Assessing data quality and the variability of source data verification auditing methods in clinical research settings

Area of Science:

• Clinical Research
• Data Management
• Quality Assurance

Background:

• Data audits are vital for identifying errors and ensuring data quality in clinical settings.
• Current clinical trial guidelines lack specificity regarding data audit frequency, timing, and nature.
• The absence of a clear data quality definition and error measurement method impacts assessment reliability.

Purpose of the Study:

• To assess the variability of source data verification (SDV) auditing methods used to monitor data quality in clinical research.
• To identify inconsistencies in current data auditing practices within clinical trials.

Main Methods:

• A systematic literature search was conducted across MEDLINE, Scopus, and Science Direct for English-language publications.
• Studies involving clinical trials or research settings that measured or reported data quality using SDV auditing methods were included.
• No date restrictions were applied to the database searches.

Main Results:

• Fifteen publications were included in the review, revealing significant variation in SDV audit methods.
• Audit methods differed based on document complexity, study type, variables, audit proportion (3-100%), and collection frequency (6-24 months).
• Inconsistent methods for error coding, classification, and calculation were observed; transcription errors and personnel inexperience were primary error sources. Repeated SDV audits improved data accuracy by ~40%.

Conclusions:

• A wide range of SDV auditing methods exist, but no single "best practice" for clinical trials could be determined.
• There is a need for published audit methodology articles to develop a standardized SDV auditing method for clinical research data quality monitoring.