Stream: implementers
Topic: Allergies with multiple reactions
Cooper Thompson (Dec 18 2018 at 19:00):
AllergyIntolerance has two different ways to represent multiple reactions to a substance. You can either have multiple AllergyIntolerance.reactions, each with one manifestation, or one reaction with multiple manifestations. I think I understand the difference between the two in the spec. The reactions are the reaction events, so if you have two reactions a year apart, and for some reason you had different symptoms during each reaction event, then you'd return multiple AllergyIntolerance.reactions. However if a system doesn't really track allergy events on the allergy record itself, then it is likely more appropriate to return a single AllergyIntolerance.reaction with the "reactions" listed as manifestation.
Epic's implementation of AllergyIntolerance isn't currently event based, so we will return a single AllergyIntolerance.reaction, where multiple symptoms (swelling, rash, etc.) are returned as multiple manifestations on the reaction.
1) Is my understanding as described above correct?
2) I'm curious how folks have actually implemented it.
We are working on a client to pull Argonaut-compliant (thus DSTU2) allergies, and are wondering how many implementers will actually return multiple reactions, rather than a single reaction with multiple manifestations. Since we need to map into a non-event based allergy record, we're trying to understand how flexible we really need to be.
Lloyd McKenzie (Dec 18 2018 at 19:27):
@Michelle (Moseman) Miller ?
Grahame Grieve (Dec 18 2018 at 20:40):
I think your understanding is correct
Rob Hausam (Jan 02 2019 at 16:32):
I've been pretty actively involved in the PCWG work and discussions on AllergyIntolerance andI agree with your understanding. I think your suggested approach for the "non event based" situation seems reasonable - but I'm not sure about the state of current and planned implementations in this regard, and, as you suggest, others could do it differently.
Lloyd McKenzie (Jan 02 2019 at 17:53):
If reaction is going to mean different things from different systems, it would be useful if we had a documented convention for differentiating them. @Cooper Thompson can you submit a change request?
Last updated: Apr 12 2022 at 19:14 UTC
