Neuropsychological test performance and cognitive reserve in healthy aging and the Alzheimer's disease spectrum: a theoretically driven factor analysis.

Accurate measurement of cognitive function is critical for understanding the disease course of Alzheimer's disease (AD). Detecting cognitive change over time can be confounded by level of premorbid intellectual function or cognitive reserve and lead to under- or over-diagnosis of cognitive impairment and AD. Statistical models of cognitive performance that include cognitive reserve can improve sensitivity to change and clinical efficacy. We used confirmatory factor analysis to test a four-factor model composed of memory/language, processing speed/executive function, attention, and cognitive reserve factors in a group of cognitively healthy older adults and a group of participants along the spectrum of amnestic mild cognitive impairment to AD (aMCI-AD). The model showed excellent fit for the control group (χ(2) = 100; df = 78; CFI = .962; RMSEA = .049) and adequate fit for the aMCI-AD group (χ(2) = 1750; df = 78; CFI = .932; RMSEA = .085). Although strict invariance criteria were not met, invariance testing to determine if factor structures are similar across groups yielded acceptable absolute model fits and provide evidence in support of configural, metric, and scalar invariance. These results provide further support for the construct validity of cognitive reserve in healthy and memory impaired older adults.