RXNORM Mapping for SNOMED
SNOMED and RXNORM are very different in how they present drug information. RXNORM is much more advanced and codes strength, route of administration, packaging information, and product names. SNOMED has mainly main active ingredients and some product names. SNOMED's coverage for durability and management paths is small.
For example SNOMED has no identifiers for Prozac
, but has for Fluoxetine
, which is the main ingredientProzac
In RXNORM, you can find the following connections:
[PROzac Weekly 90 MG Delayed Release Oral Capsule]
-- ingredient --> [Prozac]
-- precise ingredient --> [Fluoxetine]
And the ingredient concept takes you to SNOMED:
[Fluoxetine]
-->
SNOMEDCT_US:53640004
SNOMEDCT_US:372767007
Keep in mind that most of the ingredients in SNOMED are duplicated as product and substance, and each of these types gives a different relationship.
Three options for implementing this:
- Use RXNORM / NIH NLM data files https://www.nlm.nih.gov/research/umls/rxnorm/docs/rxnormfiles.html
- BioPortal API (free) http://bioportal.bioontology.org/
- Lexigram API (commercial) http://www.lexigram.io/
The important aspect here is that if you replace the RXNORM IDs with SNOMED IDs, you will lose information, but of course depending on your use case, which can be very good. I would consider using NDC .
Disclaimer: I was an engineer at BioPortal and currently the CTO at Lexigram.
source to share