Login / Signup

What Are the Determinants of the Quality of Systematic Reviews in the International Journals of Occupational Medicine? A Methodological Study Review of Published Literature.

Giuseppe La TorreRemigio BovaRosario Andrea CocchiaraCristina SestiliAnna TagliaferriSimona MaggiacomoCamilla FoschiWilliam ZomparelliMaria Vittoria ManaiDavid ShaholliVanessa India BarlettaLuca MorettiFrancesca VezzaAlice Mannocci
Published in: International journal of environmental research and public health (2023)
Objective: The aim of this study was to evaluate the methodological quality of systematic reviews published in occupational medicine journals from 2014 to 2021. Methods: Papers edited between 2014 and 2021 in the 14 open access journals with the highest impact were assessed for their quality. Studies were included if they were systematic reviews and meta-analyses, and if they were published in English. Results: The study included 335 studies. Among these, 149 were meta-analyses and 186 were systematic reviews. The values of the AMSTAR-2 score range between three and fourteen with a mean value of 9.85 (SD = 2.37). The factors that significantly and directly associate to a higher AMSTAR-2 score were impact factor (p = 0.003), number of consulted research databases (p = 0.011), declaration of PRISMA statement (p = 0.003), year of publication (p < 0.001) and performing a meta-analysis (p < 0.001).The R² values from the multivariate analysis showed that the AMSTAR-2 score could be predicted by the inclusion of these parameters by up to 23%. Conclusions: This study suggests a quality assessment methodology that could help readers in a fast identification of good systematic reviews or meta-analyses. Future studies should analyze more journals without applying language restrictions and consider a wider range of years of publication in order to give a more robust evidence for results.
Keyphrases
  • meta analyses
  • systematic review
  • randomized controlled trial
  • autism spectrum disorder
  • minimally invasive
  • machine learning