Page tree


0900 - 1230 BST

0800 - 1130 UTC 

Zoom Meeting Details

Jim Case is inviting you to a scheduled SNOMED International Zoom meeting. 

Topic: SNOMED EAG Face to face meeting London
Time: Apr 16, 2024 09:00 London

Join from PC, Mac, Linux, iOS or Android:
    Password: 848732

Meeting Files:

Meeting minutes:

The call recording is located here.


  • Obtain consensus on agenda items

Discussion items



1Call to order and role call

This meeting is being recorded to ensure that important discussion points are not missed in the minutes.  The recording will be available to the SNOMED International community.  Joining the meeting by accepting the Zoom prompt declares that you have no objection to your comments being recorded

  • Recording of meeting approved by participants.

Conflicts of interest and agenda review

None recorded

3EAG item status since October 2023

Here is the status of items brought to the EAG since October 2023:

  • Modeling of "No known X" - Ongoing. revision of Context values has not progressed.  Consensus on approach not yet reached
  • Transplantation vs. grafting and Terming of Transplantation concepts - solid tumors have been remodeled, grafts and tissue transplantation are in progress
  • Mechanical complication of device - Inactivation of grouper concepts, as recommended by the EAG is underway
  • Evaluation procedures to Observable Entity - Based on community feedback, this proposal has been retracted.  Evaluation procedures will remain, although new laboratory evaluation procedures will not be added.
  • Replacement of "Surgical approach" with "Procedure approach - testing for impact analysis has been completed
  • MRCM change and revision of Physical object hierarchy - request for access to a non-promotable project in the SNOMED authoring platform is in process
  • Updating of Neuroendocrine tumors.  Part of Histology update project - work completed
  • Inactivation of "Role groupers" from substances and products - based on community feedback, this proposal has been retracted.  Inactivation of newly constructed Product role groupers is out for community comment
  • Adverse reaction usage - Concepts that have a CAUSATIVE AGENT value from the 373873005 |Pharmaceutical / biologic product (product)| hierarchy have been remodeled and promoted.  A spreadsheet with concepts using CAUSATIVE AGENT values from the substance hierarchy has been circulated for EAG review
  • Inactivation of 370115009 |Special concept (special concept)| - Had been on hold pending focused input.   Briefing note has been prepared. 
  • Creation of new top level hierarchy for Generic products -  proposal in development
  • Inactivation reason review, change to Classification derived concept - in progress
  • Bypass graft revision proposal - document under review by the EAG, will be discussed at April business meeting

4Adverse reaction subtypes

On the January EAG call, it was agreed that all disorder concepts with a CAUSATIVE AGENT relationship value derived from the 373873005 |Pharmaceutical / biologic product (product)| hierarchy should be modeled as subtypes of 281647001 |Adverse reaction (disorder)|.  Upon further review of the content, there are a number of substances that could also fit under this grouper term.  A spreadsheet with a listing of disorders with a CAUSATIVE AGENT  value derived from the substance hierarchy was sent out for review by the EAG.  This had a column suggestions as to which concepts should or should not be subtypes of 281647001 |Adverse reaction (disorder)|.  The spreadsheet is available at:


Even based on a definition, such as from WHO, drugs can be used either therapeutically or they can be misused.  To make that distinction in the terminology consistently would be challenging.  

The challenge is that there are differing definitions of what is an adverse reaction and what may just be a side effect.  Regulatory agencies may have a differing view from a clinician.

What would be the heuristics that would allow an editor to make the assignment consistently if it was not stated in the FSN that what is being represented is an adverse reaction in the eyes of the person documenting the event? 


Based on the difficulty in determining what an adverse reaction is, we will leave the concepts where they are.  New adverse reaction concepts would only be added if the FSN specifically states Adverse reaction in the FSN, otherwise concepts would just be modeled as disorders.  A review of the current Adverse reactions is necessary to ensure that we are not overinterpreting a concept to be an adverse reaction when it is not clear stated in the FSN, or clearly implied by the disorder.

