Preparing a Cochrane review is complex and involves many judgements. In order to minimize the potential for bias in the review process, these judgements should be made in ways that do not depend on the findings of the studies included in the review. Review authors’ prior knowledge of the results of a potentially eligible study may, for example, influence the definition of a systematic review question, the subsequent criteria for study eligibility, the choice of intervention comparisons to analyse, or the outcomes to be reported in the review. Since Cochrane reviews are by their nature retrospective (one exception being prospective meta-analyses, as described in Chapter 19), it is important that the methods to be used should be established and documented in advance. Publication of a protocol for a review prior to knowledge of the available studies reduces the impact of review authors’ biases, promotes transparency of methods and processes, reduces the potential for duplication, and allows peer review of the planned methods (Light 1984).
While the intention should be that a review will adhere to the published protocol, changes in a review protocol are sometimes necessary. This is similarly the case for a protocol for a randomized trial, which must sometimes be changed to adapt to unanticipated circumstances such as problems with participant recruitment, data collection or unexpected event rates. While every effort should be made to adhere to a predetermined protocol, this is not always possible or appropriate. It is important, however, that changes in the protocol should not be made on the basis of how they affect the outcome of the research study. Post hoc decisions made when the impact on the results of the research is known, such as excluding selected studies from a systematic review, are highly susceptible to bias and should be avoided.
Protocols for Cochrane reviews are published before the completed systematic review in the Cochrane Database of Systematic Reviews (CDSR). Changes in the protocol should be documented and reported in the ‘Differences between protocol and review’ section of the completed review, and sensitivity analyses (see Chapter 9, Section 9.7) exploring the impact of deviations from the protocol should be undertaken when possible.