Content Managers Advisory Group

Page tree


Summary

The purpose of this briefing note (see attached) is to inform the SNOMED International Community of Practice and select Advisory Groups and Project Groups of the intention to inactivate the Navigational concept hierarchy (<<363743006 |Navigational concept (navigational concept)|) and to seek input from the Community of Practice on impact of that change. 

Comments on the proposal should be submitted to the Chief Terminologist, Dr Jim Case, by June 30, 2023.


Relevant documents

  File Modified
PDF File Briefing note - Inactivation of Navigational concept hierarchy v1 20230326.pdf 2023-Jun-13 by Cathy Richardson


Actions

Final outcome: 


  • No labels

4 Comments

  1. Katrien Scheerlinck

    Hi all,


    In my opinion, inactivating the Navigational concept hierarchy will not impact SNOMED CT use in Belgium as we avoid using these concepts.


    Kind regards, Katrien

  2. Elze de Groot

    Same in the Netherlands! We also avoid this content and need to explain to people why not use them (smile)

    Kind regards,

    Elze

  3. Ed Cheetham

    The arguments in this paper miss the point about the potential value of 'navigational concepts'.

    As explained in the glossary entry for navigational hierarchy,  navigational concepts exist to assist in the convenient organisation of content for navigation and selection - not for data entry and not to participate in formal definitions.

    The shortcomings of the pure subtype hierarchy for data selection and entry, and the merits of using navigational hierarchies (including the use of navigational concepts) is explained in the Search and Data Entry Guide. The SNOMED CT-native technical mechanism to achieve this is presented in both the Release File Specification and the Reference Set Practical Guide.

    The statement that "...In the past, there was an indiscriminate move of concepts in and out of the navigational concept hierarchy..." may well be true, but the earlier assertion (quoted in the briefing paper) that "...Because the Reference Set mechanism is now available, there is no longer a need for navigational concepts in the International Release..." is an odd one.

    Take, for example, the notion of 28772006 |Arthropod-borne infectious disease (navigational concept)| and a number of related ('flea', 'tick', 'louse', 'mite' etc. navigational concepts).

    If they are allowed to remain in the international release:

    1. They can be used by any member to construct a navigational hierarchy for data entry relating to the broad notion of arthropod-borne infections.
    2. It can be used by a third party derivative provider to make one or more comparable navigational hierarchy products which can be marketed to multiple SNOMED CT user jurisdictions to aid with data entry.

    If it is removed from the international release:

    1. Any member wishing to construct a navigational hierarchy for data entry relating to the broad notion of arthropod-borne infections must create their own navigational concept (making extra work for the member).
    2. Any third party derivative provider must create its own navigational concepts and require any potential user to accommodate such codes (making extra work for the derivative provider and the member.

    In either situation:

    1. the need to disallow the entry of navigational concepts into records remains whether they are internationally distributed, developed locally or developed by a third party provider
    2. agreement on what constitutes an arthropod-borne infection may be difficult to tie down with complete agreement, but supporting a requirement to present infections diseases to clinicians who expect to see things organised this way may be the difference between the product being acceptable or unacceptable to that community.
    3. forcing the creation of navigational concepts with 'broadly shared intent' to take place multiple times locally is inefficient and an extra burden on members.

    Yes, there is work to be done disentangling problems with historical UK data, but this is independent of the question of whether there are internationally useful navigational concepts.

    1. Matt Cordell

      I think the idea of this content being all "navigational concepts" is a misnomer.
      I said elsewhere, there's obvious navigation content like 312102009 |Types of drugs N-S| that are unecessary. These are holdovers from tree-view UI - which is probably not an appropriate UI component nowdays (much of the time)
      Tree views are good for "discovery". But comboboxes are better for (search) - It's better to simply search a "the set of drugs" starting with "R".

      A lot I think are legitimate concepts, but they're just hard to sufficiently define in the concept model.

      It's hard to understand why Arbovirus (navigational concept) would be excluded.
      But we have Arbovirus infection (disorder) and Antibody to Arbovirus (substance)

      Or there's "legitimate" concepts that look similar, and any distinction is almost historical luck

      Abnormal biochemistry finding (navigational concept) vs Abnormal blood test (finding)

      Topical agent adverse reaction (disorder) vs  Barrier preparation adverse reaction (navigational concept)

      Rather than sink effort into trying to remodel/identify all these. I think it's reasonable to inactivate everything and re-introduce as required in the proper content areas.
      There's too much variation in the content to solve it any one way.

      But I agree, it'd be ideal to centralise this rather than each member (or vendor) having to create their own solutions.

      The organism categories in particular would be useful.