Validation of the Tracmor triaxial accelerometer system for walking.

PURPOSE Walking is likely to contribute substantially to nonexercise activity thermogenesis. The Tracmor triaxial accelerometer system (Maastricht, The Netherlands) is the most widely validated system for detecting body movement in free-living subjects. The aim of this study was to validate the Tracmor triaxial accelerometer system for estimating the energy expenditure of walking. METHODS Experiments were conducted in healthy subjects. First, baseline variability for Tracmor output was determined for subjects standing still. Second, Tracmor output was compared for walking on a treadmill and on level ground. Third, both Tracmor output and energy expenditure were compared for walking on a treadmill and walking on level ground. Finally, the effect of gradient on Tracmor output and energy expenditure was compared for subjects walking on a treadmill. RESULTS The data demonstrated excellent reproducibility for comparing Tracmor output for standing (CV < 2%). There were excellent log-linear relationships between velocity and Tracmor output walking on a treadmill (r = 0.998) and on level ground (r = 0.999). Tracmor output and the energy expenditure of walking were inseparable for the two modalities of walking. However, the variance in response was such that to reliably derive the relationship between Tracmor output and energy expenditure, separate regression equations are needed for each subject. Finally, the Tracmor accelerometer did not detect the increased energy expenditure of walking that occurs as gradient increases. CONCLUSION The Tracmor triaxial accelerometer provides reproducible and reliable data on the body motion associated with walking regardless of whether a subject walks on a treadmill or level ground. Tracmor units can be used to predict the energetic cost of walking provided that separate regression equations are derived for each subject to convert Tracmor output to energy expenditure.