clothes fit are all reasonable companions to the number. If a precise value genuinely matters, more accurate methods exist through clinical and research facilities.

Supplements interact with prescribed medicines, so tell whoever prescribes for you what else you are taking.

Two cautions worth knowing

Devices generally advise caution or avoidance for people with implanted electrical devices such as pacemakers. The current is tiny, and the advice exists because manufacturers cannot test every implant combination. Daily weighing suits some people and worsens preoccupation in others, particularly with a history of disordered eating.

A scale reporting metabolic age is presenting a marketing construct rather than a measurement of anything. The weight itself is the one number on the display that was actually measured.

The takeaway

Only the weight was measured; everything else was calculated from how well your feet conduct electricity. Use the trend, ignore the decimals.

When the marketing is more precise than the study, believe the study.