Drug Identifier Standardization for FDA-Approved Drug Labeling Using RxNorm

Authors:: Poster Author(s)

Isaac Harris; Stephen Harris; Zhichao Liu, Leihong Wu, Taylor Ingle, Junshuang Yang, Weida Tong, and Hong Fang

Center:: Contributing Office

National Center for Toxicological Research

Abstract

Poster Abstract

Introduction

The FDA regulates and approves all prescription drugs marketed in the US. The drug labeling, which must be available at the time of approval, contains rich drug safety and efficacy information that has a wide variety of uses in the healthcare practice. Due to the inherent complexity of a drug, a drug identifier could be presented in various forms by the FDA in its systems, databases, and publications. This can present challenges when comparing data within the FDA and with outside drug sources due to conceptual inconsistencies. This study investigates the different naming systems of a drug in FDA approved drug labeling and mapping to RxNorm, a normalized drug naming system produced by the National Library of Medicine and widely used in clinical medicine and research.

Method and Results

The most prevalent identifiers used for a drug in the FDA are Application Number, National Drug Code (NDC), Unique Ingredient Identifier, active ingredients (UNII), and Structured Product Labeling (SPL) Set Identifier. The practical description of a drug consists of the active ingredients, route of administration (ROA), Dosage, and Strength/Concentration. RxNorm considers SCDs (Semantic Clinical Drug: active ingredient + strength + dose form) to be a different drug identifier than SBDs (Semantic Branded Drug: Active ingredient + strength + dose form + brand name); and provides approximately 5,830 quantified human prescription (Rx) SCD/generic drugs from a total of 11,876 human Rx drugs (SBD + SCD). We mapped from the FDA’s identifiers of a drug to RxNorm’s quantified SCD drugs to standardize the drug names. The results illustrated that (1) Approximately 45% of all human Rx application numbers (5,473 of 12,137) are associated with more than one RxNorm SCD/generic drug; (2) Approximately 25% of all human Rx SPL set ids (11,045 of 43,445) are associated with more than one RxNorm SCD/generic drug; (3) Approximately one percent of two-part sections NDC codes (436 of 68,816) are associated with more than one RxNorm SCD/generic drug. Thus, application numbers and Set IDs are not precise enough to refer to a drug product or a drug concept, and NDCs are far too specific to be associated with the concept of a drug. However, NDCs are perfect for identifying a singular drug product. We have mapped different identifiers of a drug from the FDA labeling (e.g. UNII, Application Number, NDC, SPL Set ID) to an external identifier, RxNorm’s SCDs, in a graphical representation that elucidates the relationship among the different levels of drug identifiers. These relationships will be available for the benefit of regulatory, clinical, and research applications.

Conclusion

Standardization of the drug identifier for FDA-approved drug labeling using RxNorm will unify drug names and bolster interpretations between the regulatory interpretation of drugs and the clinical application of those drugs. This will allow pharmacists, clinicians, healthcare professionals, and researchers to efficiently utilize the rich resources (e.g., drug Indications and Usage, Drug Interactions, Warnings and Precautions) included in the FDA labeling. In addition, it will allow products to be grouped under a unified drug identifier. Ultimately, the mapping results between FDA drug labeling and RxNorm will allow easier identification and searching of drug information under the standard names, codes, vocabularies, and drug product representations to advance drug related clinical decisions, research, and ultimately, drug safety for public health.