A Russian version of AMSTAR 2: validation and guidance

Abstract

Relevance. Determining the quality of a systematic review is known to be of the main criteria to consider when making a clinical decision. And various assessment tools allow you to determine this, both for the readers themselves and for those who write a systematic review. Aim. In this study we investigate characteristics of the AMSTAR 2 assessment tool to determine the methodological quality of systematic reviews of intervention research, as well as validated the Russian version of this questionnaire. Materials and methods. We used an adapted Russian version of AMSTAR 2 (our own translation) to conduct a descriptive study of systematic reviews which were made in the Russian language. The search for Russian-language studies was carried out in the electronic scientific library named «eLibrary», by keywords and their combinations, without chronological restrictions. The selection of publications (total 10) was carried out randomly. After that, the methodological quality was assessed independently by three authors. The agreement of expert assessments was measured by calculating the coefficient kappa according to the J. Fleiss’ method. Under the null hypothesis H0 we meant kappa equaled 0, that is, the observed agreement between the reviewers is tantamount to random agreement. When testing statistical hypotheses, p < 0.05 was used. Statistical analysis was carried out in R 4.1.2 (RStudio 1.1.463), using the kappaGUI and irr packages. Results. We have translated the AMSTAR 2 questionnaire (16 signalling questions) into Russian, as well as made terminological and methodological additions to it. The preliminary analysis showed that our Russian-language version of this evaluation tool allows us to properly assess the methodological quality of systematic reviews and meta-analyses, their potential value for consumers of evidence-based information regarding the effectiveness of medical interventions. Conclusion. The AMSTAR 2 methodology allows critical evaluation of systematic reviews not only of randomized controlled clinical trials, but of non-randomized trials. The extended functionality of the questionnaire, in particular its Russian version, makes it possible to unambiguously determine the correctness of the methodology for writing a systematic review and meta-analysis, which will help in finding a high-quality review and control at the writing stage.