A Study On The Association Of Childhood Maltreatment And Adolescent Non-suicidal Self-injury

Objective: Non-suicidal self-injury(NSSI)is a major public health concern worldwide.NSSI has attracted attention because it is a very important predictor of future suicidal behaviors(especially suicide attempts).However,the mechanisms that confer increased risk for NSSI remain largely unknown.This study aimed to use longitudinal study to:(1)identify unique mediation pathways from childhood maltreatment(CM)to NSSI using depressive,stress,and anxiety symptoms as individual mediators in multiple mediation analysis,(2)determine if different forms of maltreatment show unique pathways to NSSI and if any sex differences exist in the mediation pathways,(3)determine if the indirect association between CM and NSSI through depressive,anxiety or stress symptoms is moderated by whether the participants met the WHO physical activity(PA)recommendations and whether any sex differences exist,and(4)examine the moderating role of bullying victimization in the indirect association between CM and NSSI through depressive,anxiety or stress symptoms and any sex differencesMethods: A total of 848 middle school adolescents from Hunan province,China were followed up for 6 months.At baseline,the participants filled out questionnaires regarding demographic information and childhood maltreatment(CTQ-SF).At the 6 months follow up,participants filled out questionnaires regarding NSSI history,methods,and frequency in the past 6 months(self-compiled based on SITBI);depressive,anxiety and stress symptoms(DASS-21);Physical activity(YRBS);bullying victimization(self-compiled)and witnessing domestic violence(self-compiled).WHO recommends at least 60 minutes of moderate to vigorous physical activity every day of the week for children and adolescents aged 5 to 17;therefore,the physical activity variable was whether adolescents meet the physical activity recommendations or not,with physical activity for at least 60 minutes 0-6 days per week denoting not meeting the recommendations,and 7 days per week denoting meeting the recommendations.Chi square test was use for group comparisons for categorical variables and independent t test was used for continuous variables.Two-way anova was used to examine sex*maltreatment interaction effects on clinical characteristics.To define the relationship between CM and NSSI and identify unique pathways from CM and NSSI,we tested depressive,anxiety and stress symptoms simultaneously for mediating effects using a path analysis in Mplus.Further analysis was performed to identify specifically which type of CM had significant indirect association with NSSI through depressive,anxiety or stress symptoms,and sex differences were also examined.Further,we conducted moderated mediation analysis using Hayes’ s PROCESS macro to determine whether the indirect paths significant in the mediation stage were moderated by PA and bullying victimization,and if Sex differences existed.Results:(1)Multiple mediation analysis results showed that anxiety symptoms(β = 0.079,95% CI:0.034,0.136,p=0.002)and stress symptoms(β = 0.06,95% CI: 0.008,0.115,p=0.028)independently mediated the relationship between CM and NSSI when all mediator variables were included in the model,accounting for 36% and 28% of the total effects on NSSI respectively.However,depressive symptoms were not found to be a significant mediator after the effects of anxiety and stress symptoms were accounted for.(2)Further mediation analysis revealed that only emotional abuse(EA)had a significant indirect effect on NSSI through anxiety symptoms(β = 0.168,CI:0.117,0.229,p<0.001)and stress symptoms(β= 0.172,CI:0.122,0.228,p=0.002).These mediation pathways were significant in both male and female adolescents.Additionally,emotional neglect also had significant indirect effects on NSSI through stress symptoms among female adolescents(β = 0.066,CI:0.005,0.137,p<0.05).(3)Moderated mediation results showed that whether the WHO PA recommendations were met or not moderated the indirect effect of EA on NSSI through anxiety symptoms(EA →anxiety symptoms: β =-0.226,p=0.021;anxiety symptoms →NSSI: β =-0.030,p=0.046).However,the moderation effect by PA on the direct effect of EA on NSSI was not significant(EA→NSSI: β = 0.008,p=0.675).Analysis by sex revealed that among males,PA moderated the indirect effect of EA on NSSI through anxiety symptoms,(EA →anxiety symptoms: β =-0.419,p=0.001;anxiety symptoms →NSSI: β =-0.061,p<0.001),but the direct effect of EA on NSSI was not moderated by PA(EA →NSSI: β =-0.016,p=0.456).However,PA did not moderate the indirect effect of EA on NSSI through anxiety symptoms among females(EA →anxiety symptoms: β = 0.043,p=0.804;anxiety symptoms →NSSI: β = 0.037,p=0.228;EA →NSSI: β = 0.019,p=0.627).Results also showed that PA moderated the indirect effect of EA on NSSI through stress symptoms,with results similar to those of the anxiety symptoms model.The indirect association between EN and NSSI through stress symptoms significant among females was not moderated by PA(EN →stress symptoms: β =0.050,p=0.649;stress symptoms →NSSI: β = 0.029,p=0.333;EN→NSSI: β =-0.005,p=0.848).(4)Finally,moderated mediation results showed that bullying victimization moderated the indirect effect of EA on NSSI through anxiety symptoms(EA →anxiety symptoms: β = 0.211,p=0.009).The direct association between EA and NSSI was also moderated by bullying victimization(EA →NSSI: β = 0.048,p=0.003).Analysis by sex showed that among male participants,bullying victimization only moderated the indirect effect of EA on NSSI through anxiety symptoms(EA →anxiety symptoms: β = 0.385,p=0.001)while among female participants,bullying victimization only moderated the direct association between EA and NSSI(EA →NSSI: β = 0.057,p=0.018).Bullying victimization also moderated the indirect effect of EA on NSSI through stress symptoms,with results similar to those of the anxiety symptoms model.Regarding the indirect association between EN and NSSI through stress symptoms significant among females,bullying victimization only moderated the direct effect of this model(EN →NSSI: β = 0.039,p=0.031).Conclusions: This study found that NSSI is common among adolescents(15.6%)and that childhood maltreatment plays a significant role in the occurrence of NSSI;we found that childhood maltreatment(emotional abuse in particular)influences NSSI through symptoms of stress or anxiety among both male and female adolescents.These findings highlight the importance of simultaneously paying attention to childhood maltreatment(especially emotional abuse)and symptoms of anxiety or stress.This study also found that physical activity can weaken the impact of childhood maltreatment on NSSI,emphasizing the protective effect of sufficient moderate to vigorous physical activity on negative emotions and NSSI among adolescents,especially males.The findings suggest that sufficient daily physical activity can be used as an emotion regulation strategy in the individualized treatment of adolescent NSSI.The study also found that being bullied amplifies the influence of childhood maltreatment on NSSI,suggesting that prevention of school bullying victimization might reduce the impact of childhood maltreatment on NSSI;therefore,we need to pay more attention to and actively implement anti-bullying measures.These findings provide an important basis for schools and clinicians for the prevention and intervention of NSSI and have important reference significance.