Stream: conformance
Topic: data-absent-reason for choices (e.g. effective[x])
Giorgio Cangioli (Nov 02 2018 at 08:07):
Have you any suggestion about how to indicate for a choice (e.g. Observation.effective[x]) the reason for absence ?
Giorgio Cangioli (Nov 02 2018 at 08:11):
It works for single elements (e.g. effectiveDateTime) but the IG publisher fails when this extension is applied to a choices..
Michel Rutten (Nov 02 2018 at 10:44):
Seems like a tooling limitation. FHIR allows extensions everywhere.
Giorgio Cangioli (Nov 02 2018 at 12:24):
Do you suggest I should move this thread to the IG creation stream ?
Lloyd McKenzie (Nov 02 2018 at 14:49):
You always have to pick a concrete type in order to extend. You can't extend "choice" directly because it doesn't manifest as a type. So just pick one (typically the simplest one) and extend that with dataAbsentReason.
Giorgio Cangioli (Nov 02 2018 at 16:04):
Yes but if I profile the effective[x] by using effectiveDateTime (with the data-absent-reason extension); I'm no more allowed to instantiate this choice with - for example - an effectivePeriod. .. Or I misunderstood your suggestion...
Giorgio Cangioli (Nov 02 2018 at 16:08):
What I'd like to say is I'd expect you valorize the effective element as DateTime or period, but if you have not, please let me know why the value is absent
Giorgio Cangioli (Nov 02 2018 at 16:11):
and it seems that I'm not allowed either to slice the effective[x] and add the extension to each slice... (effective[x] has max=1)
Lloyd McKenzie (Nov 02 2018 at 16:28):
You must pick one of those two. In one of them, you can also allow for a reason why the data is absent. You can't indicate data being absent in a type-independent way
Last updated: Apr 12 2022 at 19:14 UTC
