A practical introduction individual participant data meta-analysis

Core Methods Workshop

Workshop category: 

  • Statistical methods
Date and Location

Date: 

Tuesday 22 October 2019 - 16:00 to 17:30

Location: 

Contact persons and facilitators

Contact person:

Facilitators:

Nevitt S1, Rydzewska L2
1 University of Liverpool, UK
2 MRC Clinical Trials Unit, University College London, UK
Target audience

Target audience: 

Review authors, members of Cochrane Editorial teams and any researcher with an interest in individual participant data meta-analysis

Level of difficulty: 

Intermediate
Type of workshop

Type of workshop : 

Training
Abstract

Abstract:

Background: systematic reviews of individual participant data (IPD), where individual-level data from all studies addressing a particular question are collected, checked and analysed, are widely regarded as the ‘gold standard’ approach to evidence-synthesis. The number of published Cochrane and non-Cochrane Reviews using IPD is increasing every year and this is likely to keep rising with the increasing prevalence of data-sharing initiatives.

Objectives: to provide a practical introduction to IPD meta-analysis. This will also consider the advantages and disadvantages of the IPD approach compared with traditional meta-analyses based on summary-level data, with a particular focus on experiences of requesting IPD and how to include those data in Cochrane reviews.

Description: experienced members of the Cochrane IPD Meta-analysis Methods Group will facilitate the workshop, which will focus on the following questions.
- What is an IPD meta-analysis?
- Why and when should I consider using IPD in meta-analysis?
- What is the process of an IPD review, and how does it differ from the summary-level approach?

We will draw on our personal experiences to provide guidance and practical tips for conducting this type of review and requesting IPD from a diverse range of sources, including both publicly funded and pharmaceutical company-sponsored studies. We will use examples based on completed IPD reviews to illustrate the issues presented. We will also provide practical advice about what can be done if a subset of the data is not available to include in a particular meta-analysis.

There will be dedicated time throughout the workshop for interactive group discussion and attendees will also have the opportunity to share and discuss their own experiences and/or their ideas for IPD reviews.

At the end of the workshop, attendees should be able to decide if the IPD approach would be appropriate for them.