Jim Case

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

Jim Case

None noted.

Jim Case 

Nicola Ingram 

A Briefing Note has been written to inform the community of practice about changes to the dose form modeling to include the use of newly published concentrate dose forms (added as required by the EDQM mapping project) in the modeling of clinical drugs in the International Edition of SNOMED CT. The intent is to gather feedback on a change to Editorial Guidance for clinical drug concepts with dose form of drops or spray (feedback to be received by 29th 22th April).

The document is attached to the meeting agenda above.

Jim Case 

Scenarios where qualifier values are not sufficient; e.g. at risk findings, coma score subscore results

From

In developing the model for defining risk findings, we have run up against an issue with describing the specific risk factor as the range of HAS INTERPRETATION is limited to 

<<  [260245000 |Finding value (qualifier value)||http://snomed.info/id/260245000]  OR
 <<  [263714004 |Colors (qualifier value)||http://snomed.info/id/263714004]  OR
 <<  [308916002 |Environment or geographical location (environment / location)||http://snomed.info/id/308916002]

There are two approaches to modeling specific risk factors.

• Use the DUE TO relationship to describe the risk factor
• Use the INTERPRETS = <<80943009 |Risk factor (observable entity)|. HAS INTERPRETATION = <Clinical finding OR <Event OR <417662000 |History of clinical finding in subject (situation)| OR <416940007 |Past history of procedure (situation)|

Example 1:

1184692002 |At increased risk due to lack of fire extinguisher in residence (finding)|

Clinical finding:

   DUE TO Inadequate fire extinguishing equipment in residence (finding)   

   {INTERPRETS =  Risk level,     HAS INTERPRETATION =  Increased}

Example 2:

1184692002 |At increased risk due to lack of fire extinguisher in residence (finding)|

Clinical finding:

   {INTERPRETS =  Risk level,     HAS INTERPRETATION =  Increased}

   {INTERPRETS =  Risk factor,    HAS INTERPRETATION = Inadequate fire extinguishing equipment in residence (finding)}

Currently, there is no way to assign values to the Observable 80943009 |Risk factor (observable entity)| due to the limitations of the range of HAS INTERPRETATION.  Existing use of the 80943009 |Risk factor (observable entity)| has been restricted to a few concepts that also use DUE TO as the relationship to the factor with no HAS INTERPRETATION VALUE (e.g. 1162398002 |Adverse health risk due to mold in residence (finding)|)

There is a similar issue with assigning the values associated with assessment instrument Observables

There is a huge risk in extending the range of HAS INTERPRETATION, but using this pattern is a better representation of the definition than the implied relationship using the DUE modeling pattern.

27 Feb 2023 

Additional issues.

Should "Decreased risk" and "Low risk" be siblings or parent-child?  In the current project (not yet promoted to MAIN), 75540009 |High (qualifier value)| has been made a subtype of 35105006 |Increased (qualifier value)|; however, the interpretation of the meaning of "high" is absolute (compared to a standard), whereas "Increased" is a relative term, i.e. compared with a prior measurement or with a "relative" norm). 

Example: High risk

Image Removed

There is currently only one concept stating decreased risk (1144845004 |Risk of suicide decreased (finding)|).  All "Low risk" concepts currently classify directly under 281694009 |Finding of at risk (finding)|

Discussion:

Based on input from Monique van Berkum (document attached to agenda above) and the Gravity project, 13 "at risk" concepts submitted by the Gravity project were inactivated.  Additionally, a recommendation to create the necessary pre-coordinated concepts for use in the HAS REALIZATION relationship for the At risk model has mitigated the current need to extend the range of HAS INTERPRETATION, for the purposes of modeling risk findings.

This remodeling has been performed and should be available in the May 2023 release.

The full model currently being used for "At risk" concepts is defined in the template located at: At [qualifier] risk of [finding/event] (finding) - Ready for review

•  Jim Case to continue modeling based on revised editorial guidance.

Document by James R. Campbell  related to risk calculators is attached to the agenda.  

The primary premise is that the "proper use of HAS_INTERPRETATION when INTERPRETATION has value of an Observable entity is that the valueset must adhere rigorously to datatype restrictions specified by the SCALE_TYPE defining the Observable concept."   Additionally, the following extensions to the concept model are proposed:

1) Expand the valueset constraints for 719722006|Has realization (attribute)|

Note: Range for this attribute is currently << 272379006 |Event (event)| OR << 404684003 |Clinical finding (finding)| OR << 71388002 |Procedure (procedure)| OR << 719982003 |Process (qualifier value)|

2) Add a Risk property to <<118598001|Property (qualifier value)|
3) Create a role hierarchy for 363713009|Has interpretation (attribute)| to support additional attributes including concrete domains and ordered refsets that are needed
for Quantitative and Ordinal or quantitative Scale types
4) Aligning with the editorial principles of LOINC, a policy for employing refsets defining the valuesets for observable entity concepts with scale type of Nominal, Ordinal and
Ordinal or quantitative. These refsets would allow interpretation of evaluation findings and conceivably could by integrated into classification.

