Risk of Bias
Figure 2 presents the results of the risk-of-bias assessment for each study separately, and Figure 3 summarizes the percentages of studies with low, unclear, and high risks of bias. The risk of bias of included trials was high or unclear for most domains (Figure 2). Random sequence generation was adequately described in 5 trials (56%) and unclear in 4 trials (44%). Allocation concealment was assessed as low risk of bias in 3 trials (33%), high risk of bias in 1 trial (11%) and unclear in the remaining 5 trials (56%). Of the 9 trials, three trials (33%) blinded the participants to the trial intervention, high risk of bias in only 1 trail (11%) and unclear in 5 trial (56%). Blinding of outcome assessors occurred in 4 trials (44%), high risk in 1 trial (11%) and unclear in 4 (44%). Six trials (67%) were considered as low risk of attrition bias and 11 trials (33%) were assessed as high risk. All trials (100%) were assessed to have low reporting bias. Eight trials (89%) were assessed as having low risk of other potential bias and unclear in 1 (11%). Overall methodological quality of evidence was considered fair. We were unable to assess publication bias due to the limited number of trials.