[Evidence-based medicine. 3. Systematic reviews: a tool for clinical practice, permanent education and health policy decisions. Italian Group on Evidence-Based Medicine-GIMBE].

The reviews of research, summarizing a great amount of studies in a manageable format, are invaluable tools for physicians, inundated with enormous amount of biomedical information. However, narrative reviews are often misleading because, mixing together opinions of authors and results of research, the relation between clinical recommendation and evidence is partial and based on a biased citation of primary studies. In contrast to narrative reviews, the systematic reviews assemble, critically appraise, and synthesize the results of primary studies addressing a specific topic. Additionally their authors use strategies for minimizing bias and random error. The science of systematic reviews is now supported by the Cochrane Collaboration, an international network established for "preparing, maintaining and disseminating systematic reviews of the effects of health care". The authors provide tools for searching, critically appraising and using in practice the systematic reviews, which use can help physicians to improve the transfer of research in clinical practice, a task obliged by limitation of financial resources to physicians of any health service.