Pedestrian Safety Strategic Plan: Background Report
PEDESTRIAN SAFETY CONCERNS IN THE U.S.
Factors Related to Pedestrian Crashes and Fatalities
According to the 2001 National Household Travel Survey (the most recent travel survey data available), walking trips accounted for 2.8 percent of commutes to work, and 8.6 percent of all trips. As pedestrians are the most vulnerable of all road users, they are overrepresented in crashes, especially fatal crashes, in comparison to their mode share of trips. In 2007, there were 4,654 pedestrians killed in traffic accidents, representing a 3 percent decrease from the 4,795 deaths in 2006. Pedestrian fatalities have fallen by 12.5 percent since 1997, and 17 percent since 1993 (see Figure 2). However, it is not clear if this decrease is due to changes in pedestrian activity (i.e., fewer people walking) or improvements in walking conditions, or a combination of both. In fact, data regarding the number of pedestrians using a facility, or pedestrian exposure data, is not included in Figures 2 and 3.
In terms of injuries, 2007 saw the highest number since 2003. The 70,000 reported injuries in 2007 represent a 15 percent increase from the previous year. However, reported injuries have fallen almost 17 percent since 1995 (see Figure 3). It is important to note that these figures represent only estimates of reported pedestrian injuries, and studies have shown that actual injury rates could be much higher due to unreported cases. In addition, the lack of pedestrian exposure data also affects the validity of the following conclusions. Following this figure are some pedestrian safety trends surrounding various issues.
Making up only 9 percent of the total U.S. population, individuals over the age of 70 accounted for 16 percent of all pedestrian fatalities in 2007. Children under the age of 15, roughly 23 percent of the population, represented 8 percent of the total pedestrian fatalities, and 25 percent of pedestrian injuries. These crash estimates may correlate with the exposure (i.e., amount of walking activity) of each age group.
Males represented 70 percent of pedestrians killed in traffic accidents in 2007. The male fatality rate of 2.19 per 100,000 population was substantially higher than the female fatality rate of 0.91 per 100,000 population. According to the 2001 NHTS data, the number of walking trips taken by males and females was roughly equal, so there is little evidence to support the notion that males are more “exposed” than females in terms of pedestrian trip-making. However, there is some evidence that males in general may walk longer distances than females, which may increase their exposure in terms of time and distance as a pedestrian (Clifton & Levi, 2005). Speculation has also been made that males may be more likely to take risks than females or to walk in more dangerous locations than females. Little information exists on this issue and more research is needed to understand the relationship between gender and pedestrian safety.
Vulnerable or Overrepresented Groups
As described in the section on age, older pedestrians are overrepresented in fatal pedestrian crashes as compared to their representation in the general population. Also, both Hispanic and African American pedestrians tend to be overrepresented in fatal crashes. In 2006, Hispanic pedestrians accounted for 17 percent of pedestrian fatalities, but were only 15 percent of the U.S. population (Chang, 2008). Similarly, African American pedestrians accounted for 14 percent of pedestrian fatalities but only 12 percent of the U.S. population (NHTSA, 2008b; U.S Census Bureau, 2006).
As described on the section on gender, males are twice as likely to be involved in fatal pedestrian crashes as females. Children are also slightly overrepresented in pedestrian crashes, but not pedestrian fatalities. From 1997 to 2006, children under the age of 15 (a group that made up 21 percent of the population) represented 8 percent of pedestrian fatalities and 23 percent of pedestrian crashes (Chang, 2008).
Alcohol was involved in 49 percent of pedestrian fatalities in 2007. Table 3 represents alcohol use by the driver, the pedestrian, or both parties involved in a fatal crash.
(Source: NHTSA, 2007).
The speed of a vehicle is a major determinant in the severity of a crash. According to one study (and several other studies have found similar results), a pedestrian hit at 40 miles per hour has an 85 percent chance of fatality, while a pedestrian hit at 20 miles per hour has only a 5 percent chance of fatality (U.K.DOT, 1987). Additionally, even though the proportion of fatalities resulting from crashes goes down as the vehicle speed decreases, there are still a large number of injuries that result from moderate and lower-speed crashes (see Figure 4).
