SNOMED International is seeking input on the foreign body disorder concepts

Hi Mukesh,
Thank you for replying to the questions, this is much appreciated.

As you mentioned this is a very good example where postcoordination could be used and I agree with you. The problem we have is the lack of support in the different Information Systems that are in use in Canada. One other challenge is to ensure that the systems also provide an "equivalency" mechanism to ensure different presentations of the same idea is recognized as being equivalent semantically. I'm hoping that with more mature implementations the postcoordination expressions will be easier to integrate, use and manage.

Once the "superficial" term is removed form the current "skin" concepts, would there still be a use case to expect the "superficial" term with the foreigh body?

Thank you for your message. Regarding this topic, we had a discussion among few clinicians and health informatics professionals in British Columbia, and I am attaching a brief summary of the comments/ opinions/suggestions on this topic:

1. The proposal to reduce the “Superficial….” To “Foreign body…” and add more “Foreign body…” as needed.

2. Agree with removing “Splinter of X”., although there is perceived benefit from coding the nature of the foreign body, but there is not practical benefit from coding it. The information will appear in the body of the note, coding provides no more value (are we going to run a query for “How many patients had wood splinters verses fiberglass splinters?”


3. Although there is a value to record things as accurately as possible, but many physicians will agree with record where splinters and foreign bodies are then it would be nice if there was a way for them to record these things. If something is of no interest to a user then that user does not need to search for that item and they will never see it.

4. Our more general concern is to have been very little movement on promoting the use of post coordination. The long list of body parts where foreign objects may be could range into the thousands. We would support the recommendation that post coordination should be used. Have the concept of "125670008 |Foreign body (disorder)|" and have a second concept for the location "31636006 |Anterior chamber of eye structure (body structure) |. It is the same situation with splinter. A search in SNOMED shows that there is no entry for 'fiberglass splinter'. There are 143 matches for 'splinter'. The only type of splinter is a wood splinter. There are 61 entries for 'glass in' various body parts. We would recommend that post coordination would be able to handle this issue in an elegant manner. Do a search for the type of object, the body part, presence or absence of an infection. By the way has there been any mention of the use of post coordination with the EMR vendors? Having this capability would make many things a lot easier, yet it seems to be ignored.


5. We think, the "superficial" modifier is legacy of ICD9/10 and the heavy use of superficial wound in a classification. Makes more sense to be specific to skin when it comes to high level generic items. We agree that it is way better to post-coordinate location anatomical structure (skin, esophagus, rectum etc.) as well as body location (hand, foot, face etc.). As we are having challenges even getting vendors and providers to use SNOMED CT at all, it is probably a good idea to go easy on post-coordination but as much as possible have the functionality present. For the BC value set, it would make sense to identify key structures and locations for some pre-coordinated entries (e.g. FB eye/cornea versus skin, rectum/vagina/stomach/esophagus/penis, hand/foot. There are use cases for reporting on these structures/locations - even clinical decision support when it comes to the eye. We don't think putting time into details of the FB composition is value add now - as that would be present in the note. Reporting on fiberglass versus wood etc. seem over the top.

6. EMR vendors are providing the capability to record concepts and also with modifiers. This is what post coordination is. When recording things in a patient's problem list many EMRs provide the ability to also record modifiers like laterality, chronicity and so on. This is post coordination. The problem is that these modifiers are not used when exporting data. This is an issue related to combinatorial explosion when post coordination is not available. On the other hand there is a convenience for things that are used commonly if the user can select just one thing rather than several things. Then again users complain of doing searches and getting too many hits.

7. Without post coordination how things are entered in a problem list results in trouble when you try to apply clinical decision support rules. Consider recording "chronic kidney disease, stage 3". There will be some clinical guidelines that refer specifically to Stage 3 while other guidelines refer to chronic disease across any stage. I think that the work we are doing should be getting things lined up in such a way to facilitate the implementation of clinical decision support systems. The magic happens when EMRs implement clinical decision support BUT a proper foundation is a necessary precursor.


As we bit discussion on this topic with clinicians and have more details from them but I am making brief summary of the discussion. Please advise if you required more information on this topic from British Columbia.


Regards,


Mukesh Kashyap

The questions are:

• Within this hierarchy (125670008 |Foreign body (disorder)|) there are 95 superficial foreign body disorder concepts e.g., 874928002 |Superficial foreign body in axilla (disorder)|. Are these concepts seen as having the same meaning as the concepts that represent a foreign body in the skin e.g., 298075005 |Foreign body of skin of axilla (disorder)|?
• If these concepts are deemed to be equivalent in meaning, then the duplicates would require inactivation.


• Note, where the structure does not have skin then the relevant superficial structure e.g. mucosa, would apply.
• In doing this the proposed concepts to be kept would be |Foreign body of skin of <x> (disorder)|. Would this be acceptable?


• At present there are more |Superficial foreign body in <x> (disorder)| concepts than |Foreign body of skin of <x> (disorder) concepts|. In addition to the inactivation, we propose |Superficial foreign body in <x> (disorder)| concepts without a duplicate have the FSN updated to Foreign body of skin of <x> (disorder) or as relevant for structures without skin e.g. mucosa, to support consistency. Would this be acceptable?


• There are 65 |Splinter of <x>, without major open wound (disorder)|concepts. These are concepts are classification constructs. Has this content been used and is it currently being used by members? If not, this content will be considered for inactivation.


• Please post your comments and feedback on this forum and we will communicate them to SI.
  Thank you!
  Responses are due by 9 July 2021.