Statistical methods for updating meta-analyses


The following videos are from a Cochrane Learning Live webinar on Statistical methods for updating meta-analyses, delivered by Dr Mark Simmonds from the Centre for Reviews and Dissemination, University of York, UK. The webinar was organized by the Cochrane Statistical Methods Group, with support from Cochrane's Learning & Support Department.

Systematic reviews and meta-analyses generally require updating as new studies become available; this requires multiple, repeated meta-analyses. This increases the risk of attaining spurious statistical significance in each analysis. As updating of meta-analyses becomes both more common and more frequent there is an increasing risk that meta-analysis results may be misinterpreted, particularly if readers are unaware of the updating process.

In this presentation, Dr Mark Simmonds describes several methods to address this problem when meta-analyses are repeatedly updated, which have been assessed using a formal simulation study, and by applying methods to a range of recently-updated systematic reviews in the Cochrane Library. The questions and answers are facilitated by Dr Joanne McKenzie from Monash University, Melbourne, Australia.

Below you will find slides from the webinar [PDF] as well as edited videos covering the presentation and the questions and answers session.


Part 1: Presentation

Part 2: Questions and answers


Author information: 
Simmonds, Mark; McKenzie, Joanne; Watts, Chris; Sambunjak, Dario