gcangioli
commented
2 years ago Not sure I captured your doubts but I'll try to give some explanations...
- PCSP FHIR API does not necessarily represent data exactly as represented in the UI (e.g. there is not a Boolean yes/no for each of the listed sacrified organs. but you can convey what are the organs that have been sacrified...)
- In most cases PCSP FHIR API are enabled for more details of what is currently provided by the platform: the current profile allows to provide evidences about predisposition as a condition, as an observation, FamilyMemberHistory or as DocumentReference, because these are the possible resources that can be used by real EHR systems to capture them. Then the profile describes how the genetic predisposition can be conveyed using an observation.
- I don't know what are the information that you are planning to send, if they are those actually captured by the UI the easiest solution is to use the same approach followed by CINECA for the export (@davisart can you share how you are doing ? )
- .. in any case I need to update the declared mapping not limited to Condition.evidence.where(code = http://snomed.info/sct#32895009 'Hereditary disease').....
davisart
In the Diagnosis logical model, definition of the hereditaryPredisposition element states that "The SurPass system provides a list of the most common Hereditary Cancer Predisposition Syndromes or other medical conditions reported to be cancer associated": http://build.fhir.org/ig/hl7-eu/pcsp/StructureDefinition-Diagnosis-definitions.html#Diagnosis.hereditaryPredisposition
This element is mapped to an evidence element of a Primary Cancer Condition with the 'Hereditary disease' code: http://build.fhir.org/ig/hl7-eu/pcsp/StructureDefinition-Diagnosis-mappings.html . In the respective primary cancer condition profile, two codes can generally be provided in the predisposition evidence slice (one being the 'Hereditary disease' code): http://build.fhir.org/ig/hl7-eu/pcsp/StructureDefinition-Condition-primaryCancer-eu-pcsp.html . In the details field of this evidence slice, a hereditary predisposition observation might also be provided: http://build.fhir.org/ig/hl7-eu/pcsp/StructureDefinition-Observation-predisposition-eu-pcsp.html
To my understanding this observation would specify the predisposition syndrome, that is mentioned in the logical model. However, in the respective profile the code for this observation is fixed to 'Genetic predisposition' (the second possible code in the predisposition evidence slice). To me this seems inconsistent, because I would have to provide a predisposition slice with the 'Hereditary disease' code in order to be consistent with the logical model, but reference an observation with a 'Genetic predisposition' code to be able to provide details.
How would a correct example to provide a value from e.g. this valueset look like: http://build.fhir.org/ig/hl7-eu/pcsp/ValueSet-vs-orpha-predisposition-eu-pcsp.html ?
I tried several combinations when sending data via the PCSP API and was never able to have a code from this valueset selected in the respective UI field. The field is visible if 'Hereditary disease' is provided as a code in the evidence slice, but it seems in this case the observation provided in the details field is ignored. On the other hand if 'Genetic predisposition' is provided as a code in the evidence slice, the field is not visible.
I'm a little bit confused how this should work and I did not find any example that covers this case. Maybe you can clarify and/or provide a working example.