Estimating comparative effectiveness using Single-Arm trials: A challenge in the field of agnosticism

Although single-arm trials (SATs) can serve as a pivotal study design for evaluating drug efficacy in specific scenarios, they are accompanied by inherent limitations and challenges. This paper explores the challenges of validity, reliability, statistical power, and the use of external control in SATs from a statistical perspective. We reflect on the rationale for employing SATs to support efficacy assessments, and provide recommendations based on their limitations and potential biases. The objective is to offer guidance for the improved design and implementation of SATs in future research.