In the previous article on WellSky’s Enterprise Patient Matching, I shared how accuracy in retrieving patient records is built. There are challenges of processing the intake of millions of health records into one of our FHIR ecosystems daily.

FHIR aims to provide a common, structured way to represent healthcare information, which is crucial for tasks like sharing patient data across many providers. Implementing this on a large scale is not a simple plug-and-play solution. A few challenges are:

  • Data ingestion variety: EHRs have diverse healthcare data with varying levels of data quality and identifiers
  • Data quality: Usefulness for downstream applications supported by accuracy and completeness
  • Scaling and responsiveness: Low latency and efficiency when handling millions of records daily

WellSky has a vision for interoperability across the care continuum with one goal being to share data to enable informed data-driven decisions. Generally, solutions store data separately, and these can be structured in vastly different ways. Each source system has unique data models that need to be mapped to corresponding FHIR elements. This system also must ensure high data quality while still processing millions of records daily.

The team has built a robust, scalable data pipeline (diagrammed below) that involves collecting data in various raw forms, and mapping into standardized formats. That data is then transformed into FHIR resources stored in centralized repositories.

Reusable data mappers and transformation logic allow for consistent mapping and data enrichment. Several “Designated Record Sets” (DRS) were created to define which data should be shared to FHIR resources. This also addresses USCDI standards requirements.

Handling high data volume is where this serverless technology shines. Individual processing tasks automatically scale. Inbound records are published to message queues, where functions spin up to transform and load the FHIR resources in a cost-efficient manner. These mechanisms make it easy for data owners to identify and correct issues in their data.

With these strategies and architecture, WellSky is prepared for the future of FHIR at scale. By combining data standards such as FHIR with the scalability of modern cloud technologies, we unlock new possibilities for improving patient care and optimizing healthcare operations across other forward-thinking organizations.

‍