Objective: The principal aim of this study was to validate a proposed new index of physical activity, the activity-related time equivalent based on accelerometry (ArteACC), in adolescents. A secondary aim was to develop regression equations for prediction of total energy expenditure (TEE) and activity energy expenditure [AEE = 0.9 x TEE - resting metabolic rate (RMR)].
Research methods and procedures: RMR and energy expenditure (EE) under standardized exercises were measured by indirect calorimetry in 36 adolescents (14 to 19 years old). TEE was measured by the doubly labeled water method, and physical activity was assessed simultaneously with an accelerometer for 14 days. AEE, AEE in relation to body weight (AEE per kilogram), and activity-related time equivalent based on energy expenditure (ArteEE = AEE/[EE reference activity - RMR]) were calculated from laboratory and free-living EE data. ArteACC was calculated as total activity counts/activity counts of reference activity.
Results: ArteACC was significantly related to AEE per kilogram (r = 0.57; p < 0.0001) and ArteEE (r = 0.68; p < 0.001). The absolute amount of time (minutes per day) spent in physical activity was significantly lower when calculated from ArteACC than from ArteEE (p < 0.001). TEE was significantly influenced by RMR, sex, and ArteACC (r(2) = 0.89). AEE was significantly influenced by sex and ArteACC (r(2) = 0.59).
Discussion: Despite an absolute difference between the two indexes, ArteEE and ArteACC, ArteACC seems to be a valid indicator of free-living physical activity. It contributed significantly, by 3.3% and 12.5%, to the explained variations in TEE and AEE, respectively.