“…As for the flow and timing of subjects, the risk of bias was high in 11 studies (32.4%) because not all subjects were included in the analysis (25,29,31,32,35,39,44,48,51,53,55) and the risk was unclear in 10 studies (29.4%) (21,26,30,33,38,40,43,47,50,52).…”