Stream: IG creation
Topic: Publisher slicing reason
Brian Reinhold (Feb 10 2021 at 12:21):
I am getting these remarks from the IG Publisher
This element does not match any known slicedefined_in_the_profilehttp://hl7.org/fhir/uv/phd/StructureDefinition/PhdCompoundNumericObservation
Observation.component[0]: discriminator = true and (code memberOf 'http://hl7.org/fhir/uv/phd/ValueSet/CodeableConcept11073MDC')
,
Observation.component[0]: discriminator = true and (code memberOf 'http://hl7.org/fhir/uv/phd/ValueSet/Quantity11073MDC')
,
Observation.component[0]: discriminator = true and (code memberOf 'http://hl7.org/fhir/uv/phd/ValueSet/Quantity11073MDC')
,
Observation.component[0]: discriminator = true and (code memberOf 'http://hl7.org/fhir/uv/phd/ValueSet/MDCnotObject')
,
Observation.component[0]: discriminator = true and (code memberOf 'http://hl7.org/fhir/uv/phd/ValueSet/Quantity11073MDC')
I am not sure why. Is it due to the fact that every one of the above value sets references the system
<system value="urn:iso:std:iso:11073:10101"/>
which the terminology server does not understand?
Lloyd McKenzie (Feb 10 2021 at 15:09):
That would do it
Brian Reinhold (Feb 10 2021 at 15:18):
Lloyd McKenzie said:
That would do it
Okay, then there is not much I can do about it until that issue gets fixed.
Lloyd McKenzie (Feb 10 2021 at 15:37):
It's unlikely to ever be fixed. The validator can't differentiate because there's no way it can tell which slice is in play
John Moehrke (Feb 10 2021 at 16:57):
This is really unfortunate. These system oids have been around for years, the terminilogy server should understand them. I get that we want to encourage NEW data to use the URL methods, but these oids will be in alot of historic information.
Lloyd McKenzie (Feb 10 2021 at 18:43):
I didn't mean to say we couldn't get support for these code systems in the terminology server - that's absolutely viable. What I was saying is that the general issue of not being able to differentiate slices where slicing is based on value sets drawing from value sets that uses from the same code system isn't something we can technically solve.
ryan moehrke (Feb 10 2021 at 19:43):
If you have a proper valueset expansion then there is no issue right? It's just making sure that your expansion is there and up to date that could pose problems
Grahame Grieve (Feb 10 2021 at 20:37):
The validator can't differentiate because there's no way it can tell which slice is in play
why not?
Grahame Grieve (Feb 10 2021 at 20:37):
These system oids have been around for years
Nothing to do with OIDs
Lloyd McKenzie (Feb 10 2021 at 21:10):
If the validator doesn't know what codes are in the value sets, it can't tell which value set a given instance is part of.
Grahame Grieve (Feb 10 2021 at 22:01):
once the code system is supported, it will
Last updated: Apr 12 2022 at 19:14 UTC
