Clinical Data Management and Standards
What the exam tests
Clinical data management and standards tests CRF design, data capture, coding, controlled terminology, CDISC standards, query handling, source verification, and database lock. You must know the difference between SDTM (submission-oriented tabulations) and ADaM (analysis-ready datasets), and when MedDRA or WHO Drug dictionaries apply.
Core rules
- ALCOA+ principles: attributable, legible, contemporaneous, original, accurate, complete, consistent, enduring, available.
- Edit checks should be pre-specified, tested, and documented; queries must be traceable to resolution.
- SDTM organizes collected data into domains such as DM, AE, LB, and VS; ADaM supports analysis with clear derivations.
- Use controlled terminology; do not invent codes when a standard term exists.
- Database lock follows resolution of critical queries, reconciliation, and audit-trail review.
Worked example
A lab result is received as 'glucose 95 mg/dL' with a local code. The mapper should not place the verbatim text only in SDTM. Map the local code to LBTESTCD 'GLUC' and LBTEST 'Glucose'; store the original result in LBORRES and unit in LBORRESU; populate LBSTRESN only after validated conversion to standard units; retain the original in the source and define the derivation. If the local code is ambiguous, raise a query rather than guessing. In ADaM, create the analysis variable with a documented derivation and provenance.
Traps
- Putting derived analysis variables in SDTM because they look convenient.
- Treating MedDRA and WHO Drug as interchangeable coding dictionaries.
- Locking the database before reconciliation or unresolved critical queries.
- Overwriting source values during cleaning instead of preserving originals.