Bulk FHIR at Scale: Why APIs Are Only the Beginning

At the Civitas Networks for Health 2026 Annual Conference, Idaho Health Data Exchange (IHDE) and Orion Health shared how they completed a bulk FHIR project for a payer partner: about 429,000 members and more than 3.4 million API calls. Health IT Answers covered the session.

The big takeaway: FHIR APIs offer a more standard way to exchange healthcare data, but moving data at scale still takes real engineering, governance and collaboration.

Taking On the Complexity for Our Partners

The payer wanted a bulk FHIR dataset for analytics. IHDE had already seen another payer struggle to use that kind of data on its own, so IHDE took on more of the integration work and delivered the data in a format the payer could use right away.

IHDE Executive Director Jesse Meldru describes the HIE's role as a “trusted third-party intermediary”: not just a pipe for data, but a partner that absorbs complexity for the organizations it serves.

When Scale Changes the Equation

The project took roughly 50 hours of continuous data pulls. The teams managed short-lived security tokens, many parallel workers and failed processes, all without overwhelming the systems underneath. They built a “heartbeat” monitor that noticed slowing response times and automatically eased the workload.

Security and Governance Built In

IHDE used patient matching so only members on the payer's roster were included. About half of those members had lost eligibility during the period, so clinical information created after each member's eligibility ended was removed.

FHIR Isn't Always the Final Format

Raw FHIR data arrives as nested JSON, which was costly and awkward for the payer's analytics tools. The team converted it to Parquet, a format built for large-scale analytics. IHDE also added information the FHIR APIs didn't readily provide, such as blood pressure readings from HL7 messages and provider identifiers matched against registry data.

The Last Mile of Interoperability

The lesson isn't that FHIR failed. It's that interoperability doesn't end when data meets a standard. As organizations gain access to more standardized data, the challenge becomes making it secure, usable and fit for its purpose. APIs provide the connection; HIEs like IHDE help handle the last mile.