Please Note: This service will be unavailable on 20 July from 06:00 UTC till 18:00 UTC for essential maintenance.
Page tree
Skip to end of metadata
Go to start of metadata

For additional Relevant links with regards to Content Development Information please click here Content Development Information

ItemsProposed improvement

Planned Release date

(*this is provisional only and is subject to change)

1

Replacement of the Stated Relationship files with the new OWL Axiom refset files

Following on from all of the previous announcements providing early visibility of the transition plan, SNOMED International would now like to provide technical details of the planned migration from Stated Relationships to OWL files.  This plan has been discussed and agreed with numerous stakeholders from the community, and will be implemented over the course of the next two International Releases, as follows:

January 2019:

  • The International Edition package (packageA) will include both active Stated Relationships plus a partial OWL file containing a subset of the OWLAxioms.  This file will contain both OWLAxiom + OWLOntology refsets combined into one OWLExpression file, in each section of the International Release package (Full, Snapshot and Delta):
    • sct2_sRefset_OWLExpressionFull_INT_20190131.txt
    • sct2_sRefset_OWLExpressionSnapshot_INT_20190131.txt
    • sct2_sRefset_OWLExpressionDelta_INT_20190131.txt
  • We will also publish a separate optional package (packageB) containing a complete set of OWL records (including all Axioms - sufficient, transitivity etc) set to January 2019 effectiveTimes, as if this was the complete OWL file that we will introduce in July 2019.  This package would also contain all stated relationships from the previous release (July 2018 International Edition) marked as inactive.  
  • packageB will therefore act as a second Demonstration release (similar to the September 2018 package), allowing everyone to thoroughly trial how the July 2019 Int Edition will work for them in practice.  It will, therefore, retain the same nomenclature and formatting as the Sept 2018 Demo release, with everything x prefixed and labelled to warn against use in Production clinical systems.
  • packageB will be published in MLDS alongside the International Edition, and will therefore be available to everyone who has access to the International Edition, so that they can choose to either download it or not.
  • Release Notes will be updated to clearly explain all of the above.

July 2019:

  • The International Edition package (packageCwill include both inactive Stated Relationships plus a complete OWL file (combined) from July 2019 effectiveTime onwards (ie) with NO history of OWL records that weren’t included in the Jan 2019 International Edition package itself (as opposed to the optional package).
  • We will also publish a separate optional package (packageD) containing an OWL (combined) Delta file, identifying which of the concepts have had modelling changes in relation to packageB.
  • No support for Stated Relationships will be provided from this point onwards, however we will continue to include the inactivated Stated Relationships in future International Edition packages, until it is decided that this is no longer required.
  • Anyone who isn’t yet ready to move forward with OWL will therefore remain on the January 2019 International Edition until they are ready to update to OWL with the July 2019 version onwards.
  • Release Notes will be updated to clearly explain all of the above.

  

The inferred relationship file will maintain the same format and structure, though it will no longer contain all necessary and sufficient conditions. The inferred relationship file is represented in Necessary Normal Form for distribution of relationships. It is a collection of all the necessary conditions and represents a subset of the full semantics from the 2018 July release and onwards. Most users will benefit from the improvements in the inferred relationships without requiring changes to their existing systems.

A set of documentations has been developed to support the Logic Profile Enhancements. 


Users should carefully analyse any potential impact to their systems (upload routines, etc) and make provisions for these changes urgently (if not already done), in order to prevent any issues when these changes come into effect in January 2019.  Please contact SNOMED International at support@snomed.org with “OWL Axiom refset files implementation question” in the subject line.    


January 2019 International Release +

July 2019 International Release

2LOINC timeline changes - please refer to Roadmap link here LOINCJuly 2018 and potentially other future releases - On hold
3

Revision of IS A relationships for anatomy concepts 

The new anatomy concept model implemented the enhanced Description Logic features, e.g. reflexive and transitive properties, additional axioms, for consistent logical modeling. The inferred IS A relationships from the OWL anatomy ontology have been reviewed and the changes have been implemented in the production. The potential impact to other hierarchies, such as disorders, procedures, observables, situations etc are also reviewed as part of the project.

The revision of sub-hierarchy 26107004|Structure of musculoskeletal system (body structure)| has been planned for the January 2019 release. The rest body systems will be revised in the future releases.

The subject area for revision of IS A relationships will be determined and updated in the following confluence page.

July 2019 International Release
4

Improvement for the representation of role groups

It is important to clearly indicate if an attribute is grouped or not because role grouping has impact to semantics and classification results. The majority of the Modeling Advisory Group members recommended explicit representation for role groups. There will be two key improvements for representation in the July 2019 release. The changes are demonstrated in the January 2019 Demo release package.

