Clinical Measures of Balance and Walking Ability in People with Stroke for Assessment via Videoconferencing

Purpose: This study modified established clinical balance and walking measures and estimated the reliability, validity, and feasibility of using these measures to assess people post-stroke via videoconferencing. Method: Twenty-eight people with chronic stroke were recruited and completed the in-person balance and mobility tests. Five clinical measures were modified as virtual assessments over videoconferencing. Feasibility was evaluated by task completion rate, occurrence of adverse events, and technical difficulties. Test-retest reliability and agreement were examined by intra-class correlations and standard error of measurement between two testing days. Convergent validity was examined by the magnitude of associations between in-person and virtual assessments using Pearson or Spearman rank correlation. Results: Twenty-one participants (52% female) participated in both in-person and virtual assessments. No adverse events occurred. Technical challenges were experienced by eight participants. Test-retest reliability for timed up and go test, 30-seconds sit-to-stand, five-times sit-to-stand, functional reach test, and tandem stance resulted in intra-class coefficients of 0.97, 0.90, 0.77, 0.54, and 0.50 respectively. The standard error of measurement was low across all virtual assessments. The timed up and go test, five-times sit-to-stand, and 30-seconds sit-to-stand showed relationship with in-person assessments (r = -0.55 to -0.81). Conclusions: Virtual assessment of walking and balance function in ambulatory people post-stroke is feasible; however, technical challenges were experienced. The test-retest reliability of virtual assessments of timed up and go test and sit-to-stand tasks for people with stroke, together with strong convergent validity of the measures compared to in-person assessments is promising.