2021-05-21 TSC WGM Agenda/Minutes

Quorum Requirement: Chair + 8 individuals comprised of at least 3 Working Group representatives and 1 National/Regional representative

Agenda

1. Housekeeping
2. Introduction of visitors (including declaration of interests)
3. Agenda review and approval
4. Discussion topics:
   1. Terminology Service Group update: 9:15 am Eastern (Austin)
      1. Terminology Services Group (TSG) Draft Mission and Charter
   2. Future of V3: 10:00 am Eastern (Austin)
   3. Re-training cochairs - what is our role? What are the training topics? - 30 min (Virginia)
      1. Training for new TSC members
   4. Management group terms/reappointments - should there be term limits? - 30 min (Austin)
   5. Increasing participant joy/reducing participant burden - 30 min (Virginia)
   6. Work Group Health metrics - task force update - 15 min (Riki)

Minutes

1. Housekeeping
   1. Temperature check after break or in chat
2. Introduction of visitors (including declaration of interests)
   1. We will have guests for the terminology service group update (Reuben, Carol, Julie, Rob Hausam) and the future of V3 discussion (Floyd, Rob Hausam, Craig)
3. Agenda review and approval
   1. Reviewed; no additions
4. Discussion topics:
   1. Terminology Service Group update: 9:15 am Eastern (Austin)
      1. Terminology Services Group (TSG) Draft Mission and Charter
         1. Austin gives the TSC some background on the concept and why it is proposed as a service group rather than a traditional management group.
         2. Reuben shares a presentation on the problem statement this group is meant to solve, the consequences of lack of action, and how the Terminology Services Group will provide the solution.
         3. Jean agrees with the problem and setting up a group, but has questions about the HTA. Seems like HTA would become one of the functions of the TSC? 
            1. Julie responds that the HTA places great demands on volunteers and it needs greater management and more volunteers. Would like to see the responsibilities reside in the TSG.
            2. Carol: The disjointedness perceived around HTA is a symptom of the lack of understanding of the intricacies of what we do. 
            3. Melva: Not convinced that the HTA has to be disbanded - it could be a subcommittee or a task force. Austin: The initial proposal is to leave the HTA intact for the time being.
            4. Julie clarifies that HTA doesn't have any responsibility of HL7 owned terminologies; it's external terminologies and bringing everything together.
            5. Melva: Have you done any thinking about financial resourcing implications to this? Austin clarifies that TSC has no budget, but the group if formed will have to determine what kind of funding they might need. Melva: The need for resourcing has to be part of this discussion.
            6. Jean: If we're going to be serious about providing terminology to it, we'll need more volunteers and funding.
            7. Rob: This thing could be called the HL7 Terminology Authority and it would cover internal and external terminologies. We need to acknowledge that there is a service requirement that has always been there and that FHIR has made evident. One of the best ways to think about this request is to think about what we're already doing for FHIR. We fund a lot of infrastructure to deliver FHIR to our customers, particularly in terms of publishing. This is a part of publishing. The terminology elements are across all of our product families. The community is asking us to ensure that we have consistency within our product families and across product families. We may need to start thinking about publishing the same way. We need to think about what we're already delivering for FHIR, and what we can deliver for our funded stakeholders. 
            8. Mary Kay: I want to clarify that I wholeheartedly support this. The costs related to startup and maintenance will be significant, as well as the cleanup.
            9. Amit proposes that we choose a name less similar to TSC. Calling it a management group might make sense, maybe TSMG. The responsibilities of HTA in the GOM is to maintain and manage external terminologies but it sounds like we need to simply empower them further to manage external and internal. Julie disagrees. HTA is a small group charged with trying to get external terminology correctly available for use in HL7 standards. We're doing what vocabulary facilitators used to do as a group of 6 for the whole organization. 
