Accident Analysis and Prevention 49 (2012) 404409

Contents lists available at SciVerse ScienceDirect

Accident Analysis and Prevention

journal homepage: www.elsevier.com/locate/aap

Risk factors for severe injury in cyclists involved in trafc crashes in Victoria,
Australia
Souane Boufous , Liz de Rome, Teresa Senserrick, Rebecca Ivers
The George Institute for Global Health, The University of Sydney, Sydney, Australia

a r t i c l e

i n f o

Article history:
Received 23 June 2011
Received in revised form 28 February 2012
Accepted 6 March 2012
Keywords:
Cyclist
Crashes
Injury
Severity

a b s t r a c t
This study examines the impact of cyclist, road and crash characteristics on the injury severity of cyclists
involved in trafc crashes reported to the police in Victoria, Australia between 2004 and 2008. Logistic
regression analysis was carried out to identify predictors of severe injury (serious injury and fatality) in
cyclist crashes reported to the police. There were 6432 cyclist crashes reported to the police in Victoria
between 2004 and 2008 with 2181 (33.9%) resulting in severe injury of the cyclist involved. The multivariate analysis found that factors that increase the risk of severe injury in cyclists involved in trafc
crashes were age (50 years and older), not wearing a helmet, riding in the dark on unlit roads, riding on
roads zoned 70 km/h or above, on curved sections of the road, in rural locations and being involved in
head-on collisions as well as off path crashes, which include losing control of vehicle, and on path crashes
which include striking the door of a parked vehicle. While this study did not test effectiveness of preventative measures, policy makers should consider implementation of programs that address these risk
factors including helmet programs and environmental modications such as speed reduction on roads
that are frequented by cyclists.
2012 Elsevier Ltd. All rights reserved.

1. Introduction
Cycling is becoming an increasingly popular mode of transport
with an estimated 800 million bicycles in use worldwide, twice
the number of cars (Peden et al., 2004). For example in Melbourne,
the second largest metropolitan city in Australia, the proportion of
cyclist journeys to work doubled between 2001 and 2006 (Victorian
Department of Transport, 2009). This increase in popularity may
partly be due to the increased promotion of cycling as a way to
improve health outcomes and reduce road congestion and environmental pollution (Andersen et al., 2000; Cooper et al., 2008;
Maibacha et al., 2009; Heinen et al., 2010), as well as to the other
benets to cyclists, such as reduced costs and ease of parking. However, in addition to the broad health benets to the individual and
the community ascribed to cycling, the activity also carries with it
the risk of serious injury and even death. It is estimated that the
risk of death while riding a bicycle is 12 times higher than when
driving a car (Pucher and Dijkstra, 2003).
The increase in cycling and the associated injury and fatality resulted in a substantial growth in the eld of cycling safety
research over the last two decades. Most previous research has
focused on quantifying the magnitude of the problem in terms of

Corresponding author. Tel.: +61 2 9657 0314; fax: +61 2 9657 0301.
E-mail address: sboufous@george.org.au (S. Boufous).
0001-4575/$ see front matter 2012 Elsevier Ltd. All rights reserved.
doi:10.1016/j.aap.2012.03.011

providing overall crash and injury rates as well as trends in the frequency and rates of cyclist crashes (Davidson, 2005; Kloss et al.,
2006; Meuleners et al., 2007; Tin et al., 2010; Boufous et al., 2011).
Other studies have examined the characteristics and risk factors of
cyclist crashes (Wachtel and Lewiston, 1994; Stutts and Hunter,
1999; Adbdel-Aty et al., 2007). These include demographic and
behavioural characteristics of riders (age, gender, socioeconomic
status, alcohol use and bicycle helmet usage), road characteristics
(road type, road geometry, speed limit, cycling infrastructure and
rural or urban location) and crash characteristics (time and day of
the crash, other vehicle involved, type of crash and environmental
conditions). However, there is limited research that systematically
examines the impact of these factors on severity of bicycle related
injury. Previous studies in this area were limited and focused only
on crashes resulting from collisions between motor vehicles and
cyclists or on collisions occurring on two-lane roads (Klop and
Khattak, 1999; Kim et al., 2007). Another study focused on factors that contributed to the severity of cyclist crashes in relation
to fatalities (Bl et al., 2010).
Given the increase in cycling and associated crash related injury
and disability, understanding factors that contribute to the severity
of injury is important in order to implement appropriate preventative programs. In this study, we investigate whether characteristics
related to the cyclist (age, gender, and bicycle helmet usage), the
road (road type, road geometry, speed limit, rural or urban location)
and the crash (time and day of the crash, other vehicle involved,

