During this step, you are deciding which articles to include or exclude for your systematic review.
These are some free tools available to help with the screening process.
Using AI Tools to Screen Articles for Systematic Reviews: Benefits and Cautions
You may choose to use Artificial Intelligence (AI)-assisted tools such as Rayyan, Abstrackr, or others to help with the article screening phase of your systematic review. These tools offer several benefits, including:
Faster screening of large numbers of abstracts
Prediction or suggestion features to identify relevant articles based on your decisions
Collaboration support for multi-reviewer projects
Tagging and labeling for organizing reasons for inclusion/exclusion
However, use caution when relying on AI-assisted screening:
AI predictions are not foolproof: Suggested inclusions/exclusions should always be verified by human reviewers.
Risk of bias: Early screening decisions may influence the algorithm's learning, introducing bias.
Lack of transparency: Some tools use opaque algorithms that don’t explain how decisions are made.
Incomplete automation: These tools are meant to assist, not replace, thorough human review.
Best Practices:
Treat AI suggestions as preliminary—final decisions should be made according to your review protocol.
Ensure at least two independent human screeners are involved when using AI-assisted tools.
Document how and when AI tools were used in your review methods section.
When in doubt, discuss discrepancies with your review team or advisor.
AI can enhance efficiency, but not replace the rigor required in systematic review methodology.
When you have completed literature searching and gathered all citations, it is time to screen the results. The purpose of screening is to remove studies that do not meet your inclusion criteria. At least two members of your team should independently screen all studies, starting with a title and abstract screening, followed by a full-text screening. If a conflict arises the reviewers discuss until an agreement is reached or an additional reviewer mediates. Whichever method you use, it needs to be consistent and described as part of the methodology.
After you complete title/abstract screening, you will need to review any conflicts between reviewers and decide if those citations should move forward to full text screening or be marked as irrelevant.
When full-text screening, for each reference, read the full-text and make a decision to:
After full-text screening, review any citations where there are different opinions about inclusion/exclusion, as well as different reasons for excluding.
Can't find the full text?
ASK US!
You can also Ask a Librarian about visiting one of our partner libraries.