U.S. Department of Transportation
Federal Highway Administration
1200 New Jersey Avenue, SE
Washington, DC 20590
Federal Highway Administration Research and Technology
Coordinating, Developing, and Delivering Highway Transportation Innovations
|This report is an archived publication and may contain dated technical, contact, and link information|
Federal Highway Administration University Course on Bicycle and Pedestrian Transportation
PDF Version (1.46 MB)
PDF files can be viewed with the Acrobat® Reader®
LESSON 3: PEDESTRIAN AND BICYCLIST SAFETY
Many professionals involved in pedestrian and bicycle programs have never received training that focuses on crash causation. Everyone—from traffic safety specialists to traffic engineers, planners, educators, and law enforcement personnel—can benefit from an understanding of how crashes occur and how to avoid them. It should be noted, however, that in some cases, the study of bicycle and pedestrian conflicts with vehicles may be more useful than crash analyses. Conflict analysis may be useful because bicycle and pedestrian crashes may be a rare occurrence on some facilities and may not be the best measure of operational analysis.
This lesson provides an understanding of crash characteristics, crash rates, exposure, crash typing, crash analysis, benefit-cost analysis, and associated countermeasures. The most significant crash types will be explained and associated with contributing factors and typical errors made. The concepts of corridor and site crash analysis and team problem solving will be emphasized. Discussion will include special conditions—especially nighttime crashes, those involving impaired drivers, pedestrians and bicyclists, and high-speed roadways. The major sections of the lesson are as follows:
3.2 What Is a Crash?
The word "crash" may be new to some people as a way to describe the event in which a bicyclist or pedestrian collides with the ground, a motor vehicle, or any other solid object in a way that can result in bodily harm and/or property damage. Historically, these events were called accidents. The term accident implies heavy doses of chance, unknown causes, and the connotation that nothing can be done to prevent them.
Crashes are preventable. Bicyclist and pedestrian crashes are not random events. They fall into a pattern of recurring crash types and occur because the parties involved make mistakes. The mistakes can be identified and counteracted through a combination of education, skill development, engineering, and enforcement measures that can substantially reduce crash occurrences.
3.3 The Crash Avoidance Process
Whether you are a pedestrian, bicyclist, or motorist, you generally go through a similar sequence of actions leading from searching for and recognizing a potential crash situation to taking steps to avoid it. The steps in this sequence are described below. If either party overlooks any of these steps, a crash may result.
3.4 Number of Bicycle and Pedestrian Crashes
According to statistics from NHTSA, approximately 4,749 pedestrians and 622 bicyclists were killed in 2003 as a result of collisions with motor vehicles (see figure 3-2).(1,2)As a group, pedestrians and bicyclists comprise about 13 percent of all highway fatalities each year. There were more than 70,000 pedestrian injuries and 46,000 bicyclist injuries reported as well in 2003 (see figure 3-3).
The reported injuries are subject to significant undercounting; many more injuries were not reported to record-keeping authorities. A study by Hunter, et al., indicates that less than two-thirds of bicycle-motor vehicle crashes serious enough to require emergency room treatment were reported on State motor vehicle crash files.(3)Another study of emergency room records confirmed that there are numerous pedestrian and bicycle injuries that are never recorded by transportation authorities.(4)This may be due, in part, to the fact that the crashes didn’t involve a motor vehicle (70 percent of reported bike injuries and 64 percent of pedestrian injuries) or that they occurred at nonroadway locations such as sidewalks, parking lots, or off-road trails (31 percent of bike injuries and 53 percent of pedestrian injuries). However, these crashes should not be overlooked or dismissed, as they still may have been caused by poor roadway facilities that could be improved by engineering or planning methods.
The number of pedestrian fatalities has dropped 16 percent from the 5,649 pedestrians killed in 1993, showing a steady decline in pedestrian-motorist crashes each year (see figure 3-2). The number of bicyclists killed in 2003 was 24 percent lower than the 816 fatalities reported in 1993. From 1993 to 2003, bicyclist fatalities per year have fluctuated, but they have not risen higher than the number reported in 1993 (see figure 3-2).
Of all nonmotorist traffic fatalities recorded in 2001, 85 percent were pedestrians, 13 percent were bicyclists, and the remaining 2 percent were inline skaters, skateboard riders, etc.
3.5 Summary of Bicycle and Pedestrian Crash Characteristics
Bicycle-Motor Vehicle Crashes
NHTSA provided the following statistics from 2003 that describe bicycle-motor vehicle crashes.(1)
Sixty-two percent of bicyclists killed and 84 percent of those injured were between 5 and 44 years of age. In 2003, the average age of those killed was 35.8 years; the average age of those injured was 26.5 years. The proportion of bicyclists 25 years of age and older killed in motor vehicle crashes has risen since 1993. In 1993, 25 to 64-year-olds made up 41 percent of all bicyclist fatalities, while in 2001, that age group made up 57 percent.
