“…In the case of three studies (Jiao, Wang, Zhou, Bai, & Li, 2017; Li et al, 2016; Zhang et al, 2018) which examined MHL in both patients and caregivers, data for each sample group contributed to the analyses independently (Lipsey & Wilson, 2001). To avoid bias, for 10 studies (Chen et al, 2011; Chen & Tan, 2010; Chen, Zuo, & Shen, 2008; Ding & Chen, 2012; Huang, 2013; Jin, Zhang, Wang, & Duan, 2006; Liu, 2009; Tan, 2005; Xing & Kang, 2017; Xing, Xi, & Xu, 2002) which investigated participants' MHL before and after an intervention, only the data obtained before the intervention were utilized. Similarly, from two studies (Liu, Xu, Yue, Liu, & Fu, 2005; Ran et al, 2003) which employed randomized controlled trails (RCTs) to test the effectiveness of psychoeducational interventions, only the datasets of the control groups (ie, drug treatment only) were extracted and used.…”