Investigating the use of FHIR in Clinical Research
- 1 Project Title
- 2 Affected Stakeholder
- 3 Project Meeting Frequency
- 4 Definition(s) of Problem/Issue/Challenge
- 5 Potential Technological Solution(s)
- 6 Other Initiatives that can be leveraged/merged
- 7 Recommended Solution(s)
- 8 Specific Actions
- 9 Deliverables and Dates:
- 10 CSS 2017
- 11 Meeting Minutes
- 12 Webinar Presentations
Investigating the use of FHIR in Clinical Research
|Geoff Low||Co-Lead||Medidata Solutionsfirstname.lastname@example.org|
|Aditya Gadiko||Participant||MMS Holdingsemail@example.com|
|Christine K Denney||Participant||Lillyfirstname.lastname@example.org|
|Jeff Abolafia||Participant||RH World||Jeff_Abolafia@RhoWorld.com|
|Lauren White||Project Coordinator||PHUSEemail@example.com|
|Mark McGilchrist||Participant||University of Dundeefirstname.lastname@example.org|
Standards Development Organisations, Software Developers, Data Managers, Regulatory Bodies, Software Vendors
Project Meeting Frequency
1 Hour / Per Week
Definition(s) of Problem/Issue/Challenge
Increasing interest in eSource keeps the issue of data integration between Research Systems (EDC, CTMS, CDMS, etc) and healthcare systems (EHR, etc) as a consistent want for Sponsors of Clinical Investigators and Regulators. Previous efforts to make this a repeatable, scalable solution have not met with widescale adoption, for a variety of reasons.
Some common historical points of view have included:
- That the quality of the data that can be retrieved from the Healthcare sites is insufficient to meet research needs.
- That uptake of electronic systems at investigative sites has been slow, expensive, and not delivering real value to healthcare practises.
- Types of data captured in healthcare have been more operational rather than clinical.
- Enabling the Necessary interfaces is an expensive and process-heavy undertaking.
- There is not a suitable, generally supported electronic exchange format, with a number of standard representations being supported in recent memory.
Many of these issues are on the path to being resolved; both the HITECH ACT and Meaningful Use Programs have accelerated adoption of EHR systems across the US and is now continuing to incentivise sites to store more clinically relevant data in their EHR Systems. As an example the ONC Common Clinical Dataset is mandating the availability of many of the core domains of interest to the clinical trials industry. However, the solution to the lack of availability of a common exchange format has not been completely resolved; the HL7 Continuity of Care Document (CCD) is a mature standard, but is not used consistently across implementations or geographies.
Potential Technological Solution(s)
Review the HL7 FHIR standard as a basis for future data integrations between Research Systems and Healthcare systems. It will include looking at the existing FHIR based Research standards (including DAF and SDC).
Other Initiatives that can be leveraged/merged
CDISC EHR to CDASH (E2C) Group
- The CDISC E2C group is the mechanics or the 'how'; it is predicated on the fact that there is a need for a migration of data from the EHR system to the research system; it has broadened from it's original scope to incorporate the view of the EHR system being the CCD (per RDF) to include FHIR. I think this is an important evolution, there is much more enthusiasm around FHIR than I've seen around any technological healthcare standard - ever!
- The PhUSE group is the why - it's taking a look at the standard (and the initiatives) around FHIR and applying them to a few scenarios within the research domain. The expected output will be a white paper with the conclusions of the group about how FHIR will (or will not, perhaps) change clinical research in the future. There will be subsequent projects that will dig into any of the use cases identified to show Pharma and other stakeholders the utility - these will utilise the rules and mappings put together in the E2C project.
- Follow on from the PhUSE Semantic Technologies keyCRF Project
- HL7 have been publishing a new draft standard; Fast Healthcare Interoperability Resources (FHIR). This is an entirely different paradigm, built around Resources and using REST as a architectural style. This resource-based view of healthcare data is much more aligned to the research 'view' of data; we consider studies using CRFs, typically aligned with clinical data element domains - which some might describe as clinical data resources. The alignment between the terminologies is a known problem and groups such as the CDISC EHR to CDASH (E2C) are looking to come up with a shared semantic layer between the data standards (as well as work previously done with the CDISC BRIDG model).
- The publication of the Federal Notice in 2015 and the updated Draft Guidance in 2016 have shown the FDA's commitment to increased adoption of eSource in Clinical Research. In particular, the Draft Guidance showed a renewed focus on bi-directional integration of data between Research and Healthcare systems; something that would not easily be achieved given current technology and processes. FHIR offers the possibility of making this a reality.
- Identify gaps and opportunities between the Healthcare and Research realms with a view to providing a shared vocabulary and platform to enable better collaboration.
- Prepare a white paper for Biopharmaceutical Companies providing sample use cases for where the adoption of FHIR will improve the experience of Sponsors, Technology Firms and most importantly sites and site users.
Deliverables and Dates:
|Review FHIR Specification for applicability as a Research Standard||Through June 2017|
|Identify 3-6 use cases which illustrate how FHIR Resources can enable effective data sharing between Healthcare and Research Systems||CSS March 2017 - August 2017|
|Prepare white paper for presentation at Annual PHUSE Conference||October 08 - 11 2017|
Last revision by Laurenwhite,01/15/2020