See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/263440264

Promoting Safe Walking and Cycling to Improve Public Health: Lessons From
The Netherlands and Germany

Article  in  American Journal of Public Health · October 2003

DOI: 10.2105/AJPH.93.9.1509

CITATIONS READS

597 735

2 authors:

John Pucher Lewis Dijkstra

Rutgers, The State University of New Jersey European Commission
141 PUBLICATIONS   7,914 CITATIONS    26 PUBLICATIONS   1,715 CITATIONS   

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

The Politics of Implementing Sustainable Transport Policies View project

Regional coordination of public transport View project

All content following this page was uploaded by John Pucher on 26 December 2016.

The user has requested enhancement of the downloaded file.

  PUBLIC HEALTH MATTERS 

Promoting Safe Walking and Cycling to Improve

Public Health: Lessons From The Netherlands and Germany
| John Pucher, PhD, and Lewis Dijkstra, PhD

implemented a wide range of policies over

Objectives. We examined the public health consequences of unsafe and inconvenient
the past 2 decades that have simultaneously
walking and bicycling conditions in American cities to suggest improvements based on
successful policies in The Netherlands and Germany. encouraged walking and cycling while dra-
Methods. Secondary data from national travel and crash surveys were used to com- matically lowering pedestrian and bicyclist fa-
pute fatality trends from 1975 to 2001 and fatality and injury rates for pedestrians and talities and injuries and keeping auto use at
cyclists in The Netherlands, Germany, and the United States in 2000. only half the American level. The Nether-
Results. American pedestrians and cyclists were much more likely to be killed or in- lands and Germany provide valuable lessons
jured than were Dutch and German pedestrians and cyclists, both on a per-trip and on for integrating more physical exercise into the
a per-kilometer basis. lives of Americans.
Conclusions. A wide range of measures are available to improve the safety of walk- This article first examines variations in
ing and cycling in American cities, both to reduce fatalities and injuries and to encour-
walking and cycling levels among North
age walking and cycling. (Am J Public Health. 2003;93:1509–1516)
American and Western European countries
and then focuses on The Netherlands, Ger-
Improving conditions for walking and bicy- exercise they need.3–7 Similarly, the US sur- many, and the United States in particular. We
cling in our cities is vital for America’s pub- geon general specifically recommends more examine differences in travel behavior, fatal-
lic health. The measures described in this ar- walking and cycling for practical, daily travel ity and injury rates, and trends over time.
ticle would not only reduce pedestrian and as an ideal approach to raising physical activ- Most importantly, we describe the 6 cate-
cycling fatalities and injuries but also allow ity levels.8 gories of policies in The Netherlands and
millions of people, many of them danger- Even in the sprawling metropolitan areas Germany that have made walking and cycling
ously overweight, to bike or walk for some of the United States, 41% of all trips in 2001 such safe and attractive alternatives to driv-
of their short trips and thus obtain healthful were shorter than 2 miles, and 28% were ing: better facilities for walking and cycling,
exercise in the course of daily life. More shorter than 1 mile.9 Bicycling can easily urban design sensitive to the needs of nonmo-
walking and cycling would yield further pub- cover distances of up to 2 miles, and most torists, the traffic calming of residential neigh-
lic health benefits by reducing the use of au- people can walk at least a mile.10 Yet Ameri- borhoods, restrictions on motor vehicle use in
tomobiles, thus diminishing air and noise cans use their cars for 66% of all trips up to cities, rigorous traffic education of both mo-
pollution and the overall level of traffic a mile long and for 89% of all trips between torists and nonmotorists, and strict enforce-
danger. 1 and 2 miles long.9 Clearly, there is enor- ment of traffic regulations protecting pedestri-
The United States is gripped by a worsen- mous potential for increased walking and cy- ans and bicyclists.
ing epidemic of obesity. Nationwide surveys cling over these shorter trip distances.
based on self-reported weight and height indi- There are 2 problems with proposals to in- DATA AND METHODS
cate an increase in obesity from 12% of crease walking and cycling: their current dan-
adults in 1991 to 20% in 2000.1 Estimates ger and inconvenience in most American We relied on secondary sources for all the
of obesity based on clinical measurements of cities. As documented in this article, walking data series discussed in this article. They are
weight and height are considerably higher, in- and cycling in the United States are much the official national sources of statistics on
dicating that in 2000, 31% of the adult popu- more dangerous than car travel, both on a travel behavior and traffic accidents in each
lation was obese (body mass index [BMI] ≥ per-trip and per-mile basis. Moreover, the lack country. For the United States, the data for
30) and 64% was overweight (BMI ≥ 25).2 of proper pedestrian and bicycling facilities travel behavior came from the 1995 Nation-
Many studies suggest that lack of physical ex- makes walking and cycling not only unsafe wide Personal Transportation Survey and the
ercise is one important reason for the alarm- but also inconvenient, slow, unpleasant, and 2001 National Household Travel Survey,
ing trend toward increased obesity. Several unfeasible in most places. both conducted by the US Department of
articles and editorials in the leading medical The good news presented in this article is Transportation (Federal Highway Adminis-
and public health journals have explicitly ad- that it is indeed possible to achieve safe and tration).11 The data on traffic fatalities also
vocated more walking and cycling for daily convenient walking and cycling conditions, as came from the US Department of Trans-
travel as the most affordable, feasible, and de- demonstrated by the experience of Germany portation (National Highway Traffic Safety
pendable way for people to get the additional and The Netherlands. Those 2 countries have Administration),12 while the injury data came

September 2003, Vol 93, No. 9 | American Journal of Public Health Pucher and Dijkstra | Peer Reviewed | Public Health Matters | 1509
  PUBLIC HEALTH MATTERS 

from the Centers for Disease Control and 50

Prevention (CDC).13
45
For Germany, the data on travel behavior Bicycling 28

came from the German Ministry of Trans- 40 Walking 20

10
port14,15 and the German Institute of Eco- 35 9
nomic Research.16 The German fatality and 4 4 10 12

