RA4: Advanced Topics in the Analysis and Reporting of Systematic Reviews

Introduction

RA4 contains a large amount of training material and activities to cover common issues which arise with continuous data types. It also covers how to deal with data in different formats, over-coming unit of analysis issues and dealing with multiplicity in reviews, as well as using GRADE and Summary of Findings Tables. The workshop aims to raise awareness of a variety of methods that are available for use in systematic reviews to overcome common pitfalls, and to empower review authors to seek statistical support when appropriate.  

Should I attend?

This workshop is suitable if you:

  •      have completed Workshops RA1 to RA3 or have the equivalent level of knowledge
  •      want to be able to deal with a wider range of data types and analysis methods
  •      want to be able to produce the summary of findings section of your review.

What will be covered?

  •      The generic inverse variance method of meta-analysis
  •      Continuous data with zero and small counts
  •      Converting other data types to binary form
  •      Calculating standard deviations when they are missing from papers, using other reported values
  •      Using change scores, time-to-event data
  •      Dealing with multiplicity - outcomes, interventions & crossover trials
  •      Assessing quality of evidence using the GRADE method
  •      Summary of findings tables

How is the workshop run?

This workshop comprises lectures and both individual and group practical exercises. Handouts of slides used in the lectures are provided.

What will I achieve?

By the end of the workshop, you should be able to:

  •      deal with continuous data that are not in the usual form of odds ratio, risk ratio or risk difference
  •      deal with continuous data where standard deviations are missing from publications
  •      understand the methods for handling other formsof data and be able to communicate with a statistician for these analyses
  •      understand and construct a summary of findings table