            10. Josh asks where HL7 draws the line between managing consensus vs. operations? Wayne notes there's no real guidance and it's becoming a problem. Rob: This is what I mentioned earlier, the consequences of creating an enterprise service. 
            11. Carol: HL7 is under fire constantly for terminology but we're not able to keep up in the current way it's set up.
            12. David: We're looking at a management group because we have governance, management, and methodology. A service provision group made up of volunteers is going beyond the mandate. If you have FHIR-I and the PMGs involved it starts to duplicate the TSC. It's reaching far beyond what a management group does. We need to keep the roles clear. Need to be very careful about who is a member of the service group - there could be a liaison function but not part of the overall structure.
            13. Reuben: When we put services in the name, we meant it in the most basic definition of the word. Some of the services can be handled on a volunteer basis and some will eventually require funding. Re: the GOM, HL7 has a history of evolving over time. We're at a point with terminology that it's time for another evolution by creating this kind of group. 
            14. Rob: FHIR at its core is wonderful because it's operational. The board should answer what they think they're building when they say FHIR is what we need to focus on? As an organization the terminology volunteers are overwhelmed. If we don't come up with an infrastructure to support what FHIR can deliver on we're going to miss an opportunity, and it will collapse and/or an external entity will take it over for us. This group is voting for the consequence.
            15. Christof: Decision making people are simplifying the picture of interoperability to technical interoperability and semantic interoperability. It's difficult to explain what HL7 does because it sits between those two. If we are putting terminology high on the visibility, we're cannibalizing the idea that we care about the data structures. 
   2. Future of V3: 10:00 am Eastern (Austin)
      1. Members of the board re-envisioning task force here to discuss.
      2. Austin notes that a number of rim-based V3 standards have come up for reaffirmation and we're not sure what to do about those. Same question has come up about the V3 RIM. The technical situation is that we can maintain and update the RIM, but we don't have much demand to do so. There is not a need in FHIR to do it and no one else is pushing for updates. Austin asks Jean when there has been a request for updates? Jean states not since face to face harmonization has an attribute been added to the RIM. Austin: So we may still have the capability to update it but the demand is apparantly gone. The RIM-derived models (RMIMs) have an issue that the tooling stack is such that only one or two individuals may be able to update it. Jean reports that it's probably four or five. Austin: But overall our ability to update and maintain is questionable, although we do have more ability than previously and would likely have associated costs. 
      3. Clinical Statement CMETs came up for reaffirmation as well as a couple others. If reaffirmed there's an implication that we can update those standards. There's still a question about whether ANSI requires us to have update ability. Floyd: ISO references it as a normative standard and we need to determine any impact there. 
      4. Christof loves V3 and one of the most satisfying sessions are where we dive into V3. It's a big learning opportunity in terms of structural vocabulary, etc. There is a lot of work in there and a big achievement and has a value in itself even if it's not implemented.
      5. David: Everyone seems to think that if it's retired from ANSI it disappears; it won't disappear, it will become a retired standard that is still usable. There is no downside to deprecating it. Let's retire it and keep the RIM, but we don't need to keep it updated and fresh.
         