Percentage
30
injury data came from the Federal Statistical
25
Office17 and the Federal Traffic Institute.18
The data for The Netherlands came from 20
4
Statistics Netherlands19 and the Dutch Min- 15 2 28 29
istry of Transport.20 Each of these surveys 10 1
24 24 24 22 21
18
and other data collection procedures relied 12
5 10
on extensive underlying methodologies that 6
cannot be discussed here. Interested readers 0

United States

Canada

England and

France

Italy

Switzerland

Germany

Austria

Sweden

Denmark

The Netherlands
can consult any of the individual sources for

Wales
detailed information. We note in the text and
figures the specific sources and any important
differences among the countries in definition
Country
or methods.
Some of these data series are more compa- Note. Modal split distributions for different countries are not fully comparable owing to differences in trip definitions, survey
methodologies, and urban area boundaries. The distributions given here are intended to show the approximate differences
rable across countries than others. The travel among countries and should not be used for exact comparisons.
surveys measuring usage of different means Source. Transportation Research Board,29 Table 2–2, p. 30.
of transportation rely on basically the same
FIGURE 1—Percentage of trips in urban areas made by walking and bicycling in North
definitions of transport modes but use varying
America and Europe, 1995.
methodologies for sampling and trip measure-
ment. The data on traffic fatalities are quite
reliable. Studies indicate that roughly 95% of
all traffic fatalities are reported to the police VARIATION AMONG COUNTRIES Perhaps even more striking are the large
and thus appear in official records.21 More- IN LEVELS OF WALKING differences in travel behavior between coun-
over, all the countries that we examined de- AND CYCLING tries as their populations get older. As shown
fine traffic fatalities as occurring within 30 in Figure 2, walking increases with age in
days of the crash. Unfortunately, trends in travel behavior in both The Netherlands and Germany, while
Traffic injury data are far less comparable. the United States could hardly be worse for cycling falls off only slightly. Indeed, the
Underreporting of pedestrian and cyclist in- public health. The journey-to-work section of Dutch and Germans who are 75 and older
juries is a problem in all countries. For the the US Census indicates that the percentage make roughly half their trips by foot or bike,
United States, we used the CDC injury esti- of all work trips made by walking fell from compared with only 6% of Americans aged
mates from WISQARS, which are based on a 10.3% in 1960 to only 2.9% in 2000.23 In- 65 and older. While cycling is almost nonex-
representative survey of injuries reported by cluding all trip purposes, the Nationwide Per- istent among the American elderly, it ac-
hospital emergency rooms.13 Even those esti- sonal Transportation Survey shows that the counts for a fourth of all trips made by the
mates underreport total injuries since they ex- percentage of urban trips made by walking Dutch elderly and for 7% of trips made by
clude minor injures not requiring a hospital and cycling fell from 10.0% in 1977 to only the German elderly. Equally stunning, walk-
visit. The Dutch and German injury estimates 6.3% in 1995, which is far lower than in ing accounts for 48% of trips by Germans
are based on police reports. One study esti- most other countries.9,10 Figure 1 shows the aged 75 and older and 24% of trips made by
mated that Dutch police reports captured percentage of all urban trips made in 1995 Dutch aged 75 and older. This not only pro-
only 15% of minor injuries to pedestrians and by walking and cycling in the United States, vides them with valuable physical exercise
cyclists but 60% of all injuries requiring a Canada, and 9 European countries.24 Even but also ensures them a level of mobility and
hospital visit.22 German police reports cap- Canada has almost twice the percentage of independence that greatly enhances their
tured 16% of minor injuries but 48% of in- walk and bike trips as in the United States. In quality of life. It also may contribute to both
juries requiring a hospital visit.21 Thus, the most European countries, at least a fourth of the longer life expectancy and the longer
Dutch and German estimates of injuries urban trips are made by walking or cycling, healthy life expectancy in the Netherlands
should be roughly doubled to make them and a few countries—like Denmark and The and Germany—2 years longer than in the
comparable to the hospital-based injury esti- Netherlands—report a nonmotorized travel United States.25 As the Dutch and German
mates of the CDC. rate of over 40%. examples clearly show, the physical and men-

1510 | Public Health Matters | Peer Reviewed | Pucher and Dijkstra American Journal of Public Health | September 2003, Vol 93, No. 9
  PUBLIC HEALTH MATTERS 

60 ient conditions faced by pedestrians and bi-

cyclists in most American cities. As shown in
Bicycling 7
the next section, the perceived risk of walk-
50 The Netherlands
Walking ing and cycling in American cities is based
Germany 11 on real dangers. Even without dramatic
40 changes in American land-use and trans-
24
portation systems, much could be done in
Percentage

