If you’re completing a review that looks only at effects of health interventions through Covidence and want to only consider studies reporting on Randomised Controlled Trials (RCTs), we now have a tool to help you sort out the non-RCTs so that you can focus your effort on studies more likely to be included in your review.

We’ve integrated with the leading RCT classifier algorithm to provide you with a prediction on whether studies in your review potentially report on an RCT. The Classifier has been endorsed by Cochrane and is also being used in the Cochrane Screen4Me workflows, which also includes Cochrane Crowd.

The classifier has shown in testing to successfully identify over 99.5% of health-related studies that potentially report on RCTs, only incorrectly classifying ~0.5% of RCTs as not being an RCT (Thomas et al, 2021).

 

How it works in Covidence

Enabling the feature

The tagging functionality can be enabled on health-related reviews when creating a review, or can be updated on existing reviews via the review settings page.

When creating a review, you will be shown a setting called “Automatically tag studies reporting on RCTs using the Cochrane RCT Classifier” when the review’s area is set to “Medical and health sciences”. By enabling the option, the review will have the feature turned on:

Tagging of studies

Once the feature is turned on for the health-related review, all references imported to the review will be run through a machine learning algorithm to predict whether each study is potentially reporting on an RCT. If references have previously been imported to the review when you enable the feature, we will automatically run these through the classifier as well.

Studies that the classifier predicts are potentially reporting on RCTs will be tagged with “Possible RCT”:

Studies that the classifier predicts are not reporting on RCTs will be tagged with “Not RCT”:

Feature limitations

The feature only works for studies with a title longer than 15 characters, and an abstract longer than 400 characters. Studies that haven’t been run through the classifier won’t have a tag applied to them.

Filtering

The studies list can be filtered by the RCT tags, allowing you to see a list of only “Possible RCTs” or “Not RCTs”:

Learn more about how filtering works here.

 

Do you have feedback or suggestions on how we can improve this feature?

Let us know by submitting your feedback here.