If you've tried a TDEE or calorie calculator, you've encountered the question: which BMR formula should I use? The answer matters — two formulas applied to the same person can produce results that differ by 150–300 kcal per day, which compounds significantly over weeks of dieting.
This article compares all four widely-used BMR prediction equations — Mifflin-St Jeor, Harris-Benedict (revised), Schofield, and Katch-McArdle — including what the research says about their accuracy, who each formula suits best, and why no formula is a perfect match for every individual.
Basal Metabolic Rate (BMR) is the estimated number of calories your body requires at complete rest to sustain basic physiological functions — breathing, circulation, cellular repair, and thermoregulation. It represents the largest component of total daily energy expenditure (TDEE) for most sedentary to moderately active individuals, typically accounting for 60–75% of total calories burned.
The 4 BMR Formulas
1. Mifflin-St Jeor (1990)
Female: BMR = (10 × kg) + (6.25 × cm) − (5 × age) − 161
The Mifflin-St Jeor equation was developed by a research team at the University of Nevada using indirect calorimetry measurements in 498 healthy adults aged 19–78 years. It was designed to improve on the accuracy limitations of earlier formulas when applied to contemporary populations.
A 2005 systematic review published in the Journal of the American Dietetic Association (Frankenfield et al.) evaluated multiple predictive equations against measured resting metabolic rate and found the Mifflin-St Jeor formula to be the most accurate for the majority of healthy non-obese and obese adults. This finding has been replicated in subsequent independent studies, and the equation is currently the preferred formula recommended by the Academy of Nutrition and Dietetics (AND) for estimating resting energy expenditure in clinical and non-clinical settings.
2. Harris-Benedict (Revised, 1984)
Female: BMR = 447.593 + (9.247 × kg) + (3.098 × cm) − (4.330 × age)
The original Harris-Benedict equation, published in 1918, was among the first attempts to mathematically model basal metabolism. It was developed from measurements in 239 subjects and remained the standard for decades. In 1984, Roza and Shizgal published revised coefficients using a larger dataset and more precise calorimetry equipment, which is the version now commonly encountered in calculators.
Research indicates the revised Harris-Benedict formula is reasonably accurate for many adults, but tends to overestimate BMR modestly compared to Mifflin-St Jeor — particularly in individuals with overweight or obesity. For very lean or very heavy individuals, these discrepancies can become clinically meaningful. Despite this limitation, Harris-Benedict remains widely used and is a reasonable estimate for the general population.
3. Schofield (1985)
18–30 yrs: BMR = (15.057 × kg) + 692.2
30–60 yrs: BMR = (11.472 × kg) + 873.1
60+ yrs: BMR = (11.711 × kg) + 587.7
The Schofield equation uses age-banded regression coefficients applied to body weight alone — it does not incorporate height. It was derived from a large meta-analysis of over 7,000 BMR measurements compiled from studies spanning several decades. The World Health Organization subsequently adopted these equations in its 1985 Technical Report Series (No. 724) on energy and protein requirements, making Schofield the standard reference in global public health and clinical nutrition contexts.
Because Schofield relies only on weight (not height), it is less precise for individuals at the extremes of height. It also tends to slightly overestimate BMR in affluent Western populations, since much of its underlying data came from tropical and lower-income country samples. Despite this, it remains important in institutional and WHO-referenced nutrition assessment frameworks.
4. Katch-McArdle
Where LBM (kg) = body weight × (1 − body fat fraction)
Example: 80 kg person with 20% body fat → LBM = 80 × 0.80 = 64 kg → BMR = 370 + (21.6 × 64) = 1,752 kcal
The Katch-McArdle formula is the only one of the four that is based on lean body mass (LBM) rather than total body weight. The rationale is physiologically sound: metabolically active tissue — primarily muscle and organ mass — drives resting energy expenditure, while stored body fat contributes very little to BMR per kilogram. This means two people with identical total weight but different body compositions will have meaningfully different BMRs.
For athletes, bodybuilders, and lean individuals, Katch-McArdle can provide a more personalised estimate than the weight-based formulas. However, its accuracy depends entirely on the accuracy of the body fat percentage input. If body fat % is estimated from a bathroom scale (bioelectrical impedance) rather than DEXA or an accurate skinfold protocol, the resulting BMR figure may be no more reliable than simpler formulas.
Accuracy Comparison at a Glance
| Formula | Inputs Required | Best For | Accuracy Notes |
|---|---|---|---|
| Mifflin-St Jeor | Weight, height, age, sex | Most healthy adults | Recommended Most validated in systematic reviews |
| Harris-Benedict | Weight, height, age, sex | General population | Slight overestimate Especially in heavier individuals |
| Schofield | Weight, age, sex | WHO/clinical contexts | WHO standard No height adjustment |
| Katch-McArdle | Lean body mass (body fat %) | Lean/athletic individuals | Best if BF% known Accuracy depends on BF% precision |
Why All BMR Formulas Have a Margin of Error
Even the most validated formula carries an inherent margin of error. Research consistently shows that predictive equations — when compared against gold-standard indirect calorimetry (measuring actual oxygen consumption and CO₂ production) — can vary by ±10–15% or more in individual cases.
This is not a flaw of any particular formula; it reflects biological variability that no simple mathematical model can fully capture. Factors that influence actual metabolic rate but are not included in any standard equation include:
- Genetics and individual metabolic variation — two people with identical weight, height, age, and sex can have meaningfully different BMRs
- Thyroid and hormonal status — thyroid hormones directly regulate metabolic rate; subclinical hypothyroidism, for example, can suppress BMR by 15–30%
- Muscle fiber composition — slow-twitch fibres are more metabolically active at rest than fast-twitch fibres
- Adaptive thermogenesis — prolonged caloric restriction causes metabolic downregulation beyond what weight loss alone would predict
- Recent weight loss history — individuals who have recently lost significant weight may have a BMR lower than predicted for their current weight
For this reason, any calculated BMR or TDEE figure is best treated as a starting estimate. The most reliable approach in practice is to track actual intake and weight change over 2–4 weeks and adjust calorie targets empirically based on observed results.