High vehicle speeds may be related to the road type (local/collector/arterial), road context (rural/urban), and road design (i.e., the presence of pedestrian infrastructure). More detail with regards to these typologies can be found in the following sections. Some tactics that have been implemented to discourage speeding have included traffic calming and citywide speed limit reductions. While these may incite resistance from politicians and residents because of concerns regarding longer travel times, studies show a relatively minor impact on travel times. Pedestrians are likely to gain the most from speed limit reductions, but benefits have been seen for drivers as well through reductions in road crashes, generally improved attitudes and awareness towards safety, a more livable environment, and increased automobile energy efficiency (Archer, 2008).
Several other studies, completed independent of each other, also support the relationship between higher speeds and more severe traffic-related injuries or death (Leaf & Preusser, 1999; TRB, 1998). This relationship may be because at higher speeds, drivers are less likely to see a pedestrian and are even less likely to be able to stop in time to avoid a collision (Harkey & Zegeer, 2004).
While the relationship between vehicle speed and injury severity is fairly well understood, it is not clear how speed contributes to the cause of a crash or how many pedestrian crashes have resulted in part because of driver speeding (or how speeding is defined or measured). Currently, most data collected by crash investigators is not reliable in assessing the role of driver speed in crashes. More work is needed to improve data collection and support research on understanding driver speed and speeding behaviors as they relate to pedestrian safety.
Several other factors also characterized pedestrian fatalities in 2007:
Common Crash Types
The following crash types were most often reported for pedestrian fatalities in 2008:
(Source: FARS, 2008). *Note: Multiple crash factors were reported for many fatalities, so the sum of percentages may not equal 100 percent. †Note: These percentages do not account for pedestrian exposure data.
Crash types for pedestrian injuries at the national level are not known at this time.
Locational Factors Related to Pedestrian Safety
High Crash States and Cities in the U.S.
There are different ways to quantify pedestrian crashes. Two common ways are: 1) raw crash numbers, or frequency, of pedestrians involved in crashes, and 2) by rates, which can normalize crash numbers by population or some other measure, such as time spent walking or distance traveled. Since we do not have consistent and accurate national data regarding time spent walking, distance traveled, or pedestrian exposure, we will present pedestrian crash statistics by frequency and as a rate per 100,000 population.
In 2008, the following states had the highest number of pedestrian fatalities:
(Source: FARS, 2008).
These states had the lowest number of pedestrian fatalities in 2008:
(Source: FARS, 2008).
(Source: Chang, 2008). *Note: This does not include consideration of pedestrian exposure data.
Fatality rates (fatalities per 100,000 population) are one way to compare different localities in terms of pedestrian crashes. However, rates may be a biased measure to use without also considering other measures, such as exposure, since the rate alone cannot show whether more/fewer people are killed or more/fewer people are walking in general in these locations. Rates should not be used to compare two locations where it is expected that pedestrian exposure (i.e., the number of people walking) differs dramatically. This is because pedestrian crashes are not thought to have a straight-line relationship with pedestrian exposure, so comparing rates between two very different exposure situations may not give an accurate comparison. The NCHRP Report 08-78 will seek to address this issue, but may fall short of creating a universally applicable guide for all situations (TRB, 2010). The FHWA should continue to lead research in the area of pedestrian exposure, just as they have done for motor vehicle volume data collection with the Traffic Monitoring Guide and Highway Pavement Management System. Pedestrian exposure data is critical for understanding and evaluating pedestrian risk.
In 2008, the following states had the highest rates of pedestrian fatalities per 100,000 population:
(Source: NHTSA, 2007).
By contrast, the following states had the lowest rates of pedestrian fatalities in 2008:
(Source: NHTSA, 2007).
Between 1997 and 2008, the following states have had the highest average rate of pedestrian fatalities per 100,000 population:
(Source: FARS, 2008).
The following states have had the lowest average rate of pedestrian fatalities between 1997 and 2008:
(Source: FARS, 2008).
