Objective: To examine older adults' performance on a newly developed tablet-PC-based cancellation test (e-CT) and to study its psychometric properties.
Methods: 94 older adults with normal cognitive functioning were recruited. The effects of age, education, sex, and experience with computer-based devices on the e-CT were examined. Construct validity was tested by correlating the e-CT with established measures of executive functions (EF) and episodic memory. Correlation coefficients were used to assess short-term test-retest reliablity.
Results: The mean age of participants was 74.6 (SD: 7.3) years and 78% were women. Sixty-nine percent had higher education level (> high school) and 76% used computer-based devices daily. The correct cancellations (CC) on the e-CT ranged from 18 to 56, with a mean (SD) of 40.3 (5.7). The CC was inversely correlated with advancing age (rs = -0.59, N = 94, p <0.001) and positively associated with higher education level (U(94) = 646.5, p = 0.02). No relationship was observed between the e-CT and sex or computer-based device experience. In multivariate analysis, only age remained significantly associated with CC (β = -0.46, SE = 0.07, t = -6.47, df = 93, p <0.001). The e-CT correlated significantly with most of measures of EF. Highest correlations were found between the e-CT and the K-T test, a paper-and-pencil cancellation test (rs = 0.63, N = 90, p <0.001) and TMT-B (rs = -0.41, N = 85, p <0.001). The e-CT did not correlate with the RL-RI 16 episodic memory test. The correlation between the first and second e-CT indicated good reliability (rs = 0.89, N = 13, p <0.001).
Conclusions: Results suggested that e-CT has good psychometric properties and may be useful for assessing EF in older adults.
Keywords: Computer experience; computerized cognitive assessment; executive functions; psychometrics.
Copyright © 2015 American Association for Geriatric Psychiatry. Published by Elsevier Inc. All rights reserved.