Test-retest reliability and responsiveness of the Stroop Color and Word test in patients with stroke

The Stroop Color and Word Test (SCWT) is commonly used to assess cognitive function in patients with stroke. The SCWT contains four scores: Word, Color, Color-Word, and Interference. The purpose of this study was to examine test-retest reliability and responsiveness of the SCWT in patients with stroke. Verifying responsiveness is important to determine whether the SCWT can detect changes resulting from disease progression or interventions. For test-retest reliability, outpatients with stroke (n = 60) completed the SCWT twice within a two-week interval. For responsiveness, inpatients with stroke (n = 56) were administered the measure at admission and discharge. For test-retest reliability, intraclass correlation coefficient (ICC) was estimated and minimal detectable change (MDC) was calculated. For responsiveness, paired t test and two forms of effect size (i.e., standardized effect size and standardized response mean) were computed. The ICC and MDC₉₅, MDC₉₀, and MDC₈₀ values of the four SCWT scores were 0.71-0.79, 9.1-17.1, 7.6-14.3, and 6.0-11.1, respectively. The paired t test showed statistically significant differences between admission and discharge in the SCWT (p < 0.05), except for the Interference score. The values of standardized effect and standardized response mean for the four SCWT scores were 0.13-0.32 and 0.69-0.71, respectively. Our results showed that the SCWT had sufficient test-retest reliability and adequate responsiveness in patients with stroke. The MDC values provided in this study help clinicians and researchers judge whether a patient's difference in score between two assessments reflects an actual change in cognitive function rather than random measurement error.