FHIR Chat · Ordinal without system

Stream: implementers

Topic: Ordinal without system

Alexander Henket (May 20 2019 at 12:37):

There are many questionnaires out there in the real world where you know what score exists, you know what each score means, but that you do not have a code/system for.

This gives specification headaches in the Questionnaires because this would be illegal logically:

<option>
   <valueCoding>
     <extension url="http://hl7.org/fhir/StructureDefinition/questionnaire-ordinalValue">
        <valueDecimal value="1" />
     </extension>
     <display value="Verscheidene dagen" />
  </valueCoding>
</option>

Or would it not be? How to go about when you encounter answer lists without proper codings

@Theo Stolker @Arianne van de Wetering @Lloyd McKenzie

Grahame Grieve (May 20 2019 at 12:40):

why use a coding?

Alexander Henket (May 20 2019 at 12:42):

Without coding no display? How would I tell what a score means and offer the Questionnaire renderer the right data for display?

Alexander Henket (May 20 2019 at 12:46):

Alexander Henket (May 20 2019 at 12:51):

Presumably this could be an alternative? If the extension had not been specified for valueCoding only:

<option>
   <valueString value="Verscheidene dagen">
     <extension url="http://hl7.org/fhir/StructureDefinition/questionnaire-ordinalValue">
        <valueDecimal value="1" />
     </extension>
  </valueString>
</option>

Grahame Grieve (May 20 2019 at 13:01):

is it? I thought we enabled it for other things too

Grahame Grieve (May 20 2019 at 13:02):

we did.... Questionnaire.item.answerOption

Alexander Henket (May 20 2019 at 13:02):

Ah... that is R4

Grahame Grieve (May 20 2019 at 13:10):

yes that's why we changed it

Alexander Henket (May 20 2019 at 13:48):

What's the best way to pre-adopt this R4 extension in STU3? Do I need to recreate it under a new local canonical? That's what I normally do

Alexander Henket (May 20 2019 at 13:49):

I cannot 'just' call it as it is not in the STU3 Core package we depend on

Grahame Grieve (May 20 2019 at 13:50):

you can just use it. make a task to figure it out in the validator

Alexander Henket (May 20 2019 at 13:51):

@Richard Ettema : suppose the validator receives an update for resolving an R4 extension from STU3 context: how would that work for Touchstone?

Alexander Henket (May 20 2019 at 14:00):

GF#22482 Validator support in STU3 for R4 extensions

Alexander Henket (May 20 2019 at 14:33):

@Arianne van de Wetering / @Theo Stolker I have created the pre-adopt for the short term based on a copy/paste from R4 where I updated .version / .fhirVersion / .context to match FHIR STU3. I've applied it in nl-core-questionnaire

Alexander Henket (May 20 2019 at 14:34):

This solves our issue in a way that should not bite with a longer term solution based on the very same extension

Richard Ettema (May 20 2019 at 17:29):

@Alexander Henket, I did a quick test where I added the R4 Questionnaire extension for ordinalValue to an R3 Questionnaire instance, ran the validator against it and it came back with just an informational issue:

  <issue>
    <extension url="http://hl7.org/fhir/StructureDefinition/operationoutcome-issue-source">
      <valueCode value="InstanceValidator"/>
    </extension>
    <severity value="information"/>
    <code value="structure"/>
    <details>
      <text value="Unknown extension http://hl7.org/fhir/StructureDefinition/ordinalValue"/>
    </details>
    <diagnostics value="Unknown extension http://hl7.org/fhir/StructureDefinition/ordinalValue"/>
    <location value="Questionnaire.item.item[2].item[3].option[1].extension (line 83, col81)"/>
  </issue>

Alexander Henket (May 20 2019 at 20:07):

That doesn't sound too bad. I guess most magic has to come from Questionnaire(Response) implementers that need to add support for the ordinal value upon processing responses into total scores

Grahame Grieve (May 20 2019 at 21:43):

SDC is commenting on that

Last updated: Apr 12 2022 at 19:14 UTC

Main menu

FHIR Chat · Ordinal without system · implementers