  • Jim Case to review existing adverse reactions for potential misclassification
5Change Surgical approach to Procedure approach

At the March 2024 EAG call, the group asked for an impact analysis of the change from Surgical approach to a more general Procedure approach.  The results of that analysis are presented in the attached document.


Why were there no classification changes when sufficiently defining a large number of currently primitive concepts?  

The group would like to see the results on this for the non-surgical procedures.  Not sure that the benefits of this generalization are sufficient.  There may be additional impacts when this attribute is applied to existing sufficiently defined concepts.

There is always an implicit approach in procedures. There are questions about reproducibility.  Are some approaches solely surgical? The current set of approaches does not seem to be organized in a way that provides any benefit to classification so adding these additional procedure approaches may not be either reproducible or provide improvements in the taxonomy.  The overlap or need to add ACCESS relationships as well, based on a particular approach has not been addressed here.  In some cases the same procedure described using different FSNs with different models do not show up as equivalences.  Will need very clear editorial guidance if this is the direction is taken.  As many of these concepts are leaf nodes, there does not appear to be a benefit to justify the resources needed to remodel the  content.

Could this be addressed by templates?

One concern was that the application of Procedure approach for non-surgical procedures would result in a loss of subtypes.  This can be mitigated by the quality reports that are now available to authors that allow the rapid identification of taxonomic changes (e.g. KPI report, Lost and found descendents report)

Another consideration is whether there is a need for a hierarchy of approaches, i.e. Procedure approach → surgical approach and whether there is a need to identify those approaches that are ONLY surgical and should not be used for non-surgical procedures?


Distribute list of non-surgical procedures affected by this change for review by EAG

  • Victor Medina to distribute a list of non-surgical procedures affected by the proposed change
6Bypass graft revision proposal

As part of the QI project, substantial changes to the representation of bypass grafts and shunts have been proposed.  A briefing note (attached) and a detailed document for the proposed changes (referenced in the briefing note) are provided.


2024-03-11 How are bypass concepts without definition of proximal and distal targets going to be modeled?  A: Need review. How are proximal and distal defined?  A: It is based on either anatomic position or flow of contents. Definitions for proximal and distal anastomosis will be provided.

Concern about creating intermediate primitives as this complicates the assignment of parents.  Also need to make sure the definitions are clear.  Where do "-stomy" concepts fit into this proposed model. 


Added document with proposed definitions for bypass, shunt, bypass graft, shunt with graft, anastomosis, etc. 

Concerns about the proposed definitions were noted.  The top level concept anastomosis included fibers or parts of a network whereas the subtypes do not include these. Need to be consistent in the definition wording to reflect the types of structures (e.g. passages, channels, etc. ) these need to be unified.  

FSNs should include the term "surgical construction" to be consistent with the definitions.

The use of "mainstream" can be unclear and we should use a clearer description, e.g. normal flow.  The definitions need to be clear in differentiating bypass and shunt as siblings as opposed to supertype/subtype.  That may be differentiated by whether the movement of contents either goes back to the normal path (bypass) or does not (shunt).  

NOTE: This may be complicated by procedures such as an intramammary artery bypass, where there is a single anastomosis from the  internal mammary artery distal to the blockage of the coronary artery, i.e. not the normal flow.  Yet this is called a bypass.

It was suggested that rather than create two new primitive concepts that distinguish between bypass and shunt, the actions be kept and new actions representing proximal and distal bypass and shunt actions be created to allow for sufficient definition of the top level grouper terms.  This needs to be tested to determine if this is a better approach than the one proposed as both options have some ambiguity of assignment due to the often conflated use of bypass and shunt.

Other questions about how to name the "connecting device" for shunts with graft.

For construction of bypass or shunt with graft, the modeling would be cleaner if we just added a USING DEVICE relationship in the graft RG as opposed to having an insertion RG with a DIRECT DEVICE or DIRECT SUBSTANCE.


2024-03-11 Comments will be added to the document by the EAG members. 

2024-04-16: Proposed definitions will be revised based on input and resubmitted to the EAG for review.  

