FHIR Chat · CDA format code vocabulary · terminology

Stream: terminology

Topic: CDA format code vocabulary

John Moehrke (Oct 07 2019 at 18:22):

I understand that there are now some vocabulary values managed by HL7 for CDA "formatCode" vocabulary. This is in relationship to GF#14104. I have been charged with making the updates to the FHIR ValueSet used in DocumentReference. I am also charged with coordinating with IHE on their stopgap creation of codes that should now be managed by HL7. What I need is details on where within HL7 these codes are now managed (what codeSystem), and to determine if they are in the tx.fhir.org terminology service so that I can compose them into a Valueset.

Grahame Grieve (Oct 08 2019 at 00:07):

according to the spec, the formatCode value set is an HL7 maintained value set that refers to an IHE maintained code system

Rob Hausam (Oct 08 2019 at 00:13):

So it seems like we need to make some changes with that. Probably John and I should coordinate a time to go over it and make sure the details are clear and then get it done.

John Moehrke (Oct 11 2019 at 19:41):

The CR GF#14104 asked for HL7 to take ownership of those FormatCode values that they have created. There is the hamonization request attached to the CR. I understand that harmonization happened, and that the list of codes now exist somewhere in an HL7 managed codeSystem. I however don't know what codeSystem. In order to create the new ValueSet that the CR asks me to create, I need to find the codeSystem into which these HL7 managed formatCodes were placed.

Rob Hausam (Oct 11 2019 at 19:44):

Do you have an idea which harmonization cycle it was?

Rob Hausam (Oct 11 2019 at 19:45):

oh, it looks like it should have been July 2019

Rob Hausam (Oct 11 2019 at 19:54):

here's the code system at terminology.hl7.org (UTG):
http://build.fhir.org/ig/HL7/UTG/branches/master/CodeSystem-hl7-document-format-codes.html

John Moehrke (Oct 11 2019 at 20:13):

ah, thanks. I will work on the ValueSet.

John Moehrke (Oct 11 2019 at 20:16):

before I put this into the build, I really need the IHE FormatCode vocabulary coordinated too...
With my IHE hat on... I want to use this as a forcing function to getting IHE publishing something using the IG publisher.... Otherwise I can just put the updated codeSystem on the IHE ftp site, which I think is where the fhir core build pulls them from...

John Moehrke (Oct 11 2019 at 20:18):

in IHE, these HL7 codes have been published for a year or more... Is there a way to mark them deprecated? Or do I just remove them?

Rob Hausam (Oct 11 2019 at 20:25):

I'm not sure what is needed for the process of making the code system updates (deprecating the HL7 codes) on the IHE side. If IHE has a mechanism for deprecating/inactivating the codes then I think you would do that. Otherwise presumably you will create a new version of the code system with those codes removed.

John Moehrke (Oct 11 2019 at 20:35):

I was mostly asking about the FHIR CodeSystem.. but it doesn't look like there is a way to leave them in a CodeSystem and marked do-not-use

John Moehrke (Oct 11 2019 at 20:36):

the GitHub I asked you to look at is the IHE codeSystem with the HL7 codes just commented out... so, removed.

John Moehrke (Oct 11 2019 at 20:37):

one thing I want you to look at is the versioning stuff I do in that IHE managed codeSystem.

Grahame Grieve (Oct 15 2019 at 13:16):

ok, catching up... it's so hard to stay up to date.

Grahame Grieve (Oct 15 2019 at 13:16):

you can make codes as deprecated - is that what you want?

Grahame Grieve (Oct 15 2019 at 13:16):

which github did you want me to look at?

Rob Hausam (Oct 15 2019 at 13:54):

it definitely is hard to stay up to date
I'm not quite sure either what the GitHub link is that John is referring to for the IHE codeSystem

Grahame Grieve (Oct 15 2019 at 21:09):

found it. see https://github.com/IHE/FormatCode/blob/master/input/resources/codesystem-IHE.formatcode.cs.xml

daniel venton (Oct 16 2019 at 11:32):

Why would you deprecate the format codes that correspond to ccda document types anyway? What are they replaced with? In other words when I'm sending an IHE transaction that contains a CCDA document, what format code do I use in the message? Are you trying to say that the format codes used in the IHE transactions come from a value set comprised of values defined by IHE and HL7? In which case what is that value set and shouldn't IHE themselves have to update their TF (or have they)?

