PT - JOURNAL ARTICLE AU - Tracy, Melissa TI - 133 An agent-based model of violence prevention across the life course AID - 10.1136/injuryprev-2022-SAVIR.122 DP - 2022 Mar 01 TA - Injury Prevention PG - A47--A48 VI - 28 IP - Suppl 1 4099 - http://injuryprevention.bmj.com/content/28/Suppl_1/A47.3.short 4100 - http://injuryprevention.bmj.com/content/28/Suppl_1/A47.3.full SO - Inj Prev2022 Mar 01; 28 AB - Levels of fatal and non-fatal assaultive violence have spiked in urban areas over the past year, while the share of violent assaults perpetrated by family members or intimate partners has been steadily increasing over the past three decades. A comprehensive approach to violence prevention calls for a variety of interventions with diverse targets across multiple levels of influence that vary across the lifespan. Complex systems approaches like agent-based modeling can be used to estimate the reductions in fatal and non-fatal violence that would be expected under different intervention scenarios, alone and in combination and considering both benefits and unintended consequences, while incorporating critical interactions between individuals and between individuals and their environment. In this presentation, I will describe the development and initial findings from an agent-based model (ABM) of violence transmission within families, romantic relationships, social networks, and communities at different stages of the life course. This model simulates a sample of the New York City (NYC) population across two generations, with ‘agents’ (representing individual people) embedded in households, romantic partnerships, peer networks, and neighborhoods. Agent dynamics in the ABM reflect the biological, behavioral, and social processes that increase risk for and follow from exposure to interpersonal violence, informed by empiric data from multiple longitudinal studies including the Fragile Families and Child Wellbeing Study and the National Longitudinal Study of Adolescent to Adult Health (Add Health). Interventions to address violence at different stages of the life course that are simulated in the ABM include home-visiting programs in early childhood, school-based violence prevention programs for adolescents, and increased access to behavioral health services for young adults. Results of the model illustrate how agent-based modeling can be used to identify the optimal timing, targeting, and combinations of violence prevention interventions across the lifespan in urban populations.