**Table 3.** Risk of bias assessments for each domain for included studies. L: low risk of bias, U: unclear risk of bias, H: high risk of bias.

### *4.1. Selection Bias*

There was a high risk of selection bias for the majority of the included studies. Only one study was at a low risk of selection bias for all three signalling questions used [27]. The generation of an adequate randomisation sequence was reported in only 8 studies [12,13,22,23,25,27,30,34], while 10 studies [1,2,4,8,11,28,29,32,36,38] stated that randomisation took place but failed to report the actual methods used, thus being assigned as "unclear". For the remaining studies, randomisation did not take place, or was not mentioned. Baseline characteristics were identified to be similar between groups for only 10 included studies [1,2,8,12,22,23,27,30,31,38], and only 3 [2,27,38] studies reported adequate methodological details on the use of appropriate and proper allocation concealment.
