Workshop Title: Individual Patient Meta-Analysis

Workshop moderator: Ashraf Fawzy Nabhan

Workshop format: Quiz game

Venue: Fairmont Hotel

Date: Thursday May 31, 2012 – at 3:30 pm

Background

In all systematic reviews, the evidence being brought together for a particular intervention needs to be as free from bias and as reliable as possible. This is why most reviews of the effects of interventions rely exclusively on randomized trials. The process of bringing together the evidence also needs to minimize bias, be rigorous and robust. The ultimate aim of any systematic review of randomized trials should be to ensure that relevant data on all randomized participants from all relevant trials are included. Increasing the number of trials or participants reduces the influence of chance effects, and produces a more tightly defined and precise estimate of the difference between the effects of the interventions under investigation.

In most systematic reviews, meta-analyses are based on aggregate data from the included studies. Individual patient data meta-analyses rely on the central collection and re-analysis of the raw data for each study.

This workshop will  highlight the rationale for individual patient data meta-analyses and the processes of obtaining individual patient data for inclusion in systematic reviews. Reviews based on individual patient data have been described as the ‘yardstick’ against which other forms of systematic review should be measured. They are the most resource-intensive and time-consuming but have, in a number of cases, produced definitive answers which might not have been obtained in any other way.