In reference to content development in the International release, the following must be considered:

• On 2020-04-29 the EAG approved an editorial policy: When defining observable entities for the international release, the SCALE TYPE attribute will not be used. If extensions would like to add specific subtypes of observable entities that include the SCALE TYPE, they are free to do so.
  When using observable entities to define clinical findings, international concepts that do not include a SCALE TYPE relationship would be used a values for the INTERPRETS relationship. The exception to this guidance are existing "vital sign" observable entities that have been defined with the SCALE TYPE of "Quantitiative".
• The majority of Observable entities in the International release of SNOMED CT are not modeled, so Scale Type is not available.
• There are currently no International concepts that require concrete values for HAS INTERPRETATION
• It is not possible to support both concept based values and concrete values in the same range for a scale type such as "Quantitative or Ordinal"

Discussion:

•  James R. Campbell to write up a more general description of the issue with additional examples.

Should "Mechanical complication of device" be a disorder or a finding?

Current situation:

111746009 |Mechanical complication of device (disorder)| has 215 subtypes, all of which refer to a failure of a device without specifying a deleterious effect on the patient. 

       e.g. 285961000119107 |Mechanical breakdown of prosthetic heart valve (disorder)|

We also have disorder concepts that refer to a patient condition due to mechanical failure of a device 

       e.g. 5053004 |Cardiac insufficiency due to prosthesis (disorder)|

Without specifying the resultant condition associated with device mechanical issues, is it appropriate that these are represented as patient disorders, or should they be findings that may be related to disorders in the patient?

A briefing note using "Leakage of device as an example is located here.

Discussion:

Without an associated condition in the patient it is unclear whether there is an adverse impact on the patient.  In general, if the device is implanted, it can be assumed that there is a negative impact. These should be retained as disorders.  Each of the subtypes would need to be evaluated as to the potential impact on the patient.

It was suggested at a recent Modeling Advisory Group meeting that SNOMED should consider inactivating the 363743006 |Navigational concept (navigational concept)| hierarchy.  A list of 635 primitive concepts that are unable to be defined due to their highly contextual use.  The concern is that because they are more or less "orphan" concepts, and provide no analytical advantage, they would be discouraged from use in medical records.  However, because many of these descriptions are those that are commonly used in clinical records, with organizationally specific meaning, they are being entered into EHRs.

UK has high usage of a few of these:

CONCEPTID             FSN                                                                                                                         USAGE 2011-2022

394617004              Result (navigational concept)                                                                                 48,227,610

160237006              History/symptoms (navigational concept)                                                             10,146,392

309157004              Normal laboratory finding (navigational concept)                                                 581,209

267368005              Endocrine, nutritional, metabolic and immunity disorders                                   160,475

243800003              Test categorized by action status (navigational concept)                                      108,176

250541005              Biochemical finding (navigational concept)                                                           100,632

It is unclear how these are used in clinical records although from the above list it appears they may be used as document headers .  An inquiry to the UK resulted in an interesting observation that some of these highly used concepts are primarily found in only one of two major primary care systems, and some that were not used much a decade ago are beginning to increase in usage. It was proposed to SNOMED that aside from the highly used concepts above, the remainder of the navigational concepts could be inactivated without much impact on users.  However, we would need to consider carefully before inactivating the above concepts without suitable replacements due to their high, albeit incorrect, usage.

Question: Should these be replaced with Record artifact concepts or something else?

Discussion:

The NL and AU have not approved the use of navigational concepts.  IMO does not recommend use of these. The more general question is whether to inactivate navigational concepts as a whole.  The consensus was that that these should be inactivated. Because there are concepts of high use in the UK, they will be contacted and informed of the impending inactivation.  Potential replacement concepts for the high use concepts include:

394617004  Result (navigational concept)    → 423100009 |Results section (record artifact)|

160237006  History/symptoms (navigational concept) → 371529009 |History and physical report (record artifact)|

309157004  Normal laboratory finding (navigational concept) → None (laboratory is a vague concept)

267368005  Endocrine, nutritional, metabolic and immunity disorders → None (arbitrary grouping)

243800003  Test categorized by action status (navigational concept)  → None (meaning unclear)

250541005  Biochemical finding (navigational concept) → None (context dependent)

Update 27 Feb 2023 

Message sent to the UKTC, no response as of yet.  Briefing note will be written once input from the UK is received.

Section of Chief Terminologist report to Content Committee June, 2010

Questions have been raised regarding editorial policy for Navigational Concepts. There are two main
questions:
1) What are the criteria for making a concept navigational?
2) Should navigational concepts be used for recording clinical data?

The current User Guide (2010)has the following information:
Navigational concept
These concept codes are to be used only as nodes in a Navigation Subset. They are not suitable for
data recording or aggregation.
The subclasses of Navigational concept have the following characteristics:
• They have no IS A subtypes.
• They have no IS A supertypes other than Navigational concept.
• They may be associated with other concept codes by the use of Navigation Links.

JTC Note: From the current SNOMED Editorial Guide: "...Navigational concepts were created to group other concepts without explicit regard for defining attributes (since there were none). Their purpose was to provide top level groupers for subsets and reference sets used in implementations.  Because the Reference Set mechanism is now available, there is no longer a need for navigational concepts in the International Release; however, they can be added at the national or lower level."

•  Jim Case to contact UKTC to inform them of the impending inactivation
•  Jim Case to write briefing note explaining the rationale for inactivation of the Navigational concept hierarchy.

Jim Case 

Penny Livesay 

•  Jim Case to send a note to the MAG inquiring about the value of a new definition status that identifies a concept with GCIs

Next meeting April 4. SNOMED business meeting