Stream: questionnaire
Topic: UPdating my Questionnaire Editor
Grahame Grieve (Sep 07 2018 at 07:57):
I'm working on releasing an updated FHIR toolkit.
Grahame Grieve (Sep 07 2018 at 07:57):
I'm looking through the questionnaire extensions - we have rather a lot of them
Grahame Grieve (Sep 07 2018 at 07:57):
I do not understand the definitions of
11179-permitted-value-conceptmap and 11179-permitted-value-valueset
Grahame Grieve (Sep 07 2018 at 07:58):
I do not know what the difference between cqf-optionCode and Questionnaire.code is
Grahame Grieve (Sep 07 2018 at 07:58):
I suspect that questionnaire-baseType no longer has value (and questionnaire-fhirType, which I'm not sure how it differs from)
Grahame Grieve (Sep 07 2018 at 07:59):
the definition on choiceColumn needs lots of work.
Grahame Grieve (Sep 07 2018 at 08:00):
is questionnaire-deMap still relevant?
Grahame Grieve (Sep 07 2018 at 08:02):
I don't understand questionnaire-enablewhen - it seems like it's in the wrong place
Grahame Grieve (Sep 07 2018 at 10:07):
I thought I had got us down to one ordinal value extension....
Grahame Grieve (Sep 07 2018 at 10:09):
does anyone have any idea at all what http://hl7.org/fhir/StructureDefinition/iso21090-verification is about?
Lloyd McKenzie (Sep 07 2018 at 17:42):
11179-permitted-value is tied to the 11179 notion of separating the permitted value from the domain value. Essentially one is what's selected and one is what's stored. It seems ludicrous to me, but the 11179 folks were quite firm that it was needed, so an extension it was.
Lloyd McKenzie (Sep 07 2018 at 17:42):
I'll let @Bryn Rhodes speak to cqf-optionCode
Lloyd McKenzie (Sep 07 2018 at 17:44):
questionnaire-baseType was yours for roundtripping your profile-generated Questionnaire instances. I think it was relevant if you had a polymorphic complex type. But you're best positioned to know if it's still relevant
Lloyd McKenzie (Sep 07 2018 at 17:44):
Feel free to submit a changeRequest on choiceColumn
Lloyd McKenzie (Sep 07 2018 at 17:44):
questionnaire-deMap can go - change request for that too.
Lloyd McKenzie (Sep 07 2018 at 17:46):
questionnaire-enablewhen is on item - where it should be. It needs to be a sibling to the existing enableWhen - you have one or the other. It can't be inside enableWhen because when you use it, the mandatories aren't relevant.
Lloyd McKenzie (Sep 07 2018 at 17:46):
I think we're using only one ordinal value extension. We need a change proposal to remove the others.
Lloyd McKenzie (Sep 07 2018 at 17:54):
21090-verification - my best guess would be II.reliability, but it's super poorly defined, has the wrong context, etc. I'm fine with us yanking it.
Bryn Rhodes (Sep 07 2018 at 18:12):
cqf-optionCode was my proposal to address tracker GF#12485, but it hasn't had activity from the submitter in a while.
Bryn Rhodes (Sep 07 2018 at 18:13):
The intent was to provide a way to support creating resources from questionnaire responses, but I think the new extensions are a better approach to addressing that.
Grahame Grieve (Sep 07 2018 at 21:08):
GF#17822, GF#17823, GF#17824, GF#17825, GF#17826, GF#17827, GF#17828
Grahame Grieve (Sep 07 2018 at 21:12):
questionnaire-enablewhen is on item - where it should be
where it can only reference values from itself....
Grahame Grieve (Sep 07 2018 at 21:12):
we have to revisit this and use the active questionnaire framework for it
Lloyd McKenzie (Sep 07 2018 at 21:19):
enableWhen can reference variables declared further upstream. The actual "enable" still needs to appear on the item that it's enabling
Lloyd McKenzie (Sep 07 2018 at 21:19):
And we can't use activeQuestionnaire for this
Lloyd McKenzie (Sep 07 2018 at 21:19):
They're two different things
Lloyd McKenzie (Sep 07 2018 at 21:19):
The source for activeQuestionnaire might well be driven by enableWhens
Lloyd McKenzie (Sep 07 2018 at 21:21):
activeQuestionnaire is just wrapping a black box around the complexity for the client. It doesn't get rid of the need to define the complexity when authoring complex questionnaires and passing them around on the maintenance side. And activeQuestionnaire only works in environments that can support talking to an external server. We can't enforce that as the only client architecture - though I suspect it will be the most popular.
Grahame Grieve (Sep 07 2018 at 21:22):
no idea what you are talking about.
Grahame Grieve (Sep 07 2018 at 21:22):
enableWhen can reference variables declared further upstream - that's what I meant by 'active questionnaire framework'
Grahame Grieve (Sep 07 2018 at 21:22):
the definition/documentation needs to describe that
Lloyd McKenzie (Sep 07 2018 at 21:24):
Sorry, I thought you meant adaptive questionnaires
Lloyd McKenzie (Sep 07 2018 at 21:24):
If the documentation isn't clear, feel free to do another change request :)
Grahame Grieve (Sep 07 2018 at 21:29):
already created it
Ville Lindholm (Sep 27 2018 at 11:09):
Speaking of reducing ordinalValue to only one extension: would it be possible to expand the context of the extension to places other than option.valueCoding? Basically, to option.value[x], since I think it is not unreasonable to assign "points" to any type of question.
Grahame Grieve (Sep 27 2018 at 11:19):
I think that's part of what we decided to do
Last updated: Apr 12 2022 at 19:14 UTC
