Test-re-test reliability of two sustained attention tests in persons with chronic stroke

Primary objective: To investigate the test-re-test reliability of the Conners' Continuous Performance Test II (CCPT) and Digit Vigilance Test (DVT) in persons with chronic stroke. Methods and procedures: Thirty-nine persons with stroke participated in this study. The participants were twice administered both the CCPT and DVT over a 1-week interval. Results: The CCPT performance was reported by two error measures (i.e. Omissions and Commissions) and three time measures (i.e. Hit Reaction Time, Hit Reaction Time Standard Error and Variability of Standard Error). These five measures of the CCPT had good-to-excellent test-re-test reproducibility (ICCs = 0.70-0.90) and limited substantial random measurement errors (SEM% = 7.6-75.2%). Only the Commissions demonstrated substantial practice effects (Cohen's effect size d = 0.38). The DVT had two indicators: the Total Error measure had moderate reproducibility (ICC = 0.55), substantial practice effects (d = 0.30) and random measurement errors (SEM% = 60.8%); the Total Time measure showed excellent reproducibility (ICC = 0.91), less practice effects (d = 0.10) and limited random measurement errors (SEM% = 10.1%). Conclusion: Only the time measures (i.e. Hit Reaction Time for the CCPT and Total Time for the DVT) had satisfactory test-re-test reliability in persons with chronic stroke, which implies that the time measures are good indicators of sustained attention on both tests.