25
the short term to improve walking and cy-
30 cling conditions to make them both safer and
9 30 22
48
19 more attractive.
10
20 United States 39
DANGERS OF WALKING AND
24
CYCLING IN THE UNITED STATES
23
10 1 19
0.2 17
0.5 13 14
0.3 12 It is much more dangerous to walk or
7 5 6
4 cycle in American cities than to travel by
0
16–24 25–39 40–64 ≥65 18–44 45–64 65–74 ≥75 18–24 25–39 40–64 65–74 ≥75 car. Per kilometer traveled, pedestrians were
Age in years 23 times more likely to get killed than car
occupants in 2001 (140 vs 6 fatalities per
Source. US Department of Transportation,11 German Ministry of Transport,14,15 and Statistics Netherlands.19
billion kilometers), while bicyclists were 12
FIGURE 2—Percentage of trips in urban areas made by walking and bicycling in the United times more likely than car occupants to get
States, Germany, and The Netherlands, by age group, 1995. killed (72 vs 6 fatalities per billion kilome-
ters).30 Walking and cycling in American
cities are much more dangerous than in
tal limitations that come with aging are not cities cannot be attributed mainly to long many other countries. As shown in Figure 3,
the main impediments to walking and cycling trip distances.28 Indeed, if distance were the nonmotorist fatality rates in the United
by the American elderly. overriding factor, one might expect more cy- States are much higher than in The Nether-
For both the elderly and the nonelderly, cling than walking in American cities, since lands and Germany. Per kilometer and per
walking and cycling are discouraged in the cycling covers longer distances faster and trip walked, American pedestrians are
United States by longer trip distances, the low easier. In fact, walk trips outnumber bike roughly 3 times more likely to get killed
cost and ease of auto ownership and use, and trips 6 to 1. than German pedestrians and over 6 times
a range of other public policies that make The much higher cost of auto ownership more likely than Dutch pedestrians. Per kilo-
walking and cycling inconvenient, unpleasant, and use in Europe also helps explain the meter and per trip cycled, American bicy-
and, above all, unsafe. higher levels of walking and cycling there. clists are twice as likely to get killed as Ger-
The more compact land-use patterns in High taxes on gasoline and new cars, as well man cyclists and over 3 times as likely as
European cities lead to average trip dis- as higher prices for parking, make the overall Dutch cyclists.
tances that are only about half as long as in cost of auto use at least double what it is in Because of the unreliability of injury data
American cities and thus are easier to cover the United States.29 In addition, roadway and in all countries, it is far more difficult to
by foot or by bike.26 As explained in previ- parking facilities are much more limited than compare differences in pedestrian and cyclist
ous studies27 and by other articles in this in American cities. From a political perspec- injury rates. Nevertheless, they also appear
issue, planning for more compact, mixed-use tive, it has been very difficult to raise taxes on to be much higher in the United States than
development in American cities would en- auto ownership and use in the United States, in The Netherlands or Germany. The CDC
hance the feasibility of walking and cycling even slightly, let alone to the dramatically data based on hospital reports capture a
by reducing trip distances to likely destina- higher levels in Europe. With over 95% of all larger percentage of total injuries than the
tions. However, that is a long-term approach parking free of charge, and with gasoline Dutch and German injury data, which are
that will take many years to implement, if it taxes, roadway tolls, licensing fees, and vehi- based on police reports. As noted earlier,
can be adopted at all. Moreover, as noted cle taxes among the lowest in the developed studies indicate that the Dutch and German
earlier, 41% of all urban trips in the United world, the United States makes driving a car police reports capture only about half of all
States are already shorter than 2 miles, and almost irresistible.29 That, in turn, discourages serious injuries requiring hospitalization.21,22
28% are shorter than 1 mile. The potential walking and cycling. Thus, the Dutch and German injury rates
for more walking and cycling already exists. Clearly, however, one of the biggest im- shown in Figure 3 should be roughly dou-
Thus, the extraordinarily low 6% of trips pediments to more walking and cycling is the bled to make them more comparable to the
made by walking or cycling in American appallingly unsafe, unpleasant, and inconven- CDC rates for the United States. Even after

September 2003, Vol 93, No. 9 | American Journal of Public Health Pucher and Dijkstra | Peer Reviewed | Public Health Matters | 1511
  PUBLIC HEALTH MATTERS 

30 such an upward adjustment, American

pedestrians are about twice as likely to get
25 injured as German pedestrians and 4 times
25 Pedestrian fatality rates per 100 million trips as likely as Dutch pedestrians. American cy-
Bicyclist fatality rates per 100 million trips clists are at even greater risk: they are 8
21 Pedestrian fatality rates per 100 million km traveled
times more likely to get injured than Ger-
Bicyclist fatality rates per 100 million km traveled
20 man cyclists and about 30 times more likely
Pedestrian injury rates per 500 000 km traveled

17 Bicyclist injury rates per 500 000 km traveled

than Dutch cyclists.
Some good news to offset that bad news is
15 14 that a great deal could be done to make walk-
ing and cycling safer in the United States.
Germany and The Netherlands, for example,
10
8.2 have drastically cut the number of pedestrian
7.2
and bicyclist deaths over the past 25 years by
5.2
5 4.4 implementing a wide range of policies to im-
3.2
2.1
2.5 prove safety. Figure 4 shows that from 1975
2.0
1.6 1.6
0.7 1.1
0.3 0.4
to 2001, total pedestrian fatalities declined
0 by 82% in Germany and by 73% in The
United States Germany The Netherlands Netherlands. Over the same period, cyclist fa-
Source. US Department of Transportation11,12; Centers for Disease Control and Prevention13; German Institute of Economic talities declined by 64% in Germany and by
Research16; German Federal Statistical Office17; German Federal Traffic Institute18; Statistics Netherlands9; and Dutch Ministry 57% in The Netherlands. The drop in cyclist
for Transport, Public Works and Water Management.20 fatalities in Germany is especially impressive
FIGURE 3—Pedestrian and bicycling fatality rates and nonfatal injury rates in the United because it came during a boom in cycling
States, Germany, and The Netherlands, 2000. there, with a doubling in the number of bike
trips and 50% growth in the share of total
trips made by bike.10 By contrast, the 27%
fall in cyclist fatalities in the United States was
due almost entirely to the sharp decline in cy-
120
cling by children.31,32

100 HOW TO MAKE WALKING AND

CYCLING SAFER

80 However dangerous walking and cycling

currently are in the United States, it is defi-
Percentage

nitely possible, and essential, to make them

60
much safer. As shown by the wide range of
coordinated policies in The Netherlands33,34
and Germany,35 the necessary techniques
40
Pedestrians USA
and programs already exist and have been
Pedestrians Germany proven to work extremely well. They include
Pedestrians Netherlands
20 Bicyclists USA better facilities for walking and cycling, traf-
Bicyclists Germany fic calming of residential neighborhoods,
Bicyclists Netherlands
urban design sensitive to the needs of non-
0 motorists, restrictions on motor vehicle use
1975 1980 1985 1990 1995 2000 in cities, rigorous traffic education of both
Year motorists and nonmotorists, and strict en-
forcement of traffic regulations protecting
Source. US Department of Transportation12; German Federal Statistical Office17; German Federal Traffic Institute18; Statistics pedestrians and bicyclists. American cities
Netherlands19; and Dutch Ministry for Transport, Public Works and Water Management.20
lack only the political will to adopt the same
FIGURE 4—Trends in pedestrian and bicycling fatalities in the United States, Germany, and strategies.
The Netherlands, 1975–2001 (1975 = 100%). Owing to space limitations, we can only
briefly summarize here the 6 categories of

