From Signals to Solutions: An Integrated Approach to Early Suicide Cluster Response

Suicide clusters arise when multiple deaths occur within specific geographic areas and time periods at rates significantly above expected levels. Current estimates indicate that approximately 10-15% of suicide deaths in the United States occur within clusters, thus a research program dedicated to suicide cluster prevention has the potential to significantly reduce the burden of suicide in the US, especially among young people. Our work has demonstrated that clusters are becoming more frequent and concentrated, creating ripple effects through communities that amplify beyond individual losses. Current surveillance systems typically detect clusters long after multiple deaths have occurred, missing critical opportunities for intervention that could save lives. We aim to create an integrated early warning and response system that shifts suicide cluster prevention from reactive documentation to proactive intervention. Our three aims establish the scientific and practical infrastructure needed to implement suicide cluster detection and prevention. Aim 1 develops an early warning system that leverages and integrates suicide attempt and death data along with over 70 contextual-level predictors and forecasts cluster emergence. By integrating contextual predictors spanning demographic, economic, healthcare access, and other social determinants with advanced statistical, machine learning, and mathematical contagion models, we will evaluate whether data patterns can accurately predict cluster emergence with sufficient lead time to support prevention. In Aim 2, we will apply machine learning classification to the suicide death data in the National Violent Death Registry to identify distinct suicide cluster typologies and their key defining features. By applying natural language processing to narratives and incorporating broader contextual factors, we will differentiate media-triggered contagion requiring rapid communication responses, structural vulnerabilities demanding long-term resource investment, method-specific clusters necessitating lethal means restriction, and seasonal patterns requiring precisely timed prevention efforts. This mechanistic understanding enables deployment of appropriate, evidence-based responses. Aim 3 carries out systematic stakeholder engagement to translate technical capabilities into actionable prevention responses. Through structured workshops with approximately 20 stakeholders, we will identify the information different decision-makers need, including the required formats, timescales, and organizational constraints, to coordinate effective responses. This includes developing differentiated communication products that balance urgency with accuracy while enabling collaboration and generating concrete specifications for how cluster intelligence can flow from detection through decision-making to action. The synergy across aims creates a comprehensive prevention framework: early warning signals from Aim 1 trigger appropriate interventions based on mechanistic classification from Aim 2, delivered through stakeholder-validated channels from Aim 3. Our team brings decades of experience in suicide cluster analysis, spatiotemporal modeling, and community engagement, including foundational work demonstrating cluster dynamics at population scale. By building systems that enable rapid detection of emerging clusters, clear understanding of their underlying mechanisms, and effective implementation of evidence-based interventions by those empowered to act, this research builds the infrastructure needed to transform how communities respond to suicide clusters, ultimately preventing contagion-driven harm and saving lives.