Systematic Review Screening Process

There are several screening stages in a systematic review:

1. Title/Abstract screening based on your inclusion/exclusion criteria.
2. Full text screening based on your inclusion criteria and quality assessment criteria (or risk of bias tool).
3. Additional searching.
4. New items identified are then screened as above.
5. Review team agree which studies to include in final review.

For further details, see Polanin et al. 'Best practice guidelines for abstract screening large-evidence systematic reviews and meta-analyses', Res Syn Meth, 10, (2019), pp. 330-342, https://doi.org/10.1002/jrsm.1354.

To reduce bias, it is best practice to have two reviewers to work through the screening stages. Some guidelines and publications require a second reviewer. Further advice: 

• Waffenschmidt et al. 'Single screening versus conventional double screening for study selection in systematic reviews: a methodological systematic review', BMC Med Res Methodol, 19, 132, (2019), https://doi.org/10.1186/s12874-019-0782-0. 
• Mahtani et al. 'Single screening or double screening for study selection in systematic reviews?', BMJ evidence-based medicine, 25, 4, (2020), pp. 149-150, https://doi.org/10.1136/bmjebm-2019-111269. 

Systematic review software

The University of Leicester does not have an institutional subscription to any specialist systematic review software. You will either need to use free/trial versions or fund an individual subscription. For further details, check the websites of the relevant software:

• Covidence (Free trial for up to 500 records).
• Rayyan (Free version available)
• RevMan (Review Manager – used for Cochrane Reviews, subscription required)

There is also other subscription software such as EPPI Reviewer and JBI Sumari.

The library can only advise on how to export references from literature databases in preparation for their use in reviewing software. We don't provide advice or training on how to use reviewing software itself.

Referencing software

Referencing (or bibliographic) software can also be used for deduplicating records and for the screening process. You would need to set up a folder structure within the software to assist you with the screening stages.

We would recommend either EndNote or Zotero for this task. If you have never used these tools before, start with our Referencing Software guides [login required].

If you are going to use EndNote in your project, we recommend this guide produced by UCL: EndNote for systematic reviews.

AI and Machine Learning tools

This is a fast moving area, but currently we recommend starting with King’s College London Guide to AI tools in evidence synthesis. This includes an overview of the tools available, research evaluating their effectiveness and relevant position statements (e.g from NICE).

If you use AI/Machine Learning tools you will need to:

• research and evaluate the tools yourselves
• understand the tools and their limitations in order to use them appropriately
• be transparent: state which tools you have used, and how you have used them, in any resulting publications and/or your thesis
• check the University of Leicester statement on The use of generative (AI) tools in research [login required].