A calculator gives you one number. A gym or app gives you another. One says BMR, the other RMR. Which is right, and which should you eat by?
This guide gives you the honest answer: how each number is measured, how far apart they really are, how good the equations are, and how to estimate your own resting metabolic rate at home.
- BMR and RMR measure the same thing: the energy your body burns while doing nothing. The only difference is how strict the test conditions are.
- The gap is small and not fixed. The popular "RMR is 10% higher" claim is hard to trace to a real measurement. In one study, simply changing the test time and fasting window moved RMR by about 100 kcal a day (6%).
- Your online "BMR" is really an RMR estimate. The Mifflin-St Jeor equation was built from resting measurements.
- Equation error is much bigger than the BMR/RMR gap. Even the best formula misses by more than 10% for roughly one person in five.
- For eating and dieting, use one resting estimate, multiply it by your activity level, then adjust using your real weight trend.
BMR and RMR in plain words
Your body spends energy all day just to stay alive: your heart beats, your lungs move, and your cells repair themselves. For most people this resting cost is the biggest part of the daily burn, around 60 to 70% of the total (Poehlman, 1989; Plaza-Florido & Alcantara, 2023).
Basal metabolic rate (BMR)
BMR is that baseline cost measured under the strictest rules: you have just woken up, you have not eaten since the night before, you have not exercised, and you are calm and lying still.
Resting metabolic rate (RMR)
RMR is the same idea with looser rules. You still rest and skip food and exercise beforehand, but you don't need to sleep at the lab. You may also see it called resting energy expenditure (REE).
"Resting and basal metabolic rate (BMR) are similar and only differ in that BMR is usually measured in the morning, after an overnight fast, no exercise for the previous 24 h, free from emotional stress, familiar with the apparatus, and the subject completely rested." McMurray et al., Medicine & Science in Sports & Exercise, 2014
Even feeling nervous about the mask can nudge the reading. That is how sensitive this number is.
How each one is measured in a lab
Most articles stop at "BMR is stricter." Here is what a real test looks like, based on the evidence reviews dietitians use.
The method: indirect calorimetry
A machine measures the oxygen you breathe in and the carbon dioxide you breathe out while you wear a mask or lie under a clear hood. Your body uses oxygen to burn fuel, so your gas exchange shows how much energy you use. This is indirect calorimetry, the reference method in clinics and research.
What an RMR test asks of you
The Academy of Nutrition and Dietetics has reviewed the evidence on testing twice, in 2006 and 2015. Their best-practice advice for a valid RMR test includes:
- Fasting: at least 7 hours without food (Fullmer et al., 2015).
- Rest first: about 30 minutes lying down in a quiet, dim, comfortable room, with no reading, music, or fidgeting (Fullmer et al., 2015). The earlier review found 10 to 20 minutes enough for everyday activity effects to fade (Compher et al., 2006).
- No stimulants: caffeine and nicotine affect the result. Their effects last longer than 240 and 140 minutes respectively, though the exact duration is unknown (Fullmer et al., 2015).
- No hard exercise beforehand: moderate or vigorous exercise has a longer carry-over, so labs control it for hours before the test (Compher et al., 2006).
- A short, steady reading: a 10-minute measurement, discarding the first 5 minutes, with the last 5 minutes stable (coefficient of variation under 10%) gives an accurate result (Compher et al., 2006).
What a true BMR test adds
A BMR test adds the stricter conditions from the quote above, which often means sleeping at the testing site. That is why almost every "metabolism test" you can book is an RMR test.
How big is the gap, really?
This is where most articles go wrong. You will read that RMR is "about 10%," "5 to 10%," or even "10 to 20%" higher than BMR. These claims disagree, and the peer-reviewed reviews that define the terms give no fixed percentage.
The reason is simple. RMR relaxes BMR's rules, and each lab relaxes them differently. A 4-hour fast and a 10-hour fast are both "RMR," so there is no single gap to measure.
What real measurements show
The best direct evidence comes from a study of 37 healthy adults at the University of Colorado (Haugen et al., 2003). Each person had four RMR tests: two in the morning after a 12-hour fast, and two in the afternoon after a 4-hour fast.
- Afternoon readings came out about 99 kcal per day higher, roughly 6%.
- Repeat tests under the same conditions matched closely (correlation r = 0.93).
A separate review found that RMR naturally varies by about 5 to 8% between tests in the same person (Donahoo et al., 2004). So the strict-vs-relaxed difference is about the size of normal day-to-day noise. Treat BMR and RMR as nearly the same number.