S. Boufous et al. / Accident Analysis and Prevention 49 (2012) 404409

type of crash) were associated with injury severity of cyclists

involved in trafc crashes. We examined these associations using
police reported crash data from Victoria, Australia.

2. Materials and methods

The state of Victoria has a population of over 5 million people,
accounting for 24% of the Australian population. Most Victorians
(3.8 million) are concentrated in the large metropolitan city of
Melbourne (Australian Bureau of Statistics, 2009).
Police reported cycling-related crashes were obtained from
VicRoads (Road Authority of Victoria) for a ve-year period:
between January 2004 and December 2008. Crashes reportable to
police are those that occurred on public roads where at least one
person was killed or injured and the crash was attributable to vehicle movement. Police data do not include crashes that occur on
private property, in car parks or similar off-road environments, but
include those that occur on footpaths if they are part of the road
reserve. A road reserve is dened as the entire width between abutting property boundaries open to the public for travel purposes.
Ethics approval for access and usage of data from these sources was
obtained from The University of Sydney Human Research Ethics
Committee.
Descriptive statistics were computed, including the proportion
of various categories in the sample and the proportion of those
with severe injury across various cyclist, road and crash characteristics. Logistic regression analysis was undertaken to examine the
impact of cyclist, road and crash characteristics on the severity of
injury in cyclists involved in crashes. A severe injury is dened as
a crash-related death or injury that required, according to police
reports, the crash victim to be transported to a hospital. Independent variables examined were selected if they were shown to have
some relationship with the outcome variable (severe injury) in univariate analyses (p < 0.2) or when there was some evidence in the
literature pointing to a possible relationship. Independent variables considered at this level included individual characteristics
of the victim: age, gender and helmet use; crash characteristics:
time and day of crash, light condition at time of crash, type of
other vehicles involved and type of crash. Other independent variables included in police reports were related to the road condition
at the time of crash, including road location (rural, urban), speed
zone, road geometry (intersection), road curvature, road condition
(paved, unpaved, etc.) and surface condition (dry, wet, etc.).
Backward stepwise regression was then used to determine the
factors that contributed to injury severity that is, we started
with a full model and variables were eliminated from the model
in an iterative process. The nal model, which contained only independent variables that signicantly contributed to injury severity,
was reached when no more variables could be eliminated. We also
examined interactions, one at a time, between a number of variables including age, gender, helmet use, light condition, crash type,
speed limit, location of the crash and road curvature but none of
the interactions were signicant. All analyses in the report were
carried out using SAS, 9.1 (SAS, 2006).

3. Results
There were 6432 cyclist crashes reported to the police in Victoria
between 2004 and 2008 with 2181 (33.9%) resulting in severe injury
of the cyclist. Tables 13 show cyclist, crash and road characteristics
along with the proportion of cyclists that were severely injured in
crashes. More than three quarters of cyclists involved in crashes
were male (76.6%) with 62.3% aged between 20 and 49 years and
74.1% reporting the use of a helmet at the time of the crash.

405

Table 1
Characteristics of cyclists in police-recorded crashes in Victoria, 20042008.
n

