We established working groups to develop minimum standards based on early proposals and groundwork by many groups and individuals within Cochrane. We agreed the need to identify methodological expectations for Cochrane protocols, reviews and updates of reviews on the effects of interventions that could be implemented across Cochrane. Six Working Groups covered six core methodological aspects of Cochrane Intervention Reviews:
- Developing a question and deciding the scope of the review,
- Searching for studies,
- Selecting studies and collecting data,
- Assessing risk of bias in studies,
- Analysing data and undertaking meta-analyses,
- Interpretation and presenting results.
For each of these areas, we set out to identify the following in respect of intervention reviews:
A. essential minimum standards (must do);
B. desirable standards (should do);
C. common errors (should not do);
D. fatal flaws (must not do) and identification of any important methodological uncertainties.
The existing standards cover A and B, whilst work on C and D is ongoing. At least one methodologist and one Co-ordinating Editor (clinical specialist) jointly led each working group. We sought to ensure that groups reflected divergent views and had access to appropriate expertise. We co-opted other people from across Cochrane as necessary to ensure co-ordination and consistency of approach (training and knowledge translation). From an initial draft set of standards, we consulted widely throughout Cochrane, after which the MECIR co-ordinating author team collated responses to produce this final set of standards. Although primarily drawn from the current (2011) version of the Cochrane Handbook for Systematic Reviews of Interventions (the
Handbook), the standards proposed have also required amendment to the Handbook. Thus, the Handbook will incorporate these standards.