Abstract/Summary In response to the public health threat that suicide poses to American youth in general and incarcerated youth in particular, we propose a multi-year, multi-site, setting-level intervention in the New York City Juvenile Justice system. In partnership with the NYC Administration for Children's Services (ACS), we will experimentally evaluate an evidence-based staff-level training (S4L) to address youth suicidality during two critical points of contact for juvenile-justice involved youth: placement and post-release Aftercare. S4L combines training in Shield of Care (SOC), an evidence-based model for acute suicide detection training for juvenile justice staff, with evidence-based suicide prevention skill-building training informed by dialectical-behavior therapy for adolescents at risk for suicide ideation, behavior, and non-suicidal self-injury (NSSI). We will compare the effectiveness of S4L to a training as usual control condition (control), and to a condition in which we provide monthly on-site coaching to staff to support the acquisition and implementation of S4L skills (S4L+). The intervention will be implemented as part of ACS's usual staff-training procedures. The need for intervention is urgent at this time due to recent state-level policy shifts that will increase the number youth in these sites and heighten their suicide risk. S4L and S4L+ will be evaluated using a 3-arm cluster-randomized design with the full population of non-secure placement (NSP)/Aftercare sites (N=30 sites; N=1800 youth, 30% girls). These sites are sole infrastructure for long-term confinement in NYC. NSP/Aftercare sites will be randomized to control (N=10 sites; 600 youth), S4L (N=10 sites; 600 youth), and S4L+ (N=10 sites; 600 youth) conditions. We test the effectiveness of S4L compared to S4L+ and control on improving suicidal behavior, suicidal ideation, and NSSI; mental health outcomes (internalizing, externalizing and substance use); and correlates of suicide/mental health (impulsivity, mood, and coping). In addition, we test the extent to which site-level implementation characteristics (e.g., proportion of staff trained; average amount of detection and prevention activities); site and staff-level structural characteristics (e.g., prior training, burnout); and individual youth characteristics (e.g., demographics; trauma history) moderate S4L and S4L+ effectiveness. Youth outcome data on suicidal behavior, suicidal ideation, NSSI, and mental health problems and their correlates will be tracked longitudinally for 1 year. In response to the FOA, outcomes will be assessed via evidence-based measures that capture dimensional facets of disorder in line with NIH's common data elements and recommendations by the National Action Alliance for Suicide Prevention. Site-level intervention moderators will be measured by staff survey and administrative data. This project aims to expand the scientific knowledge base on suicide prevention and intervention in high-stakes youth settings and provide empirically-based guidance regarding the feasibility and scalability of staff training and coaching into different juvenile justice systems and jurisdictions. This project leverages an existing collaboration and has the full support of ACS.

Project Narrative Youth between 10 and 24 years old are more likely to die by suicide than by cancer, illness, or hearth disease combined (Heron, 2016). The nearly 2 million youth in the juvenile legal system each year are at even higher risk: they experience a 3- to 5-fold increase in the probability of death by suicide and have higher rates of non- suicidal self-injury and mental health problems. There is a public health imperative to implement and evaluate scaleable suicide detection and prevention interventions that leverage existing legal system infrastructure to respond to the suicide crisis in this vulnerable population.