How to Screen Studies in Rayyan for a Systematic Review
Set up the review
- Create a new review and give it a clear name with your project code.
- Import your search results from each database (RIS or similar export files).
- Run duplicate detection and resolve duplicates before anyone starts screening.
- Invite your team as collaborators. Write your inclusion and exclusion criteria in the review description so everyone screens to the same rules.
Use Blind Mode
Blind Mode hides each reviewer's decisions from the others, so two people screen independently. This is what makes your screening "dual and independent" in the methods section. Only the review owner can turn Blind Mode on or off, so the project lead should own the review.
Screen titles and abstracts
Use the keyboard to move quickly:
- I to include
- E to exclude
- M for maybe, when you are unsure
- R to exclude with a reason
At this stage, be inclusive. If the abstract does not give enough information to exclude a study, mark it maybe or include it and decide at full text. Add exclusion reasons that match your PRISMA flow diagram categories, such as wrong population, wrong design or no comparator.
Before the full screen, run a calibration round: both reviewers screen the same 50 records, then compare and agree on how to apply the criteria.
Resolve conflicts
- When both reviewers have finished, the owner turns Blind Mode off.
- Filter by inclusion decision and choose Conflicts to see records where one reviewer included and the other excluded.
- Discuss each one against the criteria. Bring in a third reviewer if you cannot agree.
- The reviewer whose decision changes updates it in Rayyan. You cannot change another reviewer's decision. When both decisions match, the record leaves the conflict list.
Note that a "maybe" from one reviewer does not count as a conflict, so check the Maybe list separately.
Full text screening
Move included records to full text screening, attach the PDFs, and repeat the same blind, dual process. This time every exclusion needs a reason, because PRISMA 2020 asks you to report full text exclusions by reason. Use our free PRISMA flow diagram generator to build the figure from your counts.
Tips for teams
- Set a deadline for each round. Four days per batch works well for student teams.
- Agree that nobody discusses individual records until blinding is off.
- Export your decisions at the end of each stage and keep the file with your protocol.
- Check which features your Rayyan plan includes, as some features need a paid plan.
Sources
Rayyan Help Center: screening and conflict resolution guidance. Page MJ et al. The PRISMA 2020 statement. BMJ 2021;372:n71.
Frequently asked questions
Is Rayyan free?
Rayyan has a free plan with core screening features. Some features need a paid plan, so check what yours includes.
Who can turn Blind Mode off?
Only the owner of the review.
Does a maybe count as a conflict?
No. Records marked maybe by one reviewer appear in the Maybe list, not the Conflicts list, so review them separately.
How many reviewers should screen each record?
At least two, independently, for a systematic review that meets PRISMA and Cochrane standards.
Related
PRISMA flow diagram generator · PRISMA records, reports and studies · All guides
