Logical model: BeObservation

[bdc-ehealth]

see: https://github.com/hl7-be/public-health/issues/7

[costateixeira]

We must invite @ArdonToonstra for the discussions Also, see if the BE Lab Report observation becomes a child of this new profile (First impression: I think it would)

[bdc-ehealth]

this issue is handled here: #21

@ArdonToonstra , please evaluate the document provided in issue #21 and file new issues here with your remarks.

[ArdonToonstra]

@bdc-ehealth , thanks for the heads-up. I will add this as an action item for next week and try to align with healthdata.be.

[KarlienHL7Belgium]

Logical model has to be put in the IG based on the business rules document, work can be started by @bdc-ehealth

There seem to be some fields that are missing in the Business rules document (in issue 21), WG members have to list them up (@annabel-uzl ) f.e. link to encounter, together with RIZIV @annenerenhausen and Tom Andries who will meet in between and report back to the WG by 27 June

[annabel-uzl]

CareSet.BeObservation - short.docx

In the attachment above some remarks on the business rule document. I also looked at the difference with https://www.ehealth.fgov.be/standards/fhir/core-clinical/StructureDefinition-be-observation.html since they are not in sync. @KarlienHL7Belgium @annenerenhausen @bdc-ehealth I will not be present the 27th but am available for an in-between meeting :)

The remarks (but also in the document):

• RecordedDate and Recorder are not part of the international model. I know we discussed this during the WG but I cannot remember the conclusion. (Is RecordedDate the same as ‘issued’ ? - also these are not present in https://www.ehealth.fgov.be/standards/fhir/core-clinical/StructureDefinition-be-observation.html )
• Performer and patient are made mandatory. OK (but not mandatory in https://www.ehealth.fgov.be/standards/fhir/core-clinical/StructureDefinition-be-observation.html )
• ‘Autre patient concerné’ to which type of FHIR resource does this reference ?
• Category is made mandatory. We would like to see 0..1 since we don’t always know the category. (this is OK in https://www.ehealth.fgov.be/standards/fhir/core-clinical/StructureDefinition-be-observation.html )
• Date or period of observation mandatory. OK (but not mandatory https://www.ehealth.fgov.be/standards/fhir/core-clinical/StructureDefinition-be-observation.html )
• As asked in https://github.com/hl7-be/core-clinical/issues/21: What does the ‘ /’ mean in the cardinality of the value ? I don’t think Value should be mandatory as you can register a bloodpressure with components diastolic and systolic but without value.
• Status is mandatory in the international model, I don’t think we can just make it optional (OK in https://www.ehealth.fgov.be/standards/fhir/core-clinical/StructureDefinition-be-observation.html )
• Can there be only one conclusion ? Can we make it **0..*** ? And rename back to ‘interpretation’ (as in the international model) since interpretation and conclusion are not the same thing (In https://www.ehealth.fgov.be/standards/fhir/core-clinical/StructureDefinition-be-observation.html the cardinality is OK)
• BodyLocalisation : see https://github.com/hl7-be/core-clinical/issues/54
• Encounter is ommited, please add (but present in https://www.ehealth.fgov.be/standards/fhir/core-clinical/StructureDefinition-be-observation.html )
• Method is missing (but present in https://www.ehealth.fgov.be/standards/fhir/core-clinical/StructureDefinition-be-observation.html )
• ReferenceRange missing (but present in https://www.ehealth.fgov.be/standards/fhir/core-clinical/StructureDefinition-be-observation.html )
• hasMember and derivedFrom are also missing. For now this is okay, but we would like to use LHC Forms for FHIR questionnaires in our EPD and it would be nice to link the observations derived or used in the FHIR QuestionnaireResponse. So to be sure we would like to have this added (Present in https://www.ehealth.fgov.be/standards/fhir/core-clinical/StructureDefinition-be-observation.html )
• dataAbsentReason : missing, at the moment I cannot think of a use case but maybe handy to discuss in group (Present here https://www.ehealth.fgov.be/standards/fhir/core-clinical/StructureDefinition-be-observation.html )

Other remarks on CareSet document to discuss in the BWG, already mentioned in the WG

• https://github.com/hl7-be/core-clinical/issues/54 Alternation between BodyLocalisation and BodyLocation
• https://github.com/hl7-be/core-clinical/issues/53 Reference to FHIR R5 (should be R4?)
• https://github.com/hl7-be/core-clinical/issues/51 Valueset code (some guidelines)