On FHIR: Simplifying data sharing to improve public health outcomes

Big data is big business in global health care. RBC Capital Markets estimates that "by 2025, the compound annual growth rate of data for healthcare will reach 36%."

As the amount of health data increases, hospitals, healthcare providers, and public health agencies must accelerate their ability to securely share this data to manage future emergencies such as infectious disease pandemics and other significant threats like heart disease, cancer, and suicide.

The U.S. federal government is responding to that pressure. Last December, the Centers for Medicare & Medicaid Services (CMS) proposed the Advancing Interoperability and Improving Prior Authorization Processes rule, which requires payers to establish application programming interfaces (APIs)—definitions or protocols that allow two software components to communicate with each other—to share patient health information. The rule specifies that the recommended approach is to use Fast Healthcare Interoperability Resources (FHIR®) to "advance an API ecosystem based on open-industry standards across the health care value chain." 

HL7® (Health Level Seven®) FHIR include a set of best practices and open standards that have been developed and adopted by a global community to make data sharing more flexible and effective. Several federal agencies are already implementing FHIR in their data modernization efforts, and their experiences have provided valuable insights about leveraging this vital resource.

FHIR 101: Understanding the latest interoperability standards

HL7—an international organization that provides frameworks for sharing electronic health information—launched FHIR 2012 to facilitate the exchange of data across computer systems that store data differently. Earlier standards are available (V2, V3, and CDA), but FHIR has several advantages:

• FHIR uses concepts familiar to software engineers, reducing the learning curve for developers with technical skills but little experience in the healthcare industry.
• HL7, in cooperation with the healthcare community, has created several FHIR implementation guides. These guides offer agencies and organizations guidance on how to implement FHIR in their domain.
• Placing FHIR APIs on top of healthcare data increases interoperability. It allows for the creation of apps, quite like how the public release of GIS and location/point-of-sale APIs enabled the rise of Lyft, Uber, and similar apps.
• FHIR allows a single location the ability to run analytics to query data for each hospital in a system. Developers can use standard authorization, authentication, and security measures to protect sensitive data as it's moved from hospital to hub and back.

Leveraging FHIR to improve data reporting during public health emergencies

Early in the COVID-19 pandemic, public health agencies at all levels scrambled to get data out of hospitals inundated by patients. But there were no standards for American hospitals to collect that data, much less share it with the agencies. Without guidelines, hospitals created their own solutions, resulting in a slow, cumbersome process that hampered the nation's response to the crisis.

The Helios FHIR accelerator project aims to solve these issues by simplifying data sharing between hospitals and public health agencies. ICF is investigating how FHIR APIs can be integrated into existing electronic health record (EHR) systems to standardize tracking of aggregate data, like open ICU beds, ventilators, and personal protective equipment (PPE). This would free hospitals to focus on patient care and enable agencies to quickly access real-time data.

Using FHIR APIs to streamline CMS provider scoring

Under the Affordable Care Act, participating medical providers must report annual quality metrics to CMS. Although standards for reporting this data exist, the process is complicated because different providers track encounters and procedures in different ways.

ICF is working with CMS to develop a system to standardize this process. By placing FHIR APIs on top of the data in providers' EHRs, it's possible to write logic that can query the EHR data using standardized development resources and standard healthcare code systems to deliver the provider's CMS score.

For example, one CMS quality measure for diabetes requires a yearly eye exam for each patient. FHIR can query provider EHRs for patient records that include billing codes associated with a diabetes eye exam. The logic may look like this: "Give me a count of all the patients between ages X and Y, with a diagnosis of diabetes, who have recorded any of the following billing codes between Month/Year and Month/Year." This allows for the efficient collection of this data regardless of how an individual provider may code an eye exam. 

Lessons learned from real-world FHIR implementations

A FHIR implementation may seem daunting for an agency at the beginning of the process—but it doesn't have to be. Consider these proven steps that agencies and developers can take to get a head start on FHIR implementation:

• Use the HL7 registry: In creating FHIR, HL7 assembled work groups with representatives from federal, state, and local health departments to build out a variety of implementation guides for different domains. Individual agencies and organizations already implementing FHIR have also contributed resources. The HL7 registry houses these valuable resources, which can be amended for similar use cases. In short: There's no need to reinvent the wheel when starting an implementation.
• Consult open-source products: One product called HAPI FHIR is a free and fully open-source compliant server. HAPI FHIR provides instructions for how to take its server and wire it into a database. If that database isn't formatted like FHIR—which, most likely, it isn't—HAPI FHIR can help map FHIR resources to specific aspects of the database, like tables and columns. 
• Host a hackathon: Several agencies invite potential users of their data to work alongside their own data experts. The potential users bring cases and scenarios, the data experts bring their tools, and together, they figure out how best to provide easy and secure access to the data. What they discover often becomes requirements for the next generation of APIs.

FHIR is essential to the future of public health data

It's clear that FHIR is and will continue to be a critical tool in the federal technology arsenal for years to come. To thrive in this environment, agencies must understand FHIR and leverage its power. Learn more about the ways ICF is already supporting several federal agencies in their efforts to integrate FHIR to manage health data securely and effectively.