Interim monitoring of sequential multiple assignment randomized trials using partial information.

Area of Science:

• Biostatistics
• Clinical Trials Methodology
• Health Services Research

Background:

• Sequential Multiple Assignment Randomized Trials (SMARTs) are crucial for evaluating multistage treatment regimes.
• Interim monitoring in SMARTs is challenging due to participants being at different treatment stages.
• Existing methods for interim analysis in SMARTs may not fully utilize available data.

Purpose of the Study:

• To develop a more efficient estimator for mean outcomes in SMARTs during interim analyses.
• To derive statistical procedures for early stopping based on the proposed estimator.
• To compare the performance of the new estimator against existing methods.

Main Methods:

• Proposed an estimator for mean outcomes that incorporates partial data from all participants.
• Derived Pocock and O'Brien-Fleming testing procedures for early stopping.
• Conducted simulation experiments to evaluate the estimator's performance.

Main Results:

• The proposed estimator effectively controls type I error and achieves nominal power.
• The new method demonstrated a reduction in expected sample size compared to previous approaches.
• Simulations confirmed the efficiency gains from utilizing partial participant data.

Conclusions:

• The developed estimator offers a more efficient approach to interim analysis in SMARTs.
• This method allows for principled early stopping while maximizing data utilization.
• The findings have implications for designing and conducting adaptive clinical trials for complex treatment strategies.