Trial2rev: Combining machine learning and crowd-sourcing to create a shared space for updating systematic reviews

Area of Science:

• Health Informatics
• Clinical Trial Management
• Evidence Synthesis

Background:

• Systematic reviews require timely updates to reflect new clinical trial evidence.
• Manual monitoring of trial registrations and publications is time-consuming and inefficient.
• Existing methods often reactively filter published trial reports.

Purpose of the Study:

• To develop a public interface for a database linking systematic reviews with trial registrations.
• To enable faster updating of systematic reviews by monitoring trials proactively.
• To provide a resource for systematic reviewers to aid in update decisions.

Main Methods:

• Developed a server-side system using Python and PostgreSQL.
• Implemented a web-based user interface with Javascript, HTML, and CSS.
• Created a database collating information from multiple sources, including ClinicalTrials.gov.

Main Results:

• The trial2rev system offers a web interface for managing systematic reviews and trial links.
• Users can browse, search, add, and verify systematic reviews and associated trials.
• The system supports automated agents for adding and voting on relevant trials.

Conclusions:

• A publicly accessible resource was created to assist systematic reviewers with update decisions.
• The trial2rev system proactively monitors trials from registration to completion.
• This approach facilitates more efficient and timely systematic review updates.