Examining the validity of the ActivPAL monitor in measuring posture and ambulatory movement in children
Centre for Physical Activity and Nutrition, Centre for Child Health Research, Auckland University of Technology, Auckland, New Zealand
International Journal of Behavioral Nutrition and Physical Activity 2012, 9:119 doi:10.1186/1479-5868-9-119Published: 2 October 2012
Decreasing sedentary activities that involve prolonged sitting may be an important strategy to reduce obesity and other physical and psychosocial health problems in children. The first step to understanding the effect of sedentary activities on children’s health is to objectively assess these activities with a valid measurement tool.
To examine the validity of the ActivPAL monitor in measuring sitting/lying, standing, and walking time, transition counts and step counts in children in a laboratory setting.
Twenty five healthy elementary school children (age 9.9 ± 0.3 years; BMI 18.2 ± 1.9; mean ± SD) were randomly recruited across the Auckland region, New Zealand. Children were fitted with ActivPAL monitors and observed during simulated free-living activities involving sitting/lying, standing and walking, followed by treadmill and over-ground activities at various speeds (slow, normal, fast) against video observation (criterion measure). The ActivPAL sit-to-stand and stand-to-sit transition counts and steps were also compared with video data. The accuracy of step counts measured by the ActivPAL was also compared against the New Lifestyles NL-2000 and the Yamax Digi-Walker SW-200 pedometers.
We observed a perfect correlation between the ActivPAL monitor in time spent sitting/lying, standing, and walking in simulated free-living activities with direct observation. Correlations between the ActivPAL and video observation in total numbers of sit-to-stand and stand-to-sit transitions were high (r = 0.99 ± 0.01). Unlike pedometers, the ActivPAL did not misclassify fidgeting as steps taken. Strong correlations (r = 0.88-1.00) between ActivPAL step counts and video observation in both treadmill and over-ground slow and normal walking were also observed. During treadmill and over-ground fast walking and running, the correlations were low (r = 0.21-0.46).
The ActivPAL monitor is a valid measurement tool for assessing time spent sitting/lying, standing, and walking, sit-to-stand and stand-to-sit transition counts and step counts in slow and normal walking. The device did not measure accurately steps taken during treadmill and over-ground fast walking and running in children.