FHIR Chat · FHIR Validator · CARIN IG for Blue Button®

Stream: CARIN IG for Blue Button®

Topic: FHIR Validator


view this post on Zulip jason chow (Dec 08 2020 at 13:44):

We've started to run our eob's through the FHIR validator jar, and on a lark, I applied the validator to the EOBOutpatientInstitutional1 example json provided with the IG. The validator returned numerous errors. (see attached carin-op-output-1.html ).
Should we expect the examples to pass validation or are the examples to be accepted as more general guides?

view this post on Zulip Saul Kravitz (Dec 08 2020 at 19:32):

Hi @jason chow
The examples should and do pass validation -- see http://hl7.org/fhir/us/carin-bb/qa.html#C__temp_hl7.fhir.us.carin-bb_1.0.0-milestone_input_examples_ExplanationOfBenefit-EOBOutpatientInstitutional1.

The two errors are known issues with terminology resolution. All of the remaining errors were reviewed by the FMG along with my excuses.

I think you have something else going on with your validation attempts.

view this post on Zulip Josh Lamb (Dec 09 2020 at 15:47):

Are you using the latest validator? https://confluence.hl7.org/display/FHIR/Using+the+FHIR+Validator

The validator gives you line number, should be somewhat clear if the issues are valid or not.

view this post on Zulip jason chow (Dec 10 2020 at 14:26):

@josh lamb
Yes. Current version.
What is throwing me for a loop is using the ExplanationOfBenefit-EOBOutpatientInstitutional1.json example (downloaded from the IG's examples) and feeding it to the validator with the uri for the outpatient profile produces 25 errors. (output carin-eob-outpatient-validation.html attached)
While it does include line and column numbers, when the error message says:
"Error @ ExplanationOfBenefit (line 1, col2) : ExplanationOfBenefit.diagnosis: minimum required = 1, but only found 0 (from http://hl7.org/fhir/us/carin-bb/StructureDefinition/C4BB-ExplanationOfBenefit-Outpatient-Institutional)"
I can't deduce what the problem is; because, line 1 col2 is the newline after the opening curly bracket.
What's even more confusing is that the profile json does indeed have a diagnosis property with a principal diagnosis element in it.

example command:
java -jar validator_cli.jar ExplanationOfBenefit-EOBOutpatientInstitutional1.json -version 1.0.0 -ig http://build.fhir.org/ig/HL7/carin-bb -profile http://hl7.org/fhir/us/carin-bb/StructureDefinition/C4BB-ExplanationOfBenefit-Outpatient-Institutional -html-output carin-eob-outpatient-validation.html

view this post on Zulip Saul Kravitz (Dec 10 2020 at 14:47):

Not that it should make much difference, but: the IG is published, so I'm not sure you need the -ig, and if you do need the -ig, it should point to the published location -- http://hl7.org/fhir/us/carin-bb/STU1/ or http://hl7.org/fhir/us/carin-bb . I don't think you need to specify the -profile. The documentation says it is preferable to pull in the set of profiles using the -ig flag (as necessary).

view this post on Zulip Josh Lamb (Dec 10 2020 at 14:50):

Hi Jason,

Something else may be going on. My validation results are different than yours.

Here is the command I am running (latest validator):
java -jar validator_cli.jar ......\OutpatientEOB.json -ig http://hl7.org/fhir/us/carin-bb/ -output Result.json

Here is the output I receive:
Results.json

There are validation issues, but they are specific to the terminologies. The validator is having issues resolving some of the codes we use in CARIN IG.

view this post on Zulip Saul Kravitz (Dec 10 2020 at 14:53):

The error's in Josh's output are precisely those found validating the examples as part of the IG. They are expected given the current capabilities of the terminology services.

view this post on Zulip jason chow (Dec 10 2020 at 15:24):

Thanks for helping me out.
Using that command i got the same result.
Where did you pull the URI for the IG from? I didn't see an obvious URI for the published profile on the profile main page or the published profiles page. Did you copy the link address of the latest profile?

view this post on Zulip Josh Lamb (Dec 10 2020 at 15:32):

Others may be more sophisticated, but I just visit the website for the IG I want to validate against, and remove the "index.html" portion. So for CARIN IG the STU1 version is available here: http://hl7.org/fhir/us/carin-bb/index.html

view this post on Zulip jason chow (Dec 10 2020 at 16:07):

I see. I was grabbing the continuous build rather than the published version.

view this post on Zulip Saul Kravitz (Dec 10 2020 at 18:18):

@jason chow - the canonical URL for this IG is http://hl7.org/fhir/us/carin-bb . Using that gets you the most recent published version. Until recently, this was the STU balloted version. When the IG was published it because the STU1. Each publication event has a fixed URL that is immutable over time.
http://hl7.org/fhir/us/carin-bb/STU1 will forever reference the version that was initially published as STU1. http://hl7.org/fhir/us/carin-bb/2020Feb/ will forever reference the balloted version of the IG.

view this post on Zulip jason chow (Dec 10 2020 at 18:28):

Thanks Saul. I see if you go to the fhir.org guide registry the link to the CARIN for BB link points to the canonical URL.


Last updated: Apr 12 2022 at 19:14 UTC