Statistical technique for combining the results of multiple studies into a single pooled estimate
Should only be done in the context of a systematic review, not just selecting certain studies and creating a pooled estimate, which causes bias
An optional final step of a systematic review, where appropriate (but it’s not always appropriate, for example if studies are too different to combine meaningfully)
Happens at the synthesising evidence step of a systematic review
It’s not just a narrative synthesis, it’s also statistical analysis that calculates an overall estimate of effect
Comments
No comments have yet been made