About Step 5: Screen Citations
Step 5: Screen Citations
Average time (in hours) to complete
In this step, you will screen all articles to decide if they are relevant to your research question. You will:
- Use a screening (or SR) manager to organize articles
- Read the title & abstract of each article; vote to include or exclude them
- Resolve disputes with your project team
- Find the full text of all included articles
- Read the full text of included articles to decide if they should still be included
How a librarian can help with this step
During the study selection or screening step, ZSR can help you:
- Support the screening process by helping you apply your inclusion and exclusion criteria
- Use Covidence to screen citations
- Create a random sample of citations for beta testing the screening phase before full implementation
Citation screening tools
As a reminder, Covidence, the software specifically designed to help you manage a SR project, is available to anyone with a wakehealth.edu or wfu.edu email extension.
To request an invitation to join the Wake Forest University School of Medicine Carpenter Library Covidence account CLICK HERE!
Other FREE web-based options are available. They support reference importing of multiple file types, including/excluding references, exporting results, and tagging with keywords. Check out the following research guides or videos on Rayyan, abstrackr, and HAWC:
Rayyan for Systematic Reviews (University of Hawaii at Manoa John A. Burns School of Medicine)
abstrackr Center for Evidence Synthesis in Health at Brown University)
HAWC (Health Assessment Workspace Collaborative) Resources < Tutorials
Setting up your project in Covidence & importing citations
Setting up your project in Covidence is easy. Once you've created your account, select "Start a new review." In "Settings," visit each tab to input the requested data.
If members of your team don't have access to Covidence, you can invite them to join your project (in the "Reviewers" tab) where they will have full functionality in each step. They will not be able to start new projects, however.
Next, you'll import your references so you can start screening. Check out this video from Covidence on importing your citations to your new project:
Screening citations
Now that all your citations are in Covidence (or other screening manager), you're ready to screen!
This step is designed to remove studies that don't meet your inclusion criteria but still got through the literature search due to the various limitations of - even the best designed - database search engines and search strategies.
To reduce bias in the screening process at least two reviewers should screen studies INDEPENDENTLY and BLINDED.* It's idea for a third reviewer to resolve conflicts, but in the absence of a third reviewer, the project team can work together to break voting ties.
*Voting team members should be unable to see how others cast their votes. Covidence is designed to blind votes through the screening process.
During title / abstract screening, reviewers are to read the title and abstract for each article.
There are three possible decisions in this step:
After all citations have been screened, the third review (or project team) needs to review any conflicts (yes/no, yes/maybe, no/maybe) and determine whether those references should be excluded or move on to full text screening.
During full text screening, reviewers are to read the full article.
There are two possible decisions in this step:
During full text screening, reviewers should select a reason for exclusion. It is beneficial to select a few likely reasons for exclusion before the full text screening step and to place them in a hierarchy. For instance, the primary reason for exclusion may be "Wrong Study Design" but the study may have also observed the "Wrong Population." The project team should make and refine decisions on the exclusion hierarchy during a pilot or beta screening phase,* but also review any conflicts on exclusion reasons before moving on to the next step.
*Pilot or Beta Testing - Take a random sample of citations before the title / abstract and full text screening phases and have all reviewers screen the sample to ensure consistency across responses and confirm your exclusion reasons and hierarchy are appropriate. This extra step could save time and lead to fewer errors and conflicts.
PRISMA (Reporting Standards for Systematic Reviews)
Use PRISMA to document citation screening methods: 
If you'll recall, the PRISMA-P 2015 Statement & Checklist is used to help you write the SR protocol. The PRISMA Statement also consists of a 27-item checklist and four-phase flow diagram.
You will track the number of:
1. citation results from each database you use
2. duplicates you remove,
3. decisions you make on each citation, and
4. reasons you exclude a citation (in the full text screening phase)
PRISMA & Covidence
Covidence generates the PRISMA flow diagram for you while it tracks the numbers of citations screened (and included / excluded) in each stage of the review. If you don't use Covidence, you will still need to record this information by using the PRISMA diagram template.