Type of Problem Being Addressed
General Description of the Problem
While the number of annual pedestrian fatalities due to traffic accidents had generally decreased across the United States over the latter part of the 1990s (about 13 percent overall from 1992 to 2002, per NHTSA Web site), that trend seems to have changed somewhat over the early years of the new millennium (see Exhibit III-1). There were 71,000 pedestrians injured in traffic crashes in 2002 (per NHTSA Web site).
The number of conflicts and fatalities remains high in many urban areas and for specific segments of the population. In addition, results of travel surveys suggest that the observed drop in pedestrian fatalities in recent years may simply reflect reduced exposure rather than any gains in pedestrian safety.
The need to reduce pedestrian deaths and injuries (see Exhibit III-2), even in the face of ongoing efforts to increase levels of walking, continues to be an important goal for the engineering profession. Specific groups that do not or cannot drive primarily depend on walking for transportation, including children, the elderly, and low-income populations. These individuals comprise up to 30 percent of the population in many communities and are particularly in need of a safe walking environment to help lower their risk of injury and death.
The U.S. Census is the most complete information on the percent of journey-to-work trips made by walking. For the 2000 U.S. Census, the percentage of journeys to work by foot was 2.9 percent, or 3.8 million workers 16 years and over (Reschovsky, 2004). This estimate is lower than the 1990 Census data, which showed 3.90 percent of workers 16 years and over, or 4.5 million people, walking to work. About 1 in 5 trips involve travel to or from work.
The Nationwide Personal Transportation Survey, which measures travel of all kinds at the national level, also indicates there has been a decrease in the percent of trips made by walking. In 1995, approximately 20 billion trips, or 5.4 percent of all trips, were made by walking. These numbers compare to 18 billion walking trips, or 7.2 percent of all trips, in 1990.1 While the absolute number of walking trips increased by about 11 percent between the 1990 and 1995 NPTS surveys (Hu and Young, 1992,1993; U.S. Department of Transportation, 1995), it was far less than the increase in trips by private auto, creating a reduction in the percentage of total trips by walking. If walking trips had increased at the same rate as private auto trips, the observed reduction in pedestrian fatalities would likely have been much smaller. During the 5-year time period covered by the two NPTS surveys, pedestrian fatalities decreased by 13.9 percent (from 6,482 to 5,584). Engineering improvements coupled with enhanced safe behavior by pedestrians and motorists are needed to further reduce pedestrian fatalities.
States and Local Areas with the Highest Numbers of Crashes
Crash statistics differ significantly by State and local jurisdictions. States with the highest number of pedestrian crashes per 100,000 population in 2000 included Florida, Arizona, Delaware, and New Mexico; the District of Columbia also has a high rate. State pedestrian traffic fatality counts and fatality rates are presented in Exhibit III-3.
Factors Affecting the Number and Severity of Crashes
Alcohol impairment may be as serious a problem for pedestrians as it is for motor-vehicle drivers, although there is evidence the problem may be lessening, based upon fatal crash data for the year 2000. From 1980 through 1987, 37 percent to 44 percent of fatally injured pedestrians had a reported blood-alcohol concentration (BAC) of 0.10 or greater (Federal Highway Administration, 2002). In 1997, that figure was 29.5 percent, and in 2002 it decreased to 21 percent (NHTSA Web site). Alcohol involvement in pedestrian crashes continues to be a concern, however, due to the continued high percentage of either drivers or pedestrians who have some level of BAC. Alcohol involvement—either for the driver or the pedestrian or both—was reported in nearly one-half of all pedestrian fatalities (NHTSA Web site). However, care should be taken in using these results, as NHTSA cautions that BAC results reported to the Fatality Analysis Reporting System (FARS) are from state measurements and many are untested. Also, it is not clear whether the drop in pedestrian fatalities involving alcohol-impaired pedestrians may be partly the result of less reporting of alcohol involvement due to changes in police practices in 2000.
Speed is a major contributing factor in crashes of all types (see Exhibit III-4). In 2000, high vehicle speed was a contributing factor in 29 percent of all fatal crashes, a number slightly lower than in previous years—30 percent in 1994, 1996, 1997, 1998, and 1999 and 31 percent in 1995 (NHTSA Web site). Speed has serious consequences when a pedestrian is involved (see Exhibit III-4). A pedestrian hit at 64.4 km/h (40 mph) has an 85-percent chance of being killed; at 48.3 km/h (30 mph), the likelihood goes down to 45 percent, while at 32.2 km/h (20 mph), the fatality rate is only 5 percent (U.K. Department of Transport). Faster speeds
also increase the likelihood of a pedestrian being hit. At higher speeds, motorists are less likely to see and react to a pedestrian, and are even less likely to be able to stop in time to avoid hitting one (Federal Highway Administration, 2000). Speed, however, is always a factor in crashes, regardless of whether it is illegal (i.e., above the posted speed limit) or not. Speed limits that are set inappropriately high can also contribute to pedestrian crashes and injuries.
