JSR-01335; No of Pages 12

Journal of Safety Research xxx (2016) xxx–xxx

Contents lists available at ScienceDirect

Journal of Safety Research

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

1Q2 Fatal and serious injuries related to vulnerable road users in Canada
2Q3 Ward Vanlaar, a Marisela Mainegra Hing, a,⁎ Steve Brown, a Heather McAteer, a Jennifer Crain, b Steven McFaull b

F
3 a
Traffic Injury Research Foundation, 171 Nepean Street, Ottawa, Ontario K2P 0B4, Canada
4 b
The Public Health Agency of Canada, Injury Section, Centre for Chronic Disease Prevention, 785 Carling Avenue, 7th Floor, AL 6807B, Ottawa, Ontario K1A 0K9, Canada

O
5

6 a r t i c l e i n f o a b s t r a c t

O
7 Article history: Introduction: The goals of this study were to analyze possible trends of fatal and serious injuries related to vulner- 17
8

R
Received 3 July 2015 able road users in Canada (pedestrians, cyclists and motorcyclists) from 1990 to 2012 and the role of alcohol and 18
9 Received in revised form 27 June 2016 drugs in these cases. Drugs have rarely been documented with respect to vulnerable road users. Method: The Traf- 19
10 Accepted 5 July 2016 fic Injury Research Foundation's National Fatality and Serious Injury Databases and the Public Health Agency of 20
11

P
Available online xxxx
Canada's Canadian Hospitals Injury Reporting and Prevention Program databases were used. Numbers and 21
16
15
14
13
12
rates of fatalities and serious injuries among vulnerable road users were analyzed and regression models were 22
38 Keywords:
39 Pedestrians
used to assess changes over time. Results: The analyses show that while the absolute number of fatalities and 23
40 Bicyclists the rate per 100,000 population among vulnerable road users may be decreasing, no such trends are apparent
D 24
41 Motorcyclists when looking at the proportions of these road user fatalities out of all motor-vehicle fatalities. The trend for 25
42 Alcohol the proportion of motorcyclist fatalities is significantly increasing (coef. = 0.16, p b 0.001). The elderly 26
43 27
E
Drugs (76 years or older) are overrepresented among pedestrian fatalities, and serious injuries (they represent 18.5%
of all pedestrian fatalities but only 5.8% of the population), while those 15 years or younger are overrepresented 28
among cyclists (they represent 23.3% of cyclist fatalities but 19.5% of the population), and those 16 to 25 years old 29
T

are overrepresented among motorcyclists (27.2% of motorcyclists fatalities and 13.6% of population). Alcohol and 30
drug use among fatally injured vulnerable road users were significant problems, especially among pedestrians. 31
C

Among fatally injured pedestrians tested for alcohol and drugs, 39.7% and 43.4% tested positive, respectively. Con- 32
clusions: With the promotion of walking and cycling as forms of exercise and the popularity of motorcycling, the 33
safety of vulnerable road users is an important issue. The results corroborate previous research and extend our 34
E

understanding about the influence of alcohol and drugs in vulnerable road user injuries. Practical applications: 35
These findings can help better inform prevention and mitigation initiatives for vulnerable road users. 36
© 2016 Published by Elsevier Ltd. 37
R

47
45
44
R

46
48
Q4 1. Introduction deaths in low-income (57%) and middle-income (51%) countries, than 62
in high-income countries (39%; WHO, 2013). In Canada, in 2013, the 63
O

49 Modern motor vehicles have a variety of safety features that protect share of VRUs among road traffic fatalities was 29.1%, below the WHO 64
50 occupants in the event of a crash. Seatbelts, airbags, and crumple zones reported average for high-income countries. Specifically, 15.6% of all 65
C

51 are among the built-in features designed to increase survivability of ve- road traffic deaths were pedestrians, 10.3% were motorcyclists, and 66
52 hicle occupants during a motor-vehicle collision. In contrast, road users 3.2% were bicyclists according to Transport Canada and the Canadian 67
53 who do not have the benefit of this level of protection, like pedestrians, Council of Motor Transport Administrators (CCMTA, 2015). Although 68
Q5
N

54 bicyclists, and motorcyclists, face potentially devastating consequences the proportion of VRU fatalities in Canada was below the global propor- 69
55 when involved in crashes. Due to the relative safety disadvantages of tion, the 2013 numbers increased in comparison to the percentages of 70
U

