A longitudinal examination of the interpersonal theory of suicide and effects of school-based suicide prevention interventions in a multinational study of adolescents.

ABSTRACT

BACKGROUND: Predictions of two different models for suicide attempts (SA) over 12 months, as differentially impacted by specific school-based suicide prevention interventions, were compared. These were as follows (a) interpersonal theory (IPTS) and (b) a two-pathway model, one path associated with externalizing symptoms and continuum of self-harm behaviors, and the other with internalizing symptoms. METHODS: Self-report questionnaires were completed by 11,110 high school students from ten EU countries enrolled in the Saving and Empowering Young Lives in Europe (SEYLE) study. Baseline measures included perceived burdensomeness, thwarted belongingness from parents and peers, health risk behaviors, self-injury, suicide ideation (SI), and attempts (SA). SI and SA were reassessed at 12-month follow-up. Each model's predictions of SI and SA groups over time (i.e., repeated SA, remitted SA, SA onset, and no SA) were estimated in the no intervention/control group. The superior model was estimated across intervention groups. RESULTS: Interpersonal theory showed better fit compared to the two-pathway model. In partial agreement with IPTS predictions, parental low belongingness but not peer belongingness or burdensomeness predicted greater likelihood of SI. The likelihood of repeated SA versus no SA was higher among adolescents who reported SI, self-injury, risk behaviors, and particularly both SI and self-injury. All three interventions attenuated the combined effect of SI and self-injury. Youth Aware of Mental Health Program (YAM) additionally decreased the effect of risk behaviors on the likelihood of repeated SA. CONCLUSIONS: Interpersonal theory assumptions were partially supported. Perceived interpersonal difficulties with parents were primarily related with SI, and risk behaviors and self-injury were important predictors of SA. Suicide prevention interventions may be effective by mitigating the hazardous effect of varying self-harm behaviors and may be further advanced by increasing parental involvement.