% in the sample

% with severe injury

Sex
Male
Female
Unknown

4925
1381
126

76.6
21.5
2.0

34.9
32.2
15.1

Age group
09
1019
2029
3039
4049
5059
60+
Unknown

161
1116
1460
1525
1021
574
363
212

2.5
17.4
22.7
23.7
15.9
8.9
5.6
3.3

25.5
29.7
31.2
35.2
36.6
40.4
44.9
22.6

Helmet use
Yes
No
Unknown

4768
649
1015

74.1
10.1
15.8

33.8
40.2
30.3

Total

6432

100

33.9

The vast majority of cyclist crashes (78.4%) occurred during

weekdays. Morning and afternoon peak hours between 6 am to
10 am and 2 pm to 6 pm were the periods with the highest proportion of cyclist crashes (30.5% and 32.1% respectively). Overall,
three quarters of cyclist crashes occurred in day light and one in
ten occurred during dusk/dawn. Another 11.1% of crashes occurred
in the dark in areas with street lights and 2.5% occurred in the dark
on unlit streets.
The majority of cyclist crashes reported to the police (90%)
involved another vehicle. The majority (83.7%) involved car-type
vehicles (cars, SUVs, utility and light trucks), 3.6% involved heavy
vehicles and 1.4% involved other cyclists. Single vehicle crashes
made up 6.4% of cyclist crashes reported to the police. Grouped road
user movements according to DCAs showed that the most frequent
groups of cyclist crashes involved vehicles travelling in the same
direction (22.7%), including rear end and side swipe crashes, followed by those involving vehicles travelling in adjacent directions
at intersections (21.8%), The third most frequent group involved
manoeuvring type crashes (21.1%), particularly as a result of cyclists
emerging from footpaths or other vehicles emerging from pathways. It is important to note that adjacent directions is a term
commonly used in police crash reports to refer to instances when
a vehicle crosses the path of another vehicle, often at a right angle,
at an intersection.
The majority of cyclist crashes occurred in urban areas (94.6%)
and about half (48.1%) on roads with speed limits of 60 km/h with
over one-third (34.5%) occurring on roads with speed limits of
4050 km/h. Another 11.4% occurred on roads with speed limits between 70 and 90 km/h and 6% on roads with speed limit of
100 km/h and over. Most cyclist crashes occurred at an intersection (58.5%). The most common intersection types for crashes were
T-intersection (28.9%) and cross intersection (27.7%). Almost all
cyclist crashes occurred on straight sections of the road (98.7%) and
on paved roads (96.3%). Nearly nine in ten cyclist crashes occurred
on dry road surfaces (88.7%) with another 7.9% occurring on wet or
muddy roads.
The multivariate analysis showed that a number of cyclist characteristics (age, helmet use) as well as crash and road characteristics
(crash type, light condition, location of the crash, road speed limits
and road curvature) were independently and signicantly associated with the severity of injury in cyclists involved in trafc crashes
(Table 4).
Overall, injury severity of cyclists involved in trafc crashes
increased with age, with those aged 60 years and over twice as likely

406

S. Boufous et al. / Accident Analysis and Prevention 49 (2012) 404409

Table 2
Characteristics of police-recorded cyclist crashes in Victoria, 20042008.
n

% in the sample

% with severe injury

Day of the crash

Week day
Weekend

5045
1387

78.4
21.6

33.6
35.1

Time of the crash

6:0110:00
10:0114:00
14:0118:00
18:0122:00
22:010:00
0:016:00
Unknown

1962
1190
2062
956
151
102
9

30.5
18.5
32.1
14.9
2.3
1.6
0.1

33.4
32.5
32.6
36.3
49.0
39.2
55.6

Light condition at the time of the crash

Day
Dusk/dawn
Dark with streets lights on
Dark with no streets lights or street lights off
Unknown

4824
687
714
160
47

75.0
10.7
11.1
2.5
0.7

32.5
33.9
40.2
49.4
34.0

Type of other vehicles involved

No other vehicle
Car-type vehicles
Heavy vehicles
Bicycles
Others (include motorcycles)
Unknown

409
5385
229
89
86
234

6.4
83.7
3.6
1.4
1.3
3.6

43.0
32.4
50.2
46.1
40.7
29.9

1404
811
1458
1360

21.8
12.6
22.7
21.1

33.2
40.1
31.5
29.6

713
498
188

11.1
7.7
2.9

36.7
42.4
29.8

6432

100.0

33.9

Crash type
Vehicles from adjacent directions (intersections only)
Vehicles from opposite directions
Vehicles from the same direction
Manoeuvring (include emerging from footpath, or
other vehicles emerging from driveway)
On path (include striking the door of parked vehicle)
Off path (include out of control on carriageway)
Other (include pedestrian, overtaking, etc.)
Total

