OBJECTIVE: The aim of this work is to compare the basal metabolic rate determined by indirect calorimetry with the basal metabolic rate estimated by prediction equations in university students. METHODS: The sample consisted of 48 Brazilian volunteers (23 women and 25 men), aged 20-25 years. The basal metabolic rate determined by indirect calorimetry was compared with the basal metabolic rate estimated by the World Health Organization and Henry & Rees equations. The data was analyzed by the paired t-test, intra-class correlation coefficients, standard deviation of differences and Bland and Altman technique. RESULTS: There were no statistically significant differences between measured and estimated basal metabolic rates by both equations. Correlation coefficients between measured and estimated values were higher in men, however statistically significant in both genders. The magnitude of the standard deviation of differences was similar in both genders. The Bland and Altman technique suggests a similar variability behavior between the measured basal metabolic rate and that estimated by both equations. However, basal metabolic rate predicted by the Henry & Rees equation showed greater concordance than that predicted by the World Health Organization equation, especially in women. CONCLUSION: Due to the use of more conservative statistical indicators, it appears that the basal metabolic rates measured by indirect calorimetry and predicted by both equations do not differ significantly. However, when statistical procedures with an apparently higher sensitivity for detecting differences between measured and predicted rates are used, the potential of the equations to estimate the basal metabolic rate was somewhat compromised. There was a bias of 12.5% to 15.6% in comparison with the basal metabolic rate determined by indirect calorimetry.
Calorimetry indirect; Students; Basal metabolism; Forecasting