56 this group, they are often referred to as vulnerable road users (VRUs). VRUs among all road traffic deaths recorded in 2010. In that year, 71
57 The vulnerability of pedestrians, cyclists, and motorcyclists is dem- 13.7% of all road traffic deaths were pedestrians, 8.6% were motorcy- 72
58 onstrated by their overrepresentation in the total number of road user clists, and 2.8% were bicyclists, according to Transport Canada and 73
59 injuries and fatalities. According to the World Health Organization CCMTA (2015). 74
Q6
60 (WHO), 50% of global fatal road traffic injuries are sustained by VRUs Pedestrians are generally the least protected VRUs and older persons 75
61 (WHO, 2015a). VRUs account for a greater proportion of road traffic are overrepresented among injured pedestrians. Specifically, in Canada 76
during the period of 2004–2008, 35% of fatally injured pedestrians were 77
65 years of age or older, while this age group made up only 13% of the 78
total population (Transport Canada, 2011). Seniors are at a higher risk 79
⁎ Corresponding author at: Traffic Injury Research Foundation, Canada. Tel.: +1 613
238 5235. for a variety of reasons, including their relative physical frailty; natural 80
E-mail address: mariselah@tirf.ca (M.M. Hing). declines in visual and perceptual capacities; and limits to their mobility, 81

http://dx.doi.org/10.1016/j.jsr.2016.07.001
0022-4375/© 2016 Published by Elsevier Ltd.

Please cite this article as: Vanlaar, W., et al., Fatal and serious injuries related to vulnerable road users in Canada, Journal of Safety Research (2016),
http://dx.doi.org/10.1016/j.jsr.2016.07.001
 2 W. Vanlaar et al. / Journal of Safety Research xxx (2016) xxx–xxx

82 which lengthen the time required to cross the road (Robertson & contain toxicological data. Both of these databases contain the “popula- 145
83 Vanlaar, 2008). People aged 56 and older are the least likely to be in- tion” of fatalities and injuries in Canada, rather than a sample. A more 146
84 jured in a bicycle-related crash, but have the worst prognosis in the detailed description of how these data are collected and the procedures 147
85 event of a crash (Kim, Kim, Ulfarsson, & Porrello, 2007). Children and used to create both databases (including definitions) is available in 148
Q8
86 teens under 18 are most often injured in bicycle crashes, particularly Mayhew (2011). 149
87 children aged 10–16, however their prognoses are generally the best
88 (Siman-Tov, Jaffe, Israel Trauma Group, & Peleg, 2012; Kim et al., 2.1.2. Canadian Hospitals Injury Reporting and Prevention Program 150
89 2007). Of all the VRUs, motorcyclists are arguably the most susceptible (CHIRPP) 151
90 to very severe injury in the event of a collision. This is largely due to CHIRPP data from 1990 to 2010 were also analyzed. CHIRPP is an 152
91 the speeds reached by motorcycle riders compared to bicyclists and pe- emergency department based injury surveillance system operated by 153
92 destrians, and their physical proximity to motor vehicles (as opposed to the Public Health Agency of Canada (PHAC), in which there are current- 154
93 bicyclists and pedestrians who may also use bike lanes/paths and side- ly 11 pediatrics and 6 general hospitals participating (although for this 155
94 walks). Motorcyclists have been estimated to be seven times more likely study data from only 4 general hospitals and the 11 pediatrics were 156
95 to crash than drivers of motor vehicles in Alberta (Monk, Buckley, & available). Almost 3 million records have been collected nationally 157

F
96 Dyer, 2009). In New Zealand comparing new motorcycles with new since CHIRPP's establishment in 1990 (Crain et al., in press, Mackenzie 158
97 cars, the odds of fatal or serious injury to a motorcycle rider involved & Pless, 1999). This database does not contain toxicological data. Al- 159

O
98 in an injury crash were almost eight times the odds for a car driver though not population-based, a number of studies have shown 160
99 (Keall & Newstead, 2012). In the United States, the fatality rate per reg- CHIRPP to be representative of Canada in some contexts such as sport 161
100 istered vehicle for motorcyclists in 2013 was 6 times the fatality rate for and recreational injury data in Calgary, when compared to regional 162

O
101 passenger car occupants (NHTSA, 2015). health administrative data (Kang, Hagel, Emery, Senger, & Meeuwisse, 163
102 Alcohol and drug use among VRUs, particularly the involvement of 2013); injuries at Montreal Children's Hospital that were not injuries 164
103 drugs, has not been the subject of many studies. In 2010, 45.8% of fatally resulting in admissions, poisonings, and those presenting overnight 165

R
104 injured pedestrians tested positive for alcohol on Canadian roadways (Macarthur & Pless, 1999a); younger children or children with severe 166
105 (Brown, Vanlaar, & Mayhew, 2013). Thirteen percent of motorcyclists injuries at the Children's Hospital of Eastern Ontario (Macpherson 167

P
106 fatally injured in Canada between 2004 and 2006 had consumed alcohol et al., 2008); and external causes of injury and injury patterns suggest- 168
107 prior to their crash, and this was highest at 23% among the 25–34 age ing priorities for intervention in Canadian youth aged 11–15 (Pickett 169
108
Q7 group (Transport Canada, 2009). Little is known about the influence of et al., 2000).
D 170
109 alcohol in bicycle crashes in Canada (Vanlaar, Marcoux, & Robertson,
110 2009). As a comparison, in 2013 in the United States, 24% of fatally in- 2.1.3. Other data 171
111 jured pedalcyclists (cyclists powered solely by pedals) and 28% of fatally Readily available exposure data for VRUs is limited. In this regard, in- 172
E
112 injured motorcyclists were alcohol-impaired (blood alcohol concentra- formation on population and registered motorcycles was obtained from 173
113 tion (BAC) of 0.08 g/dL or higher; NHTSA, 2015). In 2011 in the United Statistics Canada (Statistics Canada, 2014, 2015) and from the Motorcy- 174
T

114 States, 37% of fatally injured pedestrians had a BAC of 0.08 g/dL or higher cle and Moped Industry Council (MMIC, 2013). To compare data for 175
115 (Eichelberger, Cicchino, & McCartt, 2013). VRUs with data for passenger vehicles (PV: automobile, vans and light 176
C

116 The goal of this paper is to investigate VRU injuries in Canada over an trucks), PV data from the TIRF databases on fatally and seriously injured 177
117 extended period of time (from 1990 until 2012) and to gain an under- were also used. 178
118 standing of possible injury trends among these road users. A previous
E

119 study that focused on off-road vehicles was conducted using similar Ca- 2.2. Data analysis 179
120 nadian data between 1990 and 2010 (Vanlaar et al., 2015). Furthermore,
R

121 this paper seeks to investigate the role of alcohol and drugs in fatal VRU The data analyses used Stata software (StataCorp, 2013) and in- 180
122 crashes. volved trend analysis of the frequencies, rates and proportions of fatal- 181
ities and serious injuries; the characterization of the victims in terms of 182
R

123 2. Methods gender, age and type of VRU; the nature of the serious injuries; and, the 183
level of alcohol or presence of drug found in fatal victims. Note that data 184
O

124 2.1. Data sources from TIRF do not include British Columbia (BC) as data for some years 185
for BC were missing. For the same reason, analysis of the serious injury 186
125 2.1.1. TIRF databases data excludes information from Nunavut (NU) and Newfoundland and 187
C

126 This study involved the analysis of data on VRUs from 1990 to 2012 Labrador (NL). 188
127 contained in two databases managed by the Traffic Injury Research Fatalities and serious injuries per 100,000 population were present- 189
Foundation (TIRF). The first, TIRF's National Fatality Database, includes
N

128 ed as crude rates as well as age-specific rates. Motorcycle fatality and se- 190
129 information on persons fatally injured in motor vehicle collisions in rious injury rates per registered motorcycles were also presented. Linear 191
130 Canada derived from both coroner medical examiner files and police regression models were estimated to assess possible significant trends 192
U

131 collision reports. This database contains toxicological data for fatally in- over time on these rates for each type of VRU and on the proportion of 193
132 jured victims. Toxicological data on alcohol use among victims are ob- VRU fatalities and serious injuries among all fatalities and serious inju- 194
133 tained from files in coroners' and medical examiners' offices. The ries in road crashes. A logistic regression model determined the associ- 195
134 alcohol data are the results of chemical tests performed on body fluid ation of VRU type, sex and age with a fatal outcome among all fatalities 196
135 samples (typically blood). Specific to drug use, the data collected con- and serious injuries. Results were considered significant if correspond- 197
136 tain approximately 500 codes for different drugs and have been consis- ing p-values were below 0.05. Age and gender distribution of VRUs in- 198
137 tently collected since 2000. The drug types can be more broadly volved in fatal and serious injuries are presented for each type of VRU. 199
138 classified into seven groups according to the Drug Classification Evalua- The most common types of serious injuries were analyzed based on 200
139 tion categories (i.e., cannabis, CNS-depressants, CNS-stimulants, narcot- CHIRPP data. Three CHIRPP narrative (free text) fields allowed for a de- 201
140 ic analgesics, hallucinogens, dissociative anesthetics, and inhalants; tailed level of classification and identification of very specific injury cir- 202
141 Brown, Vanlaar, & Robertson, 2015). The second database, the Serious cumstances. Records of injuries sustained by VRUs were identified using 203
142 Injury Database, includes information on seriously injured persons the CHIRPP's Injury Group variable, and an extensive bilingual (English 204
143 (i.e., a person admitted to a hospital after being involved in a crash) in and French) narrative-based search was used to verify that identified 205
144 Canada and is based on police-reported data. This database does not cases met the study criteria, including the removal of ineligible cases. 206

Please cite this article as: Vanlaar, W., et al., Fatal and serious injuries related to vulnerable road users in Canada, Journal of Safety Research (2016),
http://dx.doi.org/10.1016/j.jsr.2016.07.001
 W. Vanlaar et al. / Journal of Safety Research xxx (2016) xxx–xxx 3

