The prediction of athlete resting metabolic rate – is it time to reassess the method?

Effective energy prescription requires an accurate assessment of the athletes’ daily energy expenditure. Whilst the use of published prediction equations using RMR and an activity factor is common practice; there is little evidence to validate their use with athletic groups. This study compared measured resting metabolic rate (RMR) using indirect calorimetry to RMR using 17 prediction equations. Anthropometric and metabolic data was collected for 23 male rugby athletes and a literature review was conducted for evidence relating to the measurement and prediction of RMR in athlete populations. Paired samples t-tests and root mean square prediction error (RMSPE) were used to compare measured and predicted RMR and the Bland-Altman procedure was used to assess the bias for each prediction. While prediction equations significantly and systematically underestimated RMR in rugby players for all equations (p=0.001), there are several sources of error that need to be addressed. The validation of population-specific prediction equations in athlete groups requires standardised and accurate assessments of body composition (including fat and fat-free mass) and RMR by indirect calorimetry. While there is a strong, linear relationship between lean mass and RMR, research is also needed to identify the unique characteristics of athletes that can act as covariates.