Matching FHIR interoperability and population health
Sign up for news and updates in population health
March 29, 2016
Population health management comes in many forms, such as care management, risk stratification, clinical and financial analytics, quality measures and more. Credibility and true clinical effectiveness, though, demands a common requirement: a sharable, complete and accurate longitudinal patient chart. Wellcentive believes FHIR exchange could present a significant step toward standardizing how that longitudinal chart is created and maintained. FHIR (Fast Health Interoperability Resources, pronounced “fire”) is a specification from standards developer HL7.org that outlines how to support an application program interface (API) for the purpose of exchanging data among healthcare IT systems. Its goal is to provide web-based access so that pertinent elements of patient data can be extracted for expedient clinical use, rather than the ongoing exchange of larger and potentially repetitive patient chart data through other transport means.
Moving beyond episodic data exchange
To date, most interoperability frameworks have relied on a query/retrieve model of integration, which is not well suited to automatically maintaining a longitudinal patient chart. In turn, this means that query/retrieve models of integration are not good for population health. For example, a primary care provider might only query an HIE for new data when the patient is in the office for a visit. Such a manual mode of operation, which is still quite ubiquitous, does not address the idea that data relevant to that patient can be generated outside of that provider’s encounter. To continue the example, after visiting their PCP, the patient might see a pulmonologist who orders a PFT panel. Any analytics tools in use by the provider or the health system would somehow have to know to query the pulmonologist’s EMR to get that new information, without any sort of explicit triggering event. If you consider the need to perform this operation several hundred thousand times a day for all the potential patients in a given community, it’s clear how a query/retrieve model of integration would necessitate an untenable number of polling queries.
FHIR subscription and population-based care
Fortunately FHIR appears to have answer to this problem by supporting a push method of integration. The concept of push integration has proved invaluable in multiple settings already, such as ADT feeds from hospitals, or result feeds from labs. As such, Wellcentive is excited to see how we can apply it to these and other scenarios. The specific resource that FHIR uses for push integration is its “subscription” resource. Subscriptions are a mechanism in FHIR where a client can tell a server to automatically push new and updated data that matches some criteria. The search criteria for a subscription can make use of the full weight of FHIR server’s searching capabilities. That theoretically means that FHIR clients can set submit complex rules for automatic pushes of data to match their own complex use cases, without the need for custom development.
Maturing the FHIR standard
Like all new standards and use cases, though, there is maturity to be reached, elements we will be working with other industry interoperability stakeholders to fine tune. One potential shortfall here is that if a resource a client is interested in changes, such that it no longer matches the criteria the client provided to the server for the subscription in the first place, the server won’t send any more updates for that resource. For example, if a FHIR client tells a FHIR server “always send me updates to the patient chart with first name ‘John,’ ” but then that chart has its name updated to ‘Bob,’ that FHIR client would not receive any subsequent updates to that chart. Similarly, subscriptions cannot currently handle deletion of resources, on account of the fact that a deleted resource cannot match any subscription criteria.
Despite these hurdles, Wellcentive is pleased to see that FHIR is including functionality to serve integration needs beyond just the point of care.
Success in moving from fee-for-service to value-based care or population-based payments will require interoperability frameworks that push the boundaries of how we exchange data.
Advising and working with ONC
If there’s an interoperability czar within healthcare delivery and health IT, it is the Office of the National Coordinator for Health Information Technology (ONC), a division of the U.S. Department of Health & Human Services (HHS). Along with its traditional role of health IT certification for meaningful use, which is set to segment and expand for MACRA and other quality reporting programs, ONC has become the conscience, if you will, of establishing national interoperability, most workable as a public/private endeavor. To that end, Wellcentive has been interacting with ONC on several fronts. Last year, we commented on ONC’s broad interoperability roadmap for Most recently, we advised ONC’s annual data exchange standards review by promoting FHIR’s place within document sharing, messaging and its resource hierarchy. The thread throughout all of this communication is that population health management, value-based care We see our dedication to maturing FHIR and identifying population health use cases as a step in that direction.
About the author
Greg Fulton,
Industry & Public Policy Lead, Philips
Greg Fulton is Industry & Public Policy Lead for Philips PHM. He has extensive health IT experience in government relations at Congressional, Health & Human Services, state and industry organizational levels. He is a current member of the CommonWell Health Alliance Government Affairs Advisory Council and the HIMSS Government Relations Roundtable.
You are about to visit a Philips global content page
Continue