RESUMEN
Emerging research has identified sleep disturbance as an important risk factor for predicting self-harming behaviours. However, the temporality of this relationship, particularly in clinical adolescent samples remains poorly understood. This study examines the relationship between sleep disturbance and self-harming behaviours (namely nonsuicidal self-injury and suicide attempts) in clinical adolescents engaging in risk-taking and self-harming behaviours using secondary analyses from a clinical cohort study. Cross-lagged structural equation modelling was used to determine whether baseline sleep disturbance and self-harming behaviours were predictors of each other over a one-year follow-up period in a sample of adolescents (n = 238, 89.5% female) attending and receiving treatment from an outpatient clinic specializing in risk-taking and self-harming behaviours. When controlling for age, sex and depressive symptoms, greater sleep disturbance (p = 0.001) at baseline independently predicted higher numbers of suicide attempts at follow-up. No bidirectional relationship was found when sleep disturbance was modelled with the frequency of nonsuicidal self-injury. This study adds to the growing evidence that sleep disturbance may predict suicidal behaviours. Clinicians should thus regularly assess for sleep disturbances when evaluating suicidal behaviours in high-risk adolescents. Further research and clinical trials should investigate whether sleep-based interventions may be efficacious in reducing the prevalence of suicidal behaviours.