FFMI and ALMI: the two muscle indexes, explained from their sources
By The DexaExplained Team · Last verified
Educational body-composition information, not medical advice. Nothing here diagnoses, treats, or prevents any condition, and this site does not interpret bone-density results. Talk to your clinician about anything on a scan report that concerns you.
BMI scales your total mass by height and can't tell muscle from fat. The two indexes on this page fix that in different ways — and both have proper published pedigrees that most calculators never mention.
FFMI: fat-free mass index
Formula: FFMI = fat-free mass (kg) ÷ height (m)². Fat-free mass is everything that isn't fat: muscle, bone, organs, water. From a DEXA report, FFM = weight × (1 − body-fat %/100).
The normalization. Kouri and colleagues noticed taller athletes scored slightly lower, so they adjusted every value to a 1.80 m reference height:
normalized FFMI = FFMI + 6.1 × (1.80 − height in m)
(Trivia for the careful reader: the paper's abstract prints 6.3, but the methods and the literature that reproduce the formula use 6.1 — we follow the methods.)
The famous 25. The paper measured 157 male athletes — 83 steroid users, 74 non-users. Its verbatim finding: the normalized FFMI of athletes who had not used steroids "extended up to a well-defined limit of 25.0," while users "easily exceeded 25.0, and some even exceeded 30." Twenty Mr. America winners from the pre-steroid era (1939-1959) averaged 25.4. That's the entire basis of the "natural limit ≈ 25" folklore — a real, citable observation, and also just one sample from 1995.
Compute your FFMI and see it against the published landmarks — thirty seconds if your report is in front of you.
Open the FFMI calculator →ALMI: appendicular lean mass index
Formula: ALMI = (lean mass of both arms + both legs, kg) ÷ height (m)². The "appendicular" restriction is the clever part: limb lean mass is mostly skeletal muscle, so ALMI tracks the muscle you move with, excluding organs.
Where your number sits. The NHANES reference curves (Kelly 2009 — the same dataset behind our percentile charts) put the adult male median around 8.4-9.1 kg/m² and the female around 6.2-7.0, both declining with age. The interpreter computes your exact percentile from your regional data.
The published cut-points. Research on age-related muscle loss defines "low lean mass" by ALMI thresholds:
| Source | Men | Women |
|---|---|---|
| Baumgartner 1998 (Am J Epidemiol) | < 7.26 kg/m² | < 5.45 kg/m² |
| EWGSOP2 2019 (Age Ageing, Table 3) | < 7.0 kg/m² | < 5.5 kg/m² |
These are research definitions — 2 SD below a young reference in Baumgartner's case — used to study populations, not to label individuals. Sitting near a cut-point is a reason to talk with a clinician and to take training and protein seriously; it is not a diagnosis, and a single scan is not a trend.
The limits, stated plainly
- FFMI inherits your body-fat error. A ±3-point error in body-fat % (typical for consumer BIA) moves FFMI by roughly ±0.7-0.9 for an average-height man — enough to cross a "landmark." DEXA-derived inputs make the index worth reading.
- Neither index measures performance. They quantify tissue, not strength, and neither knows how the mass is distributed within a limb.
- Reference data has a scope. The NHANES curves describe US adults 20-85 measured on Hologic hardware in 1999-2004; the Kouri sample was male athletes. Context, applied with those labels attached — that's what these indexes are for. Every constant here is rendered live on the methodology page.
Frequently asked questions
What is a good FFMI?
Untrained men typically sit in the high teens; consistent lifters commonly reach the low 20s. In Kouri's 1995 dataset the normalized FFMI of athletes who hadn't used steroids extended up to a well-defined limit of 25.0, with pre-steroid-era Mr. America winners averaging 25.4 — which is why ~25 became the folk ceiling for natural physiques. It describes a sample, not a law.
What is a normal ALMI?
It's age- and sex-specific. The NHANES reference median runs around 8.4-9.1 kg/m² for adult men and 6.2-7.0 for adult women, declining with age. Research definitions of low lean mass sit at 7.0-7.26 kg/m² (men) and 5.45-5.5 (women) depending on the working group. The interpreter computes your exact percentile.
FFMI vs ALMI — which should I care about?
FFMI answers a physique question (total fat-free mass for height) and only needs weight, height, and body-fat %. ALMI answers a functional-muscle question (arm+leg lean mass for height, mostly skeletal muscle) and needs DEXA regional data. For health tracking with age, ALMI is the one the research literature uses; for training progress, either works if measured consistently.
Can FFMI tell if someone is natural?
No. The 25.0 figure describes the ceiling in one 157-athlete sample from 1995. It's useful context for setting expectations — values above ~25 are rare without pharmacology — but as a detector applied to an individual it produces both false positives and false negatives.