207 Motorcyclists include drivers and passengers of motorcycles, dirt bikes, to 0.23 (50.7% decrease), and the rate of motorcyclist fatalities de- 243
208 mopeds and scooters, on a street, highway, or other public road creased to 0.54 (38.5% decrease). Overall, the rate of VRU fatalities de- 244
209 (i.e., excludes off-road cases). Years 2004–2006 were estimated using creased 49.2% from 1990 (3.9) to 2012 (1.9). 245
210 a 3-point prior moving average, to adjust for a change in coding prac- In terms of serious injuries, there was an overall reduction in the 246
211 tices during that period. The most common trauma injuries of fatally in- number of VRUs seriously injured from 1995 to 2012, from 3,379 to 247
212 jured VRUs are also presented. 2,643 (21.8% reduction). Specifically, for pedestrians there was a reduc- 248
213 The presence and amount of alcohol and drugs in fatally injured tion of 28.3% (from 1,789 in 1995 to 1,283 in 2012) and 39.6% for cyclists 249
214 drivers and pedestrians were derived from objective tests on body (from 591 in 1995 to 357 in 2012). However, for motorcyclists the 250
215 fluids. For alcohol, percentages of victim drivers and pedestrians (pas- change was almost negligible as there was a small 0.4% increase (from 251
216 sengers excluded) with different BAC levels (in mg/dL) were reported; 999 in 1995 to 1,003 in 2010). 252
217 however, only the presence of drugs, and not specific quantities, were The regression analyses (Table 1) show that the decreasing trends 253
218 reported, as this information was not consistently available. Linear re- (negative coefficient) for the rates per 100,000 population of each 254
219 gression models were estimated in order to study the significance of type of VRU fatality, were statistically significant (p-values b 0.05) al- 255
220 changes in these percentages over time. though the decreasing trends were more pronounced (larger absolute 256

F
221 As comparison, similar analyses for fatally and seriously injured oc- values) for pedestrians. The rates per 100,000 population for seriously 257
222 cupants (passengers and drivers) of PV were also presented when injured pedestrians and cyclists also significantly decreased over time. 258

O
223 possible. However, there was no significant trend for motorcyclists. 259
In comparison, the number of fatal victims in PVs decreased from 260
224 3. Results 1,973 in 1990 to 1,368 in 2012 (31% decrease). The reduction of serious- 261

O
ly injured victims in PVs was approximately 54% (from 12,549 in 1995 262
225 3.1. Fatalities and serious injuries from 1990 to 2012 to 5,814 in 2012). The rates per 100,000 population of fatal victims 263

R
and seriously injured in PV collisions also significantly decreased over 264
226 Fig. 1 shows the number of VRUs in Canada (excluding BC) that were time, but these decreases were more pronounced than for any of the 265
227 killed every year from 1990 to 2012 by victim type; and the number of VRU types (fatally injured: coef. = −0.21, p-value b 0.001; seriously in- 266

P
228 VRUs in Canada (excluding BC, NU, and NL) who were seriously injured jured: coef. = −1.87, p-value b 0.001). 267
229 from 1995 to 2012 by victim type. The rates for these fatalities and seri- Fig. 3 shows the population-based rates for motorcyclists compared 268
230 ous injuries per 100,000 population are presented in Fig. 2. Table 1 de- to rates based on registered motorcycles. In 1990, there were 59.6 mo-
D 269
231 picts the results from linear regression models in order to study the torcycle fatal victims per 100,000 registered motorcycles and this rate 270
232 significance of the changes over time. decreased to 27.1 in 2012. The rates of seriously injured motorcyclists 271
233 According to these data, there were 898 VRUs killed in 1990. Of this per 100,000 registered motorcycles also decreased from 312.2 in 1995 272
E
234 total, 571 were pedestrians, 214 were motorcyclists, and 113 were cy- to 166.6 in 2012. Regression analyses estimated significant decreasing 273
235 clists. In 2012, there were 565 VRU deaths and of those, 333 were pedes- trends for both rates of fatally and seriously injured motorcyclists 274
T

236 trians, 163 were motorcyclists, and 69 were cyclists. These numbers per 100,000 registered motorcycles (fatally injured: coef. = −1.15, 275
237 represent an overall reduction in VRU fatalities of 37% (41.7% for pedes- p-value b 0.001; seriously injured: coef. = −8.76, p-value b 0.001). 276
C

238 trians, 23.8% for motorcyclists, and 38.9% for cyclists). With respect to In comparison, the rates of fatally and seriously injured occupants 277
239 rates, in 1990, there were 2.34 pedestrian fatalities per 100,000 popula- of PV per 100,000 registered PVs also significantly decreased (fatally in- 278
240 tion, compared to 0.88 for motorcyclists and 0.46 for cyclists. In 2012, jured: coef. = −0.48, p-value b 0.001; seriously injured: coef. = −3.82, 279
E

241 the rate of pedestrian fatalities per 100,000 population decreased sub- p-value b 0.001). The rate of fatally injured occupants of PV per 100,000 280
242 stantially to 1.1 (53% decrease), the rate of cyclist fatalities decreased registered PV decreased from 15.6 in 1990 to 7.6 in 2012. The rate of 281
R
R
O
C
N
U

Fig. 1. Fatally (1990–2012) and seriously (1995–2012) injured VRUs. Source: TIRF fatality and serious injuries databases.

Please cite this article as: Vanlaar, W., et al., Fatal and serious injuries related to vulnerable road users in Canada, Journal of Safety Research (2016),
http://dx.doi.org/10.1016/j.jsr.2016.07.001
 4 W. Vanlaar et al. / Journal of Safety Research xxx (2016) xxx–xxx

F
O
O
R
Fig. 2. Rates of fatally (1990–2012) and seriously (1995–2012) injured VRUs per 100,000 population. Source: TIRF fatality and serious injuries databases, Statistics Canada Table 051-0001.

P
282 seriously injured occupants of PV rate per 100,000 registered PVs de- 3.3. Gender and age of victims
D 298
283 creased from 95.1 in 1995 to 32.4 in 2012.
Table 3 presents the distribution of fatally and seriously injured 299
284 3.2. VRU injuries and PV injuries proportional to all injury collisions from VRUs according to age categories and gender (the table excludes 1,842 300
E
285 1990 to 2012 victims whose age or gender was unknown). The numbers show that 301
among all VRU types, fatal and serious injuries were more common 302
T

286 Further analyses (Table 2) comparing the proportion of VRUs with among males than females (the absolute number of males is higher 303
287 those of PV occupants among all fatalities and serious injuries in road than for females in every age category). The highest percentage of vic- 304
C

288 crashes revealed: tims were aged 16 to 25 (21.5%), despite this age group only 305
representing approximately 13.6% of the Canadian population from 306
289 • a significant increasing trend in the proportion of motorcycle fatalities
1995 to 2012. 307
(coef. = 0.16, p-value b 0.001);
E

290
Tables 4, 5 and 6 show the information for each type of VRU, disag- 308
291 • significant increasing trends in the proportions of each type of VRU se-
gregated by type of injury and gender. 309
292 riously injured (coef. = 0.16, 0.35, 0.03 respectively for pedestrians,
R

As seen in Table 4, among fatally injured pedestrians, 63.5% were 310

293 motorcyclists and cyclists; p-values b 0.003); and,
male, whereas 36.5% were female. Among seriously injured pedestrians, 311
294 • significant decreasing trends in the proportion of PV victims
56.5% were male, while 43.5% were female. However, among pedes- 312
R

295 (fatal and seriously injured; coef. = − 0.39, − 0.82 respectively;

trians seriously injured aged 66 and over, females outnumbered 313
296 p-values b 0.001).
males. Table 4 shows that pedestrians aged 76 and older were overrep- 314
297
O

resented and accounted for 18.5% among all fatally injured pedestrians. 315
This is a concern since this population represented approximately only 316
5.3% of the Canadian population from 1995 to 2012. This was particular- 317
C

t1:1 Table 1 ly true among women as 22.8% (539 of 2,367) of all female pedestrians 318
t1:2 Regression models for trends in rates of fatally (1990–2012) and seriously (1995–2012) fatally injured between 1995 and 2012 were aged 76 and older, versus 319
t1:3 320
N

injured VRUs (pedestrians, motorcyclists and cyclists) per 100,000 population. Source: 16% (660 of 4,122) of all male pedestrians fatally injured. Further inves-
Q1
t1:4 TIRF fatality and serious injuries databases. tigation of this issue using age-specific rates (per 100,000 age-specific 321
t1:5 Variables Coeff. S.E. p-Value population between 1995 and 2012), clearly indicates a higher rate of 322
U

t1:6 Fatalities pedestrian rate trend

fatalities (4.5) among those aged 76 and older. Among seriously injured 323
t1:7 −0.050 0.004 b0.001 pedestrians, it was the young pedestrians aged 16 to 25 who were over- 324
t1:8 Fatalities motorcyclist rate trend represented (19.6%). However, in terms of age-specific rates, the older 325
t1:9 −0.007 0.003 0.034 pedestrians (i.e., 76 years and older) had a higher rate of serious injuries 326
t1:10 Fatalities cyclist rate trend
(7.7), followed by young pedestrians in the 16 to 25 age group (7.3). 327
t1:11 −0.010 0.002 b0.001
t1:12 Fatalities pass. veh. rate trend In Table 5, it can be seen that among fatally injured cyclists between 328
t1:13 −0.21 0.017 b0.001 1995 and 2012, 83.8% were men while only 16.2% were women, and 329
t1:14 Serious inj. pedestrian rate trend among seriously injured cyclists during the same time period, 76.1% 330
t1:15 −0.17 0.013 b0.001 were men while 23.9% were women. The overrepresentation of males 331
t1:16 Serious inj. motorcyclist rate trend
among fatally and seriously injured VRUs was more pronounced in 332
t1:17 −0.002 0.013 0.853
t1:18 Serious inj. cyclist rate trend the case of cyclists (and motorcyclists, see Table 6) than for pedestrians. 333
t1:19 −0.059 0.006 b0.001 Table 6 shows the age and gender distribution for fatally and seri- 334
t1:20 Serious inj. pass. veh. rate trend ously injured motorcyclists from 1995 to 2012. Of those who were fatal- 335
t1:21 −1.87 0.074 b0.001
ly injured, 89.5% were men while 10.5% were women. Among seriously 336

Please cite this article as: Vanlaar, W., et al., Fatal and serious injuries related to vulnerable road users in Canada, Journal of Safety Research (2016),
http://dx.doi.org/10.1016/j.jsr.2016.07.001
 W. Vanlaar et al. / Journal of Safety Research xxx (2016) xxx–xxx 5

F
O
O
R
Fig. 3. Rates of fatally (1990–2012) and seriously (1995–2012) injured motorcyclists per 100,000 population and per 1000 registered motorcycles. Sources: TIRF fatality and serious

P
injuries databases, Statistics Canada, Catalog No. 53-219 and CANSIM Tables 051-0001, 405-0004.

337 injured motorcyclists, 83.5% were men compared to 16.5% who were were more prevalent among seriously injured pedestrians (17.4%) and
D 355
338 women. cyclists (15.9%) than in motorcyclists (7.6%). Analyzing the fatality data- 356
base from 1990 to 2012, major head injuries were also more prevalent 357
339 3.4. Types of injuries in fatally injured cyclists (45.3%) and pedestrians (30.7%) than in motor- 358
E
cyclists (26.5%) and PV occupants (24.5%). 359
340 Table 7 presents the average annual distribution of the types of inju-
T

341 ries sustained by VRUs based on CHIRPP data from 1990 to 2010. The 3.5. Alcohol and drugs 360
342 data exclude cases where the nature of injury was classified as either in-
C

343 jury undetected or missing information (pedestrians: 502; cyclists: 143; A description of alcohol consumption by fatally injured VRUs be- 361
344 motorcyclists: 20). Furthermore, soft tissue injuries were only collected tween 1990 and 2012 is provided in Table 8. Passengers were excluded 362
345 from the year 2000 onwards. The most common injury types across all as they are usually not tested for alcohol or drugs. Analysis of data pre- 363
E

346 VRUs were superficial injuries, fractures, head injuries, and soft tissue sented in Table 8 revealed that, within all three VRU types, males were 364
347 injuries. Among pedestrians and cyclists, the most common type of overrepresented with respect to having consumed alcohol. Among fa- 365
R

348 trauma was superficial injury (29% and 29.5%, respectively). Fractures tally injured pedestrians, 49.1% of males had been drinking compared 366
349 (36.4%) represented the injury type most commonly found among to 29.2% of females. Among fatally injured motorcyclists, 36.8% of 367
350 motorcyclists. males had been drinking compared to 18.3% of females. For fatally in- 368
R

351 Overall, the proportions of these types of injuries have been quite jured bicyclists, 25.2% of males had been drinking compared to 9% of fe- 369
352 stable over time. Fig. 4 shows the percentages per year for the most se- males. It should be noted that many fatally injured VRUs had BAC over 370
O

353 vere injuries (head injuries and fractures) for each type of VRU. No sig- 80 mg of alcohol in 100 mL of blood (which is the legal limit of alcohol 371
354 nificant changes were found using regression analyses. Head injuries for drivers in Canada). To illustrate, among fatally injured drinking 372
VRUs, 86.2% of pedestrians, 76.5% of bicyclists, and 76% of motorcyclists 373
C

t2:1 Table 2 had BACs over 80 mg/dL. Furthermore, 66.1% of fatally injured drinking 374
t2:2 Regression models for trends in proportions of fatally (1990–2012) and seriously (1995– pedestrians had BACs over twice this limit. 375
N

t2:3 2012) injured VRUs and PV occupants. Table 9 shows similar data about drug presence among fatally in- 376
t2:4 Source: TIRF fatality and serious injuries databases.
jured VRUs in Canada between 2000 and 2012 along with alcohol pres- 377
t2:5 Variables Coeff. S.E. p-Value ence data in the same years. The testing rate for drugs, 37.3% overall, is 378
U

