Social Functioning and Suicide Risk in a Community Sample of Adolescents.

Suicide is the second leading cause of death among young people. Both mental illness and social factors are associated with suicide in adolescents, and youth with mental disorders often experience social deficits, which may compound risk. The cumulative effects of mental disorders and social factors on suicidal ideation and behaviors (SIB) in adolescents have not previously been explored. Adolescents 13-18 years of age (N = 6,447; 49% female, 65% non-Hispanic White) participated in the National Comorbidity Survey Replication Adolescent Supplement. Adolescents were interviewed to assess mental health diagnoses, history of SIB, and relationship quality. Parents completed self-reports about adolescent mental health and family characteristics. Logistic regression estimated associations of friend, sibling, and family relationships with 12-month SIB and lifetime suicide attempt (SA); associations between relationships, SIB, and SA were compared across classes of mental disorders. Friendship negativity (odds ratio [OR] = 1.20, 95% confidence interval (CI) [1.04, 1.40]), and family conflict (OR = 1.26, 95% CI [1.13, 1.41]), were positively associated with SIB, beyond the risk conferred by mental disorders, particularly mood disorders (OR = 4.75, 95% CI [3.20, 7.05]). Friendship positivity (OR = 0.89, 95% CI [0.80, 0.99]); sibling relationship positivity (OR = 0.79, 95% CI [0.68, 0.91]); family cohesion (OR = 0.77, 95% CI [0.69, 0.87]); maternal care (OR = 0.76, 95% CI [0.69, 0.84]); and paternal care (OR = 0.68, 95% CI [0.59, 0.78]), were inversely associated with SIB. Sibling relationship positivity, family conflict, and paternal care were similarly associated with SA. Self-reported adolescent friend and family relationship characteristics are associated with SIB and SA beyond the risk conferred by mental disorders. This suggests that perceptions of friend and family relationships may be an appropriate target to reduce suicide risk among adolescents.