Stream: terminology
Topic: extensible / other
Grahame Grieve (Jul 10 2018 at 22:13):
hey vocab people
Grahame Grieve (Jul 10 2018 at 22:13):
https://hl7.org/fhir/2018Jan/v2/0131/index.html
Grahame Grieve (Jul 10 2018 at 22:15):
this is extensibly bound, and has a v2 'other' code - that doesn't make sense. Some questions:
- the FHIR tooling should not allow an extensible value set binding if the value set contains NullFlavor.OTH, right?
- do we have v2 mappings for code like this v3 nullFlavor?
- are there any other v3 nullfflavors that should not be in an extensible binding?
Brian Postlethwaite (Jul 11 2018 at 06:34):
Is that something we should create as a build warning?
Brian Postlethwaite (Jul 11 2018 at 06:34):
Or maybe even a structure definition invariant warning/best practice?
Michael Lawley (Jul 11 2018 at 06:40):
I don't follow - what if the extensions add detail (ie are specialisations of codes in the base valueset) or is that also banned in an extensible binding? It seems to depend on how you read the phrase "not covered by"
Grahame Grieve (Jul 11 2018 at 11:11):
if I can identify the codes, then I can make warnings etc.
Grahame Grieve (Jul 11 2018 at 11:12):
@Michael Lawley if you put 'other' in the value set, then there's no point making it extensible, because there's never any code that's not valid
Robert McClure (Jul 11 2018 at 14:30):
@Michael Lawley indeed extensible is completely governed by "not covered by." What do you mean by "extension"?
@Grahame Grieve I agree that having any form of "Other" or NOS is not appropriate for Extensible bound value sets. Using this would mean all codes not explicitly included in the value set are "covered by" Other. Even more confusing would be a concept scoped by one of the other member concepts because one could argue that such a concept could be mapped to either the scoping concept or Other.
Grahame Grieve (Jul 11 2018 at 22:38):
right. so I'm wondering whether I can catch 'other' in extensible value set in the tooling
Last updated: Apr 12 2022 at 19:14 UTC
