Stream: IPA
Topic: IPA Status
Elliot Silver (May 09 2021 at 20:22):
Hi all, is there any current work on the IPA IG? The last update I see is February 2020. Is the project still alive?
Grahame Grieve (May 09 2021 at 21:36):
it's... resting... I have no time. But strategically it's still important. I think it should be folded into IPS though. That's on my list of things to do
Elliot Silver (May 10 2021 at 02:35):
Ok. I’m working on a project that might have an interest in helping move this forward.
Rob Hausam (May 10 2021 at 11:45):
@Elliot Silver @Grahame Grieve I think that folding IPA into IPS would make sense, and we should start discussing how to make it happen. The GDHP IPS work definitley needs this (or at least something like it). We're talking about having a breakout session on IPA and IPS in the IPS Connectathon track, and that should be a good opportunity for further discussion and strategizing on it.
Elliot Silver (May 11 2021 at 01:58):
I don't think I'll be able to join that discussion -- I wasnt' going to go to the Connectathon, and it looks like registration is closed. I'll touch base after for an update.
Elliot Silver (May 11 2021 at 02:00):
@Grahame Grieve , are you taking feedback on IPA? (I don't see a Jira project for it.)
Grahame Grieve (May 11 2021 at 02:18):
sure. it should become am HL7 project, but I've been too busy with other stuff
Elliot Silver (May 11 2021 at 02:21):
Understandable. It isn't obvious to me why Patient.link is constrained from 0..* to 0..1. It would be helpful to include an explanation, or to relax the constraint.
Grahame Grieve (May 11 2021 at 02:30):
I see that I didn't document that yet. It came out the presentation I did around patient merge at DevDays. The idea is that for patient access, internal links etc are not pertinant; what Patient.link is used for is the sole purpose of directing the software to the correct record for the patient
Grahame Grieve (May 11 2021 at 02:30):
so there would be a must-support on it: if you see this, you're not using the right record. Hence the desire for simplicity on this, and ruling out any other uses of the link
Elliot Silver (May 11 2021 at 02:33):
hmm. Wouldn't that be an argument to constrain it to 0..0? It doesn't matter what the back-end system does, by the time it gets to a patient, there is only one relevant record?
Elliot Silver (May 11 2021 at 02:34):
I may need to track down that presentation.
Grahame Grieve (May 11 2021 at 02:34):
no because there'
Grahame Grieve (May 11 2021 at 02:35):
it's a question of how you tell the client what's going on. There's a strong - though not convincing - argument that you don't just redirect to the right record, you tell the client that's what it needs to do. Hence, link is 0..1
Grahame Grieve (May 11 2021 at 02:36):
Grahame-Grieve-Managing-Patient-Merge.pptx
Elliot Silver (May 11 2021 at 02:36):
https://www.youtube.com/watch?v=8BOGePClMrE
Elliot Silver (May 11 2021 at 02:37):
So it's the FHIR equivalent of a 301 error. "Go look over there."
Grahame Grieve (May 11 2021 at 02:37):
yes. a 301 might be right, but it leaves you with an information deficit.
Grahame Grieve (May 11 2021 at 02:38):
This is not a resolved thing, btw, at all, even for just me. My intention with IPA was to drive us to get a single answer for this. Because it's one of the real unsafe areas of patient access
Elliot Silver (May 11 2021 at 02:38):
Conceptually, a 301, not actually a 301.
Elliot Silver (May 11 2021 at 02:38):
Right.
Elliot Silver (May 11 2021 at 02:39):
OK, thanks.
Rob Hausam (May 11 2021 at 02:41):
Do we have or need to have plans for also discussing this at the WGM? Will you be at the WGM, @Elliot Silver?
Elliot Silver (May 11 2021 at 02:42):
I wasn't planning on it, but I can re-evaluate.
Elliot Silver (May 11 2021 at 02:44):
Are you thinking discussion of IPA or patient merge?
Grahame Grieve (May 11 2021 at 02:45):
I do think this should be on our agenda somewhere central. But Argonaut has to be in the picture
Elliot Silver (May 11 2021 at 02:48):
Let me check if we'll have anyone at the meeting already.
