Objective: To explore the mediating effect of depressive symptom on the relationship between peer bullying patterns and suicidal ideation, and to provide suggestions for school bullying and suicide prevention for adolescents. Methods: A follow up cohort was established in a junior middle school in Suixi County, Anhui Province in September 2019 (T1). The first follow up was conducted in September 2020 (T2). A total of 1 687 junior high school students participated in this study. Self designed peer bullying questionnaire,center for Epidemiological Studies Depression Scale for children,and questions regarding suicide related behaviors were administered. Results: Prevalence of suicidal ideation, victimization and bullying were 28.3%, 27.0% and 24.4%, respectively. The latent class analysis(LCA) identified three heterogeneous peer bullying classes: low aggressive victims(71.4%), high aggressive victims(3.3%), high verbal and relational aggressive victims( 25.3 %). The low aggressive victims class was used as the reference group, T1 depressive symptom mediated the association between high aggressive victims(mediating effect=0.43, P < 0.05 ), high verbal and relational aggressive victims class (mediating effect= 0.29 , P <0.05) and T2 suicidal ideation. Conclusion Depressive symptom plays a mediating role in the relationship between peer bullying and suicidal ideation. In order to prevent peer bullying and suicide related psychological behaviors of adolescents, educators should pay attention to adolescents mental health when formulating corresponding measures.

Objective: To explore the relationship between different types of bullying behavior and depressive symptoms among adolescents, and to provide scientific basis for further prevention of peer bullying. Methods: Based on the follow up data of 1 687 adolescents from Huaibei City, Anhui Province in September 2019 (T1) and September 2020 (T2), the autoregressive cross lagged analysis was employed to explore the relationship between different types of peer bullying and depressive symptoms. Results: The scores of bullying behaviors (physical bullying, verbal bullying, relational bullying and cyber bullying) and depressive symptoms at T2 were lower than those at T1,and the differences were statistically significant ( t =13.60, 8.61,7.24,3.76,8.29, P <0.01). There was a positive correlation between bullying behavior and depressive symptoms ( P <0.01). The results from cross lagged regression analysis showed that physical, verbal, relational and cyber bullying at T1 could positively predict depressive symptoms at T2 ( β = 0.06 , 0.04, 0.12, 0.05), and physical, verbal, relational and cyber bullying at T1 could positively predict depressive symptoms at T2 ( β =0.07, 0.10, 0.13, 0.10) ( P <0.05). Conclusion There were bidirectional associations between adolescent peer bullying and depressive symptoms.