3.2 Risk of bias assessment
The quality of RCT was assessed according to the Cochrane Evaluation Manual, including six aspects: randomization, allocation concealment, blindness, incomplete data bias, selective reporting of results, and other factors that may potentially affect authenticity. If the number of people lost to follow-up exceeded 10% in the included study, the possible causes of loss of follow-up were further analyzed and intention-to-treat (ITT) analysis was performed. For each included study, 3 evaluations were made for the above 6 items, namely, low bias risk, high bias risk and unclear risk (Fig. 2).