Stream: argonaut
Topic: Australian clone of Argonaut
Grahame Grieve (Apr 17 2018 at 23:29):
I have just started working on an Australian adaptation of Argonaut R2. it will be the base that US vendors use when they implement Argonaut in Australia.
Grahame Grieve (Apr 17 2018 at 23:30):
the general intent is that it will be as closely based on the existing Argonaut specification as possible, with necessary changes for the Australian regulatory context
Grahame Grieve (Apr 17 2018 at 23:31):
right now, I need to decide what version of FHIR it should be based on. From an Australian perspective, our answer would be: R3 right now, but R4 on publication, on the anticipation that everything else is R3 and will be R4 in Australia.
Grahame Grieve (Apr 17 2018 at 23:31):
that anticipates that argonaut US will update to R4 when we publish it.
Grahame Grieve (Apr 17 2018 at 23:32):
I'm interested in opinions - @Jenni Syed @Dennis Patterson @Danielle Friend @Isaac Vetter (@Brett Esler )
Isaac Vetter (Apr 18 2018 at 02:55):
Starting with R3 is very reasonable. Depending upon when the Australian guide is published, R4 probably makes sense as well. Ultimately, the goal is to support normative resources. :octopus: @Danielle Friend - you agree?
Grahame Grieve (Apr 18 2018 at 06:38):
ok: first clone: http://build.fhir.org/ig/hl7au/argonaut-au/index.html - this is simply argonaut updated to run against R3
Josh Mandel (Apr 18 2018 at 12:42):
How is different than US Core, which we notionally think of as what the Argonaut profiles led to in R3?
Grahame Grieve (Apr 18 2018 at 12:54):
So far, all it is is updating Argonaut to R3. For US, then it needs to be refactored to be based on US Core (R3), so that the base definitions are directly those in US core, and the conformance requirements are in Argonaut. I'll be doing the equivalent of that for AU Core profiles now
Eric Haas (Apr 18 2018 at 13:19):
What does " and the conformance requirements are in Argonaut" mean? US Core has its own updated set of
capstatements too?
Grahame Grieve (Apr 18 2018 at 13:22):
well, I can't speak for US, really, but I do not think that all uses of FHIR in USA are by EHRs.
Grahame Grieve (Apr 18 2018 at 13:26):
so we will keep the conformance requirements for EHR access in the EHR guide
Grahame Grieve (Apr 18 2018 at 13:26):
for Australia
Eric Haas (Apr 18 2018 at 13:33):
I am still confused since USCore and Argonaut are scoped the same.
Grahame Grieve (Apr 18 2018 at 13:39):
then what is the base US implementation guide that applies to all use of FHIR in USA?
Grahame Grieve (Apr 18 2018 at 13:43):
more generally, we've discussed this before. Argonaut (scope of EHR data access) is able to make agreements about the data that wider usage cannot e.g. there must always be a family name
Eric Haas (Apr 18 2018 at 15:46):
That is a good question, but I don't think such a guide is possible since the scope is too broad. Once FHIR touches EMRs then I think is Argo/US Core. But for fitbit data or my consumer nutrition app its wide open.
Grahame Grieve (Apr 18 2018 at 22:14):
ok, I've updated it a little, and I've added a comparison link. e.g. the page http://build.fhir.org/ig/hl7au/argonaut-au/StructureDefinition-argo-patient.html has a link to https://services.w3.org/htmldiff?doc1=http%3A%2F%2Fwww.fhir.org%2Fguides%2Fargonaut%2Fr2%2FStructureDefinition-argo-patient.html&doc2=http%3A%2F%2Fbuild.fhir.org%2Fig%2Fhl7au%2Fargonaut-au%2FStructureDefinition-argo-patient.html
Eric Haas (Apr 18 2018 at 22:54):
So you are comparing across jurisdictions and versions? I'd make that clear.
Grahame Grieve (Apr 18 2018 at 23:00):
yes we should make the kinds of differences we have clear, you're right
Jenni Syed (Apr 19 2018 at 01:07):
Late to the response... but agree that R3+ is a good target.
Danielle Friend (Apr 19 2018 at 22:48):
I also agree - R3 and above makes the most sense.
Rob Hausam (Jun 08 2018 at 13:52):
I'm bringing this back up because it's also a question that we have for International Patient Summary (IPS). Picking up on Danielle's last comment, how should we understand and apply "R3 and above"? To be working toward future IPS compatibility/alignment with Argonaut, should we do our work now in R3 and align there, and then plan to update to R4 at some future point tracking reasonably closely with Argonaut (whenever that point may be)? Or would it be best to work in R4 now and focus on aligning as closely as possible with Argonaut there in anticipation of the future convergence on R4 for both? I'm sure there may not be a single right answer for this, but I'm definitely interested in hearing thoughts/opinions.
Grahame Grieve (Jun 08 2018 at 20:16):
you have to anticipate future conversion towards R4 I think
Rob Hausam (Jun 08 2018 at 20:22):
that was my expectation - but which would likely be the best strategy for anticipating it?
get ahead of it by moving our work on IPS to R4 now and try to align as well as possible across versions
or stay with STU3 for now and follow Argonaut's progress closely and try to move to R4 at the same time
Grahame Grieve (Jun 08 2018 at 20:22):
you're already a version different anyway. So go forR4 as close a you can and just watch for conceptual alignment
Rob Hausam (Jun 08 2018 at 20:25):
sounds pretty reasonable to me - that's where I'm leaning
Last updated: Apr 12 2022 at 19:14 UTC
