Most systematic reviews reporting adherence to AMSTAR 2 had critically low methodological quality: a cross-sectional meta-research study.

To analyze the methodological quality and characteristics of systematic reviews (SRs) that reported they were conducted in line with the AMSTAR 2 (A MeaSurement Tool to Assess Systematic Reviews). This was a cross-sectional meta-research study. We searched MEDLINE and Embase. We included full reports of SRs reporting the study was conducted, prepared, or designed in line with the AMSTAR 2. Eligible SRs were those published from January 1, 2018, until May 3, 2022. We assessed the methodological quality of the included SRs using AMSTAR 2. We included a total of 45 records. There were 43 SRs and 2 SR protocols. Among them, most were SRs of interventions that included primary studies on humans. More than half had a meta-analysis. According to our overall AMSTAR 2 assessments of included SRs, 35 SRs were of critically low confidence, 7 SRs were of low confidence, and one SR was of high confidence. There were no SRs of moderate confidence. Even when authors indicate in their manuscripts that the SR was conducted/prepared/designed in line with the AMSTAR 2, it does not necessarily imply it is of high or even moderate confidence according to AMSTAR 2. A self-assessment with AMSTAR 2 could be required for submission and carefully checked by the editors/peer reviewers.

Informations de copyright

Copyright © 2023 Elsevier Inc. All rights reserved.

Déclaration de conflit d'intérêts

Declaration of competing interest The authors have no competing interests to declare.