How Accurate Is the Mifflin-St Jeor Equation?

The Mifflin-St Jeor equation is the most accurate of the common resting-metabolism formulas — a 2005 systematic review found it came within 10% of measured resting metabolism for more people than any other standard equation. It is still an estimate, and can be off by more than 10% for some individuals, so treat the number as a well-grounded starting point.

By Ash · Updated 5 July 2026

How accurate is it, exactly?

In the 2005 systematic review that established it as the default, Frankenfield and colleagues compared the four most common resting-metabolic-rate equations — Mifflin-St Jeor, Harris-Benedict, Owen, and WHO/FAO/UNU — against rates measured directly by indirect calorimetry. The Mifflin-St Jeor equation was the most reliable: it predicted measured resting metabolism to within 10% for more people, both non-obese and obese, than any other equation, and it had the narrowest range of error. That is why the Academy of Nutrition and Dietetics recommends it. The review's authors were also clear that noteworthy errors remain when the equation is applied to individuals — being the best of the group is not the same as being exact.

How was its accuracy measured?

Accuracy here means agreement with indirect calorimetry — measuring the oxygen a person actually consumes at rest to work out their real resting energy expenditure, then comparing it with the formula's prediction. The Mifflin-St Jeor equation itself was built this way: its 1990 paper derived the formula from indirect-calorimetry measurements on 498 healthy adults. A formula is judged "accurate" when its prediction lands within about 10% of that measured value.

When is it least accurate?

The equation uses only weight, height, age and sex — not body composition — so it is least reliable for people whose muscle-to-fat ratio is unusual for their weight. A very muscular person carries more calorie-hungry lean tissue than the formula assumes and may be under-estimated; a person with a lot of body fat at the same weight may be over-estimated. It is also less validated for older adults and for populations very different from its original derivation sample. This is a real limitation of any weight-based formula, and the honest reason to treat the number as an estimate rather than a measurement. (Body-composition-based formulas exist, but they need an accurate body-fat percentage most people don't have — which brings its own error.)

Who was it built on?

The equation was derived from 498 healthy adults — 251 men and 247 women, aged 19 to 78, ranging from normal weight to obese — measured at the University of Nevada School of Medicine. That broad, modern sample is exactly why it out-performs the 1919 Harris-Benedict equation, which was built on a smaller, leaner, early-20th-century group. Even so, the model explained about 71% of the variation in resting metabolism between people, which means roughly a third of the differences come from factors weight, height, age and sex don't capture.

Should you trust the number?

Trust it as a well-grounded starting point, not a precise target. The most reliable figure is always your own real-world result: set your calories from the equation's estimate, then track your weight over two to three weeks and adjust. You can see how far two respected formulas can diverge for the same person in our 840-profile study, or run your own number on the Mifflin-St Jeor calculator.