Interaction Between Previous Attempts and Diagnosed Psychiatric Disorder as a Risk Marker of Repeated Suicide Attempts among Adolescents: Results from a Prospective Hospital-based Study

Abstract

Suicide is the second most frequent cause of death among the youth and its rates among adolescents have recently risen. Up to 30% of adolescents who attempt suicide will try it again within a year. Our objective is to analyze how previous attempts and diagnosed psychiatric disorder behave as markers of risk of reattempts and their statistical interaction. We include every underage patient treated by an emergency room psychiatrist after a suicide attempt in a General Hospital between years 2010 and 2015. Patients free of relapse after 1000 days are censored. We obtain Kaplan–Meier estimates for the risk of a new attempt as a time-dependant variable, dividing them by the presence of previous suicide attempts, diagnosed psychiatric disorder or both at a time, checking the differences by using log-rank tests. Then, we perform Cox proportional risk models including both variables and a factor of their interaction and adjust them by sex and age in a non-automatically driven multivariate analysis, thus obtaining HR estimates. We present 150 cases (118 female; mean[SD] age in years: 15.8 [1.6]). Overall, 22.6% of them relapse during follow-up time. Multivariate models show interaction of previous attempts and diagnosed psychiatric disorder is associated with relapse with an HR of 1.27 × 10⁸ (95% CI: 5.51 × 10⁷ – 2.9 × 10⁸). Interaction of both factors is an outstanding risk marker of relapse after an attempted suicide and should thus be given clinical importance in tertiary prevention.

Disclosure of interest

The authors have not supplied their declaration of competing interest.