This workshop is given on behalf of the Statistical Methods Group.

Workshop type: Training.

Objective: To provide review authors with the knowledge of issues surrounding meta-analysis of continuous outcomes.

Summary:
Continuous data are commonly encountered in health care. Pooling intervention effects from continuous data presents many issues. Some of these issues will be discussed in this workshop.

A brief introduction to meta-analysis of continuous outcomes will be included, consisting of: data extraction (extraction of standard deviations from standard errors, confidence intervals, test statistics, P values); and dealing with outcomes measured on different scales. More complex issues will be discussed including: options for pooling estimates of intervention effect when a mix of results from analyses using change from baseline and final values have been reported; incorporation of r esults from analysis of covariance and regression models; and use of the generic inverse variance.

Issues will be illustrated by examples. Time will be allowed for discussion of the issues raised.

[The workshop is part of a series of five workshops on meta-analysis, but does not require attendance at any of the others in the series.]

Intended audience: Review authors and editors.

Technical level expected of participants: Basic/Intermediate.