Stream: terminology
Topic: ICD-11 intensional value set
Giorgio Cangioli (Feb 12 2021 at 17:52):
Due to my little knowledge of ICD-11 I have some doubts on how it should be defined some intensional value sets.. I mean
Giorgio Cangioli (Feb 12 2021 at 17:52):
If I should specify descendent-of "XM3KV2 Bacterial vaccines" (Foundation URI : http://id.who.int/icd/entity/1448162632) should I use the following expression ?
"system": "http://id.who.int/icd11/mms",
"filter": [
{
"property": "concept",
"op": "descendent-of",
"value": "XM3KV2"
}
Giorgio Cangioli (Feb 12 2021 at 17:53):
If this is the right approach, how to express descendent-of "vaccines" (Foundation URI : http://id.who.int/icd/entity/164949870) ?
"system": "http://id.who.int/icd11/foundation",
"filter": [
{
"property": "concept",
"op": "descendent-of",
"value": "164949870"
}
Giorgio Cangioli (Feb 12 2021 at 17:55):
in case http://id.who.int/icd11/mms#XM3KV2 will belong to that value set ?
Michael Lawley (Feb 13 2021 at 08:27):
As I understand things, XM3KV2
is a code in the CodeSystem http://id.who.int/icd11/mms
(ie ICD11 MMS) while 1448162632
is a code in the different CodeSystem http://id.who.int/icd11/foundation
(ie ICD 11 Foundation).
So, the second definition can only contain codes from Foundation and thus will not contain the MMS code.
Note also, descendent-of
excludes the code itself, so 164949870
will not be a member either. If you want to include the referenced code then you should use the op is-a
(see https://www.hl7.org/fhir/codesystem-filter-operator.html#4.3.14.416.2)
Giorgio Cangioli (Feb 13 2021 at 08:58):
As I wrote my knowledge of ICD-11 is quite low, so what I understood is that the ICD concept XM3KV2 Bacterial vaccines is also identified by the foundation identifier 1448162632 (please correct me if I'm wrong).
Thus I was wondering if filtering for all the descendants of 1448162632 this automatically includes for example also the XM29K4 Cholera vaccines (Foundation URI :http://id.who.int/icd/entity/1471383174) or not ?
Giorgio Cangioli (Feb 13 2021 at 08:59):
I'm asking that because the "vaccines" concept has an identifier in the foundation (http://id.who.int/icd/entity/164949870), but I've been not able to find a code under the MMS linearization, so I don't know how to say all the descendants of this concept using a http://id.who.int/icd11/mms code... unless making an union of all the descendants of
XM3KV2 Bacterial vaccines ; XM61M7 Viral vaccines ; XM4W25 Combination vaccines ; XM7C66 Bacterial and viral vaccines, combined ....
Robert McClure (Feb 13 2021 at 15:51):
I am also not well versed in ICD-11 but my understanding is the ICD foundation is a multidimensional collection of all ICD entities
meaning all ICD-11 concepts exist in the foundation. That is why any ICD-11 concept has a "entity" identifier of the structure http://id.who.int/icd/entity/xxx. The MMS is a subset focused on the primary use case for ICD - Morbidity and Mortality. If you want all ICD-11 Vaccines you should be able to construct a FHIR compose starting at http://id.who.int/icd/entity/164949870. I also think you can identify the system for any of these using the ICD-11 foundation system identifier.
Michael Lawley (Feb 13 2021 at 22:34):
My knowledge of ICD-11 is also quite low :). In so far as there is a map between (a subset of) Foundation codes and MMS codes, then I think you could find all MMS vaccine codes by getting the descendants of http://id.who.int/icd/entity/164949870
then finding all the corresponding MMS codes via said map.
However, there's no direct way to do this in a ValueSet definition, or even with the FHIR $translate
operation, since it only works on sngle codes.
Manisha Mantri (Feb 18 2021 at 06:53):
I am also in the learning phase but what I understand is that all the descendants of http://id.who.int/icd/entity/164949870 may not be mapped to ICD-MMS. When one wants to look at ICD as terminology then you should always refer to Foundation, whereas if one wants to specifically look for reporting codes (as a coding system) then refer to ICD-11 MMS. So in this case the expansion will only work for all descendants of http://id.who.int/icd/entity/164949870.
Peter Jordan (Feb 19 2021 at 02:04):
Does anyone out there have, or know of, a FHIR Server that returns a Code System resource where the value of the url search parameter in the request is...
http://id.who.int/icd11/mms
or http://id.who.int/icd11/foundation
?
It would be helpful to see an example.
Grahame Grieve (Feb 24 2021 at 02:18):
MMS codes are not foundation codes, nor a subset of them.
Grahame Grieve (Feb 24 2021 at 02:18):
They are a short cut for a grammatical expression in the foundation codes
Robert McClure (Feb 25 2021 at 03:59):
@Grahame Grieve that is a pretty odd way of characterizing MMS. Can you provide a reference that confirms that perspective? MY understanding, which might align with yours, is that the MMS is essentially the linearization of the foundation concepts that are to be used as "ICD-11" for WHO reporting. As such, each MMS code gets an MMS letter-number code familiar to ICD lovers. I'd like to understand the "expression" part via a link.
Grahame Grieve (Feb 25 2021 at 04:11):
hmm as I understand it "linearization" means that there's a formation definition of the MMS term in foundation concepts, and that's what I meant by expression
Robert McClure (Feb 25 2021 at 04:13):
Perhaps, which is why I'd like to see the algorithm or expression used to create the formal linearization. It may exist, I'd like to see it.
Grahame Grieve (Feb 25 2021 at 04:23):
That'd be a question for Robert, I think
Michael Lawley (Feb 25 2021 at 04:33):
The linearization includes its own (single parented) hierarchy, so that must also be encoded somehow?!?
Robert McClure (Mar 01 2021 at 18:56):
The MMS code certainly identifies that the foundation concept is in the linearization but not sure how to identify the specific MMS-based relationships. We need an 11 expert to chime in.
Carol Macumber (Mar 03 2021 at 15:51):
I was just on a call with WHO and have a contact willing to work with us on the information HTA needs, including the code system URLs. I believe he would be willing to talk you through how to identify the MMS based relationships, at least in the context of their API.
Last updated: Apr 12 2022 at 19:14 UTC