Reducing complexity: a visualisation of multimorbidity by combining disease clusters and triads

Area of Science:

• Gerontology
• Network Medicine
• Epidemiology

Background:

• Multimorbidity is common in the elderly, leading to adverse outcomes like mortality and functional decline.
• Previous studies identified multimorbidity clusters but lacked insight into internal disease structures and cluster linkages.
• Understanding these associations is crucial for managing complex health conditions in older populations.

Purpose of the Study:

• To map person-level disease associations within multimorbidity clusters.
• To identify diseases responsible for overlapping multimorbidity clusters.
• To visualize the structure of disease networks in older adults.

Main Methods:

• Analysis of insurance claims data from 43,632 females and 54,987 males aged 65+.
• Identification of disease triads using observed/expected ratios (≥2) and prevalence rates (≥1%).
• Network analysis of disease associations (edges) to link triads and multimorbidity clusters.

Main Results:

• Identified 57 triads (31 diseases, 200 associations) in females and 51 triads (29 diseases, 174 associations) in males.
• Cardiovascular and metabolic disorders formed large clusters (12-14 diseases).
• Anxiety, depression, somatoform disorders, and pain clusters involved 12-15 diseases.

Conclusions:

• Chronic low back pain was the most significant disease mediator, linking various conditions.
• Depression strongly associated with other diseases in females; hypertension central in metabolic syndrome disorders.
• Metabolic syndrome and musculoskeletal disorders showed prominent interconnections, informing clinical guideline development.