Critical interpretation of pairwise and network meta-analysis of randomized respiratory clinical trials

Abstract

A meta-analysis is a statistical technique used to assess the data of independent studies concerning the same clinical scenario, in order to synthesize results that are reported as effect estimate. The strength of a meta-analysis lies in its potential to combine the results of studies characterized by different size and that may have been underpowered to answer clinically relevant questions. Furthermore, meta-analyses can be used to clarify questions for which large randomized controlled trials have not led to consensus within the scientific community. The effect estimate resulting from a meta-analysis should be interpreted both from a statistical and clinical point of view. The clinical interpretation of the effect estimate must take into consideration the minimal clinically important differences compared to the comparator, which may be placebo and/or other active treatments. In this review we consider the key points needed to correctly and critically interpret the current meta-analyses, and to assess how reliable are the results from a statistical and clinical point of view.

Keywords

• BPCO; broncodilatatori; corticosteroidi; farmacoterapia; meta-analisi.