Meta-analyses include meta-analysis of the published literature (MPL) and meta-analysis of individual patient data (MIPD). Recursive cumulative meta-analysis is a method used to reorganize the secondary analysis data based on original studies thus to ensure a timely update, in addition, it can also be used to analyze the data from longer followup of existing trials. By using this method, with the each newly included or updated study, the change of pooled effect size in each pooled step can be detected, therefore, the bias/heterogeneity and stability of pooled results can be evaluated. In this article, we briefly introduced the concept of recursive cumulative meta-analysis and an example was used to illustrate this method.
Meta-analysis has been regarded as the critical tool of assisting the healthcare professionals to make decisions. And the theory of evidence-based medicine is widely disseminated in domestic. However, it must be noted that the increasing number of meta-analyses causes a fact that several meta-analyses investigating the same or similar clinical questions were captured commonly. More importantly, the results from these meta-analyses are often conflicting. Consequently, decision-making of those healthcare professionals who depend on those results become a thorny thing. To address this issue, Jadad et al. from McMaster University proposed an adjunct algorithm to help healthcare professionals to select the best result from conflicting meta-analyses to make decisions properly. Our article will introduce the tool briefly and explain the process of it with an example.