Actuarial Methods Help Identify Adults With High Risk for Cognitive Impairment

Emily C. Edmonds, PhD

A recently performed cluster analysis using applied actuarial methods identified differences in cognitive profiles of mild cognitive impairment (MCI) and cognitively normal (CN) subtypes, which may improve the diagnostic sensitivity relative to consensus diagnosis for locating older adults at risk for cognitive decline.

Lead author Emily C. Edmonds, PhD, neuropsychologist, University of California–San Diego (UCSD), and colleagues performed cluster analysis on baseline neuropsychological data for 738 nondemented participants from the Alzheimer’s Disease Research Center at UCSD. They used a discriminant function analysis to examine the ability of neuropsychological scores in discriminating cluster-derived groups and predict group membership. Composite z-scores were created in order to capture performance across 5 domains based on up to 19 test scores.

In total, the analysis resulted in 5 cognitive groups: optimal CN (oCN; n = 130) with above-average cognition in all domains; “typical” CN (tCN; n = 204) with average cognition in all domains; non-amnestic MCI (mMCI; n = 84) with impaired performance in the domains of executive function and visuospatial abilities; amnestic MCI (aMCI; n = 216) with impaired memory and language; and mixed MCI (mMCI; n = 84) with impairment in all 5 domains. Additionally, results on the discriminant analysis showed a correct classification rate of 86.3% when predicting group membership.

"Findings indicate that data-driven algorithms could enhance diagnostic sensitivity by identifying empirically-derived at-risk groups of individuals for enrollment in clinical trials, including those with non-amnestic forms of MCI, those with subtle cognitive deficits, and those who do not report subjective concerns but have nonetheless experienced meaningful cognitive changes," Edmonds et al wrote.

READ MORE: Understanding Biomarker Trends in Alzheimer via Roche’s NeuroToolKit

Of the 738 participants, 172 (23.3%) progressed to a diagnosis of dementia, with diagnoses made at an average of 6.3 years post-baseline (standard deviation [SD], 5.4; range, 2-30). Investigators observed a significantly increased risk of progression to dementia in the naMCI (HR, 1.98; P <.02), aMCI (HR, 3.56; P <.001), and mMCI (HR, 5.76; P <.001) groups on Cox regression analysis relative to the tCN reference group. All groups differed significantly from one another, except for the oCN and tCN groups. Notably, 84.3% (n = 145) of those who progressed to dementia received a consensus diagnosis of probable Alzheimer disease based on NINCDS-ADRDA (National Institute of Neurological and Communicative Diseases and Stroke/Alzheimer’s Disease and Related Disorders Association) criteria.

Those in the aMCI group had a higher tau/amyloid-ß (Aß) 1-42 ratio than those in the oCN and naMCI groups (P <.05). Notably, the mMCI group had the highest ratio compared to oCN, tCN, and naMCI (P <.01), but not aMCI (P = .22). When the cut point for biomarker positivity was applied to the tau/Aß 1-42 ratio, chi-square analysis with False Discovery Rate (FDR)-adjustment demonstrated that the mMCI group had a higher rate of biomarker positivity relative to oCN/tCN (P <.001) and naMCI (P = .06) groups.

There were no significant differences in cerebrospinal fluid (CSF) biomarker variables between the oCN, tCN, and naMCI groups. Of the 74 participants who were positive for the tau/Aß 1-42 ratio, 43 (58%) were classified as MCI at baseline by the cluster analysis, whereas 31 (42%) were classified as MCI at baseline by the consensus criteria.

The highest proportion of patients with a “high” level of AD pathology, based on National Institutes of Aging-Reagan consensus criteria, was found in the mMCI group relative to the oCN (P = .01) and naMCI (P = .03) group. The aMCI group showed a greater proportion relative to the oCN group (P = .030); however, these patterns fell to a trend level after false discovery rate adjustment for multiple comparisons. Investigators wrote, “this may reflect the fact that autopsies in general are more likely to occur in those with an advanced stage of dementia, regardless of which cluster classification they originated from.”

Comparisons in consensus diagnosis showed that the cluster analysis classified 54.7% of the sample with MCI compared to 25.2% of the sample classified via consensus diagnosis, corresponding to a 29.5% increase.

REFERENCE

Edmonds EC, Smirnov DS, Thomas KR, et al. Data-driven vs consensus diagnosis of MCI: enhanced sensitivity for detection of clinical, biomarker, and neuropathologic outcomes. Neurology. Published online August 10, 2021. doi: 10.1212/WNL.0000000000012600