Also I don't think the code values are supposed to have spaces for example:
<concept>
<code value="urn:ihe:pat:apsr:cancer:esophagus: 2010"/>
<display value="Anatomic Pathology Structured Report Cancer Esophagus"/>
</concept>

John Moehrke (Oct 16 2019 at 20:29):

The CDA codes are now owned by HL7, so the codeSystem is now formal. Historic use of these codes incorrectly used the IHE code system. For this historic reason they were published in the IHE codeSystem for a period of time to enable people finding them published somewhere.

John Moehrke (Oct 16 2019 at 20:30):

you can make codes as deprecated - is that what you want?

@Grahame Grieve I looked at codeSystem resource and can't figure out how to mark codes as deprecated.

Michael Lawley (Oct 16 2019 at 22:23):

Use the status property -- see http://hl7.org/fhir/codesystem.html#status

John Moehrke (Oct 17 2019 at 12:46):

The codeSystem is not deprecated. Just some of the codes in that codeSystem are deprecated. These codes now exist in a different codeSystem owned by HL7.

John Moehrke (Oct 17 2019 at 12:46):

Look at this draft codeSystem I have on GitHub. Look at the bottom of the XML to see the codes.

John Moehrke (Oct 17 2019 at 12:46):

https://github.com/IHE/FormatCode/blob/master/input/resources/codesystem-IHE.formatcode.cs.xml

John Moehrke (Oct 17 2019 at 12:47):

These are still good codes, just belong in a different code system

Robert McClure (Oct 17 2019 at 14:14):

The transition of these codes to HL7 was approved at the July harmonization meeting. I don't think the content has actually appeared yet - I'm following up about that - but when it appears, the defining url will be http://hl7.org/fhir/hl7formatcodes and shoould show up at http://terminology.hl7.org/CodeSystem/hl7-document-format-codes. The name will be HL7DocumentFormatCodes. It will have the following codes:
pasted image

John Moehrke (Oct 17 2019 at 14:15):

Yes, I am the one making that happen.

John Moehrke (Oct 17 2019 at 14:17):

I am asking how I mark these codes that have been used within the IHE FormatCode codeSystem (I am the author) such that the community understands that these same code values may appear in the IHE managed FormatCode system because of history, but should going forward use the HL7 managed FormatCode system for THESE values.

John Moehrke (Oct 17 2019 at 14:17):

I don't see a way to do anything other than comment them out of the IHE FormatCode codeSystem... as I showed in the github above

John Moehrke (Oct 17 2019 at 14:19):

if that is all I can do... then that is what I will do.

Robert McClure (Oct 17 2019 at 14:24):

@John Moehrke , Michael provided the correct guidance, you use the http://hl7.org/fhir/codesystem.html#status property. This is a concept property, not the entire code system. Note the set of elements that can (and should) be used to note the date. @Grahame Grieve , can you provide specific guidance to John on how to do this?

John Moehrke (Oct 17 2019 at 14:43):

I then am too stupid to understand how to mark a code as deprecated. Can someone provide example? I only see a .status element at the codeSystem level, not at the codeSystem.concept level

John Moehrke (Oct 17 2019 at 14:48):

I do see a handy extension for replacedBy -- http://hl7.org/fhir/extension-codesystem-replacedby.html -- does someone have an example of use of this that I can cheat off of?

Michael Lawley (Oct 17 2019 at 18:30):

What I pointed to is a concept status not the CodeSystem.status. The way to use is is to add a property called "status" with value = "deprecated" to each of your deprecated codes. You'll also want to declare that property in the CodeSystem using the appropriate URI http://hl7.org/fhir/concept-properties#status
Here's an example

