Screening your search results
Screening is the process of determining which papers are included in your review. Learn more about those next processes in the Systematic and systematic-like review toolkit which covers the whole review.
There are two screening stages, ‘Title and Abstract Screening’ and ‘Full Text Screening’. Each screening stage is typically undertaken by two independent reviewers to minimise bias.
Screening process
Step 1: Refer to your inclusion and exclusion criteria
These criteria are developed when writing your protocol and should be established at that time as not to introduce bias. Find out about review protocols and inclusion/exclusion criteria in the Systematic and systematic-like review toolkit.
Inclusion criteria: Characteristics the study must have to be eligible to be included in the review.
Exclusion criteria: Characteristics that would exclude a study from inclusion in the review.
If you are using a tool e.g. Covidence to screen, set these criteria before importing your results into the screening tool
Step 2: Title and abstract screening
Reviewers will independently screen all articles by title and abstract to remove obviously irrelevant material. At this stage you may not need to provide justification for your exclusions:
- Screen the titles and abstracts of your results against your inclusion and exclusion criteria.
- Articles that do not meet the inclusion criteria do not move to the next screening stage.
- Articles that meet inclusion criteria or those with unclear eligibility proceed to the second stage of screening, which involves a full-text review.
- Often a third reviewer is used to resolve conflicts between other reviewers. This is often the person who is the foremost expert in the subject.
Step 3: Full text screening
Articles which were not excluded in the title and abstract screening stage undergo a full text review. Reviewers examine each article in detail to assess eligibility for inclusion in final review.
- Source the full text of all articles.
- Review the full text and record reasoning for inclusion or exclusion.
- If reviewers' inclusion and exclusion decisions conflict, use a third reviewer to resolve the decision.
NOTE: When you are ready to source the full text (PDF) of records which make it to the full text screening stage in Covidence, you can export a list of these articles into a new EndNote Library. Then use a combination of EndNote’s “Find Full Text” feature and manual searching to source and attach the full text to these EndNote records. You can the upload these PDFs in bulk and Covidence will recognise and attach the documents to their corresponding record. See the Covidence guide for instructions.
PRISMA flow diagram
The PRISMA flow diagram is a tool which visually demonstrates your decisions through the screening process. The flow diagram is designed to make the selection process more transparent, clearly reporting the number of records in the search results, each screening stage, and data extraction.
You can download your own copy of the PRISMA Flow diagram from the PRISMA statement website. We have developed a completed Example of the PRISMA Flow diagram so that you can see how it might be filled in.
Full documentation available in PRISMA 2020 Explanation and Elaboration paper, and PRISMA statement website.
Activity: Explore the PRISMA 2020 Flow Diagram
Click on the plus (+) icons to learn more about each stage in the PRISMA screening process for the identification of studies via databases and registers (registries generally refer to registers of trials).
*Consider, if feasible to do so, reporting the number of records identified from each database or register searched (rather than the total number across all databases/registers).
**If automation tools were used, indicate how many records were excluded by a human and how many were excluded by automation tools.
Covidence - a screening tool
Covidence is a fantastic online tool that helps you screen citations, abstracts, and full-text studies for systematic and advanced reviews. It also supports risk of bias assessment, data extraction, and allows you to export data and references in CSV format or directly to Excel.
This tool can save you valuable time by enabling multiple reviewers to access the search results, streamlining your decision-making and helping managing conflicts throughout the screening stages.
To get started, you can create an unlimited account by following the instructions below on how to Create a Covidence account and link to Deakin.
Once you have created a Deakin account, you can access Covidence directly via https://www.covidence.org/.
Create a Covidence account and link to Deakin
Follow the steps below to create a Covidence account and then link it to Deakin's subscription:
- Request an invitation to Deakin University's Covidence account by completing the invitation request form using your Deakin email address.
- You will receive an invitation by email. Select the link in the email to confirm your account creation.
- You will be asked to login to Covidence to accept the invite. If you don't already have a Covidence account, you will need to sign up. Please use your Deakin email address as your Covidence username.
If you have waited a reasonable period and still not received the invitation or have other issues with requesting an invitation using your Deakin email address, please email support@covidence.org with details of the difficulties.
Learn more about Covidence
For more detailed guidance, check out our dedicated Covidence page which includes step-by-step instructions and information about its various features.
We also recommend exploring The Covidence Academy, which offers extensive resources on both using Covidence and the systematic review process.
Tip
LibKey Nomad provides you with fast and easy access to full-text articles while using Covidence. Follow these steps to install LibKey Nomad in your browser and it will automatically integrate with your Covidence review.
Continuing your review
From this point you will continue going through the steps to appraise the studies, extract the data, write the analysis and publish the study. For support for these steps, refer to the Systematic and systematic-like review toolkit or speak to your research team, colleagues and supervisors.