Systematic reviews (SR) are a type of clinical research, and thus they follow the hypothetical-deductive or scientific method. Their objective is to answer clinical questions and they do it based on a specific structure (protocol) and working through inferences. Interestingly, a SR is a type of study design that can answer all types of questions, from efficacy to incidence. They manage the information from previous studies in a pre-established reproducible way, as unbiased as possible. They provide quick answers. However, they should be performed and interpreted with caution: A poor systematic review is much worse than a narrative review, as it gives the false impression of “science”. This is even worse if the studies were combined in a meta-analysis. In this session, we will review the protocol of a SR, the importance of rephrasing the question, the search strategy, selection criteria, procedures of studies selection, primary endpoints, and quality and risk of bias. The last part of any SR is the analysis, which is always qualitative, supported by the evidence table, and following a synthetic discourse with the PICO order, and sometimes quantitative (meta-analysis). We will review some methods to evaluate publication bias, to combine results, and to explore heterogeneity. In this sense, it is essential to note that if we cannot find a valid explanation for heterogeneity, our results may not be valid.
Disclosure of Interest None declared
Statistics from Altmetric.com
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.