search

OHDSI / Vocabulary-v5.0

Build process for the OHDSI Standardized Vocabularies. Currently not available as independent release.
The Unlicense
214 stars 75 forks source link

use of Unit versus International Unit #906

Open tiozab opened 10 months ago

tiozab commented 10 months ago

Describe the problem in content or desired feature For me as a pharmacist, "unit" is not a clinically relevant unit, but "international unit" is because it has a precise definition. However, it seems that "unit" is used in drug concepts. For example the RxNorm Extension of the drug enoxaparin has the unit "UNT".

How to find it (https://athena.ohdsi.org/search-terms/terms?conceptClass=Branded+Drug&conceptClass=Branded+Drug+Box&conceptClass=Branded+Drug+Comp&conceptClass=Branded+Drug+Form&page=1&pageSize=15&query=enoxaparin+UNT&boosts)

Expected adjustments I think in RxNorm Extension, "international unit" (concept id: 8718) instead of "unit" (concept id: 8510) should be used for drugs. Because for enoxaparin for example: 100 IU (international units) translate to 1 mg of enoxaparin,

tiozab commented 10 months ago

@cgreich I add you here, this will come to your table anyway when you will read through the dosing manuscript that we will share soon with co-authors ;-)

cgreich commented 10 months ago

.. and I thought I could stay under the radar. But no!!! :)

I think you are right. But that would be a big surgery. Why don't we treat them both as synonymous for the time being?

tiozab commented 10 months ago

@Tsemharb @ajoedicke do you agree to include "unit" as a clinically relevant unit for our patterns? I think it would make datapartners happy!! It would add quite some patterns though (14 additional patterns as unit would qualify in "amount" (with numeric amount value) and "numerator", and "denominator"))

ajoedicke commented 10 months ago

Happy with that approach for now

tiozab commented 10 months ago

Thank you @ajoedicke ! For completeness I post here @Tsemharb's reply as well: "I totally agree. One follow-up thought though. I think you are proposing switching from Units to International Units at some point. I think it will work for the vast majority of the drugs. But I remember a few cases while building vocabularies when that pattern can break. Some of the source drugs have pretty weird units. Here is an example. Source concept unit is SQ-T which is mapped to "Unit" in RxExtension. I don't think that treating that one as an international unit will be a valid approach. So in some cases "Unit" is just used as an umbrella term for a lack of a better standard unit, contrary to clinical practice where IU and U are interchangeable. But on the other hand such drugs are the minority and won't affect the results to a substantial degree."