See PRs for description.
What if we don’t add a new category, and define a new attribute in the RM to tell what context the persistence is about.
The default value is persistent within the EHR.
Then we add the possibility to add a terminology which defines the contextual persistence semantic as i.e. a DV_CODED_TEXT with values like episode of care, periode of care, referral, a surgery, etc.
Persistence will be defined different in the systems and applications. Thus it’s needed to leave the different scopes to the application.
“Persistence will be defined different in the systems and applications. Thus it’s needed to leave the different scopes to the application.“
Agreed but that’s why I’m not sure that having a ‘flavours of persistence’ attribute takes us much further - I can tag it as ‘period of care’ but how that is actually defined and implemented will be very context/application-specific, so in terms of documenting the behaviour vs. ‘just ‘episodic’ I’m not sure it takes us much further.
Let's find out from other CDR implementers how they are using/enforcing ‘persistent’ right now. I’m not totally opposed to your suggestion!! We could do both, of course - add ‘episodic/contextual’ and a new attribute for the reason.
Perhaps we should just vote on it.
I think does have a useful point here, but I’m inclined to finish this CR as originally envisaged, even if it turns out a) implementers don’t use ‘episodic’ much in the end and/or b) we decide we want some extra ‘temporal_context’ attribute added.
The mere fact that it is not obvious now means that we don’t have enough hard evidence or more solid model idea to propose.
Paragraph 8.1.3 could also use an update now that there’s episodic compositions.
e.g. “two kinds of Composition are identified: persistent and event” is outdated. And so is the list of persistent data types. E.g. pregnancy is now probably an episodic composition.