Learning health system linchpins: information exchange and a common data model.

Area of Science:

• Health Informatics
• Population Health Management
• Learning Health Systems

Background:

• Centrally managed health information exchanges (HIEs) aggregate data from multiple partners.
• Standardizing HIE data to a common data model (CDM) is crucial for interoperability.
• Learning health systems (LHS) require seamless semantic data flow for continuous improvement.

Purpose of the Study:

• To demonstrate how a health information exchange common data model (HIE-CDM) facilitates semantic data flow for a learning health system (LHS).
• To showcase the potential of standardized HIE data for observational population health research.
• To illustrate leveraging existing health IT infrastructure for LHS advancement.

Main Methods:

• Operated a statewide HIE aggregating data from over half the state's population across 47 partners.
• Standardized HIE data to the Observational Medical Outcomes Partnership (OMOP) Common Data Model (CDM).
• Analyzed Atherosclerotic Cardiovascular Disease (ASCVD) risk and primary prevention practices from 2013-2023.

Main Results:

• Calculated longitudinal 10-year ASCVD risk for 62,999 individuals, revealing risk factors from multiple data partners for many.
• Enabled granular tracking of individual ASCVD risk, primary prevention (statin therapy), and disease incidence.
• Found suboptimal statin adherence and identified disparities in ASCVD risk profiles and statin use among Federally Qualified Health Center patients. CDM transformation unified data and reduced heterogeneity.

Conclusions:

• Demonstrated the potential of HIE-CDM to support observational population health research.
• Showcased the ability to overcome LHS barriers by utilizing existing health IT infrastructure and data best practices.
• Confirmed that HIE-CDM facilitates knowledge curation and intervention development at multiple levels (individual, health system, population).