Crashes in Rural Versus Urban Areas
In 2007, about 73 percent of fatalities occurred in urban areas (NHTSA, 2008a). A study using police reports from 2001 to 2002 and 2001 NHTS data in New York State looked at the urban and rural variation in walking patterns and pedestrian crashes. The study estimated the rates of pedestrian/motor vehicle crashes according to miles walked and resident age. The study found that in urban New York City, “the rate of pedestrian crashes per resident year was about four times that in rural areas; in small and mid-size urban areas it was double that of rural areas. However, pedestrian–vehicle collision rates based on miles walked were similar in all urban areas and about twice those in rural areas” (Zhu, 2008). Thus, for New York (and probably other states), it is likely that the higher numbers (and rates per capita) of pedestrian crashes in urban areas are correlated to the higher rates of walking in those areas, as compared to rural areas lacking facilities or a high enough density of destinations within walking distance.
While rural areas may experience fewer crashes or lower pedestrian crash rates, the crashes that do occur tend to be more severe. Regardless of age or sex, pedestrians are 2.3 times more likely to die from a crash with a motor vehicle in a rural setting than in an urban setting (Mueller, 1988). Approximately 57 percent of all automobile fatalities in 2007 occurred in rural roads, even though travel on rural areas comprises only 40 percent of total vehicle miles traveled (Federal Highway Administration, 2007). According to PEDSAFE, this trend could be attributed to the presence of higher vehicle speeds on rural roads in combination with the absence of sidewalks, paths, or shoulders to separate pedestrians from traffic (Harkey & Zegeer, 2004). Exacerbating these existing conditions, crashes in rural areas are often located farther away from quality emergency care, requiring more time for Emergency Medical Services to arrive (Mueller, 1988).
Crashes by Road Type
In urban areas in 2007, the majority of pedestrian fatalities (more than 50 percent of all fatalities) occurred on arterial roads, with a smaller percentage (14 percent of all fatalities) on local roads or streets. For injuries that occurred in urban areas, the trend was similar: close to 50 percent of all pedestrian injuries occurred on arterial roads, with an additional 17 percent of all pedestrian injury crashes occurring on local roads.
In rural areas in 2007, almost 16 percent of all pedestrian fatalities occurred on arterial roads, with another 6 percent on local roads or streets. The trend is similar for injury crashes.
Overall, for both fatality and injury crashes involving pedestrians, the majority of crashes occur on various types of arterial roads. Thus, future research focusing on these road conditions and potential solutions is needed.
(Source: FARS, 2008). *Note: May not add up to 100 percent as blank responses were not available for the urban and rural strata.
Future Demographic, Social, and Policy Changes
Significant population and other trends have been observed in growth, development, and transportation patterns in the United States and are expected to continue. Most apparent among these changes is the growth of the population on the whole and, more specifically, the growth of immigrant and older adult populations. In addition, the increasing urbanization of the country, the recent decrease in vehicle miles traveled, economic crisis, climate change, gas price fluctuations, and changes in transportation and housing policy may result in potential shifts in travel behavior that should be considered in pedestrian safety research.
One of the most notable demographic trends in the United States is the growing senior population. The growth of the older adult age group will likely be one of the most significant factors affecting changes in travel behavior in the near future (Litman, 2009). In 2000, approximately 35 million adults aged 65 and over lived in the United States. By 2025, the number of people aged 65 and older in the U.S. is projected to top 63 million (Shrestha, 2006). The growth of this age group represents an overall aging of the country as a whole: the 65 and up age group will make up 18.2 percent of the population in 2025 (up from 12.4 percent in 2000) while children and adolescents will decrease from 28.5 percent to 26.3 percent and the proportion of 20-64 year olds will decrease from 59 percent to 55.5 percent. The aging of the U.S. is already noticeable—from 2000 to 2008, the median age in the United States increased from 35.3 years to 36.8 years (U.S. Census Bureau, 2000 and 2008). It is expected that the population as a whole will continue to age for several decades due to increased life expectancies, low birth rates, and the aging of the baby boomer generation, which will begin turning 65 in 2010 (Shrestha, 2006).
