Stream: IG creation
Topic: New ICD-10-CM warnings
Jean Duteau (Jul 08 2020 at 15:43):
Today I am getting a new set of warnings about a bunch of ICD-10-CM codes that are not valid. These codes do actually exist in the ICD-10-CM code system (I just re-checked) and I didn't get warnings before today. I'm also getting warnings on CVX and NDC codes that are new.
Rob Hausam (Jul 08 2020 at 15:50):
I'm not sure what warnings you are getting, but Grahame keeps making the Publisher and Validator better at finding issues. :) Probably they are being generated because tx.fhir.org doesn't have support presently for those particular codes. If you would like to have those codes supported you can let me know that they are and we can arrange to do that. That definitely may occur for ICD-10-CM and NDC, as the available codes for them in the terminology server are quite limited. Possibly the CVX version is outdated and needs to be updated.
Jean Duteau (Jul 08 2020 at 15:54):
it looks like all of my ICD-10-CM codes are now being flagged as invalid. If tx.fhir.org doesn't have all of the ICD-10-CM codes (and CVS and NDC and whatever) maybe we shouldn't be flagging those as invalid (even though they are warnings)?
Rob Hausam (Jul 08 2020 at 18:25):
I suspect I know what they are, but can you post a sample of the messages that you are getting? I think that possibly adjusting the warning level and text is something that we may need to look at further.
Jean Duteau (Jul 08 2020 at 18:32):
ValueSet/AnaphylaxisSymptomVS: ValueSet.compose[0].include[0].concept[0] warning The code H10.2 is not valid in the system http://hl7.org/fhir/sid/icd-10-cm
ValueSet/VaccineCodeVS: ValueSet.compose[0].include[0].concept[60] warning The code 197 is not valid in the system http://hl7.org/fhir/sid/cvx
Saul Kravitz (Jul 13 2020 at 21:28):
I'm having the same problem with this code: "Code "S06.0x1A" not found; The code "S06.0x1A" is not valid in the system http://hl7.org/fhir/sid/icd-10-cm; "
https://icd.codes/icd10cm/S060X1A#:~:text=a%20principal%20diagnosis.-,S06.,minutes%20or%20less%2C%20initial%20encounter.
See https://saulakravitz.github.io/carin-bb-publish/qa.html
Eric Haas (Jul 21 2020 at 20:40):
I have this NDC warnings for us core:
Immunization/imm-1: Immunization.vaccineCode.coding[1] (l17/c5) error The code "49281-0631-15" is not
valid in the system http://hl7.org/fhir/sid/ndc; The code provided (http://hl7.org/fhir/sid/ndc#49281-0631-15) is
not valid in the value set All codes known to the system for "http://hl7.org/fhir/sid/ndc#49281-0631-15" (see Tx
log)
https://ndclist.com/ndc/49281-631/package/49281-631-15
@Rob Hausam how do I access what is on the term server?
Eric Haas (Jul 21 2020 at 20:44):
Basically every code I have for NDC is invalid
Rob Hausam (Jul 22 2020 at 12:47):
@Eric Haas Terminology server support for NDC is still a problem. We're trying to work on it. If you have a list that you can provide of all of the NDC codes that you need, we can get those codes (at least) supported. The latest issue with it that we've been trying to address is that there are both the 10-digit (with dashes) formats on the packages and from the FDA and the 11-digit format that is used for prescribing (NCPDP) and by the drug knowledge databases. Which format do you need/want - or do you need both?
Vassil Peytchev (Jul 22 2020 at 15:51):
Isn't there an algorithmic conversion between 10- and 11-digit NDC codes?
Gino Canessa (Jul 22 2020 at 16:32):
Wow that's a painful rabbit hole. As far as I can tell you need to see the original grouping pattern to modify from 10 to 11. That, or you need a current list of manufacturers, forms, and package codes in 10-digit form to lookup and do the same conversion.
Eric Haas (Jul 22 2020 at 16:38):
@Rob Hausam thanks for the response. It would be great if the CDC vaccine list could be added.
https://www2a.cdc.gov/vaccines/iis/iisstandards/ndc_crosswalk.asp
Vassil Peytchev (Jul 22 2020 at 16:44):
@Gino Canessa AFAIK 10-digit NDC codes are always represented with dashes, so the grouping pattern is always there. I am not sure whether it is better to have two views of the same code, or having two codes, and then indicate that they are equivalent....
Gino Canessa (Jul 22 2020 at 17:19):
If you have access to the dashes for a 10-digit, it is quite simple to convert to 11 based on the grouping pattern. For the coding, I generally lean towards listing both as equivalent so that others downstream don't have to ask the same question again.
Rob Hausam (Jul 22 2020 at 17:24):
@Vassil Peytchev Yes, there is a generally accepted (but still unofficial from the NDC "publisher" perspective) algorithm for transforming the 10-digit codes to 11-digit format (as Gino just said). But you can't do that in reverse, as there is no algorithmic way to determine which of the three patterns of the dashes to apply, and they result in different (actual or potential) codes. Your second option of having two codes and declaring that they are equivalent is what I believe we will need to do in FHIR - unless, instead for convenience we decide to actually represent them as two separate code systems. We're still having some discussions in HTA (who said previously it should be a single code system) and Vocab WG about this.
Rob Hausam (Jul 22 2020 at 17:27):
You can still declare the equivalence (using ConceptMap) if they are in separate code systems (and it is probably even more "normal" and obvious if they are in different systems), so we can do that in either case.
Vassil Peytchev (Jul 22 2020 at 17:44):
What I want is the ability to say that "0002-7597-01" and "00002759701" are the same concept. I think it might be less worse to have a single concept within the code system, with Concept.code="00002759701", and (possibly) the Concept.property.code="NDC10", and Concept.property.valueString="0002-7597-01".
Mark Kramer (Jul 23 2020 at 19:18):
+1. All 6000 of my ICD-10-CM codes are getting warning messages.
Eric Haas (Jul 30 2020 at 20:31):
@Rob Hausam where are we with this with re my request to add all the immunization codes from CDC:
https://www2a.cdc.gov/vaccines/iis/iisstandards/ndc_crosswalk.asp
Last updated: Apr 12 2022 at 19:14 UTC
