“…Three papers (Boter, 2004; Burton & Gibbon, 2005; Ostwald et al, 2013) did not explain the method of random sequence generation in detail, two papers (Deyhoul et al, 2020; Mei et al, 2017) did not specify how to implement allocation hiding and three papers (Forster & Young, 1996; Johnston et al, 2009; Tielemans et al, 2015) did not describe these two parts in detail, which indicated the possibility of selection bias in these studies. Fourteen papers (Boter, 2004; Burton & Gibbon, 2005; Deng et al, 2020; Deyhoul et al, 2020; Eames et al, 2013; Forster & Young, 1996; Johnston et al, 2009; Lin et al, 2022; Mei et al, 2017; Minshall et al, 2020; Ostwald et al, 2013; Tielemans et al, 2015; Wang et al, 2015; Zhao, 2021) did not blind the study subjects and/or the intervention provider and had a high risk of implementation bias. One paper (Clark et al, 2003) did not mention whether the study subjects or intervention providers were blinded, indicating possible implementation bias.…”