Many CSRs are obtained through unsealed litigation documents, repositories (e.g. The US Food and Drug and Administration had historically avoided releasing CSRs but launched a pilot programme in 2018 whereby selected portions of CSRs for new drug applications were posted on the agency’s website. Review authors could request CSRs from the European Medicines Agency (Davis and Miller 2017). CSRs are often difficult to access, and are usually not publicly available. CSRs often contain more data about trial methods and results than any other single data source (Mayo-Wilson et al 2018). Because CSRs also incorporate tables and figures, with appendices containing the protocol, statistical analysis plan, sample case report forms, and patient data listings (including narratives of all serious adverse events), they can be thousands of pages in length. To obtain marketing approval of drugs and biologics for a specific indication, pharmaceutical companies submit CSRs and other required materials to regulatory authorities. ) catalogue trials that have been planned or started, and have become an important data source for identifying trials, for comparing published outcomes and results with those planned, and for obtaining efficacy and safety data that are not available elsewhere (Ross et al 2009, Jones et al 2015, Baudard et al 2017).Ĭlinical study reports (CSRs) contain unabridged and comprehensive descriptions of the clinical problem, design, conduct and results of clinical trials, following a structure and content guidance prescribed by the International Conference on Harmonisation (ICH 1995). However, the information presented in conference abstracts is highly variable in reliability, accuracy, and level of detail (Li et al 2017).Įrrata and letters can be important sources of information about studies, including critical weaknesses and retractions, and review authors should examine these if they are identified (see MECIR Box 5.2.a). design, main results, and other results).Ĭonference abstracts are commonly available. Note that a study can be reported in multiple journal articles, each focusing on some aspect of the study (e.g. Journal articles are the source of the majority of data included in systematic reviews. For guidance on searching for and selecting reports of studies, refer to Chapter 4. The relative strengths and weaknesses of each type of source are discussed in Section 5.2.2. 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.
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |