Institute of Emergency Medicine, Split, Croatia.
Department of Research in Biomedicine and Health, University of Split School of Medicine, Split, Croatia.
Res Synth Methods. 2022 Jul;13(4):478-488. doi: 10.1002/jrsm.1556. Epub 2022 Mar 22.
Some Cochrane reviews were declared stable or closed, that is, not in need of updating. For some of them, it has been declared that conclusions will not (or it is unlikely they will) change with further studies. We explored whether there is a discernable decision-making pattern for decisions about the conclusiveness and stabilization of these reviews. We analyzed Cochrane reviews published until April 2020 labeled as stable or closed. We extracted the rationale leading to the decision declaring that the conclusion is not expected to change with further studies. Furthermore, we assessed whether the reviews used GRADE analysis. We extracted data from summary of findings (SoF) tables on the direction of effect, statistical significance, and I values for the first and primary outcomes in SoFs, conclusions in the abstract and review, and implications for practice and future research. We included 40 stable/closed Cochrane reviews. Rationales for their stabilization did not enable any insight into the Cochrane's decision-making algorithm for considering the evidence as conclusive. Among 191 outcomes presented in the SoFs, 70% were rated with either low or very low certainty evidence. None of the reviews mentioned in the text that the review should be stabilized or closed, or that there is sufficient evidence on the subject. Reasons for stabilizing/closing Cochrane reviews were unclear, and we could not discern any pattern of "conclusive review" traits. Definition of systematic review conclusiveness is still lacking, which may contribute to research waste.
有些 Cochrane 评价被宣布为稳定或已关闭,也就是说无需更新。对于其中一些评价,已经宣布其结论不会(或不太可能)随着进一步的研究而改变。我们探讨了这些评价的结论性和稳定性决策是否存在可识别的决策模式。我们分析了截至 2020 年 4 月发表的被标记为稳定或已关闭的 Cochrane 评价。我们提取了导致做出进一步研究不会改变结论的决定的理由。此外,我们评估了这些评价是否使用 GRADE 分析。我们从汇总结局数据(SoF)表中提取了方向、统计显著性和 I²值的数据,这些数据来自 SoF 的首项和主要结局、摘要和评价中的结论以及对实践和未来研究的影响。我们纳入了 40 项稳定/已关闭的 Cochrane 评价。其稳定的理由并不能洞察 Cochrane 考虑证据具有结论性的决策算法。在 SoF 中呈现的 191 个结局中,有 70%的结局被评为低或极低确定性证据。在文本中没有提到任何稳定/关闭的 Cochrane 评价,或者在这个主题上有足够的证据。稳定/关闭 Cochrane 评价的理由不明确,我们无法辨别任何“结论性评价”特征的模式。系统评价结论性的定义仍然缺乏,这可能导致研究浪费。