1512 | Public Health Matters | Peer Reviewed | Pucher and Dijkstra American Journal of Public Health | September 2003, Vol 93, No. 9
  PUBLIC HEALTH MATTERS 

public policy measure implemented in The Traffic Calming of Residential almost always include other uses such as cul-
Netherlands and Germany. For detailed de- Neighborhoods tural centers, shopping, and service establish-
scriptions and illustrations of the Dutch and Traffic calming limits the speeds of motor ments that can easily be reached by foot or
German measures, readers can consult a vehicle traffic, both by law—30 km per hour bike. Both residential and commercial devel-
range of publications about walking and cy- (19 mph) or less—and through physical barri- opments have sidewalks and bicycle paths to
cling in Europe.10,26,36–39 ers such as raised intersections and cross- serve nonmotorists. Parking lots almost never
walks, traffic circles, road narrowing, zigzag surround buildings, as in the United States;
Better Facilities for Walking routes, curves, speed humps, and artificial instead, they are built next to or behind
and Cycling dead ends created by midblock street clo- buildings, thus permitting easy access to
One emphasis of Dutch40 and German35 sures.10 Traffic calming gives pedestrians, bi- pedestrians and bicyclists. When an obstacle
policy has been to improve the transportation cyclists, and playing children as much right to such as a highway, railroad, or river must be
infrastructure used by pedestrians and bicy- use residential streets as motor vehicles; in- traversed, Dutch and German cities usually
clists. For pedestrians, that has included ex- deed, motor vehicles are required to yield to provide safe and attractive pedestrian and bi-
tensive auto-free zones that cover much of these other users. In both The Netherlands41 cyclist crossings. By comparison, strip malls in
the city center; wide, well-lit sidewalks on and Germany, traffic calming is area-wide and American suburbs are difficult and dangerous
both sides of every street; pedestrian refuge not for isolated streets. That ensures that to reach by foot or bicycle, and most bridges
islands for crossing wide streets; clearly faster through traffic gets displaced to arterial lack provisions for pedestrians and bicyclists.
marked zebra crosswalks, often raised and routes designed to handle it and not simply In the United States, the separation of resi-
with special lighting for visibility; and shifted from one local road to another. dential from commercial land uses increases
pedestrian-activated crossing signals, both The most important safety impact of traf- trip distances and makes the car a necessity.
at intersections and midblock crosswalks. fic calming is the reduced speeds of motor Suburban cul-de-sacs further discourage walk-
Dutch and German cities also have in- vehicles. This is crucial not only to the mo- ing and bicycling by making trips circuitous
vested heavily to expand and improve bicy- torist’s ability to avoid hitting pedestrians and excessively long. Residential roads often
cling facilities. From 1978 to 1996, the and bicyclists but also to the survival of feed directly into high-speed traffic arteries,
Dutch more than doubled the extent of their nonmotorists in a crash. The British Depart- increasing the danger of any trips outside the
already massive network of bike paths and ment of Transport, for example, found that neighborhood. The lack of sidewalks in most
lanes (from 9282 km to 18 948 km). From the risk of pedestrian death in crashes rises American suburbs further exacerbates the
1976 to 1995, the Germans almost tripled from 5% at 20 mph to 45% at 30 mph and problem.
the extent of their bikeway network (from 85% at 40 mph.42
12 911 km to 31 236 km).10 In addition, Area-wide traffic calming in Dutch neigh- Restrictions on Motor Vehicle Use
there are an increasing number of so-called borhoods has reduced traffic accidents by Dutch and German cities restrict auto use
“bicycle streets,” where cars are permitted 20% to 70%.43 Traffic calming in German not only through traffic calming, auto-free
but cyclists have strict right-of-way over the neighborhoods has reduced traffic injuries zones, and dedicated rights-of-way for pedes-
entire breadth of the roadway. Unlike the overall by 20% to 70% and serious traffic trians and cyclists.10,26,29 They also enforce
sparse and fragmented cycling facilities in the injuries by 35% to 56%.44 A comprehensive lower general speed limits for motor vehicles
United States, the bike paths, lanes, and review of traffic calming impacts in Denmark, in cities—usually 50 km per hour (31 mph).
streets in The Netherlands and Germany Great Britain, Germany, and The Nether- Parking is much more limited and more ex-
form a truly coordinated network covering lands found that traffic injuries fell by an av- pensive than in American cities. In addition,
both rural and urban areas. Importantly, erage of 53% in traffic-calmed neighbor- most Dutch and German cities prohibit truck
Dutch and German bikeway systems serve hoods.45 In short, traffic calming greatly traffic and through traffic of any kind in resi-
practical destinations for everyday travel, not reduces the danger of traffic deaths and in- dential neighborhoods. Motor vehicle turn re-
just recreational attractions, as with most juries in residential neighborhoods. Traffic strictions are widespread; moreover, right
bike paths in the United States. calming greatly improves not only pedestrian turns on red are illegal.
The provision of separate rights-of-way is safety but also the safety of bicycling, since
complemented by various other measures: much bike use—especially by children—is in Traffic Education
special bike turn lanes leading directly to in- residential neighborhoods. Driver training for motorists in The
tersections, separate bike traffic signals with Netherlands and Germany is much more ex-
advance green lights for cyclists, bicyclist- Urban Design Oriented to People and tensive, thorough, and expensive than in the
activated traffic signals at key intersections, Not Cars United States.46,47 A crucial aspect of that
and modification of street networks to create New suburban developments in The training in The Netherlands and Germany is
deliberate dead ends and slow, circuitous Netherlands and Germany are designed to the need to pay special attention to avoiding
routing for cars but direct, fast routing for provide safe and convenient pedestrian and collisions with pedestrians and cyclists. Mo-
bikes.10 bicycling access.10 Residential developments torists are required by law to drive in a way

