Stream: argonaut
Topic: Race Complex Extension
Jenni Syed (Feb 07 2017 at 19:41):
@Eric Haas Is this extension as documented what we think is final? http://build.fhir.org/ig/Healthedata1/Argo-DSTU2/StructureDefinition-argo-race.html
I ask because I don't think the "sub extension" url fields look like a valid URI
Jenni Syed (Feb 07 2017 at 19:41):
I would have expected it to at least be #ombcategory, for example
Jenni Syed (Feb 07 2017 at 19:41):
But could be wrong :)
Eric Haas (Feb 07 2017 at 20:36):
See this discussion regarding complex extension urls. Does that help?
Jenni Syed (Feb 07 2017 at 20:58):
It does, but it's surprising to me from a spec standpoint now. If I was running URI validation, I don't think those would validate.
Jenni Syed (Feb 07 2017 at 20:58):
But I can test to make sure :)
Eric Haas (Feb 07 2017 at 21:33):
They validate using the FHIR validator. but not sure that is the same thing
Grahame Grieve (Feb 07 2017 at 23:19):
looks all good to me
Jay Lyle (Feb 07 2018 at 13:58):
Patient.argo-race contains extension.valueCoding (0..5) with type Coding, which contains both an OMB code and a display string.
Also contains extension.valueCoding (0..*) for detailed codes, with codes & strings.
And, it requires a separate extension.valueString (1..1).
Assumption: concatenate the values from the codings into the valueString?
Ron Shapiro (Feb 10 2018 at 06:09):
@Jay Lyle the examples seem to indicate that if there is one category, that categories display string becomes the text valueString. If there are multiple categories, then the example just has "Mixed" as the value. Perhaps the spec could be updated to make this more clear on what is expected?
Ron Shapiro (Feb 10 2018 at 06:09):
The CDC code system that is used is a hierarchal code system - see http://build.fhir.org/ig/HL7/US-Core/CodeSystem-cdcrec.html
@Eric Haas , to make it more obvious to implementers how this complex extension works, what do you think of changing the description on this page http://build.fhir.org/ig/HL7/US-Core/ValueSet-detailed-race.html to state something like:
. Include the "level 3" codes from urn:oid:2.16.840.1.113883.6.238 for Race (concept is-a 1000-9)
. Exclude the "level 2" codes as defined in urn:oid:2.16.840.1.113883.6.238 (these are the categories used for http://build.fhir.org/ig/HL7/US-Core/ValueSet-omb-race-category.html )
instead of:
. Include codes from urn:oid:2.16.840.1.113883.6.238 where concept is-a 1000-9
. Exclude these codes as defined in urn:oid:2.16.840.1.113883.6.238
This same logic applies to the ethnicity codes as well.
Eric Haas (Feb 11 2018 at 02:51):
I don't see how that makes it more obvious unless you speak OID
Jay Lyle (Feb 12 2018 at 17:42):
Thanks @Ron Shapiro ; didn't see that example in the body.
Jay Lyle (Feb 15 2018 at 15:37):
Second example: Asian/Filipino. It's not mixed; it's just refined. How should we represent that? (US Core also illustrated 'mixed' but not refined.)
Eric Haas (Feb 15 2018 at 16:38):
assuming you are you talking about what to put into the text field? Argonaut provides no guidance on that because is left up to the implementer what they want to display. I am not aware of any ONC guidance either.
Last updated: Apr 12 2022 at 19:14 UTC
