Modeling family members / next of kin participants and "other clinicians"

Clinical (archive)
1

Hi,

I’m checking on the RM 1.0.2 how to add information about family members (if the patient is a child or an elder), and maybe other clinicians that participate of an emergency care event (composition).

In 1.0.2 the extra participations appear at the ENTRY level as PARTICIPATION http://openehr.org/releases/RM/Release-1.0.2/docs/ehr/ehr.html#_entry_and_its_subtypes

But it is not possible to add that kind of information at the COMPOSITION level http://openehr.org/releases/RM/Release-1.0.2/docs/ehr/ehr.html#_overview_3

I think a family member that participates through all the event should be linked at the COMPOSITION level, and maybe some clinicians and nurses also need to be at that level.

  1. One idea is to use the EVENT_CONTEXT.other_context to add DV_IDENTIFIERs of DEMOGRAPHIC entities that represent family, clinicians, nurses, etc. But that is a little “hacky” since I want to reference DEMOGRAPHIC entities and DV_IDENTIFIERs are not for that, the reference should be PARTY_REF that is used from PARTY_PROXY, but can’t add PARTY_PROXY to EVENT_CONTEXT.other_context because the RM doesn’t support it, since PARTY_PROXY is not LOCATABLE.

  2. Another idea is to add ADMIN_ENTRY and do the same there, also a little hacky and has the same problem with the RM types.

  3. The third is to add all the participants to each ENTRY of the COMPOSITION, duplicating the same data by the number of entries.

  4. An idea between 2 and 3 is to add a dummy ADMIN_ENTRY with no data, that includes the participants in the ENTRY.other_participants collection.

How do others model this?

Would it be useful to propose a change to the model to support “global” participants at the COMPOSITION level?

Thanks!

2

Hi Pablo,

there is COMPOSITION.event_context.participations, of type List (see here). It seems to me that does what you need.

  • thomas
3

Hi Thomas,

Thanks, I’m on 1.0.2. Looking at the PDF, in the UML the EVENT_CONTEXT has participations. http://www.openehr.org/releases/1.0.2/architecture/rm/ehr_im.pdf page 35

But in the HTML, the UML of EVENT_CONTEXT doesn’t have participations, nor health_care_facility (that are in the PDF UML). http://www.openehr.org/releases/RM/Release-1.0.2/docs/ehr/ehr.html#_overview_3

The tabular view includes those attributes http://www.openehr.org/releases/RM/Release-1.0.2/docs/ehr/ehr.html#_event_context_class

My confusion was because I use the UML as reference and go to the tabular definition when I need detailed info.

Should be fix the UML in the generated HTML?

I detected many errors in the generated HTML for 1.0.2. I like the HTML more but I should go back to the PDF as the trusted source.

Thanks,

Pablo.

4

Hi Pablo,

Unfortunately there were various unintended changes in the UML - I think mostly not errors, but view problems like this one. The UML model was correct (the tables are generated from the same model), but I had to do some manual changes over the last couple of years to make the MaigcDraw UML diagrams clearer, and this is one, which appears only to be fixed in the current working baseline.

I’ll have a look into add in retrospective changes - it will require some Git + script magic.

  • thomas