In clinical nutrition settings — such as calculating nutritional requirements for hospitalised patients or pre-operative assessment — indirect calorimetry remains the reference standard when precision is critical. Predictive equations are used when direct measurement is not available or practical. For healthy individuals using calculators for weight management goals, a ±15% understanding of the estimate's accuracy is a clinically reasonable and honest framing.
Which Formula Should You Use?
For most people: Mifflin-St Jeor. It is the most rigorously validated across diverse adult populations and is the default recommendation in major dietetics literature. If you're using a TDEE calculator without specific reasons to do otherwise, Mifflin-St Jeor gives you the most reliable starting estimate.
If you know your body fat percentage accurately: Katch-McArdle. If you've had a DEXA scan or a reliable skinfold assessment — not a bathroom scale bioimpedance reading — Katch-McArdle will account for your actual metabolically active tissue and may provide a more personalised result, particularly for athletes with high muscle mass relative to body weight.
For WHO-referenced or institutional contexts: Schofield. If you're working within a healthcare or public health framework that references WHO dietary guidelines, the Schofield equation is the appropriate standard.
Harris-Benedict is a reliable secondary option when Mifflin-St Jeor is unavailable, and remains broadly accurate for most of the population. Its historical prevalence means it is still widely referenced in older literature and some clinical tools.
Comparing All 4 Formulas on the Same Person
To illustrate how results can diverge, here is an example calculation for a 35-year-old male, 80 kg, 178 cm tall, with an estimated 18% body fat:
- Mifflin-St Jeor: (10 × 80) + (6.25 × 178) − (5 × 35) + 5 = 1,743 kcal
- Harris-Benedict: 88.362 + (13.397 × 80) + (4.799 × 178) − (5.677 × 35) = 1,816 kcal
- Schofield (30–60 male): (11.472 × 80) + 873.1 = 1,791 kcal
- Katch-McArdle (LBM = 80 × 0.82 = 65.6 kg): 370 + (21.6 × 65.6) = 1,787 kcal
In this example, the range spans from 1,743 to 1,816 kcal — a difference of 73 kcal per day, or approximately 510 kcal per week. Harris-Benedict estimates the highest, while Mifflin-St Jeor produces the lowest result. At a 500 kcal daily deficit, a 73 kcal discrepancy in your baseline estimate translates to a meaningful difference in actual progress over weeks — demonstrating why formula choice matters in practice.
More Free Tools on BodyMetric
Once you have your BMR and TDEE, the following calculators help you act on that number:
- Calorie Deficit Calculator — find a safe daily calorie target with minimum-threshold safety checks
- Macro Calculator — protein, carbs, and fat targets across 5 dietary approaches
- Calorie Surplus Calculator — muscle gain targets with estimated muscle-to-fat ratio
- Weight Loss Timeline Calculator — week-by-week projections with metabolic adaptation simulation
- BMI Calculator — WHO classification with healthy weight range
- Body Fat % Calculator — US Navy method using body measurements
- Ideal Weight Calculator — four clinical formulas (Devine, Robinson, Miller, Hamwi)
- Water Intake Calculator — daily hydration target adjusted for activity, climate, and pregnancy
This article is provided for general educational and informational purposes only. BMR calculations are estimates and do not constitute medical or dietetic advice. Individual metabolic rates vary considerably and are influenced by factors beyond the scope of any predictive equation. If you have a medical condition affecting metabolism (such as thyroid disease, diabetes, or a history of eating disorders), are pregnant or breastfeeding, or have specific clinical nutritional requirements, please consult a qualified healthcare professional or registered dietitian before making significant changes to your caloric intake.
References
- Mifflin, M. D., St Jeor, S. T., Hill, L. A., Scott, B. J., Daugherty, S. A., & Koh, Y. O. (1990). A new predictive equation for resting energy expenditure in healthy individuals. American Journal of Clinical Nutrition, 51(2), 241–247.
- Frankenfield, D., Roth-Yousey, L., & Compher, C. (2005). Comparison of predictive equations for resting metabolic rate in healthy nonobese and obese adults: A systematic review. Journal of the American Dietetic Association, 105(5), 775–789.
- Harris, J. A., & Benedict, F. G. (1918). A biometric study of human basal metabolism. Proceedings of the National Academy of Sciences, 4(12), 370–373.
- Roza, A. M., & Shizgal, H. M. (1984). The Harris Benedict equation reevaluated: Resting energy requirements and the body cell mass. American Journal of Clinical Nutrition, 40(1), 168–182.
- Schofield, W. N. (1985). Predicting basal metabolic rate, new standards and review of previous work. Human Nutrition: Clinical Nutrition, 39(Suppl. 1), 5–41.
- Food and Agriculture Organization of the United Nations, World Health Organization, & United Nations University. (1985). Energy and protein requirements: Report of a joint FAO/WHO/UNU expert consultation (Technical Report Series No. 724). World Health Organization.
- McArdle, W. D., Katch, F. I., & Katch, V. L. (2015). Exercise physiology: Nutrition, energy, and human performance (8th ed.). Wolters Kluwer Health/Lippincott Williams & Wilkins.
- Weijs, P. J. (2008). Validity of predictive equations for resting energy expenditure in US and Dutch overweight and obese class I and II adults aged 18–65 y. American Journal of Clinical Nutrition, 88(4), 959–970.