臨床試験における主要評価項目を評価するための図表示の提案　−対立仮説、ベイズファクター、信頼区間に基づく−

抄録

An alternative hypothesis, defined as the assumed clinically meaningful treatment effect at the primary endpoint, is determined when a randomized controlled trial (RCT) is designed. Nevertheless, most researchers conducting RCTs for superiority very often disregard evaluating the alternative hypothesis. Since most clinical trials are designed to demonstrate the statistical significance of the primary endpoint, in trials that fail to show this, aspects of the results that differ from what researchers initially intended are often inappropriately emphasized. The evaluation of treatment effects from clinical trials should be based on the primary endpoint. Therefore, we propose a quantitative statistical graph, referred to as the “ABC plot,” representing the alternative hypothesis, the Bayes factor, and the confidence interval, which enhances the visual evaluation of the treatment effect based on the result of the primary endpoint. The ABC plot was applied to three actual RCTs intended to show superiority. A discussion based only on statistical significance is insufficient for interpreting the primary endpoint. The ABC plot enables researchers to evaluate the alternative hypothesis on the estimation of the treatment effect.