The same discipline, sustained through conduct
This is Double Programming / Independent QC applied continuously as conduct data accrues, not a one-time check.
What's included
What goes into ongoing dataset QC
- Independent re-verification of SDTM and ADaM datasets against the locked mapping and derivation specs on each conduct data cut
- Discrepancies between production and QC outputs logged, investigated, and resolved before the cut is signed off
- Edit checks and derivation logic re-tested whenever new data patterns surface issues not seen in earlier cuts
- QC scope adjusted in line with the study's risk-based QC plan as enrollment and visit volume grow
- Findings communicated back to data management when QC surfaces upstream data quality issues
- QC evidence retained continuously, supporting inspection readiness without a scramble at database lock
Sets up a clean close out
Datasets QC'd throughout conduct carry directly into Final SDTM/ADaM QC and Validation once the database locks.