Stream: cds hooks
Topic: valueset-detectedissue-category
Erich Schulz (May 26 2016 at 12:15):
so... I'm looking at this: https://www.hl7.org/fhir/valueset-detectedissue-category.html
Erich Schulz (May 26 2016 at 12:16):
it seems like a something pulled straight out of a pharmacy system
Erich Schulz (May 26 2016 at 12:17):
they all seem like ok issue categories but it is very incomplete
Erich Schulz (May 26 2016 at 12:18):
the issue is this is currently classified as preferred
Erich Schulz (May 26 2016 at 12:18):
i'm kinda thinking it needs downgrading to example ??
Erich Schulz (May 26 2016 at 12:20):
the sort of categories i'd like a CDSS to report are things like "missing examamination finding" (ie listen to their chest doctor!)
Erich Schulz (May 26 2016 at 12:21):
and "test not required" - (ie no this 42yo woman does not need and ECG or a coag profile)
Erich Schulz (May 26 2016 at 12:22):
and also "missing test result" (you've told me this person is having major surgery and and cant walk up stairs, we need some kind proper cardiac assessment)
Erich Schulz (May 26 2016 at 12:24):
also not sure where blood test - drug warnings would go... "climbing creatinine|LFTs" with ongoing NSAID|paracetamol
Erich Schulz (May 26 2016 at 12:42):
looking at "operation outcome" - arguably some of these things could go in there but the issue type codes dont really seem relevant
Lloyd McKenzie (May 26 2016 at 23:38):
Preferred says "we'd encourage people to use these codes if they apply and there isn't a strong need to use something else". Preferred doesn't say the code set is necessarily complete.
Erich Schulz (May 26 2016 at 23:39):
oh right sweet
Erich Schulz (May 26 2016 at 23:42):
but the wording says If the data type is CodeableConcept, then one of the coding values SHOULD be from the specified value set, but another code and/or text can be used in its place.
Erich Schulz (May 26 2016 at 23:43):
I'm still thinking SHOULD be suggests that FHIR.org reckons the existing list is adequate
Erich Schulz (May 26 2016 at 23:43):
and it really isnt (IMHO)
Erich Schulz (May 26 2016 at 23:45):
inadequate in that it is clearly derived from pharmacy and lab ordering
Josh Mandel (May 27 2016 at 00:03):
BTW, this isn't quite a CDS Hooks issue.
Erich Schulz (May 27 2016 at 00:11):
oh - is there a better stream @Josh Mandel ? this resource seems to be primary mechanism by which a CDS could express itself... but happy to be directed :-)
Lloyd McKenzie (May 27 2016 at 03:01):
By default, spec issues should be in the Implementer's stream
Erich Schulz (May 27 2016 at 04:02):
cool thanks. i had a little read on the cds hooks project - looks very interesting @Josh Mandel
Last updated: Apr 12 2022 at 19:14 UTC
