Abstract
PURPOSE: The purpose of this article was to evaluate the accuracy of 2 physical activity monitors, monitors 1 and 2, for measuring weight-bearing activity in persons with prior diabetic foot ulcers.
DESIGN: Cross-sectional design.
INSTRUMENTS: Two recently developed monitors were used to differentiate anatomical postures such as lying, sitting, and standing upright. One monitor was designed to distinguish between duration of standing and walking and the other combines duration of standing and walking into 1 measure.
SUBJECTS AND SETTING: Thirty-one subjects were recruited; all participants had experienced a diabetic foot ulcer and completed participation in a previous cohort study. The study setting was 2 medical centers in the Midwestern United States.
METHODS: Subjects simultaneously wore the 2 monitors while performing 14 weight-bearing (ie, walking and standing) and non-weight-bearing (ie, sitting and lying) activities. The duration spent on each activity and the total number of steps taken for each walking activity were directly observed and recorded with each monitor. The accuracy of monitors 1 and 2 was assessed via direct observation as a reference standard. Paired-samples t tests were used to examine the difference in accuracy between the 2 monitors.
RESULTS: For measuring duration of activity, the accuracy of monitor 1 ranged from 73% to 100% for walking, 50% for standing, and from 42% to 100% for sitting/lying. In contrast, the accuracy of monitor 2 ranged from 98% to 100% for walking, 100% for standing, and from 97% to 100% for sitting/lying. The accuracy of monitor 1 for counting the number of steps ranged from 43% to 81%, while the accuracy of monitor 2 ranged from 91% to 99%. Monitor 2 was significantly more accurate than monitor 1 in measuring duration of standing still, slow walking, pedaling while sitting, lying on the left, and lying on the right, as well as measuring steps across different kinds of walking activities. Differences in monitor accuracy between subjects with and without foot pain and between subjects with and without foot amputation were not statistically significant.
CONCLUSION: These findings suggest that monitor 2 is a more accurate measure of weight-bearing activities than monitor 1 among patients with previous diabetic foot ulcers. Additionally, the 2 monitors differ in terms of function; monitor 2 distinguishes standing from walking, whereas monitor 1 combines standing and walking into 1 measure. We recommend monitor 2 to examine the impact of weight-bearing activity on foot ulceration in patients with diabetic neuropathy.