Department of Prevention and Evaluation, Leibniz Institute for Prevention Research and Epidemiology - BIPS GmbH, Bremen, Germany.
Brandenburg Medical School Theodor Fontane (MHB), Center for Health Services Research (ZVF-BB), Brandenburg an der Havel, Germany.
BMC Med Res Methodol. 2023 Mar 16;23(1):63. doi: 10.1186/s12874-023-01879-8.
'A Measurement Tool to Assess Systematic Reviews, version 2' (AMSTAR 2) is a validated 16-item scale designed to appraise systematic reviews (SRs) of healthcare interventions and to rate the overall confidence in their results. This commentary aims to describe the challenges with rating of the individual items and the application of AMSTAR 2 from the user perspective.
A group of six experienced users (methodologists working in different clinical fields for at least 10 years) identified and discussed the challenges in rating of each item and the general use of AMSTAR 2 to appraise SRs. A group discussion was used to develop recommendations on how users could deal with the identified challenges. We identified various challenges with the content of items 2-16 and with the derivation of the overall confidence ratings on AMSTAR 2. These challenges include the need (1) to provide additional definitions (e.g., what constitutes major deviations from SR protocol on item 2), (2) to choose a rating strategy for multiple conditions on single items (e.g., how to rate item 5 if studies were selected in duplicate, but consensus between two authors was not reported), and (3) to determine rules for deriving the confidence ratings (e.g., what items are critical for such ratings). Based on these challenges we formulated specific recommendations for items 2-16 that AMSTAR 2 users could consider before applying the tool. Our commentary adds to the existing literature by providing the first in-depth examination of the AMSTAR 2 tool from the user perspective. The identified challenges could be addressed by additional decision rules including definitions for ambiguous items and guidance for rating of complex items and derivation of confidence ratings. We recommend that a team consensus regarding such decision rules is required before appraisal procedure begins.
Not applicable.
“系统评价的评估工具，第 2 版”（AMSTAR 2）是一个经过验证的 16 项量表，旨在评估医疗干预措施的系统评价（SRs），并对其结果的总体可信度进行评分。本评论旨在从用户的角度描述评价个别项目和应用 AMSTAR 2 的挑战。
六名经验丰富的用户（在不同临床领域工作至少 10 年的方法学家）确定并讨论了评价每个项目和使用 AMSTAR 2 评估 SRs 的一般用途的挑战。小组讨论用于制定关于用户如何处理已确定的挑战的建议。我们确定了项目 2-16 的内容以及 AMSTAR 2 总体可信度评分方面的各种挑战。这些挑战包括需要（1）提供额外的定义（例如，项目 2 中构成对 SR 协议的主要偏离的内容），（2）为单个项目的多种情况选择评分策略（例如，如果研究重复选择，但未报告两位作者之间的共识，如何对项目 5 进行评分），以及（3）确定得出置信度评分的规则（例如，对于此类评分哪些项目是关键）。基于这些挑战，我们为 AMSTAR 2 用户在应用该工具之前可能考虑的项目 2-16 制定了具体建议。本评论通过从用户角度对 AMSTAR 2 工具进行首次深入检查，补充了现有文献。可以通过额外的决策规则来解决这些挑战，包括对模糊项目的定义和对复杂项目的评分以及置信度评分的指导。我们建议在评估程序开始之前，需要就此类决策规则达成团队共识。
不适用。