Rob Hausam (May 11 2021 at 02:49):
My thought for discussion in the Connectathon was focusing on IPA and IPS - including potentially folding it in, as Grahame mentioned.
Elliot Silver (May 11 2021 at 02:51):
I'll check tomorrow whether I can swing either the meeting or Connectathon, and check with HL7 if they'll take a late Connectathon registration (I don't see why not--I'm not bringing a test system.)
Elliot Silver (May 11 2021 at 18:04):
@Rob Hausam , looks like my best option for discussion will be a regular Patient Care call.
Rob Hausam (May 11 2021 at 18:45):
I don't think we've decided on an owning WG for IPA (@Grahame Grieve?). But discussing in Patient Care makes sense for the relationship with IPS - and potentially folding it in. So I think that should work. We will give an update on wherever we get to with discussions in the Connectathon and WGM.
Grahame Grieve (May 13 2021 at 00:13):
PC decided in Sydney to own it; I just haven't submitted it to them as a project
Grahame Grieve (May 13 2021 at 00:13):
guilt is with me
John Moehrke (May 17 2021 at 17:52):
IHE already has Implementation Guides that are international and cover this space -- See IHE-QEDm and others
John Moehrke (May 17 2021 at 17:52):
Here is a good overview from IHE that has links to the various deeper topics https://profiles.ihe.net/ITI/HIE-Whitepaper/index.html
John Moehrke (May 17 2021 at 17:52):
there is efforts to convert that IHE QEDm to the IG builder format.
Mikael Rinnetmäki (May 23 2021 at 14:56):
Noting here too that there's no a PSS for IPA at https://jira.hl7.org/browse/PSS-1803.
Rob Hausam (May 23 2021 at 22:30):
Yes, we discussed it in the IPS and IPS session at the Connectathon and Isaac Vetter volunteered to submit the "project proposal" (the first step now before getting to the actual PSS). I have a few additions/edits that I've thought of and we'll have an opportunity to further flesh some of it out for the actual PSS.
Vassil Peytchev (May 27 2021 at 19:49):
Is the first discussion at Noon Eastern (4pm UTC) on Wednesday, June 2?
Rob Hausam (May 28 2021 at 11:22):
@Vassil Peytchev Yes, except that it's 11:00 AM Eastern (15:00 UTC) on June 2 (here). I'm volunteering that we start by dedicating time for IPA on the regular weekly Wednesday IPS calls (11:00 - noon Eastern). I expect that we will want to extend the call time (likely to 90 min.) to be able to address what we need for IPA, but we can discuss that when we meet next week.
Rob Hausam (Jun 01 2021 at 22:19):
Just a reminder - we'll plan to have some initial discussion of IPA tomorrow on the IPS call at 11:00 AM Eastern (15:00 UTC). I'm working with Grahame to get a new GitHub repo for IPA set up in the HL7 space - that should be there soon.
Rob Hausam (Jun 02 2021 at 11:40):
The new IPA GitHub repository is here. I've made some additional commits so that it builds successfully with the latest IG Publisher (with 2 errors, plus warnings for missing examples).
Rob Hausam (Jun 09 2021 at 14:59):
The IPS/IPA call is happening now at https://us02web.zoom.us/j/5328571160.
Elliot Silver (Jun 09 2021 at 15:00):
Unfortunately, I won't be able to make it today.
Jens Villadsen (Jun 30 2021 at 14:10):
@Rob Hausam - https://github.com/HL7/fhir-ipa/blob/637e913331d031abbff6198ab7317a5ea00ca664/input/resources/CapabilityStatement-ipa-server.json#L46 why does IPA say something specific to US Core?
Jens Villadsen (Jun 30 2021 at 14:14):
Is it a hard requirement that SMART SHALL/MUST be followed for IPA (https://github.com/HL7/fhir-ipa/blob/main/input/pagecontent/access.md)?
