Comprehensive analysis of vehicle–pedestrian crashes at intersections in Florida
Introduction
As population and traffic volume increase, the conflicts between pedestrians and vehicles on roads are more frequent and consequently, vehicle–pedestrian crashes have become a major concern in improving traffic safety. According to the National Highway Traffic Safety Administration (2004), approximately 4700 pedestrians were killed and 70,000 pedestrians were injured in the United States in the year 2003. Although the number of pedestrian victims has continuously decreased over the last 16 years (1988–2003), actual pedestrian crash risk may not have been reduced considering the fact that the number of pedestrian trips has been reduced as more people own and drive cars. In particular, the current intersection design guidelines mainly focus on the operation and safety of vehicles rather than pedestrian safety (Pietrucha and Opiela, 1993).
For the improvement of pedestrian safety at intersections, understanding the effects of crash-related factors on pedestrian crashes can help develop more effective countermeasures as the types of pedestrian crashes vary with these factors (Stutts et al., 1996). Thus, the past studies on frequency and injury severity of pedestrian crashes have generally focused on the effects of pedestrian and driver characteristics, vehicle characteristics and conditions, and road geometric and traffic characteristics of intersections.
A number of studies identified that there exists strong relationship between demographic factors of pedestrians (particularly age) and crash risk. For example, Zegeer et al. (1993) observed high fatality of older pedestrians in daytime, on weekdays, and in winter. They also found that since older pedestrians reacted slowly with reduced vision and wore dark clothing at nighttime, dark lighting conditions were more hazardous to older pedestrians. Fontaine and Gourlet (1997) found that children and the elderly were the most vulnerable to pedestrian crashes among age groups. Oxley et al. (1997) found that when cars approached closely, older pedestrians crossed more frequently and adopted unsafe road crossing strategy (e.g. slow walking speed, delay in reaction). Similarly, Tarawneh (2001) observed that old pedestrians walked slower than younger pedestrians and their level of exposure to vehicle traffic increased. Al-Ghamdi (2002) found that the fatality rate of the old-age pedestrian group (age 60 and over) was the highest whereas the fatality rate of the young-age pedestrian group (age 20–29) was the lowest. These results suggest that old pedestrians are most likely to be involved in a crash and also severely injured when they are involved in a crash.
Some studies claimed that a pedestrian's alcohol use is also an important factor affecting pedestrian crashes. Miles-Doan (1996) suggested that alcohol-impaired pedestrians were more involved in pedestrian crashes and their odds of dying relative to surviving were higher than non-alcohol-impaired pedestrians. Öström and Eriksson (2001) found that intoxicated pedestrians were more severely injured and suffered more head injuries than non-intoxicated pedestrians. Some studies considered the combined effect of pedestrian age, gender and alcohol use on crash risk. For example, Holubowycz (1995) reported that young and middle-age intoxicated males are high-risk pedestrian groups.
Vehicle characteristics and conditions such as vehicle speed, vehicle types, and vehicle movement are also closely associated with pedestrian crashes. For instance, Anderson et al. (1997) observed that when the speed limit was reduced, the number of fatal pedestrian crashes was also reduced. Some researchers compared the injury severity of pedestrian crashes caused by different vehicle types and vehicle movement. Lefler and Gabler (2004) found that the pedestrian fatality rate when struck by light trucks and vans (LTV) was two to three times greater than the fatality rate when struck by passenger cars since LTVs have higher bumpers and more blunt frontal profiles. Preusser et al. (2003) found that turning vehicles often caused pedestrian crashes because drivers failed to yield the right of way to pedestrians at intersections.
A few studies examined the effect of road geometric and traffic characteristics on pedestrian crash risk. Given that median not only blocks vehicle interactions in different directions but also provides safe refuge area for pedestrians, Bowman et al. (1994) demonstrated that different types of median have different effects on pedestrian crashes. LaScala et al. (2000) observed that injuries in pedestrian crashes were greater in the areas with higher population density, average daily traffic, and number of cross-streets per kilometer roadway through a spatial analysis using a geographic information system. On the other hand, Garber and Lienau (1996) reported contradicting results that fatality rate of pedestrian crashes in rural areas with lower population density was higher than the fatality rate in urban areas. Similarly, Zajac and Ivan (2003) found that pedestrian injury severity was higher in village and downtown fringe areas than downtown and low-to-medium density commercial areas.
