Factor Structure and Internal Consistency of the National Alzheimer Coordinating Center's Uniform Data Set Version 3...

Construct validation of cognitive batteries across Africa is imperative to understanding dementia in the region. We examined construct validity and internal consistency of the neuropsychological battery of Uniform Data Set version 3 (UDSNB 3.0) of the Alzheimer Coordinating Center in Nigeria older adults.

Three hundred forty-nine (220 females; age: 65 to 85) community dwellers were recruited. UDSNB 3.0 with 12 subscales were used to measure cognition. Two sets of data were collected. First was for exploratory factor analysis (EFA) and second was confirmatory factor analysis (CFA). Four models were specified for CFA.

EFA principal axis factor with varimax rotation yielded 4 factors: Executive function, memory, visual-spatial ability, and processing speed. Four CFA were performed based on 4 specified models, with only model 3 showing good model fit: CMIN/DF=2.13; confirmatory fit index=0.94; root mean square error of approximation=0.07. Model 3 had 5 latent variables: working memory, language, verbal memory, visual-spatial ability, and processing speed. UDSNB 3.0 had an overall Cronbach alpha of 0.73, suggesting strong internal reliability with ANOVA model F134,1619​​​=183.65 significant at P<0.001 level of testing.

Our study showed that UDSNB 3.0 has construct validity and good internal consistency in our older adult population.