September 2003, Vol 93, No. 9 | American Journal of Public Health Pucher and Dijkstra | Peer Reviewed | Public Health Matters | 1513
  PUBLIC HEALTH MATTERS 

that minimizes the risk of injury for pedestri- torists is far more severe in The Netherlands fighting the current obesity epidemic. Why
ans and cyclists even if they are jaywalking, and Germany than in the United States.45 not try integrating walking and cycling into
cycling in the wrong direction, ignoring traffic the daily travel routines of Americans? That
signals, or otherwise behaving contrary to CONCLUSIONS clearly would be the cheapest, most reliable,
traffic regulations. and most practical way to ensure adequate
Traffic education of children has high pri- The neglect of pedestrian and bicycling levels of physical exercise.
ority in both The Netherlands and Ger- safety has made walking and cycling danger- Walking and cycling also help alleviate traf-
many.46,47 By the age of 10, all schoolchildren ous ways of getting around American cities. fic congestion, save energy, reduce air and
have received extensive instruction on safe Walking and cycling can be made quite safe, noise pollution, conserve land, and produce
walking and bicycling practices. They are however, as clearly shown by the much lower various other environmental benefits. It is the
taught not just the traffic regulations but how fatality and injury rates in The Netherlands broad spectrum of benefits from walking and
to walk and bicycle defensively, to anticipate and Germany. There is no good reason why cycling that explains the widespread public
dangerous situations, and to react appropri- American cities could not adopt many of the support in The Netherlands and Germany for
ately. That sort of safety education is com- same measures to enhance safety. The neces- the impressive range of policies they have
pletely lacking in the United States. sary methods and technology are already adopted to make walking and cycling safer,
available, with decades of successful experi- more convenient, and more pleasant.
Traffic Regulations and Enforcement ence in Europe. The same synergistic benefits have the po-
Traffic regulations in Germany and The It is important to package safety-enhancing tential for energizing a broad coalition of
Netherlands strongly favor pedestrians and programs in a way that dramatizes their bene- groups in the United States to advocate better
bicyclists. Even in cases where an accident re- fits to everyone. The most obvious benefit walking and cycling conditions in American
sults from illegal moves by pedestrians or cy- would be the reduced risk of death and injury cities. Public health experts should be work-
clists, the motorist is almost always found to from walking and cycling. The safety issue ing together with bicyclist and pedestrian ad-
be at least partly at fault. When the accident must be brought home to Americans by pub- vocates, traffic engineers, urban planners,
involves children or the elderly, the motorist lic campaigns emphasizing the direct impacts environmentalists, architects and private de-
is usually found to be entirely at fault. In al- on individuals, their families, and their velopers, community leaders, and government
most every case, the police and the courts friends. Improved safety also would encour- officials at all levels. The public health com-
find that motorists should anticipate unsafe age more people to walk and cycle on a regu- munity probably has the most potential to en-
and illegal walking and cycling. lar basis, providing them with valuable exer- courage the necessary changes at the grass-
In addition, German and Dutch police are cise, mobility options, independence, and roots level. Unless individual Americans can
far stricter in ticketing motorists, pedestrians, even fun. be convinced that they will directly benefit
and cyclists who violate traffic regulations. The European countries with the highest from better walking and cycling conditions,
Thus, walking against the light is not allowed levels of walking and cycling have much politicians are unlikely to support the neces-
in any German city and can easily result in a lower rates of obesity, diabetes, and hyperten- sary policies. Self-interest is likely to be the
ticket and fine. Likewise, cyclists caught riding sion than the United States.25,48 The Nether- strongest motivation to effect changes in
in the wrong direction, running red lights, lands, Denmark, and Sweden, for example, travel behavior. Getting enough physical exer-
making illegal turns, or riding at night without have obesity rates only a third of the Ameri- cise is quite literally a matter of life or death.
functioning lights can expect at least a warn- can rate, while Germany’s rate is only half as Health care professionals must convince their
ing notice and possibly a ticket and fine. high.48 Moreover, the average healthy life ex- patients that walking and cycling on a regular
The most significant contrast with the pectancies in those 4 European countries are basis for daily travel will help them live
United States is the much stricter enforce- 2.5 to 4.4 years longer than in the United longer and healthier lives.
ment of traffic regulations for motorists in States,25 although their per capita health ex- Of course, the public health community
Germany and The Netherlands. Penalties can penditures are only half those of the United cannot do it alone. Transportation profession-
be high even for minor violations. Not stop- States.49 als, urban planners, architects, and private de-
ping for pedestrians at crosswalks is consid- Of course, many factors affect differences velopers must provide the improvements in
ered a serious offense and motorists can get between Europe and the United States. Nev- walking and cycling conditions so desperately
ticketed for noncompliance, even if pedestri- ertheless, the dramatically higher levels of needed to reduce the dangers of walking and
ans are only waiting at the curb and not actu- walking and cycling for daily travel certainly cycling in American cities. Those efforts will
ally in the crosswalk.10 Similarly, red traffic contribute to better public health in countries require the support of local, state, and federal
signals are strictly enforced, and some inter- such as The Netherlands, Denmark, Ger- government officials. Public policymakers at
sections in German and Dutch cities have many, and Sweden. Repeated waves of fad all levels must not only provide the necessary
cameras that automatically photograph cars diets, rising memberships in health clubs, ex- funding for better bicycling and pedestrian fa-
running red lights and stop signs.37 Finally, ercise equipment in more homes, diet pills, cilities but also adopt and implement a range
the punishment for traffic violations by mo- and liposuction have all been total failures in of policies to encourage more compact,