BMR vs RMR vs TDEE
TDEE (total daily energy expenditure) is a different number: everything you burn in a day, adding digestion and movement to your resting burn. See our guide on how to read BMR and TDEE on a calorie calculator.
| Feature | BMR | RMR | TDEE |
|---|---|---|---|
| What it means | Minimum energy to stay alive, under strict conditions | Energy used at rest, under practical conditions | All energy used in 24 hours |
| Test conditions | Morning, on waking, overnight fast, no exercise for 24 h, fully calm | At least about 7 h fast, 20–30 min rest, no stimulants | Not a resting test; measured in research with doubly labelled water |
| Includes digestion? | No | Very little, if the fast was long enough | Yes (the thermic effect of food, around 10% of the day) |
| Includes movement? | No | No | Yes: exercise, walking, chores, fidgeting |
| Typical size | Lowest | Close to BMR, often slightly higher | Roughly 1.4 to 2.0 times your resting burn for most people |
| Where you see it | Research labs | Clinics, gyms, sports labs, most calculators | Calorie apps and meal plans |
| Best use | Science and comparison | Your starting point for daily needs | Your actual eating target |
Table sources: Poehlman (1989), Westerterp (2004, 2013), FAO/WHO/UNU (2004).
Protein costs the most to digest, as our guide to the thermic effect of food explains.
How the equations estimate your number
Few people ever get a lab test, so calculators use equations built from people who were tested. Here are the four you will meet most often. For more detail, see every BMR formula explained.
Mifflin-St Jeor (1990)
Researchers built this equation from 498 healthy adults aged 19 to 78, including normal-weight and obese people. They measured each person with indirect calorimetry (Mifflin et al., 1990). The paper's title says "resting energy expenditure," so it predicts RMR, even when a website labels it BMR.
Women: (10 × weight kg) + (6.25 × height cm) − (5 × age) − 161
Harris-Benedict (1919, revised 1984)
This is the oldest classic. Roza and Shizgal re-checked the original data in 1984 and reported a precision of about ±14% for healthy people. The Mifflin study found that the original version overestimated measured resting burn by about 5%.
Women: 447.593 + (9.247 × kg) + (3.098 × cm) − (4.330 × age)
Cunningham (1980 and 1991)
Cunningham uses lean body mass (weight minus fat). The 1980 version is 500 + 22 × lean mass in kg. The 1991 revision, based on a review of many studies, is 370 + 21.6 × fat-free mass and explained 65 to 90% of the differences between people (Cunningham, 1991).
Katch-McArdle
A detail most sites miss: the "Katch-McArdle" formula on calculators is 370 + 21.6 × lean mass, the same as Cunningham's 1991 equation. Both need a body fat estimate, and any error in it carries straight into the result.
How accurate are BMR calculators?
This is the question that matters most, and the honest answer is: good for groups, rough for individuals. Researchers usually call a prediction "accurate" when it lands within 10% of the measured value.
| Equation | Inputs | What studies found | Best suited to |
|---|---|---|---|
| Mifflin-St Jeor | Weight, height, age, sex | Within ±10% in more people than other common equations, with the narrowest error range (Frankenfield et al., 2005). Missed by more than 10% in 22% of 130 adults (Frankenfield et al., 2003). Accurate in 79% of US adults with BMI 25–40 (Weijs, 2008). | Most adults, including people with overweight |
| Harris-Benedict | Weight, height, age, sex | Missed by more than 10% in 33% of adults (Frankenfield et al., 2003). Original version ran about 5% high (Mifflin et al., 1990). Precision about ±14% (Roza & Shizgal, 1984). | Comparison with older research |
| Cunningham | Lean body mass | Most accurate of the tested equations in endurance athletes (Thompson & Manore, 1996). Confirmed as accurate in 90 recreational athletes (ten Haaf & Weijs, 2014). | Active, muscular people who know their body fat |
| Katch-McArdle | Lean body mass | Same formula as Cunningham (1991), so the same strengths and limits apply | Same as Cunningham |
Two more findings keep this honest:
- In athletes, Harris-Benedict, Mifflin, and several others were accurate for less than half of the group (ten Haaf & Weijs, 2014). Weight-based formulas tend to underestimate people with lots of muscle.
- One cross-validation study found individual errors (95% limits of agreement) of roughly ±314 to ±445 kcal a day across the equations tested, even when the group average looked fine (Flack et al., 2016).
"The Mifflin-St Jeor equation is more likely than the other equations tested to estimate RMR to within 10% of that measured, but noteworthy errors and limitations exist when it is applied to individuals." Frankenfield, Roth-Yousey & Compher, Journal of the American Dietetic Association, 2005
The same review warned that older adults and some ethnic groups were under-represented in the research, so give your result a wider margin if that applies to you. More reasons your number can drift are in why your BMR calculator might be wrong.
Worked example
Let's meet Sara. She is 35, 165 cm tall, weighs 70 kg, and a smart scale says she has about 30% body fat. That gives her about 49 kg of lean mass (70 × 0.70).
| Equation | Calculation | Result (kcal/day) |
|---|---|---|
| Mifflin-St Jeor | 700 + 1,031 − 175 − 161 | about 1,395 |
| Harris-Benedict (revised) | 447.6 + 647.3 + 511.2 − 151.6 | about 1,455 |
| Cunningham 1991 / Katch-McArdle | 370 + (21.6 × 49) | about 1,428 |
| Cunningham 1980 | 500 + (22 × 49) | about 1,578 |
Same woman, four numbers, a spread of about 180 kcal. None is "true" for sure. Here is how to read them:
- Pick one starting point. Mifflin-St Jeor (about 1,395) is the best-tested choice for most adults.
