Severity of injury resulting from pedestrian–vehicle crashes: What can we learn from examining the built environment?

Abstract

This paper examines the impact of personal and environmental characteristics on severity of injuries sustained in pedestrian–vehicle crashes using a generalized ordered probit model. The data covers 2000–2004 of pedestrian–vehicle crashes taken from police incident reports for Baltimore City and supplemented with local land use, urban form and transportation information specific to the individual crash locations. The results on personal and behavioral variables confirm previous findings. Women pedestrians involved in crashes tend to be injured less frequently than their male counterparts; children have an increased likelihood of sustaining injuries and older persons are more likely to be fatally injured. Pedestrians who cross against the traffic signal, are not in a crosswalk and are involved in a crash after dark are associated with greater injury risk. Of the built environment policy variables of interest, transit access and greater pedestrian connectivity, such as central city areas, are significant and negatively associated with injury severity. These results suggest that the environmental conditions should be given more scrutiny and be an important consideration when evaluating and planning for pedestrian safety.

Introduction

Although pedestrians are involved in a relatively small proportion of vehicle crashes −2 to 3%, they represent a much higher proportion of crash fatalities – around 11–13% of those killed in a collision (National Highway Traffic Safety Administration, 2001). In more than half of these deaths, improper pedestrian behavior, such as crossing against a traffic light, or impairment by alcohol or drugs, were contributing factors. For this reason, the majority of research on pedestrian safety has focused on pedestrian culpability, including risk taking behaviors, drug or alcohol impairment or signal disobedience (Preusser et al., 2002). There have been also been investigations into the effects of these behaviors on the severity of injuries sustained (Sze and Wong, 2007).

With respect to pedestrian safety, planners and engineers are concerned primarily with modifying the physical environment to reduce risk and improve the walkability of the environment, including design features that may reduce potentially dangerous behaviors by pedestrians. However, research examining the influence of the attributes of the environment, including land use, urban form and transportation facility characteristics on crashes and their outcomes is less common in pedestrian safety research. Because over 70% of all pedestrian crashes in the US occur in urban environments (Hebert-Martinez and Porter, 2004; National Highway Traffic Safety, 2002), it is important to understand the interaction between pedestrian crashes and various characteristics of the urbanized environment in order to better direct public safety investments.

Intuitively, it appears that the built environment could have significant influence on pedestrian crash outcomes, particularly micro-scale design features such as the design of the intersection and presence of crosswalks. But other aspects of the built environment may also have an influence. Characteristics of buildings such as building height and bulk and set backs from the lot line may influence lines of sight or the visual order or degree of visual confusion along a corridor. Higher order transportation facilities, such as arterials and highways, are designed for higher vehicle speeds and volumes and tend to be less hospitable to pedestrians. Areas with pedestrian generators, such as high-density development and commercial land uses, would expect to see greater numbers of crashes because by nature, they attract greater numbers of pedestrians. Yet, the injuries sustained in these areas may be less severe because congestion and specific built environment attributes result in lower vehicle speeds.