Validity and reliability of measurements obtained with an "activity monitor" in people with and without a transtibial amputation

Background and Purpose. In this study, the validity and reliability of measurements obtained with an "Activity Monitor" (AM) were examined. The instrument is designed to monitor ambulatory activity by use of accelerometer signals, and it detects several activities associated with mobility (standing, sitting, lying, transitions, movement-related activities). Subjects. Four men with a transtibial amputation and 4 men without a transtibial amputation participated. Methods. The subjects performed normal daily activities, during which accelerations were measured and videotape recordings were made (reference method). Validity was assessed by calculating agreement scores between the AM output and the videotape recordings and by comparing the number of transitions and the duration of activities determined by both methods. Results. The overall agreement between the AM output and the videotape recordings was 90%. Other agreement scores, in addition to the determination of the number of transitions and the duration of activities, were generally within a range of error of 0% to 10%. Conclusion and Discussion. The reliability and validity of the AM measurements appeared to be good, which supports its potential use in rehabilitation and physical therapy.