Body Health CalculatorMifflin-St Jeor vs Harris-Benedict: Which BMR Formula Is Better?
Mifflin-St Jeor is more accurate for modern adults and is the default in most clinical and fitness settings. Harris-Benedict remains useful for comparison but often overestimates resting calories by 5 to 10 percent.
Last updated: September 2026
Markdown version for AI agents and developers
Summary
Mifflin-St Jeor and Harris-Benedict are the two most cited predictive equations for basal metabolic rate. Both estimate resting calorie needs from weight, height, age, and sex without direct calorimetry. They differ in coefficients and the populations used to derive them.
Evidence from the past three decades favors Mifflin-St Jeor for general adult populations. Harris-Benedict, especially the revised 1984 version, still appears in textbooks and older clinical protocols but tends to overestimate expenditure in modern cohorts.
Mifflin-St Jeor vs Harris-Benedict
| Mifflin-St Jeor (1990) | Harris-Benedict revised (1984) | |
|---|---|---|
| Men formula base | 10×kg + 6.25×cm − 5×age + 5 | 88.362 + 13.397×kg + 4.799×cm − 5.677×age |
| Women formula base | 10×kg + 6.25×cm − 5×age − 161 | 447.593 + 9.247×kg + 3.098×cm − 4.330×age |
| Validation era | 1990 obese and non-obese adults | 1984 updated from 1919 original |
| Typical error vs measured REE | Within 10% for ~70% of adults | Overestimates ~5 to 10% in many modern samples |
| Common use today | Default in most apps and clinics | Historical comparison, some legacy systems |
When to use Mifflin-St Jeor
Choose Mifflin-St Jeor for general BMR and TDEE estimation when you do not have a measured body fat percentage. The Academy of Nutrition and Dietetics and multiple validation reviews recommend it as the first-line predictive equation for healthy adults.
The equation was derived from 498 healthy subjects with indirect calorimetry and tested in separate obese and non-obese groups. It performs well across BMI ranges and both sexes without the systematic overprediction seen in older formulas.
Most modern calorie calculators, hospital nutrition software, and fitness apps default to Mifflin-St Jeor for this reason. If you are starting a new diet plan and want one consistent baseline, this is the standard choice.
When to use Harris-Benedict
Harris-Benedict still has a role when you need to compare against legacy protocols, older research papers, or institutional guidelines that were written around the revised 1984 equation. Some long-running nutrition studies report Harris-Benedict values, and matching their formula keeps comparisons valid.
Running both equations side by side shows how sensitive your calorie target is to formula choice. If Harris-Benedict and Mifflin-St Jeor differ by more than 150 kcal, treat the range as uncertainty rather than trusting either number exactly.
The original 1919 Harris-Benedict equation is obsolete and should not be used. Always confirm you are using the 1984 revision with updated coefficients when comparing to published Harris-Benedict results.
What validation studies show
Frankenfield and colleagues pooled multiple studies comparing predictive equations to measured resting energy expenditure. Mifflin-St Jeor predicted within 10 percent of measured REE for the highest proportion of individuals across weight categories.
Harris-Benedict ranked lower in accuracy for contemporary American and European adults. Overestimation matters because diet plans built on inflated BMR can stall fat loss or produce slower-than-expected weight change, leading people to cut calories further than necessary.
Neither equation accounts for lean mass directly. Athletes with known body fat percentage may get better results from Katch-McArdle, which uses lean body mass instead of total weight.
Practical workflow for diet planning
Start with Mifflin-St Jeor BMR. Multiply by an activity factor (1.2 sedentary through 1.9 very active) to estimate TDEE. Adjust intake based on real-world weight trend over two to four weeks.
If you are comparing historical Harris-Benedict targets to a new Mifflin-based plan, expect the new BMR to run 50 to 200 kcal lower for many adults. That is usually a correction, not a sign you were under-eating before.
Recalculate after significant weight change because both equations use body weight as a primary input. A 20 kg loss can reduce predicted BMR by 100 to 150 kcal independent of metabolic adaptation.
Online macro calculators that let you toggle between formulas are useful for sensitivity analysis. If your deficit plan only works under Harris-Benedict but not Mifflin-St Jeor, the plan may be too aggressive.
Limitations
Predictive equations cannot replace indirect calorimetry when precise resting energy expenditure is required for ICU nutrition, complex metabolic disorders, or research.
Thyroid disease, corticosteroids, beta blockers, and major illness alter true metabolic rate in ways no anthropometric formula captures.
Age, sex, and weight explain much but not all of resting metabolism. Treat any BMR output as an estimate to refine with feedback, not a fixed biological constant.
Postmenopausal women and adults over 65 may see larger gaps between predicted and measured resting expenditure. Clinical dietitians sometimes adjust predictive outputs using measured RMR when available.