OBJECTIVES: To develop normalization methods for managing the variation in clinical drug names. METHODS: Manual examination of drug names from RxNorm and local variants collected from formularies led to the identification of three types of drug-specific normalization rules: expansion of abbreviations (e.g., tab to tablet);reformatting of specific elements (e.g., space between number and unit); and removal of salt variants (e.g., succinate from metoprolol succinate). RESULTS: After drug-specific normalization, recall of 3397 previously non-matching names from formularies reaches 45% overall (70% of some subsets), compared to 10-20% after generic normalization. Ambiguity has not increased significantly in the RxNorm dataset. CONCLUSIONS: A limited number of drug-specific normalization operations provide significant improvement over general language normalization.
OBJECTIVES: To develop normalization methods for managing the variation in clinical drug names. METHODS: Manual examination of drug names from RxNorm and local variants collected from formularies led to the identification of three types of drug-specific normalization rules: expansion of abbreviations (e.g., tab to tablet);reformatting of specific elements (e.g., space between number and unit); and removal of salt variants (e.g., succinate from metoprolol succinate). RESULTS: After drug-specific normalization, recall of 3397 previously non-matching names from formularies reaches 45% overall (70% of some subsets), compared to 10-20% after generic normalization. Ambiguity has not increased significantly in the RxNorm dataset. CONCLUSIONS: A limited number of drug-specific normalization operations provide significant improvement over general language normalization.
Authors: Sunghwan Sohn; Cheryl Clark; Scott R Halgrim; Sean P Murphy; Christopher G Chute; Hongfang Liu Journal: J Am Med Inform Assoc Date: 2014-03-17 Impact factor: 4.497