For rare disease drug development, a single-stage hypothesis testing procedure is often considered for clinical investigation regarding safety and efficacy of a test treatment study. Under the hypothesis testing framework, "p-value less than 0.05 with at least 80% power" is often used as the gold standard in the regulatory review/approval process. This regulatory standard, however, may not be feasible in rare disease drug development due to the small patient population available and the large variability associated with the clinical outcome. To overcome this controversial issue, alternatively, it is suggested that an innovative probability-based confidence interval approach based on precision analysis be considered. The proposed probability-based confidence interval approach can not only stay away from the limitations and concerns regarding the use of p-value but also provide valuable information regarding whether the observed results are reproducible.
Chow et al. (Fri,) studied this question.