Rob Hausam (Jun 30 2021 at 14:18):
It's probably a copy-paste oversight, I expect. We definitely should adress it. I'm not sure that IPA is actually in Jira yet (though it could be, so we should check). If it isn't, we could log a GitHub issue in the meantime so we don't lose track of fixing this (I'll try to add a note in the minutes, too - even though we're not actually doing IPA today).
Jens Villadsen (Jun 30 2021 at 16:26):
@Rob Hausam here you go: https://github.com/HL7/fhir-ipa/issues/3 and https://github.com/HL7/fhir-ipa/issues/2
Rob Hausam (Jun 30 2021 at 16:58):
Great. Thanks, @Jens Villadsen.
Jens Villadsen (Jun 30 2021 at 16:59):
Np
Grahame Grieve (Jun 30 2021 at 19:25):
I missed removing the US core reference there, but it is definitely a hard requirement for SMART. I think IPA would be meaningless without that
Rob Hausam (Jun 30 2021 at 19:33):
I'll look at it further so I know the details - and I guess we'll need to think further about the implications.
Jens Villadsen (Jun 30 2021 at 21:31):
FYI - SMART itself references US Core AFAIK
Josh Mandel (Jun 30 2021 at 21:38):
References like "mentions the existence of"? Or references like "takes a dependency on"?
Josh Mandel (Jun 30 2021 at 21:39):
SMART is a universal realm spec. The former kind of reference seems innocuous to me, and it's good context, but now's the time to raise an issue if there's something we can improve.
Grahame Grieve (Jun 30 2021 at 21:41):
I don't see any reference. There are some extremely US perspectives in there, but nothing actually US specific
Jens Villadsen (Jul 01 2021 at 07:44):
its a textual reference
Jens Villadsen (Jul 01 2021 at 07:45):
Jens Villadsen (Jul 01 2021 at 07:45):
https://docs.smarthealthit.org/ is the first hit on a Google Search (at least for me)
Grahame Grieve (Jul 01 2021 at 08:39):
http://hl7.org/fhir/smart-app-launch
Jens Villadsen (Jul 01 2021 at 08:41):
fair enough - it does start out mentioning the argonaut project in the first section, but I guess thats fair since it has been the main catalysator for it
Grahame Grieve (Jul 01 2021 at 08:42):
yes. the documentation is definitely slanted ot US and that's one way, but it doesn't actually depend on it
Jens Villadsen (Jul 01 2021 at 08:43):
right
Jens Villadsen (Jul 01 2021 at 08:48):
I may still be of the oppinion that mandating the use of SMART is 'setting the bar too high'. For some authorities it may not be an option to support SMART eventhough it brings a lot of clarity and standardization to the table.
Grahame Grieve (Jul 01 2021 at 08:48):
why would it setting the bar too high?
Jens Villadsen (Jul 01 2021 at 08:50):
They may already have existing auth flows that they arent willing to change
Jens Villadsen (Jul 01 2021 at 08:50):
but it is mainly (semi qualified) speculations from my side
Grahame Grieve (Jul 01 2021 at 08:53):
then they won't be able to conform to IPA. But smart app launch is a low bar to clear - client hands over to web site for whatever auth the website wants to do.
Jens Villadsen (Jul 01 2021 at 08:59):
sure
Rob Hausam (Jul 07 2021 at 14:33):
I'm not able to get in to the Patient Care Zoom meeting today. I send an alternate in a moment.
Rob Hausam (Jul 07 2021 at 14:35):
Let's use this: https://us02web.zoom.us/j/8488242144?pwd=ZDNjeWhkenYxb09adlZiRXoxL1N3dz09
Mikael Rinnetmäki (Jul 07 2021 at 15:12):
http://build.fhir.org/ig/HL7/fhir-ipa/branches/main/index.html
Mikael Rinnetmäki (Jul 14 2021 at 14:30):
@Ardon Toonstra drawing your attention to this, since I think you scheduled the "Argonaut International" session back in FHIR DevDays Amsterdam at 2019, correct?
Mikael Rinnetmäki (Jul 14 2021 at 14:31):
Also, I've heard @Alexander Henket might be a good contact in the Netherlands regarding International Patient Access. True?
Last updated: Apr 12 2022 at 19:14 UTC
