Skip to Main Content

Systematic reviews

A brief overview of systematic reviews and resources to support producing one.

Selection (screening)

After databases (and any other sources) have been searched, and duplicates removed, the next step is to select the papers that will go on to be included in your review project. This selection phase is also known as screening.

Firstly the title and abstract of the citations are checked by two reviewers independently to decide whether each citation should be included or excluded (this is often called title/abstract screening). The process for resolving any disagreements should be specified before beginning this phase. 

The citations which are included from the title/abstract screening step go on to the next stage, full text screening. The full text of these papers is obtained, and they are again checked for inclusion or exclusion according to the review criteria. At this stage having the full text of the paper available enables a final decision to be made for each paper. The papers included at this stage are the ones that will be included in the review (i.e. they are the ones that will have data extracted from them if this is a project that will synthesise the data from multiple papers).

The numbers of citations included and excluded for this process need to be recorded for the review. There is a free tool called the PRISMA flow diagram which is commonly used to demonstrate the process of inclusion and exclusion and to record the numbers of references. For reviews to be PRISMA compliant they must also adhere to the other requirements presented at that website. The PRISMA flow diagram is available from the PRISMA statement website:

Rethlefsen ML, Page MJ. PRISMA 2020 and PRISMA-S: common questions on tracking records and the flow diagram. J Med Libr Assoc. 2022;110(2):253-7.
 

Using software tools for selection

Various software tools are available to help with screening. 

If you are undertaking a project or assignment, or are looking for a solution other than Covidence, we'd recommend using one of these tools:

  • EndNote. Screening can be done just using EndNote. An example workflow might be:
    • Make a copy of the library containing all of your search results. This is so you can keep the original library of search results as a record
    • Deduplicate this copied library. As you already have your results in EndNote you may find it easier to remove duplicates within EndNote
    • Make a copy of this deduplicated library. Again, this is so you can keep the deduplicated library as a record 
    • In this latest copy of the library (the one you will use for screening), make groups called 'Include', 'Exclude', 'Unsure'. Also, make sure all the references are in the 'Unfiled' area, by deleting any references that are in groups (when you delete a reference from a group, it goes to the Unfiled area) 
    • Make a copy of this library for the other person who will be screening
    • Each screener works through their library, dragging the references from the Unfiled area into the appropriate groups
    • The two screeners then come together to compare decisions and resolve any disputes
  • Screenatron. This is one of the free tools available as part of the Systematic Review Accelerator. The SRA tools work well when used together with EndNote. You will first need to deduplicate your results (the SRA also has a Deduplicator so you may wish to first use that before using the Screenatron)
  • Rayyan. Has a free plan with limited features, for 3 active reviews. On the free plan deduplication is a more manual process, so you might wish to deduplicate your search results before uploading.
  • JBI SUMARI. Full access is available as part of the JBI EBP database (find SUMARI under the EBP Tools menu). You will first need to deduplicate your search results before uploading to SUMARI

Covidence

For UQ researchers undertaking an evidence synthesis project (including Cochrane reviews), Covidence is recommended. UQ Library subscribes to Covidence - an online software product designed to improve the efficiency and experience of creating and maintaining systematic reviews. Covidence can help you with deduplication, and the screening and extraction phases of your project.

UQ researchers, please follow the steps on our signing up to Covidence page to ensure that you get full access to UQ's subscription.

The following resources will help you navigate Covidence: