An assessment of the quality of current clinical meta-analyses

Abstract

Background: The objective of this study was to assess the overall quality of study-level meta-analyses in high-ranking journals using commonly employed guidelines and standards for systematic reviews and meta-analyses. Methods: 100 randomly selected study-level meta-analyses published in ten highest-ranking clinical journals in 2016-2017 were evaluated by medical librarians against 4 assessments using a scale of 0-100: the Peer Review of Electronic Search Strategies (PRESS), Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA), Institute of Medicine's (IOM) Standards for Systematic Reviews, and quality items from the Cochrane Handbook. Multiple regression was performed to assess meta-analyses characteristics' associated with quality scores. Results: The overall median (interquartile range) scores were: PRESS 62.5(45.8-75.0), PRISMA 92.6(88.9-96.3), IOM 81.3(76.6-85.9), and Cochrane 66.7(50.0-83.3). Involvement of librarians was associated with higher PRESS and IOM scores on multiple regression. Compliance with journal guidelines was associated with higher PRISMA and IOM scores. Conclusion: This study raises concerns regarding the reporting and methodological quality of published MAs in high impact journals Early involvement of information specialists, stipulation of detailed author guidelines, and strict adherence to them may improve quality of published meta-analyses.