High Performance Computing on Flat FHIR Files Created with the New SMART/HL7 Bulk Data Access Standard

Area of Science:

• Health Informatics
• Data Science
• Clinical Systems Engineering

Background:

• Population-level data extraction from clinical systems is complex and resource-intensive.
• Current methods often require specialized Extraction, Transform, and Load (ETL) expertise at each healthcare site.
• Standardized data access is crucial for large-scale health research and analytics.

Purpose of the Study:

• To introduce a uniform capability for population-level data exports from clinical systems.
• To leverage existing authentication and authorization standards for secure data access.
• To simplify the process of defining and accessing patient cohorts from diverse data sources.

Main Methods:

• Utilizing the Fast Healthcare Interoperability Resources (FHIR) Bulk Data API.
• Implementing the SMART (Substitutable Medical Applications, Reusable Technologies) backend services authentication and authorization profile.
• Developing a "Flat-FHIR" file format for standardized data output.

Main Results:

• Achieved a uniform capability for exporting population-level data.
• Enabled "push button" simplicity for accessing cohorts from electronic health records and payor claims data.
• Significantly reduced the reliance on highly skilled ETL efforts for data extraction.

Conclusions:

• The FHIR Bulk Data API offers a substantial advance in accessing population-level health data.
• This approach enhances efficiency and accessibility for healthcare providers and organizations.
• Standardized, simplified data access can accelerate health research and improve patient care insights.