Intraindividual variability in cognitive test performance has the potential to be a good marker of preclinical Alzheimer's disease status (S. C. Li & U. Lindenberger, 1999). Using cross-sectional community data from 2,317 individuals aged 60-64 years, the authors of this study found that variability was greater in individuals who met criteria for mild cognitive impairment or aging-associated cognitive decline but not for age-associated memory impairment. Higher variability was associated with lower education and a non-English-speaking background. In contrast to previous findings, variability in this study did not contribute uniquely to meeting criteria for mild cognitive impairment. The reasons for the differences may reside in the authors' method of estimating mean independent variability, the use of an occasion-specific measure, or the relatively younger age of the participants. Follow-up of the cohort in 4 years will yield data on the prospective validity of variability as a risk factor for impairment.
Copyright (c) 2005 APA, all rights reserved.