Related Experiment Video
Updated: May 15, 2026
A Machine Learning Approach to Design an Efficient Selective Screening of Mild Cognitive Impairment
Published on: January 11, 2020
Defining MCI in the Framingham Heart Study Offspring: education versus WRAT-based norms
Richard E Ahl1, Alexa Beiser, Sudha Seshadri
1*Framingham Heart Study †Department of Neurology, Framingham Heart Study, Boston University School of Medicine ‡Department of Biostatistics, Boston University School of Public Health, Boston, MA.
Introduction:
Psychometric definitions of mild cognitive impairment (MCI) typically use cut-off levels set at 1.5 SDs below age-adjusted and education-adjusted norms, assuming that the education adjustment accounts for premorbid abilities. However, noncognitive factors impact educational attainment, potentially leading to incorrect categorization as MCI. We examined whether using an adjustment based on reading performance [Wide Range Achievement Test (WRAT) Reading] improved MCI diagnostic accuracy.
Methods:
A total of 935 Framingham Offspring (mean age, 72±5 y) underwent tests of memory, executive function, abstraction, language, and visuospatial function as part of a neuropsychological test battery. Domain-specific test scores were regressed onto age and WRAT score, or education, to define MCI. Survival analyses were used to relate baseline MCI to incident dementia.
Results:
The 2 MCI definitions differed most for the lowest and highest education groups. The WRAT definition was more strongly associated with incident dementia for all 5 tests. MCI level abstraction performance was associated with incident dementia using the WRAT definition (HR=3.20, P=0.033), but not the education definition (HR=1.19, P=0.814).
Discussion:
The WRAT should be considered along with the standard measure of years of education, as it may be a better surrogate marker of premorbid abilities.
