The Mifflin St Jeor equation

The equation, worked through

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

For a man of 30 who is 170 cm and 70 kg: (10 × 70) + (6.25 × 170) − (5 × 30) + 5 = 700 + 1,062.5 − 150 + 5 = 1,618 kcal a day.

In pounds and inches the same equation becomes 4.536 × weight in lb + 15.875 × height in inches − 5 × age, then +5 or −161. The numbers are the metric ones converted, so you get the same answer.

This is energy at rest. Everything you do, from walking to digesting, comes on top, which is what the TDEE calculator adds.

What the 1990 study did

Mifflin, St Jeor and four colleagues, led from the University of Nevada School of Medicine, measured the resting energy use of 498 healthy adults: 251 men and 247 women, aged 19 to 78, average age 45. Nearly half of them, 234, were obese. Resting energy was measured by indirect calorimetry, which works out how much energy the body is using from the oxygen it takes in and the carbon dioxide it breathes out.

They then fitted an equation to those measurements. Weight, height, age and sex together explained 71% of the differences in resting energy between the people they measured. Adding how heavy someone was for their height, or where they carried their fat, did not significantly improve the predictions, so neither is in the equation.

The familiar version is a rounded one

The equation the paper actually fitted is 9.99 × weight + 6.25 × height − 4.92 × age + 166 for men − 161. The authors rounded it to the familiar 10, 6.25 and 5, split it by sex, and reported that this did not affect how well it predicted. The difference is a few kilocalories:

If two calculators using "Mifflin St Jeor" give you answers a few kilocalories apart, one of them may be using the unrounded form. Neither is wrong.

What it found about Harris Benedict

Before 1990 the standard was the Harris Benedict equations, from measurements Harris and Benedict published in 1918 and 1919. Tested against the 498 people in the new study, Harris Benedict overestimated their measured resting energy by 5% on average. As a sense of scale, 5% of 1,618 kcal is about 81 kcal a day, before activity multiplies it.

The best single predictor was fat-free mass

On its own, fat-free mass, everything you weigh except fat, predicted resting energy better than weight did: it explained 64% of the differences between people, against 56% for weight. That matches Cunningham's 1980 reanalysis of the Harris and Benedict data, which found lean mass was the single predictor of basal rate.

The equation uses weight, most likely for a practical reason: few people have a measured fat-free mass, and everybody has a weight. If you do have a body fat figure you trust, theBMR calculator shows a second estimate worked from lean mass.

How it has held up

In 2005, a systematic review in the Journal of the American Dietetic Association, using the association's own evidence analysis method, looked at the published tests of the four equations most used in practice: Harris Benedict, Mifflin St Jeor, Owen and the WHO, FAO and UN University equation. Mifflin St Jeor predicted resting rate to within 10% of the measured figure in more people than any of the others, obese or not, and had the narrowest spread of errors.

Where it misses

The same review also sets out the limits. All four equations can produce noteworthy errors for an individual, because two people with the same weight, height, age and sex can burn different amounts at rest. And older adults and ethnic minorities living in the US were underrepresented both in building the equations and in testing them, so the authors advise a high level of suspicion about accuracy for those groups.

If the figure matters, for example in a clinical setting, resting energy can be measured with indirect calorimetry, which is what the review recommends when an equation is failing somebody.

Why this site uses it

Because of all four equations in the 2005 review it had the best record for both obese and non-obese adults, and because it needs only four things everyone knows about themselves. TheBMR, TDEE,macro andcalorie deficit calculators all start from it. The one exception is the optional second line on the BMR calculator, which uses Katch McArdle when you give it a body fat figure.