Miriam D. Cohen

A novel approach to measuring activity in chronic obstructive pulmonary disease: using 2 activity monitors to classify daily activity.

PURPOSE Patients with chronic obstructive pulmonary disease with a low profile of daily activity have poor health outcomes. Although walking is a common activity, it may not be the most relevant physical activity to measure in this population. It was the purpose of this study to determine the accelerometer-defined thresholds that discriminate a range of daily activities and use these thresholds to assess activity profiles among stages of disease severity. METHODS Subjects with chronic obstructive pulmonary disease (N = 57) completed a standardized sequence of activities that comprised sitting, standing, and walking while wearing an accelerometer on the waist and an actigraph on the wrist. Using a calibration procedure, accelerometer output was translated into speeds. Speeds were estimated for each interval in the testing sequence. Walking and nonacceleration thresholds were derived from the intervals to define 4 activity categories: walking, slow/intermittent walking, active-not-walking, and rest. Subjects wore the 2 devices for 2 days. Accelerometer output was then classified into 1 of the activity categories. Percent time spent in activity categories and speeds generated were compared among Global Initiative for Obstructive Lung Disease (GOLD) stages. RESULTS The waist-worn accelerometer accurately estimated speeds. Speed thresholds for walking and nonacceleration were 0.70 mph and 0.25 mph, respectively. Among GOLD stages, those with more severe obstruction spent less time in the walking categories and generated slower speeds. CONCLUSIONS The accuracy of these methods to detect a range of physical activities enhances the utility of accelerometers in comparing daily activity in sedentary populations. Measurements of the more subtle activities offer an appealing new area of study.