Metabolic Physiology
Katch-McArdle Formula
Also known as: Katch-McArdle equation, lean-mass RMR equation
A resting metabolic rate equation calculated from fat-free mass rather than total body weight, more accurate in principle and dependent on a body-composition measurement.
Key takeaways
- RMR = 370 + (21.6 x fat-free mass in kg), with no terms for age, height or sex.
- More accurate than weight-based equations in principle, because lean tissue accounts for most resting energy expenditure and adipose tissue for little.
- Its accuracy is entirely inherited from the body-composition measurement it is given.
- A consumer bioimpedance scale is not an adequate input — its readings shift several percentage points with hydration alone.
- Fed a guessed body-fat percentage it produces apparent precision from a bad input, which is worse than using Mifflin-St Jeor.
The Katch-McArdle formula predicts resting metabolic rate from fat-free mass instead of total body weight:
RMR = 370 + (21.6 × fat-free mass in kg)
Notice what is absent: no age term, no height term, no sex term. All three exist in weight-based equations as proxies for lean mass. Given lean mass directly, they become redundant.
Why it is better in principle
Resting energy expenditure is dominated by metabolically active tissue — organs and skeletal muscle. Adipose tissue contributes comparatively little per kilogram.
So two people at 80 kg with meaningfully different body-fat percentages do not have the same RMR, and a weight-based equation cannot distinguish them. Katch-McArdle can, which is why it tends to outperform in populations at the extremes: very lean athletes and people with high body fat, where a weight-based equation errs in opposite directions.
Why it is usually worse in practice
Because it has a hard dependency, and the dependency is normally unmet:
Katch-McArdle's accuracy is entirely inherited from the body-composition measurement you feed it. It cannot be better than that input, and it will not tell you the input was bad.
| Input source | Suitable? |
|---|---|
| DEXA | Yes — the practical reference for consumers |
| Air displacement plethysmography | Yes |
| Skinfold calipers, skilled operator | Acceptable |
| Consumer bioimpedance scale | No — shifts several points with hydration alone |
| Estimated from a photograph | No |
A bathroom bioimpedance scale reading is not a body-composition measurement in the sense this equation requires. It is repeatable rather than accurate, and it moves with how much water you drank.
The failure mode worth naming
Fed a guessed body-fat percentage, Katch-McArdle produces a number that looks more authoritative than Mifflin-St Jeor while resting on a worse foundation. Its shorter form and single input read as precision.
The decision rule is therefore simple:
- Have a DEXA or equivalent? Use Katch-McArdle.
- Guessing your body fat, or reading it off a scale? Use Mifflin-St Jeor. Its inputs — weight, height, age — are things you actually know.
And either way, replace the prediction after three weeks with maintenance derived from your own intake and weight trend. An equation estimates a population; three weeks of your own data measures you.
Frequently asked
Is Katch-McArdle more accurate than Mifflin-St Jeor?
In principle yes, in practice usually no. It is better in principle because it calculates from fat-free mass, and lean tissue accounts for most resting energy expenditure while adipose tissue accounts for little — so it distinguishes two people of the same weight with different body composition, which a weight-based equation cannot. In practice it is only as good as the body-composition figure you give it, and most people do not have a reliable one. With a guessed body-fat percentage it produces apparent precision from a bad input, which is worse than using Mifflin-St Jeor.
Can I use my smart scale body-fat reading with Katch-McArdle?
No. Consumer bioimpedance readings shift by several percentage points with hydration status alone, which means they are repeatable rather than accurate. Feeding one into an equation whose entire advantage is the quality of that single input defeats the purpose. Use DEXA, air displacement plethysmography, or skilled skinfold calipers if you want to use this formula; otherwise use a weight-based equation whose inputs you actually know.
Why does Katch-McArdle have no age or sex term?
Because age, height and sex appear in weight-based equations as proxies for lean mass. Once lean mass is supplied directly, those proxies become redundant — the equation already has the variable they were standing in for. That structural simplicity is a genuine strength of the formula and also the reason it looks more authoritative than it is when the lean-mass figure was guessed.
References
- Katch VL, McArdle WD, Katch FI. "Essentials of Exercise Physiology". Lippincott Williams & Wilkins , 2011 .
- Frankenfield D, Roth-Yousey L, Compher C. "Comparison of predictive equations for resting metabolic rate in healthy nonobese and obese adults: a systematic review". Journal of the American Dietetic Association , 2005 .
Related terms
- Body Fat Percentage Fat mass expressed as a percentage of total body mass — a headline metric for body composi…
- Lean Body Mass Total body mass minus fat mass — everything non-fat: muscle, bone, organs, connective tiss…
- DEXA A low-radiation X-ray scan that measures body composition with high accuracy — the clinica…
- BIA A body-composition method that estimates fat-free mass by measuring the body's resistance …