267 - Association of Urban Neighborhood Built Environments with Firearm Injuries to Children

Poster Number: 267
Publication Number: 267.1423

Background: Firearm injuries, now the leading cause of death among children, disparately impact areas with greater socioeconomic disadvantage. Less is known about how child firearm injuries are influenced by neighborhood built environments, which may be more readily modifiable than socioeconomic factors.

Objective: To examine census tract-level associations of built environment characteristics with incidences of firearm injuries to children in urban areas in the US.

Design/Methods: We conducted a cross-sectional study examining large central metropolitan areas as classified by the National Center for Health Statistics. Data from the Gun Violence Archive on firearm injuries between 2019 and 2022 to victims age under 18 years were geocoded to census tracts. Built environment characteristics were derived from Google Street View images retrieved in 2019 using computer vision techniques. Census tracts were categorized into tertiles by prevalence of each of the following characteristics: green streets; dilapidated buildings and chain link fences (markers of physical disorder); crosswalks (walkability); streetlights, multiple cars, and single lane roads (urban development level). Mixed effect negative binomial regression, with random intercepts for metropolitan area and logged census tract population of children as the offset, was used to estimate adjusted incidence rate ratios (aIRRs) with 95% confidence intervals (CIs) of combined fatal and nonfatal firearm injuries, adjusted for census tract population density and socioeconomic measures and state firearm laws.

Results: We examined 9,985 firearm injuries to children, 2,315 fatal, in 21,111 census tracts across 51 metropolitan areas. Built environment characteristics of the census tracts are summarized in Table 1. Green streets were associated with a 45% lower rate of firearm injuries (aIRR 0.55, 95% CI 0.48–0.62 for highest tertile of prevalence relative to lowest). Conversely, chain link fences (aIRR 2.11, 95% CI 1.88–2.37) and crosswalks (aIRR 1.86, 95% CI 1.62–2.12) were associated with approximately double the injury rates. Markers of urban development had mixed associations, streetlights with a higher injury rate (aIRR 1.22, 95% CI 1.09–1.38) and multiple cars with a lower injury rate (aIRR 0.76, 95% CI 0.66–0.88) (Table 2).

Conclusion(s): Built environment characteristics of urban neighborhoods, including lower street greenness and greater physical disorder, are associated with increased firearm injuries to children. Place-based interventions to address disparities in neighborhood conditions may represent a potential structural approach to child injury prevention.