Meta-analysis is an important and valuable tool for summarising data from multiple studies to provide researchers with a new pooled estimate of how much more beneficial a particular treatment is as compared to a different treatment.

Although meta-analysis does not involve human subjects or experimental animals directly, it is often considered an easy study that can be done with the minimum of effort and little attention is often paid to its design and implementation. A valid meta-analysis however, requires the same careful planning in the protocol stage as for any other research. The criterion for selecting which studies are included in the meta-analysis is important since the pooled result of the meta-analysis depends on which studies are included.