Nonsuicidal Self-Injury and Suicidal Behaviors in Girls: The Case for Targeted Prevention in Preadolescence

Abstract

Non-suicidal self-injury (NSSI) affects 15-20% of adolescents-disproportionately girls-and is a strong predictor of eventual suicide attempts and suicide. Many girls now initiate NSSI before age 10. These early-starters exhibit greater frequency of NSSI, use more diverse methods, and are hospitalized more often, yet there are no empirically supported prevention programs for preadolescents. Obstacles to prevention include ascertaining who is sufficiently vulnerable and specifying mechanistic intervention targets. Recent research indicates that (1) preadolescent girls with ADHD who are also maltreated are at alarming risk for NSSI and suicide attempts by adolescence, and (2) the conjoint effects of these vulnerabilities are sufficiently potent for targeted prevention. Research also indicates that existing interventions are effective in altering child- and family-level mechanisms of NSSI. These interventions alter neurobiological markers of vulnerability, which can be used as proximal efficacy signals of prevention response, without waiting for NSSI and suicide attempts to emerge.

Keywords: impulsivity; maltreatment; prevention; self-harm; self-injury; suicidal behavior; suicide.

Figures

Figure 1.

Progression of early life impulsivity for boys (blue semi-circles) versus girls (pink semi-circles) given similar environmental risk exposures. Impulsive girls who incur adversity—particularly physical and/or sexual abuse—are likely to progress to NSSI, SBs, and eventual BPD (see text). In the absence of these adversities, emotion dysregulation is unlikely to develop, and impulsivity is unlikely to progress beyond ADHD. Adapted with permission from Beauchaine (in press).

Figure 2.

Rates of self-injury and suicide attempts among late adolescent/young adult females by risk status. See text for details.

Figure 3.

Conjoint effects of sex, ADHD status, and maltreatment on prospective vulnerability to NSSI and SBs. None of these characteristics—either singly or in pairs—confers sufficient vulnerability for recruitment into prevention programs. In combination, however, they mark considerable vulnerability.