Comparison of activity monitors to estimate energy cost of treadmill exercise

Purpose: To evaluate the validity of five physical activity monitors available for research: the CSA, the TriTrac-R3D, the RT3, the SenseWear Armband, and the BioTrainerPro. Methods: A total of 10 healthy men and 11 healthy women performed 10 min of treadmill walking at 54, 80, and 107 m.min(-1) and treadmill running at 134, 161, 188, and 214 m.min(-1). The CSA, TriTrac-R3D, RT3, and BioTrainer-Pro accelerometers were placed side by side bilaterally at the waist in the axillary position, and the SenseWear Armband monitors were placed bilaterally on the posterior portion of each arm in the mid-humeral position. Simultaneous measurements of body motion and indirect calorimetry were continuously recorded during all exercise. Data were analyzed with repeated measures ANOVA and pairwise Bonferroni-adjusted estimated marginal means. Results: There was no significant difference in the mean energy expenditure (EE) recorded bilaterally by any of the monitors (P>0.05) at any treadmill speed. The SenseWear Armband, the TriTrac-R3D, and the RT3 had significant increases in mean EE across all walking and running speeds (P<0.05). Below 161 m.min(-1), the mean EE recorded by the BioTrainer-Pro and the CSA increased significantly (P<0.001); however, there was no significant difference (P>0.10) in mean EE recorded by either monitor for speeds above 161 m.min(-1). In general, all monitors overestimated EE at most treadmill speeds when compared with indirect calorimetry (P<0.001), except for the CSA which underestimated EE at the lowest and highest speeds. Conclusion: The CSA was the best estimate of total EE at walking and jogging speeds, the TriTrac-R3D was the best estimate of total EE at running speeds, and the SenseWear Armband was the best estimate of total EE at most speeds.