Exploring Multimorbidity Patterns in older hospitalized Norwegian patients using Network Analysis modularity

Area of Science:

• Gerontology
• Network Science
• Public Health

Background:

• Understanding multimorbidity patterns (MPs) is essential for effective healthcare planning and resource allocation.
• Identifying these patterns can significantly improve patient outcomes.

Purpose of the Study:

• To demonstrate the application of Network Analysis (NA) for exploring multimorbidity patterns in hospitalized elderly Norwegian patients.
• To identify and interpret clusters of co-occurring chronic conditions.

Main Methods:

• Utilized data from the Norwegian Patient Registry (2017-2019) for patients aged 65+ with multiple chronic conditions.
• Defined multimorbidity and identified chronic conditions using the Chronic Condition Indicator Refined (CCIR) list.
• Employed Relative Risk and Phi-correlation to assess condition associations, constructing a multimorbidity network analyzed with Louvain community detection.

Main Results:

• A network of 539 chronic conditions revealed distinct multimorbidity patterns, including cardiorenal, metabolic-cardiovascular, and respiratory disorders.
• Identified influential conditions within each pattern using network centrality measures.
• Developed interactive network and sunburst graphs for public access to visualize these patterns.

Conclusions:

• Network analysis and modularity detection effectively identify multimorbidity patterns in older adults.
• The study underscores the complex interplay of chronic conditions in the elderly.
• NA methodology offers a powerful approach to exploring these intricate disease relationships.