Challenges and opportunities in developing a web-based systematic review data repository

Article type
Authors
Lau J1, Hadar N1, Parkin C1, Kwong M1, Earley A1
1Tufts Medical Center, USA
Abstract
Background: Systematic reviews have become increasingly critical to informing healthcare policy. However, they remain a time consuming and labor-intensive activity. These needs resulted in the Systematic Review Data Repository (SRDR) initiative. With funding from the US Agency for Healthcare Research and Quality, our team has developed a collaborative Web-based tool to extract and archive data from primary studies used in systematic reviews. This tool is an open-access resource intended for use by researchers and consumers of healthcare evidence worldwide.

Objectives: SRDR seeks to: increase the efficiency of conducting and updating systematic reviews by minimizing unnecessary replication; enhance the transparency of systematic reviews by making data accessible to all users; and improve the quality of archived data via collaborative review and commentary. The repository will also serve as a rich data source for methodological research.

Methods: A team of programmers, methodologists, and systematic reviewers as well as an international external advisory committee of stakeholders and potential users has met face-to-face and by teleconferences since 2010 to develop specifications, program, test, plan deployment and discuss issues and challenges.

Results: The current main SRDR features include: flexible and reusable data extraction templates, linkages with external databases such as Medline, and facility to reconcile discrepancies in duplicate extractions. SRDR is scheduled to be launched in June 2012 with access initially limited to select individuals and organizations. A committee of diverse group of international stakeholders will be formed to advise the governance of SRDR.

Conclusions: SRDR will continue to be refined to meet the needs of the community. For SRDR to achieve its objectives, many non-technical issues need to be addressed. These include data ownership, quality assurance, appropriate use of data, governance structure, and long term funding. Meeting these challenges can help SRDR to become a leading tool in conducting systematic reviews.