<![CDATA[This study develops a Bayesian extension of the Skip-Lot Sampling Plan V (SkSP V) and demonstrates its utility in adaptive clinical trial monitoring. The proposed approach incorporates prior information, probabilistic decision rules, and a skipping mechanism to minimize patient exposure while preserving statistical efficiency. Methodologically, the plan defines operating characteristic (OC) curves, average outgoing quality (AOQ), average outgoing quality limit (AOQL), and expected sample number (EN) under a Bayesian framework. A case study in a Phase II clinical trial context illustrates the adaptability of SkSP V to real-world monitoring, where patient cohorts are treated as lots and treatment failures as defectives. Simulation results show that Bayesian SkSP V achieves higher acceptance probability at acceptable efficacy levels while maintaining strong protection against poor treatments at limiting efficacy levels. Compared with alternative skip-lot designs such as SkSP III, the Bayesian SkSP V achieves a lower AOQL and reduced expected sample number, thereby offering both ethical and statistical advantages. These findings suggest that Bayesian SkSP V provides a robust, efficient, and ethically favorable framework for modern clinical research, bridging the gap between industrial quality control methods and biomedical trial monitoring.]]>
Copyrights © 2026
