Reliability and validity of a wheelchair collision test for screening behavioral assessment of unilateral neglect after stroke

Objective: To examine the reliability and validity of a new wheelchair collision test (WCT) for screening behavioral unilateral neglect in people with stroke. Design: A total of 19 patients with left hemiplegia resulting from first-ever stroke (mean ± SD, 61.9 ± 25.8 days after onset) were examined with the WCT, whereby the patient was asked to propel a wheelchair to pass four round chairs arranged in two rows. Separate tests were done using two different distances between the chairs: 120 and 140 cm. The number of collisions with the chairs was recorded. The Catherine Bergego Scale was used to evaluate behavioral unilateral neglect. Because unilateral neglect has an unfavorable impact on rehabilitation, the FIM™ instrument was used to determine the function of daily activities. Results: Test-retest reliability ranged from 0.68 to 0.97 in different situations in the WCT. The rate of agreement tested by Kappa statistics between the WCT and Catherine Bergego Scale was higher when the distance between chairs was 120 cm than when the distance was 140 cm (0.68 vs. 0.58). Spearman's rank-correlation coefficient between the WCT and Catherine Bergego Scale was 0.72 (P < 0.001) if the distance between the chairs was 120 cm and 0.75 (P < 0.001) if the distance was 140 cm. The WCT and Catherine Bergego Scale correlated well with the FIM instrument. Spearman's rank-correlation coefficient between the Catherine Bergego Scale and FIM total score was -0.70 (P < 0.001). Conclusion: The WCT is a simple screening test with high test-retest reliability and validity to evaluate behavioral unilateral neglect.