Digital self-harm among adolescents in China: Latent profiles, correlates, and its relationship with suicidal behaviors

To explore the latent profiles of digital self-harm among Chinese adolescents and to examine between-profile differences in sample characteristics and psychological variables, as well as their associations with suicidal behaviors.

A total of 5655 adolescents (mean age = 16.26 ± 0.82 years; 53.2% male) were recruited from 27 April to 19 May 2025. All participants completed self-report web-based questionnaires covering demographic characteristics, lifestyle variables, depressive symptoms, cyber-victimization, digital self-harm, physical self-harm, and suicidal behaviors. Latent profile analysis and logistic regression analysis were used to analyze the data.

Three latent profiles of digital self-harm in adolescents were identified: high digital self-harm (7.0%), inner self-emotion harm (11.2%), and low digital self-harm (81.8%). Multivariate logistic regression indicated that sex, grade, parental marital status, history of mental disorders, and smoking status distinguished the digital self-harm subtypes. Short sleep duration, depressive symptoms, and cyber-victimization were risk factors for digital self-harm (p < 0.05). After controlling for covariates, both high digital self-harm and inner self-emotional harm were associated with physical self-harm and suicidal behaviors. Notably, inner self-emotional harm demonstrated a unique association with suicidal ideation (OR = 2.00, 95% CI = 1.60-2.51) and suicide attempts (OR = 1.70, 95% CI = 1.10-2.61) even after additionally controlling for physical self-harm.

Healthcare and educational professionals should prioritize early screening for digital self-harm, promptly identify high-risk groups, and provide targeted psychological counselling and support to safeguard adolescent mental health and prevent suicide.