Predictive ability of a simple subjective memory complaints scale for incident dementia: Evaluation of Japan's national checklist, the "Kihon Checklist"

AimSubjective memory complaints scales are expected to be useful for the prediction of future cognitive decline. In Japan, the Kihon Checklist-Cognitive Function (KCL-CF), which consists of three items, is used for primary screening of high-risk older adults. However, the predictive validity of the KCL-CF remains unknown. The aim of the present cohort study was to examine whether the KCL-CF can predict the incidence of dementia. MethodsInformation on the KCL-CF score (0-3 points) was collected from community-dwelling older adults (65 years) through a questionnaire. Data on incident dementia were retrieved from the Long-term Care Insurance database. The Cox model and receiver operating characteristic curve analysis were used. ResultsAmong 13 974 participants, the 5.7-year rate of incident dementia was 8.8%. All KCL-CF items significantly predicted the risk of incident dementia even after adjustment for age and sex (P-trend < 0.0001). A higher KCL-CF score was associated with a higher risk of dementia; the age- and sex-adjusted hazard ratios (95% confidence interval) were 1.00 (reference) for a KCL-CF score of 0 points, 1.89 (1.65-2.15) for 1 point, 3.01 (2.59-3.50) for 2 points, and 6.20 (4.87-7.90) for 3 points (P-trend < 0.0001). A cut-off score of 1 points had a specificity of 65.1% and a sensitivity of 60.2%, and the area under the receiver operating characteristic curve was 0.65 (95% confidence interval 0.63-0.66). ConclusionsThe KCL-CF was able to predict incident dementia. However, because a false-negativity rate of approximately 40% would be expected, the KCL-CF score alone might not be sufficient for screening of dementia incidence. Geriatr Gerontol Int 2017; 17: 1300-1305.