As a technologist, I am always skeptical when hearing about a brand new technology that is going to solve all of our current problems. I feel it is my responsibility to warn people about these types of promises.
Claims of “solving all healthcare information exchange (HIE) problems” are often being made about the emerging HL7 FHIR standard. According to Gartner’s hype cycle, the standard is at the “Peak of Inflated Expectations,” which I view as a stage where wishful thinking fills the gap created by a lack of knowledge and understanding.
FHIR, which stands for Fast Healthcare Interoperability Resources, is a draft standard from HL7 International. This standard is designed to be easier to implement, more open and more extensible than its predecessors, V2 and V3 standards.
So what’s the big deal here? Aren’t we simply replacing one transport protocol with another?
FHIR claims to bring a change in paradigm to the way health information can be accessed and consumed. Since FHIR is a RESTfull API, it allows access to the source of truth on demand and in real time. This capability has many advantages over the message-based, push-and-deliver model of other HL7 protocols.
Applications that build on FHIR do not have to copy and sync data from one system to another, a process that is often cumbersome and lengthy. There is no risk of data being stale, which is crucial in many clinical scenarios. Bi-directional communication, which was always a hassle in previous versions, is an integral part of FHIR. The fact that FHIR is an open standard (all documentation can be found on the web) and leverages modern web-based technologies such as HTTP/S, JSON, OAuth for authorization and ATOM for query results, means that most developers can adopt it easily.
Talk of FHIR includes all the right buzz words, but like any standard, it is just a tool, and the success of the tool depends on its adoption. So far, the adoption of FHIR has been quite remarkable. Although the current 3.x release is still defined as a “Trial Use Specification,” and only Release 4 is slated to include content that will be considered stable and ‘locked’, many of EHR vendors like Epic, Cerner, Meditech, eCW and iMDsoft announced their support for the standard. FHIR-based solutions have been piloted in many mobile health, pharma, precision medicine and public health sectors.
FHIR, however, faces the same challenges as the popular V2 messaging protocol: inconsistent implementations, multiple sub versions, and partial adoption by the community.
The FHIR standard is very broad and includes over 100 resources which are considered the core elements of healthcare. It covers not only clinical data but billing and payments as well. The patient resource, for example, supports animals and species. This wide coverage becomes a disadvantage when companies want to achieve complete adherence to the standard.
In order to mitigate these challenges, FHIR profiles, which are subsets of the standard with extended definitions, are being created. One particularly important profile is the Argonaut Project, which is a joint project of major US EHR vendors and hospitals such as Beth Israel Deaconess Medical Center and the Mayo Clinic. The INTEROpen group also provides clinical validation of FHIR profiles for use in the NHS and social care.
So what will the future HIE look like if everyone makes the switch to FHIR? What practical solutions will be available to care delivery organizations (CDO) in this new world that will truly make a difference?
In my opinion, the main benefit of FHIR is the ability to tap into a clinical repository on demand and interact with it. This capability brings interesting opportunities for mobile apps and other “add-on” products that can complement an enterprise EHR and that are sorely missing in today’s world. CDOs are looking to extract more value out of their expensive IT investments. FHIR has the potential to create an ecosystem of value-added products that EHR vendors are not willing or do not know how to provide. Startups and small companies are struggling to penetrate the market due to the complex and often costly integration work required.
On that note, it is important to mention the “SMART on FHIR” project sponsored by Boston Children’s Hospital. SMART, which stands for “Substitutable Medical Applications, Reusable Technologies”, is an open-source standard designed to enable “plug-in apps” to run natively inside any compliant EHR. SMART builds on FHIR and adds the necessary elements, such as context sharing and security, for developing clinical apps. In the SMART gallery there are already 41 apps, like nomogram and score calculators, that can supplement the capabilities of an enterprise EHR.
The issue of interoperability was solved in industries such as banking, airlines, and tele-communication, but we have yet to find a satisfactory answer for healthcare. Interoperability tears down the walls and allows innovation to flourish. FHIR meets the healthcare IT industry at a fortuitous moment. CDOs are starting to evaluate new technologies beyond what is provided by their EHR, such as predictive algorithms, machine learning, mobile access from any device, new decision-support methods, and real-time surveillance tools. Like PCs or smartphones, EHRs can become the platform for running all kinds of interoperable products, and the value of that interoperable ecosystem can surpass the value provided by the core platform.
FHIR still has a long way to go and many hurdles to overcome if it is to meet the high expectations of the HCIT community and fulfill the promise of seamless integration. At this point, I can only speculate about whether seamless integration will become a reality and when. If it does, however, I feel confident that FHIR will play an important role in that new world.
Comments
Post a Comment