Mild cognitive impairment: an operational definition and its conversion rate to Alzheimer's disease

Abstract

Objective: Because of discrepant findings regarding the accuracy of mild cognitive impairment (MCI) in predicting Alzheimer's disease (AD), further study of this construct and conversion rates is essential before use in clinical settings. We aimed to develop an operational definition of MCI consistent with criteria proposed by the Mayo Alzheimer's Disease Center, and to examine its conversion rate to AD.

Methods: Patients were identified from an inception cohort of patients with at least a 3-month history of memory problems, and referred to a 2-year university teaching hospital investigation by primary care physicians. We classified 161 nondemented patients at baseline using MCI criteria. Diagnostic workups were completed annually, and patients were classified as meeting criteria for AD or showing no evidence of dementia after 1 and 2 years.

Results: Of 161 patients, 35% met MCI criteria at baseline. Conversion rates to AD were 41% after 1 year, and 64% after 2 years. Logistic regression analyses to examine predictive accuracy of MCI after 1 and 2 years, with age and education as covariates, were significant (p < 0.0001). After 1 year, MCI showed an optimal sensitivity of 91% and specificity of 79%, and after 2 years, these values were 88 and 83%, respectively.

Conclusions: MCI is an accurate predictor of AD over 1 and 2 years in patients referred by their primary care physicians. Discrepancies in conversion rates may be due to the manner in which patients are recruited to studies as well as the use of different measures to operationalize the construct.