Evidence-based medicine (EBM) is the judicious and conscientious incorporation of the best available evidence into clinical decision making while considering the values and preferences of the patient [1
]. This definition of EBM invokes a hierarchy of evidence.
Systematic reviews of the literature occupy the highest position in currently proposed hierarchies of evidence[2
]. It is argued that systematic reviews should occupy this top position because of two fundamental premises. First, clinical reviews systematically search, identify, and summarize the available evidence that answers a focused clinical question, with particular attention to methodological quality. Second, reviews that include a meta-analysis provide precise estimates of the association or the treatment effect. Clinicians can apply the results of meta-analyses to a wide array of patients – certainly wider than those included in each of the primary studies – that do not differ importantly from those enrolled in the primary studies.
Narrative reviews are summaries of research that lack an explicit description of a systematic approach. Despite the emerging dominance of systematic reviews, narrative reviews persist. A study by Antman et al
] found that narrative reviews, which frequently reflected the opinion of a single expert, lagged behind the evidence, disagreed with the existing evidence, and disagreed with other published expert opinions. Mulrow[4
] and later McAlister et al
] found that these reviews lacked methods to limit the intrusion of bias in the summary or the conclusions.
Because of the importance of systematic reviews in summarizing the advances of health care knowledge, their number is growing rapidly. The Cochrane Collaboration, a world-wide enterprise to produce and disseminate systematic reviews of effectiveness, has published in excess of 1000 systematic reviews since its inception [6
]. Collectively, other groups and individuals have likely contributed three to five times that number in the past 20 years and these reviews are dispersed throughout the medical literature [8
]. Researchers wanting to define the frontier of current research and clinicians wanting to practice EBM should be able to reliably and quickly find all valid systematic reviews of the literature. Nonetheless, researchers have reported difficulty finding systematic reviews within the mass of biomedical literature represented in large bibliographic databases such as MEDLINE [9
If systematic reviews in fact represent the best available evidence, they are likely to have great clinical importance. It follows that they be cited often in the literature. The Institute for Scientific Information (ISI) impact factors reflect (albeit far from ideally [13
]) the prestige of a journal and the importance of the articles it publishes. The impact factor for a journal is the number of citations to articles published in the journal in the past two years, divided by the number of articles published during that period. ISI also reports the number of citations that individual articles receive. Since the impact factor relates to the overall citation performance of the articles a journal publishes and not to any individual article type, and since systematic reviews are a relatively small fraction of all articles published in journals, we did not expect a strong association between impact factors and the frequency of publication of systematic reviews. However, we hypothesized that the number of citations for systematic reviews would be greater than the number of citations for a "look alike" article, in this case, a narrative review published in the same journal.
Thus, we sought to answer the following research questions: (1) Where are systematic reviews published? (2) What is the relation between journal impact factor and journal yield of systematic reviews? (3) Do systematic reviews receive more citations than narrative reviews?
Answers to our first question may lead to definition of journal subsets in MEDLINE within which most systematic reviews will reside. Answers to our second and third questions will indicate whether the literature reflects the hierarchy of evidence, one of the basic tenets of EBM.