Cross-Validation of Resting Metabolic Rate Prediction Equations

Area of Science:

• Metabolic research
• Nutritional science
• Clinical assessment

Background:

• Resting metabolic rate (RMR) estimation is crucial for nutritional management.
• Current prediction equations offer convenience but vary in accuracy.
• Identifying factors influencing prediction accuracy is key to improving RMR estimation.

Purpose of the Study:

• To evaluate the validity of commonly used RMR prediction equations in healthy adults.
• To compare the accuracy of Harris-Benedict, World Health Organization, Mifflin-St Jeor, Nelson, Wang, and Sabounchi meta-equations against indirect calorimetry.

Main Methods:

• Indirect calorimetry was used to measure RMR in 30 healthy adults (aged 18-65).
• RMR predictions from seven different equations were compared to measured RMR.
• Bland-Altman analysis assessed bias and limits of agreement; repeated-measures ANOVA tested for significant differences.

Main Results:

• The Harris-Benedict and World Health Organization equations showed the lowest bias at the group level.
• These two equations did not significantly differ from measured RMR for the group.
• Prediction bias was inversely related to RMR magnitude and fat-free mass, with underestimation increasing with higher fat-free mass.

Conclusions:

• While Harris-Benedict and World Health Organization equations are group-level accurate, they lack individual precision.
• Existing RMR prediction equations underestimate metabolic rate as fat-free mass increases.
• Further research is needed to develop more accurate RMR prediction models, particularly for individuals with varying body composition.