Conformance Weekly Meeting 2017 06 05

Meeting Information

• Date: Monday, June 5, 2016
• Time: 10:00 am, North America Eastern Time (New York, GMT-04:00)
• Phone Number: +1 770-657-9270
• Pass Code: 644843
• Web Meeting: Uber Conference

Proposed Agenda Topics

• Request from Templates for new Project (John Roberts)
• Birth and Fetal Death Reporting IG: Two conformance questions raised by Mead Walker on listserv (Mead Walker)
• Requiring all occurrences within a data type (Craig Newman)
• Review and approve WGM minutes
• Cochair Election
• Data Types Project
  • Any updates?
• Review FHIR Resource QA Tracker
  • DSTU QA Guidelines
  • Conformance has been asked to look at this list and see if anything needs to be added:
    • Google Spreadsheet
• FHIR Conformance Review
  • First proposed topic (looking at Slide 2): Usage & Cardinality
  • Original document FHIR Constraints and Rules

Attendees

• Mead Walker
• Nathan Bunker
• Craig Newman
• Rob Snelick
• John Roberts

Discussion

• Request from Templates for new Project (John Roberts)
  • Are you willing to add a reflection of this language to your M&C? " Conformance Work Group to ensure that the rules for defining constraints in templates are consistent with those for other HL7 artifacts, especially the alignment of expressing constraints and the underlying constraint specification methodology for HL7 version 2 profiles."
    • Yes
    • Templates and FHIR has been stated as a fairly easy transition, but probably not in the way the terms have been defined. Need to go through the exercise and prove how well it maps over.
    • Need to review Conformance Mission and Charter to mention Templates
  • Templates is proposing to write a PSS. The work needs to be done by Templates and/or Conformance.
    • Yes we should do it, Conformance was heading down this path anyways. Probably most appropriate for Conformance to lead the first bullet point. Two distinct things here, but we can throw them into one.
    • Draft PSS: Conformance will draft the PSS. Kai may have already started a draft. At the next call John will bring this up and then send us what we have.
• Birth and Fetal Death Reporting IG: Two conformance questions raised by Mead Walker on listserv (Mead Walker)
  • There needs to be a way for the sending system to assert that it's not providing a value. There is a format today that has a slot for every piece of data.
    • Conformance believes that the right answer is to not send the OBX if there is no value for OBX-5. Although in cases where the value is coded the "unknown" value can be added to the code set. Unknown should only indicate the user indicating unknown and should NOT represent that the system did not collect or receive this data.
  • Second part of the question. Best to be phrased as example: If you are recording whether a birth was at home, then if it was then was it intentional? Should there be a conformance statement to indicate that the second OBX is required when the previous OBX has a certain value. Should the dependent question be made conditional.
    • It's subjective on where to draw the line. It can be defined in the profile and made to be very rigid. Or you could provide the LOINC and the observations and not force the observation.
  • Question #2: Units-of-measure. Read somewhere that if you indicated the amount you had to indicate the units. Currently setting OBX-6 as 1, is this right?
    • Seems that probably need to send something like 1^^UCUM or do we make it RE and then create test cases that verify proper support? Both solutions are valid, but hate to see one group pick one solution and another pick the other. Would like to have further research on this. Put this on the agenda for the next call.