Stream: Orders and Observation WG
Topic: Vital Signs IG
Javier Espina (Aug 04 2021 at 17:10):
https://jira.hl7.org/browse/FHIR-29050 was resolved by explanation - basically clarifying that HeadCircumference is optional within the panel (as stated in the Vital Signs IG and in the LOINC panel itself). That clarification made me however wonder whether http://hl7.org/fhir/us/vitals/2020Sep/StructureDefinition-VitalSignsPanel.html#tabs-diff is correct. All Observation.hasMember slices show a cardinality of "1..*. Doesn't that mean that they have to be present? [Note that this is a new comment, now that it is clear that all panel members are optional]
Eric Haas (Aug 05 2021 at 01:10):
at least one has to be present. (in other words - no empty panel)
Javier Espina (Aug 05 2021 at 13:47):
Exactly - the panel should not be empty while all its members are optional. But after consulting http://build.fhir.org/profiling.html#slice-cardinality (and the entire section explaining slicing, which is a rather advanced part of FHIR) I would expect those requirements to be expressed as:
- hasMember:vitalSignsElements 1..* (Slice Definition)
-hasMember:OxygenSaturationArterialBloodPulseOx 0..*
-hasMember:.... 0..*
Or am I misunderstanding how to express slicing?
Lloyd McKenzie (Aug 05 2021 at 14:57):
In theory you could have an empty panel if work was still in progress - it just couldn't have a status of 'complete'
Eric Haas (Aug 05 2021 at 17:02):
I mispoke earlier, because I assumed you were referencing the FHIR base profile. I don't know what that slicing pattern means, either its empty or all the slices ?? which I think this is overly prescriptive approach is unworkable in the field.
Javier Espina (Aug 09 2021 at 12:52):
@Eric Haas - yes, also I find it a bit confusing... Moreover I just realized that http://hl7.org/fhir/R4/vitalspanel.html (with which the Vital Signs IG should be aligned) describes it in a different way - without slicing and requiring 1..* "hasMember". I find this approach easier to understand.
Richard Townley-O'Neill (Aug 10 2021 at 22:45):
I've posted some questions about the VitalSIgns profiles to the VitalSIgns stream, but is is a rather dead stream. I now think I should have asked them in this OO stream. Could someone move them to this stream for me please?
They are
- https://chat.fhir.org/#narrow/stream/235632-Vital-Signs/topic/VitalSigns.20slice.20category.3AVSCat.20is.20misconstrained
- https://chat.fhir.org/#narrow/stream/235632-Vital-Signs/topic/Must.20Support.20on.20VitalSigns.2Ecomponent.20in.20wrong.20place
- https://chat.fhir.org/#narrow/stream/235632-Vital-Signs/topic/Use.20of.20Must.20Support.20in.20VitalSigns.20varies
Javier Espina (Sep 21 2021 at 17:41):
@Nathan Davis - any views on the above discussion?
Javier Espina (Sep 21 2021 at 17:47):
Another thing I just discovered. The Vital Signs IG BP profile (http://hl7.org/fhir/us/vitals/2020Sep/StructureDefinition-BloodPressurePanel.html) makes the the systolic comppnent mandatory and the diastolic component optional. Whereas I see a practical reason in that, it breaks what the BP core profile (http://hl7.org/fhir/R4/bp.html - of mandatory application). The latter makes both components mandatory. BTW, the US Core BP profile does the same as BP core profile. @Nathan Davis
Javier Espina (Sep 22 2021 at 07:48):
Javier Espina said:
Another thing I just discovered. The Vital Signs IG BP profile (http://hl7.org/fhir/us/vitals/2020Sep/StructureDefinition-BloodPressurePanel.html) makes the the systolic comppnent mandatory and the diastolic component optional. Whereas I see a practical reason in that, it breaks what the BP core profile (http://hl7.org/fhir/R4/bp.html - of mandatory application). The latter makes both components mandatory. BTW, the US Core BP profile does the same as BP core profile. Nathan Davis
This specific discussion point continues at https://chat.fhir.org/#narrow/stream/179166-implementers/topic/Vital.20Signs.20Profile.20-.20clarity.20on.20applicability
Last updated: Apr 12 2022 at 19:14 UTC