Types of Pedestrian Crashes
In order for engineers and planners to address specific pedestrian hazards and high-crash locations, information is needed on where the pedestrian crashes occur (city, street, intersection, two-lane road, etc.), when they occur (time of day, day of week, etc.), characteristics of the victims involved (age, gender, injury severity, etc.), and the events that precipitated the crash (child chasing ball onto road, motorist swerving around a parked car, etc.).
Where Crashes Occur
Pedestrian crashes occur most frequently in urban areas where both pedestrian activity and traffic volumes are greater than in rural areas. The National Safety Council estimates that 85.7 percent of all nonfatal pedestrian crashes in the United States occur in urban areas and 14.3 percent occur in rural areas. However, 25 percent of pedestrian fatalities occur in rural areas, where vehicle speeds are higher than on city streets (Zegeer et al., 1992, 1993). In addition, many rural areas have no sidewalks, paths, or shoulders to serve as separated pedestrian facilities, and no lighting to increase the visibility of pedestrians at nighttime.
According to the NHTSA, “most pedestrian fatalities in 2000 occurred in urban areas (71 percent), at nonintersection locations (78 percent), in good weather conditions (91 percent), and at night (64 percent).” Additionally, “more than two-thirds (68 percent) of the 2000 pedestrian fatalities were males.” While all age groups are more likely to be killed at nonintersection locations, the numbers are higher for children primarily because of dartouts into the street. Likewise, the oldest age groups are more likely to be struck at intersections since older pedestrians tend to cross at intersections more often than younger ones. Moreover, some older pedestrians have physical, visual, and/or hearing impairments that place greater demand on intersection design (Zegeer et al., 1992). Studies have shown that older pedestrians are particularly over-represented in crashes at intersections involving vehicles turning left and right (National Highway Traffic Safety Administration, 1990b) (see Exhibit III-5 and Exhibit III-6).
When Crashes Occur
Exhibit III-7 and Exhibit III-8 show the time of day for when crashes occur.
Characteristics of the Victims
Specific populations that are heavily represented in crash injury and fatality statistics are children under the age of 16 and older pedestrians. Both of these groups deserve special attention because for many of them driving is not an option and, in the case of older pedestrians that no longer drive, their numbers will increase dramatically as a result of the “graying of the population.” “Older pedestrians (ages 70+) accounted for 17 percent of all pedestrian fatalities and 6 percent of all pedestrians injured. The death rate for this group, both males and females, was 3.18 per 100,000 population—higher than any other age group” (NHTSA Web site). The pedestrian age group that is most likely to be involved in a crash is 5- to 9-year-old males, who tend to dart out into the street, a problem that can be aggravated by higher vehicle speeds in areas where children are walking and playing (U.S. Department of Transportation, 2001).
To address pedestrian motor-vehicle safety problems, agencies must have information on factors precipitating a crash. Exhibit III-9 below contains information on factors related to fatal collisions involving a pedestrian and a single motor vehicle. The percentages in the graph total more than 100 percent because in some instances more than one related factor was identified. Most frequently cited were improper crossing of a roadway or intersection and walking, playing, or working in the roadway.
The National Highway Traffic Safety Administration developed a methodology for typing pedestrian crashes in the 1970s (National Highway Traffic Safety Administration, 1971). The method was refined in the early 1990s and used to determine the crash types for more than 5,000 pedestrian crashes in the States of California, Florida, Maryland, Minnesota, North Carolina, and Utah (Hunter et al., 1995; National Highway Traffic Safety Administration, 1971) (see Exhibit III-10).
Key findings of this study, which pertained to both fatal and nonfatal crashes, included the following:
Crash types that were the most severe as measured by the percentage of pedestrians seriously injured or killed were
Least severe crashes included
Based upon these findings and additional research, 13 crash type groupings (12 specific types and 1 miscellaneous type) have been identified for use with crash data to identify safety problems and corresponding countermeasures (see Exhibit III-11 for the 12 specific types). They can also be used to help educate safety professionals, as well as the general public, about the types of situations that pose dangers to pedestrians. These crash types form the basis for the Pedestrian and Bicycle Crash Analysis Tool software known as PBCAT (Harkey et al., 2000).
Appendix 1 presents a matrix of these 12 major crash types showing which strategies might be considered to help mitigate each crash type.
1 The NPTS survey methodology changed between 1990 and 1995 from a telephone survey to a travel-diary survey. This resulted in an increase in the reported number of trips overall in 1995, a change which does affect the accuracy of comparisons between different year NPTSs.