- Accept a band. A ±10% band gives roughly 1,255 to 1,535 kcal. Her measured RMR could sit anywhere in there.
- Add activity. Sara works at a desk and walks a little. The FAO/WHO/UNU sedentary range is a multiplier of 1.40 to 1.69. Using 1.5 gives a TDEE of about 2,090 kcal.
- Let her weight decide. If she eats about 2,090 kcal for a few weeks and her weight stays flat, that is her real maintenance, whatever the formula said.
Notice that we did not add 10% to "convert" BMR into RMR. Mifflin already estimates resting burn, so that would double-count.
How to estimate your RMR at home
You can't match lab precision at home, but you can get a solid working number in five steps.
- Measure your inputs well. Weigh yourself in the morning, after the bathroom and before breakfast. Use your real height, not the one on an old ID card.
- Choose your equation. Use Mifflin-St Jeor by default. If you are very muscular or train hard and have a decent body fat estimate (for example from a DEXA scan), also run Cunningham.
- Run the numbers. Our free BMR calculator does the math for you in seconds.
- Multiply by an honest activity level. Most office workers fit the 1.40–1.69 range; very active lifestyles reach about 2.0–2.4 (FAO/WHO/UNU, 2004; Westerterp, 2013).
- Test it for 2 to 4 weeks. Track your intake and morning weight, and watch the weekly average. If your weight drifts while you eat at "maintenance," adjust by 100 to 200 kcal and check again.
Step 5 is the part most guides skip, and it turns a guess into your own number. Your weight trend already includes your true resting burn, digestion, and movement.
Which number should you use? A checklist
- ✅ Planning meals or a diet? Use your resting estimate (RMR) × activity level = TDEE. Eat around TDEE to maintain.
- ✅ Want to lose weight? Start from TDEE, not BMR, and eat a modest amount below it. Our guide to how many calories to eat to lose weight walks through the math.
- ✅ Got a lab RMR result? Trust it over any equation, as long as you followed the prep rules.
- ✅ Muscular or an athlete? Compare Mifflin with Cunningham. Weight-only formulas often run low for you.
- ✅ Two apps disagree? They probably use different equations or activity levels. Pick one and stick with it.
- ✅ Older adult, or from a group under-represented in research? Use a wider error band and lean on your weight trend.
- ⚠️ Never "convert" a calculator BMR into RMR by adding a fixed percentage. Most calculators already give you a resting estimate.
- ⚠️ Unexplained changes? If your weight changes a lot without a clear reason, talk to a doctor. Thyroid problems and some medicines can affect metabolism.
Frequently asked questions
Is BMR or RMR better for weight loss?
RMR is more practical, and most calculator BMR results are RMR estimates anyway. What matters more is turning that number into TDEE and adjusting it with your real weight trend.
How much higher is RMR than BMR?
Usually only a little, with no fixed figure. In one controlled study, an afternoon test after a 4-hour fast came out about 99 kcal a day (6%) higher than a morning test after a 12-hour fast.
How do I figure out my resting metabolic rate?
The most accurate way is an indirect calorimetry test after at least a 7-hour fast and 30 minutes of rest. At home, use Mifflin-St Jeor, multiply by your activity level, and check it against your weight trend for 2 to 4 weeks.
What is the basal metabolic rate formula?
The most widely used is Mifflin-St Jeor: 10 × weight (kg) + 6.25 × height (cm) − 5 × age, then + 5 for men or − 161 for women. Harris-Benedict, Cunningham, and Katch-McArdle are common alternatives.
How accurate are BMR calculators?
The best equation, Mifflin-St Jeor, lands within 10% of a measured value for roughly 4 in 5 adults. That leaves about 1 in 5 with a bigger error, so treat the result as a starting point.
What is the difference between BMR and TDEE?
BMR (or RMR) is what you burn at rest. TDEE is everything you burn in a day, including digestion and movement, usually about 1.4 to 2 times your resting burn.
Sources
- Compher et al., 2006, J Am Diet Assoc
- Fullmer et al., 2015, J Acad Nutr Diet
- McMurray et al., 2014, Med Sci Sports Exerc
- Haugen et al., 2003, Am J Clin Nutr
- Donahoo et al., 2004, Curr Opin Clin Nutr Metab Care
- Mifflin et al., 1990, Am J Clin Nutr
- Roza & Shizgal, 1984, Am J Clin Nutr
- Cunningham, 1980, Am J Clin Nutr
- Cunningham, 1991, Am J Clin Nutr
- Frankenfield et al., 2003, J Am Diet Assoc
- Frankenfield et al., 2005, J Am Diet Assoc (systematic review)
- Weijs, 2008, Am J Clin Nutr
- Flack et al., 2016, J Acad Nutr Diet
- Thompson & Manore, 1996, J Am Diet Assoc
- ten Haaf & Weijs, 2014, PLoS One
- Poehlman, 1989, Med Sci Sports Exerc
- Plaza-Florido & Alcantara, 2023, Metabolites
- Westerterp, 2004, Nutr Metab
- Westerterp, 2013, Front Physiol
- FAO/WHO/UNU, Human energy requirements, 2004
Comments
Post a Comment