Stream: committers
Topic: CodeableReference
Grahame Grieve (Dec 12 2019 at 10:51):
I have added a new type to the R5 build: CodeableReference. You can see it here: http://build.fhir.org/references.html#codeablereference. And you can see it in action here: http://build.fhir.org/medicationrequest-definitions.html#MedicationRequest.reason
For this type, you can specify both target resource types and a value set binding.
MnM proposes to replace all elements that are a choice between CodeableConcept and Reference (either explicitly, or implicity, such as reasonCode / reasonReference) with CodeableReference in R3. I will do this between Xmas and New Year if committers / committees don't object (will be heaps of work).
Lloyd McKenzie (Dec 12 2019 at 12:44):
What's the process if a particular profile wants to constrain down to only one of the two options? They have to create a distinct profile on CodeableReference - they can't just substitute Reference or CodeableConcept as their type, correct?
Grahame Grieve (Dec 12 2019 at 18:52):
right. though you don't have to create a distinct profile - you can just walk into it
Richard Townley-O'Neill (Dec 13 2019 at 01:18):
It would be nice if the structure diagram showed type as CodeableReference(Condition | Observation), the way it appears in the detailed description
Grahame Grieve (Dec 13 2019 at 01:20):
agree
Eric Haas (Dec 13 2019 at 02:31):
< right. though you don't have to create a distinct profile - you can just walk into it
What does ‘walking into it’ mean exactly?
Grahame Grieve (Dec 13 2019 at 02:36):
what we already do
medication.reason.concept 0..0
Rob Hausam (Dec 19 2019 at 19:21):
It looks like the Constraints text here was probably copied originally from Reference and wasn't quite updated to match the final form of the datatype:
At least one of reference, identifier and coding SHALL be present (unless an extension is provided).
I presume it should be:
At least one of reference or concept SHALL be present (unless an extension is provided).
@Grahame Grieve?
Vassil Peytchev (Dec 19 2019 at 19:39):
What is meant is reference.reference, reference.identifier, or concept.coding
Vassil Peytchev (Dec 19 2019 at 19:42):
There is a problem with the link on concept - it points to references definition, and the link is non-existent
Grahame Grieve (Dec 20 2019 at 01:50):
thanks. Actually, that is not a constraint anymore - your rewrite is correct, but doesn't say anything not already implicit. I just missed removing that fragment.
Last updated: Apr 12 2022 at 19:14 UTC