{
   "resourceType" : "CodeSystem",
   "name" : "DeprecationExample",
   "text" : {
      "status" : "generated",
      "div" : "<div xmlns=\"http://www.w3.org/1999/xhtml\"><h2>DeprecationExample</h2><tt>urn:code-deprecation</tt><p></p></div>"
   },
   "meta" : {
      "versionId" : "1",
      "lastUpdated" : "2019-10-18T04:29:40.245+10:00"
   },
   "url" : "urn:code-deprecation",
   "content" : "complete",
   "compositional" : false,
   "count" : 3,
   "id" : "9d5e05f6-af76-45e8-8dd7-35032bb50950",
   "status" : "draft",
   "property" : [
      {
         "code" : "status",
         "type" : "code",
         "uri" : "http://hl7.org/fhir/concept-properties#status"
      }
   ],
   "concept" : [
      {
         "display" : "OLD",
         "code" : "old",
         "property" : [
            {
               "code" : "status",
               "valueCode" : "deprecated"
            }
         ],
         "definition" : "my old code"
      },
      {
         "display" : "NEW",
         "definition" : "my new code",
         "code" : "new"
      }
   ]
}

Rob Hausam (Oct 17 2019 at 18:54):

yes, that's the way to do it

John Moehrke (Oct 17 2019 at 21:53):

can we get this into the spec as an example on codeSystem?

John Moehrke (Oct 17 2019 at 21:54):

also would like an example that uses the replacedBy extension

John Moehrke (Oct 17 2019 at 22:05):

I just pushed my updates with the above deprecated status. Anyone willing to check my work? The IG does build without failure. See my IG in prototype form (imagine it with a blue background and IHE logo) http://build.fhir.org/ig/IHE/FormatCode/branches/master/index.html

Michael Lawley (Oct 17 2019 at 22:08):

Feel free to POST your CodeSystems to https://r4.ontoserver.csiro.au/fhir/CodeSystem/$validate (although more useful for ValueSets as we do extra validation there with respect to implicit ValueSet URI and SNOMED ECL parsing etc)

John Moehrke (Jan 02 2020 at 21:49):

The transition of these codes to HL7 was approved at the July harmonization meeting. I don't think the content has actually appeared yet - I'm following up about that - but when it appears, the defining url will be http://hl7.org/fhir/hl7formatcodes and shoould show up at http://terminology.hl7.org/CodeSystem/hl7-document-format-codes. The name will be HL7DocumentFormatCodes. It will have the following codes:
pasted image

This formatcode vocabulary from HL7 is still needed to be published.. right? @Grahame Grieve

Grahame Grieve (Jan 04 2020 at 09:27):

it's in UTG. we can use it there

John Moehrke (Jan 04 2020 at 16:24):

Thanks. The IHE codesystem is updated and a valueset that combines the IHE codes and the HL7 codes is now published for draft purposes at http://build.fhir.org/ig/IHE/FormatCode/branches/master/index.html
A few tweaks and approvals before it gets moved over to the permant home on http://ihe.net/fhir/ihe.formatcode.fhir
Comments, Suggestions, Complaints... are welcomed by me.

Rob Hausam (Jan 24 2020 at 21:21):

I know this has taken a long time, but here's an update on the hl7-document-format-codes code system. It has been committed and pushed to the FHIR Github 'packages' repository, where eventually it will be picked up by the tx.fhir.org server (for r4 - and we can add r3, if needed). I say "eventually", because there is another manual step that first needs to fixed and then completed before the code system will show up on the server. In regard to the canonical url, @Robert McClure's comment in October says "the defining url will be http://hl7.org/fhir/hl7formatcodes and should show up at http://terminology.hl7.org/CodeSystem/hl7-document-format". The code system does show up now in HL7 Terminology at http://build.fhir.org/ig/HL7/UTG/branches/master/CodeSystem-hl7-document-format-codes.html, and its canonical (defining) url as shown there is http://terminology.hl7.org/CodeSystem/hl7-document-format-codes. I think that is the canonical url that we would want (not http://hl7.org/fhir/hl7formatcodes)? - but we should clarify that. @John Moehrke @Grahame Grieve

John Moehrke (Jan 27 2020 at 02:30):

lets discuss and make a plan in sydney.

John Moehrke (Feb 04 2020 at 21:56):

When will this topic be discussed so that I can attend? Most open for me is Friday

Rob Hausam (Feb 04 2020 at 22:10):

I don't think we have it on the agenda anywhere at the moment. It might fit with the UTG discussion this morning in Q1, if there is time (and if you might have any availability)? Or I'm sure that we can do it in Vocab Friday Q4 (if that's not too late to do it).

