May 2021 - HL7 WGM - Thursday Q3 Minutes

From Davera in the chat: 

We use value sets to support analytics in research

the requirement being that the research / analyses are reproducible

so there is a requirement for persisting value set resources

and these are not currently in the VSAC or quality measure libraries. TSO there's an alternative use case

Where did we land at the end of the last quarter?

NamingSystem will continue to be used for performing translations (e.g. from uri's to oids). This could be done through CodeSystem, but do not want to force systems to support both resources.

GG: its reasonable to say that identifier translations would have to examine CodeSystems. 

Davera G: described her use case - adhere to the principles of open science (different than traditional research). One of the important principles is reproducibility of scientific results. They are publishing code systems, value sets, and sometimes sets of relationships. Some of this work is going on in FHIR (e.g. ontologies - xxxxx). 

They are managing DOI - document object identifiers. Using this to assign/manage? DOIs https://zenodo.org/ 

https://academic.oup.com/jamia/article/28/3/427/5893482     This community also has a need to manage identifiers as they are creating and managing concept/value sets.

GG: the key question is about the publishing pathway - suggests submitting her questions on Zulip

Workflow process for use of internal code system identifiers in FHIR and beyond

Push back from some attendees on establishing a code system identifier and name in THO, IG authors should be able to choose to reserve a THO namespace. (referencing the 2 step process)

THO inclusion/"registration" should not be mandatory. IG authors should be able to have their own content defined elsewhere but expose the contents in THO.  Not agreement on this. 

the url shouldn't matter - its just an identifier - 

LM contends that creating a NameSpace early on creates a burdensome overhead. Pre-judges. 

TK: creating the NamingSystem to establish a url and a name is not burdensome. This will be a pro-forma process - no consensus review. We are using the THO website in this capacity as a sign board, posting to notify the community.  These will be displayed as Pending Code Systems.

LM; concern about putting burden on implementers who are playing around. 

LM: is not against transparency. 

LM: does tx.fhir.org have access to all code systems across all IGs? 

GG: not really - 

GG: could do an analysis across IGs and THO and people could examine and address - this would be part of every build. 

Frank M: getting feedback from the community is valuable - ensuring the UTG process doesn't happen too late, encourage IG developers to start terminology efforts early on, rather than later. Avoid learning too late through the review process that something different should have been done.

LM: there is a presumption that early in the design, people will be looking at external codes. There is zero expectation that they will be doing this. People won't look for codes when preparing for a connectathon.  Maturity level 3 is where the quality criteria indicate the terminology work begins. 

CC comment: this is an issue - FHIR itself does not promote good terminology practices. We're supporting 

LM: things should not be in UTG until it is well thought out. Use names in UTG for things that are real. Don't use it for discovery of works in progress. There are other ways to promote discoverability.

RD: are we all agreed that HL7 terminology artifacts code system and value set resources should be in THO? 

LM: he does not agree with that. Vocab, M&M and FHIR-I have agreed that there are certain terminologies (datatype = code) will not be in THO.  Those will live in the IG only. 

RD: what namespace do the FHIR core ones use? either fhir or IG namespace

************

FHIR IG Quality Project Proposal

Carol M: would terminology practices be part of this guide

LM: its reasonable for Vocab to be involved from a perspective of vocabulary practices. Vocab will also have input from the THO perspective. 

CM: informing IG implementers about CodeSystem licensing requirements, consistently presenting terminology artifacts, etc. 

TK: there are several JIRA tickets entered with requested enhancements for IG publishing/tooling (HRSC).  

Once the ticket is vetted, it will be moved to the appropriate queue. LM mentions that these tickets end up in Git. 

Triage responsibility is not yet determined. 

Andrea M: ??? accessibility has been an issue in Canada for some time. Is that what we're talking about here?

LM: possibly. funding necessary.