Health systems explore Bulk FHIR to unlock pop health insights

LAS VEGAS – Healthcare organizations are gaining a new way to exchange and analyze clinical data at scale, as Bulk Fast Healthcare Interoperability Resources APIs begin to mature.

While many providers are familiar with patient-level APIs created under the 21st Century Cures Act, the law also introduced Bulk FHIR capabilities that allow organizations to access standardized data across entire patient populations.

During a session this past week at HIMSS26 here, Anna Taylor, assistant VP of population health & value-based care at MultiCare Health System, and Dr. Don Rucker, chief strategy officer at 1upHealth, discussed how Bulk FHIR could reshape population health analytics and value-based care strategies by enabling providers and payers to retrieve large datasets through standards-based APIs.

"What Bulk FHIR does is move the power of FHIR to a population of patients," Rucker said during the session, which was moderated by Michael Westover of Providence Health & Services.

Historically, most FHIR implementations have been limited to one patient at a time, which creates challenges for organizations trying to analyze trends across thousands or millions of records.

Bulk FHIR introduces a batch-style approach that allows providers and payers to extract large datasets, stream them into analytics platforms and use them as the starting point for data pipelines and population-level analysis.

The approach also standardizes authentication and authorization, creating a consistent way for covered entities and business associates to negotiate data exchange agreements.

"In the transition to AI, Bulk is a huge building block in getting more complete and more longitudinal data," Rucker said. "You get consistent data across broader ranges to really see the patient holistically."

Taylor explained initiatives such as the HL7 Da Vinci Project are helping accelerate adoption of these capabilities across the healthcare ecosystem.

The project brings together providers, payers, EHR vendors and technology companies to build FHIR-based standards designed specifically for value-based care programs.

The goal is to enable secure business-to-business data exchange between providers and payers so organizations can access coverage information, benefits data and clinical insights when they are needed.

"We want higher quality outcomes, lower costs and higher levels of patient satisfaction," Taylor said.

Standards such as Bulk FHIR play a key role in that effort because they allow organizations to exchange data consistently across systems and organizations.

"Standards can be an electrical grid for the future in healthcare, especially for Bulk FHIR APIs," Taylor said.

The broader objective, she added, is to move healthcare data exchange beyond today's fragmented environment.

"We want to transform out of controlled chaos through collaboration," Taylor said. "If we can all get there through this accelerator, we reach a new floor for data liquidity across healthcare."

However, technical barriers remain. Westover noted some EHR systems still limit how patient populations can be managed dynamically, and restrictions may exist on how many records can be included in a Bulk FHIR export.

"It's very important to be able to do that," he said.

In some cases, extracting data in bulk can even be slower than retrieving information one patient at a time because of limitations in underlying EHR infrastructure.

"The foundational bottleneck to get all this data sharing to work in bulk is getting clinical data at scale out of the EHRs," Westover said.

For organizations pursuing advanced analytics or AI applications, that scalability challenge becomes even more important. Health systems must be able to pull large datasets covering entire patient populations to support risk modeling, predictive analytics and population health strategies.

"At Providence we should be able to pull data at scale," Westover said. "We need to be able to get it for the entire population in one block so we can do risk profiles and population analysis."

WATCH NOW: How nurses can improve AI deployments