Stream: terminology / utg
Topic: Setting compose.inactive =false still displays code
Abdullah Rafiqi (Feb 04 2022 at 15:28):
I've updated the FHIR MeasureType Valueset in my local build to not display inactive codes:
<compose>
<inactive value="false"/>
But when running the publisher, the code still appears ("composite") even though its noted as being 'retired' within the codesystem
Should the code still appear in the valueset?
Grahame Grieve (Feb 04 2022 at 19:39):
I don't believe that 'retired' is the same as 'inactive'
Abdullah Rafiqi (Feb 04 2022 at 19:57):
I found the following within the codesystem concept status :
HL7 uses "active" (and considers this the default status), and deprecated and retired (= inactive)
Rob Hausam (Feb 04 2022 at 20:26):
I agree that "retired" and "inactive" are treated as (essentially) synonymous - so a "retired" code should be considered as "inactive". It sounds like that's not happening in this case - and if so, I think that's a problem? But I think the consensus in Vocab generally has been that "deprecated" codes are still considered to be "active" (just discouraged from any new use, and are expected to be retired in the future). As far as I know the deprecation policy group that's meeting (had a meeting earlier today) hasn't changed that - but we could verify. @Ted Klein?
Ted Klein (Feb 04 2022 at 20:37):
for many many years HL7 has considered the notion of 'retired' to be synonymous with 'inactive' for vocabulary items. This was ratified with the notion that 'deprecated' is NOT retired or inactive, it is a usage guidance around 'active'. Because of differences in interpretation of these human language words, I am a proponent of making the controls for whether codes having certain values as concept properties should be included in value set expansions to be very explicit so the we do not get differences in server behaviors around the globe. For sure if there is an indication in the CLD to include only active codes in an expansion, then absolutely ones that are not active (retired or any of the other states that are not active ) should absolutely not be included in the expansion I would think.
Abdullah Rafiqi (Feb 04 2022 at 20:55):
So in this instance is this a publisher issue that the retired code is still being displayed when compose.inactive=False ?
Ted Klein (Feb 04 2022 at 21:22):
I should think so. @Grahame Grieve ?
Grahame Grieve (Feb 20 2022 at 11:14):
I guess so, yes
Abdullah Rafiqi (Feb 23 2022 at 13:24):
@Grahame Grieve Would a Jira ticket submission be needed to address the issue?
Grahame Grieve (Feb 25 2022 at 02:00):
fixed next release
Abdullah Rafiqi (Feb 28 2022 at 20:08):
@Grahame Grieve The current valueset version for the FHIR MeasureType Valueset = 0.2.0
Would the change to compose.inactive=false result in the Valueset version= 0.2.1 or 1.0.0?
Grahame Grieve (Feb 28 2022 at 20:21):
I do not know. That's a UTG policy question. @Jessica Bota I guess
Jessica Bota (Feb 28 2022 at 20:28):
@Grahame Grieve , Ted was concerned about the versioning because you initially set the versions for the FHIR value sets. Will changing them following the UTG policy affect anything in the FHIR space?
Grahame Grieve (Feb 28 2022 at 23:08):
no. UTG is in control now, as long as semver is followed it's all good, I think
Jessica Bota (Mar 01 2022 at 16:03):
Thanks Grahame. @Abdullah Rafiqi , you can update the version to 1.0.0.
Last updated: Apr 12 2022 at 19:14 UTC
