Bullying Victimization and Non-Suicidal Self-Injury Among Adolescents from Community and Inpatient Mental Health Settings in Ontario, Canada

Background and objective: Bullying victimization is a major global public health problem among adolescents that has been linked to a number of behavioral and emotional problems including anxiety, depression, loneliness, social withdrawal, low self-esteem, poor psychosocial adjustment, as well as non-suicidal self-injury (NSSI). Conservative estimates show that as many as 16-38% of children of school-going age will experience bullying at some point. Some scholars have speculated that individuals with a history of bullying victimization may engage in NSSI as a cry for help, as a form of self-punishment, or to release the stress associated with being victimized. Although bullying victimization has been linked to a number of behavioral and emotional problems among adolescents, few studies have investigated the potential mechanism through which bullying victimization affect NSSI. Therefore, drawing on a large dataset on adolescents referred to community and inpatient mental health settings in Ontario, Canada, and based on vulnerability-stress theory, the objectives of this study were to investigate the effect of bullying victimization on NSSI and the mediating effect of depressive symptoms on the relationship between bullying victimization and NSSI among adolescents.

Methods: This study used data from the interRAI Child and Youth Mental Health dataset. A total of 1,650 adolescents aged 12-18 years (M = 14.56; SD= 1.79; 54.2% males) were analyzed. The dependent variable investigated was NSSI and the independent variable was bullying victimization. Depressive symptoms was the mediator variable. Binary logistic and Poisson regression models were conducted to identify the mediating effect of depressive symptoms on the relationship between bullying victimization and NSSI following the procedure recommended by Baron and Kenny (1986).

Results: Overall, 37% (n = 611) of the 1,650 adolescents engaged in NSSI and 26.7% were victims of bullying. The effect of bullying victimization on NSSI was partially mediated by depressive symptoms after adjusting for the effect of demographic characteristics, history of childhood abuse, social support, and mental health diagnoses. Adolescents who were bullied had 63% higher odds of engaging in NSSI (AOR=1.63 95% CI=1.26-2.11). This significant effect was attenuated upon the inclusion of depressive symptoms in the final model where the adjusted odds ratio went from 1.63 to 1.50 (95% CI=1.16-1.95) and the p-value increased moderately from p < 0.001 to p=0.002. A unit increase in scores on depressive symptoms increased the odds of engaging in NSSI by 5% (95% CI=1.03-1.07). Compared to males, odds were 3.02 times higher for females to engage in NSSI (AOR=3.02, 95% CI=2.37-3.85). Each additional type of childhood abuse experienced increased the odds of NSSI by 11% (AOR=1.11, 95% CI=1.02-1.20). Social support had a protective effect on NSSI such that adolescents who had someone they could turn to for their emotional needs had 26% lower odds of engaging in NSSI, net all other factors included in the model (AOR=0.74, 95% CI=0.55-0.98).

Conclusions and Implications: The finding that depressive symptoms partially mediate the effect of bullying victimization on NSSI offers important clinical implications for clinicians, psychologists, school social workers, and other mental health care professionals towards suicide prevention.