What PRISMA 2020 asks of systematic review authors

A systematic review is only as useful as its report. If readers cannot see how studies were found, selected and appraised, they cannot judge whether the conclusions follow. The PRISMA statement exists to make that visible.

PRISMA 2020, published in The BMJ in 2021 by Page and colleagues, replaced the original 2009 statement. It reflects a decade of change in how reviews are done: new search methods, new risk-of-bias tools, and the routine use of certainty-of-evidence ratings.

What the update provides

PRISMA 2020 includes a 27-item checklist for the full report, a separate checklist for abstracts, and revised flow diagrams that show how records moved from identification to inclusion — including records found outside databases, such as in trial registries.

It also asks authors to report more about their methods: the full search strategy for every database, how risk of bias was assessed, how certainty of the evidence was rated, and where the protocol, data and code can be found.

How we use it

The Evidence Review and Scientific Integrity Unit reports its own reviews to PRISMA 2020, appraises others’ reviews with AMSTAR 2, and assesses included studies with RoB 2 or ROBINS-I. Certainty of evidence is rated with GRADE.