Within the healthcare industry, there is much discussion about the promise of Health Level Seven’s (HL7) Fast Healthcare Interoperability Resource (FHIR) standard and the role it will play in health IT’s future to improve interoperability and data exchange. Transitioning to FHIR, and enabling the use of web technology to manipulate data within the framework of FHIR, will transform healthcare organizations, essentially making it easier for providers to share health data, according to Russell Leftwich, M.D., adjunct assistant professor of biomedical informatics at Vanderbilt University and senior clinical advisor for interoperability at Cambridge, Mass.-based InterSystems, And, adoption of the FHIR standard will help bring healthcare forward, essentially catching up healthcare IT to the technology and domains used by Facebook, Google and Amazon.
However, Leftwich, who is an HL7 International board member, also says that for the foreseeable future, healthcare providers will need to translate between different standards—HL7 v2, HL7 v3 and Consolidated-Clinical Document Architecture (C-CDA) and FHIR—as there will be a need to transform back to earlier standards so that legacy systems can consume data.
Leftwich chairs the HIMSS Office of the National Coordinator for Health IT (ONC) Interoperability Standards Advisory Task Force and has previously served on the HIMSS Interoperability and Standards Committee. In addition, he is co-chair of the IHE USA Implementation Committee. He recently founded and is co-chair of the HL7 Learning Health Systems Workgroup and a past co-chair of the HL7 Patient Care Workgroup. Most recently he served as chief medical informatics officer for the State of Tennessee Office of eHealth Initiatives. Leftwich recently spoke with Healthcare Informatics Assistant Editor Heather Landi about the promise of FHIR in healthcare and the near-term “hybrid” environment that healthcare organizations will be operating in as the FHIR standard continues to be built out.
There is a lot of discussion about the promise of the FHIR standard to transform healthcare organizations. What role do you see FHIR playing in healthcare IT?
There’s no doubt, it will transform organizations. It allows the same type of information exchange that we’re used to in other domains and accessing information on multiple different servers, and that might be information for a particular individual that exists in different electronic health records (EHRs). FHIR will make it much easier to access all that information. However, the other reality is that, for the foreseeable future, we will live in a hybrid world of standards. There’s the existing standards, HL7 v2 and C-CDA, and they are not going away, not in the next 10 or 20 years. They will still be in use, and the HL7 version 2 standard is the most widely used standard, and it’s being used in countless systems out there and people aren’t going to simply replace those systems. They are going to continue to use those systems because that standard does what it does very well, as far as exchanging lab orders and lab results, between the EHR and the lab system. FHIR, as it’s built out over the next two or three years, will be a standard that can be used to translate between standards and allow those systems that are legacy systems using an older standard to still exchange information within the ecosystem of health information data. But there will still be a huge portion of the information ecosystem that use older standards, so having FHIR as one means of translating between different standards will be the reality of the future.
Russell Leftwich, M.D.
InterSystem has created a health information exchange (HIE) platform that’s enabled it to translate between HL7 v2 and C-CDA documents and FHIR, in both directions, so that people can use that as way of aggregating data about an individual and as a patient. If they turn that data into FHIR-based data, then it’s much easier to search and manipulate the data. I think that is one way FHIR will transform things at the point of care because once that data is in FHIR format, people will then use mobile devices to access that data, for things like decision support. FHIR is very adaptable to mobile devices such as smartphones and tablets, but also to mobile devices that are monitoring as well as medical devices, and the Internet of Things will be enabled by FHIR being the mechanism of interoperability. I think it will all feed into this hybrid ecosystem of different standards co-existing and FHIR will be the glue that ties all that together.
What is the latest progress on the FHIR standard?
We’re already starting to see FHIR-based apps and FHIR development going on. There was a panel at a meeting that I attended yesterday where people from four different organizations talked about how they are using FHIR in their organizations. In July, HL7 will have a two day event where different individuals and organizations will demonstrate what they’ve already done with FHIR within their organization. Right now, people have these apps that are enabled to use FHIR to access the data in that organization’s EHRs system but because these apps have been developed with earlier versions of FHIR, they won’t be easily carried across different organizations. Over the next year or two, as FHIR becomes stable enough and more developed, then apps will be developed. Organizations will have an app that you can get from an FHIR app store and actually be able to use that wherever your organization is.
