Screening, graded
DEXA body composition scan: evidence grade B, conditional on same-scanner use
Evidence grade B. Whole-body dual-energy X-ray absorptiometry reporting fat mass, lean mass, body-fat percentage, estimated visceral fat, and bone density context. Graded as a MEASUREMENT TOOL: the question is whether its numbers are accurate enough to act on.
- The condition on the grade: same-scanner, same-prep use. Cross-brand trend lines void the grade.
- Accuracy: group means match criterion models within about half a point of body fat; one person's printed number carries error of roughly -3 to +4 points.
- Precision: repeat scans on the same scanner vary by under 1%, the best repeatability of any field method, which is what makes a trend on one scanner worth reading.
- Cost: Tens of dollars to about $150 (BodySpec lists $40 to $60; vendor figure).
- Repeat cadence: Quarterly at the fastest for composition tracking (basis: measurement science, labeled on the record).
Already have a result that worries you?
That includes a bone-density flag on a DEXA report, or a lean-mass trend that does not match how you feel, train, and perform. Your next step is the clinician who ordered the test, or your own doctor, before anything else on this page. Bring the result itself, not a grade from a website.
The grade, axis by axis
Every screening grade on this site is built from the same five axes. The headline grade is outcome-change evidence discounted by harms: for this test, Graded as a measurement tool: individual error around plus or minus 4 points against the criterion bars the A range, because the printed percentage is not the buyer's true body-fat percentage. Class-leading precision, trivial physical harm, low cost, and direct program actionability hold it at B. CONDITIONAL on same-scanner, same-prep use; cross-brand trend lines void the grade.
1. Does early detection change outcomes?
Not applicable as disease screening; axes 1 and 2 are reframed as accuracy and precision (see detectionPerformance).
2. How well does it detect, in people without symptoms?
Accuracy: against four-compartment criterion models, group means land within about half a point while individual error spans roughly minus 3 to plus 4 points of body fat (Clasey et al., 1999, and successor validation work). Precision: same-scanner repeatability is class-leading (GE Lunar iDXA precision CV 0.86% for percent fat; Hind et al., 2011), so fat-mass changes of roughly 0.5 kg and up are resolvable on the same scanner with consistent preparation; treat body-fat moves under about 1 point as noise. Cross-brand numbers are NOT interchangeable: Hologic and GE Lunar differ by about 1.4 points even after harmonization (Shepherd et al., 2017).
3. What does the harm ledger look like?
| Harm | Rate or scale | Source |
|---|---|---|
| Cross-brand trend corruption: a trend line that switches scanner brands measures the scanner change | 1.4 points absolute difference after harmonization; 2.13x difference in sarcopenia classification before standardization (2024) | Shepherd J et al., 2017; 2024 Hologic-GE comparison study |
| State effects: hydration, recent food, and recent exercise move lean-mass readings and manufacture fake change | Per-scan, controllable by standardized prep | Longevity Lab Report screening evidence review, 2026-10-18 |
| Radiation | About 4 to 5 microsieverts, under a day of natural background | ISCD-cited figures |
| Access limits: table weight limits (about 159 kg Hologic, about 182 kg GE) exclude some buyers | Structural | Longevity Lab Report screening evidence review, 2026-10-18 |
4. Does a positive result connect to an action that helps?
HIGH FOR ITS ACTUAL JOB. A same-scanner trend verifies whether a training or nutrition block changed fat and lean mass beyond noise, and flags low lean mass for follow-up. It changes programs, not prescriptions, and does not pretend otherwise.
5. What does it cost, including the follow-through?
Per screen: Tens of dollars to about $150 (BodySpec lists $40 to $60; vendor figure)
Downstream: No meaningful cascade standalone; an unexpected bone-density flag is the main referral path, and it is a legitimate one.
Repeat cadence: Quarterly at the fastest for composition tracking (basis: measurement science (least-significant-change math) plus common practice; monthly scanning is a vendor pattern. Bone-density DEXA intervals are a separate guideline world and must not be imported here.)
On the same scale
DEXA is graded as a measurement tool, not as disease screening: it answers a question about your own body composition, accurately enough to steer training and nutrition, and it claims nothing about finding disease early. The other measurement sold to healthy people on a repeat cadence is the glucose monitor for people without diabetes, graded C- on the same axes, where the accuracy story runs the other way. Every graded test sits in the screening comparison table.
Grades are research summaries, not medical advice. Decisions about your own stack belong with your clinician.
These pages grade a test in people without symptoms, as a population question. They never grade a reader, and they cannot read your result.
Longevity Lab Report
Screening news, graded when it happens.
The grades on this page stay free. The email is the week's longevity and biotech news in plain English: biotech news, expert moves, what is working, and new supplements and peptides, each story linked to the grades on this site.
Questions, answered plainly
Is a DEXA scan worth it?
As a measurement tool, graded B: the strongest-graded item in our screening set alongside the calcium score. What you are buying is a trustworthy trend, whether a training or nutrition block actually changed fat and lean mass. The grade is conditional on same-scanner, same-prep use; a trend line that switches scanner brands measures the scanner change instead.
How accurate is the body-fat percentage?
Against four-compartment criterion models, group means agree within about half a point of body fat, but an individual's printed number carries error of roughly -3 to +4 points. Repeat scans on the same scanner vary by under 1% (a coefficient of variation of 0.86% in one precision study), which is why the trend is trustworthy and the single number is not.
How often should you get a DEXA scan?
Quarterly at the fastest for composition tracking, and most interpretable change takes about six months to detect. The monthly cadence marketed by scan chains is a vendor pattern, not a measurement need: scanning faster than real change accumulates mostly measures noise.
Does the scanner brand matter?
Yes. Hologic and GE Lunar scanners differ by about 1.4 points of body fat even after published harmonization equations, and lean-mass figures differ alongside. Pick one scanner, prepare the same way each time (fasting state, time of day, no recent hard workout), and trend on that scanner only.
What about the bone-density number in the same scan?
It rides along as the companion measurement, and that side carries the guideline support: it finds fracture risk while osteoporosis treatment still helps. A low bone-density flag is a clinical finding, not a biohacking metric, and it belongs with your clinician.
Sources
- Hind K et al. GE Lunar iDXA precision study. European Journal of Clinical Nutrition, 2011. (2011)
- Clasey JL et al. Four-compartment comparison. Journal of Applied Physiology, 1999. (1999)
- Shepherd JA et al. Hologic and GE Lunar harmonization. 2017. (2017)
Last reviewed October 18, 2026
- 2026-10-18|Initial grading from the 2026-10-18 screening evidence review.