FHIR for Clinical Research
HL7 FHIR in Clinical Research and Biopharma
[FHIR] (Fast Healthcare Interoperability Resources) is an HL7 platform standard that defines a set of rules that describe how computers use the web to share and understand healthcare information. By having a common set of rules, computer programmers can write programs that will reliably share healthcare information without having to deal with the variances and complexities of each individual system they’d like to connect with. That saves a lot of time (and money) that can be applied to develop better software to more rapidly provide more effective solutions.
FHIR is about "Healthcare" - a wide net. It covers:
¥ Descriptive information about the people, places, and things involved in healthcare (patients, clinicians, clinics, devices, drugs, etc.) ¥ Clinical observations and summary information (allergies, problems, medications, lab results, imaging data, vital signs, etc.) ¥ Appointments, referrals, procedures and other processes that help ensure the right care gets delivered at the right time ¥ Care protocols, decision support, and clinical quality & consistency measures - to help improve care quality and consistency ¥ Financial processes such as billing and insurance ¥ Clinical research and other biopharmaceutical processes ¥ and a whole lot more, with new uses showing up continuously.
Fundamentally, FHIR is a standard that defines the kinds of health-related information a patient would like to have about themselves and what a physician or researcher would like to know - their record of medical care, as well as the wider processes that are necessary to provide that care.
FHIR defines a consistent, structured format that can offer a consistent view of health data for caregivers, payers and researchers, and potentially allows patients to load additional health-related data into their own structured personal health records. More importantly, that information is defined in such a way that it's easy to find, sort, manage and filter much more efficiently than having a great big list of documents that have to be read each time. In the words of John Halamka, “FHIR is the HTML of healthcare.” Grahame Grieve, the architect of FHIR, refers to it more succinctly as “The web, for healthcare.”
Unlike many other healthcare standards, FHIR is free. No payment or license agreement is required to use it (whether commercially or otherwise), and the standard is posted free as open source to the web for everyone’s benefit. FHIR is developed using standard open community processes and the rapidly expanding FHIR community (which includes thousands of implementers worldwide) works hard to make high quality information and support freely available. The aim of FHIR is to reduce the cost and effort of exchanging and using healthcare information, so that data sharing becomes simple, and healthcare providers can and will set up processes that enable sharing of higher quality health information in a more consistent manner - which will lead to increased insights and better health outcomes.
Flexibility and Fitness to Task
The healthcare system is extremely fragmented - different countries and cultures have very different views of what 'health' is, and how care should be provided, funded and used. Different clinical disciplines have very different approaches to how to represent, collect, store, share and use information in support of their care processes. There's no single authority that can make rules about what clinicians do, and clinicians are taught to be highly self-accountable. HL7 International, which publishes FHIR, has no authority to tell healthcare providers how they should provide care.
The consequence of this is that the FHIR specification is extremely flexible and there's a set of features in FHIR that make it possible to manage this variability. But that comes with a price: programmers that work with FHIR have to deal with this variability - and that's harder (and costlier) than it could be with a more proscriptive standard. So FHIR provides mechanisms such as profiles and extensions to tailor the specification in more proscribed ways for specific uses such as research.
Data Availability, Security, and Compliance
FHIR defines how things can be done. But because it's an international standard, and used in all sorts of contexts, it can't make any hard and fast rules about what information has to be shared, when it has to be shared. Nor can it dictate security policies and approaches. Instead, making those rules about how FHIR should be used in a particular context (called 'implementation guides') is left to a patchwork of regulators, standards bodies, implementation communities and vendor consortiums around the world. However, FHIR utilizes prevalent security methods commonly used in today’s web-centric world and provides building blocks, called resources, for representing key concepts such as provenance, audit trail and consent.
In spite of the challenges around lack of standardization of the overall healthcare process, FHIR is a key to building a patient-focused healthcare system. The ability to reliably share information with patients at an affordable price creates an environment where the patient and their own record can be the pivot around which their care is coordinated. The ability to become more deeply integrated in the care process (or, by extension the research process) - through better information exchange - allows the healthcare system to be more patient focused and responsive rather than provider focused.
Note that the healthcare system needs more than just better information flow to make these kinds of changes. However, better information flow is a critical pre-condition, and that's the focus of the FHIR community.
Applicability to Research
The increased availability of digital healthcare data presents both a challenge and opportunity to clinical researchers in the biopharmaceutical industry. As the industry tries to become more efficient with risk-based and remote monitoring and reducing the time and cost for clinical trials in general, it must be recognized that Electronic Health Record (EHR) data is e-Source. And the essential building blocks of FHIR, called resources, can be thought of as roughly equivalent to CRFs.
FHIR provides the best mechanism to date for feeding EHR and patient-originated data directly to clinical trial databases, and for keeping the study database in sync with the EHR database, to reduce the likelihood of discrepancies with source documents. FHIR can also make it possible to design better protocols that match up with available patient populations, identify higher performing investigators and sites, and allow researchers to more closely monitor how protocols are being executed at the site. FHIR provide real world evidence to support product marketing applications for faster approvals of new drugs and new indications, and give regulators greater insight into the patient record to develop more confidence in the safety and efficacy of new products and to gain greater insight into the safety of approved drugs.
Current State of FHIR To date, the FHIR community has focused on the basic building blocks, working towards being the standard that best enables healthcare data exchange globally. Since its introduction in 2011, FHIR, now in its third major release, has steadily attracted more interest and become the platform of choice for technology solutions centering on health data interoperability. As of 2017, the FHIR team has made significant progress in building that foundation - the standard itself, and more attention is being devoted to focus on supporting use of the standard to provide real world solutions. Hundreds of FHIR implementations and applications are already in use throughout the world, and the FHIR community has become a magnet for start-ups and developers seeking to quickly develop robust solutions for multiple purposes throughout the healthcare ecosystem. As Levin Brown said at a recent Partners in Interoperability conference, “for new projects, there is no reason to do anything else but start with FHIR.”