Sze Yen Tan1, Marijka Batterham2, Linda Tapsell2. 1. Smart Foods Centre, University of Wollongong, Northfields Avenue, Wollongong, NSW 2522, Australia. Electronic address:syt153@uow.edu.au. 2. Smart Foods Centre, University of Wollongong, Northfields Avenue, Wollongong, NSW 2522, Australia.
Abstract
BACKGROUND AND AIMS: Studies examining diet-induced thermogenesis and substrate oxidation in a whole room calorimeter (WRC) benefit from a state of energy balance in the measurement period. We found reports on four approaches to assessing energy needs in this context, and they vary considerably. This study assessed the utility of three simple alternatives for predicting energy requirements for calorimeter studies. METHODS: Energy expenditure (EE) data on 37 males and 104 females was obtained from the baseline measurements of three weight-loss trials. The EE data was from 3-d RT3 accelerometer measurements in a free-living environment, and 24-h WRC measurements. A linear regression model was developed from Study 1 and Study 2 (N = 97), using mean 3-d EE as the independent variable. Twenty-four hours WRC EE (N = 29) was compared with three prediction methods: (a) Schofield equation, (b) regression model, and (c) free-living EE × 85%. RESULTS: Energy expenditure measured by accelerometers was 2454.1 ± 491.9 kcal/d in the free-living condition and 2022.7 ± 295.8 kcal/d during the 24-h calorimeter stay (83.8% ± 10.6% of free-living). Energy requirements in the whole room calorimeter was estimated as 0.405 × mean [3-d RT3 energy expenditure] + 1009.6 kcal/d (N = 97, R(2) = 0.479, P < 0.001). Predicted energy requirements were not significantly different from the WRC EE but the Schofield method produced the lowest mean bias and standard deviation. CONCLUSION: Schofield equations are an inexpensive and convenient alternative for predicting energy requirements in WRC studies. Â
BACKGROUND AND AIMS: Studies examining diet-induced thermogenesis and substrate oxidation in a whole room calorimeter (WRC) benefit from a state of energy balance in the measurement period. We found reports on four approaches to assessing energy needs in this context, and they vary considerably. This study assessed the utility of three simple alternatives for predicting energy requirements for calorimeter studies. METHODS: Energy expenditure (EE) data on 37 males and 104 females was obtained from the baseline measurements of three weight-loss trials. The EE data was from 3-d RT3 accelerometer measurements in a free-living environment, and 24-h WRC measurements. A linear regression model was developed from Study 1 and Study 2 (N = 97), using mean 3-d EE as the independent variable. Twenty-four hours WRC EE (N = 29) was compared with three prediction methods: (a) Schofield equation, (b) regression model, and (c) free-living EE × 85%. RESULTS: Energy expenditure measured by accelerometers was 2454.1 ± 491.9 kcal/d in the free-living condition and 2022.7 ± 295.8 kcal/d during the 24-h calorimeter stay (83.8% ± 10.6% of free-living). Energy requirements in the whole room calorimeter was estimated as 0.405 × mean [3-d RT3 energy expenditure] + 1009.6 kcal/d (N = 97, R(2) = 0.479, P < 0.001). Predicted energy requirements were not significantly different from the WRC EE but the Schofield method produced the lowest mean bias and standard deviation. CONCLUSION: Schofield equations are an inexpensive and convenient alternative for predicting energy requirements in WRC studies. Â