Uploaded image for project: 'FHIR Specification Feedback'
  1. FHIR Specification Feedback
  2. FHIR-27809

Some constraints on definition resources are not clearly stated

    XMLWordPrintableJSON

Details

    • Icon: Change Request Change Request
    • Resolution: Not Persuasive
    • Icon: Medium Medium
    • FHIR Core (FHIR)
    • R4
    • FHIR Infrastructure
    • MeasureReport
      QuestionnaireResponse
    • Hide

      First, the formal rules are dependent on whether a Questionnaire or Measure are actually specified (there are use-cases where they wouldn't be).  Second, it's not a given they'll be resolvable.  Both of these would add complexity to the FHIRPath.  Finally - particularly in the case of Questionnaire, the number and complexity of the FHIRPath assertions would be significant.  Also, there's ongoing development introducing extensions, etc. that are also relevant in the process.  In general, it's not clear that the juice is worth the squeeze.

      If the submitter really feels this is viable, they can define the set of FHIRPaths they deem to be sufficient/desirable and we can re-evaluate

      Show
      First, the formal rules are dependent on whether a Questionnaire or Measure are actually specified (there are use-cases where they wouldn't be).  Second, it's not a given they'll be resolvable.  Both of these would add complexity to the FHIRPath.  Finally - particularly in the case of Questionnaire, the number and complexity of the FHIRPath assertions would be significant.  Also, there's ongoing development introducing extensions, etc. that are also relevant in the process.  In general, it's not clear that the juice is worth the squeeze. If the submitter really feels this is viable, they can define the set of FHIRPaths they deem to be sufficient/desirable and we can re-evaluate
    • Rick Geimer / Gino Canessa: 13-0-2

    Description

      Resources that have definitions (e.g., MeasureReport -> Measure, QuestionnaireResponse -> Questionnaire), have constraints applied to them which are not clearly stated in FHIR.  These are effectively invariants between the resource (e.g., MeasureReport or QuestionnaireResponse) and its definition (e.g., Measure, Questionnaire).  We’d like to see these constraints expressed computationally in the FHIR specification, rather than just in the validation code.  We believe that the existing mechanism supporting invariants with FhirPath expressions is sufficient to represent these constraints and would recommend they be incorporated into a new section.

      Attachments

        Activity

          People

            Unassigned Unassigned
            david.pyke David Pyke
            David Pyke, John Moehrke
            Watchers:
            4 Start watching this issue

            Dates

              Created:
              Updated:
              Resolved: