Remapping of codes (and decodes) in analysis datasets for electronic submissions

AbstractFor electronic submissions, the U.S. Food and Drug Administration requests, according to the Study Data Specifications (version 1.6, June 2011), that variable names and codes in analysis datasets should be consistent across studies and, where feasible, the NCI CDISC Vocabulary should be used. For integrated analysis, it is necessary to harmonize codes across all studies. Unfortunately, during the life cycle of a product, codes may change either due to project needs or due to external circumstances, e.g. CDISC introduces new controlled terminology. For analysis datasets following the Analysis Data Model, there are often pairs of corresponding variables containing a code and a decode which both have to be updated. This paper describes a workflow to remap codes and also decodes for a study or an integrated database with a macro using metadata and standard SAS format datasets with a few additional variables.