Within-person across-neuropsychological test variability and incident dementia

Context: Neuropsychological tests are used to predict and diagnose dementia. However, to our knowledge, no studies to date have examined whether within-person across-neuropsychological test variability predicts dementia. Objective: To examine whether within-person across-neuropsychological test variability predicts future dementia. Design: The Einstein Aging Study (EAS) is a population-based longitudinal study of aging and dementia located in Bronx County, New York. We used Cox proportional hazards models using age as the time scale to estimate hazard ratios (HRs) for performance on individual neuropsychological tests (Free and Cued Selective Reminding Test, Digit Symbol Substitution subtest of the Wechsler Adult Intelligence Scale Revised, and the Vocabulary subtest of the Wechsler Adult Intelligence Scale Revised) and for within-person across-neuropsychological test variability as predictors of incident dementia. Analyses were stratified by sex, and controlled for education and medical illness. Setting and Participants: A total of 1797 participants (age ≥70 years) enrolled in the EAS between October 1993 and December 2007. Participants seen for the baseline visit only (n=750), prevalent dementia cases (n=72), and those with missing follow-up information (n=78) were excluded. A total of 897 individuals were included in this investigation. Participants had follow-up visits every 12 to 18 months. Main Outcome Measure: Incident dementia. Results: Sixty-one cases of incident dementia were identified during follow-up (mean [SD], 3.3 [2.4] years), of which 26 were in the highest quartile of within-person across-neuropsychological test variability. Adjusting for sex, education, and medical illness, variability was associated with incident dementia (HR for 1-point difference in variability, 3.93 [95% confidence interval {CI}, 2.04-7.56]). The association persisted even after adjusting for level of performance on individual neuropsychological tests (HR for 1-point difference in variability, 2.10 [95% CI, 1.04-4.23]). Comparing Cox models using neuropsychological tests with and without within-person across- neuropsychological test variablity showed that the former improved the prediction of dementia. Sensitivity in a model predicting dementia at 1 year also improved when neuropsychological test variability was included. Conclusions: In this population, within-person across-neuropsychological test variability was associated with development of incident dementia independent of neuropsychological test performance. This finding needs to be confirmed in future studies.