Status Update on Data Required to Build a Learning Health System.

Area of Science:

• Health Informatics
• Clinical Data Management
• Health Services Research

Background:

• Electronic health records (EHRs) are widely adopted, creating expectations for large-scale real-world data analysis.
• Real-world data from EHRs is anticipated to enhance patient outcomes and optimize healthcare resource utilization.
• A learning health system relies on accessible, high-quality, and structured clinical data.

Purpose of the Study:

• To detail the current state of point-of-care data within EHR systems.
• To identify necessary changes for leveraging EHR data to establish a learning health system.

Main Methods:

• Analysis of the current landscape of EHR data.
• Review of challenges in data accessibility and quality.
• Description of requirements for a functional learning health system.

Main Results:

• The majority of EHR content is unstructured and resides in proprietary systems.
• Significant challenges exist in conducting accurate clinical outcome analyses.
• Lack of standardized, structured data impedes large-scale research and system improvement.

Conclusions:

• Transforming EHR data from unstructured to structured formats is crucial.
• Interoperability and data quality improvements are essential for building a learning health system.
• Systemic changes are required to fully realize the potential of EHRs in healthcare improvement.