Logistics:


  • Overview for participants of other tracks: Friday at 11AM on main GoToMeeting


Submitting WG/Project/Implementer Group

FHIR-I / Argonaut Bulk Data Workgroup


Justification and Objectives

Argonaut has selected Bulk Data as a 2020 project and work has begun on v1.2 of the IG. Participants in this track will prototype and test some of these proposed improvements.


FHIR Bulk Data Resources


This track will use FHIR R4


Track Leads

  • Dan Gottlieb
  • Josh Mandel


Participants

Sign up sheet


System Roles

Data Provider: Bulk Data server implementing the draft file export enhancements to v1.0 of the IG.

Data Consumer: Bulk Data client with capability to retrieve binary and text files following the draft file export enhancements to v1.0 of the IG.


Scenarios - based on current draft of Bulk Data IG

Scenario 1: Bulk data export with retrieval of referenced files on an open endpoint

  1. Data Consumer issues the following request

    GET [base]/Patient/$export?_outputFormat=ndjson
Accept: application/fhir+json
Prefer: respond-async
  2. Data Provider and Data Consumer follow the flow outlined in the Bulk Data Export IG to generate and retrieve a dataset. This dataset must include DocumentReference resources with a populated attachment.url element. The requiresAccessToken property in the Bulk Data manifest should be set to false.
  3. Data Consumer retrieves binary files and text files referenced from DocumentReference.attachment.url


Scenario 2: Bulk data export with retrieval of referenced files on a protected endpoint

  1. Data Consumer registers with Data Provider, per SMART Backend Services Authorization
  2. Data Consumer obtains an access token, per SMART Backend Services Authorization

  3. Data Consumer issues the following request with the access token

    GET [base]/Patient/$export?_outputFormat=ndjson
Accept: application/fhir+json
Prefer: respond-async
  4. Data Provider and Data Consumer follow the flow outlined in the Bulk Data Export IG to generate and retrieve a dataset. This dataset must include DocumentReference resources with a populated attachment.url element. The requiresAccessToken property in the Bulk Data manifest must be set to true.
  5. Data Consumer attempts to retrieve files referenced from DocumentReference.attachment.url without an access token and fails
  6. Data Consumer retrieves binary files and text files referenced from DocumentReference.attachment.url with an access token and succeeds


Bonus Scenarios:

  1. Prototype group membership enhancements outlined on Github (also see notes here)
  2. Prototype deleted resource enhancements outlined on Github (also see notes here)


Security and Privacy Considerations

SMART Backend Services Authorization will be required to participate in Scenario 2



  • No labels