Citation

Dai D, Taquechel E, Steward J, Strasser S. West. J. Emerg. Med. 2010; 11(3): 294-301.

Affiliation

Georgia State University, Department of Geosciences, Atlanta, GA.

Copyright

(Copyright © 2010, California Chapter of the American Academy of Emergency Medicine)

DOI

unavailable

PMID

20882153

PMCID

PMC2941370

Abstract

OBJECTIVES: Motor vehicle-pedestrian crash is a significant public health concern. The urban campus of Georgia State University poses unique challenges due to a large number of students and university employees. The objectives of this study are twofold: 1) to examine the correlation between specific features of the built environment on and around the University campus and pedestrian crashes; and 2) to identify crash clusters in the study area using network-based geospatial techniques. METHODS: We obtained pedestrian crash data (n=119) from 2003 to 2007 from Georgia Department of Transportation and evaluated environmental features pertaining to the road infrastructure, pedestrian infrastructure and streetscape for each road segment and intersection. Prevalence rate of each feature with pedestrian crashes present was calculated. We used network-based Kernel Density Estimation to identify the high density road segments and intersections, then used network-based K-function to examine the clustering of pedestrian crashes. RESULTS: Over 50% of the crosswalk signs, pedestrian signals, public transit, and location branding signs (more than three) at intersections involved pedestrian crashes. More than half of wider streets (greater than 29 feet), two-way streets, and streets in good condition had pedestrian crashes present. Crashes occurred more frequently in road segments with strong street compactness and mixed land use present and were significantly (p<0.05) clustered in these high-density zones. CONCLUSION: Findings can be used to understand the correlation between built environment and pedestrian safety, to prioritize the high-density zones for intervention efforts, and to formulate research hypotheses for investigating pedestrian crashes.

Language: en