John Moehrke (May 21 2020 at 14:57):

I looked again, it still does not seem to be in the terminology server

http://terminology.hl7.org/CodeSystem/hl7-document-format-codes

Rob Hausam (May 21 2020 at 19:19):

No, it's not, John. Sorry about that. The code system is there, but there's been the issue with loading the package on the server and I've been waiting for a fix from Grahame on that.

Rob Hausam (May 21 2020 at 19:21):

On the 18th he said "fixed next update". I don't think the update has happened yet, but I can check again.

John Moehrke (May 22 2020 at 11:10):

I will keep checking

Rob Hausam (May 22 2020 at 11:11):

OK. I am, too, and will let you know when it is working.

John Moehrke (Jul 29 2020 at 14:04):

I still can't find these terminologies published. I am approaching getting the IHE codes published, so it is not like I am moving faster. I would like to have my IHE valueSet be able to include these HL7 defined formatCodes.

Rob Hausam (Jul 29 2020 at 14:55):

There is still a bottleneck with getting this content from the updated package (which has been done) to actually be loaded and operational in the terminology server. I will be spending more time on outstanding terminology service tasks over the rest of this week. My last discussion with Grahame on this was in late May, and I will check with him again on it now.

Rob Hausam (Jul 29 2020 at 15:39):

@Grahame Grieve Are we (close to being) ready to move this forward?

Grahame Grieve (Jul 29 2020 at 20:01):

I don't remember the issues?

Rob Hausam (Jul 29 2020 at 23:17):

Back on Jan. 24 the CodeSystem resource with url = http://terminology.hl7.org/CodeSystem/hl7-document-format-codes was committed to the FHIR/packages repo, but the content has never become available on tx.fhir.org. It was showing up in the UTG site at http://build.fhir.org/ig/HL7/UTG/branches/master/CodeSystem-hl7-document-format-codes.html, but it's not there now (and it also doesn't show up at https://terminology.hl7.org/CodeSystem-hl7-document-format-codes.html). Since quite a bit of time has passed and with the recent changes in publisher 1.1.4 to include UTG, we may want to rethink how we want to handle this. If the code system (and value sets) become available in UTG and are accessible by the publisher that would fulfill the need.

John Moehrke (Oct 26 2020 at 12:10):

Any update on the availability of the HL7 format codes @Rob Hausam ? Now @Brett Marquard and @Eric Haas are asking about it.

Robert McClure (Oct 26 2020 at 18:55):

@Rob Hausam I suspect that this needs to be redone, now as a UTG proposal to get things into UTG/Terminology.hl7.org (THO). I'll bet this got lost with the movement of things from FHIR into THO.

Rob Hausam (Oct 27 2020 at 06:22):

@John Moehrke @Robert McClure It's been stalled waiting for the terminology server to be able to correctly pick up the changes. Grahame has been doing a lot of work on updating the server recently, and I expect that we are there now (or at least will be finally very soon). I will check it during the day and see where we are with it and whether it's finally ready.

John Moehrke (Oct 27 2020 at 12:25):

will that include pulling in the IHE FormatCode vocabulary from the IHE published implementation guide into the hl7 terminology server? I think we need both the IHE codes pulled in, and the HL7 defined codesystem created.

Rob Hausam (Oct 28 2020 at 05:12):

@John Moehrke This wasn't working the last time I had checked it (a good while back), but it is now. You can see the code system here:

http://tx.fhir.org/r4/CodeSystem?url=http://terminology.hl7.org/CodeSystem/hl7-document-format-codes

It has the HL7 C-CDA 1.1, 2.1 and 2.2 and PACP 1.0 codes. It's been so long that I don't recall (or maybe don't even know) if there is anything else that is supposed to be there in the code system. I don't recall exactly where we are with the IHE codes, but I believe that is (or will be) a separate code system and the value set(s) would include codes from both code systems. Update me if that isn't correct and with what is. I'm sure we can get it the way that it needs to be, as long as we are clear on what that is. Further updates to the code system, if needed, I think should no longer be problematic.

