Project Highlights

The project developed a simulation model to examine adverse childhood experiences (ACEs) and their impact on adolescent suicide. As the project lead, I facilitated collaboration with subject matter experts and external partners in the fields of computer science, economics, systems modeling, adverse childhood experiences (ACEs) and suicide prevention, statistics, information security and 508 compliance. The project involved systems mapping to conceptualize relationships between ACEs and suicide, network analysis of the systems map, data gaps analysis to identify data challenges with operationalizing map components, scoping reviews of existing research, and design of the agent-based model. The project has made substantial contributions to the fields of simulation modeling, ACEs and suicide prevention, and has generated seven published manuscripts to date. The model assists in strategic decision making for choosing suicide prevention strategies. This project also serves as a roadmap for future modeling projects that can assist decision makers in understanding the prevention impacts of other injury and violence prevention strategies.

This project estimated the cost of program implementation for 23 programs funded under CDC’s Core State Injury Prevention Program (Core SIPP). I led the development and design of the study and all scientific and technical deliverables. This project involved developing a user-friendly cost collection tool, pilot testing of the tool, holding feedback discussions on tool enhancements, technical assistance for all 23 programs, and developing a dashboard for visualizing cost results. This project provides leadership with information useful for budgeting and performance monitoring of federally funded injury prevention programs.

This project used advanced statistical methods for measuring the relationship between ACEs and suicide and the role that associated risk and protective factors play in changing the strength or direction of this relationship over time. I managed the data use application process for accessing restricted data from the National Longitudinal Study of Adolescent to Adult Health (Add Health). This project examined multiple measures of economic well-being as potential moderators in the relationship between ACEs and suicidal behaviors. The project expands the evidence base for prevention strategies to reduce suicide ideation, attempt, and fatality.

This project used machine learning methods to identify a set of social ecological factors associated with the risk of violence victimization during adolescence. Machine learning algorithms encompassed a spectrum of linear and non-linear models with varying levels of complexity, including logistic regression, ridge regression, support vector machine (SVM) with both linear and non-linear (radial) kernels, decision trees, random forests, and extreme gradient boosting (XGBoost). The insights gained from this study have significant implications for researchers and policymakers in developing tailored intervention strategies to prevent violence and promote community safety.