Publicly-available data from more than 130 systematic reviews containing more than 13,000 studies – The Systematic Review Data Repository (SRDR)

ID: 

OS10.4

Session: 

Oral session: Innovative solutions to challenges of evidence production (2)

Date: 

Tuesday 22 October 2019 - 14:00 to 15:30

Location: 

All authors in correct order:

Saldanha I1, Smith B1, Ntzani E2, Jap J1, Balk E1, Lau J1
1 Center for Evidence Synthesis in Health, Brown University School of Public Health, USA
2 University of Ioannina, Greece
Presenting author and contact person

Presenting author:

Ian Saldanha

Contact person:

Abstract text
Background: the global systematic review (SR) enterprise involves a great deal of effort and resources. Much of these are spent gathering large amounts of data about included studies from multiple sources that often extend well beyond journal articles, such as conference abstracts, study registries, regulatory documents, clinical study reports, and information gathered from websites or contact with study authors. Making data gathered during SRs publicly available could offer many benefits, including reducing unnecessary duplication of effort, standardizing data, supporting analyses to address secondary research questions, and facilitating methodological research. Funded by the Agency for Healthcare Research and Quality (AHRQ), the Systematic Review Data Repository (SRDR) is a free, web-based, open-source, data-management and archival platform for SRs.

Objectives: to describe:
1) the current extent of usage of SRDR; and
2) the characteristics of all SRs with publicly available data on the SRDR Website.

Methods: we examined all SRs with data made publicly available through SRDR (on the SRDR Published Projects Website) as of 26 February 2019. We extracted information about the characteristics of these SRs. Two investigators extracted and verified data.

Results: SRDR has had 1291 individual accounts belonging to users from 93 countries. Since AHRQ’s launch in 2012, data have been made publicly available in SRDR for 132 SRs, at an average of 20 SRs per year (Figure 1); these 132 SRs include data from 13,334 studies. Most SRs are in clinical fields (127/132; 96%) and most have evaluated interventions (therapeutic or preventive; 101/132; 77%). The most frequent health areas addressed are mental or behavioural disorders (30/132; 23%) and diseases of the eye or ocular adnexa (23/132; 17%). One-sixth of the SRs (22/132; 17%) are Cochrane Reviews, and two-thirds (89/132; 67%) are funded by AHRQ. The 132 SRs each address a median of three research questions (interquartile range: 1–5) and include a median of 69 studies (interquartile range: 17–130).

Conclusions: until we arrive at a future in which the SR and broader research communities are comfortable with the accuracy of automated data extraction, re-use of data extracted by humans has the potential to help reduce redundancy and costs. The 132 SRs with publicly available data through SRDR, and the more than 13,000 studies therein, are freely available to anyone who might be working on similar SRs or updates of SRs or those who want access to the data for decision making, meta-research, or other purposes. While SR data sharing offers many advantages, there are challenges, such as concerns about data completeness and accuracy and intellectual property and proprietary considerations.

Patient or consumer involvement: we did not involve patients or consumers in the conduct of this study, but many SRs described herein are developed with active involvement of these (and other) stakeholders.

Attachments: