FHIR Chat · Idealistic/generic Condition resource · implementers

Stream: implementers

Topic: Idealistic/generic Condition resource


view this post on Zulip Mattias Flodin (Apr 26 2018 at 09:39):

I'm mapping data from an EHR system to Encounter. The source system contains a list of diagnosis codes for each encounter. They're standardized codes used for statistics and billing, and there's no additional information apart from the list of codes. Would it be poor use of FHIR to create a single Condition resource for each canonical code, and have multiple encounters for multiple patients refer to the same Conditions? I get the sense that the idea is to create a new Condition instance for every encounter or episode, but it's not clear to me if that is the case. If it is, I'd have to make up arbitrary unique Condition IDs for every time a diagnosis is referenced.

view this post on Zulip Grahame Grieve (Apr 26 2018 at 09:56):

in general making up new ids in a facade to the database is an anti-pattern.

view this post on Zulip Mattias Flodin (Apr 26 2018 at 11:06):

Then what would you suggest Grahame?

view this post on Zulip Mattias Flodin (Apr 26 2018 at 11:07):

The fact is, I've realized, I might not have a lot of choice in this case since Condition has a mandatory subject patient. So it would have to be one instance per patient at least, unless I make up an imaginary generic patient also.

view this post on Zulip Grahame Grieve (Apr 26 2018 at 11:37):

so all you have is a code?

view this post on Zulip Mattias Flodin (Apr 26 2018 at 11:49):

Yes

view this post on Zulip Mattias Flodin (Apr 26 2018 at 11:50):

and a description, e.g. J50E = "Cellulitis"

view this post on Zulip Grahame Grieve (Apr 26 2018 at 11:52):

ok so I think that this is the kind of situation for which we defined contained resources

view this post on Zulip Mattias Flodin (Apr 26 2018 at 11:53):

Ah, that's an interesting angle. Sounds like a good idea.


Last updated: Apr 12 2022 at 19:14 UTC