• Recording of meeting approved by participants.

Conflicts of interest and agenda review

A revised version of the briefing note related to QI work being done on primary and metastatic neoplasms.  Comments from members have resulted in a revision of the original proposal. See attached briefing note.

Discussion:

No adverse comment on the proposal for resolution of primary malignancies, nor terming using "metastatic".  We did receive comments expressing concern about the potential loss of "Metastatic neoplasm (morphologic abnormality)" concepts.  For this reason, new proposal is to retain existing metastatic neoplasm morphologies and add new ones as needed.

The CSRWG has approved this proposal. 

The use case for retaining the metastatic morphologies is for some implementations of cancer synoptic reporting.  The variation of how these codes are implemented in systems.

Decision:

No additional comments from the EAG members.

Will circulate the revised briefing note to the MF and CMAG for additional review.

The current definition of the Occurrence attribute is:

"This attribute refers to the specific period of life during which a condition first presents. However, conditions may persist beyond the period of life when they first present."

In many cases, the presentation of a condition occurs after the period of life in which it is observed.  This is especially true when there is a continuum of time periods; where it can be determined that a disorder actually began prior to the time in which it is observed.  New additions to the 282032007 |Periods of life (qualifier value)| hierarchy to support maternal pregnancy and fetal development phases warrants a revision of the definition of this attribute.

A proposed new definition for the Occurrence attribute is:

"This attribute refers to the specific period of life determined to be the period of onset of a condition.  This may be prior to the actual initial observation and conditions may persist beyond the period of life when they first present."

This issue arose during a remodeling of Birth trauma, where it can be determined that the trauma occurred prior to or during delivery of the child, but is not necessarily noticed until after complete delivery.

Proposed to change the last words of the definition from "when they first present" to something representing when the condition actually started (is realized). 

4/12/2022: revised wording:

OCCURRENCE: The specific period of life determined, possible retrospectively, to be the period of onset of a condition, as opposed to the period when the condition first presents or is diagnosed.  The condition may persist beyond the period when it was first determined to have occurred. 

5/16/2022: revisions based on comments from EAG members.

OCCURRENCE: Denotes the period of life during which the onset of a finding or condition is determined to have occurred.  "Onset" may be determined based on the time of the diagnosis, time of symptom presentation, or other objective or subjective information that informs selection of period of life. Thus, the “period of life” may be fairly accurate (specific) or estimated (vague). A finding or condition may persist beyond the period of life in which it had its “onset”.

Does this represent a definition for the class or the instance of a condition? 

There is a problem with the attribute itself.  In the case of conditions related to pregnancy, this represents a co-occurrence as opposed to onset of a condition.  Thus it is interpreted at least two ways; initial onset vs. co-occurrrence.  This definition does not address both interpretations.  Suggested that we have examples that represent the types of OCCURRENCE, that we are trying to represent.  Most of the issues relate to conditions during pregnancy.   The attempts to use this to define maternal and fetal conditions has obscured its real meaning.  The OCCURRENCE attribute had initial problems and these have been exacerbated.  

Investigate the need for a new attribute to represent maternal and fetal conditions.  Evaluate the overall usefulness of this attribute as currently used.

Decision:

The attribute itself is currently overloaded and the potential for a new attribute should be considered.  Currently the OCCURRENCE attribute fails URU and resists a comprehensive text definition that would allow for consistent use.

Background:

EAG approved the updated inactivation reasons and historical associations in September 2021 and these were implemented in the Authoring Platform in January/February 2022. An outstanding item at that time was the issue of how we deal with concepts that are potentially ambiguous but not all of the individual elements of the ambiguity could or should be provided with a replacement concept.

Proposed solution:

1. For the existing reason for inactivation "Ambiguous concept" enforce a minimum of 2 "Possibly Equivalent To" historical associations (e.g. ensure that sum of the individual elements is semantically equivalent to the inactivated concept)
2. Create a new inactivation reason which states that the concept is ambiguous (label to be decided) but the tooling will allow the user to express less than the full set of associations that make up semantic equivalency with the inactivated concept.
3. This probably requires a new historical association.

Discussion:

The creation of new inactivation reason that would allow the partial set of POSSIBLY EQUIVALENT TO  historical associations might undermine the use of AMBIGUOUS.  There is a level of subjectivity in selecting ALL of the POSSIBLY EQUIVALENT TO  historical relationships.  The enforcement of at least 2 historical relationships might be too large a burden for editors.  There are a very large number of existing inactive AMBIGUOUS concepts with only one POSSIBLY EQUIVALENT TO  historical relationships.  

The tooling currently does not enforce a minimum of two historical relationships.  There is no guarantee that the full set of historical relationships would be assigned.  The excessive use of AMBIGUOUS previously may have been due to the lack of other more appropriate inactivation reasons.  With the new set of reasons, we may want to re-evaluate the use of AMBIGUOUS.  

Because there are a number of concepts that an AMBIGUOUS concept might point to would never be added and there may not be a complete set of concepts identified assigned so it would not be fully represented.  This is especially true with grouper concepts.  

Decision:

Retain the current tooling that allows for a single POSSIBLY EQUIVALENT TO relationship for AMBIGUOUS inactivation.  Editorial guidance would need to be updated to reflect this change as well as emphasizing that multiple historical relationships should be assigned, if possible.  It is not necessary to represent all of the semantic meaning if the concepts needed should not exist in SNOMED CT.  Where appropriate, new concepts should be created if they are clinically valid. 

Measurement Findings:

Proposed changes to FSNs

• Proposed changes to FSNs: Please read the attached Briefing before the meeting and if possible return your comments for discussion at the meeting.

So far, with limited input, there has been a positive response to the proposal.  One area where there is an issue is where existing concepts that refer to "increased" are referencing a relative increase as opposed to an above or below reference range. 

Another concern...ambiguity based on the source of the component being measured.  E.g. increased albumin. There are a number of "agnostic" groupers (with regards to specimen) that would need to be addressed as to whether they are ambiguous or not. Should these be resolved at the same time as the remodeling and reterming or handled as a separate phase of the project.  

Consensus that the notion of "relative to last time (or time period)" should not be represented in the international release. There were some in favor of inactivating those that use "increased" or "decreased" as inherently ambiguous.    "Normal" in terms of substances/drug is also ambiguous based on the use case.  Given the many examples of how the terms can be interpreted, the group determined that these all need to be inactivated and replaced with more explicit.  The issue is that there is a problem with determining what the inactivation reason would be. It is suggested to inactivate the entire hierarchy without replacement and create a new hierarchy with more explicit concepts.

A revised briefing note has been distributed.   The main change is to accept a decision made in 2010 that "increased" in the context of measurement findings means "above reference range" (which is the way most concepts are currently modeled).  If so, can the FSNs be changed as opposed to inactivation and replacement?

High and low concepts should also be interpreted WRT the reference range.  There is an extremely low possibility that a user would interpret these as "changed relative to a previous measurement".  This would allow for a simple change to the FSN as opposed to inactivating and replacing these.

Re: Within reference range.  Does this mean "normal"? Can "normal" represent a case-specific representation of reference range?  It is incumbent upon other aspects of the record to determine which reference range would be used.  The notion of normal is context dependent and is something that should be viewed in the context of the entire medical record.  Should "normal" be included as a synonym?  it is so commonly used that should it be included as a synonym?

Decision:

None