An international register of ongoing systematic reviews

Article type
Authors
Booth A1, Clarke M2, Ghersi D3, Moher D4, Petticrew M5, Stewart L1
1Centre for Reviews and Dissemination, University of York, York, UK
2UK Cochrane Centre, Oxford, UK
3International Clinical Trials Registry Platform, WHO, Geneva, Switzerland
4Clinical Epidemiology Program, Ottawa Hospital Research Institute, Ottawa, Canada
5Public and Environmental Health Research Unit, London School of Hygiene and Tropical Medicine, London, UK
Abstract
Background: Support is growing for the prospective registration of protocols for systematic reviews. As for clinical trials, registration could be an important means of combating publication and selective outcome reporting biases. Registration could also help avoid unnecessary duplication; encourage collaboration; and create opportunities for new methodological and other research. Until now, there has been no international centralised, comprehensive register of ongoing reviews. Existing registration of protocols has been limited to individual organisations, such as the Cochrane and Campbell Collaborations. However, more than 80% of systematic reviews are generated outside these organizations, and an international register, open to all, has the potential to have a major impact. Objectives: Our intention is to develop, implement and manage an international web-based register of ongoing systematic reviews with health-related outcomes. Progress to date: The Register uses the existing IT platform and infrastructure that supports production of the three publicly accessible databases at the Centre for Reviews and Dissemination (CRD): DARE, NHSEED and HTA. To ensure that the Register is fit for purpose and acceptable to potential contributors and users, a Delphi exercise was undertaken in 2010 to define a minimum dataset. The Register offers free public access, is electronically searchable and open to all prospective registrants. Registration requires provision of the minimum data set. After acceptance, the register entry is loaded as a permanent entry and a unique identification number issued. If a protocol is available for the review, this is loaded as well. Each record will also include an audit trail of amendments and links to resulting publications. Conclusions: As the Register develops, it will contribute to improvements in the quality of systematic reviews and the decisions that rely upon them; supporting the efficient use of funding and timely updating, and providing a way of assessing the risk of bias in systematic reviews.