Untitled

Abstract

Objective-To compare physical activity, via accelerometers secured in three different attachement site, and determine the correlation between estimates of METs by accelerometry and indirect calorimetry in adults playing different active videogames. Methods-Participants (n =54) played four active videogames that simulate sports activities using an Xbox 360 Kinect™ (10 minutes/each): athletics, bowling, boxing and bach volleyball. The accelerometers (ActiGraph® GT3XE-Plus) were positioned in three different location (wrist, hip, and ankle) to determine the counts and predicted METs values. The Cosmed K4b2 portable metabolic analyzer (indirect calorimetry) measured the Actual METs. Analysis of variance (multi-factor model), Student's t-test for paired samples, and Pearson's correlation were used to test the data (level of significance of 5%). Results-The correlation between counts and Actual METs depended on the game type and accelerometer positioning. Athletics and beach volleyball video games presented positive correlations between the Actual METs and counts in all axis (vertical, horizontal, perpendicular, and vector's magnitude). There was an interaction between the accelerometer location and the different games (F=575.324, p<0.001). The hip accelerometer presented the biggest mean counts values in all the sports games. There wasn´t a significant difference between the games according to the hip accelerometer. In all games, the Actual METs values were superior (p<0.05) to those estimated by the accelerometer (hip). There was a positive correlation only for beach volleyball game. Conclusion-The counts versus Actual METs relationship for accelerometry are exclusive to the attachement site and the games. The accelerometry underestimate the values of METs.