Chair: @Kathleen Connor
Scribe: @Joel Bales
John Moehrke , Co-Chair
Alexander Mense, Co-Chair
Kathleen Connor , Co-Chair
|VA (Book Zurman)||@Trish Williams, Co-Chair||Flinders University|
|@Christopher Shawn, Co-Chair||VA||x||@Mike Davis||VA|
|@David Staggs||VA (Electrosoft)|
|SRS||@Francisco Jauregui||VA (Electrosoft)|
|@Matt Blackmon||Sequoia||HL7 FHIR|
|@Dave Silver||VA (Electrosoft)||x||Kaiser|
|VA (Book Zurman)||@Ken Salyards||SAMHSA|
|@Peter VanLiesdonk||Phillips||@Adam Wong email@example.com||HHS|
|@Mohammad Jafari firstname.lastname@example.org||VA (Book Zurman)|
@Ricky Sahu, @1up.health
|x||Joel Bales||DoD/VA Interagency Program Office||Eric Kettler||DoD/VA IPO|
|x||Carlos Polk||DoD/VA IPO||Gary Dickinson||CentriHealth/UnitedHealth Group|
|Michael Donnelly||Epic||Didi Davis||Sequoia|
|Matthew Rahn||x||Laura Bright|
Decision Link(if not child)
|Invariant from prior week|
Share updates of new invariant - ask question about requiring that on device also
Inside FHIR profile is requirement of organization: who (author). Determined onBehalfof SHALL be present when Provenance.agent.who is a Practitioner. The same approach will be used for device. Brett Marquard will coordinate with Argonaut community for concurrence. See link to Argonaut R4 Implementation Guide.
Argonaut R4 Implementation Guide CI Build (v3.1.0)
Dig a deeper into the use cases we agreed to previously. (Section 2)
Existing use case works, Brett Marquard worked through a different use case to improve the guide. When a provider in their own local data store accepts the problem into their data store, they are accepting a new level of responsibility. When it is carried forward, then the person who accepts it into the EHR is then responsible for the content.
This depends on the type of data (some is inherently reconciled, and other data is not reconciled). For example, lab results don’t change authorship, but allergies do get reconciled (change authorship). Must specify that this is a special case for a limited number of items, which are currently reconciled.
Another example: if you accept a care plan, then accept ownership for the care plan. However, reconciliation is not required for care plans.
Other use cases are related to clinical information reconciliation process.
3 Implementation guidance
3.2: Authorship considerations. Group accepted the proposed language. Kathleen Connor suggested that legal expertise is needed. For now, we can’t do more than Best Practices unless HL7 publishes guidance.
3.3.1: C-CDA Best Practice – top three EHR vendors concurred with the proposed approach. Third bullet: “When generating a C-CDA document, if only a free text provide is available, the organization of the provider SHALL be present. If the organization of the free text provider is unknown, the source of information, or sender, SHALL be recorded as the organization.” This requires input during the ballot.
3.4.1: FHIR-Specific Best Practices – “All systems that support the ONC USCDI SHALL be capability of providing a Provenance Resource that conforms to the Argonaut, or subsequent, US Core Profile. (May be controversial).
Gaps: Appendix – Advanced Provenance Efforts and 3.5: Resources for Advanced Implementers. Mr. Marquard will contact John Moehrke.
3.5 Resources for Advanced Information
Kathleen Connor will add information on these topics:
Volume 2 Provenance (supporting Mike Davis)
Data Segmentation for Privacy (DS4P)
ONC Provenance Work was folded into Gary’s work (CDA R2 Provenance)
Use of Block Chain to store Provenance information. Federated Provenance information has value and was included in the ONC challenge.
Mike Davis: confirmed that links in 3.5 is a useful approach.
Mr. Marquard and Ms. Connor agreed that he will clean up existing content and submit to HL7 to meet the ballot deadline. Then clean-up can continue through Aug 2.
Review a draft of the Basic Provenance Informative Guide and the latest Argonaut Provenance Profile.
4:00 PM eastern time