1514 | Public Health Matters | Peer Reviewed | Pucher and Dijkstra American Journal of Public Health | September 2003, Vol 93, No. 9
  PUBLIC HEALTH MATTERS 

mixed-use development that naturally permits ing walking and cycling conditions, thus ex- Endnotes
and encourages walking and cycling as a part plicitly fostering the necessary partnership 1. A. H. Mokdad, B. A. Bowman, E. S. Ford, F. Vivi-
cor, J. S. Marks, and J. P. Koplan, “The Continuing Epi-
of daily life. and teamwork. demics of Obesity and Diabetes in the United States,”
If for no other reason than their large However admirable these initial efforts JAMA 286 (2001): 1195–1200.
numbers and extensive network of contacts, are, they remain exceedingly modest com- 2. K. Flegal, M. Carroll, C. Ogden, and C. Johnson,
public health experts have a crucial role to pared with the enormity of the problem. “Prevalence and Trends in Obesity Among Adults,
1999–2000,” JAMA 288 (2002): 1723–1727.
play in mobilizing political support for the Public health organizations should publicize
3. C. Dora, “A Different Route to Health: Implica-
necessary policy changes. At the very least, far more widely the worsening obesity epi- tions of Transport Policies,” British Medical Journal 318
they should publicize more prominently the demic in the United States as the national (1999): 1686–1689.
disastrous public health consequences of an crisis it is. They need to mount massive 4. J. Koplan and W. Dietz, “Caloric Imbalance and
automobile-dependent transportation system media campaigns to encourage and improve Public Health Policy,” JAMA 282 (1999): 1579–1581.

and a land-use pattern that make walking the conditions for walking and cycling. Only 5. D. Carnall, “Cycling and Health Promotion,” Brit-
ish Medical Journal 320 (2000): 888.
and cycling dangerous, inconvenient, un- when the public and politicians become
6. S. P. Wolff and C. J. Gilham, “Public Health Versus
pleasant, and, in some cases, impossible. fully aware of the severity of the obesity Public Policy? An Appraisal of British Urban Transport
In fact, the public health community has problem—and the huge potential of walking Policy,” Public Health 105 (1991): 217–228.
already begun developing programs and part- and cycling to mitigate the problem—will 7. M. Hillman, “Health Promotion: The Potential of
nerships to achieve more walkable and bik- public policies change enough to make a Non-Motorized Transport,” in Health at the Crossroads:
Transport Policy and Urban Health, ed. T. Fletcher and
able communities that encourage higher real difference. A. J. McMichael (London: Wiley and Sons, 1997):
levels of physical exercise. The CDC, for ex- Some studies predict that obesity will soon 177–186.
ample, has developed the Active Community overtake smoking as the most important 8. Physical Activity and Health: A Report of the Sur-
Environments (ACE) program, a multidiscipli- cause of premature death in the United geon General (Atlanta, Ga: Centers for Disease Control
and Prevention, 1996); Healthy People 2010: Under-
nary initiative to promote walking and cy- States.53–55 It is time for the public health standing and Improving Health, 2d ed. (Washington,
cling through better urban design, transporta- community to undertake as vigorous a cam- DC: US Department of Health and Human Services,
tion, and land-use policies.50 The federal paign to promote more physical exercise and November 2000).

CDC program links up with state health de- improved diet as the decades-long campaign 9. J. Pucher and J. Renne, “Socioeconomics of Urban
Travel: Evidence From the 2001 NHTS,” Transporta-
partments to encourage similar efforts at the against smoking. tion Quarterly 57 (2003): 49–77; based on data from
state and local level. For example, the ACE 2001 National Household Travel Survey (Washington,
program in the California Department of DC: US Department of Transportation, Federal High-
way Administration, 2003). Article also available at
Human Services funds organizations such as About the Authors http://nhts.ornl.gov/2001/Documents/Pucher-Renne-
California Walks, the California Bicycle Coali- John Pucher is with the Bloustein School of Planning and TQarticle.pdf. Accessed July 18, 2003. These calcula-
tion, the Local Government Commission, and Public Policy, Rutgers University, New Brunswick, NJ. tions are for urban areas only, but they include all trip
Lewis Dijkstra is with the European Commission, Brussels, purposes and all means of transportation. The survey
the Rails-to-Trails Conservancy to encourage Belgium. description and detailed data are available at: http://
local community policies that promote walk- Requests for reprints should be sent to John Pucher, nhts.ornl.gov/2001/index.shtml. Accessed December
ing and cycling.51 California’s ACE also coor- PhD, Bloustein School of Planning and Public Policy, Rut- 15, 2002.
gers University, 33 Livingston Ave, Room 363, New 10. J. Pucher and L. Dijkstra, “Making Walking and
dinates the state’s Safe Routes to School ini- Brunswick, NJ 08901-1900 (e-mail: [email protected]. Cycling Safer: Lessons From Europe,” Transportation
tiative by providing technical assistance and edu). Quarterly 54 (2000): 25–50. A version without photos
funding community-based projects that pro- This article was accepted April 16, 2003. is available as a pdf file at: http://www.vtpi.org/
mote walking to school. puchertq.pdf. Accessed December 15, 2002.

