Suicide is the second leading cause of death among adolescents. In fact, the national suicide rate has increased 24% over the last 15 years, and hospital visits for child and adolescent suicidality have doubled in the last 10 years. Although considerable research has been devoted to identifying risk factors for suicidal behavior, the research to date has focused on long-term risk, with average follow-up intervals of 10 years. Indeed, there are currently no studies of short-term risk for this behavior in adolescents. Addressing this gap in suicide research is important because long-term risk factors are poor predictors of short-term risk. Therefore, clarifying processes relevant to short-term risk for suicidal behavior is important for advancing our knowledge from who is at risk to when they are most at risk. Neurocognitive function has been identified as having particular potential as indicators of short-term risk. Although core components of neurocognitive function have been associated with suicidality, short-term longitudinal studies of adolescent suicide attempts, are notably lacking. Additionally, although traditional statistical methods have been used to model these neurocognitive processes, recent developments in computational psychiatry involving cognitive process models may more realistically reflect these processes. Furthermore, computational psychiatry allows for simultaneous consideration of multiple variables in a manner that produces a complex prediction model that is often substantially superior to those derived from traditional statistical approaches. Our long-term goal is to improve detection of short-term risk and to identify potential modifiable targets for clinical intervention. Toward this end, the objective of the current application is to evaluate candidate neurocognitive markers of short-term risk for suicide attempts and suicidal events (i.e., over one-month follow-up) in a sample of adolescent psychiatric inpatients, supplementing traditional statistical methods with novel advances in computational psychiatry. Importantly, the primary reason for the absence of studies investigating short-term risk for suicidal behavior in adolescents is that the low base rate for this behavior poses a significant challenge for feasibility. Large samples drawn from high-risk or clinically severe populations are required to overcome this obstacle. We are uniquely positioned to address this challenge by leveraging existing data from our current studies. We will pool our current data with data collected on 200 new patients for a final sample of 414. This application is innovative by: (i) being the first statistically powered study of risk factors for suicide attempts over one month; (ii) featuring a comprehensive neurocognitive battery that includes general and self-harm-specific neurocognitive processes; and (iii) applying novel computational psychiatry approaches to modelling neurocognitive indices of short-term risk. This application addresses the research objectives of Aspirational Goal 1 in the NIMH/National Action Alliance for Suicide Prevention's Prioritized Research Agenda for Suicide Prevention: identifying neuropsychiatric profiles and cognitive dysfunction associated with suicide risk.

Public Health Relevance

The purpose of this application is to clarify the neurocognitive processes underlying short-term risk for suicidal behavior in a high-risk adolescent clinical sample. The aims of this application are clinically significant because there is an absence of studies on short-term risk factors for suicide attempts in adolescents. This work has the potential to identify (i) pertinent risk factors for short-term risk assessments, (ii) potential risk assessment tools for use in clinical settings, and (iii) modifiable targets for intervention.

National Institute of Health (NIH)
National Institute of Mental Health (NIMH)
Research Project (R01)
Project #
Application #
Study Section
Child Psychopathology and Developmental Disabilities Study Section (CPDD)
Program Officer
Murphy, Eric Rousseau
Project Start
Project End
Budget Start
Budget End
Support Year
Fiscal Year
Total Cost
Indirect Cost
Massachusetts General Hospital
United States
Zip Code