Stream: CCDA / FHIR mapping stream
Topic: C-CDA author cardinality vs.FHIR target cardinality
David Riddle (May 12 2020 at 18:12):
I apologize in advance for the redundant posting; however, I don't understand the difference between the CCDA/FHIR mapping stream and the #cda to fhir stream. I have posted this same question there.
Use case: 'Extraction' of clinical concepts (e.g., Problems) from a C-CDA document and representing them as FHIR resource instances (e.g., Conditions).
I am wondering if there are any recommended/standard rules for resolving multiple C-CDA author elements when the target FHIR resource attribute has a 0..1 or 1..1 cardinality? For example, the Model-Based Transformation Service (MBTS) DRAFT Mappings for CDS Hooks Use Case has multiple instances in which the source C-CDA author element has a cardinality of 0..* according to a given C-CDA template, but the target FHIR resource attribute for the author information has a cardinality of 0..1. Specifically the C-CDA Problem Observation template that the CCDA to CCDA-on-FHIR ProblemObservation to Condition.xlsx mapping pertains to indicates that the Problem Observation CCD Element should have its descendant author/assignedAuthor element data mapped to the FHIR US-Core-Condition.asserter; however, the cardinality of the author element within the C-CDA Problem Observation is 0..* vs. the cardinality of the FHIR US-Core-Condition.asserter which is 0..1. I don’t see anything in the CCDA to CCDA-on-FHIR ProblemObservation to Condition.xlsx mapping that indicates how to resolve these differences in cardinality. Am I missing something? Have any standard rules been recommended for selecting the ‘preferred author’ when multiple author elements are encountered in the source C-CDA document? For example:
• When mapping from C-CDA Problem Observation authors to the FHIR US-Core-Condition.asserter, always prefer authors based on type (prefer PractitionerRole over Practitioner, and either of these over Patient…) and the most recent date/time (author/time/@value).
• Some C-CDA author elements can be converted into a PractitionerRole (e.g., the author element includes an assignedAuthor/representedOrganization &/or a Healthcare Provider Taxonomy code) while others may only be converted into a Practitioner.
• When resolving between PractitionerRole authors vs. Practitioner authors, always select the ‘practitioner’ (PractitionerRole or Practitioner) with the most recent author/time/@value. If all author/time/@values are equal, prefer a PractitionerRole author over a Practitioner author. If multiple PractitionerRole authors share the same most recent date/time, prefer the one that was 1st in order of appearance under the Problem Observation.
• Always prefer an author that can be converted into a ‘practitioner’ (PractitionerRole or Practitioner) over those that can be converted into a Patient or RelatedPerson.
cc @Michelle (Moseman) Miller, @Brett Marquard, @Drew Torres
Last updated: Apr 12 2022 at 19:14 UTC
