Stream: patient empowerment
Topic: Diagrams
Lloyd McKenzie (Dec 03 2020 at 00:10):
Breaking this out from the project-patient corrections chat. In the longer term, it would be good to migrate the state transition diagrams to plantUml. It'll allow us to maintain the source as text files in Git and will also allow automatic generation of the diagrams as SVG - which will be searchable.
Jose Costa Teixeira (Dec 03 2020 at 19:04):
I have a conflict with the PE calls and sadly I'm missing out. I can assist in this if needed
Dave deBronkart (Dec 03 2020 at 19:06):
Thanks, @Jose Costa Teixeira ! cc @Virginia Lorenzi @Debi Willis
We recorded the call and the massive chat log. Will publish those links tonight.
Jose Costa Teixeira (Dec 03 2020 at 19:15):
In case anyone wants to see how to add a diagram to the IG, this is how it's done:
- add plantuml source in images-source
- update your html page and refer to the svg which will be generated from the source.
if the html page is new, add it to the implementationguide and to the menu if you want (This is optional).
Here's an example:
https://github.com/costateixeira/devdays-covid19-vaccine/commit/a6a818f930960114457a19d33bfebdc869699003
Jose Costa Teixeira (Dec 03 2020 at 19:16):
I'll try to catch up with the group because of 2 topics I'm working on:
- Data access control
- medication lists/summaries/records (whatever we want to call those)
Abbie Watson (Dec 03 2020 at 20:45):
Do we have the plantUML for the existing Task state diagram? On one hand, I think it's important that we build off the existing state diagram if we're going to use the TaskStatus value set and profile/extend it. On the other, we do need to think about novel value sets specific to the patient corrections workflow.
PatientCorrection.Task.StateMachine.png
John Moehrke (Dec 03 2020 at 20:54):
do you have the plantuml code for this diagram?
Vassil Peytchev (Dec 03 2020 at 21:15):
it's important that we build off the existing state diagram if we're going to use the TaskStatus value set and profile/extend it.
Note that the TaskStatus value set is Required. The state of the Task will be represented as a tuple of Task.status, Task.businessStatus, and Task.statusReason - it is the last two that can have any value set that the Profile comes up with.
Abbie Watson (Dec 03 2020 at 21:37):
Correct. But we can profile the TaskStatus down, and highlight key states that the IG uses, and which ones are optional to the IG though, right?
Lloyd McKenzie (Dec 04 2020 at 00:01):
The source for the Task state machine is not PlantUML unfortunately. It was produced about a year before we started using PlantUML. However, if someone were to put in a change request, we could try to change it...
Dave deBronkart (Dec 04 2020 at 12:44):
Lloyd McKenzie said:
The source for the Task state machine is not PlantUML unfortunately. It was produced about a year before we started using PlantUML. However, if someone were to put in a change request, we could try to change it...
Sounds like a good idea, but I'm not the right sort of person to do that, because if any questions are asked I'll topple like a forbidden statue :smile:
Jose Costa Teixeira (Dec 04 2020 at 14:36):
https://jira.hl7.org/browse/FHIR-29933
Virginia Lorenzi (Dec 07 2020 at 19:38):
Jose would you like to participate :smile: in the connectathon? @Mikael Rinnetmäki @Timon Grob Jose/Timon/Mikael - would like a perspective on our designs from your country. And are any of you able to participate in the connectathon?
Virginia Lorenzi (Dec 07 2020 at 19:50):
The novel one includes business status - didn't expect to extend value set on Task (don't know if we could). We can't change the values of Task.status we can only constrain the value set as far as I understand. Looking back at the original diagram and based on our discussions I now see a use for Cancelled and On-Hold. I do not see a use for Draft, Received, or Ready and am not sure about Failed. Maybe we use the base diagram and explain what the statuses mean or show their use in the context of corrections. @Michele Mottini @Lloyd McKenzie @Vassil Peytchev
Lloyd McKenzie (Dec 07 2020 at 23:09):
Draft is for a Task that you're working on but isn't yet ready to be processed. Relevant if the 'app' that's authoring the correction request doesn't have its own persistence layer. Agree Received and Ready aren't relevant. 'Failed' would be in a situation where a provider says "we will make this correction" and then discovers there's a barrier that prevents them from doing so (e.g. if there's a legal hold placed on records that keeps them from being altered or something).
Timon Grob (Dec 08 2020 at 09:04):
@Virginia Lorenzi I am intending to join the connectathon.
Jose Costa Teixeira (Dec 08 2020 at 14:07):
I'd like to participate in the connectathon but I don't have a system to test..
Abbie Watson (Dec 08 2020 at 14:59):
@Jose Costa Teixeira - We may have a surprise in the works for folks like yourself who want to participate but don't have a system to test. Stay tuned. We're working on something to address that need/interest.
Dave deBronkart (Dec 08 2020 at 15:00):
We?? :wink:
Abbie Watson (Dec 08 2020 at 15:02):
Stay tuned. I got a little somethin-somethin in the works, beyond the test server.
Virginia Lorenzi (Dec 10 2020 at 03:51):
@Jose Costa Teixeira Do you program? A simple client that can post and read task resources would do but even if you can't I m sure we could use your input If you or you know anyone who has requested their medical record to be corrected let us know the process. Similarly if you know a clinician or hospital folk that you can ask about the process on their side. Want to make sure this would work in Belgium!
Jose Costa Teixeira (Dec 10 2020 at 07:01):
Hi. I think if I look at the code I type and say "I do program", I'll go to jail or worse, but I could do something. I am still learning what is the scope - to me, one thing that needs to be standardized is the patient's request - regardless of whether that patient's request has merit. That is something I'd like to see working. Correction of demographics, or correction of procedure results, or statements of medical errors that are not in the record..
Then we can discuss and try many workflows which IMO will not be standardized.
Last updated: Apr 12 2022 at 19:14 UTC
