Predicting the incidence of mild cognitive impairment with a computer-based cognitive assessment tool in...

This study examined the ability of a computer-based cognitive assessment tool (CompBased-CAT) to predict mild cognitive impairment (MCI) in community-dwelling older adults. A two-year longitudinal study was conducted using data from 2016 to 2018 from the Otassha study cohort of community-dwelling older adults. MCI was defined as a Mini-Mental Status Examination (MMSE) score of <27. The CompBased-CAT was used at baseline, with each subtest score converted into a Z-score. Subsequently, the total Z-scores were calculated. Participants were divided into robust and MCI groups, and all variables were compared using the t-test or χ2 test. Receiver operating characteristic (ROC) curves and logistic regression analyses were conducted, with MCI and total Z-scores as dependent and independent variables, respectively. Among the 455 participants (median age, 72 years; range, 65-89 years; 282 women and 173 men), 32 developed MCI after two years. The participants in the MCI group were significantly older. They had lower maximal gait speed, baseline MMSE scores, subtest Z-scores, and total Z-scores than those in the robust group. The area under the ROC curve was 0.79 (95% confidence interval: 0.70-0.87; P <0.01). The sensitivity was 0.76, and the specificity was 0.75. The logistic regression analysis showed an odds ratio of 1.34 (95% confidence interval: 1.18-1.52; P <0.01). This study showed that CompBased-CAT can detect MCI, which is an early stage of dementia. Thus, CompBased-CAT can be used in future community health checkups and events for older adults.