Reliability of spatiotemporal asymmetry during overground walking for individuals following chronic stroke

Rehabilitation research for individuals with chronic stroke is increasingly addressing spatiotemporal asymmetries (STA). Understanding the reliability of STA between sessions is essential for determining whether treatment responses exceed day-to-day variation. The purpose of this study was to determine the minimal detectable change (MDC) and test-retest reliability in STA and gait speed measures for individuals with chronic stroke.

Twenty-six individuals with chronic (>6 months) stroke participated in 2 visits. At each visit, participants walked across a 14-ft (4.3-m) GAITRite mat at comfortable gait speed (CGS) and fast gait speed (FGS). Spatiotemporal asymmetries ratios (paretic/nonparetic limb) were calculated for step length, stance time, and swing time. Intraclass correlation coefficients (ICC, 2,1) were calculated to determine reliability between sessions. Minimal detectable change values were calculated to determine the smallest change that is considered "real."

Spatiotemporal asymmetry measures were consistent between sessions, with ICCs ranging from 0.93 to 0.98. Asymmetry ratio MDC values were calculated for step length (CGS: 0.15; FGS: 0.19), swing time (CGS: 0.26; FGS: 0.20), and stance time (CGS: 0.09; FGS: 0.10). Gait speed MDC was 0.20 m/s and 0.22 m/s, respectively, for the CGS and FGS walking conditions.

There were considerable differences among MDC values for the various STA measures, suggesting that larger changes will be necessary to show improvement for certain measures (eg, swing time asymmetry). These data will assist with setting clinical goals for patients with chronic stroke and will be useful for evaluating interventions designed to minimize temporal and spatial interlimb asymmetries during walking.