VALIDITY OF THE COMPUTER-SCIENCE AND APPLICATIONS (CSA) ACTIVITY MONITOR IN CHILDREN

Purpose: The purpose of this study was to evaluate the validity of the CSA activity monitor as a measure of children's physical activity usi ng energy expenditure (EE) as a criterion measure, Methods: Thirty sub jects aged 10 to 14 performed three 5-min treadmill bouts at 3, 4, and 6 mph, respectively. While on the treadmill, subjects wore CSA (WAM 7 164) activity monitors on the right and left hips. (V) over dot O-2 wa s monitored continuously by an automated system. EE was determined by multiplying the average (V) over dot O-2 by the caloric equivalent of the mean respiratory exchange ratio. Results: Repeated measures ANOVA indicated that both CSA monitors were sensitive to changes in treadmil l speed. Mean activity counts from each CSA unit were not significantl y different and the intraclass reliability coefficient for the two CSA units across all speeds was 0.87. Activity counts from both CSA units were strongly correlated with EE (r = 0.86 and 0.87, P < 0.001). An E E prediction equation was developed from 20 randomly selected subjects and cross-validated on the remaining 10. The equation predicted mean EE within 0.01 kcal.min(-1). The correlation between actual and predic ted values was 0.93 (P < 0.01) and the SEE was 0.93 kcal.min(-1). Conc lusion: These data indicate that the CSA monitor is a valid and reliab le tool for quantifying treadmill walking and running in children.