Many Reviews Are Systematic but Some Are More Transparent and Completely Reported than Others

T he past decade has seen the establishment of the systematic review (SR) as one of the cornerstones of evidence-based medicine. The value of SRs to researchers, practitioners, and policy makers is well established; when done well, they are considered the highest level of evidence for medical decision making. Potentially, they are also a resource for patients seeking to make sense of the relative value of different treatments. It is important to distinguish SRs from the traditional, narrative reviews also often published in medical journals. A helpful defi nition of a SR, which also clarifi es the difference between an SR and meta-analysis, comes from the Web site (http://www.cochrane. org/resources/glossary.htm) of the Cochrane Collaboration (named after Archie Cochrane, a British medical researcher), an international organization that publishes rigorous SRs evaluating the effectiveness of a wide range of health care interventions. Their defi nition of a SR is this: " A review of a clearly formulated question that uses systematic and explicit methods to identify, select, and critically appraise relevant research, and to collect and analyze data from the studies that are included in the review. Statistical methods (meta-analysis) may or may not be used to analyze and summarize the results of the included studies. " The increasing number of new SRs being published (currently estimated as being around 2,500 per year [1]) have become an essential part of the biomedical literature. But SRs that are of low quality or out of date have the potential to mislead, and selective publication of SRs that support particular agendas—or failure to publish those with " undesirable " results—could undermine the literature's reliability in the same way that biased publication of primary research can. How rigorously are these reviews being performed? How consistent are they in reporting their methods and their results? In this issue we publish an article [1] by David Moher and colleagues that examines both the " epidemiological aspects " of published SRs and their " reporting characteristics. " It is not the fi rst time SR quality has been examined, but this study is different in that it does not focus on a particular area of health care, one type of intervention, or a sample of journals, but on all SRs indexed on PubMed within a specifi ed period of time—one month, November 2004. The authors included both SRs published in the Cochrane Database of Systematic Reviews (CDSR) and those …