Stream: terminology
Topic: ICD-10 ValueSet of all codes
Michael Lawley (Oct 20 2020 at 05:45):
https://www.hl7.org/fhir/icd.html does not provide any guidance on a value for CodeSystem.valueset for ICD-10-[x]. ie. what is the URI for the ValueSet that contains all ICD-10-[x] codes? Should it be http://hl7.org/fhir/sid/icd-10-[x]/vs
Rob Hausam (Oct 20 2020 at 05:51):
I assume that's as reasonable as anything, and we should say something about it. Would be good to have @Grahame Grieve's opinion.
Grahame Grieve (Oct 20 2020 at 05:59):
I have no particular opinion except that we should define a URL
Peter Jordan (Oct 20 2020 at 06:42):
Do we have an official CodeSystem URI for ICD-10-[x]? It's possibly not worth deciding on the ValueSet URI until/unless we have one.
Jim Steel (Oct 20 2020 at 06:43):
Mostly http://hl7.org/fhir/icd.html (which speecifies http://hl7.org/fhir/sid/icd-10 and
http://hl7.org/fhir/sid/icd-10-[x] (see below) (where the below specifies the suffixes for -gm, -nl and -cm. I assume the fact its in the spec means it is (they are) official?
Peter Jordan (Oct 20 2020 at 07:08):
I don't think that those are 'official' ICD-10 -[x]URIs and I know that @Reuben Daniels was attempting to obtain one for ICD-10-AM, at least.
Grahame Grieve (Oct 20 2020 at 08:45):
those ones are official. But the extrapolation is not official
Lin Zhang (Oct 20 2020 at 10:37):
Does WHO officially recognize ICD-x' urls?
Peter Jordan (Oct 20 2020 at 19:43):
Grahame Grieve said:
those ones are official. But the extrapolation is not official
Seem strange to have an HL7, and FHIR -specific, URI for an external Code System.
Vassil Peytchev (Oct 20 2020 at 20:10):
Why is it strange? It's an identifier, to be used in the context of HL7 FHIR, isn't it?
Jim Steel (Oct 20 2020 at 21:28):
Hard to say. We have plenty of people using the FHIR terminology service outside of the FHIR context (with system urls embedded in non-FHIR, non-HL7 data structures)
Peter Jordan (Oct 20 2020 at 22:24):
Possibly relevant is the ICD-11 API documentation, which uses a base URI of http://id.who.int/icd/.
Grahame Grieve (Oct 20 2020 at 23:27):
for ICD-11, we are using a URI we agreed with WHO '=
Lin Zhang (Oct 21 2020 at 00:13):
Our national/regional variants of ICD-10 and ICD-9-CM Vol3 are lack of good versioning. Like a nightmare.
Bryn Rhodes (Jan 24 2021 at 20:00):
What is the URI for ICD-11? I don't see it in the currently published terminology.hl7.org, nor on current FHIR build?
Michael Lawley (Jan 24 2021 at 20:04):
Which ICD-11? The MMS or foundation ?
Bryn Rhodes (Jan 24 2021 at 20:11):
The particular code I'm looking at right now appears to be in both, though my guess is they mean the Foundation one: http://id.who.int/icd/entity/1851857465
Robert McClure (Jan 25 2021 at 04:35):
@Carol Macumber @Julie James We need ICD-11
Bryn Rhodes (Jan 26 2021 at 01:17):
FWIW, I used http://hl7.org/fhir/sid/icd-11 and the tooling seemed happy with it. Grahame's comment above indicates there is a URI that was agreed to with WHO, but it doesn't seem to be documented anywhere I can find.
Lin Zhang (Jan 26 2021 at 02:32):
Could we have/permit multiple synonymous URIs just like OIDs?
Michael Lawley (Jan 27 2021 at 04:25):
I think that (synonymous URIs) would be a terrible idea in general and to be avoided if at all possible.
Michael Lawley (Jan 27 2021 at 04:27):
The URI I have previously (March 2020) seen @Grahame Grieve use is http://id.who.int/icd11/mms
Lin Zhang (Jan 27 2021 at 06:52):
I agree. But if there is no standard group or gov agency to do this ...
Lin Zhang (Jan 27 2021 at 06:58):
We have been facing such an agonizing situation in the HIT domain.