In spite of significant research efforts in the past studies, there still remain some unanswered questions: (1) do we need to analyze pedestrian crashes at driver's fault and pedestrian crashes at pedestrian's fault separately since their causal factors may be different? (2) Does higher frequency of pedestrian crashes actually reflect higher crash risk by pedestrians? As daSilva et al. (2003) suggested, due to lack of information on walking patterns by different age groups, most studies on pedestrian crashes could not properly reflect risk by age groups. (3) Are there any other important road geometric and traffic characteristics affecting pedestrian crashes, such as average traffic volume at intersections, that we are unable to identify since they are not readily available in typical crash reports?
Thus, this study has three objectives: (1) to investigate the relationship between frequency/injury severity of pedestrian crashes at intersections and various driver, pedestrian, traffic, and environmental characteristics; (2) to explore exposure of pedestrian walking on the roads for estimating the risk of pedestrian crashes; and (3) to examine the effect of average traffic volume at intersections on the occurrence of pedestrian crashes.
Section snippets
Frequency and injury severity of pedestrian crashes
This study analyzed pedestrian crashes that have occurred at intersections in Florida over 4 years (1999–2002) using the crash information compiled in Florida Traffic Crash Records Database (Florida Department of Highway Safety and Motor Vehicles, 2002). Approximately 7000 pedestrian crashes occurred at intersections or are influenced by intersections. The crashes at mid-block crosswalks and unmarked locations are not included in this study. Five percent of these crashes were fatal crashes. The
Exposure of pedestrian crashes
Despite statistically significant results in the previous sections, it is still uncertain whether the model estimates properly reflect the risk of pedestrian crashes in the absence of exposure measures. In fact, crash risk from a pedestrian's perspective is more influenced by pedestrian volume than vehicle volume (Gårder, 2004). In this regard, population has been often used as exposure to calculate pedestrian crash rate in the form of the number of crashes per capita. Using the classification
Effects of average traffic volume at intersections
It should be noted that all the possible factors affecting pedestrian crashes were not considered in this study, since the factors were limited to only the information included in the crash database. In other words, there may exist more important factors affecting crashes but they could not be captured in the earlier analysis due to lack of data. Thus, this study obtains average traffic volume at intersections that is not typically available in crash reports and investigates its effect on
Conclusions and recommendations
This study analyzed vehicle-pedestrian crashes in Florida from different perspectives. First, the study identified the groups of drivers and pedestrians, and traffic and environmental characteristics that are correlated to pedestrian crashes using a log-linear models. It was found that middle-age male drivers and pedestrians were correlated to more pedestrian crashes than the other age and gender groups, the passenger cars were correlated to more crashes than trucks, vans and buses, and more
References (28)
Pedestrian-vehicle crashes and analytical techniques for stratified contingency tables
Accid. Anal. Prev.
(2002)- et al.
Vehicle travel speeds and the incidence of fatal pedestrian crashes
Accid. Anal. Prev.
(1997) - et al.
Fatal pedestrian accidents in France: A topological analysis
Accid. Anal. Prev.
(1997) The impact of speed and other variables on pedestrian safety in Maine
Accid. Anal. Prev.
(2004)Age, sex, and blood alcohol concentration of killed and injured pedestrians
Accid. Anal. Prev.
(1995)Pedestrian exposure to risk of road accident in New Zealand
Accid. Anal. Prev.
(1995)- et al.
Demographic and environmental correlates of pedestrian injury collisions: A spatial analysis
Accid. Anal. Prev.
(2000) - et al.
The fatality and injury risk of light truck impacts with pedestrians in the United States
Accid. Anal. Prev.
(2004) Alcohol use among pedestrians and the odds of surviving an injury: Evidence from Florida Law Enforcement Data
Accid. Anal. Prev.
(1996)- et al.
Pedestrian fatalities and alcohol
Accid. Anal. Prev.
(2001)