Michelle Miller is one of Health Data Management’s 2016 Most Powerful Women in Health Care IT.
I still remember attending my first Health Level Seven (HL7) work group meeting in Paris, wearing a bright pink “First-Time Attendee” ribbon. I was not new to HL7 Fast Healthcare Interoperability Resources (FHIR), as I had been participating in the HL7 ballot process, but I was new to the face-to-face HL7 work group meetings. Luckily, the HL7 and FHIR communities are extremely welcoming to first-time attendees, which served as the foundation for me to continue to provide feedback and ultimately, influence the HL7 FHIR specification.
HL7 is an American National Standards Institute-accredited, consensus-driven organization whose mission is to empower global health data interoperability by developing standards and enabling their adoption and implementation. The in-person HL7 work group meetings occur three times a year. Each work group is responsible for defining their specific agendas, but generally speaking, ballot resolution is a top priority because work groups must resolve negative votes (i.e., determine if the vote is persuasive enough to warrant changes to the standard).
HL7 FHIR has a focus on those who adopt and implement the standard specification, such that implementer involvement and input is very much respected and encouraged. The HL7 FHIR specification is a working standards framework, which means it’s not frozen and continues to evolve.
Each FHIR resource, which is a building block of exchangeable content, has its own designated maturity level. A FHIR resource maturity level increases with implementer adoption and investment. As the maturity level increases, fewer non-passive changes will be made to the standard specification. Even with my bright pink “First-Time Attendee” ribbon, I quickly learned that my input was valued.
To increase adoption, HL7 encourages participation and collaboration across the community, which leads to requested standard changes. Changes can be submitted during or outside of a formal ballot process. I’ve requested many edits, which have resulted in direct changes to the FHIR specification. These changes range from a simple clarification needed within the specification to mitigate ambiguity to proposed new resources and everything in between, such as element cardinality changes, value set binding changes or new elements within an existing resource.
After getting energized by the fantastic collaboration that occurred during the HL7 work group meetings, I started attending weekly work group conference calls to continue the discussion. Within months, I was elected as an HL7 Patient Care co-chair and granted FHIR committer privileges to edit, build and commit changes within the FHIR specification.
During the conference calls, we often discuss proposed FHIR changes to determine if the change is persuasive. I feel strongly that all changes, big and small, help make the FHIR specification that much better for the next implementer or developer to adopt. Adoption is what makes a specification a standard because without adoption, we haven’t really achieved interoperability.
It’s easier to gain consensus that the change is persuasive when the use case pertains to a common scenario impacting 80 percent of the implementations. The more targeted or customized requests that don’t apply broadly, which we all know exist, can be handled via profiles, which are extensions and/or restrictions of a resource. Providing input on the profiles and implementation guides is equally important. For example, I participate in the Argonaut community providing feedback on the Argonaut implementation guide, which is the FHIR-based data query API implementation guide, heavily influenced by the U.S. Data Access Framework implementation guide, to support the 2015 Edition API certification requirement.
So, what does this all mean for Cerner? We have developed and continue to expand our Cerner Millennium FHIR services. Through our firsthand experience developing those electronic health record FHIR services, we have contributed feedback to the standard. Additionally, we use FHIR resources for inspiration in designing vendor-agnostic data models for Cerner’s population health platform, HealtheIntent, which aggregates, normalizes and standardizes data from disparate source systems.
If you want to connect and collaborate with other FHIR implementers, be sure to check out chat.fhir.org. Alternatively, the work group conference calls are open to anyone who wants to participate, so I encourage you to participate, too!
I have been so impressed with the knowledge, collaboration and overall friendliness of the HL7 FHIR community. The discussion is always thoughtful and respectful, such that I have high confidence in FHIR’s ability to maximize interoperability.