Methodology & sources

Body-fat & ALMI percentiles — NHANES via Kelly 2009

Percentiles come from Kelly TL, Wilson KE, Heymsfield SB. "Dual energy X-ray absorptiometry body composition reference values from NHANES." PLoS ONE 2009;4(9):e7038 (PMID 19753111) — the standard reference publication of NHANES 1999-2004 whole-body DXA data (Hologic QDR-4500A), adults aged 20-85.

The paper publishes smoothed LMS parameters (skew L, median M, population SD σ) at 5-year age anchors rather than percentile tables. We store its Table S2 (%fat) and Table S6 (appendicular lean mass ÷ height²) rows verbatim and compute percentiles with the paper's own equation, interpolating parameters linearly between anchors:

z = M · ((X/M)^L − 1) / (L·σ)      percentile = Φ(z)
X = M · (1 + L·σ·z/M)^(1/L)         (inverse, for reference rows)

Honest simplification, disclosed: the paper reports separate curves for White, Black, and Mexican-American participants and no pooled column. It notes the White curves rest on the most observations and "should provide more robust classification estimates," so the tools use those as the single general-population reference. Body-composition distributions differ across ethnic groups; treat percentiles as population context, not a personal grade.

FFMI

Fat-free mass index = FFM(kg) ÷ height(m)² (VanItallie TB et al., Am J Clin Nutr 1990;52:953-9). Normalized FFMI adds 6.1 × (1.80 − height in m) per Kouri EM, Pope HG, Katz DL, Oliva P. Clin J Sport Med 1995;5:223-8 (PMID 7496846). That paper's non-steroid-using athletes reached normalized FFMI values up to a well-defined limit of 25.0, and pre-steroid-era Mr. America winners averaged 25.4. (The abstract prints a 6.3 normalization slope; the paper's methods and the literature that reproduces it use 6.1, which we follow.) We deliberately do not repeat the widely circulated "user vs non-user group means" because they aren't verifiable from the primary abstract.

ALMI reference cut-points

The interpreter reports ALMI (appendicular lean mass ÷ height²) percentiles from Kelly 2009 above, and mentions two published low-lean-mass reference cut-points for context only:

• Baumgartner RN et al., Am J Epidemiol 1998;147:755-63 (PMID 9554417): <7.26 kg/m² (men) / <5.45 kg/m² (women) — 2 SD below a young reference group.
• Cruz-Jentoft AJ et al. (EWGSOP2), Age Ageing 2019;48:16-31 (PMID 30312372), Table 3: <7.0 kg/m² (men) / <5.5 kg/m² (women).

These are research definitions used in studies of low muscle mass. Falling under a cut-point is a reason to talk to a clinician, not a diagnosis — and the tools say so wherever the numbers appear.

Visceral fat (VAT)

DXA visceral-fat estimates were validated against CT by Kaul S et al., Obesity 2012;20:1313-8 (PMID 22282048; r² ≈ 0.95 vs CT in 124 adults). The ≥100 cm² elevated-risk area threshold is primary-cited to the Examination Committee of Criteria for "Obesity Disease" in Japan, Circ J 2002;66:987-92 (PMID 12419927), derived on CT. ~160 cm² as "high" is the common convention in DXA vendor reporting rather than a single primary paper, and the interpreter labels it that way. Thresholds are CT-derived areas applied to DXA via the validated calibration; DXA also reports VAT as mass (g) and volume (cm³), which are not interchangeable with area cut-points.

ACE body-fat categories

As a secondary interpretive scale, some guides reference the American Council on Exercise body-fat norms:

The composition band visual

The three-segment band renders your real body-fat share; when you haven't entered bone-mineral content, the bone segment is drawn at a typical 4% of body mass (adult whole-body BMC on DXA is roughly 2-3 kg) and labeled as typical. The bone segment is illustrative; no statistic uses it.

Price data

Every price in the cost calculator was read from the provider's own published page on the date shown. National provider-type ranges are derived from these points — nothing is invented:

Provider-type ranges currently shown: Specialty scan chain / marketplace $69-$330; University human-performance lab $125-$210; Imaging center / clinic (self-pay) $75-$250; Mobile van service $40-$89.

The boundary

Related honesty rules: no invented averages, no fake reviewer credentials, no sponsored input into any number — see about and disclosure.