Publication Details

Using integrated data to examine characteristics related to pedestrian injuries

Type: Paper

Subtype: Final report

Author(s): Harmon, Katherine J.; Sandt, Laura S.; Hancock, Kari; Rodgman, Eric A.; Thomas, Libby J.

Publisher: Collaborative Sciences Center for Road Safety

Url: https://www.roadsafety.unc.edu/research/projects/2019r22/

Publication Date: Sept-2021

Number: CSCRS-R22

Address: Chapel Hill, NC

Abstract: This study seeks to provide a comprehensive examination of pedestrian motor vehicle crash injuries in North Carolina (NC) using linked crash and emergency department visit data for the period October 1, 2010 - September 30, 2015. Unlike the individual data sources, linked crash-ED visit data provide detailed information on the crash circumstances and the pedestrian health outcomes. Approximately, 50% of the crash data were linked to ED visit data using hierarchical deterministic methods for a study population of 6,923 injured pedestrians. This study used categorical analytic techniques (bivariate and multivariate logistic regression analysis) to examine person, crash, environment, roadway, and vehicle characteristics associated with pedestrian injury severity. This study found that pedestrian age, pedestrian gender, pedestrian race/Hispanic ethnicity, pedestrian comorbidities, striking driver age, striking driver gender, crash hour-of-day, pedestrian/driver suspected alcohol use, ambient light levels, pedestrian crash type, intersection-relatedness, and vehicle type were related to pedestrian injury severity, among other factors. In addition, linking crash and ED data facilitated a greater understanding of the types of injuries (e.g., traumatic brain injuries) associated with being struck by a motor vehicle.