      6. Melva agrees with David and notes that it is no longer included in HL7 essentials education. 
      7. Paul: V3 is still used because CDA is used. The RIM is an information model of healthcare that is widely used. We have a history of not removing things from normative status. The cost of maintaining it is trivial but it is still used in the marketplace
      8. Austin: There are clear examples of normative standards in the V3 space that were built using draft CMETs, so they were built partially on sand. Paul: The CMETs were a plug in piece, and we have exactly the same thing in FHIR.
      9. Jean: We also have V2 IGs that are normative based on versions of V2 that are no longer normative or will be retired soon. When we looked at the RIM we tried to reach out to Oracle and Accenture to find out what they would like HL7 to do with the RIM. Oracle didn't seem future requirements to add to the RIM.
      10. Austin: So we need to consider the status of the RIM and the status of the RIM-derived products. There is a recent release of CDA with little or no uptake, and the next release will be a FHIR-based construct instead of a V3-based construct.
      11. Melva notes that it's still an HL7 standard even if it's not an ANSI standard. Paul is concerned it would no longer be a standard that you could claim conformance to.
      12. Jean: There's a belief that it's a house of standards and if you pull out one from the bottom then it makes your whole house not normative, and another belief that it's not the case. If standard X is based on standard A, does standard A have to remain normative?
      13. Melva: Pharmacy withdrew the IDMP standards a few years ago and retired them, but they are still ISO standards. 
      14. Jean: We need to know ANSI's position on normative specs that reference normative specs that are normative no longer.
          1. MOTION to table the issue until we have a response from ANSI: Wayne/Sandy
          2. VOTE: All in favor
          3. Melva: That seems to be about the RIM, but what about the standards coming up for reaffirmation? Jean: MnM has made a statement that unless you plan on updating it or it needs to be normative for regulatory purposes, V3 RIM-based standards should be withdrawn
          4. MOTION: TSC should adopt MnM's statement with TSC edits noted below in 4b on the withdrawal and reaffirmation of V3 messaging standards: Melva/Jean
             1. MnM Statement: HL7 workgroups should have a serious reason for reaffirming their V3 content. If no serious reason exists, then the workgroup should strongly consider retiring (withdrawing) their V3 content by the next expiration date. Serious Reason Questions to Consider: Does the committee intend to keep maintaining this content? or Does the implementation community require this content to maintain its ANSI certification status?
                
