Variables to support datapoint traceability should be included whenever practical and feasible. Primary candidates for datapoint traceability, when used in conjunction with USUBJID, include the dataset or domain name, the name of the source variable, and the relevant sequence number (SDTM domain -‑SEQ value or the ADaM ASEQ value). The ADaM ASEQ variable (see Section 2.9.6.1 ,Identifier Variables for BDS Datasets) facilitates datapoint traceability by providing sequence numbers that are unique within a subject within an ADaM dataset, ensuring uniqueness of a record when used in combination with USUBJID.

The following table defines additional variables useful in certain situations to facilitate datapoint traceability. These variables are useful in situations where a single ADaM dataset or multiple SDTM datasets and/or ADaM datasets were used to create an ADaM dataset. Section 2.9.9.3, Inclusion of Input Data that Are Not Analyzed but that Support a Derivation in the ADaM Dataset, contains an example of how to use these variables.

Variables used for datapoint traceability may also include any other variables that facilitate transparency and clarity of derivations and analysis.

Variable Name

Variable Label

Type

Codelist/ Controlled Terms

Core

CDISC Notes

SRCDOM

Source Data

Char


Perm

The SDTM domain name or ADaM dataset name that relates to the analysis value (i.e., AVAL or AVALC in a BDS dataset). If the source data is a supplemental qualifier in SDTM, this variable will contain the value of RDOMAIN in SUPP-- or SUPPQUAL.

SRCVAR

Source Variable

Char


Perm

The name of the column (in the domain or dataset identified by SRCDOM) that relates to the analysis value (i.e., AVAL or AVALC in a BDS dataset). In the event that SRCDOM is a SUPPQUAL, then SRCVAR will be populated with the value of the related QNAM.

SRCSEQ

Source Sequence Number

Num


Perm

The sequence number --SEQ or ASEQ of the row (in the domain or dataset identified by SRCDOM) that relates to the analysis value (i.e., AVAL or AVALC in a BDS dataset). In the event that SRCDOM is a SUPPQUAL, then this variable will contain the sequence number of the relevant related domain record.

If the value of AVAL or AVALC in the ADaM dataset is taken from another ADaM dataset, SRCDOM, SRCVAR, and SRCSEQ will contain the name of the source ADaM dataset, the variable name, and the ASEQ value of the row where the source datapoint is located, respectively. If the source SDTM dataset is a supplemental qualifier dataset, the 2-letter domain prefix of --SEQ in the ADaM dataset will be the related domain abbreviation (the value of RDOMAIN in SUPP-- or SUPPQUAL), and SRCVAR will contain the value of QNAM.

SRCDOM, SRCVAR, and SRCSEQ are permissible and, if included in a dataset, do not all need to be populated on every row. For example, if the value of AVAL or AVALC is a summary of multiple rows from a single ADaM or SDTM dataset, then SRCDOM will be populated, SRCVAR will be populated if the same variable is summarized across all of the rows, and SRCSEQ will be left blank. In another example, if multiple SDTM and/or ADaM datasets are used to create AVAL or AVALC, then SRCDOM, SRCVAR, and SRCSEQ will all be left blank.

If all values of AVAL or AVALC in the ADaM dataset are taken from a single SDTM domain, all records in the ADaM dataset would have the same value for SRCDOM and SRCVAR. Instead of using SRCDOM, SRCVAR, and SRCSEQ in that case, the producer may elect to simply include the --SEQ variable from the source SDTM domain, which is sufficient to provide the needed traceability within a subject.

  • No labels