QUANTITATIVE SYNTHESIS IN SYSTEMATIC REVIEWS

The final common pathway for most systematic reviews is a statistical summary of the data, or meta-analysis. The complex methods used in met a-analyses should always be complemented by clinical acumen and common sense in designing the protocol of a systematic review, deciding whic h data can be combined, and determining whether data should be combine d. Both continuous and binary data can be pooled. Most meta-analyses s ummarize data from randomized trials, but other applications, such as the evaluation of diagnostic test performance and observational studie s, have also been developed. The statistical methods of meta-analysis aim at evaluating the diversity (heterogeneity) among the results of d ifferent studies, exploring and explaining observed heterogeneity, and estimating a common pooled effect with increased precision. Fixed-eff ects models assume that an intervention has a single true effect, wher eas random-effects models assume that an effect may vary across studie s. Meta-regression analyses, by using each study rather than each pati ent as a unit of observation, can help to evaluate the effect of indiv idual variables on the magnitude of an observed effect and thus may so metimes explain why study results differ. It is also important to asse ss the robustness of conclusions through sensitivity analyses and a fo rmal evaluation of potential sources of bias, including publication bi as and the effect of the quality of the studies on the observed effect .