Uncovering multimorbidity patterns linked to disability in aging populations: A machine learning analysis

Area of Science:

• Gerontology
• Public Health
• Biostatistics

Background:

• Multimorbidity and disability are significant challenges in aging populations.
• Understanding specific multimorbidity patterns is crucial for effective interventions.

Purpose of the Study:

• Identify clinically significant multimorbidity patterns in older adults.
• Assess the impact of these patterns on disability levels.

Main Methods:

• Utilized cross-sectional survey data from Chinese adults aged 60+.
• Employed machine learning (self-organizing maps and K-means clustering) to identify patterns.
• Applied ordered logistic regression and 3D surface modeling for analysis and visualization.

Main Results:

• Identified 10 distinct multimorbidity patterns.
• Older adults with multimorbidity faced higher disability risks (OR=1.679).
• Allomnesia-Arthritis (OR=3.976) and Hypertension-Stroke (OR=3.745) patterns showed the highest disability risks.

Conclusions:

• Specific multimorbidity patterns substantially impact disability in older adults.
• Machine learning provides nuanced insights into disease combinations and disability.
• Tailored interventions are essential for managing multimorbidity in aging populations.