Potential limitations of measuring ambulatory activity of part-time wheelchair users: a comparative study of two research grade activity monitors

Objective. Research grade activity monitors such as the StepWatch Activity Monitor (SAM) and Actigraph have been shown to be highly accurate for the assessment of ambulatory activity, but some individuals function in the community using a combination of both walking and wheelchair activity. The purpose of this study was to assess the accuracy of the SAM and Actigraph at not detecting ambulatory activity, during wheelchair activities at two intensity levels. Approach. Two independent cohorts were tested while simultaneously wearing a SAM and Actigraph on the ankle. Group 1 consisted of twelve adults performing a six-minute push test (6MPT). Group 2 were eighteen members of an adolescent wheelchair basketball team tested during a scrimmage/practice and regional tournament. Visual observation was used to confirm the number of ambulatory steps taken during testing was zero. Main results. The average number of ‘steps’ detected during a 6MPT was 19 ± 25 (range: 0-85) with the Actigraph. No ‘steps’ were detected using the SAM in Group 1. In Group 2, the average ‘step’ count during basketball practice was 1421 ± 750 (range: 612-2798) using the Actigraph and 1363 ± 1238 (range: 84-3462) using the SAM, and correlated with playing time, despite the leg/ankle being strapped down. Similar findings were observed during the wheelchair basketball regional tournament. Significance. Caution should be heeded when utilizing these activity monitors in individuals with disabilities who participate in wheelchair sports. Additional written activity logs should be used to exclude periods of time during which high acceleration wheelchair sporting activity occurs, or study participants should be fully educated in removing the activity monitors during non-ambulatory wheelchair sports.