• English
  • Español

How should we write up the statistical results of meta-analyses in the text and abstract of a Cochrane systematic review?

Topic category Knowledge translation and communicating evidence
Date and Location
Date: 
Friday 21 October 2011 - 16:00 - 17:30
Location: 
Methods Group
Methods Group: 
None
Contact person
Contact person: 
Elaine Beller (Contact this person)
Sally Hopewell (Contact this person)
Facilitators
First nameLast nameAffiliation and Country
First name: 
Elaine
Last name: 
Beller
Affiliation and Country: 
Bond University, Australia
First name: 
Sally
Last name: 
Hopewell
Affiliation and Country: 
UK Cochrane Centre, UK
Other contributors
First nameLast nameAffiliation and Country
First name: 
Kate
Last name: 
Cahill
Affiliation and Country: 
Tobacco Addiction Review Group
First name: 
Paul
Last name: 
Glasziou
Affiliation and Country: 
Bond University, Australia
Target audience
Target audience: 
review authors, managing editors, coordinating editors
Is your workshop restricted to a specific audience or open to all Colloquium participants?: 
Open
Level of knowledge required: 
Intermediate
Type of workshop
Type of workshop: 
Training
Abstract text
Abstract: 
Objectives:To give participants confidence and practical skills in taking a statistical result and ensuring the non-statistical reader understands the meaning.

Description: When we obtain the numerical result of a meta-analysis, we end up with an estimated effect size, measure of precision (confidence interval) and a P-value. How do we take those statistical results, and convey them in words in the text or the abstract of A Cochrane review, so that readers will understand what they mean? This workshop will take examples of statistically significant and non-significant results, using the common effect size measures such as odds ratio, risk ratio, mean difference and standardised mean difference, and guide participants through translating those results into meaningful text, hopefully without imparting ‘spin’ on the results! We will begin with the easier cases, such as when the result is highly significant, and easy to explain, and work our way to the more difficult cases, such as when the result is ‘almost significant’, or has a wide confidence interval, or uses standardised mean difference. The talks will be brief, and the emphasis will be on participation in the writing tasks. All examples are taken from recent systematic reviews in the Cochrane and non-Cochrane literature.