Abstract: RxNorm contains concepts which represent the drugs currently marketed in the U.S. With each monthly RxNorm release, new drug products are added and obsolete drug products are removed. While this curation supports e-prescribing and drug information exchange use cases, it is detrimental to analytics use cases. The issue in this case is that some of the RxNorm drug identifiers (RxCUIs) stored in clinical data warehouses have become obsolete and can no longer be interpreted in reference to the current RxNorm dataset. To address this issue, we started developing an RxNorm concept history service as part of the NLM drug APIs (https://rxnav.nlm.nih.gov/. This service is similar to what we developed for managing obsolete identifiers from the National Drug Code (NDC). For each obsolete RxCUI, the service returns a canonical representation of the drug concept, making it possible to relate an obsolete drug product to a class through its ingredient or to find a similar active drug product based on ingredient, strength and dose form information.

Learning Objective 1: Describe the main use cases for RxNorm

Learning Objective 2: Describe issues with the use of RxNorm for analytics purposes

Learning Objective 3: List the main properties in the canonical representation of historical RxNorm concepts


Olivier Bodenreider (Presenter)

Lee Peters, NLM

Presentation Materials: