Accelerometry as a measure of walking behavior in multiple sclerosis.

Abstract

Accelerometry has been identified as a possible ecologically valid and objective approach for measuring community ambulation in multiple sclerosis (MS). This study provides a validation of accelerometer output based on associations with Expanded Disability Status Scale (EDSS), Patient Determined Disease Steps (PDDS) Scale, and Multiple Sclerosis Walking Scale-12 (MSWS-12) scores, timed 25-foot walk (T25FW) and 6-min walk (6MW) performance, oxygen cost (O(2) cost) of walking, and spatial and temporal parameters of gait. 256 persons with MS completed the PDDS and MSWS-12, underwent an examination for the generation of an EDSS score, undertook two T25FW tests and a 6MW while wearing a portable metabolic unit for measuring the O(2) cost of walking, completed two trials of comfortable walking on a GAITRite electronic walkway for measuring spatial and temporal parameters of gait, and then wore an Actigraph accelerometer during the waking hours of a 7-day period. The accelerometer output was significantly correlated with EDSS (ρ = -0.522), PDDS (ρ = -0.551), and MSWS-12 (ρ = -0.617) scores, T25FW (ρ = -0.595) and 6MW (ρ = 0.630) performance, and O(2) cost of walking (ρ = -0.457). Regarding gait parameters, the accelerometer output was significantly correlated with velocity (ρ = 0.420), cadence (ρ = 0.349), step time (ρ = -0.353), step length (ρ = 0.395), double support (ρ = -0.424), and single support (ρ = 0.400). We provide comprehensive evidence from a large sample of persons with MS that further supports accelerometry as a measure of walking behavior.