Authors: Lesley A Stewart, Jayne F Tierney and Mike Clarke on behalf of the Cochrane Individual Patient Data Meta-analysis Methods Group.
Key points
In an individual patient data (IPD) meta-analysis, the original research data for each participant in each study are sought directly from the researchers responsible for that study.
Having access to the ‘raw’ data for each study enables data checking, thorough exploration, and re-analysis of the data in a consistent way.
IPD meta-analysis has particular benefits when the published information does not permit a good quality review, or where particular types of analyses are required that are not feasible using summary data.
Most IPD meta-analyses are carried out and published by a collaborative group, comprising a project team or secretariat, the researchers who contribute their study data, and often also an advisory group.
An IPD approach usually takes longer and costs more than a conventional systematic review relying on published or aggregate data.
There may be circumstances where the benefits of obtaining IPD are marginal; other where it could be vital.
18.2 The collaborative nature of IPD meta-analyses
Box 18.6.a: The Cochrane Individual Patient Data Meta-analysis Methods Group