Risk of bias assessment

In order to analyze the bias in each RCT, we used the revised Cochrane risk-of-bias tool for randomized trials (RoB2) (Sterne et al., 2019). It contains bias across the following domains: the randomization process, deviations from intended interventions, missing outcome data, measurement of outcomes, and selective reporting. Some signal problems are set up in each domain. Considering all the information comprehensively, RoB2 divided the overall bias risk into three grades: “high,” “some concerns,” and “low.” Two independent evaluators (LK Yue and L Xiao) conducted the entire bias risk assessment process. When their opinions were inconsistent, the consensus was reached or solved by a third researcher (GW Li).

The evaluation of the certainty of the evidence obtained from this NMA was conducted utilizing the Confidence in Network Meta-Analysis (CINeMA) application, providing the opportunity to assess the levels of confidence (Nikolakopoulou et al., 2020).