Doing More Good Than Harm

A revolution in the quality and efficiency of health care is in the making. The rational design of diagnostics, drugs, and vaccines made possible by molecular biomedical science, in conjunction with statistical methods designed to eliminate bias in the evaluation of these interventions, will dramatically change health care delivery by the turn of the century and beyond. The purpose of this conference is to consider the strengths and weaknesses of the principal means of evaluating interventions: these range from consensus among "distinguished physicians" to rigorously organized trials and overviews of related trials. Sections: Types of evidence; Observational methods: Major achievment and future potential; Randomized trials: Major achievements and future potential; Combining evidence: Reviews, overviews, and meta-analyses; Disseminating evidence; using evidence.