Stoking the Flames: Virtual Workshop on Advancing the Use of FHIR in Research Summary

Stoking the Flames: Virtual Workshop on Advancing the Use of Fast Healthcare Interoperability Resources (FHIR®) in Research brought together a select group of leaders in data science and research from across the federal agencies to develop a framework for increasing the use of FHIR for research. Since the Office of the National Coordinator’s (ONC) Final Rule was released, driving vendors to adopt FHIR as a way to provide patients with access to their healthcare data, government agencies including the National Institutes of Health (NIH) have begun efforts to support the use of FHIR. These efforts included initiatives and use cases which leverage the promise of FHIR’s open access to support research and engage patients in the research process.

The workshop had three main sessions, all focused on the use of FHIR at various federal agencies and how these federal agencies could support the use of FHIR for clinical research. Key themes that carried through the discussions were 1) the interplay needed between policy and technical advances such as FHIR, 2) the opportunities for FHIR to expand the sources of data that can be integrated into the larger ‘system of care’ to support both clinical care and clinical research, 3) the opportunity that FHIR presents to increase data reuse across both clinical care and research settings, and 4) how FHIR is enabling patient access to their own clinical data and how to support this ‘emancipation’ of clinical data while protecting patients’ rights and data security.

A number of outcomes and next steps emerged from the workshop’s discussions:

• NIH should continue to direct investigators to standards and increase community awareness of these standards, and they should continue to encourage broad use of FHIR.
• Agencies should create relationships between FHIR developers and clinical researchers in order to create new research and educational opportunities.
• Agencies need to build sufficient FHIR expertise in order to accurately evaluate incoming proposals, and work to create a pool of awardee expertise as well. 
• Funding opportunities need to include smaller institutions, smaller businesses, and underserved communities.
• A database or registry of the current FHIR efforts across agencies would be helpful.
• There is a need for Implementation Guides that discuss linking patient data to important agency data sources, such as databases on birth, mortality, and public health.
• Mapping between various existing common data models (e.g. OMOP) and FHIR was a common area of activity, and there is a clear need to automate mappings.
• We should identify the key “communities of practice” across agencies that will share resources, lessons learned, and working sessions (e.g. connect-a-thons).