Robust longitudinal neuropsychological norms in Spanish individuals with nonpathological Alzheimer's disease biomarkers.

Abstract

Objective: Neuropsychological norms serve to identify cognitive impairment and monitor neurodegenerative disease progression. However, longitudinal data are limited, and conventional approaches do not account for biomarkers to exclude underlying Alzheimer's disease (AD) pathology, reducing sensitivity to detect subtle cognitive decline in preclinical AD. To address these limitations, this research provides robust longitudinal neuropsychological norms derived from Spanish individuals with nonpathological levels of AD biomarkers.

Method: We analyzed 3-year follow-up data from 350 cognitively unimpaired individuals (Alzheimer's and Families+ cohort). A subset of 228 individuals with normal cerebrospinal fluid amyloid-β42/40, phosphorylated-tau181, and total-tau biomarkers defined the robust reference group. Neuropsychological assessment encompassed the Trail Making Test, Free and Cued Selective Reminding Test, Memory Binding Test, Wechsler Adult Intelligence Scale-IV, Wechsler Memory Scale-IV, Semantic Fluency Test, and Judgment of Line Orientation from the Repeatable Battery for the Assessment of Neuropsychological Status. Using the Reliable Change Index and Standardized Regression-Based methods, robust longitudinal neuropsychological norms were derived from the reference group. Validation analysis assessed the sensitivity of robust versus conventional normative procedures.

Results: Models, equations, percentiles, and cutoffs for identifying significant cognitive decline were provided. Robust norms showed modestly greater sensitivity in certain measures, particularly in episodic memory. The Free and Cued Selective Reminding Test: Total Free Delayed Recall showed the most notable result at the 5th percentile in Reliable Change Index (χ² = 16.06, p = 6.15E-05).

Conclusions: Robust longitudinal neuropsychological norms may improve the detection of subtle cognitive decline, modestly enhancing sensitivity at the lower ends of distributions. (PsycInfo Database Record (c) 2025 APA, all rights reserved).