Stream: terminology
Topic: ICD-10-pcs
Eric Haas (Mar 02 2017 at 20:15):
For the US Core IG we have defined icd-10-pcs as a code system. where would I get the source for icd-10pc to install in GG teminology server? What licensing or copyright things are involved.
David Teirney (Dec 19 2017 at 02:44):
I've got the same question. I'd guess it will be published as http://hl7.org/fhir/sid/icd-10-pcs. Is there someone that can confirm?
David Teirney (Dec 19 2017 at 03:03):
I found "http://www.icd10data.com/icd10pcs" listed in http://build.fhir.org/ig/Healthedata1/US-Core/ValueSet-us-core-procedure-icd10pcs.html. Is that what we should be using for the PCS codes?
Rob Hausam (Jan 17 2018 at 15:23):
@David Teirney I don't think that an icd10data.com url is the right way to go with this, as that's a commercial service and definitely not the code system owner or an officially authoritative source. That still leaves the question of what the right url is, though - I guess with the precedent we've already been following and as you suggested it would be http://hl7.org/fhir/sid/icd-10-pcs. Any different suggestions from anyone?
Joshua Bell (Jan 17 2018 at 18:14):
I have been using http://hl7.org/fhir/sid/icd-10-cm in our application as well.
I think my only confusion would be the fact that there are different region versions of icd10. For example, I am using the US version, but that doesn't change the uri at all. I see good and bad reasons to keep the url the same, but haven't really bothered too much as it seems everyone is just using that uri for any use of icd10.
Rob Hausam (Jan 17 2018 at 18:43):
This is icd-10-pcs, which is a US-specific standard for procedure coding. It has some relation (at least in spirit) to ICD-10, but none of it is from WHO and it's really a different thing, and it doesn't have anything directly in common or related to icd-10-cm at all (other than that CMS publishes both).
Robert McClure (Jan 17 2018 at 18:51):
So far as I am aware, and seen, for ICD code systems across all varients, the string identifiaction is always ICD-9-xxx or ICD-10-xxx (and presumabaly will be ICD-11-xxx.) Where xxx is the particular varient defined. In the USA there are two - ICD-10-CM, and ICD-10-PCS. In Canada it is ICD-10-CA, In Germany it is ICD-10-GM, etc. Until the USA CMS decides to take ownership of the ICD url for FHIR (and beyond?) it makes sense that HL7/FHIR provide it. This does raise the issue that these urls can change over time when the "owner/authority" decides to take over hosting. I think we've already had that occur for something.
David Teirney (Feb 02 2018 at 10:15):
@Rob Hausam sounds like there is support for http://hl7.org/fhir/sid/icd-10-pcs being the URL. Is http://build.fhir.org/icd.html where that information would end up being specified before being formally published down the track?
Jay Lyle (Nov 13 2018 at 00:51):
Whatever the FHIR uri, @Eric Haas , did you ever get an answer? The codes are here, but I can't find anything on permissions.
Rob Hausam (Nov 13 2018 at 01:13):
What exactly are you wanting/needing at this point, Jay? I don't think we've done much, if anything, further with this, but we may need to.
Jay Lyle (Nov 13 2018 at 01:36):
I'm just wondering what the license angle looks like. They look free, smell free, but there's no open license or any other language that I can see telling people what they're allowed to do with them.
Zabrina Gonzaga (Jan 27 2022 at 20:18):
Is there specific guidance on how to reference ICD-10-pcs? Do you know if this is supported by FHIR? I was unable to locate details on this page: https://www.hl7.org/fhir/icd.html
Rob Hausam (Jan 28 2022 at 16:38):
@Zabrina Gonzaga I think this is still languishing. Do we need to bring it forward?
Robert McClure (Jan 31 2022 at 00:04):
I'm a bit surprised that we've still not made clear that HTA is responsible for referencing external code systems and everyone should look on the HTA page to find the latest information, not the FHIR build. Someday we'll get it all there but not yet. Please follow the guidance at https://confluence.hl7.org/display/TA/ICD-10+Procedure+Coding+System and if you can not find what you need there, enter a comment at the bottom of the page and ask HTA to help. And when you go there you will find the official FHIR url of https://www.cms.gov/Medicare/Coding/ICD10 which we all are less than pleased with.
Rob Hausam (Jan 31 2022 at 02:20):
I agree with that. But I don't think the url alone is all that is needed.
Lloyd McKenzie (Jan 31 2022 at 03:13):
We're still working on process and cleanup of the HTA content. Where the content has been defined in the FHIR spec, that's what has to be used. The intention is eventually to have the THO site be the source of truth - once we ensure it's: a) correct; b) maintained by a process that ensures URLs don't change once issued; and c) meet the criteria that ensures they're appropriate for implementer use.
Zabrina Gonzaga (Jan 31 2022 at 13:42):
Thank you @Rob Hausam , @Robert McClure and @Lloyd McKenzie . I will use the URL referenced in the link above and see if it will validate. A pointer to the confluence page above would be helpful to those seeking answers here: https://www.hl7.org/fhir/icd.html. Perhaps adding a statement in section 4.3.11.2 ICD-10 variants for ICD-10-pcs? Just a thought. I appreciate your guidance.
Robert McClure (Jan 31 2022 at 13:46):
@Lloyd McKenzie As we've discussed multiple times, these two locations for content should not be in conflict. Are you suggesting they are? In addition, the FHIR community has been notified in every way that FMG has allowed, that external code system information is managed by HTA. Perhaps we should be allowed to update all the "using FHIR" pages as Zabrina suggests.
Lloyd McKenzie (Jan 31 2022 at 15:22):
There were places where there was inconsistency. There was supposed to be a review to update the HTA recommendations to align with what FHIR mandates, but I haven't heard that that's happened yet. Once the process is finalized and there's trust that HTA won't change URLs - regardless of the desires of the terminology source and once URLs are all accurately reflected in THO, then the FHIR specification should simply point to THO.
Last updated: Apr 12 2022 at 19:14 UTC
