CancerEpisode [Improvement]

[fiona43]

• [ ] Biopsy done by:

  1. is marked as Mandatory into DM while is 0..1 in the FHIR IG. So, seems that there is an inconsistency between the two entries. It could impact the ETL process: if this parameter is mandatory (@notnull) then an error could be raised if this parameter is missing, otherwise (0..1) a null value is allowed.
  2. is a Custom Code into DM while in the FHIR IG, it is modeled as Observation.performer (Reference Organization). So, we can use only the display item to persist the information? here is an example: OrganizationCustomCode.display -> (OrganizationFHIR) Reference.display

• [x] Biopsy mitotic count

  1. there is also a misalignment with the OHDSI code in the DM.

• [x] Plasmatic EBV DNA at baseline

  1. there is also a misalignment with the OHDSI code in the DM.

• [x] HPV status

  1. there is also a misalignment with the OHDSI code in the DM.

• [x] CRP - C reactive protein

  1. there is also a misalignment with the OHDSI code in the DM.

• [x] hnClassification (CancerEpisode Model to guide map)

  1. there is no mention of the histologyMorphologyBehavior (s. screenshot here below)

• [x] CancerEpisode.radiotherapyInducedSarcoma (Radiotherapy induced sarcoma)

  1. in the CancerEpisode Model to guide map is a valueCodeableConcept (s. screenshot here below)
  2. in the Condition: Primary Cancer is a valueCodeableConcept and a Reference (s. screenshot here below) Could it be a typo?

• [ ] Histology subgroup - Subsite group

  1. For each sarcoma/h&n pathology (listed in the DM) there is a Condition: Primary Cancer with its code and a specific subsite. In the case of two or more sarcoma/h&n pathologies, currently, in the DM no rule, label, or specification defines the association between the pathology to its subsite. In this scenario, when a CSV is received, a Condition: Primary Cancer should map the specific pathology (Condition.code) with all body sites present in the CSV. Another aspect to consider is that it could be not compliant with the clinical information and the quality check. So, the question is how the Condition Primary Cancer and the ETL can manage the relationship between the condition.code and the condition.bodysite?

[gcangioli]

Biopsy done by. we need to identify what it is actually meant as Mandatory in the model..

[unai-zulaika]

Regarding Biopsy mitotic count. The response from INT:

[ENG] check the code - according to the SNOMED code in the FHIR IG, a OHDSI code could be: "4155612", "Mitotic count score" [INT] this is not a good code because the HPF are not scpecified. for different diseases you use different HPF (eg, for sarcoma 10 HPF, for GIST 50 HPF). thus without this info you cannot compare data.

I will keep the original code from the DM, maybe we should use that also in the FHIR IG.

Do we agree @fiona43 @gcangioli?