Quantitating physical activity in COPD using a triaxial accelerometer

Study objective: To determine the reliability, validity, and stability of a triaxial accelerometer for walking and daily activity measurement in a COPD sample. Design: Cross-sectional, correlational, descriptive design. Setting: Outpatient pulmonary rehabilitation program in a university- affiliated Veterans Affairs medical center. Participants: Forty-seven outpatients (44 men and 3 women) with stable COPD (FEV₁, 37% predicted; SD, 16%) prior to entry into a pulmonary rehabilitation program. Measurements and results: Test-retest reliability of a triaxial movement sensor (Tritrac R3D Research Ergometer; Professional Products; Madison, WI) was evaluated in 35 of the 47 subjects during three standardized 6-min walks (intraclass correlation coefficient [rICC] = 0.84). Pearson correlations evaluated accelerometer concurrent validity as a measure of walking (in vector magnitude units), compared to walking distance in all 47 subjects during three sequential 6-min walks (0.84, 0.85, and 0.95, respectively; p < 0.001). The validity of the accelerometer as a measure of daily activity over 3 full days at home was evaluated in all subjects using Pearson correlations with other indicators of functional capacity. The accelerometer correlated with exercise capacity (maximal 6-min walk, r = 0.74; p < 0.001); level of obstructive disease (FEV₁ percent predicted, r = 0.62; p < 0.001); dyspnea (Functional Status and Dyspnea Questionnaire, dyspnea over the past 30 days, r = - 0.29; p < 0.05); and activity self-efficacy (Activity Self-Efficacy Questionnaire, r = 0.43; p < 0.01); but not with self-report of daily activity (Modified Activity Recall Questionnaire, r = 0.14; not significant). Stability of the accelerometer to measure 3 full days of activity at home was determined by an rICC of 0.69. Conclusions: This study provides preliminary data suggesting that a triaxial movement sensor is a reliable, valid, and stable measure of walking and daily physical activity in COPD patients. It has the potential to provide more precise measurement of everyday physical functioning in this population than self-report measures currently in use, and measures an important dimension of functional status not previously well- described.