Validation of a Web-Based Digit Symbol Substitution Task for Detection of Mild Cognitive Impairment in Older Adults

Given the global aging of society and projected increases in cognitive impairment, rapid and sensitive cognitive screening tests are needed to facilitate early detection of mild cognitive impairment (MCI) in older adults. The purpose of this study was to explore the effectiveness of a web-based digit symbol substitution task, Speeded Matching (SM), including a second administration within-session to determine the added utility of practice effects, for potential use as a cognitive screening measure in older adult populations.

Forty-seven older adults with MCI and 87 cognitively normal controls (NC) completed SM twice on an iPad, as well as a brief paper-and-pencil cognitive screening test, the Quick MCI (Qmci). Receiver operating characteristic (ROC) curve analyses were used to investigate the classification accuracy of both administrations of the SM and the Qmci. The incremental validity of the second administration of SM was examined using logistic regression analysis.

Patients with MCI performed significantly worse than their NC counterparts on all measures. NC participants also demonstrated significantly greater practice effects than those with MCI. ROC curve analyses indicated that a cutoff score of <24 for the first administration (AUC = 0.82) and < 27 for the second administration (AUC = 0.85) of SM had slightly better classification accuracy than the Qmci (AUC = 0.77), and a second administration of the SM task within-session significantly improved overall classification accuracy.

Findings suggest that a rapid web-based digit symbol substitution test is a valid screening measure that performs as well as a traditional cognitive screening test.