sdtm-map

Describe the target domain (such as DM or AE) and your raw data variables, then use the generated template built on assign_no_ct, assign_ct, and derive_seq calls. The script reads raw XPT data with haven, maps variables in a pipeline, and exports the result with xportr_write.

assign_no_ct copies raw values directly without transformation, used for dates, numbers, and free text. assign_ct maps values through a CDISC controlled terminology codelist, used for standardized variables like SEX, RACE, or AESEV.

Use tidyr pivot_longer to reshape measurement columns like SYSBP and DIABP into VSTESTCD and VSSTRESN pairs, then add VSTEST labels, VSSTRESU units, and derive VSSEQ per subject and test code.

This happens when the piped target data and raw_dat share column names, causing sdtm_join to rename duplicates so the raw variable cannot be found. Start the pipeline with a skeleton containing only oak tracking columns, and reference raw variables from the outer raw data object.

Yes, the pharmaversesdtm package ships built-in example XPT files such as dm.xpt and ae.xpt that load via system.file. This lets you run the mapping templates end to end before substituting your actual raw data paths.

Check structure with glimpse and summary, confirm required variables like STUDYID, DOMAIN, and USUBJID exist, and inspect missing values with colSums. For formal compliance, run the XPT files through Pinnacle 21 Community.