Interim guidance on the inclusion of Clinical Study Reports and other regulatory documents in Cochrane Reviews: a project summary

Article type
Authors
Jefferson T1, Doshi P2, Boutron I3, Golder S4, Heneghan C1, Hodkinson A5, Jones M1, Lefebvre C6, Stewart L7
1Cochrane Acute Respiratory Infections Group
2University of Maryland School of Pharmacy
3Cochrane Bias Methods Group, France
4Cochrane Adverse Effects Methods Group
5CRD
6Lefebvre Associates/ Cochrane Information Retrieval Methods Group
7CRD/Cochrane IPD Meta-analysis Methods Group
Abstract
Background: Publication and other reporting biases may pose serious threats to the validity of systematic reviews. There is growing support for greater research transparency and improved access to regulatory documents including clinical study reports (CSRs) - which can provide more data than are included in published articles. Using these documents has the potential to change and improve the way that Cochrane Reviews of pharmaceutical interventions are done in the future.
Objectives: To increase awareness among Cochrane authors of reporting bias and the potential of regulatory documents to address this; to seek input from the Cochrane community about how important they believe these issues to be, to identify what support Cochrane authors may need to enable them to use these sources of information.
Methods: This project supported a number of activities including: two sessions at Cochrane Colloquium; a summary of research comparing the impact of using regulatory documents in systematic reviews; production of criteria for judging priority for including regulatory data and a glossary to help Cochrane members understand more about regulatory processes and documents associated with this. We also conducted a Cochrane-wide survey to find out how often authors seek information from regulatory documents, and issues related to including them in their reviews.
Results: We discuss 18 studies that shed light on the reporting biases present in publicly available reports of trials of licenced pharmaceuticals. We list key criteria (or triggers) for consideration in the selection of reviews for priority inclusion of regulatory documents (Table 1). Our glossary currently contains over 60 terms related to regulatory documents.
160 respondents completed our survey. The survey found a lack of familiarity with regulatory sources of data and barriers to accessing these, including lack of resources. The main rationale authors gave for seeking regulatory data was to minimise bias.
Conclusions: Understanding within the Cochrane Community is limited and guidance and support is required if authors are to engage with this source of evidence. Guidance on how to use data from regulatory sources is specifically needed.
Patient or consumer involvement: For this stage of our project, we did not involve patients or healthcare consumers. We consider Cochrane authors to be our consumers.