Data files (available at data.world)
- Format 1:
drugnames_withclasses.csv
- Format 2:
drugnames_withclasses.feather
- string: a sequence of characters
- integer: whole numbers
Name | Type | Description |
---|---|---|
drugname_brand | string | Brand or trade name of the drug |
drugname_generic | string | Generic or nonproprietary name of the drug (seems to contain specific formulation information such as type and amount of active ingredient) |
rxnorm_rxcui | integer | Numeric identifiers including missing/unknown. Each represents a unique drug name which is separately provided |
drug_major_class | string | This is a set of 30 possible alpha-numeric codes including missing/unknown. It indicates the major class of the drug, such as cardiovascular medications and central nervous system medications |
dmc_name | string | String description of drug_major_class (seems mainly unknown/missing or NA) |
drug_class | string | This is a set of 263 possible alpha-numeric (two-letter plus three-digit) codes. It indicates the class of the drug, such as antidepressants and analgesics |
dc_name | string | String description of drug_class |
Data provenance Drug information comes from Part D Prescription Drug Event (PDE) data and is available for a subset (~70%) of Medicare beneficiaries. Unique drug identifiers and classes were from CMS 2010 Medicare Prescription Drug Profiles Public Use Files which represents 100% of prescription drug claims.
Data calculation methods
TODO:
- Add information on how drugnames were mapped to RXCUI
- Add how major class and class were matched to their description
This is an older dataset / dictionary that needs to be verified