FHIR Chat · Triggering consequences · workflow

Stream: workflow

Topic: Triggering consequences


view this post on Zulip Grahame Grieve (Jan 22 2018 at 23:28):

I'm looking at the eICR connectathon track for this weekend.

view this post on Zulip Grahame Grieve (Jan 22 2018 at 23:28):

one of the bonus points is ":Bonus point: eICR is created in response to evaluating trigger codes against patient data. Triggering may occur if an EHR diagnosis or problem code matches a code in the diagnosis trigger code value set, if a lab result code matches a code in the lab result trigger code value set, or if a lab order code matches a code in the lab test order trigger code value set."

view this post on Zulip Grahame Grieve (Jan 22 2018 at 23:28):

abstractly:
- client posts/puts content to the server
- server instructs client to take a follow up action

view this post on Zulip Grahame Grieve (Jan 22 2018 at 23:29):

HATEOS, presumeably.... but how?

view this post on Zulip Grahame Grieve (Jan 22 2018 at 23:29):

I don't think we have a canned answer to this?

view this post on Zulip Vassil Peytchev (Jan 23 2018 at 03:34):

Has HATEOAS ever been considered for workflow? Maybe as part of services?
The current answer, I believe, is that the client either need to do polling, use subscription, or the server post something to the "client" to indicate the subsequent action.

view this post on Zulip Grahame Grieve (Jan 23 2018 at 15:46):

we've talked about HATEOS a little. but there's no magic; you need a language to describe the actions... and then you get back to discussing the balance between meta and data level actions.

view this post on Zulip Grahame Grieve (Jan 23 2018 at 15:47):

but this is clearly a case for a header in the return: the server ..suggests... that the client now [something] this URL...


Last updated: Apr 12 2022 at 19:14 UTC