Retrieving Form Data from Epic Using RFD Standard Enhancement
Amy Harris Nordo RN/BSN, CPHQ, LNC, MMCi 2016
Duke Office Of Research Informatics
Supported in part by Duke’s CTSA grant (UL1TR001117)
▪ 3 Integrated hospitals using Epic 2014 (16,513 employees)– Duke University Hospital 957 beds– Duke Raleigh Hospital 186 beds – Duke Regional Hospital 369 beds
• Duke School of Medicine Ranked 8th 2015
• Duke Clinical Research Institute is the largest Academic CRO in the world with over 900 employees
• Duke is the Coordinating Center for PCORI, NIH CollaboratoryGrants
• Our CTSA was renewed in 2013
• 2015 Clinical revenues $2.98B
– 1.35m outpatient visits
• 2014 NIH funding $293.2m
• 2013 Industry funded research $177m
• 1462 investigators, 400+ coordinatorsin site based research
• ~2100 open enrolling IRB studiesduring FY15
• ~300 NEW clinical trial studies/yearopen at Duke University Hospital
• 7,000 simultaneous users of Epic(Epic’s largest big bang go-live)
– Integrated clinical researchmanagement into Epic.
• Duke’s REDCap approximately 4000projects and 4000 users
What can we put in between EHR and EDC that’s not a person to facilitate data migration?
What about RFD?
RFD – What is it?
• Integration profile co-developed by Integrating the Healthcare Enterprise (IHE) and CDISC
• A standard for research data collection from EHRs
• Allows EHR platform to display a data collection form sourced from an EDC platform
• Pre-populates form with data sourced from Continuity of Care Document (CCD)
• CCD contains snapshot of patient chart (XML format)
Value Proposition – Why RFD?
1- Facilitates Data Provenance• Specification dictates archiving/auditing – RFD Form Archiver
• Archive Viewer Web Application… stay tuned
2- Improve Data Quality Minimize transcription errors
3- Time: More efficient use of Clinical Research Coordinator time
Provides the opportunity to change workflow process
4- SecuritySecure, single-point registration of study participants data into EDC database from Epic context
What is it?
Why Integrate REDCap?• Significant Duke & CTSA adoption
• Duke ~4000 active projects; ~4000 users
• Investigator initiated studies – less barriers to adoption
• Integrate Duke enterprise EHR with Duke enterprise EDC
• (
• Enterprise level services, resources, expertise for Duke site-based research
• Epic for Research @ Duke services
• REDCap @ Duke services – eCRF design, Data Management, etc.
• Existing delivery channel for RFD services…
– Consortium Partners
Key
EHR and EDC must be RFD Capable
Architecture – How?
• Epic supports the RFD specification – REDCap DOES NOT support RFD
specification
• Was not on product roadmap for Vanderbilt – feature priority, funding, demand,
etc…
• Early “Proof-of-Concept” in collaboration with Epic – confident we could “make it
work” with REDCap
• Developed middleware – makes REDCap “RFD aware”
• Messaging – Brokers RFD messages between Epic & REDCap
• Data Mapping – CCD data elements to REDCap data elements
• We call it the “RFD-REDCap Adapter”…
Research Use Case
Architecture – Messaging
Duke has Winning Teams on the Court…..
And in Research Application Development
RFD Workflow Demo
Initiate RFD
REDCap Forms Menu
Populated REDCap Form
Populated REDCap Form
Populated REDCap Form
Form Saved to REDCap via RFD
Saved
!
Other study
Specific forms
that perhaps
don’t use the
CCD
Architecture – Archive ViewerData Provenance
• Comprehensive audit trail for RFD workflow
• Track snapshot of CCD
• Track form state transitions
• “Retrieved”, “Pre-populated”, “Submitted”, etc.
• Track parsed CCD data elements
• eCRF data elements populated/CCD redaction
• Mapping rules utilized
Archive Viewer Demo
Value Proposition – Why RFD?
1- Facilitates Data Provenance• Specification dictates archiving/auditing – RFD Form Archiver
• Archive Viewer Web Application… stay tuned
2- Improve Data Quality Minimize transcription errors
3- Time: More efficient use of Clinical Research Coordinator time
Provides the opportunity to change workflow process
4- SecuritySecure, single-point registration of study participants data into EDC database from Epic context
Observational Comparative Effectiveness Study
Evaluation of the effect of RFD functionality on completion of Clinical
Research data collection
Continuing Challenges, Lessons learned, & Next Steps
• Extend data mapping coverage for maximum utility
• Exhaust what is in the standard Epic CCD
• Work with Epic on strategy to extend the current CCD, create a research CCD v2.0, view & parse a subject's CCD archive to find extended history
• The RFD source document does NOT have to be CCD
• Connect with additional EDCs
• Publish Results of Observational Comparative Effectiveness Study
• Multi-Center Trial
In Conclusion:
• Auto-populate data transmission
• Surface forms within EHR
• Redact the CCD
• Audit Trail
• Security
Time, Data Provenance, Quality and
Security