At the same time, the proportion of bicyclists under age 16 involved in fatal crashes has declined since 1993. In 1993, young bicyclists (16 and younger) accounted for 38 percent of all bicyclists killed. In 2003, they accounted for 23 percent of all bicyclists killed and 37 percent of those injured. Twenty-three percent of the bicyclists killed in 2003 were between 5 and 15 years of age.
Most of the bicyclists involved in motor vehicle crashes in 2003 were males—88 percent of those killed and 78 percent of those injured. The male bicyclist fatality rate was almost eight times as high as the female rate, and the male injury rate was more than three times higher than the rate of injury for females.
Alcohol involvement, with either the driver or bicyclist, was reported in more than 33 percent of the fatal bicyclist crashes in 2003. In 32 percent of the fatal incidents, either the driver or the bicyclist was considered legally intoxicated (blood alcohol concentration of 0.08 grams per deciliter or greater). Lower alcohol levels were reported in an additional 7 percent of fatal bicycle-motor vehicle crashes. More than one-fourth of the bicyclists killed were intoxicated. This may be an emerging problem. Alcohol-drug crashes were more frequent on weekends and during hours of darkness.
Time of Day and Year
Almost one-third (31 percent) of the fatal bicyclist crashes occurred during late afternoon and early evening hours between 5 p.m. and 9 p.m. Bicyclist and pedestrian activity (i.e., exposure) is likely quite high during these hours, and visibility can be a problem. Thirty-five percent of all bicyclist fatalities occurred during the months of June, July, and August, months when bicycle riding peaks and exposure is high.
About 69 percent of the fatal bicyclist crashes were categorized as urban, and 71 percent of the fatal crashes occurred at nonintersection locations. However, for nonfatal crashes, more collisions occurred at roadway intersections than nonintersection locations. Roads with narrower lanes, no shoulders, and higher speed limits were associated with more than their share of serious and fatal injuries to bicyclists.
Factors Contributing to Bicycle-Motor Vehicle Crashes
The 1996 FHWA investigation studied 3,000 bicycle crashes in 5 States to determine factors contributing to the crashes.(3)Bicyclists were judged to be at fault in about half of these crashes with motor vehicles. Failure to yield, riding against traffic, stop sign violations, and safe movement violations were the bicyclist factors most frequently cited as contributing to crashes. Bicyclists need training about how to ride in traffic. The likelihood of the bicyclist being responsible for the crash was greatest for younger bicyclists. When the bicyclist involved in the crash was older, the motor vehicle driver was more likely to be at fault.
Motor vehicle drivers were judged to be solely at fault in 28 percent of the cases. Failure to yield, hit-and-run, and failure to see bicyclists were the driver factors most frequently cited as contributing to crashes.
Pedestrian-Motor Vehicle Crashes
NHTSA provided the following statistics that characterize pedestrian crashes:(2)
Compared to their representation in the overall U.S. population, young persons (under 25 years of age) were overrepresented in pedestrian crashes with motor vehicles, whereas older adults (ages 50 to 69) and the elderly (age 70+) were underrepresented. However, elderly pedestrians who were involved in crashes were more likely to be killed. Older pedestrians (ages 70+) accounted for only 6 percent of pedestrians injured but 16 percent of pedestrians killed. Pedestrians aged 0 to 16, on the other hand, accounted for 27 percent of all pedestrians injured in 2003, but only 9 percent of those killed.
Sixty-nine percent of the pedestrian fatalities that occurred in 2003 were males. The male pedestrian fatality rate was more than double (2.27) the female pedestrian fatality rate; the pedestrian injury rate was considerably higher for males than females as well.
Alcohol is often a key factor in pedestrian-motor vehicle crashes that lead to fatalities. In 2003, 46 percent of the traffic crashes that resulted in pedestrian fatalities involved alcohol intoxication by either the driver or the pedestrian. Thirty-four percent of the total number of pedestrians involved in fatal crashes were considered legally intoxicated, while 13 percent of the drivers involved were intoxicated. Six percent of the fatal crashes involved both an intoxicated driver and pedestrian.
Time of Day and Week
Pedestrian crashes occurred most frequently during the late afternoon and early evening hours, times when exposure is probably highest and visibility may be a problem. Forty-four percent of the pedestrian fatalities under 16 years of age occurred between 3 p.m. and 7 p.m. Sixty-five percent of all pedestrian incidents occurred at night. Fatality counts also rose on weekends as well, with nearly one-half of all pedestrian fatalities occurring on a Friday, Saturday, or Sunday.
About 72 percent of the crashes were categorized as urban. The elderly were overrepresented in commercial parking lot crashes, young adults in noncommercial parking lot crashes, and children under age 10 in collisions occurring in driveways, alleys, or yards.
