Ability of different physical activity monitors to detect movement during treadmill walking.

This study assessed the ability of four different activity monitors to discriminate changes in treadmill walking velocity. The relationships between walking velocity and bodily movement and between bodily movement and energy expenditure determined by indirect calorimetry (IC-EE or METs) were determined. Twenty-eight subjects walked at 3.2, 4.0, 4.8, 5.6, and 6.4 km/h (0 % grade) for 30 min on separate occasions. The Tritrac-R3D (TT), Computer Science & Applications, Inc. (CSA), and Mini-Logger (ML) activity monitors that measure bodily acceleration in one or three planes, and a Yamax Digiwalker-500 (YX) that records footsteps, were secured at the waistline of each subject. CSA monitors were also worn at the wrist and ankle. Walking velocity and bodily movement were significantly related (r = 0.89 to 0.93) for TT, CSA, ML, and YX. Importantly, changing each walking velocity produced significant changes in bodily movement that was detected by each monitor. Bodily movement and IC-EE were significantly related for TT, CSA, ML, and YX (r = 0.47 to 0.94). Compared to IC-EE, and at all walking speeds, EE was significantly overestimated by the TT, and EE was significantly underestimated by the YX. These results indicate that the activity monitors can differentiate bodily movement associated with walking at slow speeds better than they can estimate energy expenditure associated with walking at slow speeds.

Evaluation of methods to assess physical activity in free-living conditions.

PURPOSE: The purpose of this study was to compare different methods of measuring physical activity (PA) in women by the doubly labeled water method (DLW). METHODS: Thirteen subjects participated in a 7-d protocol during which total daily energy expenditure (TDEE) was measured with DLW. Body composition, basal metabolic rate (BMR), and peak oxygen consumption were also measured. Physical activity-related energy expenditure (PAEE) was then calculated by subtracting measured BMR and the estimated thermic effect of food from TDEE. Simultaneously, over the 7 d, PA was assessed via a 7-d Physical Activity Recall questionnaire (PAR), and subjects wore secured at the waist, a Tritrac-R3D (Madison, WI), a Computer Science Application Inc. activity monitor (CSA; Shalimar, FL), and a Yamax Digi Walker-500 (Tokyo, Japan). Pearson-product moment correlations were calculated to determine the relationships among the different methods for estimating PAEE. Paired t-tests with appropriate adjustments were used to compare the different methods with DLW-PAEE. RESULTS: There was no significant difference between PAEE determined from PAR and DLW. The differences between the two methods ranged from -633 to 280 kcal.d(-1). Compared with DLW, PAEE determined from CSA, Tritrac, and Yamax was significantly underestimated by 59% (-495 kcal.d(-1)), 35% (-320 kcal.d(-1)) and 59% (-497 kcal.d(-1)), respectively. VO2peak explained 43% of the variation in DLW-PAEE. CONCLUSION: Although the group average for PAR-PAEE agreed with DLW-PAEE, there were differences in the methods among the subjects. PAEE determined by Tritrac, CSA, and Yamax significantly underestimate free-living PAEE in women.