Abstract
Substantial interest currently focuses on the earliest stages of Alzheimer's disease (AD) and identification of informative biomarkers. The onset of mild cognitive impairment (MCI) often serves as a key classification entity, with MCI defined by the presence of subjective concerns regarding declines in cognition and memory together with observed performance below expectations on population-normed neuropsychological assessments. These criteria have served their purpose for identifying the needs of “typical” adults for increased supports, but may be problematic for staging when individual abilities depart from “average” levels. In the case of individuals performing well-above average, declines in cognition associated with early AD progression would not be recognized as MCI until that progression has advanced further, lowering estimated specificity in biomarker studies. For individuals well-below average, such as individuals with intellectual disability, MCI could be inferred from test results prior to true AD progression, lowering estimated sensitivity even in cases where exceptionally low norm-referenced performance is found, as for adults with Down syndrome. Recognizing that abilities within the overall elderly population are distributed approximately normally, over 30% of individuals will differ from age-referenced means by at least one standard deviation. Therefore, the use of any norm-referenced criteria to define MCI risks introducing substantial misclassifications in staging of underlying AD pathological progression. Participants were older adults with Down syndrome (n=360) who are part of a multidisciplinary longitudinal study focused on aging and dementia [1]. They received comprehensive evaluations at approximately 18-month intervals which included direct assessment of selected cognitive functions, informant interviews, and review of medical records. Our study of older adults with Down syndrome illustrates the necessity of factoring in lifelong cognitive functioning, where agreement was significant but far from perfect between presence/absence of relevant caregiver concern and presence/absence of assessed decline on neuropsychological measures commensurate with MCI, Φ2 = 0.19, p < .0001. The clear implication is that an alternative working definition of MCI is needed, referenced to individual decline rather than population-normed performance for both biomarker studies and for early recognition of AD progression more generally. [1] Silverman, W. et al. (2004), American Journal on Mental Retardation.
Published Version
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call