A comparative evaluation of three accelerometry-based physical activity monitors.
WELK, GREGORY J.; BLAIR, STEVEN N.; WOOD, KHERRIN; JONES, SHELBY; THOMPSON, RAYMOND W.
Medicine & Science in Sports & Exercise.
32(9) Supplement:S489-S497, September 2000.
(Format: HTML, PDF)
WELK, G. J, S. N. BLAIR, K. WOOD, S. JONES, and R. W. THOMPSON. A comparative evaluation of three accelerometry-based physical activity monitors. Med. Sci. Sports Exerc., Vol. 32, No. 9, Suppl., pp. S489-S497, 2000.
Purpose: Accelerometry-based activity monitors offer promise for the assessment of free-living physical activity. They provide an objective record of frequency, intensity, and duration of physical activity with minimal burden on participants. The purpose of this study was to evaluate the absolute and relative validity of three contemporary activity monitors (Computer Science and Applications, Inc. [CSA], Tritrac, and Biotrainer) under both laboratory and field conditions.
Methods: Fifty-two participants completed two 30-min choreographed routines designed to simulate a variety of lifestyle physical activities. Three different treadmill paces were completed in both routines to evaluate reliability and validity under laboratory conditions. Six different lifestyle activities were also examined to evaluate the validity of the monitors under field conditions. During each routine, the activity levels of participants were monitored with the three activity monitors as well as by indirect calorimetry systems.
Results: The correlations between the monitors and measured O2 were higher for treadmill activity (mean r = 0.86) compared with lifestyle activity (mean r = 0.55). Correlations among the different monitors were high for both treadmill (r = 0.86) and lifestyle activities (r = 0.70), suggesting that the monitors provide similar information under both conditions. Under laboratory conditions, the CSA yielded accurate predictions of energy expenditure (EE), whereas the Tritrac and Biotrainer tended to overestimate the EE (101-136% of measured value). The Tritrac, however, was found to have less error in individual estimates of EE. Under field conditions, all of the monitors underestimated EE (range: 42-67% of measured value).
Conclusion: The observed differences among the monitors were attributed primarily to differences in the accuracy of the calibration equations rather than to the monitors themselves. Further research is needed to better understand how to use these devices for field-based assessments of physical activity.
(C) 2000 Lippincott Williams & Wilkins, Inc.