NutritionDex

Metabolic Physiology

Mifflin-St Jeor Equation

Also known as: MSJ equation, Mifflin equation

The predictive equation for resting metabolic rate that most calculators default to, derived from body weight, height, age and sex.

By Marcus Chen · Former Fitness-Tech Product Lead ·

Key takeaways

  • For men: RMR = 10 x weight(kg) + 6.25 x height(cm) - 5 x age + 5. For women, the same expression with -161 instead of +5.
  • The sex term is the only structural difference, so at identical height, weight and age the predicted difference is exactly 166 kcal.
  • Published in 1990 on a more representative sample than Harris-Benedict, and it generally predicts closer to measured RMR in modern populations.
  • Prediction error for an individual is commonly cited around 10%, meaning a 1,600 kcal estimate carries a plausible range of roughly 1,440 to 1,760.
  • The activity multiplier applied afterwards introduces more error than the choice of equation does.

The Mifflin-St Jeor equation predicts resting metabolic rate from four inputs, and it is the default in most consumer calculators.

Men: RMR = (10 × kg) + (6.25 × cm) − (5 × age) + 5
Women: RMR = (10 × kg) + (6.25 × cm) − (5 × age) − 161

Multiply the result by an activity factor to reach TDEE.

Why it displaced Harris-Benedict

Harris-Benedict dates from 1919 and was derived from a small, narrow sample. Mifflin-St Jeor was published in 1990 from a larger and more representative one, and in modern populations it generally predicts closer to measured RMR — particularly at higher body weights, where Harris-Benedict tends to over-predict because it treats every kilogram as metabolically equivalent, and adipose tissue is not.

The practical gap between the two equations is usually 50–150 kcal.

The 166-calorie sex term

The only structural difference between the two forms is the constant: +5 against −161. So at identical height, weight and age, the equation predicts exactly 166 kcal of difference — and nothing more.

Larger observed differences between men and women in practice come from differences in average height and lean body mass, not from the equation treating the bodies differently. This is worth stating precisely because the point is routinely overstated in both directions.

Individual prediction error

Roughly 10% for an individual is the figure usually quoted. On a predicted 1,600 kcal that is a plausible band of about 1,440 to 1,760.

The sources of that error are exactly the things four inputs cannot capture: lean mass at a given body weight, organ mass, thyroid status, and how well adaptive thermogenesis has moved you from population average after a period of dieting.

The larger error is downstream

This is the part that makes the equation argument mostly beside the point. The activity multiplier applied after the equation ranges from 1.2 to 1.9, and adjacent steps in a standard table differ by roughly 200–280 kcal — more than the gap between any two prediction equations.

And the user selects that step, subjectively, with a documented tendency to over-rate their own activity. Someone training three times a week and sitting the rest of the time is closer to 1.375 than to 1.55, and picks 1.55.

Choose the equation in thirty seconds, choose the multiplier conservatively, then replace the whole estimate with three weeks of your own data.

Replacing the estimate

After three weeks of consistent logging and weighing, the following is strictly better than any equation, because it measures you:

maintenance = mean daily intake − (weight change in kg × 7700 ÷ 21)

Its advantage is that it absorbs your own logging bias. If you systematically under-record by 15%, the figure it returns is your maintenance measured on your own scale, which is the number you will actually be working with. Mifflin-St Jeor is the starting point; this is the replacement. See weekly average weight for how to get a usable weight change term.

Frequently asked

What is the Mifflin-St Jeor equation?

A predictive formula for resting metabolic rate from weight, height, age and sex. For men it is 10 times weight in kilograms, plus 6.25 times height in centimetres, minus 5 times age, plus 5; for women the same expression ends in minus 161. The result is what your body burns at rest, and it is multiplied by an activity factor to reach total daily energy expenditure. It is the default in most calculators because it predicts closer to measured RMR in modern populations than the older Harris-Benedict equation does.

How accurate is Mifflin-St Jeor for one person?

Around 10% for an individual, which on a predicted 1,600 kcal means a plausible range of roughly 1,440 to 1,760. The error comes from what four inputs cannot see: lean mass at a given body weight, organ mass, thyroid status, and any adaptive change from previous dieting. Treat the output as a starting point and replace it after three weeks with maintenance calculated from your own intake and weight trend, which measures you rather than the population.

Does the equation choice matter much?

Less than the activity multiplier applied afterwards. The gap between Mifflin-St Jeor and Harris-Benedict is typically 50 to 150 calories. The gap between adjacent activity factors in a standard table is 200 to 280 calories, and the user picks that step subjectively with a well-documented tendency to over-rate their own activity. Spend thirty seconds on the equation, choose the multiplier one step lower than feels right, and move on.

References

  1. Mifflin MD, St Jeor ST, Hill LA, Scott BJ, Daugherty SA, Koh YO. "A new predictive equation for resting energy expenditure in healthy individuals". American Journal of Clinical Nutrition , 1990 .
  2. 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