Mifflin-St Jeor vs Harris-Benedict: Which BMR Formula Is More Accurate?

A Century Apart

Two equations dominate BMR calculation in nutrition and clinical practice. Harris-Benedict, published in 1919, held the field for over seventy years. Mifflin-St Jeor, published in 1990, was specifically designed to correct its predecessor's known flaws. Understanding both — their origins, their formulas, and their accuracy — helps you know which number to trust.

Harris-Benedict (1919): The Original

James Arthur Harris and Francis Gano Benedict published their landmark study at the Carnegie Institution of Washington in 1919. They measured the basal metabolism of 239 men and 103 women using a respiration calorimeter — a technology that was cutting-edge at the time. From these measurements, they derived regression equations relating BMR to weight, height, age, and sex.

The original Harris-Benedict equations (metric) are:

These equations were revised in 1984 by Roza and Shizgal using a larger dataset, and the revised version (sometimes called the "revised Harris-Benedict") is what most modern calculators that still use this formula actually implement. The structure and coefficients differ slightly from the 1919 original.

Mifflin-St Jeor (1990): The Modern Standard

By the 1980s, evidence was accumulating that Harris-Benedict systematically overestimated BMR in modern populations — likely because the 1919 sample was drawn from a population with higher average physical activity and different body composition than late 20th-century adults. Mark Mifflin, Sachiko St Jeor, and colleagues set out to derive a more accurate equation using contemporary measurement technology.

Their 1990 study measured resting energy expenditure via indirect calorimetry in 498 healthy adults (251 men, 247 women), ranging from normal weight to obese. The resulting equations are structurally simpler than Harris-Benedict:

Note how the formulas share identical structure and coefficients — the only difference is the final constant (+5 for males, −161 for females). This elegance reflects a cleaner regression fit to the data.

Worked Comparison

Consider a 40-year-old man, 80 kg, 178 cm:

The difference here is approximately 164 kcal — nearly 10%. Over a week, that gap could represent more than 1,000 kcal of miscalculated intake, enough to meaningfully affect weight loss or gain projections.

Accuracy: What the Studies Show

The most comprehensive head-to-head comparison was published by Frankenfield, Roth-Yousey, and Compher in the Journal of the American Dietetic Association in 2005. They reviewed 29 studies comparing predictive equations against indirect calorimetry (the gold standard) in healthy adults. Their findings:

• Mifflin-St Jeor was the most accurate equation for estimating resting metabolic rate, predicting measured values within 10% for approximately 82%of non-obese subjects.
• Harris-Benedict overestimated RMR by an average of about 5% in non-obese adults and by up to 8% in obese subjects.
• Both equations perform less well in obese populations, where body composition (high fat to lean mass ratio) diverges more from the sample used to derive the equations.

This ~5% average overestimation from Harris-Benedict translates to roughly 75–120 extra kcal per day for a typical adult — small enough to be invisible week-to-week, but meaningful over months of structured dieting.

When Harris-Benedict Is Still Used

Despite Mifflin-St Jeor's superior accuracy, Harris-Benedict remains in use for several reasons:

• Clinical inertia: Many clinical nutrition protocols and hospital systems were built around Harris-Benedict and have not been updated. Registered dietitians trained before 2000 often learned Harris-Benedict as the standard.
• Research continuity: Studies that began using Harris-Benedict often continue to do so for consistency within the research program, even as newer equations have emerged.
• Specific populations: Some research suggests Harris-Benedict may perform comparably to Mifflin for certain subgroups, and the difference is small enough in practice that either equation is a reasonable starting point given the inherent variability of individual metabolism.

Why Mifflin Became the Clinical Standard

In 2005, the American Dietetic Association (now the Academy of Nutrition and Dietetics) formally recommended the Mifflin-St Jeor equation as the preferred method for estimating resting energy expenditure in healthy adults. This endorsement, backed by the systematic review evidence, shifted clinical practice. Most modern TDEE calculators, nutrition apps, and clinical decision tools now default to Mifflin-St Jeor.

The practical implication: if you have previously used a TDEE or BMR calculator and are unsure which formula it used, check — and if it used Harris-Benedict, your estimate may have been running 5–10% high. Our calculator uses Mifflin-St Jeor by default.