to be severely injured as a result of a crash as the youngest age group

(09 years). Those aged 5059 years were 72% more likely to be
severely injured as a result of a crash than the youngest age group.
The analysis also showed that not wearing a helmet increased the
odds of severe injury by 56%. Compared to crashes involving vehicles from the same direction, those involving vehicles travelling
in opposite directions as well as off path crashes which include
losing control of vehicle, and on path crashes which include striking door of a parked vehicle, were more likely to result in cyclist
severe injury. Compared to daytime, crashes that occurred in the
dark even with street lights on increased the risk of severe injury by
40%. Cyclists who crashed in dark areas with no street lights were
almost twice as likely to be severely injured compared to those who
crashed in day light.
The likelihood of being severely injured increased with increasing road speed limits with the risk of severe injury 51% higher when
crashes occurred on roads with speed limits of 100 km/h or more
compared to those occurring on roads with 4050 km/h speed limits. Cyclist crashes that occurred on curved sections of the road were
86% more likely to lead to severe cyclist injuries compared to those
that occurred on straight sections of the road. Compared to crashes
occurring in urban areas, those occurring in rural areas were 28%
more likely to result in severe injury.
4. Discussion
This study found that factors that increase the risk of severe
injury in cyclists involved in trafc crashes were older age (50 years
and older), not wearing a helmet, riding in the dark on unlit roads,
riding on roads zoned 70 km/h or above, on curved road sections

of the road, in rural locations and being involved in crashes with

vehicles travelling in opposite directions, in off path crashes often
resulting from loss of control of the vehicle involved in the crash,
and on path crashes which include striking the door of a parked
vehicle.
While almost two thirds of cyclists involved in trafc crashes
in Victoria during the study period were aged 2049 years, injury
severity increased with age, with cyclists aged 50 and over twice as
likely to be severely injured as a result of a crash than the youngest
age group. Similar results were found in a study that examined factors inuencing injury severity of cyclists involved in collisions with
motor vehicles on two-lane undivided roads in the USA (Klop and
Khattak, 1999). Older cyclists are particularly vulnerable to severe
injury mainly due their overall physical fragility often compounded
by pre-existing conditions (Anstey et al., 2005).
Not wearing a helmet increased the risk of severe injury in
cyclists involved in police-recorded trafc crashes by 56%. Nonusage of helmets has been consistently shown to lead to serious
injury, particularly head injuries, in cyclists (Povey et al., 1999;
Richter et al., 2007; Hansen et al., 2003; Amoros et al., 2012). A
Cochrane Review of ve case controlled studies from different
countries concluded that cycle helmets decreased the risk of head
and brain injury by between 63% and 88%, decreased the risk of
facial injury by 65%, and that helmets seemed equally effective in
reducing injuries in crashes involving motor vehicles as well as
cyclist only crashes (Thompson et al., 2000). While the majority
of cyclists involved in crashes had been wearing a helmet (74.1%),
this proportion was only 57.1% in children aged under 10 years
and 60.2% among adolescents aged 10 to 19 years (results not
shown) despite the existence of compulsory helmet wearing laws

S. Boufous et al. / Accident Analysis and Prevention 49 (2012) 404409

Table 3
Road characteristics of police-recorded cyclist crashes in Victoria, 20042008.
n

% in the sample

% with severe injury

407

Table 4
Logistic regression analysis of factors contributing to the severitya of injury of
cyclists involved in police-recorded crashes in Victoria, 20042008.
Adjusted ORc 95 CI

Road location
Urban
Rural
Unknown

6083
316
33

94.6
4.9
0.5

33.3
46.5
21.2

Speed zone
4050
60
7090
100+

2219
3092
735
386

34.5
48.1
11.4
6.0

30.7
34.0
37.6
44.6

Intersection
Not at intersection
Cross intersection
T-intersection
Y-intersection
Multiple intersections
Unknown

