Shifts in fatal and severe injury pedestrian crashes by census tract across eight states: 2008–2012 vs. 2017–2021

Purpose US pedestrian fatalities increased from 4,302 in 2010 to 6,565 in 2020 (+53%) and have exceeded 7,000 each year since 2021, preventing progress toward Sustainable Development Goals to reduce traffic deaths. This paper gives practitioners new insights into the types of census tracts where fatal and severe injury (FSI) pedestrian crashes have occurred so that they can prevent them in the future. Design/methodology/approach We analyze the number of FSI pedestrian crashes in more than 20,000 census tracts across eight states in 2008–2012 and 2017–2021. We use negative binomial and Poisson-Gamma conditional autoregressive models to find statistically-significant associations between tract characteristics and FSI pedestrian crashes in each study period. We also identify significant shifts between the two study periods. Findings FSI pedestrian crashes were positively associated with certain census tract characteristics in both study periods: larger proportions of Black, Hispanic, or low-income populations; higher unemployment rates; higher proportions of retail jobs; having transit nearby; and containing at least one high-traffic arterial roadway. Between 2008–2012 and 2017–2021, FSI pedestrian crashes became even more prominent in census tracts with higher proportions of Black or Hispanic residents, freeways or expressways, and higher proportions of retail jobs. Originality/value Most pedestrian crash trend studies have summarized data at the national or state level and have relied only on fatality data. Our analysis provides new insights about the pedestrian safety problem by analyzing both fatal and severe injury pedestrian crashes at a neighborhood-level geographic scale, incorporating socioeconomic, land use, and roadway variables.