A Data-Driven Approach to Risk-Based Source Data Verification.

Area of Science:

• Clinical Trials
• Data Management
• Regulatory Science

Background:

• Source Data Verification (SDV) is crucial for ensuring clinical trial data integrity.
• Traditional 100% SDV is resource-intensive, prompting exploration of reduced, risk-based approaches.
• Previous discussions on reduced SDV have been largely theoretical, lacking empirical data.

Purpose of the Study:

• To investigate the feasibility of using historical data and simulation to evaluate reduced SDV scenarios.
• To assess the risks and benefits, including cost reductions, associated with specific reduced SDV strategies.
• To bridge the gap between theoretical discussions and data-driven methodologies in SDV.

Main Methods:

• Utilized historical data from 30 electronic data capture (EDC) clinical trials completed between 2005 and 2010.
• Employed simulation methodology based on 4 hypothetical risk-based monitoring approaches from a 2010 Drug Information Journal paper.
• Replicated proposed scenarios in simulation algorithms to explore real data application.

Main Results:

• Confirmed that real study data can successfully simulate reduced SDV scenarios.
• Demonstrated the potential for a data-driven approach to determine efficient and effective reduced SDV strategies.
• Provided empirical evidence supporting the transition from theoretical to data-driven SDV methodologies.

Conclusions:

• Historical clinical trial data is a viable resource for simulating and evaluating reduced SDV approaches.
• A data-driven methodology enhances the understanding of risks and benefits in SDV strategy selection.
• This research facilitates more efficient and effective clinical trial oversight through optimized SDV.