Data-driven insights into racial disparities in violent deaths in the United States: predictive models for risk assessment and business solutions

Racial disparities in violent death rates across the United States present a critical challenge for organizations involved in healthcare, insurance, and public safety. These disparities, shaped by demographic factors, mental health, substance abuse, and geographic variability, complicate efforts to develop effective risk assessments and targeted interventions. This study leverages data from the National Violent Death Reporting System to analyze racial differences in violent mortality rates, focusing on suicides, homicides, and other violent deaths (2017-2021). By employing logistic regression models, the analysis assesses the effects of race, age, sex, mental health, substance use, and state-level variability on violent deaths, delivering valuable predictive insights. The study is guided by the Social Determinants of Health framework and structured using the Design Science Framework to ensure methodological rigor. Several machine learning models are used for comparison to evaluate the trade-off between predictive performance and interpretability. Results underscore the importance of demographic and mental health factors, along with geographic disparities, in shaping violent death outcomes. The study advances understanding of how violent mortality varies across racial and geographic groups, supporting the development of targeted risk profiles. These insights can inform data-driven policies and interventions, enabling organizations to more effectively address the needs of high-risk populations.