Incorporating data received after a sequential trial has stopped into the final analysis: implementation and...

Area of Science:

• Clinical Trial Design
• Biostatistics
• Medical Research Methodology

Background:

• Sequential clinical trials may continue data accrual beyond the predetermined stopping criterion, a phenomenon termed 'overrunning'.
• Overrunning commonly occurs when patient outcomes require an extended observation period for measurement.

Purpose of the Study:

• To compare two distinct statistical methods designed to accommodate overrunning in sequential clinical trials.
• To assess the performance of these methods in preserving the intended type I error rate.
• To evaluate the impact of incorporating overrunning data on the statistical power of the trial.

Main Methods:

• The study employed simulation studies to rigorously evaluate the proposed overrunning adjustment procedures.
• Performance was assessed based on the ability to maintain the specified type I error rate.
• The effect on statistical power when including overrunning data was also analyzed.

Main Results:

• Simulation results indicate that the evaluated methods differ in their effectiveness at controlling the type I error rate.
• The incorporation of overrunning data can influence the statistical power of sequential trials, with variations observed between the two procedures.

Conclusions:

• The choice of method for handling overrunning data in sequential trials is critical for maintaining statistical integrity.
• Further investigation into these methods is warranted to optimize clinical trial design and data analysis, ensuring reliable results.