FHIR: another step toward interoperability
It’s the “holy grail” of healthcare IT: an easy-to-use, plug-and-play solution to the technological challenges of clinical integration.
If accountable care organizations could flip a switch to connect their providers and enable seamless communication between disparate EHRs? Well, you can bet we’d find ways to occupy our time worrying about issues other than interoperability.
The problem, of course, is this chalice doesn’t exist, and for now, anyway, clinical integration is hard work. And while one proposed standard for healthcare-data exchange does appear poised to make interoperability more attainable, it is just a step in the right direction—not the panacea some have made it out to be.
A Matter of Visibility
That standard, known as FHIR—for “Fast Healthcare Interoperability Resources”—is part of a framework for data exchange and integration developed by the organization Health Level Seven International. (“HL7” members include healthcare providers, payers, EHR vendors, and many other industry stakeholders.) The promise of FHIR is that it could ensure there is semantic interoperability between clinical data—that different EHRs at different institutions could “talk” to each other using common languages and vocabularies.
The technology would make it possible for physicians, for example, to access information at the “granular” level they might need to make decisions about individual patients. It would also allow consumers, for instance, to see their personal data on a mobile app they might use to help them manage their various health conditions. Such visibility is difficult today because the current data-exchange standard of choice centers on the transfer of entire documents. What if the lab results for a patient are stored on EHR A and the doctor wants to see them on EHR B? In today’s world those results might come through, but only in a form that is essentially static—not one permitting interaction with the data.
It Takes a Team
All of that’s to say that FHIR is real progress, and exactly the kind of fix that healthcare needs if we’re going to make interoperability a reality. Still, knowing it’s possible to do something is one thing; having the expertise to pull it off is another matter altogether.
For as Cerner’s senior vice president of medical informatics David McCallie points out in a recent article in Health Data Management, “It’s not so simple as to say, ‘Go use FHIR and we’re done.’” The technology isn’t ready “out of the box”; it must be tweaked and tailored to each health system’s needs.
At Mercy Health, where our ability to effectively manage population health hinges on successful clinical integration with the EHRs of the affiliate Primary Care Providers in our ACO, we’ve determined that this semantic normalization will be far too difficult to tackle on our own. So instead we’ve turned to the team at Watson Health and the functionality of their IBM Explorys Platform. Their team of clinical content experts are able to do the complex mapping between different EHR vendors’ data models and the semantic normalization standards like SNOMED-CT and LOINC. Then, the discrete EHR data and paid claims data lands in the IBM Explorys Platform, where it can be used to generate risk scores and calculate care gaps. This work is painstaking and lengthy – but it is precisely what separates the wheat from the chaff in terms of ACO data aggregation. With Watson Health as a strategic partner, the complexities of clinical integration won’t be ours alone; they become solvable challenges.
Maybe in the future clinical integration will be as simple as “plug and play.” Until then, I’m thankful we have others to help us get the job done.
This article was first posted at Watson Health Perspectives.