glow-mdsol/fhirton

A Piece of Code illustrating the moving of Data from a FHIR endpoint to an EDC study in Medidata Rave. Put together as part of the FHIR connectathon (Sep 2016)

FHIRTON

This is the code from the HL7 Connectathon September 2016. Feel free to do as you will with it, ;-)

Use Cases

Possible Use cases to develop for the Connectathon.

Study Registration

• FHIR Gateway advertises Clinical Study Resources
• EHR system consumes resources

Study Design Model

Study Design Model to ResearchStudy

• Take an ODM-SDM message and generate a ResearchStudy resource

Study Design Model to PlanDefinition

• Take an ODM-SDM model and generate a PlanDefinition
• Note: PlanDefinition is independent of a ResearchSubject and would be ~ the Protocol per ODM

SDC

• Form
  • Transform Form ODM into Questionnaire Resource
• Form response
  • Process QuestionnaireResponse Resource into ODM ClinicalData

DAF

• DM
  • POST - Transform DAF-Patient resource into a DM record, Push to Rave
  • GET - Transform DM panel to DAF-patient (will probably be non-compliant?)
• VS
  • POST - Transform DAF-VitalSigns resource into a VS record, Push to Rave
  • GET - Extract a VS CRF and transform to DAF-vitalsigns

Demographics

• DAF-Patient
  • Race => us-core (US Realm)
  • Ethnicity
• ~ Patient-clinicalTrials

Research (2017)

• Prepopulating study research data from an EHR
  • Action: Prepopulate eCRFs in an EDC clinical database with data pulled via FHIR from an EHR using loose matching criteria. Multiple versions of a CRF type (e.g. vital signs) can be used to represent different types of clinical studies (e.g. observational vs. regulated).
  • Success Criteria: Data from the FHIR resources are extracted through the FHIR API and mapped to clinical research data element fields on CRFs. Each FHIR resource data element has been successfully imported for display in the CRFs, or documented in the mapping table as not for use in clinical research. The matching criteria is intentionally loose in this step, meaning the data may not be usable as-is and the assessment of data content suitability for research will be performed in step 2.
• Determine the FHIR resource EHR data content and quality assessed against the clinical research requirements
  • Action: evaluate the test data suitability for use in clinical research databases by assessing what actions would be needed to (1) load the data into non-regulated clinical research databases including those supporting observational studies, and (2) load the data into regulated clinical research database that requires CDISC standards alignment.
  • Success Criteria: The resulting table should function as a map highlighting potential data content issues to be addressed in future profile development activities. Summary statistics will be derived to assess the results recorded in the table.