Probability of Inconclusiveness

For review and approval of new drug products, substantial evidences regarding safety and effectiveness of the drug products under investigation are necessarily provided. A traditional approach is to test a null hypothesis of ineffectiveness against an alternative hypothesis of effectiveness at the 5% level of significance. The rejection of the null hypothesis of ineffectiveness is in favor of the alternative hypothesis of effectiveness. This approach, however, is somewhat misleading because the rejection of the null hypothesis of ineffectiveness leads to the conclusion of not-ineffectiveness, which consists of the proportion of inconclusiveness and the proportion of effectiveness. In this chapter, we explore the potential use of the concept of demonstrating not-ineffectiveness and then effectiveness for regulatory approval of new drug products, especially for rare disease drug products. For rare disease drug product development, one of the major obstacles and challenges is how to use small patient population available for achieving the same standards for regulatory approval. To address this problem, a two-stage approach by first demonstrating not-ineffectiveness and then ruling out (or controlling) the probability of inconclusiveness for demonstrating effectiveness is proposed. The proposed two-stage approach is useful with small patient population available for achieving the same standards for regulatory approval of rare disease drug products.