2651
1784
1860
21
105
11

41.2
27.7
28.9
0.3
1.6
0.2

34.4
32.0
34.9
19.0
40.0
27.3

Road curvature
Straight
Curve
Unknown

6348
82
2

98.7
1.3
0.1

33.7
48.8
50.0

Road condition
Paved
Unpaved
Gravel
Not known

6192
24
119
97

96.3
0.4
1.9
1.5

33.8
54.2
40.3
29.9

Surface condition
Dry
Wet/muddy
Unknown

5702
508
220

88.7
7.9
3.4

34.0
34.6
29.1

Total

6432

100.0

33.9

in Victoria. This indicates the need for further promotion and

enforcement of helmet use, particularly among children. Evidence
points to the effectiveness of a combination of enforcement and
education programs designed to increase the use of bicycle helmets
in children, particularly those that are community or school based
and those that involve distribution of free helmets (Karkhaneh
et al., 2006; Royal et al., 2005; Hall et al., 2004).
While most cyclist crashes occurred during daytime, the multivariate analysis found riding at night, to be a predictor of injury
severity. The impact of night riding on injury severity was found
to be even greater in the absence of street lights. This was supported by other studies that found cyclist crashes resulting from
collisions with other vehicles and occurring at night, particularly
in areas without effective street lighting, to have a higher injury
severity risk compared to those occurring during the day (Klop and
Khattak, 1999; Kim et al., 2007). Another study that examined all
cyclist crashes reported to the police in the USA found that while
the vast majority of cyclist crashes occurred during the day, 35%
of those resulting in cyclist deaths occurred after dark (Rodgers,
1995). Dark and poor lighting conditions clearly limit visibility and
increase the risk of a crash but also of injury severity due to reduced
reaction time and ability to take evasive action for those involved
in the crash leading to greater impact and higher levels of severity
(Kim et al., 2007).
It is important to note that no information was available in
the data used in this study on illumination equipment of bicycles or high conspicuity clothing for cyclists that have the potential
to increase visibility, particularly in the dark (Hagel et al., 2007).
A systematic review of visibility aids for pedestrians and cyclists
found that uorescent materials in yellow, red and orange improve
driver detection during the day; while lamps, ashing lights and
retro-reective materials in red and yellow improve visibility and
recognition at night (Kwan and Mapstone, 2006). While visibility

Age group of cyclist

09b
1019
2029
3039
4049
5059
60+

1
1.11
1.13
1.37
1.44
1.72
2.14

0.75
0.77
0.93
0.97
1.14
1.40

Helmet use
Yesb
No

1
1.56

1.30 1.87

Light condition
Dayb
Dusk/dawn
Dark with streets lights on
Dark with no streets lights or street lights off

1
1.08
1.40
1.92

0.91 1.28
1.19 1.66
1.39 2.66

Crash type
Vehicles from the same directionb
Vehicles from opposite directions
Vehicles from adjacent directions (intersections only)
Manoeuvring
On path (include striking door of parked vehicle)
Off path (include out of control on carriageway)
Other (include pedestrian, overtaking, etc.)

1
1.53
1.11
1.03
1.38
1.60
0.89

1.28
0.95
0.86
1.14
1.29
0.63

Speed limit (km/h)

40-50b
60
70-90
100+

1
1.13
1.29
1.51

1.00 1.27
1.08 1.55
1.16 1.97

Location of the crash

Urbanb
Rural

1
1.28

1.00 1.71

Road curvature
Straightb
Curve

1
1.86

1.17 2.97

a
b
c

1.63
1.67
2.03
2.14
2.59
3.28

1.84
1.31
1.23
1.67
1.99
1.27

Dependent variable: severe vs not severe injury.

Reference category.
Adjusted for all other factors in the model.