             2. TSC edits (in italics): HL7 workgroups should have a serious reason for reaffirming their V3 content. If no serious reason exists, then the workgroup should strongly consider retiring (withdrawing) their V3 content by the next expiration date. Serious Reason Questions to Consider: Does the committee have the intent and capability to keep maintaining this content? or Does the implementation community require this content to maintain its ANSI certification status?
                
          5. VOTE: Riki abstains. Remaining in favor.
   3. Re-training cochairs - what is our role? What are the training topics? - 30 min (Virginia)
      1. Training for new TSC members
         1. Virginia presents
         2. There was a 3 part live webinar for new cochair training that included how to facilitate/lead a WG (AMS), Intro to HL7 processes (Melva), and review and reflection (Virginia). Recordings are available and the cochair guide was updated. May re-do before Jan meeting.
         3. What training is missing right now?
            1. Amit: If you're not a cochair, can you attend this training? Virginia confirms that yes, non cochairs can attend and watch the recordings. Directs to the HL7 Essentials page on Confluence
            2. Paul suggests that it's a current cochair refresh as well as new co-chair training. So many processes and procedures are changing (balloting, new PSS process, vocab, org structure, updates to GOM) that everyone needs training.
            3. Austin notes that one problem is it's optional. People should have to assert in their nomination that they've attended or viewed the training; if it's optional, people will opt out.
            4. Josh echoes that it should be a requirement, but asking people to review the same content every year would not be idea. Breakout groups might be more useful.
            5. Paul suggests including facilitators as well. Virginia notes it would be open to non cochairs
            6. Christof mentions that some people are not happy with the balance between US Realm stuff and other stuff, so perhaps an overview could be given on global projects to open their minds
            7. Virginia: one way we've successfully gotten cochairs to be trained is to invite them to dinner. Is there anything in a virtual setting that could be a similar incentive? Austin suggests when we go back to in person, we could turn the dinner into more of a cochair training session since we'll continue our periodic cochair webinars as well.
            8. Amit suggests putting together videos that are short bite size trainings.
            9. Next steps:
               1. Virginia will summarize, discuss with Education and come back with recommendations
   4. Work Group Health metrics - task force update - 15 min (Riki)
      1. Initial recommendation is to use the existing metrics minus the SD related points
      2. Next four weeks the committee will look at new metrics
      3. Task force should have some recommendations for the June cochair meeting. Paul notes that we usually announce changes at one WGM and implement by the next. Austin suggests we could be more agile given the frequency of cochair meetings.
      4. Austin will announce current changes to metrics due to SD dissolution and that more changes will be forthcoming.
   5. Management group terms/reappointments - should there be term limits? - 30 min (Austin)
      1. Carry forward
   6. Increasing participant joy/reducing participant burden - 30 min (Virginia)
      1. Virginia presents
      2. Goal is to develop recommendations on how to improve the participation experience for leaders and participants so their work is more meaningful/less burdensome
      3. Focus on making participation a positive experience. 
      4. Virginia solicits and collects brainstorm answers in chat re: HL7 participation
      5. Virginia will compile and come back with a summary view
5. Terminology Service Group, continued
   1. Paul states he is completely supportive. Would prefer to see it as a management group, fundamentally. Paid staff could be ex officio members. They can set policy and it's managed by paid staff.
   2. Melva is supportive in principle. Concerns around the proposal where it implies the services group will take on the management of the infrastructure and making decisions around infrastructure and requirements. Melva agrees with the notion of them setting policy and the policy is executed by staff.
   3. Rob: When you have a committee that has operational commitments, I would expect that this group would be responsible for setting the goals, setting the set of expected deliverables, and then not being the group that has to get that done. There would be an alignment with those who dictate what staff does and make sure that there is agreement. The problem we've been having is inconsistency in the publishing process that supports FHIR and tooling development.
   4. Amit: Sounds like there are a couple of different challenges being blended together: jurisdiction/responsibility and bandwidth/execution. It would make sense to separate those. The people stating what needs to be done shouldn't be the people paid to do it.
   5. Christof proposes subgroups including external terminologies, HL7-owned terminologies, value sets, tooling/publishing, and more as needed such as SNOMED with release management
   6. Rob: I share concerns about the people being paid deciding what they're going to do, which is already happening. TSC can say this is what needs to happen, even if it needs board approval or it needs funding. We can suggest it regardless. Maybe we need to have someone knowledgeable on how we're delivering in FHIR to provide guidance. We are not delivering on the requirements to meet FHIR needs or the other product families. Need to figure out how we communicate the urgency to the board.
   7. Wayne: In terms of next steps, a statement from the TSC to the board would be useful. A problem I have with the proposal is we're looking at terminology in a vacuum as a unique problem - in reality it's much deeper than that. Terminology is a key part of it but it's not the only part. Austin states that the task force can put together a statement.
   8. Riki: We need a really clear statement on the consequences of not doing this
   9. David: We've done a lot with a little for too long. What's being presented here is creating a whole new infrastructure, which goes far above what we need as a management group. We need to invest in infrastructure but creating a TSG that has membership of the product management groups and FHIR-I creates a duplicate governance structure.
   10. Christof: If we forget the HL7 owned terminologies for a moment, if we don't have funding and capacity to integrate that with tooling, we could motivate the external organizations to offer those services to connect with the HL7 way of publishing those. What would remain is more like a branch of FHIR-I or FMG to give some rules on how to integrate those external artifacts into FHIR resources. Could be along the lines of a FHIR accelerator.
   11. Rob: We do need to think outside the box but there are immediate needs. Well over 50% of the content in the guides is HL7 content in terms of terminology. The TSC does not function as a governance body across the different product families. It does serve as administrative governance for things that are administratively important across product families. FHIR-I and FMG operate independently. If simplification is at issue, take terminology governance out of TSC. If we think the TSC will be able to determine the details necessary to deliver on the service functionality, it will take up all our time. 
   12. Next steps
       1. Subgroup of TSC members will participate with the exploratory group and formalize it as a task force. Need to increase the calls from twice a month to weekly.
       2. Group will develop a proposal to the board on what to do around this issue
       3. Task force will examine how this works in FHIR
       4. Rob and Paul volunteer for the committee
       5. Anne to send Doodle to TSC list
6. Adjourned at 1:00 pm Eastern