ABSTRACT
I. INTRODUCTION Before administration of the first dose of a new compound to human subjects, a toxicological safety assessment has to be performed. Statisti cal analysis plays a fundamental part in the interpretation of the corre sponding experiments. Usually, the traditional null hypothesis of no difference in the effect between the treatment and a vehicle or negative control group is tested. Failure to reject the null hypothesis (e.g., if the corresponding /7-value is greater than 0.05), often leads to the conclusion of evidence in favor of safety. The major drawback of this indirect procedure is what is controlled by the prespecified level is the probability of erroneously concluding hazard (producer risk). However, the primary concern of safety assessment is the control of the consumer risk, that is limiting the probability of erroneously concluding safety. Thus, the adequate test problem should be formulated by reversing the null hy pothesis and the alternative and incorporating an a priori-or a posteri ori-defined threshold. This direct approach will be demonstrated for the
two-sample and ^-sample many-to-one problem relative to nonmono tonic and monotonic response relations.
