RESUMO
INTRODUCTION: The NIH Toolbox Cognition Battery (NIHTB-CB) is a computer-based protocol not yet validated for clinical assessment. METHODS: We administered the NIHTB-CB and traditional neuropsychological tests to 247 Memory Disorders and Alzheimer's Prevention Clinic patients with subjective cognitive decline, mild cognitive impairment, mild dementia due to Alzheimer's disease, and normal cognition. Principal component analysis, partial correlations, and univariate general linear model tests were performed to assess construct validity. Discriminant function analyses compared classification accuracy. RESULTS: Principal component analysis identified three conceptually coherent factors: memory (MEMNIH), executive function (EFNIH), and crystallized intelligence (CINIH). These factors were strongly associated with corresponding traditional tests and differed across diagnostic groups as expected. Both NIHTB and traditional batteries yielded strong overall discriminative ability (>80%). DISCUSSION: The NIHTB-CB is a valid method to assess neurocognitive domains pertinent to aging and dementia and has utility for applications in a memory clinic setting.