Social support and relapse tendency in Chinese women experiencing heroin addiction: a moderated mediation model of active coping strategies and openness to experience
Description
The present study The evidence reported above supports the notion that social support both directly affects the relapse tendency of women experiencing heroin addiction, but also may have an indirect impact through the active coping strategies. In addition, the role of social support may also be moderated by openness to experience. Further, previous studies of heroin addiction found that the age of addict and their parents' level of education level were significantly correlated with their rates of heroin abuse (Aggarwal et al., 2015; Kolodny et al., 2015). Therefore, it is necessary to control for these factors in the present study. Based on the above analysis, this study proposes the hypotheses listed here and summarized in the model in Fig. 1. (1) Active coping strategies would mediate the relationship between social support and the relapse tendency. (2) Openness to experience would moderate the relationships between social support and active coping strategies, and between social support and the relapse tendency. (3) Openness to experience would moderate the mediating effect of active coping strategies in the relationship between social support and the relapse tendency. Statistical analyses Descriptive analyses and Pearson’s correlations were used by SPSS 22.0 for all variables. To test the moderated mediation model, we have adopted Stride's advice, and constructs are measured by latent variables as opposed to observed variables (Stride, Gardner, Catley, & Thomas, 2015). The analysis process of the entire model corresponds to the SPSS macro PROCESS (http://www.afhayes.com) suggested by Hayes (2018) but applying Mplus 7.4. The mediating (indirect) effect with 5000 bootstrap samples. In order to better reveal the relationship between latent variables, we used the item parceling strategy (Hall, Snell, & Foust, 1999; Little, Cunningham, Shahar, & Widaman, 2002). The critical value of the statistical test includes p value under the standard 0.05 level, and 95% bias-correction bootstrap confidence interval.
Files
Steps to reproduce
The current study used a convenience sampling method, and subjects (in drug treatment) were recruited through the archives of a Compulsory Drug Treatment Center in southwestern China. Our criteria for participation in this survey were that subjects must be (1) women; (2) only heroin users (not users of other similar drugs); and (3) voluntary participants in the study. Measures Social support Participants’ level of social support was assessed using the Chinese version of the Perceived Social Support Scale (K. Zhou et al., 2015). The scale assesses the social support felt by participants from family, friends, and other sources. Participants were asked to answer 12 items on scales ranging from 1 (very strongly disagree) to 7 (very strongly agree). Higher scores indicated higher levels of social support. The scale used have good reliability (Cronbach's α = .88) supported by previous research (D. Zhang et al., 2018), and the observed reliability for this scale was satisfactory (Cronbach's α = .86). Active coping strategies Participants’ active coping strategies were assessed with the Chinese version of the Simply Coping Style Scale (Xie, 1998). Participants were asked to answer 12 items on a scale ranging from 0 (Not used) to 3 (Often used). Higher scores indicated higher probabilities that participants will engage in that active coping strategy. The scale used have good reliability (Cronbach's α = .83) supported by previous research (Guo, Fu, Xing, Qu, & Wang, 2017), and the observed reliability for this scale was satisfactory (Cronbach's α = .81). Openness to experience Participants' openness to experience was assessed using the Chinese version of the Big Five Personality Questionnaire Short (Luo & Dai, 2018). The openness to experience subscale is a part of the Big Five Personality Questionnaire. The openness subscale consists of four items, responded to using a six-point bipolar scale (e.g., What kind of person am I? Conservative-Open, 1= Fully agree (Conservative), 6= Fully agree (Open)). The higher the total score, the more obvious the individual's openness. The scale used have good reliability (Cronbach's α = .84) supported by previous research (Luo & Dai, 2018), the observed reliability for this scale was satisfactory (Cronbach's α = .72). Relapse tendency Participants' tendency to relapse was assessed using the Chinese version of the Relapse Tendency Questionnaire, prepared by Yan Wenxiu (R. Wang, 2001). Participants were asked to respond to 11 items and on a scale ranging from 0 (Fully possible) to 5 (Completely impossible). Higher scores indicated higher tendencies to relapse into drug use. The scale used have good reliability (Cronbach's α = .86) supported by previous research (He, Kuang, Feng, Huang, & Ma, 2019), the observed reliability for this scale was satisfactory (Cronbach's α = .90).