Skip to main content
The Egyptian Journal of Internal Medicine

Table 3 Bias, accuracy, and precision of predictive equations relative to indirect calorimetry

From: Associations of measured resting energy expenditure with predictive equations, NUTRIC score, and patient outcomes

Equation

Biasa (kcal/day)

Accuracyb

Precisionc

Incidence of large errorsd

Weight-based (23 kcal/kg/day)

− 73.35–64.60

58.0%

8.75–15.46%

28%

PSUm

− 48.12–70.71

56.0%

7.53–13.12%

26.0%

Faisy–Fagon

170.50–323.34

30.0%

15.43–25.25%

54.0%

  1. aBias assessed by the 95% CI of the difference between estimated and measured metabolic rates. An interval including 0 indicates no bias
  2. bAccuracy assessed as percentage of estimated values within 10% of corresponding measured values
  3. cPrecision assessed as 95% CI of root mean squared prediction errors considered precise if ≤ 15%
  4. dLarge errors were 15% above or below the measured value
Back to article page