Predictive value and weight of factors associated with cognitive performance in Hispanics/Latinos enrolled in the Health and Aging Brain Study: Health Disparities.

ABSTRACT:

Introduction

In this analysis of cognitively unimpaired (CU) Hispanic participants from the Health and Aging Brain Study: Health Disparities (HABS-HD), we aimed to identify the main predictor factors for cognitive performance and their relative importance (weight).

Methods

The HABS-HD is a community-based longitudinal cohort study. Data from 952 CU Hispanics, enrolled from 2017 to February 2024, were analyzed. Random forest, an assembly learning method based on decision trees, was used to cross-sectionally forecast the predictive value of 42 risk factors (4 demographic variables, 4 socioeconomic variables, 6 psychosocial variables, 17 health variables, and 11 plasma and magnetic resonance imaging biomarkers) together, and the weighting of each factor for different cognitive

SUBMITTER: Vintimilla R 

PROVIDER: S-EPMC11865711 | biostudies-literature | 2025 Jan-Mar

REPOSITORIES: biostudies-literature