As discussed in Section 5.2.1, it is important to link together multiple reports of the same study. Studies are reported in a range of sources which are detailed later. Here we describe approaches that should be used in systematic reviews for collecting data, including extraction of data directly from journal articles and other reports of studies. Methods used for these decisions must be transparent they should be chosen to minimize biases and human error. Data collected for systematic reviews should be accurate, complete, and accessible for future updates of the review and for data sharing. Consequently, the findings of a systematic review depend critically on decisions relating to which data from these studies are presented and analysed. Systematic reviews aim to identify all studies that are relevant to their research questions and to synthesize data about the design, risk of bias, and results of those studies. Cochrane Handbook for Systematic Reviews of Interventions version 6.3 (updated February 2022). In: Higgins JPT, Thomas J, Chandler J, Cumpston M, Li T, Page MJ, Welch VA (editors). The key to successful data collection is to construct easy-to-use forms and collect sufficient and unambiguous data that faithfully represent the source in a structured and organized manner.Įffort should be made to identify data needed for meta-analyses, which often need to be calculated or converted from data reported in diverse formats.ĭata should be collected and archived in a form that allows future access and data sharing.Ĭite this chapter as: Li T, Higgins JPT, Deeks JJ (editors). Review authors are encouraged to develop outlines of tables and figures that will appear in the review to facilitate the design of data collection forms. trials registers, regulatory documents, clinical study reports), review authors should decide on which sources may contain the most useful information for the review, and have a plan to resolve discrepancies if information is inconsistent across sources. Systematic reviews have studies, rather than reports, as the unit of interest, and so multiple reports of the same study need to be identified and linked together before or after data extraction.īecause of the increasing availability of data sources (e.g.

Cite engauge digitizer