John Moehrke (Oct 28 2020 at 14:12):

This is the HL7 wiki page with the codes https://wiki.hl7.org/CDA_Format_Codes_for_IHE_XDS
so it holds those, it looks like some more have come along...

John Moehrke (Oct 28 2020 at 14:13):

there is no provenance (history) but I guess that this codeSystem was created before that

John Moehrke (Oct 28 2020 at 14:15):

The canonical URI doesn't resolve https://terminology.hl7.org/CodeSystem/hl7-document-format-codes Is that something that should be expected?

John Moehrke (Oct 28 2020 at 14:25):

I was able to add this codeSystem to my IHE formatcode ValueSet.. and it worked... YEAH!!!! Looks like I need to do a revision of the IHE formatCode IG.

John Moehrke (Oct 28 2020 at 15:15):

@Rob Hausam how do I get HL7 to pull in the IHE vocabulary from this publication? Would this go to terminology or UTG?

John Moehrke (Oct 28 2020 at 15:27):

I have updated the IHE formatCode IG in the continuous build, will move to formal publication later... see http://build.fhir.org/ig/IHE/FormatCode/branches/master/ValueSet-formatcode.html

Rob Hausam (Oct 29 2020 at 07:01):

@John Moehrke, let's touch base on this and figure out exactly what is still needed. If there isn't one already, I'm sure we need a UTG proposal. This one was started before UTG was ready, of course.

John Moehrke (Dec 14 2020 at 12:57):

I still can't human-navigate to this codeSystem on the terminology server... The canonical URI does not resolve or redirect. The only success I have is the query you offered against tx.fhir.org; and my IG can find it in an IG build. Why are these codes still not showing up in terminology.hl7.org?

Rob Hausam (Dec 14 2020 at 17:27):

I'm not sure if it is supposed to show up in terminology.fhir.org. Can you share your rationale for why you think it should be there (I'm not saying you are wrong about that, just want to know what your reasoning is)? The work that I've done on it was only for getting it on tx.fhir.org.

John Moehrke (Dec 15 2020 at 14:25):

because it is HL7 defined vocabulary. These are the HL7 defined FormatCode vocabulary. The IHE ones are elsewhere, and I think we are equally trying to get the IHE ones to show up in UTG. right?

Rob Hausam (Dec 15 2020 at 16:18):

We may need to revisit where we are or where we want to be with this now that UTG/THO is "official". And see if any UTG proposals are needed (or are still to be acted on).

John Moehrke (Dec 16 2020 at 15:56):

These have been in the queue for well over a year... Let me know how I can help. I just am frustrated that simple things keep getting delayed, and that delay turns into more work.

John Moehrke (Mar 22 2021 at 17:38):

poke... when will this be fixed?

Rob Hausam (Mar 22 2021 at 21:51):

@John Moehrke I think we're still going to need a UTG (THO) Jira issue for this (I don't believe there is one yet?). @Ted Klein?

John Moehrke (Mar 22 2021 at 21:53):

I have no idea what needs to be said in a jira ticket. All I know is that somehow my IG that was able to expand the codes into my valueset, yet I can't find any way to navigate to the codesystem.

Rob Hausam (Mar 22 2021 at 21:56):

I think that I or @Ted Klein or @Jessica Snell (Jess Bota) can help you figure out what needs to be in the Jira issue. I know this has been dragging out forever, and I'm sure that everyone (especially you) wants to get this done.

Jessica Bota (Mar 23 2021 at 13:37):

@Rob Hausam @John Moehrke , I've been catching up on the latest chatter across several topics about UTG vs. FHIR and how certain artifacts are maintained. This formatCodes value set appears to fall under the FHIR ballot-bound content, which we've documented here: https://confluence.hl7.org/display/VMAH/Ballot-Bound+Terminology+Content . More specifically, these are the lists of ballot-bound FHIR value sets and this one is on the list: http://build.fhir.org/valueset-formatcodes.html . It seems we need to do some brainstorming on how to make it easier to know where to access certain content and how it is maintained.
@Ted Klein

Ted Klein (Mar 23 2021 at 15:55):

