BMR

BMR — basal metabolic rate — is the energy required to keep a human body alive at complete physiological rest. It covers respiration, cardiac output, thermoregulation, cellular maintenance, and baseline organ function. For a sedentary adult, BMR is typically 60–70% of total daily energy expenditure.

The strict definition

True BMR is measured under a specific protocol: the subject has fasted for 12 hours, is lying down in a thermoneutral room (around 24°C), has not performed any physical activity, and the measurement happens immediately upon waking. Anything looser than this is technically measuring RMR (resting metabolic rate), which runs 5–10% higher.

How it's measured (in a lab)

Indirect calorimetry — measuring oxygen consumption and CO₂ production through a ventilatory hood or chamber — is the reference method. A research-grade measurement takes 20–40 minutes and produces a value accurate to within a few percent of actual energy expenditure.

How consumer apps estimate it

Every consumer tracking app — PlateLens, MyFitnessPal, Cronometer, MacroFactor, Lose It!, Yazio — plugs user height, weight, age, and sex into one of a handful of predictive equations. The three most common:

• Mifflin-St Jeor (1990): the current consumer-app default. Tends to slightly underestimate in lean athletic populations.
• Harris-Benedict (1919, revised 1984): still used in some older apps. Tends to overestimate in sedentary populations.
• Katch-McArdle: uses lean body mass instead of total weight. More accurate if the user knows their true body-fat percentage; less accurate if they've guessed it.

None of these is a measurement. All of them are population-derived regression equations with an r² somewhere around 0.70–0.80 against indirect calorimetry. That translates to a standard error of roughly ±5–10% for any individual prediction.

Why the number on your app is probably not your real BMR

Three reasons:

• Body composition isn't captured. Two people at the same height, weight, age, and sex can have meaningfully different fat-free mass — and fat-free mass is the strongest single predictor of BMR.
• Thyroid status, medication, and genetics aren't captured. Hypothyroidism, stimulants, GLP-1 agonists, and pregnancy all shift BMR; predictive equations assume population-average values.
• BMR adapts. Extended caloric deficit, weight loss, and very-low-calorie dieting can reduce BMR by 10–15% below predicted — a phenomenon called adaptive thermogenesis.

What to do with the number

Use the app-displayed BMR as a starting estimate for your TDEE calculation, then calibrate empirically by tracking intake against weight trend over 2–4 weeks. The number your app shows is the model's best guess; your actual maintenance calories are what they turn out to be.