Resting Metabolic Rate Estimates

What resting metabolic rate is

Resting metabolic rate (RMR) is the energy the body uses at rest, usually given in calories per day. It can be measured, using indirect calorimetry, or it can be estimated by predictive equations, and estimation is by far the more common method. [1] A calculator that says "your BMR is X" is almost always the second kind: an estimate from a formula, not a measurement of you.

This page explains accuracy. It does not give calorie targets, and it does not tell anyone how much to eat. See Estimated Energy Expenditure and Calorie Burn for wearable "calories burned" figures.

Direct, estimated and calculated

How accurate are prediction equations?

A systematic review of equations for healthy non-obese and obese adults found four in common clinical use: Harris-Benedict, Mifflin-St Jeor, Owen and the WHO/FAO/UNU equation. [1] Of these, Mifflin-St Jeor was the most reliable, predicting RMR within 10% of the measured value in more people than any other equation. [1] That is a relative result: the same review concluded that noteworthy errors and limitations exist when an equation is applied to an individual. [1]

The review also noted that older adults and ethnic minorities living in the US were underrepresented both in developing the equations and in checking them. [1] So a formula can be less trustworthy for some groups than for others. In another group, adults with chronic spinal cord injury, equations built for able-bodied people overestimated RMR by 4% to 15% in seven studies and underestimated it by 2% to 17% in two. [2]

Evidence label: established that equations can miss for an individual (seen in two systematic reviews, in different groups). "Within 10%" for a 1,800-calorie person still means an error of up to roughly 180 calories either way, and some people fall outside that.

How accurate are handheld calorimeters?

Handheld indirect calorimeters have been studied in clinical and research settings. Label: mixed.

• A review of handheld devices described research findings to date as mixed, with one model more consistent against a reference method and another less accurate against standard metabolic carts. [3]
• A systematic review of 12 studies of one handheld device found it comparable to metabolic carts in most comparisons, though two studies found it read lower. The authors concluded it was valid and reliable for most adults and children. [4]
• In 100 community-living adults, a handheld device and a standard cart had similar average values, but agreement between the two was only 43%, and the limits of agreement ran from -240 to +300 kcal/day. [5]
• In 24 stable home nutrition-support patients, repeated measures from one handheld device were reproducible, but its readings were frequently lower than a traditional calorimeter and needed further validation. [6]

These are results for specific models and groups. They cannot be generalized to every device, and none shows that using a handheld device improves health.

What about smartwatches and apps?

A meta-analysis of 56 studies of one smartwatch family found heart-rate and step counts were generally accurate, but accuracy for energy expenditure was limited, with error above a 10% validity threshold in every subgroup studied. [7] That concerns energy expenditure during activity, not resting metabolic rate, and it applies only to the watches studied. See Smartwatches: What They Measure and Estimate. Proper evidence not found for the accuracy of consumer "RMR" scores from wearables or apps in the sources I checked (PubMed searches on wearable resting energy expenditure and handheld calorimeter validation).

Accuracy is not benefit

Even a well-measured RMR is only one input to a person's energy needs, which also include activity and daily variation. A number for RMR has not been shown to produce better health outcomes by itself. Compare it with the ideas in Metabolism Myths.