Seventy-nine percent of the fatal pedestrian-motor vehicle crashes occurred at nonintersection locations. However, for nonfatal collisions, the percentage at nonintersection locations was significantly lowered, and more crashes occurred at roadway intersections. Serious and fatal injuries to pedestrians were directly proportional to speed limit and number of lanes.
Factors Contributing to Pedestrian-Motor Vehicle Crashes
In a 1996 study conducted by the FHWA, 5,000 pedestrian crashes in 5 States were studied extensively in order to determine contributing factors associated with the crashes.(3)The report judged that pedestrians were solely at fault in 43 percent of the crashes. Running into the road, failure to yield, alcohol impairment, stepping out from between parked vehicles, and walking or running in the wrong direction (with traffic) were the pedestrian factors most frequently cited as contributing to crashes. Younger pedestrians were more likely to be at fault.
Motor vehicle drivers were judged to be solely at fault in 35 percent of the crashes. Driver hit-and-run and failure to yield were the driver factors most frequently cited as contributing to crashes, followed by improper backing, safe movement violations, and exceeding safe speed. Only 3 percent of motor vehicle drivers striking pedestrians were judged to have been impaired by alcohol in the 1996 study.
3.6 Common Pedestrian and Bicycle Crash Types
This portion of the lesson presents information specific to the types and characteristics of pedestrian and bicycle crashes. The information on crash types was generated from a 1996 FHWA study and a study conducted by the Highway Safety Research Center (HSRC) at the University of North Carolina.(3,5,6)In the FHWA study, 8,000 pedestrian and bicycle crashes in 5 States were studied extensively in order to code crash types, determine the specific factors associated with the crash types, and identify how countermeasures could be used to reduce the frequency of crashes. The purpose of the HSRC study was to update 1970s era crash data to reflect more recent crash types, paying particular attention to roadway and locational factors so that designers can reduce crash frequency through engineering methods and other interventions.
Pedestrian Crash Types
The FHWA study concluded that more than three-fourths of pedestrian crashes fell into one of the following eight crash-type categories (see table 3-1):(3)
Table 3-1. Pedestrian crash types.
These major crash-type categories varied with respect to the pedestrian, driver, location/environmental, and roadway factors that characterized them. It is critically important for individual States and communities to develop a better understanding of the particular traffic situations endangering their residents. Figures 3-4 through 3-11 display the eight most common pedestrian crash types.
Bicycle Crash Types
The FHWA study divided the bicycle-motor vehicle crashes into three general categories (see table 3-2):(3)
Table 3-2. Bicycle crash types.
The most frequent parallel-path crashes were motorist turn/merge into bicyclist’s path (12.2 percent), motorist overtaking the bicyclist (8.6 percent), and bicyclist turn/merge into motorist’s path (7.3 percent). The most frequent crossing-path crashes were caused by motorist failure to yield to bicyclist (21.7 percent), bicyclist failure to yield at an intersection (16.8 percent), and bicyclist failure to yield midblock (11.8 percent). These six individual crash types accounted for almost 80 percent of all bicycle-motor vehicle crashes. Figures 3-12 through 3-19 show the eight most common bicycle crash types.
3.7 Pedestrian and Bicycle Crash Analysis Tool (PBCAT)
FHWA, in cooperation with NHTSA, developed a Pedestrian and Bicycle Crash Analysis Tool (PBCAT) through the University of North Carolina HSRC. PBCAT is a user-friendly analysis tool that determines the crash type through a series of onscreen questions about the crash. It provides tabular and graphing functions of various factors (such as age, sex, light conditions, etc.) associated with the crash. The system also recommends countermeasures and related resources and reference information.
3.8 Bicycle and Pedestrian Countermeasures
Sections 3.6 and 3.7 discussed common crash types and how to determine these crash types by using PBCAT. This section provides a list of countermeasures that might be employed to offset a specific pedestrian crash type (see figure 3-20). Table 3-3 lists more than the general crash types previously provided and was adapted from thePedestrian Facilities Users Guide.(7)Bike countermeasures may overlap with some pedestrian countermeasures. An exhaustive list of bike countermeasures can be found in the FHWA Bicycle Safety Tools and Technology page located on the FHWA Safety Web site. There, roadway countermeasures are listed for numerous types of bicyclist and motorist errors; these are organized by the type of entity that could implement the change.
Table 3-3. Crash types and associated countermeasures.