t2:6 Fatalities pedestrian proportion trend

t2:7 0.001 0.038 0.98
t2:8 Fatalities motorcyclist proportion trend Table 3 t3:1
t2:9 0.16 0.023 b0.001 Age distribution of fatally and seriously injured VRUs, 1995–2012. t3:2
t2:10 Fatalities cyclist proportion trend Source: TIRF fatality and serious injuries databases. t3:3
t2:11 0.01 0.017 0.73
Age category Females Males All Percent Percent of all population t3:4
t2:12 Fatalities pass. veh. proportion trend
t2:13 −0.39 0.035 b0.001 0–15 3,013 5,711 8,724 14.6 19.5 t3:5
t2:14 Serious inj. pedestrian proportion trend 16–25 3,269 9,536 12,805 21.5 13.6 t3:6
t2:15 0.16 0.027 b0.001 26–35 2,108 6,750 8,858 14.9 14.3 t3:7
t2:16 Serious inj. motorcyclist proportion trend 36–45 2,424 6,383 8,807 14.8 15.8 t3:8
t2:17 0.35 0.02 b0.001 46–55 2,531 5,945 8,476 14.2 14.4 t3:9
t2:18 Serious inj. cyclist proportion trend 56–65 1,706 3,287 4,993 8.4 10.2 t3:10
t2:19 0.03 0.009 0.003 66–75 1,514 1,948 3,462 5.8 6.9 t3:11
t2:20 Serious inj. pass. veh. proportion trend 76 & over 1,725 1,729 3,454 5.8 5.3 t3:12
t2:21 −0.82 0.063 b0.001 Total 18,290 41,289 59,579 100 100 t3:13

Please cite this article as: Vanlaar, W., et al., Fatal and serious injuries related to vulnerable road users in Canada, Journal of Safety Research (2016),
http://dx.doi.org/10.1016/j.jsr.2016.07.001
 6 W. Vanlaar et al. / Journal of Safety Research xxx (2016) xxx–xxx

t4:1 Table 4
t4:2 Number, percentage and rate of fatally and seriously injured pedestrian by age and gender, 1995–2012.
t4:3 Source: TIRF fatality and serious injuries databases.

t4:4 Age Pedestrian fatalities Pedestrian serious injuries

t4:5 Females Males Total % of all ped. fat. Age-spec. rates × 100 K pop. Females Males Total % of all ped. ser. inj. Age-spec. rates × 100 K pop

t4:6 0–15 271 405 676 0.68 1999 2896 4895 4.92
t4:7 40.1% 59.9% 100% 10.4% 40.8% 59.2% 100% 19.3%
t4:8 16–25 282 723 1005 1.46 1966 3009 4975 7.25
t4:9 28.1% 71.9% 100% 15.5% 39.5% 60.5% 100% 19.6%
t4:10 26–35 180 437 617 0.85 1146 1897 3043 4.21
t4:11 29.2% 70.8% 100% 9.5% 37.7% 62.3% 100% 12.0%
t4:12 36–45 210 504 714 0.88 1202 1854 3056 3.75
t4:13 29.4% 70.6% 100% 11.0% 39.3% 60.7% 100% 12.1%
t4:14 46–55 259 513 772 1.07 1347 1686 3033 4.22
t4:15 33.5% 66.5% 100% 11.9% 44.4% 55.6% 100% 12.0%

F
t4:16 56–65 280 419 699 1.40 1113 1205 2318 4.65
t4:17 40.1% 59.9% 100% 10.8% 48.0% 52.0% 100% 9.1%
t4:18 66–75 346 461 807 2.34 1089 917 2006 5.81

O
t4:19 42.9% 57.1% 100% 12.4% 54.3% 45.7% 100% 7.9%
t4:20 76 & over 539 660 1199 4.54 1168 864 2032 7.70
t4:21 45.0% 55.0% 100% 18.5% 57.5% 42.5% 100% 8.0%

O
t4:22 All ages 2367 4122 6489 1.29 11,030 14,328 25,358 5.03
t4:23 36.5% 63.5% 100% 100% 43.5% 56.5% 100% 100%

R
379 lower than the testing rate for alcohol, 66.1%. Among fatally injured fatally injured VRUs were less likely to be tested than PV fatalities. 404
380 VRUs tested for the presence of drugs, 39.5% of pedestrians, 31.4% of bi- There were not significant differences with respect to testing positive 405

P
381 cyclists, and 30.9% of motorcyclists tested positive. The gender differ- for drugs among all VRUs and drivers of PVs fatalities. However, fatally 406
382 ences were more pronounced in the consumption of alcohol than for injured pedestrians and PV drivers are more likely to test positive for al- 407
383 drug use. Males were significantly more likely than females to test pos- cohol than cyclists are, and the odds are larger for pedestrians than for
D 408
384 itive for alcohol (two-sample tests of proportions p-values b 0.05). PV occupants. 409
385 However, in terms of testing positive for drugs, the gender differences Linear regression models were estimated in order to study the sig- 410
386 were not significant. nificance of changes in the percentages of fatally injured VRUs testing 411
E
387 Table 10 summarizes four logistic regression models. These models positive for alcohol or drugs over time (see Table 11). The percentages 412
388 estimated the odds of being tested and the odds of testing positive for of fatally injured VRUs testing positive for drugs have not significantly 413
T