measures, including high conspicuity clothing, may help drivers see

cyclists, more research is needed to determine whether increased
rider visibility reduces the risk of crash and severe injury in cyclists.
The current study also found that injury severity of cyclists
involved in trafc crashes increased proportionally with road
speed limits. High speed limits have also previously been found to
increase the risk of cyclist crashes as well as serious injury and fatality (Adbdel-Aty et al., 2007; Stone and Broughton, 2003; Bl et al.,
2010). These ndings suggest the importance of separating bicyclists from high-speed trafc, for example separate bicycle paths
on roadways that have a speed limit of over 60 km/h. It also lends
support to low speed limit in residential neighbourhoods with signicant bicycle trafc (Kim et al., 2007).
While a higher proportion of cycle crashes occur in urban
areas of Victoria, those that occur in rural areas are more likely
to lead to severe injury. The increased injury severity of cyclists
involved in crashes in rural areas compared to urban areas is likely
to be partly due to higher travel speeds of all vehicles in rural
areas. Other factors that might explain the higher severity level
in cyclists injured in trafc crashes in rural areas include greater
alcohol levels among various road users involved in crashes, lack
of appropriate cycling infrastructure, as well as greater time delays
for emergency services to provide medical intervention to cyclists
when crashes occur (Meuleners et al., 2007).

408

S. Boufous et al. / Accident Analysis and Prevention 49 (2012) 404409

Compared to straight roads, curved roads were found to significantly increase the likelihood of severe injury in cyclists involved
in trafc crashes by 86%. Similar results were found in the study of
cyclist crashes resulting from collisions with other vehicles by Kim
et al. (2007) who argued that the ndings might be explained by
reduced visibility and ability to control the bicycle or the vehicle.
The study also found that compared to crashes involving vehicles travelling in the same direction, those involving vehicles
travelling in opposite directions, which include head-on type
crashes, and off path crashes, such as loss of control of the bicycle
or other vehicle involved, signicantly increased the risk of severe
injury in cyclists. Head-on collisions were found to increase the
probability of cyclist fatal injury (Kim et al., 2007), as this type
of crash often involves high speed and greater impact between
the cyclist and the other vehicle. The loss of control of the bicycle
has been found to be associated with poor road surface, including
debris, cracked or pot holed surface and other hazardous obstructions (Turner et al., 2009), and indicate the need for maintenance of
roads in areas frequently used by cyclists. This may be particularly
important on high speed roads where cyclists are travelling at high
speeds and any small objects on the road way may have potentially
serious consequences.
While 58% of cyclist crashes occurred at intersections and previous studies indicated that cyclists are more likely to crash at
intersections, mainly due to the failure of drivers of other vehicles
to see the cyclist approaching (Herslund and Jorgensen, 2003), this
study showed that intersections did not increase the risk of severe
injury in cyclists involved in trafc crashes. This may possibly be
explained by the fact that both cyclists and other vehicles tend to
slow down while approaching intersections resulting in less severe
injuries in the event of a crash.
It is important to note that relevant behavioural factors (i.e.
alcohol use) and road characteristics (i.e., the presence of cycling
lanes and paths) that can potentially have an impact on injury
severity in cyclist crashes (Noland and Quddus, 2004; Kim et al.,
2007; Reynolds et al., 2009), could not be explored in the multivariate analysis because of the lack of or the incompleteness of
relevant information in the police data. In addition, cyclist crashes,
particularly single vehicle crashes involving only cyclists, tend to
be under-reported in police crash data (Lujic et al., 2008; Langley
et al., 2009). Therefore, there is a need for well designed population
based studies (i.e., case control) that include all cyclist crashes, and
not only those reported in routinely collected data sources such as
police and hospital records, to examine risk factors such as alcohol
use, high conspicuity clothing and road infrastructure (cycle lanes).

5. Conclusions
This study examined the associations between cyclist, vehicle and road characteristics and the severity of injury in cyclists
involved in police reported crashes in Victoria, Australia between
2004 and 2008. The analysis found that factors that increase the
risk of severe injury in cyclists involved in trafc crashes were age
of 50 years and older, not wearing a helmet, riding in the dark on
unlit roads, riding on roads zoned 70 km/h or above, on curved sections of the road, in rural locations and being involved in head-on
collisions as well as off path crashes, which include losing control
of vehicle, and on path crashes which include striking the door of a
parked vehicle.
The ndings indicate that in order to reduce the severity of
injury in cyclists involved in trafc crashes, there is a need for
further promotion and enforcement of helmet use, particularly
among younger cyclists. Environmental modications also need
to be considered and assessed, including the separation of cyclists
using bicycle paths on roadways that have a speed limit of over

