Meta-analysis: the fashion of summing-up evidence. Part I. Rationale and conduct.

Meta-analysis is the process of formally combining the quantitative results of separate studies in order to increase the statistical precession of estimated effects. It has become fashionable when individual studies fail to yield conclusive results. A recent update of a meta-analysis (overview) for treatments of patients with early breast cancer motivated this commentary to discuss the rationale, principles of conduct, statistical methods, and graphical display for summing up results. In part II we will approach the issue of interpretation and use of overview data.