My original design called for a look=aside in the Jira scripting so users who selected artifacts (CS, VS, etc.) would be directed to where they need to go to submit change requests (if not maintained in THO through UTG), but this, like the scores of other small useful tool development tasks, is buried in the queue and voluminous task list. We will always have certain vocabulary that may be rendered in THO but is not maintained through UTG, but through ballot or other processes. We need to address this for the community, but again it is a matter of gathering the lists in a machine readable form and developing some Jira programming to make it happen. Take a number, here you have number 358...now calling number 7. And if we continue to just keep throwing stuff over the wall to the nascent terminology governance group which does not appear to be moving towards formation with any degree of alacrity, that will ensure we will all be long retired before it gets done. I don't know if @Robert McClure is keeping a list of things tossed there in the past few months or not; I have no such list.

John Moehrke (Mar 23 2021 at 16:01):

two very different things. I am FIRST looking for the HL7 authored CodeSystem for the HL7 defined FormatCodes. This is a small set of codes.
https://wiki.hl7.org/CDA_Format_Codes_for_IHE_XDS

John Moehrke (Mar 23 2021 at 16:01):

I manage the codeSystem from IHE
https://profiles.ihe.net/fhir/ihe.formatcode.fhir/

John Moehrke (Mar 23 2021 at 16:02):

I have a ValueSet in my IHE based IG that does succeed to somehow find the HL7 defined FormatCodes
https://profiles.ihe.net/fhir/ihe.formatcode.fhir/ValueSet-formatcode.html

John Moehrke (Mar 23 2021 at 16:02):

IG build magjic, as I can't find them on terminology.hl7.org with my browser

John Moehrke (Mar 23 2021 at 16:03):

SO, how do I as a human find this CodeSystem that seems to exist in the ether?

John Moehrke (Mar 23 2021 at 16:04):

SECOND, there is a FormatCode valueset in FHIR. I am also the FHIR core team member responsible for this... it is intended to be replaced with the FormatCode valueset from IHE.. but I can't do that until all the ducks are in a row. I can make this change. This is not the problem

Ted Klein (Mar 23 2021 at 18:45):

'in the ether'. sigh. so when you build an IG it is my understanding that the process looks for referenced code systems in the following places:

Ted Klein (Mar 23 2021 at 18:45):

tx.fhir.org for stuff that is already loaded into the terminology server

Ted Klein (Mar 23 2021 at 18:45):

FHIR core

Ted Klein (Mar 23 2021 at 18:45):

THO (terminology.hl7.org)

Ted Klein (Mar 23 2021 at 18:45):

local IG

Ted Klein (Mar 23 2021 at 18:46):

other sources for terminology indicated with a parameter reference in the ig control file; sources can include other IGs or other organizations where the content may be found

Ted Klein (Mar 23 2021 at 18:47):

This list may not be 100% complete, but this is what I know based upon what I have learned and been told

John Moehrke (Mar 25 2021 at 13:30):

okay, so where did it find these codes? They are NOT local.

John Moehrke (Apr 06 2021 at 14:44):

so, my IHE IG builds fine, pulling in the HL7 codes... No idea where the IG builder finds the HL7 codes.
But, when I tried to replicate my valueset in the HL7 FHIR Core build; the fhir core build failed to find both the IHE codes and the HL7 codes.

John Moehrke (Apr 06 2021 at 16:37):

@Grahame Grieve I am noticing that the FHIR build still pulls the IHE FormatCode vocabulary from the IHE FTP site... can this be done with the IHE published IG now?

John Moehrke (Oct 01 2021 at 00:59):

@Grahame Grieve terminology called me to talk about this again... too much Grahame magic for us to get far

Rob Hausam (Oct 01 2021 at 20:21):

The http://terminology.hl7.org/CodeSystem/hl7-document-format-codes code system in in tx.fhir.org (from the packages repository). I looked at that with Ted earlier this afternoon.

John Moehrke (Jan 31 2022 at 14:06):

What is the current situation around 'support' for the IHE published FormatCode vocabulary, and the HL7 published FormatCode vocabulary?

Grahame Grieve (Jan 31 2022 at 23:10):

I think it should work now

Last updated: Apr 12 2022 at 19:14 UTC