An evaluation of the NINCDS-ADRDA neuropsychological criteria for the assessment of Alzheimer's disease: a confirmatory factor analysis of single versus multi-factor models.

Neuropsychological test batteries are frequently used to assess the nature and severity of cognitive deficits among patients with early Alzheimer's Disease (AD) and related disorders. The NINCDS-ADRDA criteria are among the most widely used guidelines to diagnose dementia (McKhann et al.,1984). These criteria specify eight distinct areas of neuropsychological function that should be evaluated in patients with suspected cognitive impairment. Recent studies have suggested that neuropsychological deficits observed in AD may be explained by a single general factor related to memory deficits or to executive dysfunction. In contrast, the results of other investigations have indicated that multiple qualitatively different factors underlie cognitive abilities in AD. In the present study, we used confirmatory factor analysis to examine the structure of cognitive abilities in AD and to assess the extent to which single and multiple ability factors accurately represent neuropsychological test data obtained from patients with AD. Results indicated that the NINCDS-ADRDA model fit the data better than a single factor model. However, a more parsimonious model specifying memory, verbal abilities, visuospatial skills, executive function, and higher as well as lower functional activities of daily living fit the data better than the NINCDS-ADRDA model. These results have important theoretical and practical implications for diagnostic evaluation.