The methodological quality of animal studies: A cross-sectional study based on the SYRCLE’s risk of bias tool

Abstract

AbstractObjectiveTo assess the methodological quality of animal studies published in China and abroad using the SYRCLE’s risk of bias tool, and to provide references to improve the methodological quality of animal studies to encourage high quality preclinical studies.MethodsAn electronic search was performed in the Chinese Scientific Citation Database (CSCD) and Web of Science from 2014 to October 2018. Document screening and data extraction were performed independently by four researchers. The methodological quality of the included studies was assessed using the SYRCLE’s risk of bias tool. Statistical analysis was performed using SPSS23.ResultsA total of 2764 animal studies were included. Of the studies, 984 were published in English and 1780 were in Chinese. The citation frequency of more than 90% of the included studies was less than 5. The results of methodological quality assessment showed that 36.36% (8/22) of the sub-items were rated as “low risk” in more than 50% of the included studies, of which 75% (6/8) were rated as “low risk” in more than 80% of the included studies. A total of 59.09% (13/22) of the sub-items were rated as “low risk” in less than 30% of the included studies, of which 92.31% (12/13) were rated as “low risk” in less than 10% of the included studies. The incidence of “low risk” Chinese studies regarding performance bias, detection bias and reporting bias were lower than English studies. For foreign studies, more attention should be paid to selection bias, attrition bias, and reporting bias.ConclusionWe identified limitations in the methodological quality of animal experiment studies published in China and abroad. We therefore suggest that it is necessary to take targeted measures to popularize the SYRCLE’s risk of bias tool to effectively improve the design and implementation of animal experiments, and guide study development.