Probability and Distribution of the Number Needed to Treat / 医薬品情報学

ABSTRACT

<b>Objective: </b>The number needed to treat (NNT) is an index for determining the number of patients who need to be treated in order to prevent the occurrence of an adverse event when a new treatment instead of the standard one is used.<br><b>Methods: </b>The properties of NNT were examined by using the geometric distribution formula as the probability for preventing the occurrence of an adverse event in the NNT.<br><b>Results: </b>When the NNT was enlarged infinitely, the probability for the prevention of an adverse event by NNT was found to be 63.2% (=1−e⁻¹), and the number of patients who needed to show adverse event prevention at the probability of 95% was about 3 times as that of the NNT (Rule of Three).<br><b>Conclusion: </b>When the effect of new treatment needs to be evaluated based on NNT, one should take these properties of NNT into consideration.