A-15 Validity of the NIH Toolbox Cognitive Battery in a Healthy Oldest-Old 85+ Sample

Abstract

Abstract Objective The NIH Toolbox (NIHTB) Cognitive assessment battery has not been validated in the oldest-old population (85 years and older). Factor analysis of the NIHTB cognitive measures with “gold-standard” measures of the same cognition domains has revealed convergent and discriminant validity of these measures in a young adult population. Here we investigate if there is convergent and discriminant validity among the NIHTB and validated measures in the oldest old. Methods An exploratory factor analysis using the principal-axis factor extraction with Varimax rotation was conducted to determine the factor structure of the NIH Toolbox Cognitive Battery and other validated measures of cognition in the healthy oldest-old population. The sample, collected as part of the McKnight Brain Aging Registry, included 170 independent, healthy, community-dwelling adults aged 85+. Results Results indicated four factors with eigenvalues exceeding one with 58.6% variance explained. These four factors were interpreted as Memory, Working Memory, Speed/Attention, and Language. NIHTB Dimensional Change Card Sort test was removed due to low correlations with other measures. Conclusion Most tests from the NIH Toolbox converged with other tests of the same domain and were discriminable from tests of different domains, especially for memory, working memory, and language. The first factor, Speed/Attention, included attention and processing speed measures, but also canonical executive functioning measures such as Trails B. This overlap indicates that executive functioning tasks that rely on motor speed task demands (like the NIHTB DCCS) may mask preserved aspects of executive function in the oldest old. Further work is needed to confirm this interpretation.