Complementing such government initia- 11. 1995 Nationwide Personal Transportation Survey
Contributors (Washington, DC: Federal Highway Administration,
tives, the Robert Wood Johnson Foundation J. Pucher had primary responsibility for the study de- 1997); 2001 National Household Travel Survey (Wash-
has already spent $84 million financing a sign, data collection for Germany and the United States, ington, DC: Federal Highway Administration, 2003).
and writing the text of the article. L. Dijkstra shared in
family of Active Living programs, all of which the study design, collected data for The Netherlands, 12. Traffic Safety Fact Sheets (Washington, DC: US De-
are intended to increase physical activity.52 and produced the graphics. partment of Transportation, National Highway Traffic
Safety Administration, annual). Available at: http://
Active Living by Design, for example, is fund- www-nrd.nhtsa.dot.gov/pdf/nrd-30/NCSA/TSF2001/
ing 25 communities throughout the country 2001pedestrian.pdf and http://www-nrd.nhtsa.dot.gov/
Acknowledgments pdf/nrd-30/NCSA/TSF2001/2001pedal.pdf. Accessed
to promote changes in urban design, architec- The research for this article was funded partially December 15, 2002.
ture, land use, and transportation that encour- through a grant from the Robert Wood Johnson Foun-
dation to Rutgers University (research project: “Chronic 13. WISQARS: Web-Based Injury Statistics Query and
age more walking and cycling. Active for Life Reporting System (Online) (Atlanta, Ga: National Center
Disease and the Built Environment,” September
focuses on incorporating increased exercise 2001–April 2002). for Injury Prevention and Control, annual). Available
into the daily life of adults. The foundation’s The authors thank Michael Greenberg, Peter Jacob- at: http://www.cdc.gov/ncipc/wisqars. Accessed De-
sen, Catherine Staunton, Anne Seeley, John Renne, cember 15, 2002.
Active Living Network is intended to inte-
Charles Komanoff, Andy Clarke, Charlie Zegeer, Rich 14. W. Broeg and E. Erl, Einflussgrössen der Personen-
grate the public health agenda into a wide Killingsworth, Jim Sallis, and Martin Wachs for their mobilität (Berlin: German Ministry of Transport, Build-
range of other professions crucial to improv- many helpful suggestions on improving the article. ing, and Housing, 2002).

September 2003, Vol 93, No. 9 | American Journal of Public Health Pucher and Dijkstra | Peer Reviewed | Public Health Matters | 1515
  PUBLIC HEALTH MATTERS 

