Association Between Neighborhood Opportunity and Pediatric Injury Rates

Ashley Hollo, MD MPH
Sivani Manchu, MPH
Carisa Bergner, PhD
Kaitlyn McQuistion, MD MSHS
Katherine Flynn-O’Brien, MD MPH
Michael Levas, MD, MS

Injury is the leading cause of death among US children aged 1-18. Where a child lives influences their risk of injury. The Child Opportunity Index (COI) measures neighborhood resources and has been linked to poor pediatric health outcomes, including risk of violent death. This study examined whether improvements in COI quintile over time were associated with reductions in pediatric injuries.

This retrospective cross-sectional study included children ages 0-18 presenting to emergency departments with injury-related ICD-10 codes between 2018 and 2022 identified through the state hospital association database. Neighborhood resources were quantified using COI. Injury-related ICD-10 codes were defined using the 2025 National Trauma Data Standard (NTDS). Additional injuries excluded from NTDS, including burns, drownings, hangings/asphyxiation, and late effects of injury, were included to better capture pediatric injury burden. Kruskal-Wallis testing evaluated differences in median injury rates among census tracts with increased, decreased, or unchanged COI quintiles from 2018 to 2022. Sub-analyses stratified injuries by mechanism and intent using ICD-10 external cause codes.

Fifty (17%) of 287 census tracts had a change in COI quintile from 2018 to 2022. Of these, 34 (11%) increased and 16 (6%) decreased. During this period, 238 (73%) of 328 census tracts experienced changes in injury rates, with 174 (53%) increasing and 64 (20%) decreasing. Kruskal-Wallis testing demonstrated a significant relationship between change in COI quintile and change in injury rates (p=.011, Figure 1). Post hoc pairwise comparisons showed that census tracts with increasing COI quintiles had higher median injury rates compared to tracts without COI change (p=.0042). When stratified by injury mechanism and intent, significant relationships were observed for “struck by” (p=.007) and “unintentional” injuries (p=.0051). No statistically significant associations were identified for cut/pierce, fall, burn, nature/environmental, overexertion, pedal/cycle, firearm, motor vehicle traffic/transport, pedestrian, or assault injuries. Analysis was limited by the small number of census tracts with COI change.

Census tracts with improvement in COI over time were associated with increased median rates of “struck by” and “unintentional” injuries. No association was observed for other mechanisms or intents. These findings suggest that factors beyond the educational, health, environmental, social, and economic indicators included within COI contribute to pediatric injury risk. Interventions focused solely on COI-related factors may therefore be insufficient. Additional longitudinal data are needed to better understand the relationship between COI and pediatric injury over time.

• Understand the relationship between neighborhood conditions and pediatric injury risk
• Differentiate between place-based social resource indices such as COI, SVI, and ADI
• Recognize limitations of relying solely on place-based social resource measures to address child injury