Testing for efficacy with multiple endpoints is an important statistical problem. In accordance with the Prescription Drug User Fee Act of the U.S. Congress (PDUFA IV), the Food and Drug Administration (FDA) will issue a guidance on Multiple Endpoints in 2009. Such statistical problems have defined paths to decision-making. With primary and secondary endpoints, efficacy in a secondary endpoint is only relevant if efficacy in the primary endpoint has been shown. A current approach is based on closed testing, testing all possible intersection hypotheses, and collating the results. For decision-making to follow pre-defined paths, strategic choices of test statistics and critical values must be made. As the number of doses and endpoints increase, such strategic choices become increasingly difficult.

Partition testing is a fundamental multiple comparison principle on which step-down, step-up, and fall-back FWER and gFWER controlling methods are based. Partition testing can be used to implement the Decision Path Principle that we propose. This principle states null hypotheses should be formulated so that decision-making automatically respects pre-defined paths. With primary and secondary endpoints, the number of path-respecting partition hypotheses is far fewer than the number of all intersection hypotheses. Less is More with partition testing, in the sense that testing fewer hypotheses is less mistake-prone, while the method can be more powerful than certain gate-keeping methods.