7Remodeling of 1263452006 |Anesthesia and/or sedation procedure (procedure)| hierarchy

The current concepts 1263452006 |Anesthesia and/or sedation procedure (procedure)| and 410011004 |Administration of anesthesia AND/OR sedation (procedure)| are disjunctive groupers.  The attached document provides details on proposed changes to this hierarchy that improves the modeling and specificity of anesthetic and sedation procedures.  The proposed changes have been reviewed and accepted by the Anesthesia CRG.  Key elements include:

  • Inclusion of a HAS INTENT relationship to specify the nature of the anesthesia or sedation (e.g. local, regional, general)
  • Inactivation of "anesthesia for X" type concepts
  • Inactivate disjunctive groupers
  • Remodeling concepts to adhere to templates


This proposal has been reviewed and approved by the Anesthesia CRG.

Each concept would be  reviewed as to the intent for the administration.  The proposed HAS INTENT relationship will be grouped in the modeling with the action.

Q: Is it duplicative to have both an intent and a substance in the model when they are the same, e.g. administration of sedative with HAS INTENT = Sedation and DIRECT SUBSTANCE = Sedative? 

A: At the top level this would appear to be duplicative, and given the issues with therapeutic role groupers in the substance hierarchy,  we would test whether there is a need for a DIRECT SUBSTANCE relationship at the grouper level and add specific substances to subtypes where specified.

There are concerns about the extension of HAS INTENT.  There have been many inactivations of intent values because every procedure has an intent.  What would be the guidance for adding this to the sufficient set of relationships.  The challenge is that many substances can be used in many different ways depending on the dose and location of administration.

Regional anesthesia is challenging because it is dependent of the area blocked.  Many of the incorrect classifications are due to incorrect assignment of parents.  Also proposed that we stay away from dose levels. What is the plan for "X under anesthesia"?  The intents are really dependent on what we do with substance groupers.  Need to take this back to the Anesthesia CRG

Acupuncture for anesthesia is another concept to consider, but it is not an administration of an anesthetic substance. 

Concern that this opens a door for abuse and misuse of HAS INTENT.  

In this case the INTENT can represent the role that the substance is playing.  Also, these are generally administrations of a product instead of a substance and that should be considered.  **This is something that needs to be tested. 

Should the FSNs be made more explicit, such as "Administration of X for anesthesia"? This would allow the HAS INTENT to be applied.  Administration of anesthetic (substance) only states the administration and not the intent.  Q:  How would the current "Administration of X" be modeled if the intent was not known.  


Agreed that the "Administration of anesthesia for X procedure" concepts should be inactivated.

Additional recommendations on a proposed approach related to HAS INTENT, using Products or Substance and the use of substance groupers.

8USING DEVICE vs. DIRECT DEVICE use in procedures

During QI review of procedures, inconsistencies were revealed in the use of USING DEVICE and DIRECT DEVICE in procedures.  The current definitions for these attributes from the editorial guide: 

Direct device

Direct device (attribute) represents the device on which the method directly acts.

Using device

Using device (attribute) refers to the instrument or equipment utilized to execute an action. It is used when the device is actually used to carry out the action, that is the focus of the procedure. If the device is simply the means to access the site of the procedure, then Using access device is the appropriate attribute.

Examples of inconsistencies:
175254001 |Repair of pulmonary artery using prosthesis (procedure)| = Using device -> Prosthesis device.
771713005 |Open repair of inguinal hernia using sutures (procedure)| = Using device -> Surgical suture, device.
719263004 |Repair of parastomal hernia using mesh patch (procedure)| = Direct device -> Hernia surgical mesh.
719290008 |Repair of umbilical hernia using surgical mesh (procedure)| = Direct device -> Hernia surgical mesh.
238190005 |Repair of incisional hernia using synthetic patch (procedure)| = Using device -> Prosthetic patch.

We would like to come up with more specific editorial guidelines related to what is meant by "the device on which the method directly acts".  In general, parameters could include things such as "materially changed by the procedure", "remains in the body following the procedure", etc.  The use of the word "using" in FSNs can bias a content author to use the attribute USING DEVICE when the more appropriate DIRECT DEVICE should be used.  A broader application of DIRECT DEVICE in procedure modeling could have large impacts on the number of concept affected, but would resolve the current inconsistencies.