Firstly, role groups are explicitly stated and represented by the concept 609096000|Role group (attribute)| as an object property in the OWL axiom refset.

Secondly, for the inferred relationship file, role group 0 will be only applied to attributes that are not logically grouped, e.g. not a value of 609096000|Role group (attribute)| in an OWL axiom or grouped = 0 in the MRCM. This addresses the potential confusion whether an attribute is grouped or not in the role group 0. 

The improvements will provide consistent representation for role groups in the OWL axioms, inferred relationship file, and diagramming of the concept model.

July 2019 International Release
5

MRCM Changes

MRCM changes in the July 2019 release

An update on changes for the January 2020 release will be posted here as soon as it is available.

January 2020 International Release
6

Quality Initiative

Information about the project can found here - Quality Initiative - Progress

Note: There are changes to the Service Level Agreement for the CRS as a result of the Quality Initiative:

https://www.snomed.org/SNOMED/media/SNOMED/documents/Early-Visibility-of-Changes-to-Service-Level-Agreement-International-Content-Request-Service-(CRS).pdf

January 2020 International Release
7

Planned changes to 'Co-occurrent and due to' pattern

During the implementation of the new Description Logic features a conflict was uncovered between the modeling of 'Co-occurrent and due to' and General Concept Inclusions (GCIs). This has resulted in the need to reconsider the modeling of "Co-occurrent and due' and update the Editorial Guide for this area.

There are a number of new concepts in the January 2019 release that are based upon the existing guidance for 'Co-occurrent and due to'. The plan is to update the Editorial Guide and all concepts that are currently modeled as 'Co-occurrent and due to' starting from the July 2019 release.

January 2020 International Release

8

WAS_A

Inactivation reason of LIMITED/WAS_A is not allowed for any new content inactivations after the July 2018 release. The WAS_A association refset has not been updated thereafter.

At the Editorial Advisory Group meeting in April 2019, agreement was reached to discontinue the maintenance of WAS_A relationships when inactivating concepts that have a historical association to an inactive concept. When changes are made to a historical relationship for a concept that was previously inactivated using WAS_A, effort will be made to assign a new historical relationship that facilitates traceability of the concept (e.g. DUPLICATE or AMBIGUOUS) as opposed to NON-CONFORMANCE TO EDITORIAL POLICY.

Existing WAS_A relationships will be inactivated in a future release once a plan for batch reassignment of historical relationships has been developed. Until then, SNOMED International will not continue to use or maintain WAS_A relationships.

January 2020 International Release and future releases
9

Known Content Issues July 2019 Release

Approximately 149 text definitions are missing either a US or GB spelling variation. These are all legacy issues that will be rectified in the January 2020 release.

A small set of disorder concepts have been identified with descendants that have a semantic tag of clinical finding. A further small set of concepts have been identified that have a semantic tag of disorder but are not descendants of 64572001|Disease (disorder)|. These concepts are all legacy issues that will be rectified in the January 2020 release.

Release validation rules have been enhanced to prevent recurrence of legacy issues for future content additions.

January 2020 International Release
10ICD-O project work

IACR have released ICD-O-3.2 as a listing of all new additions, changes and revisions to ICD-O-3.1 in Excel format. The new version, ICD-O-3.2, is recommended for use from 2020.  ICD-O-3.2 is not yet available online or in book format. In preparation for the update, the changes have been divided into types and are being investigated/addressed in the ICD-O-3.2 update project. The priority is being given to those that have an impact on clinical safety, e.g., changes of behaviour (benign, malignant etc.) and adding new content for SNOMED CT morphologies, this refers to the 400177003 |Neoplasm and/or hamartoma (morphologic abnormality)| hierarchy.

The descriptions with [obs] have been inactivated as outdated because they are artefacts of descriptions from the ICD-O-3 print version for their history tracking. 

The maps to ICD-O-3.1 are being updated only for error correction purposes for the January 2020 release. The maps to ICD-O-3.2 are out of scope for the January 2020 release and are planned for a future release.

January 2020 International Release
11

Over the Counter

Concepts referring to “over the counter” status have been inactivated. 

Concepts referring to regulatory status or characterization (e.g. over the counter) are out of scope for the International Release since the meaning may vary by jurisdiction and is not consistent internationally.

January 2020 International Release
12

Vaccine Poisoning

Concepts referring to vaccine poisoning have been inactivated with historical relationships to the corresponding Adverse reaction caused by X vaccine and to Adverse reaction caused by vaccine constituent.