Several characteristics of the older adult population are worth noting. Unlike previous generations, this generation of older adults is residing primarily in Sunbelt, suburban, and exurban settings—a new trend sometimes called “aging in place” (Frey, 2007). At the same time, older adults are often dependent on alternative forms of transportation, such as walking and public transit, as they tend to live up to 10 years beyond the age when they are able to operate a motor vehicle (Foley, 2002). Consequently, the mobility of older adults in automobile oriented suburban settings may be compromised and their safety and ability to walk in these settings will be increasingly important. In addition, pedestrian infrastructure should consider the needs of older walkers, who are more likely to have a physical disability or walk at a slower pace than younger pedestrians (Lynott, Haase, Nelson, Taylor, Twadell, Ulmer, et al., 2009).
The safety of older pedestrians currently presents a significant concern. Older adults are more likely to be involved in pedestrian vehicle crashes that result in a fatality. In 2007, 70,000 pedestrian injuries and 4,654 pedestrian fatalities were recorded in the United States. Older adults made up 13 percent of the population in 2007, but accounted for 9 percent of pedestrian injuries and 21 percent of pedestrian fatalities (NHTSA, 2008a; U.S. Census Bureau, 2007).
Immigrant Population and Population Growth
Like older adults, the immigrant population in the United States is growing, residing increasingly in the suburbs, and is often dependent on walking and public transit for travel (FHWA, 2006b; Singer, 2009; Wilson, 2009). Immigration accounted for 42 percent of population growth between 2000 and 2005 (Wilson, 2009). In 2007, Latin American immigrants accounted for over half of all immigrants living in the U.S. (Wilson, 2009). On the whole, immigrants tend to have lower unemployment rates, higher poverty rates, and lower levels of education than U.S. born citizens (Wilson, 2009). In 2007, over half of the immigrants in the United States settled in the suburbs and just over a third lived in primary cities (Wilson, 2009). The Southeast is home to most of the top 10 cities with the fastest growing immigrant populations, which are listed in Table 13.
(Source: Wilson, 2009).
The U.S. population on the whole is expected to grow 19 percent by 2020 and 29 percent by 2030 and the Hispanic population will likely account for much of that growth. While the white, non-Hispanic population is anticipated to increase by 5 percent in 2020 and 7 percent by 2030, the Hispanic population is expected to grow at a much faster rate—68 percent by 2020 and 105 percent by 2030 (U.S. Census Bureau, 2004). In general, Hispanic people living in the U.S. are younger than the U.S. population on the whole. In fact, 25 percent of children under five are Hispanic (U.S. Census Bureau, 2009). Recent immigrant households are less likely to own a car and are more dependent on transit and walking as forms of transportation than native born residents (FHWA, 2006b). While the Hispanic population makes up 15 percent of the population, Hispanic people are involved with 17 percent of pedestrian-vehicle crashes (Chang, 2008). Because of their travel behavior, immigrant populations as well as native born Hispanic populations may merit particular attention in pedestrian safety planning and research.
Other Demographic Trends
As a result of this rapidly growing immigrant population and other demographic changes, the United States is expected to see minority groups become the majority versus the Caucasian demographic by 2042 (Frey, 2008). Based on historic travel patterns of minority populations, it is possible that the growth in these groups may result in noticeable changes in travel behavior, including higher rates of walking nationally. Despite rising per capita income, the U.S. has also seen recent increases in poverty. Rising rates of poverty pre-dated the current recession and poverty has become increasingly prevalent in the suburbs (Frey, Berube, Singer, & Wilson, 2009). Low-income households are less likely to own a car and more likely to use alternative forms of transportation, such as walking and public transit (Litman, 2009). In addition, the current economic climate and high rate of unemployment across the nation may have implications for travel behavior in the near future (e.g., such as an increase in the number of people walking or using transit and crossing busy roads where the transit stops are located, etc.).
As of 2000, 80 percent of U.S. residents lived in a metropolitan region; more than half of these metropolitan residents live in suburban settings (Litman, 2009; U.S. Census Bureau, 2000). Of the 100 fastest growing metropolitan areas, 71 were located in the South and 26 were located in the West (U.S. Census Bureau, 2009). Over half of the fastest growing counties were located in, Georgia, or (U.S. Census Bureau, 2009). The top 10 fastest growing metropolitan statistical areas (MSA) are shown in Table 14.
