The effect of years of schooling and age on CERAD‐MX performance in Mexican preclinical carriers of the APPV717I mutation: Randomized data simulation

Abstract

Abstract INTRODUCTION We aimed to determine the effect of years of schooling (YoS) and age on the Mexican adaptation of the Consortium to Establish a Registry for Alzheimer's Disease (CERAD‐MX) scores in preclinical carriers group (PCG) and non‐carriers group (NCG) of the APPV717I mutation. METHODS We included 39 first‐degree Mexican relatives of APPV717I carriers (PCG = 15; NCG = 24). We report eight CERAD‐MX tasks: Mini‐Mental State Examination (MMSE), Word List Learning (WLL), Delayed Recall (WLD) and Recognition (WLR), Constructional Praxis Copy (CPC) and Recall (CPR), Semantic Verbal Fluency (SVF), and Verbal Boston Naming (VBN), comparing both groups’ performance and simulating new samples’ random vectors by inverse transform sampling. RESULTS PCG and NCG performed similarly on CERAD‐MX. In both groups, YoS and age influence all z scores. A positive age effect resulted for PCG on CPC and SVF; for the NCG on MMSE, SVF, and VBN. DISCUSSION All tasks are influenced by YoS. Higher YoS/younger age or YoS/older age interactions affected different tasks, suggesting that YoS confounds outcomes. Highlights Years of schooling (YoS) and age affect the Mexican adaptation of the Consortium to Establish a Registry for Alzheimer's Disease scores of APPV717I preclinical carriers. Preclinical carriers underperformed non‐carriers on Constructional Praxis Recall. Fewer YoS emerges as a confounding variable when detecting cognitive failures. Younger participants in both groups overperformed the older ones in the Memory tasks. Randomized data simulation increases statistical power when analyzing rare diseases.

Keywords

• age
• autosomal dominant Alzheimer's disease
• cognition
• cross‐cultural
• dementia
• early onset