A comparison of error detection rates between the reading aloud method and the double data entry method.

Area of Science:

• Clinical data management
• Health informatics
• Medical research methodology

Background:

• Accurate data entry and verification are critical in clinical studies.
• The reading aloud (RA) method is a common data verification technique in Japan, often used alongside or instead of double data entry (DDE).
• The RA method involves one operator reading data aloud while another compares it to the source document.

Purpose of the Study:

• To compare the efficiency of the RA method versus the DDE method for data verification in clinical studies.
• To evaluate error detection rates and time expenditure for both methods.
• To assess the impact of operator consistency (same vs. different) on verification efficiency.

Main Methods:

• A comparative study was conducted using data from the Japanese Registry of Renal Transplantation.
• Five hundred sixty case report forms (CRFs) were randomly assigned for single data entry.
• Two types of DDE and RA methods were performed by operators, with variations in whether the same or different operators performed the verification.
• Error detection rates and time expended were recorded and analyzed.

Main Results:

• The overall per-field error rate was 0.34%.
• Double data entry (DDE) demonstrated significantly higher error detection rates (88.3% for different operators, 69.0% for same operator) compared to the RA method (59.5% for different operators, 39.9% for same operator).
• The RA method required less time (57.9 hours) than the DDE method (74.8 hours).

Conclusions:

• The DDE method is superior to the RA method for detecting single data entry errors in clinical data management.
• While the RA method is more time-efficient, its lower error detection rate makes it less suitable for ensuring high data quality.
• The choice between DDE and RA should consider the trade-off between accuracy and time efficiency based on study requirements.