3.9 Bicycle and Pedestrian Safety Goals
Based on the data from the FHWA study, the following safety goals and research recommendations were made in 1996:(3)
3.10 Geographic Information System (GIS) Crash Frequency Analysis
Geographic information system (GIS) software turns statistical data (e.g., crashes) and geographic data (e.g., roads and crash locations) into meaningful information for spatial analysis and mapping. GIS is an invaluable tool being applied to many pedestrian and bicycle safety issues. It assists in mapping crashes to identify hot spots, planning the safest route to walk or bike based on roadway and traffic elements (sidewalk, curb lane width, crosswalk locations, traffic volume, etc.), and illustrating the relationships between land use, traffic patterns, and bicycle and pedestrian safety. This tool is valuable in helping visualize the data accumulated in crash and roadway databases.
3.11 Benefit-Cost Analysis
Once the crash types have been identified and the appropriate countermeasures selected, the resulting benefit-cost ratio can be calculated. A common formula used for this calculation is the Safety Improvement Index (SII), a formula developed by the Texas Department of Transportation (TxDOT) as part of the Federal Hazard Elimination and Safety (HES) Program to assess the attributes of a safety improvement project and convert it to an index ratio by which each project can be ranked or prioritized in order of importance.(9,10)In its most basic form, SII is a formula that computes the ratio of the potential reduction in crash costs (the "benefit") to the cost of implementing the countermeasure (the "cost"). This formulated value is weighted heavily on the percentage of the expected reduction in the number of crashes that occurred prior to the improvement. Reduction factors have been determined through previous studies, such as the TxDOT HES Program and a national survey of comparative crash reduction factors conducted by Texas Transportation Institute (TTI).(9,11)SII greater than or equal to 1.0 is considered to be cost effective. However, the ratio is not designed to measure the effectiveness of individual projects. Rather, it is a method by which many projects can be compared using the same set of criteria. By way of this comparison, a prioritization list of improvement projects can be formed. With this prioritized list, the projects can be funded beginning with the most important project, and each subsequent safety improvement project is then funded individually and sequentially thereafter.
Some of the other factors included in this index are:
3.12 Student Exercise
The following ideas are suggested for student exercises.
Design a program that specifically provides countermeasures aimed at reducing one (or more) common bicycle and/or pedestrian crash types. Countermeasures can include physical changes to the bicycle/pedestrian environment (engineered and constructed solutions), or education programs aimed at a particular audience that may be susceptible to certain crash types. Be specific about what the program would include, and how it would be implemented throughout a community. Include an explanation of how you would propose to evaluate the effectiveness of your program.
Using the data provided for the case study location, Piedmont Park in Atlanta, GA, develop some conclusions regarding the crash data obtained through the Georgia Department of Transportation for 1995, 1996, and 1997 (see figures 3-21 through 3-23 and tables 3-4 through 3-7). Cross-tabulations of crashes by time of day, location, and causation factors are helpful in gaining insight into safety problems and possible countermeasures. Data available for such evaluations are often limited due to the low percentage of reported pedestrian and bicycle crashes. However, important information can be obtained by a thorough analysis of available data.
For some general background on the case study location, the following descriptive information is provided:
For conducting a case study evaluation of pedestrian and bicycle conditions at Piedmont Park, the data provided include the following:
Figure 3-21. Illustration. Bicycle crash locations.
Figure 3-22. Illustration. Pedestrian crash locations.
Figure 3-23. Illustration. Site location map.
Table 3-4. Tabulation of pedestrian crash data.
Table 3-5. Tabulation of bicycle crash data.
Table 3-6. Usage data collected at major park entrances.
Table 3-7. Summary of major roadways.
Obtain pedestrian and bicycle crash data for an area of interest. You should obtain a minimum of three years of data to support your crash analysis. Although local city agencies sometimes maintain crash data, the State DOT is the most reliable source of available data. Most States maintain their crash data in a computerized database system and sorts of the data can be conducted on various field entries to list crashes associated with either pedestrians or bicycles. These types of crashes will only constitute a very small percentage of the total crashes occurring along a roadway and it may be useful to receive a full listing of all the crashes associated with your location of interest. DOTs may only maintain data along the more significant roadways and often do not include subdivision/residential streets.
Most DOT personnel are very helpful and willing to work to get you the data you need. You should clearly explain your intentions, location of interest, and type of data that you would like to obtain. Submitting a request in writing is typically required so that your data request can be efficiently processed through their system. In addition to the crash data, you may need other information that will allow you to decode the crash data and to physically link the crash to a location on the roadway network. A crash investigation manual is usually available that lists all of the coded entries used in creating aggregated crash tabulations. Also, a roadway features log is typically available to link milepost listings to physical map features such as intersections, bridges, and street names. In the initial phases of conducting an analysis of crashes, it is seldom necessary to access the actual crash reports. It is much more useful to utilize aggregated crash records that are available through the crash data system. Allow ample time for DOT personnel to accommodate your request within their day-to-day workload. Generally, data can be received in two to three weeks after submitting a request.
3.13 References and Additional Resources
The references for this lesson are:
Additional resources for this lesson include: