RMA DelaysOur Admin Portal website is currently experiencing technical difficulties, and it could result in delays with RMAs being processed. We are currently working to resolve these issues. We apologize for the inconvenience.
Join us on August 11th for an ActiGraph webinar hosted by Xtalks:
Oncology Research and Care: Reimagining Digital InnovationRegister Now
Comparisons Of Prediction Equations For Estimating Energy Expenditure In Youth
- Presented on May 30, 2013
Various sets of prediction equations for the Actigraph have been developed to estimate energy expenditure (EE) in youth, but there is no consensus about the best approach. A set of two 2-regression models (2RM) have been proposed as alternative to traditional single regression models (1RM), but they have not been directly compared.
Purpose The purpose of this study was to compare the VM (VM2RM) and VA (VA2RM) and 1RM models from Freedson/Trost (FT), Trost (TR), Puyau (PU) and Treuth (TH) to indirect calorimetry for estimating EE in youth.
Methods Fifty nine participants (7- 13 yrs; male: 41) performed 12 different activities (randomly assigned from a set of 24) that would mimic “free-living” activities in youth. While performing these activities, the participants were concurrently measured with a metabolic gas analyzer (Oxycon Mobile; OM) and an Actigraph accelerometer. Each activity was performed for 5-min, with a 1-min rest between activities. Estimates of METs were obtained from the ActiGraph prediction methods and were compared to OM measured METs (measured VO2/predicted RMR). Comparisons were first made using the aggregated data from the whole trial (i.e. between-subject comparisons; n = 59) and then for each individual activity (i.e. activity-by-activity comparisons; n=24). Agreement with OM values was evaluated using repeated measures of ANOVA (with Tukey-Kramer pairwise comparisons), and absolute percent errors (APE).
Results For the whole trial comparisons, estimated EE from each of the ActiGraph prediction methods significantly underestimated measured OM EE (P<0.05). For the same comparisons, large APEs were observed for both the VM2RM (31.3%) and VA2RM (36.8%) and were similar to APEs from the 1RM methods (range: 25.7% to 49.2%). The evaluation of individual activities revealed that the Actigraph prediction methods underestimated almost all activities (except 1 activity by FT; 5 by TR; and 1 by TH): APEs ranging from 27.8% to 44.0%
Conclusions The two different 2RM methods significantly underestimated EE and the error was similar in magnitude as standard 1RM methods. Results indicate considerable error for assessing free living energy expenditure in youth with the ActiGraph.