DEXA, BIA and Calipers, Compared
By Longevity Lab · Published: August 23, 2026
Based on peer-reviewed research — full source list at the end of this article. This is educational information, not medical advice.
- DEXA is the practical reference standard and the only common method giving a regional breakdown.
- Calipers are cheap and track one person well, but between-tester differences are large.
- Impedance is the most convenient and the most affected by hydration.
- Every method has a systematic bias, so never compare numbers from two different methods.
- For most purposes, a tape measure answers the underlying question adequately.
There is no cheap, accurate, convenient way to measure body composition — you get two of the three. DEXA is accurate and gives a regional breakdown but requires a clinic visit and a fee. Skinfold calipers are cheap and reproducible in one skilled pair of hands, and unreliable between different pairs. Bioelectrical impedance is instant and moves several points with your hydration. Choosing well means starting from what you actually need to know.
The four practical options
| Method | Error vs reference | Cost | Gives you |
|---|---|---|---|
| DEXA | Reference standard (1–2%) | Clinic fee | Fat, lean and bone by region |
| Air displacement (Bod Pod) | 1–3 points | Clinic fee | Whole-body fat and lean only |
| Skinfold calipers | 3–5 points | Very low | Subcutaneous fat at specific sites |
| Bioelectrical impedance | 4–8 points | Low | Whole-body estimate, hydration-sensitive |
The error column is against DEXA, which is itself an estimate rather than a truth — it assumes fixed tissue densities and is affected by hydration too, just far less. Underwater weighing and multi-compartment models are more accurate still, and confined to research.
DEXA: what the extra money buys
The regional breakdown, mainly. DEXA reports fat and lean mass for each limb and the trunk separately, and most modern software estimates the visceral fat area within the abdomen. That is information no other accessible method provides, and it is the information that matters most — Visceral fat is the compartment tied to cardiometabolic risk, and total body fat percentage says nothing about it.
It also measures bone mineral density in the same scan, which is worth having in its own right for anyone over 50 or with risk factors for osteoporosis.
The caveats are real but manageable. Hydration still shifts results by a point or so, different manufacturers’ machines are not interchangeable, and repeat scans should be done on the same machine at the same time of day, fasted, without exercise beforehand.
Calipers: the underrated option
Skinfold calipers measure the thickness of a pinched fold of skin and Subcutaneous fat at standardized sites — usually three or seven — which a regression equation converts to a body fat percentage.
Their weakness is well known: results depend heavily on who is doing the pinching, on finding the same site each time, and on the equation chosen. Between two testers the same person can differ by several points.
Their strength is less appreciated. For one person measured by the same tester over time, calipers are highly reproducible, and the raw skinfold sum in millimeters is a better tracking measure than the derived percentage, because it skips the equation and its assumptions entirely. If your seven-site sum falls from 96 mm to 74 mm, fat has been lost — no model required.
They cannot see abdominal fat, which is the limitation that matters. Subcutaneous folds tell you nothing about what sits behind the abdominal wall.
Impedance: convenience and its price
Covered in detail elsewhere in this cluster, but the short version: it estimates total body water from electrical resistance and derives fat from that, so anything affecting hydration affects the result. Dehydration makes the reading higher; a recent meal or workout makes it lower.
Eight-electrode devices using both hands and feet are meaningfully better than foot-to-foot bathroom scales, which largely sample the legs and extrapolate the trunk. Segmental medical-grade devices are better still and approach DEXA for tracking, though not for absolute accuracy.
The rule that saves the most confusion: never compare a number from one method with a number from another. Each carries its own systematic bias, so a caliper 19% and a scale 25% are not a contradiction to be resolved — they are two different instruments with two different offsets.
What most people actually need
A tape measure. The question behind “what is my body fat percentage” is almost always “am I carrying too much fat, and is it in the dangerous place” — and waist circumference divided by height answers that for nothing, with no equipment, no equation and no hydration sensitivity.
Add a body composition measurement when there is a specific reason: a BMI that disagrees with how you look, tracking whether a deficit is costing muscle, an athlete needing a documented figure, or an older adult monitoring muscle loss.
Body fat assessment estimates the fat fraction from circumferences using the validated US Navy method, which is free and about as accurate as a consumer impedance scale. If the visceral compartment is what you are worried about, Lipid Accumulation Product (LAP) combines waist with fasting triglycerides into an index that tracks it considerably better than any whole-body percentage does.
Frequently asked questions
What is the most accurate body composition test?
Among accessible options, DEXA. It reports fat, lean and bone by region and estimates visceral fat area, at roughly 1–2% error against multi-compartment reference models.
Are calipers better than impedance scales?
For tracking one person over time with the same tester, generally yes. For a single absolute number from an untrained user, neither is reliable.
Why do different methods give different numbers?
Each has its own systematic bias from the assumptions in its model. Never compare across methods — pick one and stay with it.
How often should I measure body composition?
Quarterly at most. Real change is slow relative to measurement error, so more frequent testing mostly measures noise.
Sources
Sources (4)
- Kyle UG, Bosaeus I, De Lorenzo AD, et al. Bioelectrical impedance analysis — part II: utilization in clinical practice. Clin Nutr. 2004;23:1430–1453. PubMed →
- Jackson AS, Pollock ML. Generalized equations for predicting body density of men. Br J Nutr. 1978;40:497–504. PubMed →
- Shepherd JA, Ng BK, Sommer MJ, Heymsfield SB. Body composition by DXA. Bone. 2017;104:101–105. PubMed →
- Sun Q, van Dam RM, Spiegelman D, et al. Comparison of dual-energy X-ray absorptiometric and anthropometric measures of adiposity in relation to adiposity-related biologic factors. Am J Epidemiol. 2010;172:1442–1454. PubMed →
Disclaimer. This article is for information only and does not replace medical advice. Talk to a qualified clinician before changing anything about your health.
