Estimating Clinically Important Change in Gait Speed in People With Stroke Undergoing Outpatient Rehabilitation

Abstract

Gait speed is commonly used to assess walking ability in people with stroke. It is not clear how much change in gait speed reflects an important change in walking ability. The purpose of this study was to estimate clinically important changes in gait speed by using 2 different anchors for what is considered "important": stroke survivors and physical therapists perceptions of change in walking ability. Participants underwent outpatient physical therapy (mean 56 days post-stroke) after a first-time stroke. Self-selected gait speed was measured at admission and discharge. At discharge, participants and their physical therapists rated their perceived change in walking ability on a 15-point ordinal Global Rating of Change (GROC) scale. Estimated important change values for gait speed were calculated by using receiver operating characteristics curves, with the participants and physical therapists GROC as the anchors. The mean (SD) initial gait speed of all participants was 0.56 (0.22) m/s. The estimated important change in gait speed ranged from 0.175 m/s (participants perceived change in walking ability) to 0.190 m/s (physical therapists perceived change in walking ability), depending on the anchor. During the subacute stage of recovery, individuals poststroke who experience improvements in gait speed of 0.175 m/s or greater are likely to exhibit a meaningful improvement in walking ability. The estimated clinically important change value of 0.175 m/s can be used by clinicians to set goals and interpret change in individual patients and by researchers to compare important change between groups.