Stream: implementers
Topic: Allergy - Code - Translations?
Eric Whitley (May 05 2020 at 17:05):
Hello -
I'm sorry if this is a naive question. Am I correct in reading that AllergyIntollerance is intended to support one and only one substance/product through code and does not support multiple codes? EX: I can have a single RxNorm term, but not a mapped NDC in addition. If I wanted to do that I'd instead create a second AllergyIntolerance record or express more granular (or even potentially translations) through the reaction entries through reaction.substance?
Mainly asking so I can ensure I'm understanding this correctly.
Per https://www.hl7.org/fhir/allergyintolerance-definitions.html
AllergyIntolerance.reaction.substance
Note: the substance for a specific reaction may be different from the substance identified as the cause of the risk, but it must be consistent with it. For instance, it may be a more specific substance (e.g. a brand medication) or a composite product that includes the identified substance. It must be clinically safe to only process the 'code' and ignore the 'reaction.substance'.
...
The 'code' element allows for the use of a specific substance or pharmaceutical product, or a group or class of substances. In the case of an allergy or intolerance to a class of substances, (for example, "penicillins"), the 'reaction.substance' element could be used to code the specific substance that was identified as having caused the reaction (for example, "amoxycillin"). Duplication of the value in the 'code' and 'reaction.substance' elements is acceptable when a specific substance has been recorded in 'code'.
Am I understanding that correctly?
Lloyd McKenzie (May 05 2020 at 17:12):
AllergyIntolerance.code allows multiple codings. So you can express a single allergy in as many different code systems as you like
Eric Whitley (May 05 2020 at 17:15):
Am I misreading the FHIR spec? code is listed as 0..1. Is that not where you'd record the related concept identifier?
Jean Duteau (May 05 2020 at 17:16):
one code has multiple Codings. So it can contain multiple "translations" or "encodings" of the allergy, i.e. use SNOMED, use vendor-specific code, etc.
<code>
<coding 1 with SNOMED/>
<coding 2 with vendor-specific code/>
<etc./>
</code>
Eric Whitley (May 05 2020 at 17:22):
OH!. Wow. OK, I was really off there. Thank you, @Lloyd McKenzie and @Jean Duteau . You really helped me correct a much larger understanding of CodeableConcept. VERY much appreciated!
Jean Duteau (May 05 2020 at 17:24):
No problem! One thing I should have added is that the different Codings are intended to be different representations of the same concept, albeit at potentially different granularities. But it is not kosher to use the different Codings to send two different concepts, i.e. <code><coding for allergy to allopurinol> <coding for allergy to cat dander></code>
Eric Whitley (May 05 2020 at 17:49):
Jean Duteau said:
No problem! One thing I should have added is that the different Codings are intended to be different representations of the same concept, albeit at potentially different granularities. But it is not kosher to use the different Codings to send two different concepts, i.e. <code><coding for allergy to allopurinol> <coding for allergy to cat dander></code>
Perfect. My intention is not to represent two totally different concepts - more translations of the same concept when we receive multiple terminologies.
Thanks again for cluing me in on CodeableConcept. I was just looking at the ordinality reference and didn't think past 0..1 to notice CodeableConcept and recognize it included multiple codes. /facepalm
Last updated: Apr 12 2022 at 19:14 UTC
