FHIR Chat · Result/Score in QuestionnaireResponse · questionnaire

Stream: questionnaire

Topic: Result/Score in QuestionnaireResponse


view this post on Zulip Raheel Sayeed (Feb 27 2018 at 21:15):

Is there a way to include Result/Score that could be derived score based on some logic from Questionnaire and the score being included in QuestionnarieResponse.

As it stands: My understanding is that such a questionnaire would have two outputs -> QuestionnaireResponse, Observation. Is this the ideal way to go about it?

view this post on Zulip Lloyd McKenzie (Feb 27 2018 at 21:48):

So what you're wanting is someone to fill out a score, the information be analized to determine a score and the score then reported as a result? That's certainly possible. It's not really that the Questionnare has two results - the Questionnaire is a definition containing a list of questions. The two results would/could be associated with the request to complete the Questionnaire - as a ServiceRequest and/or Task. The logic to calculate the score and display the score could be included in the Questionnaire design. As could the logic to allow extraction of either that one final "answer" or possibly all answers into Observation or other appropriate resources.

view this post on Zulip Raheel Sayeed (Feb 28 2018 at 19:46):

So what you're wanting is someone to fill out a score, the information be analized to determine a score and the score then reported as a result? That's certainly possible. It's not really that the Questionnare has two results - the Questionnaire is a definition containing a list of questions. The two results would/could be associated with the request to complete the Questionnaire - as a ServiceRequest and/or Task. The logic to calculate the score and display the score could be included in the Questionnaire design. As could the logic to allow extraction of either that one final "answer" or possibly all answers into Observation or other appropriate resources.

The Questionnaire is pretty simple to address in my case. It has a bunch of questions with "Choice" type answers.
The Scoring of the questionnaire based on each item response is based on a complicated algorithm (in my case Calibrated IRT). As such it wouldn't be possible to store the scoring-logic in Questionnaire. (Im curious though, if the logic were simple math, where would it be in Questionnaire?)

But lets say that I do arrive on a score number. The question is: Is it wise, to store that score in the QuestionnaireResponse or should I create an Observation and have that be linked with the Questionnaire & QuestionnaireResponse in some way, Bearing in mind that the score is not the user-generated response but a result calculated by the requesting party (in this case the hospital).

view this post on Zulip Lloyd McKenzie (Feb 28 2018 at 20:53):

The logic would be expressed in either FHIRPath or CQL which are pretty sophisticated languages. We'd be happy to have your logic as an example to test with if you wish :) The logic would be expressed in an extension. If the score is something that would ideally be printed on the completed form, it can be in the QuestionnaireResponse as a read-only field that is calculated as the form is filled out and/or when the form is complete. If it's a separate result determined by someone else, then an Observation that has a "derivedFrom" relationship to the QuestionnaireResponse would be appropriate.

view this post on Zulip Raheel Sayeed (Feb 28 2018 at 21:40):

If the score is something that would ideally be printed on the completed form, it can be in the QuestionnaireResponse as a read-only field that is calculated as the form is filled out and/or when the form is complete.

Thanks @Lloyd McKenzie
I really like some example as to how I could include a read-only field in the QuestionnaireResponse would that be a read-only questionnaireItem in Questionnaire that is carried over with a score to QuestionnaireResponse?

view this post on Zulip Brian Postlethwaite (Feb 28 2018 at 22:13):

You just use the readOnly property on Questionnaire.item to mark the property as non editable in the QR

view this post on Zulip Raheel Sayeed (Mar 01 2018 at 14:13):

Great. thanks!

view this post on Zulip Alex Goel (Mar 01 2018 at 17:58):

Curious what kind of scores are we talking about here? Would it be a score like a TNM or BI-RADS score? Because in those cases I don't think they can be read only fields. Often we have physicians elevate or lower a score based on things that are difficult to calculate. For example a TI-RADS score is dictated mostly on measurement and a few other factors and a measurement, but a radiologist may elevate a score based on something that is in a grey area, such as family history. Therefore the a score like this can only be used as a suggestion to a radiologist and if it's in a read-only field that could be a problem for the clinical content

view this post on Zulip Paul Lynch (Mar 01 2018 at 18:06):

I think it is up the Questionnaire designer to set item.readOnly, or not, as appropriate for the case.

view this post on Zulip Alex Goel (Mar 01 2018 at 18:10):

I think it is up the Questionnaire designer to set item.readOnly, or not, as appropriate for the case.

Sounds reasonable

view this post on Zulip Raheel Sayeed (Mar 11 2018 at 07:32):

Curious what kind of scores are we talking about here? Would it be a score like a TNM or BI-RADS score? Because in those cases I don't think they can be read only fields. Often we have physicians elevate or lower a score based on things that are difficult to calculate. For example a TI-RADS score is dictated mostly on measurement and a few other factors and a measurement, but a radiologist may elevate a score based on something that is in a grey area, such as family history. Therefore the a score like this can only be used as a suggestion to a radiologist and if it's in a read-only field that could be a problem for the clinical content

It is a calculated Pain score. PROMIS-NIH.


Last updated: Apr 12 2022 at 19:14 UTC