Introduction to meta-analysis: meta-analysis of continuous outcomes

2.45 - 4.15pm

Objectives

To provide review authors with the knowledge and skills that they need to abstract continuous data from trial reports and perform a meta-analysis of such data.

Description

The workshop will cover the theoretical aspects: what are continuous data; standard deviations; standard errors; extracting standard deviations from trial reports (directly and indirectly from standard errors, confidence intervals and P values); and methods for pooling continuous outcomes (weighted mean difference, standardised mean difference). It will illustrate these with practical work abstracting data from papers and performing a meta-analysis in RevMan5. Approximately 20% of Cochrane reviews consider continuous outcomes. The proposed workshop will help to improve the quality of such reviews by covering the basic statistical skills which they require.

Target audience

Review authors and students

Level

Basic

Presenter

Andrew Bryant, Institute of Health and Society, Newcastle University, UK.

Andrew Bryant