The notion of vaccine poisoning has persisted in ICD 10 under T50 but it is not included in ICD-11; the poisoning aspect seemingly referred to the adjuvants and carriers that are included in the vaccines as opposed to the biological component itself.

January 2020 International Release
13

Revision of 298180004 |Finding of range of joint movement (finding)

The concepts under the subhierarchy 298180004 |Finding of range of joint movement (finding)| are being remodeled to represent the finding site of <<785818007 |Structure of joint region (body structure)| instead of <<39352004 |Joint structure (body structure)|. 

Range of movement of a joint region can be impacted by both bone and soft tissues, therefore the use of 39352004 |Joint structure (body structure)| was too restrictive and caused issues in other areas of the terminology.  These changes will be made over the course of two releases due to the number of affected concepts.

July 2019 International Release

January 2020 International Release

14

Planned changes to Implantation and Insertion procedures

Work commenced on content tracker IHTSDO-175 IHTSDO-175 - Getting issue details... STATUS

Ongoing structural changes for the Procedure concepts related to the content tracker:

It was agreed that not all Insertion and Implantation procedures are surgical.

As a result, further work in the hierarchy under 129284003 |Surgical action (qualifier value)| has been undertaken for the January 2020 release and this includes:
Reactivation of 129338005 |Surgical implantation - action (qualifier value)|
Addition of a new concept 788288006 |Surgical reimplantation - action (qualifier value)|
Remodel of 'Reimplantation procedures' replaced METHOD = 129337000 |Reimplantation - action (qualifier value)| with new 788288006 |Surgical reimplantation - action (qualifier value)|.

Work on this project will continue for future releases after January 2020.

January 2020 International Release

15

Concrete domains and numeric representation

Numerics are represented by concepts in the Medicinal product concept model in SNOMED CT, which is an interim solution before the implementation of concrete domains to support data types, such as decimal, integer, string and date/time. The implementation of concrete domains depends on the OWL expression reference set and the update of inferred relationship file. The first complete OWL expression refset will be available in the July 2019 international release. It will take time to update the specification for the inferred relationship file and consult with the community of practice. When they are ready, strength in medicinal product model can be transformed to concrete domains. The transformation will be completed by technical changes without impact to classification results.

July 2019 International Release



  • No labels

4 Comments

  1. Andrew Atkinson - can we re-arrange in Chronological Order?

  2. I see efforts around ICD-11 and ICD-O. Can someone from SNOMED CT comment on whether there are any plans to harmonize SCT with ICHI.  https://en.wikipedia.org/wiki/International_Classification_of_Health_Interventions

    The overlap of SNOMED CT procedures with ICHI is mind provoking.

    link to existing themes https://confluence.ihtsdotools.org/display/CDR/Content+Project+Details

  3. Hello Vojtech

    We are currently discussing with WHO how we will progress with linking SNOMED CT to the family of classification going forward, with the current focus on ICD-11 since it is released. We hope that the work we are doing on exploring how to more efficiently create a link with ICD-11 is work that can then be reused for the other WHO classifications.

    It would be helpful to understand the approach countries are going to be taking to implementing ICHI along with timelines, so that we can factor that in to our planning. Can you give us some information about this if you have it, or knowledge of any work already going on in this space

    Best wishes

    Jane Millar

    Collaboration Lead

     

  4. Thanks Jane Millar and Vojtech Huser for raising this message thread.

    Not being very familiar with ICHI myself, and knowing that we use CCI (Canadian Classification of Health Interventions) here in Canada, I reached out to Sharon Baker, the Manager of Classifications at CIHI based on Jane’s request to understand more.  Here is Sharon’s response: 

    Interesting email thread. ICHI is structured with various degrees of specificity for use at the different levels of the health systems, and uses a common accepted terminology in order to permit comparison of data between countries and services. To understand the differences between ICHI and CCI, Classifications and Terminologies have been undertaking mapping projects, as well as participating in field-trials. ICHI contains around 7,500 codes, whereas CCI has >16,000 codes so there definitely less coverage, and in some cases less specificity. However, ICHI covers all parts of the health system, and contains a wide range of new material not found in CCI e.g. Functioning and public health interventions, so we will asses if there is a need to incorporate this additional content into CCI. However, the long term implications of ICHI will require substantial discussion among interested parties in Canada.

    Of note, I am a member of the ICHI Task Force and we have a meeting in September so I can certainly bring this forward as a discussion point.

    If anyone is interested, more information about CCI can be found here: https://www.cihi.ca/en/submit-data-and-view-standards/codes-and-classifications