4.2.8 Data extraction

This section of the review protocol should specify the data extraction process and instruments that will be used in the review process, as well as the procedures for solving disagreements between reviewers.  Complete and accurate data extraction is essential for a good quality systematic review. Reviewers should carefully consider all the relevant data that should be extracted for the review given the focus of the review, the review objectives/questions, and the inclusion criteria. Details regarding the publication and the study, the participants, settings, the interventions, the comparators, the outcome measures, study design, statistical analysis and results, and all other relevant data (funding; conflict of interest etc.) should be carefully and accurately extracted from all included studies. In a review assessing effectiveness, thorough extraction of details of the intervention is essential to allow for reproducibility of an intervention that is found to the effective (Munn et al. 2014). In a JBI systematic review data extraction is performed by two or more reviewers, independently, using the standardized data extraction form developed by JBI. Any disagreements about data extraction are solved by consensus or by the decision of a third reviewer. If non-JBI data extraction forms are used these should be briefly described and the justification for their use should be explicitly indicated. The review protocol should specify if authors of studies will be contacted by reviewers in order to clarify existing data, to request missing data or additional data. The review protocol should specify the pre-planned approach for the situations when there are multiple reports (publications) for the same study, and for missing data and for data conversion/transformation.