Nov 7, 2018
Blockchain could displace FHIR for a subset of healthcare integration
FHIR and blockchain will co-exist in healthcare, but which use cases will blockchain be a better choice for?
Co-authored with Ian Fish, IBM Services Partner & Canadian Healthcare Services Leader. This material is our own and don’t necessarily represent IBM’s positions, strategies or opinions.
Global healthcare interoperability has been an ongoing challenge. Health data is more complex than many other data sets and communities of practice exist within different areas of health care which have disparate standards for recording data. Blockchain solutions have the potential to help in a portion of interoperability use cases.
In the wake of the global failure of HL7v3 as an interoperability standard, FHIR (Fast Healthcare Interoperability Resources) has emerged as a much nimbler, lower cost integration standard built on modern technologies and more realistic expectations of what to expect from health integration layers.
But now there’s blockchain. And one of its most exciting use cases is to apply it to the age old dilemma of healthcare interoperability to support the sharing of health data between systems in a very different way. It’s a repository and related synchronizing messages in a single easy-to-use and secure bundle. The promise is high for a subset of health integration requirements, but not for all. FHIR is still an invaluable part of the toolkit.
There are several potential lenses to look at the two technologies in context of healthcare integration solutions. We’ve identified ten.
Let’s look at each of these lenses in turn.
Persistence: FHIR is a collection of transient messages useful for moving data between persistent data stores or for on demand consumption. Blockchain is a persistent data store of information. It’s an immutable record of not only the data but the messages which change the data.
Synchronization: FHIR assists registries to synchronize themselves but has no inherent features beyond the messages. Blockchain solutions inherently create synchronized registries in multiple locations.
Identity: FHIR supports Provider and Client message sets but is a transient messaging medium about them. Blockchain solutions are an immutable shared ledger and are considered a very strong solution for shared electronic identity. The two primary emerging solutions for Canadian healthcare are the new SecureKey VerifyMe service with Hyperledger Fabric for identity and the Hyperledger Indy supported by it’s developer the Sovrin Foundation and adopted by the World Health Organization.
Provenance: FHIR messages are point-in-time interactions between systems so could be used to query a provenance system. Blockchain solutions contain the history of interactions through time and are effectively a provenance solution in and of themselves.
Consent: FHIR messages depend on external systems for consent management. Blockchain solutions enable internalized consent management with easier Client ownership of consent for use through public-private key encryption. There are several consent-oriented blockchain initiatives under way.
Standards: FHIR is laying out and gaining consensus on standard message payloads. They are in the middle of their five year maturation process and are at maturity level 3 of 5 as of 2018. For blockchain, no payload standards have been set, however, adopting and adapting FHIR payloads may be easy. No standards organization is working on blockchain registry persistence standards and the maturity level has to be considered at level 1.
Architecture: FHIR integrates different components in an interoperability architecture. FHIR is a messaging solution alternative to HL7v2, HL7v3 and CDA. Blockchain is a component in an interoperability architecture which automatically refreshes itself across multiple sites. Blockchain is an alternative to distributed databases and message sets which assist them to remain whole and accurate.
Fiscal vs medical: FHIR is completely focused on medical data and does not attempt to address billing or claims solutions.Blockchain solutions are inherently distributed ledgers suitable for accounting and claims solutions, but are general purpose and can support medical data easily.
Data obfuscation: FHIR messages depend on external cryptography and security to shield data from inappropriate eyes. Blockchain solutions are inherently based on public-private key encryption schemes and internal data can be obfuscated by encryption and only seen by permitted parties.
Performance: Individual messages with FHIR are much more rapidly delivered. In blockchain solutions, most parties effectively have caching of complete information locally, but new transactions can take minutes to hours (depending on blockchain technology) to work through consensus, finalization and distribution. Where a new piece of information must be delivered to a non-local system immediately, FHIR has a significant advantage.
Through these lenses, a subset of the potential ways we could look at blockchain and FHIR as we look at different requirements, some clarity of when to select FHIR versus when to select blockchain emerges in the form of high-value use cases.
Identity registries: Provider and Client registries are an obvious area where blockchain solutions can provide immediate value. The data within them is not medically specific, does not have medical language set challenges and must be immediately available in multiple systems in any jurisdiction for best outcomes. It’s easy to see blockchain and FHIR co-existing.
Claims management: Fiscal claims management solutions are well addressed by differing blockchain technologies. As the IBM Institute for Business Value’s Healthcare Rallies for Blockchain report identified, a majority of early adopters viewed claims as one of the key disruptive opportunities.
Patient-centric consent: The ability to obfuscate data internally to a patient’s set of data, to have the patient give consent by providing their public key to healthcare providers and to easily differentiate between identifying information and non-identifying information has significant potential for both better protecting patient data and making it more available for research purposes.
Pharmaceutical provenance: As a persistent and shared record of every interaction and temperature variation throughout a medical supply chain from manufacturing to delivery, blockchain solutions have the strong potential to improve outcomes. Biological pharmaceuticals such as vaccines are an obvious use case as we discussed in an earlier post.
The analysis here is intended to provide the start of a more fruitful discussion of where it has value compared to other healthcare interoperability solutions. There is tremendous opportunity in blockchain to solve existing healthcare interoperability problems more effectively, quickly and simply. And FHIR and blockchain are different solutions which have both have merit for different subsets of the requirements of healthcare.
We are strong proponents of FHIR, but we are excited by blockchain’s potential in use cases where FHIR may not add substantial value.
A version of this post formerly appeared on IBM’s Institute For Business Value’s Healthcare & Life Sciences Industries Blog.
