Important notice: We have taken the decision to no longer update this register after January 2024. This was not an easy decision to make. We would like to take this opportunity to thank all those who helped to contribute to the register’s rapid development at the start of 2020, and to those who have helped to feed the register to the end of 2023. We currently plan to keep the public interface accessible for 2024. If you have any queries or questions, please do not hesitate to contact us: support@cochrane.org

What is the Cochrane COVID-19 Study Register?

The Cochrane COVID-19 Study Register is a freely-available, continually-updated, annotated reference collection of human* primary studies on COVID-19, including interventional, observational, diagnostic, prognostic, epidemiological and qualitative designs.

* Note: the register does not include in-vitro study references.   

Cochrane's COVID-19 Study Register is "study-based," meaning references to the same study (e.g. press releases, trial registry records, preprints, journal pre-proofs, journal articles, retraction notices and expressions of concern) are all linked to a single study record. Search results display study records which contain one or more references for that study. Export files contain all the references, meaning export files may contain more records that search results. Tip: In the .csv export file, sort by the column "CRS Study record ID" to bring all the references for the same study together.  

The aim of the register is to support rapid evidence synthesis, including Cochrane's work on Rapid Reviews in response to COVID-19.

Where do we source our references?

Data sources for Cochrane's COVID-19 Study Register currently include:

• PubMed 
• Embase.com, provided under license from Elsevier**
• Cochrane Central Register of Controlled Trials (CENTRAL) via the Cochrane Register of Studies
• ClinicalTrials.gov
• WHO International Clinical Trials Registry Platform (ICTRP)
• medRxiv

We also monitor Retraction Watch. Retraction notices and expressions of concerns are added as references and linked to studies in the register.

** Search tip: To retrieve Embase.com references, conduct a search for the database name with an asterisk: Embase*  

How frequently do we conduct our searches? 

We search ClinicalTrials.gov daily (Monday-Friday); PubMed, Embase, ICTRP and medRxiv weekly; and CENTRAL monthly (after CENTRAL publishes at the end of each month). 

Please note: while we aim to publish eligible records to the register as quickly as possible, separate searches of our sources may be needed to ensure maximum currency of search results for review production. 

What are our search methods? 

Search Strategies: Search strategies are designed for maximum sensitivity, to retrieve all human studies on COVID-19. No language limits are applied. Strategies are updated to account for changes to the COVID-19 study register's eligibility criteria, changes to database interfaces and ongoing search performance assessments. Please contact us to request previous versions of the strategies and the dates they were implemented. 

The PubMed strategy was designed by Robin Featherstone MLIS, Information Specialist Cochrane Central Executive Team, and peer reviewed by Douglas Salzwedel MLIS, Information Specialist and Assistant Managing Editor Cochrane Hypertension. The PubMed search was validated Aug 2020 against a "gold standard" sample of 184 known PubMed study references on COVID-19. Sensitivity was 1.0 (i.e. all references in the sample were retrieved). 

The Embase.com strategy was designed by Robin Featherstone and peer reviewed by Julie Glanville MSc, Independent Consultant and Co-Convenor, Cochrane Information Retrieval Methods Group. The Embase.com strategy was validated Aug 2020 with a sensitivity of 1.0 using a sample of 177 known Embase study references on COVID-19.

How do we screen our records? 

Our search results are screened for eligibility by our machine learning classifier, Cochrane data curation specialists and information specialists, and volunteer citizen scientists. To assist with our screening work, please sign up for the COVID Quest! and COVID Quest Lite! tasks on Cochrane Crowd. 

How do we define our study characteristics? 

To enable filtering of search results, we add study characteristics (study types, aims, design and intervention assignment) to records in the Cochrane COVID-19 Study Register. For more information, please see our working definitions of study characteristics. 

How do we use machine learning?

In January 2021, we implemented a machine learning classifier to help screen our search results. The classifier was developed, calibrated and validated using three independent datasets from the Cochrane COVID-19 Study Register. In the current workflow, records identified from PubMed and Embase are run through the classifier. Those that receive a score below the calibrated cut-point are tagged as ineligible. The remaining records are sent for manual screening. A detailed description of this workflow is available as an open-access publication: 

Shemilt I, Noel-Storr A, Thomas J, Featherstone R, Mavergames C. Machine learning reduced workload for the Cochrane COVID-19 Study Register: development and evaluation of the Cochrane COVID-19 Study Classifier. Systematic Reviews. 2022 Dec;11(1):1-8. doi: 10.1186/s13643-021-01880-6

How do we evaluate the register?

We conducted an internal sensitivity evaluation of the Cochrane COVID-19 Study Register in May 2020 and a broader evaluation of the sensitivity, currency and accuracy of the register in November 2020. The November 2020 evaluation (full report and data) is available as an open-access publication: 

Metzendorf MI, Featherstone RM. Evaluation of the comprehensiveness, accuracy and currency of the Cochrane COVID-19 Study Register for supporting rapid evidence synthesis production [published online ahead of print, 2021 Jun 5]. Res Synth Methods. 2021;10.1002/jrsm.1501. doi:10.1002/jrsm.1501

Acknowledgements

The National Library of Medicine is the source of PubMed bibliographic records.

Embase.com records are provided under license by Elsevier. All rights are retained by the license holder.

Trial registry records are sourced from ClinicalTrials.gov and the World Health Organization's International Clinical Trial Registry Platform (ICTRP).

medRxiv preprint records are sourced from Cold Spring Harbor Laboratory, BMJ, Yale. We acknowledge the Stephen B. Thacker CDC Library for compiling and providing the medRxiv preprint records prior to September 2021.

Our development partners are Data Language and Metaxis. We also acknowledge Ian Shemilt at the EPPI-Centre for his assistance in developing the machine learning classifier. 

This work was conducted within the CEOsys project (www.covid-evidenz.de), funded under a scheme issued by the Network of University Medicine (Nationales Forschungsnetzwerk der Universitaetsmedizin (NUM)) by the Federal Ministry of Education and Research of Germany (Bundesministerium fuer Bildung und Forschung (BMBF)), grant number 01KX2021.

Notes on Modifications

Study design characteristics and controlled vocabulary descriptors are added to source records to enable advanced search features.

Terms and Conditions of Use / Disclaimer

All persons reproducing, redistributing, or making commercial use of this information are expected to adhere to the terms and conditions asserted by the copyright holder. Transmission or reproduction of protected items beyond that allowed by fair use as defined in the copyright laws requires the written permission of the copyright owners.

Cochrane does not warrant or assume any legal liability or responsibility for the accuracy, completeness, or usefulness of any information, apparatus, product, or process disclosed.

Questions? 

Please direct any questions or comments about Cochrane's COVID-19 Study Register to support@cochrane.org

Page last updated: January 2024