Predicting Suicidal Thoughts and Behaviors by Tracking Digital Experiences in Adolescents who were Depressed as Preschoolers

Child suicide rates are at a 30 year high in the US with death by suicide in children ages 10-14 more than doubling between 2007 and 2014, making suicide the third leading cause of death in children 14 and younger.  To begin to address this problem, we are conducting an NIMH funded study of suicidal thoughts and behaviors (STB) in a sample of children many of whom expressed STB as preschoolers who are now being followed into school age and adolescence.  We propose a two-site collaboration between M-PIs Luby and Barch (Washington University in St. Louis_ and Allen (University of Oregon) to add the collection of mobile sensing data at the 2nd follow-up point.  This will occur when study subjects will range in age from 12-15 when we expect > 90% will have cell phones.  The availability of these data will allow us to capture patterns of naturalistic behavior, including spontaneous interactions that adolescents have with peers. It will also allow us to deliver questions at contextually specific times to capture emotional states that are sometimes fleeting but may be highly specific markers of risk for STB.  Importantly, this information would be collected in the context of a rich longitudinal dataset with much relevant data, including numerous measures of STB, psychopathology, gender and sexual orientation and other key emotional and behavioral constructs thought to be key to risk for STB. We hypothesize that peer acceptance/rejection as well as contagion effects, evident through daily digital interactions will interact with other child characteristics to predict STB. The data set is also enriched by several prior and concurrent waves of neural data collected using Event Related Potentials (ERP) during reward and peer acceptance/rejection tasks. Therefore, the proposed AFSP grant would allow the rare opportunity to add on this critical and novel data to maximize the informativeness of the mobile sensing data in this unique sample followed since the preschool period.