Uploaded image for project: 'FHIR Specification Feedback'
  1. FHIR Specification Feedback
  2. FHIR-34384

Why is SDOH Assessment's Observation.component.code MS? Why is Observation.component?

    XMLWordPrintableJSON

Details

    • Icon: Change Request Change Request
    • Resolution: Persuasive with Modification
    • Icon: Medium Medium
    • US Core (FHIR)
    • 4.0.0
    • Cross-Group Projects
    • STU
    • USCoreObservationPatient
    • Hide

      Background

      See commenters Description:

      The commenter is referring to this Observation Profile:

      Resource Profile: US Core Screening Response Observation Profile

      ...

      This profile sets minimum expectations for the Observation resource to record, search, and fetch retrieve observations that represent the questions and responses to form/survey and assessment tools such as the Protocol for Responding to and Assessing Patients’ Assets, Risks, and Experiences (PRAPARE) Survey. This profile encompasses single, multipart, and derived responses. ...

      ...

      Each Observation must have:

      1. ..

      Each Observation must support:

      1. ...
      2. component results

      Note to Balloters

      We are seeking feedback on the inclusion of Observation.component as a Must Support element for systems to record multiple answers for a screening question in a single Observation. Based on how broadly this is supported in current implementations, we may remove the constraint on Observation.component and recommend multiple Observations.

      As the commenter correctly points out it component results are marked as Must Support  to support multi-select answers.

      Prior Art

      The SDOH Clinical Care IG enables Enabling Survey Instruments by standardizing the capture, coding and output of SDOH risk surveys. This guide provide the necessary framework for preserving "the context for the individual parts of a single screening instrument." It leverages the SDC guide to extract the specific survey results into FHIR Observation and other resources.


      These design decisions were made after soliciting feedback from ONC, the Gravity project SDOH clinical Care IG development team, and feedback from EHR vendors and FindHelp.org, a provider of social service assistance tool used by several EHRs. After considering other options, the decision was made to follow these SDOH models with changes to simplify the structure terminology:

      http://build.fhir.org/ig/HL7/fhir-sdoh-clinicalcare/StructureDefinition-SDOHCC-ObservationScreeningResponse.html

       http://build.fhir.org/ig/HL7/fhir-sdoh-clinicalcare/StructureDefinition-SDOHCC-ObservationAssessment.html

      Reasoning

      We agree with the commenter that  there is limited testing and real world experience is lacking this approach is untested and we are sensitive to burden to implementers and will remove the  component structure from the profile.

      1) We disagree with the commenter assertion that "Observation.component was changed to MS late in the IG development process for entirely arbitrary reasons, having nothing to do with interoperability or usage of complex assessment answers" and "Gravity isn't even doing so".

      2) Regarding  " What justification or research or implementation testing has US Core done to validate a real-world need for multi-select SDOH screening answers?":

      Based on ONC’s objectives, The CGP WG decided these functional requirements are in scope:

      1. Every US Core system SHOULD be able to represent multi-question SDOH surveys (e.g., instruments like PRAPARE)
      1. Every US Core system SHOULD be able to represent “check all that apply” SDOH questions (e.g., “In the past year, have you or any family members you live with been unable to get any of the following when it was really needed? Check all that apply.”)

      (https://confluence.hl7.org/pages/viewpage.action?pageId=90342002)

      Proposed Changes

       See proposal for FHIR-35364

      Show
      Background See commenters Description: The commenter is referring to this Observation Profile: Resource Profile: US Core Screening Response Observation Profile ... This profile sets minimum expectations for the  Observation  resource to record, search, and fetch retrieve observations that represent the questions and responses to form/survey and assessment tools such as the  Protocol for Responding to and Assessing Patients’ Assets, Risks, and Experiences (PRAPARE) Survey . This profile encompasses single, multipart, and derived responses. ... ... Each Observation must have: .. Each Observation must support: ... component results Note to Balloters We are seeking feedback on the inclusion of Observation.component as a Must Support element for systems to record multiple answers for a screening question in a single Observation. Based on how broadly this is supported in current implementations, we may remove the constraint on Observation.component and recommend multiple Observations. As the commenter correctly points out it component results are marked as Must Support  to support multi-select answers. Prior Art The  SDOH Clinical Care IG  enables Enabling Survey Instruments by standardizing the capture, coding and output of SDOH risk surveys. This guide provide the necessary framework for preserving "the context for the individual parts of a single screening instrument." It leverages the SDC guide to extract the specific survey results into FHIR Observation and other resources. These design decisions were made after soliciting feedback from ONC, the Gravity project SDOH clinical Care IG development team, and feedback from EHR vendors and FindHelp.org, a provider of social service assistance tool used by several EHRs. After considering other options, the decision was made to follow these SDOH models with changes to simplify the structure terminology: http://build.fhir.org/ig/HL7/fhir-sdoh-clinicalcare/StructureDefinition-SDOHCC-ObservationScreeningResponse.html   http://build.fhir.org/ig/HL7/fhir-sdoh-clinicalcare/StructureDefinition-SDOHCC-ObservationAssessment.html Reasoning We agree with the commenter that  there is limited testing and real world experience is lacking this approach is untested and we are sensitive to burden to implementers and will remove the  component structure from the profile. 1) We disagree with the commenter assertion that "Observation.component was changed to MS late in the IG development process for entirely arbitrary reasons, having nothing to do with interoperability or usage of complex assessment answers" and "Gravity isn't even doing so". 2) Regarding  " What justification or research or implementation testing has US Core done to validate a real-world need for multi-select SDOH screening answers?": Based on ONC’s objectives, The CGP WG decided these functional requirements are in scope: Every US Core system  SHOULD  be able to represent multi-question SDOH surveys (e.g., instruments like PRAPARE) Every US Core system  SHOULD  be able to represent “check all that apply” SDOH questions (e.g., “In the past year, have you or any family members you live with been unable to get any of the following when it was really needed? Check all that apply.”) ( https://confluence.hl7.org/pages/viewpage.action?pageId=90342002 ) Proposed Changes  See proposal for  FHIR-35364
    • Floyd Eisenberg / Bob Dieterle: 8-0-2
    • Enhancement
    • Compatible, substantive

    Description

      http://build.fhir.org/ig/HL7/US-Core/StructureDefinition-us-core-observation-screening-response.html (as of 2021-11-26). 

      Observation.component was changed to MS late in the IG development process for entirely arbitrary reasons, having nothing to do with interoperability or usage of complex assessment answers: https://chat.fhir.org/#narrow/stream/179175-argonaut/topic/Must-support.20on.20child.20properties.20on.20a.20parent.20that.20is.20not.20MS

       

      Gravity's Observation Assessment profile doesn't MS component.code: http://build.fhir.org/ig/HL7/fhir-sdoh-clinicalcare/StructureDefinition-SDOHCC-ObservationAssessment.html

      And Gravity's Observation Screening response does: http://build.fhir.org/ig/HL7/fhir-sdoh-clinicalcare/StructureDefinition-SDOHCC-ObservationScreeningResponse.html 

       

      1) Why is US Core forcing support for multi-select answers for SDOH screenings when Gravity isn't even doing so?

      2) What justification or research or implementation testing has US Core done to validate a real-world need for multi-select SDOH screening answers?

      Attachments

        Activity

          People

            Unassigned Unassigned
            Isaac.Vetter Isaac Vetter
            Isaac Vetter
            Watchers:
            2 Start watching this issue

            Dates

              Created:
              Updated:
              Resolved: