Making best use of existing evidence when planning trials

Objectives Although meta-analyses are typically viewed as retrospective activities, they are increasingly being applied prospectively to justify planned research by providing up-to-date evidence on specific research questions. Since healthcare policy makers often base decisions on systematic reviews of reliable evidence rather than on single clinical trials, trialists should consider planning additional research with regard to the meta-analytic result. Since adding a new study to an existing meta-analysis creates a multiple testing scenario, which standard methods do not well address, nominal significance levels need to be adjusted to preserve the overall type-I error rate. Sequential approaches to meta-analysis have been proposed, although these do not lead directly to recommendations on planning new studies.