60 km/h and the introduction of low speed limits in residential

neighbourhoods with signicant bicycle trafc. Other possible
environmental modications may include better lighting and
modication of curved sections in terms of curve length, degree
of curvature and placement of street furniture (i.e.; poles), particularly on roads frequently used by cyclists. Further studies are
needed to examine the effectiveness of conspicuous clothing and
illumination equipment of bicycles, as well as those that explore
road and bicycle lane congurations that allow for safer sharing of
the road with other road users.
Acknowledgments
This study was funded by VicRoads (Victorias road authority). The authors wish to thank Maria Ali for her assistance with
the project and VicRoads staff, particularly Barry Scott, for their
support and assistance with access to the data. Souane Boufous,
Teresa Senserrick and Rebecca Ivers receive salary funding from the
National Health and Medical Research Council of Australia. Liz De
Rome is supported by the NRMA-ACT Road Safety Trust.
References
Adbdel-Aty, M., Chundi, S.S., Lee, C.L., 2007. Geo-spatial and log-linear analysis of
pedestrian and bicyclist crashes. Journal of Safety Research 38, 571579.
Amoros, E., Chiron, M., Martin, J., Thelot, B., Laumon, B., 2012. Bicycle helmet wearing
and the risk of head, face, and neck injury: a French case-control study based on
a road trauma registry. Injury Prevention 18, 2732.
Andersen, L.B., Schnohr, P., Schroll, M., Hein, H.O., 2000. All-cause mortality associated with physical activity during leisure time, work, sports and cycling to work.
Archives of Internal Medicine 160, 16211628.
Anstey, K.J., Wood, J., Lord, S., Walker, J.G., 2005. Cognitive, sensory and physical
factors enabling driving safety in older adults. Clinical Psychology Review 25,
4565.
Australian Bureau of Statistics, 2009. Population by Age and Sex, Australian States
and Territories. ABS, Canberra.
Bl, M., Blova, M., Mllerb, I., 2010. Critical factors in fatal collisions of adult cyclists
with automobiles. Accident Analysis and Prevention 42, 16321636.
Boufous, S., De Rome, L., Senserrick, T., Ivers, R.Q., 2011. Cycling crashes in children,
adolescents and adults a comparative analysis. Trafc Injury Prevention 12 (3),
244250.
Cooper, A.R., Wedderkopp, N., Jago, R., et al., 2008. Longitudinal associations of
cycling to school with adolescent tness. Preventive Medicine 47, 3243288.
Davidson, J.A., 2005. Epidemiology and outcome of bicycle injuries presenting to an
emergency department in the United Kingdom. European Journal of Emergency
Medicine 12 (1), 2429.
Hagel, B.E., Lamy, A., Rizkallah, J.W., Belton, K.L., Jhangri, G.S., Cherry, N., Rowe, B.H.,
2007. The prevalence and reliability of visibility aid and other risk factor data
for uninjured cyclists and pedestrians in Edmonton, Alberta, Canada. Accident
Analysis and Prevention 39 (2), 284289.
Hall, M., Cross, D., Howat, P., Stevenson, M., Shaw, T., 2004. Evaluation of a schoolbased peer leader bicycle helmet intervention. International Journal of Injury
Control and Safety Promotion 11 (3), 165174.
Hansen, K.S., Engester, L.B., Viste, A., 2003. Protective effect of different types of
bicycle helmets. Trafc Injury Prevention 4 (4), 285290.
Heinen, E., Van Wee, B., Maat, A., 2010. Commuting by Bicycle: An Overview of the
Literature. Transport Reviews 30 (1), 5996.
Herslund, M.B., Jorgensen, N.O., 2003. Looked-but-failed-to-see-errors in trafc.
Accident Analysis and Prevention 35 (6), 885891.
Karkhaneh, M., Kalenga, J.C., Hage, B.E., Rowe, B.H., 2006. Effectiveness of bicycle
helmet legislation to increase helmet use: a systematic review. Injury Prevention
12, 7682.
Kim, J.K., Kim, S., Ulfarsson, G.F., Porrello, L.A., 2007. Bicyclist injury severities
in bicyclemotor vehicle accidents. Accident Analysis and Prevention 39 (2),
238251.
Klop, J.R., Khattak, A.J., 1999. Factors inuencing bicycle crash severity on two-lane,
undivided roadways in North Carolina. Transportation Research Record 1674,
7885.
Kloss, F.R., Tuli, T., Haechl, O., Gassner, R., 2006. Trauma injuries sustained by cyclists.
Journal of Trauma 8, 7784.
Kwan, I., Mapstone, J., 2006. Interventions for increasing pedestrian and cyclist visibility for the prevention of death and injuries. Cochrane Database of Systematic
Reviews 4, CD003438.
Langley, J.D., Dow, N., Stephenson, S., Kypri, K., 2009. Missing cyclists. Injury Prevention 15, 1318.
Lujic, S., Finch, C., Boufous, S., Hayen, A., Dunsmuir, W., 2008. How comparable are
road trafc crash cases in hospital admissions data and police records? An examination of data linkage rates. Australian and New Zealand Journal of Public Health
32 (1), 2833.

