Validation and calibration of the activPAL™ for estimating METs and physical activity in 4-6 year olds

Objectives: Examine the predictive validity of the activPAL (TM) metabolic equivalents equation, develop an activPAL (TM) threshold value to define moderate-to vigorous-intensity physical activities, and examine the classification accuracy of the developed moderate-to vigorous-intensity physical activities threshold value in 4- to 6-year-old children. Design: A sample of forty 4- to 6-year-old children from the Illawarra region in New South Wales, Australia were included in data analysis. Methods: Participants completed a similar to 150-min room calorimeter protocol involving age-appropriate sedentary behaviors, light-intensity physical activities and moderate-to vigorous-intensity physical activities. activPAL (TM) accelerometer counts were collected over 15 s epochs. Energy expenditure measured by room calorimetry and direct observation were used as the criterion measure. Predicted metabolic equivalents were calculated using the activPAL (TM) metabolic equivalents equation (activPAL (TM) software version 5.8.0). Predictive validity was evaluated using dependent-samples t-tests. Participants were randomly allocated into two groups to develop and cross:validate an intensity threshold for moderate-to vigorous-intensity physical activities. Receiver operating characteristic curve analysis was used to determine moderate-to vigorous-intensity physical activities threshold. The classification accuracy of the developed threshold was cross-validated using sensitivity, specificity, and area under the receiver operating characteristic-curve. Results: The activPAL (TM) metabolic equivalents equation significantly overestimated metabolic equivalents during sedentary behaviors and significantly underestimated metabolic equivalents for light-intensity physical activities, moderate-to vigorous-intensity physical activities and total metabolic equivalents compared to measured metabolic equivalents (all P < 0.001). The developed threshold of >= 1418 counts per 15s resulted in good classification accuracy for moderate-to vigorous-intensity physical activities. Conclusion: The current activPAL (TM) metabolic equivalents equation requires further development before it can be used to accurately estimate metabolic equivalents in preschoolers. The developed threshold exhibited acceptable classification accuracy for moderate-to vigorous-intensity physical activities; however studies cross-validating this moderate-to vigorous-intensity physical activities threshold in free-living preschool-aged children are recommended. (C) 2013 Sports Medicine Australia. Published by Elsevier Ltd. All rights reserved.