[Condition] !!! ValueSet 'Code' BeProblem

[annabel-uzl]

I reference to issues #45 and #22
We would like to define a valueset for BeProblem code element such that we are sure mapping on the right codes (EPD codes to code of the valueset).

At the moment we are looking at SNOMED codes under the 404684003 (Clinical finding) hierarchy or '160245001 No current problems or disability'. We try to keep it as narrow as possible in the hope that what we define on BE level is more broad ;)

Opinions / Concerns / ...

[bdc-ehealth]

WG: we will organize an additional meeting with more clinicians. The clinicians in the meeting should be aware of the concept of structured registration, SNOMED CT, and the impact on the UI and the backend of an HIS. Anne will organize the meeting and take suggestions from the participants in the WG. Philippe Vandenbergh and Filip Veldeman will limit the group to a workable audience.

[AlexisVZ-MDS]

I checked with my reference GP for terminology and structure here (Manu Berquin), and she thinks 'finding' codes should be possible in the 'problem' careset, o.a. when there is no diagnostic/condition yet.

[annabel-uzl]

@bdc-ehealth I would like to join the meeting as well
We will try to gather some more examples (the ones given during the meeting are Disorientated in place (finding) and Incontinence of feces (finding))
About the impact on the UI I'm not following. FHIR is used to exchange data, outside of openEHR I don't know of any EPD that has FHIR as a data model on the backend...

[antoine-earlytracks]

Hello,
@bdc-ehealth Could you invite us to too to this meeting? At EarlyTracks we have built and improved ValueSets for problems and specialties for a few years now and we would gladly share our experience. These ValueSets are used by a large amount of institutions in Belgium already.

@annabel-uzl I think what Bart means is that the UI to search and select the right code is a key factor to adoption. This does not depend on the data model used under the hood (which should not be FHIR imo).
We share that opinion based on our experience in the field, this had lead us to make a "context aware recommendation engine" that suggest codes based on user text search but also a series of additional context elements such as the specialty, past user behavior and very soon, the existing patient data. The system also learns to provide recommendation with the right level of granularity (e.g. we start with Diabetes Type 2 and not a super complicated version of it).

[bdc-ehealth]

@annenerenhausen ,

Medispring, UZ Leuven and EarlyTracks would like to participate in the meeting. Can you invite them?

[antoine-earlytracks]

Thank you Bart. Please also include @rosen Cren ***@***.***> from our team to join the meeting. Best regards, Antoine

…

On Mon, Jun 3, 2024 at 11:55 AM Bart Decuypere (eHealth) < ***@***.***> wrote: @annenerenhausen <https://github.com/annenerenhausen> , Medispring, UZ Leuven and EarlyTracks would like to participate in the meeting. Can you invite them? — Reply to this email directly, view it on GitHub <#50 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AW3IVNJD7Y65B4KU3PPUMSTZFQ4QZAVCNFSM6AAAAABHK2ROXWVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDCNBUG43TKNJUHA> . You are receiving this because you commented.Message ID: ***@***.***>