Examining Traffic Safety across Communities: A National Perspective and Case Studies in Arizona, Maryland, North Carolina, and Oregon
This project identifies factors contributing to inequities in traffic safety and provides practical recommendations to mitigate disparities where they exist and improve traffic safety across communities.
October 2025
Abstract
Introduction
Road safety continues to pose a public health challenge in the United States, with persistent disparities in crash outcomes across diverse demographic and geographic groups. These disparities are influenced by a combination of historical developments in transportation infrastructure, uneven resource allocation, and long-standing policy decisions. In addition, sociodemographic factors such as gender, sex, age, race, and socioeconomic status often interact in complex ways that may affect individuals’ exposure to risk and severity of injury in traffic crashes. Understanding these relationships is critical for identifying at-risk populations and informing equitable strategies that improve safety outcomes for all communities.
This study examines and addresses disparities in traffic safety through three primary objectives. It begins by identifying and analyzing differences in traffic safety outcomes across various sociodemographic and socioeconomic groups. Next a comprehensive quantitative analysis explores the interconnected sociodemographic, infrastructural, and systemic factors that may contribute to these disparities. Finally, the study develops targeted policy recommendations and solutions based on both interview data and quantitative analysis, with the goal of promoting more comprehensive safety outcomes.
Methodology
The study framework consists of three main components: a review of existing literature to establish context, a quantitative analysis to identify and measure disparities, and a qualitative assessment that uses interviews to gain deeper insights and inform policy recommendations. The quantitative and qualitative analyses are conducted at the national level and in four states: Arizona, Maryland, North Carolina, and Oregon.
The literature review focused on studies examining crash disparities across various sociodemographic and socioeconomic groups in the United States, primarily covering the years 2019 to 2024, and was further divided into crashes involving motor vehicles, pedestrians, and bicyclists. The quantitative analysis began by analyzing crash rates nationally and within the four selected states, disaggregated by sex, age group, and race for drivers, pedestrians, and bicyclists. Crashes were mapped to geographic boundaries using an area-based method. To identify crash risks, double-bootstrap data envelopment analysis and machine learning models were applied at the county level nationwide and the census tract level in four selected states. Finally, interviews were conducted with individuals from federal and state agencies, industry, and the academy to explore underlying factors contributing to these disparities and possible policy interventions.
Key Findings
- The study revealed significant variations in traffic safety outcomes across demographic groups and geographic regions. Nationally, male fatality rates were approximately 3.5 times higher than female rates per 100,000 population, with notable state-by-state differences observed across the United States.
- American Indian and Alaska Native (Native American) populations experienced the highest fatality rates among racial groups
- Arizona consistently showed higher pedestrian and bicyclist fatality rates across nearly all demographic categories compared to both national averages and other benchmark states (Oregon, Maryland, and North Carolina).
- Data envelopment analysis results indicated that crash risks and injury severity patterns differ substantially by both demographic composition and geographic location.
- Higher poverty rates were consistently associated with higher crash risks
- Associations between crash risk and proportion of neighborhood by racial characteristics differed across state- and national-level findings.
- In Arizona, higher proportions of White, Black, and Hispanic residents were linked to greater risks.
- In Maryland, risks rose with Black, Asian, Hispanic, and multiracial populations, while White populations experienced safer outcomes.
- In Oregon, risks were higher in areas with more White, Asian, Hispanic, and multiracial residents, but lower in areas with higher Black and Pacific Islander residents.
- Machine learning analysis confirmed that no single demographic group or community maintained consistently higher or lower risk levels across all regions, with injury severity and crash-related costs showing strong regional dependence.
- Qualitative findings highlighted data limitations related to tribal communities and emphasized the importance of community-centered engagement approaches.
- The interview findings also revealed institutional biases toward motorist safety in transportation planning practices, along with the compounded challenges faced by lower-income communities. These populations frequently experience both deficient transportation infrastructure and greater reliance on higher-risk transportation modes, creating a dual disadvantage in traffic safety outcomes.
Suggested Citation
