In a 6-month follow-up study of 119 hospitalized adolescents, Yen and colleagues found that many traditional risk factors including psychiatric diagnoses and past attempts failed to prospectively predict suicidal behavior. Other factors, which the authors called "cross-cutting" (because they cut across many disorders) were more potent.
These findings have direct clinical implications and indirect prevention implications. From a clinical perspective, clinicians must be cautious in applying population-generated risk factors to clinical risk formulation. Clinical training in risk formulation should emphasize dynamic factors over diagnoses and history and involve thoughtful synthesis of a wide range of factors and individual circumstances. From a broader prevention perspective, the study provides additional building blocks in the argument for focusing on cross-cutting constructs such as emotion self-regulation in suicide prevention (see our recent population-based study identifying emotion self-regulation as a critical construct for youth suicide prevention). This emphasis on "cross-cutting" constructs has interesting intersections with NIMH's effort, represented by the Research Domain Criteria (RDoC) to shift research away from DSM diagnostic categories toward dimensional assessment of more fundamental and biologically verifiable constructs. These findings are also congruent with (though they do not directly support) strategies that reach further "upstream" in adolescent development to build core "cross-cutting" protective factors.
Yen, S., Weinstock, L. M., Andover, M. S., Sheets, E. S., Selby, E. A., & Spirito, A. (2012). Prospective predictors of adolescent suicidality: 6-month post-hospitalization follow-up Psychological Medicine. Advance online publication. doi:10.1017/S0033291712001912