15. W. Broeg and E. Erl, Kenngrössen für Fussgänger- walk trips in American cities (8.6% in 2001 compared 40. Sign up for the Bike—Design Manual for a Cycle-
und Fahrradverkehr (Berlin: German Ministry of Trans- with 5.4% in 1995), but the increase is solely due to a Friendly Infrastructure (Ede, The Netherlands: Informa-
port, Building, and Housing, 1999). very important change in survey methodology, which tion and Technology Centre for Transport and Infra-
16. German Institute of Economic Research, Verkehr in introduced multiple prompts by survey interviewers to structure, 1998).
Zahlen (Berlin: German Ministry of Transport, annual). capture all possible walk trips. Thus, the 2001 walk 41. Traffic Calming in the Netherlands (Ede, The
share figure for the United States is not comparable Netherlands: Information and Technology Centre for
17. Central German Statistical Office (Statistisches with earlier NPTS surveys in 1977, 1983, 1990, and
Bundesamt), Stuttgart, special disaggregate tabulations Transport and Infrastructure, 1998).
1995. For full details of both the earlier NPTS surveys
of population, travel, and accidents done for the au- 42. Mean Streets: Pedestrian Safety and Reform of the
and the 2001 NHTS survey, as well as detailed analy-
thors for western and eastern portions of Germany. Nation’s Transportation Law (Washington, DC: Surface
sis of socioeconomic variations and time trends, see
Transportation Policy Project, 1997), Table 4, p. 4.
18. Federal Traffic Institute (Bundesanstalt für Pucher and Renne, “Socioeconomics of Urban Travel.”
Strassenwesen), IRTAD: International Road Traffic Acci- For that reason, and also because we have only 1995 43. J. Kraay and A. Dijkstra, Safety Aspects of Urban
dent Database (Bergisch-Gladbach, Germany: Federal walk share data for most other European countries, we Infrastructure (Leidschendam, The Netherlands: Insti-
Traffic Institute, annual). Available at: http://www.bast. have used the 1995 modal split shares for interna- tute for Road Safety Research, 1989).
de/htdocs/fachthemen/irtad/english/englisch.html. Ac- tional comparisons in this article. 44. C. Hass-Klau, Civilized Streets: A Guide to Traffic
cessed December 15, 2002. 29. Transportation Research Board, Making Transit Calming, Environment, and Transport Planning
19. Personenvervoer over de weg (Voorburg/Heerlen: Work: Insight from Western Europe, Canada, and the (Brighton, England: Brighton Press, 1992).
Statistics Netherlands, annual). Available at: http:// United States (Washington, DC: National Academy 45. B. Preston, “Cost-Effective Ways to Make Walking
www.cbs.nl. Accessed December 15, 2002. Press, 2001). Safer for Children and Adolescents,” Injury Prevention
20. Dutch Ministry of Transport, Public Works, and 30. The authors calculated fatality rates based on offi- 1 (1995): 187–190.
Water Management, Nederland-Slachtoffers naar cial mode-by-mode fatality and travel statistics of the US 46. German Traffic Safety Council, Kind und Verkehr:
vervoerwijze en aard letsel (Rotterdam/Heerlen: AVV Department of Transportation (National Highway Traffic Nationales Programm des Deutschen Verkehrssicherheit-
Transport Research Center, quarterly). Available at: Safety Administration and Federal Highway Administra- srat zur Verkehrserziehung in Deutschland (Berlin:
http://www.rws-avv.nl. Accessed December 15, 2002. tion). The total numbers of car occupant, cyclist, and Deutscher Verkehrssicherheitsrat, 2001). Available at:
pedestrian fatalities were derived from National High- http://dvr.de/asp/dms.asp?url=programme/kind/
21. International Road Traffic and Accident Database,
way Traffic Safety Administration, Traffic Safety Facts inhalt.htm. Accessed December 15, 2002.
Underreporting of Road Traffic Accidents Recorded by the
2001 (Washington, DC: US Department of Transporta-
Police: An International Comparison (Bergisch-Gladbach, 47. Kultusministerium, “Beschluss der Kultusminist-
tion, 2003). The total number of passenger kilometers
Germany: Organization for Economic Cooperation and serkonferenz vom 17.6.1994,” Zeitschrift für
of travel by private motor vehicle, bicycle, and foot were
Development, Road Research Transport Program, No- Verkehrserziehung 45 (1995): 4–6.
calculated for the authors by Mary Ann Keyes of the
vember 1994). 48. World Health Organization, Obesity in Europe:
Federal Highway Administration using unpublished data
22. Dutch Institute for Road Safety Research, Victims, from the 2001 National Household Travel Survey. The Case for Action (London: International Obesity
Road Accident Victims: General Developments, Official Taskforce of the World Health Organization, Septem-
31. N. Templin, “The Bicycle Loses Ground as a Sym-
Statistics and the Real Numbers (Leidschendam, The ber 2002). Available at: http://www.iotf.org/media/
bol of Childhood Liberty,” Wall Street Journal (Septem-
Netherlands, 1997). Available at: http://www.swov.nl/ globalprev.htm. Accessed December 15, 2002.
ber 10, 1996), A1.
en/kennisbank/index.htm. Accessed December 15, 49. Total Expenditures on Health per Capita,
2002. 32. J. Pucher, C. Komanoff, and P. Schimek, “Bicycling
1960–2000, in US Dollars, Purchasing Power Parity
Renaissance in North America? Recent Trends and Al-
23. Pucher and Renne, “Socioeconomics of Urban (Washington, DC: Organization for Economic Coopera-
ternative Policies to Promote Bicycling,” Transportation
Travel,” Table 1. Available at http://nhts.ornl.gov/ tion and Development, 2002). Available at: http://
Research A 33 1999: 625–654. Available at: http://
2001/Documents/Pucher-Renne-TQarticle.pdf. Ac- www.oecd.org/xls/M00031000/M00031378.xls. Ac-
policy.rutgers.edu/papers, accessed December 15,
cessed August 10, 2003. cessed December 15, 2002.
2002.
24. We used the 1995 Nationwide Personal Trans- 50. Centers for Disease Prevention and Control, Active
33. P. Van Vliet and G. Schermers, Sustainable Safety:
portation Survey (NPTS) modal split figures here for Community Environments (Atlanta, Ga: National Center
A New Approach for Road Safety in the Netherlands (Rot-
the United States even though 2001 National House- for Chronic Disease Prevention and Health Promotion,
terdam: Ministry of Transport, Public Works and Water
hold Travel Survey (NHTS) statistics are available. 2003). Available at: http://www.cdc.gov/nccdphp/
Management; Traffic Research Centre, 2000). Available
Comparable 2000 or 2001 travel survey data are not dnpa/aces.htm. Accessed December 15, 2002.
at:http://www.rws-avv.nl. Accessed December 15, 2002.
yet available for the other countries shown in Figure 1. 51. Physical Activity and Health Initiative (Sacra-
Moreover, the 2001 walk share reported by the NHTS 34. National Traffic and Transport Plan (The Hague: mento: Department of Health and Human Services,
is not even comparable with the earlier NPTS surveys Ministry of Transport, Public Works and Water Man- 2002). Available at: http://www.dhs.ca.gov/cdic/
of American walking levels, because 2001 survey agement, 2000). Available at: http://www.minvenw.nl. cdcb/epidemiology/Physical_Activity/index.htm. Ac-
methodology was drastically changed to capture more Accessed December 15, 2002. cessed December 15, 2002.
walk trips, definitely exaggerating any real increase in 35. Nationaler Radverkehrsplan 2002–2012, FahrRad! 52. Active Living Resource Center (Princeton, NJ: Rob-
walking between 1996 and 2001. Massnahmen zür Förderung des Radverkehrs in Deutsch- ert Wood Johnson Foundation, 2003). Available at:
25. Healthy life expectancy refers to the expected land. Berlin: Bundesministerium für Verkehr, Bau- und http://www.rwjf.org/programs/physicalActivity.jsp. Ac-
years of life without a major disability or illness that Wohnungswesen, April 2002). Available at: http:// cessed December 15, 2002.
significantly impairs the quality of life. C. Dathers, Esti- www.bmvbw.de/Fahrradverkehr-.423.htm. Accessed
December 15, 2002. 53. J. M. McGinnis and W.H. Foege, “Actual Causes of
mates of Healthy Life Expectancy for 191 Countries in the
Death in the United States,” JAMA 270 (1993):
Year 2000: Methods and Results (Geneva, Switzerland: 36. V. Vuchic, Transportation for Livable Cities (New 2207–2212.
World Health Organization, October 2001), Annex Brunswick, NJ: Center for Urban Policy Research,
Table A; Global Programme on Evidence for Health 1999). 54. A. Peeters, J. J. Barendregt, F. Willekens, J. P. Mack-
Policy Discussions Discussion Paper No. 38. enbach, A. Al Mamum, and L. Bonneux, “Obesity in
37. C. Zeeger, FHWA Study Tour for Pedestrian and Adulthood and Its Consequences for Life Expectancy:
26. P. Newman and J. Kenworthy, Sustainability and Bicyclist Safety in England, Germany, and The Nether- A Life-Table Analysis,” Annals of Internal Medicine 138
Cities: Overcoming Automobile Dependence (Washington, lands (Washington, DC: US Department of Transporta- (2003): 24–32.
DC: Island Press, 1999), 68–125. tion, Federal Highway Administration, 1994).
55. R. Sturm, “The Effects of Obesity, Smoking and
27. R. Ewing and R. Cervero, “Travel and the Built 38. R. Cervero, Transit Metropolis: A Global Inquiry Drinking on Medical Problems and Costs: Obesity Out-
Environment,” Transportation Research Record 1780 (Washington, DC: Island Press, 1998). ranks Both Smoking and Drinking in Its Deleterious Ef-
(2001): 87–114. 39. R. Tolley, ed, Greening of Urban Transport fects on Health and Health Costs,” Health Affairs 21
28. The 2001 NHTS reports a higher percentage of (Chichester, England: Wiley Press, 1997). (2002): 245–253.

1516 | Public Health Matters | Peer Reviewed | Pucher and Dijkstra American Journal of Public Health | September 2003, Vol 93, No. 9

View publication stats