Conventional and neuropsychological criteria for mild cognitive impairment show similar prognostic value for dementia across 12 years in a non-clinical setting

Abstract

Making early and informative diagnoses of mild cognitive impairment (MCI) is highly important for planning timely and appropriate interventions aimed at dementia risk reduction. However, there is currently no agreement on the MCI criteria, leading to wide heterogeneity in the prognosis of MCI patients and high reversion rates. Our study aimed to compare the prognostic value of Conventional (Petersen/Winblad) and Neuropsychological (Jak/Bondi) criteria for the diagnosis of MCI. We directly compared the ability of each classification method to predict progression to dementia and the stability of the diagnosis over 12 years in a population-based sample of 1021 older adults without dementia. The relative impact of subjective complaints and objective impairment on clinical progression was further evaluated. Baseline MCI diagnosis with the Neuropsychological and Conventional criteria was associated with a comparable risk of dementia over time. Across the study period, the Neuropsychological criteria led to more consistent diagnoses (63.2% vs. 43.2%). The copresence of subjective memory complaints and objective impairment at baseline was associated with increased dementia risk within both diagnostic frameworks. These results further support the use of comprehensive neuropsychological assessment to make timely and appropriate MCI diagnoses and show the added prognostic value of subjective complaints.

Keywords: Conventional criteria; MCI diagnosis; Neuropsychological criteria; Progression to dementia; Stability; Subjective Memory Impairment.

Fig. 1

Flow chart of case selection.

Fig. 2

In the plot we see the comparison of the ROC curves relating to the classification of incident dementia by the two criteria. (Dashed: Random classifier, red: Conventional criteria, green: Neuropsychological criteria).