Stream: terminology
Topic: local systems
Jay Lyle (Oct 29 2018 at 18:50):
VA needs an allergy substance code associated with the source file it was drawn from (class, ingredient, etc.).
If the source systems are unique to a file (RxNorm values occur only in the ingredient file, NDF in class, etc.) then we can use the code system to infer the file. That's probably not be the case.
Another option: we can include in our bag of codings both a standard (code 83367 for atorvastatin, system = http://www.nlm.nih.gov/research/umls/rxnorm) and a local system value (code 83367 for atorvastatin, system = http://www.va.gov/VAIngredientFile).
Is there anything wrong with that?
Lloyd McKenzie (Oct 29 2018 at 18:53):
Agree that relying on RxNorm only applying in certain cases is risky. A local code is one possibility, but there's still a risk of the local code getting propagated and showing up places it shouldn't. The only "safe" way to track the source file is probably an extension.
Jay Lyle (Oct 29 2018 at 18:56):
Good point. I thought it sounded messy but not dangerous, until we start proliferating mess and cluttering screens.
Last updated: Apr 12 2022 at 19:14 UTC
