“CDA addresses interoperability for clinical documents, mixing narrative and structured data. FHIR provides granular access to data, a contemporary, streamlined approach to interoperability, and is easy to implement. FHIR can be the future of CDA, but it is not there yet.”A few years ago, we suggested a way to simplify implementation which we called “greenCDA”. This approach broke away from the V3 XML, provided a straightforward XML schema which was one automated transform away from a full CDA Implementation. It wasn’t brilliant, just best-practice, and got some attention, most notably from Grahame Grieve who argued forcefully that the green representations should be reusable. We built a CDA-FHIR/FHIR-CDA transform for our first FHIR Connectathon and found, as so many have, that it was a pleasure to work with the new spec. At the same time, our explorations have pressed up against the limitations of Release 1. We invited Grahame to come spend time with us at our Spring Academy, to teach and to sit with us to think through to the good outcome – a future where the CDA can leverage the same great framework to sustain a full patient record. We have, today, with FHIR, an opportunity to get things right. It is also clear to us that FHIR is not yet ready to answer the call to sustain a full patient record. The vision of seamless flow of data elements from applications to persistent documents to applications is powerful. Our paper asks “Can ‘CDA on FHIR’ replace ‘CDA on HL7 V3’ (the current specification) for all clinical document use cases?” and answers “Yes, it can, here’s what we need to do to make it so”. Please take a look at our Call to Action and let us know what you think.
At Lantana, we are all pretty comfortable working with CDA. I think we are also pretty open to understanding its limitations. The current CDA, Release 2, was developed ten years ago. It was designed to meet a range of requirements from simple, transformed-from-dictation documents to fully coded, semantically interoperable reports. A key design consideration, a requirement for passing ballot within HL7 at that time, was full compatibility with Version 3 messaging and an explicit tie-in to the Reference Information Model (RIM). These latter considerations, and policy, have stressed implementation of CDA R2 as a fully structured data set and given less weight to its low-end, simpler, capabilities. At the same time, the failure of V3 messaging to answer the call has meant that CDA stood alone, sometimes being asked to do things for which it was not designed. The paper we are publishing today, Position Statement: Clinical Documents & FHIR, jointly with Grahame Grieve, Health Intersections, lays out our position this way: