A personal critique: evidence-based medicine, methodology, and ethics of randomised clinical trials

Evidence-based medicine may influence our approach to clinical trials. When preparing a systematic review, the quality of individual trials is of far greater importance than their individual results. Unbiased randomisation, a ttention to the treatment protocol and to the rules of good clinical practi ce and honest evaluation of experience are essential; less important is the power of an individual trial and the statistically significant difference between the treatment arms. The recruitment period should be short, followe d by timely publication of a report. Since systematic reviews and meta-anal yses include and quote all available information, clinical researchers and editors should be less influenced by publication bias. These changes in met hodology open clinical trials to new innovative ideas difficult to test in large multi-institutional trials, rend clinical investigators less dependen t on commercial sponsors and might bring more patients into clinical resear ch. (C) 2001 Elsevier Science Ireland Ltd. All rights reserved.