SNOMED CT Editorial Advisory Group (AG) 

Teleconference via GoToMeeting

Attendees:

Chair: Jim Case (JCA)

AG members:

• Bruce Goldberg (BGO)
• Keith Campbell (KCA)
• Paul Amos (PAM)

Presenter: Toni Morrison (TMO)

Observers: Monica Harry, others 

Apologies: 

AG member Guillermo Reynoso (GRE)

Minute-taker: Juliet Gole Krarup, from a recording.

Welcome

JCA welcomed the group. He encouraged everyone (including staff and observers) to go suggest April face-to-face topics via the Discussions page in Confluence. 

Laterality

JCA noted that Robert Turnbull had added some comments about adding QA to the document, and there had also been some comments from Olivier Bodenreider from NLM, to which JCA would respond. 

As only one out of four AG members was on the call at that point (BGO), JCA deferred discussion on laterality and asked TMO to present on Product Strength.

Product Strength

TMO showed some slides and noted that she, JCA and Ian Green had gone back and forth on various options. 

Slides:

On Option #3, JCA said non-human content was moved to an active, maintained veterinary extension.

KCA said he supported Option #4. Option #5, he said, that would not add new concepts was a bad idea. On option 4, the team could minimize the editing burden by using content from RxNorm. The VA had invested in programs that took RxNorm and transformed it into an extension to SNOMED. It could be automated, though with human QA, and build upon the maintenance already going on with RxNorm. 

JCA asked if the tooling took variability in the naming and transformed it into something regular. KCA replied no, but it took naming from RxNorm that was regular. NLM was okay with their being overlap so the VA was not spending time on reconciling overlap, but the VA could update the FSNs with an algorithmically-generated name that was consistent. 

KCA referred to the RxNorm webpage about how the naming convention involving strength and dose.

JCA agreed that automation would be an advantage.

KCA said the VA would be happy to collaborate bilaterally or multilaterally, although it was not involved in the IHTSDO drug model project because he did not have the ability to attend regular meetings. JCA asked if the drug modeling group was aware of the VA work. KCA said no, he had spoken to Dion McMurtrie and Michael Lawley but had not made a presentation to the group.

• TMO to invite KCA to present the VA RxNorm project to the drug modeling group.

PAM said that the majority of content demonstrated in TMO's presentation was from Read Codes v.2, which was based around branded products. HSCIC had abandoned that Read Codes v.2 drugs dictionary in favor of the dictionary of medicines and drugs from the UK. The only other group possibly still using the Read Codes v.2 was New Zealand, so he was hesitant about putting a lot of effort into dealing with concepts that may not be in use.

TMO said she had spoken to Jo Goulding and Emma Melush, and the UK had mappings to about 4000 concepts. PAM took an action to speak to Jo and Emma about it and report back.

• Paul Amos to speak to Jo Goulding and Emma Melush about the use of mappings to the drug concepts being considered in the drug strength project. 

TMO asked KCA if he viewed it as a 2-step approach. KCA said yes.

BGO asked how many of the drug concepts in question represented vaccines. TMO said she could send him an estimage.

TMO said the vision for the International Release/core was to include the active ingredient plus dose, but put strength elsewhere.

• TMO to revise the proposal document after getting input from Paul Amos about Read Codes v.2.

KCA asked if there was a clear consensus on option 4. PAM, BGO and JCA agreed that option 4 was the best option. 

Decision: The Advisory Group agreed that Option 4 was the best option for the Product Strength project.

Plan for concepts representing role

TMO presented some slides:

TMO said the main question for the group was how should the team inactivate the concepts?

KCA said he recommended updating, because it was not really an error, but rather it reflected an older modeling style that included role. He said he did not advocate creating and maintaining WAS A relationships.

JCA said the AG had decided in Montevideo that WAS A was more for RF1 and not needed in RF2. 

JCA said when deciding whether to inactivate large numbers of concepts, the team needed to notify Members of the intent to do that. There was a need for a communications plan. TMO said that she had the data necessary for the communications plan, and the drug product group was aware that it was coming. JCA said notification needed to be broader than that. There had been some push-back from the Member Forum when the MF had found out about retirement of substances. Inactivating 1200 terms, JCA said, qualified as a large enough number to require Member notification. TMO said it probably would not be 1200 terms (that was just the number she was starting with for a deep dive). She asked if notification to the MF, CMAG, Editorial AG and Drug Project Group was appropriate.

• JCA said that TMO should collect input on the inactivation of concepts needing role from CMAG, then brief the MF. TMO should talk to IGR about how to do that.

KCA recommended adding sufficient advice to the communications plan. If you take out all the role representations, you should point people to alternative ways of doing it. 

PAM spoke about problems resulting from the UK getting rid of the Read Codes v2 drug dictionary involving a need to group drugs by their use. JCA asked how realm specific those groups were. PAM guessed that the branded products were the ones that were realm specific. KCA guessed that there would be a 95 percent agreement across countries, but there would be things in the margins. JCA asked if the project was dependent on finding a replacement for those groupings before it could move forward. KCA said it depended on resistance from IHTSDO stakeholders and how much "tough love" would be shown. PAM said he would speak to Jo and Emma about the UK solution to see if it was interoperable or just UK specific.

JCA said the team would have to ask the MF who was using those groupers and what the use cases were. If there were no substantive use cases that couldn't be easily replaced by something else, then that would be different from the whole system being dependent on them.

• JCA instructed TMO to write up an inquiry to the MF and Content AG stating what the plan was, outlining benefits and detriments, and requesting information on how it would impact users.

TMO asked “Is it still the opinion of this group that that’s knowledge that shouldn’t be within the hierarchy? These certainly shouldn’t be an IS A relationships.” JCA agreed that it should not be IS A relationships. KCA said he completely agreed and it would be fine to put them in a ref set. PAM and BGO agreed.

Laterality Proposal

JCA briefed the group on the feedback received from the other AGs on the laterality proposal. He had received a comprehensive response from the Modeling AG, a partial one from CMAG, but nothing from Tooling AG or Release AG. JCA said the proposed changes in laterality would have a big impact on QA rules for tooling, but he was not sure whether it would impact the release process. He said he would encourage those groups to provide input so that a report could go forward in April and the changes might occur in the Jan. 2017 release.

Use of Slashes in FSNs

JCA said the final recommendation on use of slashes in FSNs was documented on the discussions page. Editorial guidance would be in the next revision of the Editorial Guide.

Potential Agenda Items for April 

JCA encouraged everyone to add topics to the discussion page. KCA suggested what modularization might look like. JCA agreed.

Next Meeting

The group agreed to meet the next time the first Monday in April at the regular time.

Adjournment

JCA adjourned the meeting after 52 minutes.