(Source: U.S. Census Bureau, 2009).
The top 10 fastest losing MSAs are:
(Source: U.S. Census Bureau, 2009).
The concentration of growth in the Southern and Western portions of the country may indicate geographic areas for further study focusing on traffic and pedestrian safety. Additionally, the increase of Hispanic and elderly populations living in the suburbs, as well as the rising prevalence of suburban poverty, may have implications for mode choice, mobility, and traffic safety in a setting that is typically considered auto-centric.
Nationally, the journey to work trip in 2000 was usually made by private car for 75.7 percent of people, by carpooling for 12.2 percent of people, by transit for 4.6 percent of people, by foot for 3.3 percent of people, and by other means for the remaining 4.2 percent. In 2001, the travel mode for all trip purposes shows travel behavior that is quite different than the journey to work trip. Driving alone accounts for 41.5 percent of trips; driving with others accounts for 43.7 percent of trips; transit accounts for 1.8 percent of trips; walking accounts for 9 percent of trips; and other means make up the remaining 3.9 percent of trips (FHWA, nd).
The national mode share of journey to work trips taken by public transit has shown steady and significant increases of 0.1 percent each year since 2006. However, the distribution of walking trips has not shown significant increases, holding at a steady 2.8-2.9 percent of the total commuting trips over the same time period (U.S. Census Bureau, 2008). Federal Transit Administration (FTA) investment in transit has been increasing steadily from 2000-2007, with the number of grants being awarded for major programs more than doubling and the amount of funds being offered increasing by approximately 36 percent (FTA, 2007). The steady increase in transit investment in conjunction with the plateau in walking trips may suggest a need for better pedestrian accessibility to transit.
Low-income populations travel less frequently, have the lowest income groups (household incomes of less than $20,000 a year), and are much less likely to own an automobile (Pucher & Renne, 2003). Car ownership is highly associated with travel mode choice. Households without a vehicle make 41 percent of their trips by foot, compared with 12.5 percent for households that own one car (Pucher & Renne, 2003). For African Americans, Asians, and Hispanics, walking accounts for 12-13 percent of trips; whites walk for 8.6 percent of trips. Measured together, low-income nonminority households, African Americans, and Hispanics account for 63 percent of transit users in general and 73 percent of bus riders (Pucher & Renne, 2003). Recent immigrants are also five times as likely to use transit as settled immigrants and U.S. born residents and less than half of recent adult immigrants are licensed to drive (FHWA, 2006b). Fatality rates per mile for pedestrians are much higher than other travel modes (on a per mile basis, pedestrians are 36 times more likely to be involved in a fatal crash than car drivers and passengers) (Pucher & Renne, 2003).
While the private car is the dominate form of travel for the vast majority of trips, vehicle miles traveled (VMT) per capita plateaued in 2000 and, since 2005, has even begun to decline (Puentes & Tomer, 2008). Many of the factors thought to have contributed to past increases in VMT have ceased to have an impact on travel; specifically, the entry of women into the workforce and vehicle ownership rates have likely reached a saturation point (Puentes & Tomer, 2008). There are indications that travel behavior is sensitive to gas price fluctuations in the long run; however, the recent drops in VMT cannot be entirely attributed to recent record high gas prices as the VMT decreases preceded the gas price volatility (Puentes & Tomer, 2008; Litman, 2008). VMT is also affected by income, development patterns, the presence of transportation alternatives, and cultural norms (Litman, 2008; Pucher & Renne, 2003).
It is likely that expected increases in gas prices will change travel patterns, but it is unclear exactly what shape those changes will take. Historically, travelers react to sustained rising gas prices by purchasing more fuel efficient cars and reducing the number of trips they take. There are indications that, when possible, travelers who typically drive in a private car will switch to public transit in the face of high gas prices (Austin, 2008; Litman, 2009).
