A Hybrid Pipeline for Mapping French UCD Drug Codes to RxNorm with Dosage Preservation

Area of Science:

• Pharmacovigilance and Drug Informatics
• Health Data Standardization
• Computational Pharmaceutics

Background:

• The Observational Medical Outcomes Partnership Common Data Model (OMOP CDM) necessitates RxNorm drug encoding for interoperable and dose-aware analyses.
• French medication data uses UCD and CIP codes, lacking direct mapping to RxNorm, hindering international data harmonization.
• Accurate drug coding is essential for reliable real-world evidence generation and comparative effectiveness research.

Purpose of the Study:

• To develop and validate an automated pipeline for mapping French UCD codes to RxNorm Clinical Drugs.
• To ensure dosage information is preserved during the UCD to RxNorm mapping process.
• To provide a scalable and reproducible framework for drug terminology harmonization within OMOP ETL processes.

Main Methods:

• A three-stage automated pipeline combining reference-based mapping (RUIM, OHDSI Standardized Vocabularies), rule-based similarity matching, and AI reasoning.
• Expert validation was integrated at each stage of the mapping pipeline.
• The pipeline was applied to 38,345 French UCD codes, with evaluation on the 500 most frequently used codes.

Main Results:

• The automated pipeline achieved a 97.4% mapping rate for French UCD codes to RxNorm.
• A precision of 86.0% was obtained when evaluated on the 500 most frequently used codes.
• The developed method successfully preserved crucial dosage information during the mapping.

Conclusions:

• The automated three-stage pipeline offers an effective solution for harmonizing French drug terminology to RxNorm.
• This approach facilitates interoperable and dose-aware analyses within the OMOP CDM for French healthcare data.
• The framework supports reproducible and scalable drug data standardization in observational research.