An example of where this is an issue is described in the attached document related to the modeling of hernia repair.

USING DEVICE and USING ACCESS DEVICE are also inconsistently used and sometimes incorrectly.  In a review of procedure subhierarchies (e.g. arthroplasty), USING ACCESS DEVICE was modeled in the surgical action RG, when in many cases the device does not act as an access to the surgical area, but plays another role, (e.g. illumination).  An evaluation of the current benefits of USING ACCESS DEVICE over USING DEVICE  is underway.  

While it is not part of this topic, we expect the same issues to be present related to USING SUBSTANCE and DIRECT SUBSTANCE


The guidance is clear in that it is the method that makes the determination of whether it is USING DEVICE or DIRECT DEVICE.  It is the focus of the action that helps makes that determination.  The changes made to account for Prosthetic device insertions and implantations have caused some confusion.  It is really a case-by-case basis to evaluate the focus of the action.  Need to be clearer in the editorial guidance related to the focus of the action.

Most of the USING ACCESS DEVICE concepts are used in relation to endoscopy.

Repair is problematic as this is a result of the action(s) as opposed to the action itself.  WRT repair of hernia using mesh, it may be overmodeling to include another role group to include  the introduction of the mesh itself.  


More precise definitions with examples will be developed and submitted to the EAG.  Impact on potential remodeling will also be determined.

Provide examples of where both a USING DEVICE and USING ACCESS DEVICE  are needed to help define the use of the latter as solely providing the access to the procedure site.

9Nontraumatic vs spontaneous injuryJim Case 

There is an inconsistency in our review of nontraumatic vs. spontaneous injury. In some cases SNOMED makes a distinction between them:
1296953008 |Nontraumatic rupture of extensor tendon of left hand (disorder)|
321371000119100 |Spontaneous rupture of extensor tendon of left hand (disorder)|

whereas in other cases we treat them as synonymous:

In determining the meaning of nontraumatic vs. spontaneous, we have found that nontraumatic means injury not caused by an external force and could occur with or without premonitory signs or symptoms, whereas "spontaneous" injuries usually refer to acute injuries without premonitory signs or symptoms.

The primary question is whether in general clinical practice this distinction has value, and if so, 1) should SNOMED explicitly separate out concepts where spontaneous injury is expressed as a synonym for nontraumatic and 2) should SNOMED create nontraumatic injury concepts as parents to spontaneous injuries where a nontraumatic parent does not currently exist?

A discussion page has been set up to solicit comments from the EAG (Re: Nontraumatic vs. Spontaneous injury)

The discussion brought out a number of meaningful points.

  • These terms are most often used interchangeably to mean an injury that does not occur as a result of external force
  • While a differentiating definition can be developed based on premonitory signs or symptoms, that distinction is often ignored by clinicians
  • It is unclear what the additional benefit is of making the distinction from a clinical perspective

If it is thought that these should be distinct, then much more specific definitions need to be developed and published.

If it is not thought that the distinction is clinically meaningful, then what should SNOMED select as the FSN and PT and should all existing differentiated concepts be merged?  Should a nontraumatic and spontaneous description be a consistent feature of these concepts?


These may just be terms that are used and although we can determine what the differences are.  

Need to be clear that we are talking about spontaneous injury as not occurring due to an external force. If we decide to keep both, then we need to make them supertype/subtype relationship.  


Create a list of concepts that are purely non-traumatic, those that are purely spontaneous and those that have both descriptions on the same concept.

Asked the EAG to add additional comments to the discussion page.


It is planned to increase the use of the Discussion section of the EAG Confluence site to allow for a more straightforward consolidation of comments solicited from the EAG regarding topics under consideration as well as publishing the outcome of the discussion and the planned implementation of decisions made, if applicable.  This will require some modifications to the discussion section to allow for the assignment of statuses to discussion pages, which will be requested from the SI technical team.

11Next meeting

May 27, 2024