Martin Bellander

Research Problem and Specific QuestionsCurrently, no reliable methods exist to predict self-harm and suicide attempts among teenagers. Additionally, evidence-based treatments targeting suicidality are lacking. For over a decade, our research group has pioneered interventions, including digital ones, for youth who self-harm. Using digital phenotyping, we aim to improve short-term prediction of self-harm and leverage this information to develop and test a Just-in-Time Adaptive Intervention (JITAI) for adolescents at high risk.Data and Method This project will be conducted within a planned randomized controlled trial (RCT) evaluating a treatment for adolescents who have recently attempted suicide and are at high risk for future self-harm. By collecting data via a smartphone app, we will estimate sleep quality, social contexts, and activity levels through digital phenotyping. This enables statistical models to detect increased risk of self-harm and suicide attempts on shorter timescales. We will also examine geographical proximity to caregivers as a protective factor. This data will be used to develop a JITAI, where participants receive personalized support via smartphones during high-risk moments. The feasibility of JITAI will then be evaluated in a pilot study. Conducting the study within an RCT provides ethical and practical advantages without compromising the trial’s research questions.Societal Relevance and UtilisationSelf-harm is the leading cause of death among youth in Sweden, with each suicide or attempt having profound consequences for individuals, families, and society. Understanding risk factors and the process leading up to self-harm is crucial for timely treatment and reducing preventable deaths. Further, there is no established treatment for adolescents at high suicide risk, making it essential to evaluate interventions that can mitigate this risk.Plan for Project RealisationThe RCT, involving 282 participants, is fully funded, with recruitment starting in Aug 2025. Digital phenotyping data will be collected for six months. Passive data collection in this high-risk population is feasible given that all participants are engaged in treatment. Our JITAI will be based on findings from digital phenotyping and focus groups with patients and clinicians. Study costs include the salary of a postdoc, software for passive data collection, intervention development, and participant reimbursment.