The issue of healthcare interoperability has long been the elephant in the room when it comes to the Healthcare Information Technology for Economic and Clinical Health (HITECH) Act, and beyond. This is no longer the case. In the US in particular, interoperability now tops the agenda as healthcare transitions to new ecosystem-based approaches to support value-based care.
This has shone the spotlight on the limitations of the most widely used standard, HL7, as well as the ocean of additional unnecessary complexity caused by coding confusion and extensive customization. The new standard, Fast Healthcare Interoperability Resources (FHIR), is emerging as a critical tool in addressing these challenges, not only in its function as an agile standard capable of supporting healthcare IT modernization, but also in the constructive collaboration it is fueling between vendors, developers, and providers.
FHIR is a more pragmatic, modern standard for healthcare
At the heart of the matter is the extent to which healthcare is inherently more complex than other activities and therefore requires unique approaches to interoperability. The last iteration of HL7, HL7v3, could be said to exemplify this view. One of the goals of HL7v3 was to convey the “mood” of healthcare information. However, the end result is complexity across the board, ranging from much denser messages to extremely high levels of nesting and limited documentation for developers.
The new FHIR is a specification that includes data models, serialization formats in XML and JSON, and a RESTful API for querying clinical data. Its creators approach the interoperability issue with the view that HL7v3 tried to accommodate too much abstraction, and its underlying design is no longer fit for purpose in an environment where healthcare information sharing is a fundamental requirement of application development, rather than an add-on. Perhaps the most pertinent (and pragmatic) aspect of FHIR is its 80/20 rule, which states that elements should only be added to the standard types if 80% of the systems use them. Its architects argue that the “mood” restraints, as a result, are mitigated by well-defined extension capabilities. It’s also worth noting that improvements in coding and documentation, combined with greater use of natural language processing (NLP) and unstructured data analytics, will contribute to greater medical information subtlety where it is needed.
The prospects for FHIR have always been promising, particularly as it was partly borne out of the HL7 consortium and has many strong advocates within the healthcare IT community. Momentum is now growing, as stakeholders – HL7, healthcare providers, vendors, and the Office of the National Coordinator for Health IT (ONC) – drive further development. Recent initiatives, such as the Argonaut Project, are significant as they address the challenge of developing a viable roadmap for FHIR deployment in areas such as consolidated clinical document architecture (CCDA) mapping and security guidance. FHIR is also becoming more integral to other programs such as Substitutable Medical Apps, reusable technologies (SMART), which has created an API platform for healthcare application developers.
2015 Trends to Watch: Healthcare Technology,IT0011-000328 (November 2014)
Effective Information Sharing in Healthcare: Challenges and Opportunities, IT0011-000324 (July 2014)
Charlotte Davies, Lead Analyst, Healthcare