Meta-analysis combines the results of multiple studies to obtain an overall answer. Recent technical advances in meta-analysis enable comparisons between multiple treatments even in the absence of direct comparisons within individual studies, and provide ways to evaluate whether conducting randomised controlled trials (RCTs) and meta-analyses differently could reduce the risk of bias. The current project will exploit these new techniques in two ways. First, we will apply advanced meta-analytical methods to a series of specific clinical problems in disciplines including psychiatry, cardiology and surgery, to help decide the best ways of treating patients. Second, we will examine a number of different methodological approaches to the conduct, analysis and interpretation of meta-analyses and RCTs. The project has potential to provide better information about the risks and benefits of specific treatments, and to improve the validity and usefulness of meta-analysis as a tool for informing clinical decision-making.