Network-medicine framework for studying disease trajectories in U.S. veterans

Area of Science:

• Health Informatics
• Network Science
• Clinical Epidemiology

Background:

• Understanding disease progression and co-occurrence is crucial for improving healthcare interventions and patient outcomes.
• Electronic Health Records (EHRs) contain vast amounts of data on patient health trajectories.
• Existing methods may not fully capture the temporal dynamics of disease relationships.

Purpose of the Study:

• To develop and present a novel network-based framework for analyzing the sequential and temporal aspects of disease relationships.
• To leverage a large-scale EHR dataset to construct and analyze a Temporal Disease Network.
• To identify disease groups, patient flow patterns, and disease trajectories for improved healthcare insights.

Main Methods:

• Utilized a network-based framework to analyze disease relationships from EHR data.
• Constructed a Temporal Disease Network incorporating over 9 million patient records from the Veterans Health Administration (VHA).
• Analyzed network properties to reflect clinical aspects and identified disease groups and patient flow.

Main Results:

• Demonstrated that network properties of the Temporal Disease Network correlate with clinical characteristics of diseases.
• Identified distinct disease groups based on co-occurrence patterns and patient transitions between diagnoses.
• Developed a strategy for identifying disease trajectories, mapping pathways from one condition to another.

Conclusions:

• The Temporal Disease Network framework provides valuable insights into disease co-occurrence and progression.
• This approach can enhance disease treatment and prevention strategies within large healthcare systems.
• The findings support the use of network analysis on EHR data for understanding complex health patterns.