3.6. Researches quality assessment

The studies included in this meta-analysis are randomized controlled trials, and the literature quality is evaluated according to the evaluation criteria of Cochrane Handbook, including the following 6 items: A. sequence generation; B. allocation concealment, equence generation and allocation concealment are both designed to prevent selection bias, which is caused by different selection conditions, and it may lead to overestimation or underestimation of the effect of intervention measures; C. blinding, in order to prevent the occurrence of implementation bias and measurement bias, the subjects and researchers do not know the grouping of trials, but the designer arranges and controls all trials. D. incomplete outcome data, incomplete result data will lead to follow-up bias, which refers to the exclusion of data due to withdrawal or loss of follow-up in the course of the trial or in the late follow-up period, the relevant data and reasons not stated in the literature or stated but not analyzed when included in the study; E. no selective outcome reporting will lead to reporting bias, which is caused by the existence of selective reporting results in the article; F. other sources of bias refer to other issues that may lead to a high risk of bias. Each bias is assessed according to 3 risk categories in Cochrane Handbook: unclear risk of bias, low risk of bias and high risk of bias.