3.2. Risk of Bias
The quality of the four studies was assessed as high risk, six studies were assessed as unclear. Participants were selected, and both the baseline and follow-up conditions are well described. Random sequence generation was deemed to be adequate in five of the 10 trials16,18,35-37. Blinding of the outcome assessment is unknown across five trials20,32,33,37,38. One trial33 helped assess the high risk of bias in the selective outcome since the pre-specified primary outcomes were not reported. Three 33,36,37of the studies reported conflicts of interest. The risk of bias assessment is indicated in Figure 2a, b.