Skip to Main Content

Systematic Reviews

On This Page

On this page you will find:

  • Information on data management for Systematic Reviews
  • An overview of the deduplication and screening processes
  • Useful automation tools and platforms to manage studies and streamline the deduplication and screening process.
  • Relevant AI tools 

Data Management

Data Management

It is not uncommon for researchers to deal with hundreds if not thousands of studies at the end of their Systematic Review search. The records for each of these studies will need to be stored in a way that will make them accessible and manageable by all members of the review team. There are a number of data management tools and platforms  

Further reading:

Deduplication

Deduplication

A systematic review search is expected to be extensive, and it is standard practice to search across multiple databases. As a result, it is common to find the same study multiple times by the end of the searching process. While it is required to report the total number of studies found in each database, it will become necessary to de-duplicate the total number of studies to reduce the amount of time required later during the screening process.

Further reading:

Screening

Screening

Once the set of records has been deduplicated, they will need to be screened against the pre-defined eligibility criteria. This process requires a team of at least 2 and is completed in phases focusing first on the titles and abstracts, and then on the full-text. 

Further reading:

Tools & Platforms

Manually managing, deduplicating and screening search results for a Systematic Review can be time-consuming, inaccurate, and ill-advised.

There are various tools and platforms available to help researchers manage, deduplicate and screen search results.

For a comprehensive list of Systematic Review tools, see the Systematic Review Toolbox.

Covidence

Covidence is a screening and data extraction tool which can be used to store and deduplicate records. Covidence is endorsed by Cochrane.

NSLHD does not provide access to Covidence. If you or a member of your review team has a university affiliation you may be able to gain access to Covidence.

Rayyan

Rayyan's is a workflow management platform which can store and deduplicate records.

It has a subscription cost but is free for basic use.

The Joanna Briggs Institutes’ premier software for the systematic review of the literature. Registration is required.

EndNote is a reference management software that assists in storing and organising references. For systematic reviews, EndNote can also be used for deduplication, screening and eligibility assessment. EndNote is available for NSLHD staff.

See: EndNote guide

AI for Systematic Reviews - Screening

Many of the tools listed above include AI features. These features include:

Covidence

NSLHD does not provide access to Covidence. If you or a member of your review team has a university affiliation you may be able to gain access to Covidence.

  • Randomized Controlled Trial (RCT) Classifier
    The Covidence RCT classifier is a machine learning tool that automatically categorises studies by whether or not they are a randomized controlled trial. It can correctly identify up to 99.5% of RCTs, screening out non-RCTs before title/abstract screening begins. 
    Read more.
  • Title/abstract screening
    When selecting "Sort by: Most Relevant" during screening, this feature begins identifying patterns in screening behaviour. Once the first 25 studies are screened, the algorithm begins automatically refreshing after each study screened, allowing the system to determine and display studies that are most likely to be included first. This reduces the number of records needing to be screened by up to 80%.
    Read more.

Rayyan

  • Rayyan Prediction Classifier
    After 50 screening decisions with a minimum of 5 included and excluded studies each, Rayyan's classifier begins identifying patterns based on screening behaviour. It calculates a confidence score for each unscreened title, providing a rating on whether the article should be included or excluded. Articles to be screened can be sorted by Most Relevant, Least Relevant, and Swap. 
    Read more.