Stream: genomics
Topic: Variant valueCodeableConcept too constrained
May Terry (Oct 18 2021 at 18:06):
Recent changes in US Core USCoreLabResultObservation (http://hl7.org/fhir/us/core/STU4/StructureDefinition-us-core-observation-lab.html) has added an invariant on value[x] such that for codeableConcepts, us-core-4: SHOULD use SNOMED CT for coded Results.
This broke validation of mCODE examples for CancerGenomicVariant profile which tried to align with the CGWG Genomics Reporting IG whereby the Variant valueCodeableConcept is tightly bound to LOINC Answer List LL1971-2 (required) but also derives from US Core Lab Result Observation.
ASK: would it be possible for CGWG to relax the binding for Variant:Observation.valueCodeableConcept to (preferred) instead?
Q: Should I create a JIRA ticket for this against the CG Reporting IG?
Kevin Power (Oct 18 2021 at 18:27):
Starting with a ticket is always fine. I am not sure what the best answer is, perhaps making it preferred is fine. But it could be that we create our own ValueSet and add the SNOMED terms to that ValueSet (is that an option?)
May Terry (Oct 18 2021 at 18:33):
@Kevin Power - I suppose that would work as well, although we'd have to make sure that the terms are included in the SNOMED Global Patient Dataset so as not to introduce SNOMED licensing issues.
I created a new issue, https://jira.hl7.org/browse/FHIR-34144.
Kevin Power said:
Starting with a ticket is always fine. I am not sure what the best answer is, perhaps making it preferred is fine. But it could be that we create our own ValueSet and add the SNOMED terms to that ValueSet (is that an option?)
Last updated: Apr 12 2022 at 19:14 UTC
