Stream: terminology
Topic: Profiles vs terminology
Domenico Corapi (Mar 06 2020 at 16:10):
We have been having a discussion internally on the proliferation of profiles. The problem is we ended up with a proliferation of human created profiles of the type BloodFerritinProfile, BloodCreatineLevel, BloodCreatinineLevel etc.. that enforce the pair <code, unitofmeasure>. There's a similar pattern on medication statement.
One of the ideas being circulated is to use terminology to autogenerate or to validate these profiles (e.g. there's a triple <"Chloride measurement, blood", has_unit, "mmol/L"> that we can use too generate the profile, or to check that a blood pressure is well formed).
Has anyone done something similar or do you think this is the wrong approach (e.g. we can just use constraints instead of creating multiple profiles)
Grahame Grieve (Mar 06 2020 at 20:26):
yes we have and I think it's the right approach - you should only have different profiles when the shape of the data changes. It's on my to do list to make this manageable
Domenico Corapi (Mar 09 2020 at 12:05):
Great to hear that! Is there anything on confluence or some early drafts I can look at?
Grahame Grieve (Mar 09 2020 at 12:29):
no it's on my todo list right now
Last updated: Apr 12 2022 at 19:14 UTC
