Harris-Benedict vs Mifflin-St Jeor vs Katch-McArdle: Why BMR Formulas Disagree by 100-200 Calories

There isn't one BMR formula — there are several, developed decades apart using different study populations, and they can disagree by 100-200 calories for the same person, which matters when that difference represents the entire "calorie deficit" of a weight-loss plan

BMR calculators (including the one on this site) typically use a specific formula to estimate basal metabolic rate from height, weight, age, and sex — but multiple formulas exist, developed at different times using different study populations, and they don't always agree. Understanding which formula is being used, and why the differences exist, matters because a 100-200 calorie discrepancy between formulas can represent a meaningful fraction of a typical weight-management calorie target.

The Harris-Benedict equation: the original, and still widely referenced

Originally developed in 1919 (and revised in 1984 to update the coefficients based on more recent data) — the Harris-Benedict equation was, for decades, the most widely-used BMR estimation formula, and remains commonly referenced even today.

General form (revised version):

• Men: BMR = 88.362 + (13.397 × weight in kg) + (4.799 × height in cm) − (5.677 × age in years)
• Women: BMR = 447.593 + (9.247 × weight in kg) + (3.098 × height in cm) − (4.330 × age in years)

Limitation: the original (1919) study population was relatively small and may not represent the broader, more diverse population that BMR calculators are now applied to — the 1984 revision improved on this, but Harris-Benedict, even revised, has generally been found, in later validation studies, to overestimate BMR for some populations compared to more-recently-developed formulas.

The Mifflin-St Jeor equation: developed specifically to improve on Harris-Benedict

Developed in 1990, specifically as an attempt to provide a more accurate estimate than Harris-Benedict, using a larger, more contemporary study population.

General form:

• Men: BMR = (10 × weight in kg) + (6.25 × height in cm) − (5 × age in years) + 5
• Women: BMR = (10 × weight in kg) + (6.25 × height in cm) − (5 × age in years) − 161

Validation studies comparing Mifflin-St Jeor against measured (via indirect calorimetry — a direct measurement method, considered a "gold standard" for individual BMR measurement, though not practical for everyday, general-public use) BMR have generally found Mifflin-St Jeor to be more accurate, on average, across a broad population, compared to Harris-Benedict — which is part of why Mifflin-St Jeor has, over time, become the more-commonly-recommended formula in many clinical/nutritional contexts, though Harris-Benedict remains widely-referenced (including, sometimes, in older resources/tools that haven't updated to reflect more recent validation findings).

The Katch-McArdle formula: based on lean body mass, not total weight

A different approach entirely: rather than using total body weight (which includes both fat mass and lean mass), the Katch-McArdle formula uses lean body mass (total weight minus estimated fat mass) as its primary input:

BMR = 370 + (21.6 × lean body mass in kg)

The rationale: lean tissue (muscle, organs) is metabolically more active than fat tissue — two people with the same total weight, but different body-fat percentages, would have different amounts of metabolically-active lean tissue, and (the theory behind Katch-McArdle suggests) correspondingly different BMRs — **a formula based on total weight (Harris-Benedict, Mifflin-St Jeor) might not capture this difference as accurately as a formula based directly on the metabolically-relevant tissue (lean mass).

The practical limitation: Katch-McArdle requires knowing (or estimating) body-fat percentage — which, unlike height/weight/age/sex (the inputs to Harris-Benedict/Mifflin-St Jeor), isn't something most people know precisely without a specific body-composition measurement (covered in the previous article on body-fat measurement methods — DEXA, BIA, calipers) — each of which has its own accuracy limitations. Using Katch-McArdle with an inaccurate body-fat-percentage estimate could produce a BMR estimate that's less accurate than simply using Mifflin-St Jeor (which doesn't require this additional, potentially-error-prone input) — Katch-McArdle's theoretical advantage (accounting for body composition) is only realized if the body-fat-percentage input is itself reasonably accurate — for people without access to reasonably-accurate body-composition measurement, Mifflin-St Jeor may be the more practically reliable choice, despite not directly accounting for body composition.