Lloyd McKenzie (Jan 27 2021 at 14:56):
HTA has the authority to do this in HL7 and they've defined a formal process. Whether they'll be able to keep up with the demand, we'll have to see.
Robert McClure (Jan 27 2021 at 16:03):
No one likes multiple identifiers for the same thing, but they occur in the wild due to siloed activities. We have lots when considering different identifier systems but unfortunately we also have this occur within a single identifier system. NamingSystem is how we deal with this.
Lloyd McKenzie (Jan 28 2021 at 00:32):
NamingSystem defines the one 'conformant' identifier for that thing. And once HL7 establishes it, we promise not to change it. Change creates far too much pain for the implementer community.
Lloyd McKenzie (Jan 28 2021 at 00:32):
(The one conformant identifier typically being different for v2, CDA and FHIR)
Lin Zhang (Jan 28 2021 at 04:34):
I think HL7 should leverage a kind-of "synonymous identifier" mechanism to alleviate/resolve this pain.
Carol Macumber (Jan 28 2021 at 15:27):
The request for HTA to start it's process for ICD-11 has been noted and we're working on it. Per the HTA process, the temporary URL can be assigned as follows: http://terminology.hl7.org/temporary-uri/codeSystem/ICD11MMS
As both Rob and Lloyd note, part of the process is to use NamingSystem to record the known identifiers for the corresponding CodeSystem resource at terminology.hl7.org
@Julie James @Reuben Daniels
The HTA page has been started here:
https://confluence.hl7.org/pages/viewpage.action?pageId=97481986
To @Michael Lawley 's point, I have made this page specific to the ICD-11 MMS. If it is the Foundation Component that we are aiming to reference, additional conversation will need to be had around representing the ICD Linearizations (aka tabular lists) as part of the Foundation, akin but not exactly the same as SNOMED and it's editions.
Lloyd McKenzie (Jan 28 2021 at 21:35):
There is no way to avoid pain if you allow more than one thing. It requires every application to implement a translation function on all of their data, including when performing search. It adds tons of technical complexity and risk. Much better to simply declare a single official string and do everything in our power to ensure that it never changes, no matter what else happens in the world.
Grahame Grieve (Feb 04 2021 at 20:55):
http://id.who.int/icd11/mms is what WHO decided on. There's not much for us to say on that matter.
Grahame Grieve (Feb 04 2021 at 20:56):
thats for MMS codes. http://id.who.int/icd11/foundation is for the foundation codes
Grahame Grieve (Feb 04 2021 at 20:56):
@Carol Macumber
Carol Macumber (Feb 04 2021 at 21:02):
Sounds fine to me. I have a note out to Robert Jakob. I'll reiterate the URLs you have above as what has been established.
I'll update the HTA page.
Grahame Grieve (Feb 04 2021 at 21:05):
thanks. Good to get Robert to confirm.
Carol Macumber (Mar 10 2021 at 17:05):
Can Çelik (Technical Officer Classification, Terminologies & Standards Health @ WHO) responded to our request to verify the URIs for ICD-11, this is his response:
"URI for the foundation is : http://id.who.int/icd/entity
URI for MMS is : http://id.who.int/icd/release/11/mms (please note the correction from your version)"
@Grahame Grieve do you have something I can send back to him stating that WHO had agreed to the ones you state above
MMS - http://id.who.int/icd11/mms
Foundation - http://id.who.int/icd11/foundation
Bryn Rhodes (Sep 08 2021 at 18:56):
Any update on this? @Carol Macumber @Grahame Grieve I'm trying to figure out what URI to use, based on searches here it seems that http://id.who.int/icd11/mss is the one to use, but I'm using that and the publisher says it doesn't know what code system that is. The HTA page says "http://id.who.int/icd/release/11/mms" (https://confluence.hl7.org/display/TA/International+Classification+of+Diseases+11th+Revision)
Paul Denning (Sep 09 2021 at 18:22):
http://hl7.org/fhir/uv/shc-vaccination/2021Sep/ValueSet-vaccine-icd-11.html is using http://id.who.int/icd/release/11/mms
Bryn Rhodes (Sep 09 2021 at 21:44):
Thanks @Paul Denning!
Last updated: Apr 12 2022 at 19:14 UTC