S. Boufous et al. / Accident Analysis and Prevention 49 (2012) 404409

Maibacha, E., Steg, L., Anablec, J., 2009. Promoting physical activity and reducing climate change: opportunities to replace short car trips with active transportation.
Preventive Medicine 49 (4), 326327.
Meuleners, L.B., Lee, A.H., Haworth, C., 2007. Road environment, crash type and hospitalisation of bicyclists and motorcyclists presented to emergency departments
in Western Australia. Accident Analysis and Prevention 39 (6), 12221225.
Noland, R.B., Quddus, M.A., 2004. An analyses of pedestrian and bicycle casualties
using regional panel data. Transportation Research Record 1897, 2833.
Peden, M., Scureld, R., Sleet, D., et al., 2004. World Report on Road Trafc. World
Health Organization, Geneva.
Povey, L.J., Frith, W.J., Graham, P.G., 1999. Cycle helmet effectiveness in New Zealand.
Accident Analysis and Prevention 31 (6), 763770.
Pucher, J., Dijkstra, L., 2003. Promoting safe walking and cycling to improve public
health: lessons from the Netherlands and Germany. American Journal of Public
Health 93, 15091516.
Reynolds, C.C., Harris, M.A., Teschke, K., Cripton, P.A., Winters, M., 2009. The impact
of transportation infrastructure on bicycling injuries and crashes: a review of
the literature. Environmental Health 8, 47.
Richter, M., Otte, D., Haasper, C., et al., 2007. The current injury situation of bicyclists
a medical and technical crash analysis. Journal of Trauma 62 (5), 11181122.
Rodgers, G.B., 1995. Bicyclist deaths and fatality risk patterns. Accident Analysis and
Prevention 27 (2), 215223.

409

Royal, S., Kendrick, D., Coleman, T., 2005. Non-legislative interventions for the promotion of cycle helmet wearing by children. Cochrane Database of Systematic
Reviews 2, CD003985.
SAS Institute, 2006. SAS: statistical software. Version 9.1. SAS Institute Inc., Cary
(NC).
Stone, M., Broughton, J., 2003. Getting off your bike: cycling accidents in Great Britain
in 19901999. Accident Analysis and Prevention 35 (4), 549556.
Stutts, J.C., Hunter, W.W., 1999. Motor vehicle and roadway factors in pedestrian
and bicyclist injuries: an examination based on emergency department data.
Accident Analysis and Prevention 31 (5), 505514.
Tin, S.T., Woodward, A., Ameratunga, S., 2010. Injuries to pedal cyclists on New
Zealand roads, 19882007. BMC Public Health 10, 655.
Thompson, D.C., Rivara, F.P., Thompson, R., 2000. Helmets for preventing head
and facial injuries in bicyclists. Cochrane Database of Systematic Reviews 2,
CD001855.
Turner, S., Binder, S., Roozenburg, A., 2009. Cycle safety: reducing the crash risk.
New Zealand transport agency research report 389. Wellington, NZ Transport
Agency.
Victorian Department of Transport, 2009. The Victorian cycling strategy. DOT, Melbourne.
Wachtel, A., Lewiston, D., 1994. Risk factors for bicyclemotor vehicle collisions at
intersections. Institute of Transportation Engineers Journal 64 (9), 3035.