Why formulas tend to diverge more for people at the extremes

For people with body compositions close to the "average" of the populations used to develop these formulas (which, broadly, skew toward "typical," non-extreme body compositions for the populations studied) — different formulas tend to produce relatively similar estimates — the gap between Harris-Benedict and Mifflin-St Jeor, for such individuals, might be modest (tens of calories, not hundreds).

For people at the extremes — very muscular individuals (high lean mass relative to total weight — athletes, bodybuilders), or individuals with substantially higher-than-average body-fat percentages — the gap between formulas can widen substantially — because formulas based on total weight (Harris-Benedict, Mifflin-St Jeor) don't distinguish between "high weight due to high muscle mass" and "high weight due to high fat mass" — both would produce the same "weight" input, and therefore the same BMR estimate from these formulas — despite potentially having meaningfully different actual BMRs (given the metabolic-activity difference between lean and fat tissue that Katch-McArdle attempts to account for).

For such individuals, Katch-McArdle (if a reasonably accurate body-fat-percentage is available) may provide a meaningfully more accurate estimate than total-weight-based formulas — this is precisely the population for whom the "theoretical advantage" of Katch-McArdle is most likely to translate into practical accuracy improvement, assuming the body-fat-percentage input itself is reasonably reliable.

Why the choice of formula matters for practical calorie planning

A BMR estimate is typically multiplied by an "activity factor" (covered in the original BMR article) to estimate TDEE (Total Daily Energy Expenditure) — and TDEE, in turn, informs calorie targets for weight loss/gain/maintenance (typically via a deficit/surplus relative to TDEE).

A 100-150 calorie difference in the underlying BMR estimate (which can occur between Harris-Benedict and Mifflin-St Jeor, particularly for individuals at the extremes discussed above) propagates directly into the TDEE estimate, and therefore into the calorie target — for weight-management plans that involve a deficit of, say, 500 calories/day — a 100-150 calorie discrepancy in the BMR estimate represents 20-30% of that deficit — a meaningful fraction, given that the entire premise of a 500-calorie deficit plan relies on that 500-calorie figure being reasonably accurate relative to the individual's actual energy expenditure.

*The practical takeaway: BMR/TDEE estimates from any formula are estimates, with inherent uncertainty — real-world outcomes (actual weight change over time, at a given calorie intake) provide more, and more individually-relevant, information than any single formula's initial estimate — using a formula's estimate as a starting point, and adjusting based on observed, individual results over several weeks, is generally more reliable than treating any single formula's output as a precise, fixed target — consistent with general guidance in this area, which emphasizes formulas as starting estimates, not precise, individually-calibrated figures.

How to use the BMR Calculator on sadiqbd.com

1. Use the default formula (Mifflin-St Jeor, generally the more contemporary/validated choice) for most purposes — providing a reasonable starting estimate for most body compositions
2. If you're at the "extremes" (very high muscle mass, or substantially higher-than-typical body-fat percentage) and have a reasonably reliable body-fat-percentage estimate (from DEXA, BIA, or careful caliper measurement, per the previous body-fat article) — consider whether a Katch-McArdle-style, lean-mass-based estimate might better reflect your specific situation
3. Treat any formula's output as a starting point — adjusting based on observed, individual results over time, rather than treating the initial estimate as precisely correct

Frequently Asked Questions

Which formula does this calculator use, and can I see results from multiple formulas? [This section can be customized based on the actual implementation — generally, clearly indicating which formula is used by default, and whether alternative formulas' results are also available for comparison, helps users understand what they're seeing and why it might differ from a different tool's output using a different formula.]

Is the BMR Calculator free? Yes — completely free, no sign-up required.

Try the BMR Calculator free at sadiqbd.com — calculate your basal metabolic rate using validated formulas.