This section describes how the SDTM will be implemented to represent relationships between records within a domain, between records in different domains, and between datasets.
When applicable, variables IDVAR and IDVARVAL will represent variable and value pairs used as record-level keys to merge/join records between datasets. In such cases, variables represented in IDVAR may have values that are unique for:
- Each observation within a domain.
- Groups of observations within a domain
The uniqueness of values by observation or groups of observations can be used to represent relationships efficiently. In cases where values are unque to observations
When possible the variable represented in IDVAR will have values from collected data (e.g., SPID, RECID), but variables with derived values may also be used (e.g., --SEQ).
- When relationship datasets (e.g., SUPPxx, FAxx, CO, RELREC) relate back to split parent domains, the value of IDVAR will be from a variable with unique values for each observation.
Overview
Content Tools