Integrating System Dynamics and Machine Learning to Model Social Determinants of Cognitive Decline: Evidence from the SHARE Dataset
Keywords
Loading...
Authors
Issue Date
2025-07-02
Language
en
Document type
Journal Title
Journal ISSN
Volume Title
Publisher
Title
ISSN
Volume
Issue
Startpage
Endpage
DOI
Abstract
The global population is rapidly aging, and cognitive decline remains a serious health concern affecting the lives of older adults. While biological and demographic determinants of cognitive decline are well established, the influence of social factors remains understudied, particularly in terms of their complex and dynamic interactions. In this regard, this research aims to model how social factors influence cognitive decline over time in adults aged 60 years or older. For this purpose, we employed a combination of System Dynamics (SD) and Machine Learning (ML) methods, applying them to data from the Survey of Health, Ageing and Retirement in Europe (SHARE). The study captured the dynamic feedback relationships between social determinants, including social engagement, social network size, lack of loneliness, social contact, social support, social participation, satisfaction with social ties, and daily activities, through a literature review. Clustering and longitudinal machine learning methods, including Random Effects Expectation-Maximization Tree, Mixed-Effects Random Treeand Mixed-Effects Random Forest, were implemented and employed to assess how these factors affect cognitive decline. Findings showed that daily activities and social participation are the most influential predictors of cognitive trajectories, and these factors participate in dominant reinforcing feedback loops that shape cognitive decline.
Description
Citation
Supervisor
Faculty
Faculteit der Managementwetenschappen