Where does FHIR go from here?
That’s the future that’s coming very soon, the apps that people are using now are more proof of concept within the organizations. They’re using the apps and they are tremendously useful because they allow practitioners to do things and access the data in a way they couldn’t previously. And there’s work to be done with FHIR, as different parts of FHIR are not yet built out, particularly those parts that deal with very complex clinical data elements. The part of FHIR that deals with the more basic things like what’s the information around a patient or information around a particular condition, the way we specify a condition or a lab test, those things are fairly far along in FHIR. But more complicated concepts that relate to specialties and genetics and things like that are yet to be fully built out in FHIR. What needs to happen is that the entire FHIR community and the clinicians in particular need to get together, one data element at a time, and agree on a model, or how to represent that piece of data. You have to have that to really have interoperability across organizations. You have to agree that this is how we’re going to define this particular clinical data element or a particular diagnosis, but it’s not as complicated as some of these more sophisticated data elements.
So, FHIR will not eliminate the HL7 version 2 standard for now. What should health IT leaders be doing to ensure their information systems are prepared for this hybrid ecosystem of different standards?
Organizations certainly need to be planning their information architecture, if you will, for that future, which is probably not that far off, perhaps the next two to three years. I think they need to be strategic about the architecture they build out and the capabilities that are in the information architecture to adopt new standards like FHIR, but also to allow access to data that exists in these legacy systems that use older standards. And that’s because organizations can’t just go out and replace all these systems that are in use now. The average hospital in the US has something approaching 100 different IT systems within their hospital, and many of those systems are legacy systems that use older standards. But those standards work very well for what they are doing. They need to enable to make the best use of that data from these legacy systems in the future, so organizations need to have a strategy to bring these different types of data together.
One of the advantages of FHIR is that, in the past, to create an interface between two of those legacy systems using the existing standards like HL7 v2 and HL7 c-CDA, it took weeks to develop those interfaces. Those same interfaces leveraging FHIR can be developed in a matter of days, sometimes even hours, and having a strategy to adopt FHIR in a way that makes that data aggregation between old systems and new systems is going to be very important to organizations.
So it will be important for healthcare providers to be able to translate between these different standards to be successful, leveraging interoperability to achieve care coordination and value-based care, is that right?
Absolutely, with care coordination being the idea of having a virtual team that’s defined around an individual patient, and of those individuals some will be professionals and some family and community members. And that’s difficult to do right now because there’s no way to put data where it can be accessed by all of that virtual team. But with the FHIR standard, because of its adaptability with mobile devices and because all of the data could be put in a FHIR format, that will enable everybody on the virtual team to then access the data for that patient using a mobile device and using FHIR. And that’s the kind of thing that is already starting to happen within organizations. I would foresee in the next couple of years where it’ll start to happen more and more across organizations where there’s an HIE where they can access data with mobile devices and FHIR. And another angle to this that I think is very exciting is that each member of the virtual team will be able to have their own app, because app development with FHIR is relatively simple. And, in that sort of information ecosystem of the future, an app that’s developed for one place should be usable across the world.
So that’s a big part of the promise of FHIR, to create the universal platform that we have with Android and iOS in order to get an app from an app store that anybody can use on their device, anywhere, so that really is the promise of FHIR. And the idea that different individuals can have an app that suits their purposes instead of having to have a constant interface developed for a single EHR, which is prodigiously expensive. As an alternative, they can have an app that they got relatively inexpensively from an app store that will work wherever they are. I think a lot of people don’t understand that what’s been done with FHIR to this point is more proof of concept than it is proven out. What we’ll see in the next year or two with FHIR will be the realization of that promise of FHIR being a standard that can tie things together and give us access to all of the data about individuals as opposed to the data that’s in a single system.