SCDM Data Reconciliation and Validation 2 — Questions and Answers
Question 1: What is the purpose of a 'soft edit' (or soft stop) in an EDC system?
- To permanently block invalid data from being entered
- To allow data entry to continue after the user acknowledges a potential issue (Correct answer)
- To automatically delete duplicate records
- To encrypt sensitive patient data during transmission
Correct answer: To allow data entry to continue after the user acknowledges a potential issue
Soft edits generate a warning to alert the data entry user of a potential issue but allow them to proceed and save the data after providing a justification or acknowledgment.
Question 2: Patient profile reviews during data validation are primarily conducted to:
- Assign subjects to treatment groups
- Identify data anomalies or patterns at the subject level that edit checks may miss (Correct answer)
- Generate financial invoices for site payments
- Review investigator credentials and training records
Correct answer: Identify data anomalies or patterns at the subject level that edit checks may miss
Patient profile reviews allow data managers to assess data coherence and plausibility at the individual subject level, catching issues not flagged by automated edit checks.
Question 3: Which activity is performed as part of 'external data reconciliation' in a clinical trial?
- Comparing CRF data entered by two different data entry operators
- Matching data in the clinical database against data from third-party vendors such as labs or IRT (Correct answer)
- Reviewing audit trails within the EDC system
- Validating the statistical programming code
Correct answer: Matching data in the clinical database against data from third-party vendors such as labs or IRT
External data reconciliation involves comparing and aligning clinical trial data from the primary EDC/database with data received from external sources such as central labs, IRT systems, or PK labs.
Question 4: A cross-form validation check would flag an error if:
- A numeric value falls outside the normal reference range
- A mandatory field is left blank
- A subject's adverse event start date precedes the date of informed consent (Correct answer)
- A date field is entered in the wrong format
Correct answer: A subject's adverse event start date precedes the date of informed consent
Cross-form validation checks relationships between fields across different CRF forms—in this case, verifying that an adverse event cannot start before the subject consented to the trial.
Question 5: What is 'database lock' in clinical data management?
- Encrypting the database with a password to prevent unauthorized access
- The formal process of restricting further edits to the clinical database after data cleaning is complete (Correct answer)
- A system failure that prevents data entry for a period of time
- Archiving individual subject data files at study end
Correct answer: The formal process of restricting further edits to the clinical database after data cleaning is complete
Database lock is a formal, documented step marking the end of data cleaning where the database is set to read-only, preventing further modifications before statistical analysis.
Question 6: What is the key difference between data validation and data verification in clinical data management?
- Validation ensures data is collected; verification ensures it is encrypted
- Validation checks data accuracy and integrity using rules; verification confirms data was transcribed correctly from source documents (Correct answer)
- Validation is performed by biostatisticians; verification is performed by monitors
- Validation occurs before enrollment; verification occurs after database lock
Correct answer: Validation checks data accuracy and integrity using rules; verification confirms data was transcribed correctly from source documents
Validation applies programmatic rules and logic checks to ensure data quality and integrity, while verification (SDV) confirms that data entered in the CRF matches the original source documents.
Question 7: In an audit trail review during data reconciliation, which of the following would be considered a red flag?
- A field corrected by the site after receiving a data query
- A value changed multiple times without associated queries or explanations (Correct answer)
- A date format standardized by the system automatically
- A missing field that was completed after site monitoring
Correct answer: A value changed multiple times without associated queries or explanations
Multiple unexplained changes to a data field without associated queries or justification may indicate data integrity issues or improper data manipulation.
SCDM Data Reconciliation and Validation 2
What is the purpose of a 'soft edit' (or soft stop) in an EDC system?