Policy and Funding Changes
In spring 2009, the U.S. Department of Transportation (DOT) and U.S. Department of Housing and Urban Development announced a partnership and inter-agency task force that would focus on creating sustainable communities. Key priorities of this partnership include bringing affordable housing and employment opportunities closer together, enhancing transportation options and reducing travel times, coordinating transportation and land use planning, and supporting the livability and health of neighborhoods and communities (U.S. DOT, 2009). These priorities indicate a national level focus on creating places that are dense, mixed-use, and walkable. This major shift in policy focus may eventually impact travel patterns around the country, including potential modal shifts. This livability agenda may provide new opportunities for pedestrian safety research.
Policy priorities and funding for pedestrians are inherently intertwined. The FHWA, FTA, and NHTSA all play an important role in pedestrian safety research and should establish a cooperative partnership to facilitate further research. At the national level, there are several sources of federal funding currently available for pedestrian issues. They include, but are not limited to the following:
Safe, Accountable, Flexible, Efficient Transportation Equity Act: A Legacy for Users (SAFETEA-LU): http://www.walkinginfo.org/funding/history.cfm
Transportation Investment Generating Economic Recovery (TIGER) Discretionary Grants (TDR): http://www.dot.gov/recovery/ost/
Congestion Mitigation and Air Quality (CMAQ):http://www.fhwa.dot.gov/environment/cmaq/cmaqbroc.pdf
Transportation Enhancements (TE): www.enhancements.org
Federal Transit Administration New Starts and New Freedom grants:http://www.fta.dot.gov/planning/planning_environment_5221.html andhttp://www.fta.dot.gov/laws/circulars/leg_reg_6624.html
Community Development Block Grants (CDBG):http://www.hud.gov/offices/cpd/communitydevelopment/programs/
Communities Putting Prevention to Work:http://www.hhs.gov/recovery/programs/cdc/chronicdisease.html
Safe Routes to School Program (SRTS): http://www.saferoutesinfo.org/
Through these and other sources, the total amount of pedestrian (and bicycle) funding from federal sources has increased by over $300 million over the past 10 years. However, this proportion has only accounted for about 1-1.5 percent of total annual Federal Aid Highway Program funds reserved for transportation projects (FHWA, 2006a) (see Figure 4). While federal funding available for pedestrian improvements has been increasing steadily over time, it still remains a small portion of the overall transportation budget.
Though state and local sources of funding for transportation projects exist, federal support is perhaps the most critical to creating, improving, and maintaining pedestrian infrastructure and facilities (Cradock, 2009). The use of funds for pedestrian projects and research has not been studied extensively, yet several trends in practice have been revealed through what has been researched. Counties with lower educational attainment, low populations, higher poverty levels or a greater proportion of households owning two or more cars were less likely to implement a pedestrian-related project. Additionally, among counties that implemented a project, low educational attainment was associated with lower levels of per capita funding. This gap presents an issue of equity and safety, as individuals in areas with low socioeconomic levels are more likely to need to walk to destinations safely and could benefit most from pedestrian-related projects (Cradock, 2009).
Additional funding for pedestrian projects has increased in recent years with SAFETEA-LU, the stimulus bill, SRTS, and TIGER. However, though this has increased nonmotorized investments, studies show states have not taken full advantage of available funding. Nationally, only 80.4 percent of the $9.4 billion available through the TE program since 1992 has been utilized, and only 35 percent of the SRTS program funding available has been obligated since 2005 (Ernst & Shoup, 2009). While this is important to note because it leaves possible projects and improvements unrealized, it also makes the programs liable for federal rescission requirements, where states return any unspent funds. In FY 2008, states returned over $98.5 million through rescissions, equal to a 12 percent reduction in 2008 TE apportionments (Ernst & Shoup, 2009).
The forthcoming CLEAN-TEA bill targets and promotes clean and alternative forms of transportation, such as walking and bicycling, by providing additional funding from future cap-and-trade programs. It is still unclear to what extent communities will take advantage of this new opportunity, or how they will modify their budgets to make better use of already available funding sources. Studies and the upcoming funding options suggest there should be a greater focus on providing adequate education and opportunity for communities so they are aware of what funding options are available, especially areas that are less likely to implement a project. This would minimize funding concerns, a major barrier to implementing projects. Also, more efforts should be taken to document the patterns of federal funding for pedestrian improvements so that trends and gaps in practice may be recognized and addressed.