389 alcohol and drugs, respectively, after a fatal crash while controlling for changed over time for any VRU type and drivers of PVs. In the case of al- 414
390 sex, age, and VRU type. Please note that since victims in the age category cohol found in victims, the change for cyclists was not significant (p- 415
C

391 0 to 15 years are usually not tested, this age category was excluded from value N 0.05). However, the trend for the percentage of fatally injured 416
392 the analyses. The odds of being tested for alcohol after a fatal crash, for a victims testing positive for alcohol significantly decreased for motorcy- 417
393 male driver were 1.3 times the odds for a female driver. The odds of a clists, pedestrians, and PV drivers. The decreasing trend of the percent- 418
E

394 positive alcohol test for a male driver were 2.6 times the odds of a pos- age of fatally injured victims testing positive for alcohol was more 419
395 itive test for a female driver. pronounced (larger absolute value for the coefficient) for motorcyclists 420
R

396 The odds of being tested for drugs after a fatal crash for a male driver than for cyclists and PV drivers. 421
397 were 1.4 times the odds for a female driver. The odds of a positive drug Despite these decreasing trends, alcohol consumption remained 422
398 test for a male driver were 1.2 times the odds of a positive test for a fe- present in a large number of fatally injured VRUs. In 2012, 39.7% of fatal- 423
R

399 male driver. In these models (for alcohol and drugs), age was significant ly injured pedestrians had been drinking, 26.6% of fatally injured motor- 424
400 (p-values b 0.05), with the age group 16 to 25 being more likely to be cyclists, and 26.5% of fatally injured cyclists. The percentage of fatally 425
O

401 tested and the age group 26 to 35 more likely to test positive. With re- injured VRUs that had used drugs was higher than those who had con- 426
402 spect to VRU type in general, pedestrians and motorcyclists were sumed alcohol in 2012 for each VRU type: 43.4% of pedestrians, 29.7% of 427
403 more likely to be tested (for drugs and alcohol) than cyclists, but all cyclists, and 36.8% of motorcyclists. In comparison, the percentage of PV 428
C

t5:1 Table 5
N

t5:2 Number, percentage and rate of fatally and seriously injured bicyclist by age and gender, 1995–2012.
t5:3 Source: TIRF fatality and serious injuries databases.
U

t5:4 Age Cyclist fatalities Cyclist serious injuries

t5:5 Females Males Total % of all cyc. fat. Age-spec. rates × 100 K pop. Females Males Total % of all cyc. ser. inj. Age-spec. rates × 100 K pop.

t5:6 0–15 58 204 262 0.26 515 1,548 2,063 2.07

t5:7 22.1% 78% 100% 23.3% 25.0% 75.0% 100% 31.6%
t5:8 16–25 36 138 174 0.25 356 1,034 1,390 2.02
t5:9 20.7% 79.3% 100% 15.5% 25.6% 74.4% 100% 21.3%
t5:10 26–35 21 98 119 0.16 218 647 865 1.20
t5:11 17.6% 82.4% 100% 10.6% 25.2% 74.8% 100% 13.2%
t5:12 36–45 24 116 140 0.17 192 705 897 1.10
t5:13 17.1% 82.9% 100% 12.4% 21.4% 78.6% 100% 13.7%
t5:14 46–55 18 135 153 0.21 167 531 698 0.97
t5:15 11.8% 88.2% 100% 13.6% 23.9% 76.1% 100% 10.7%
t5:16 56 & over 25 252 277 0.25 111 507 618 0.56
t5:17 9.0% 91.0% 100% 24.6% 18.0% 82.0% 100% 9.5%
t5:18 Total 182 943 1,125 0.22 1,559 4,972 6,531 1.29
t5:19 16.2% 83.8% 100% 100.0% 23.9% 76.1% 100% 100.0%

Please cite this article as: Vanlaar, W., et al., Fatal and serious injuries related to vulnerable road users in Canada, Journal of Safety Research (2016),
http://dx.doi.org/10.1016/j.jsr.2016.07.001
 W. Vanlaar et al. / Journal of Safety Research xxx (2016) xxx–xxx 7

t6:1 Table 6
t6:2 Number, percentage and rate of fatally and seriously injured motorcyclist by age and gender, 1995–2012.
t6:3 Source: TIRF fatality and serious injuries databases.

t6:4 Age Motorcyclist fatalities Motorcyclist serious injuries

t6:5 Females Males Total % of all mot. fat. Age-spec. rates × 100 K pop. Females Males Total % of all mot. ser. inj. Age-spec. rates × 100 K pop.

t6:6 0–15 8 37 45 0.05 162 621 783 0.79

t6:7 17.8% 82.2% 100% 1.6% 20.7% 79.3% 100% 4.5%
t6:8 16–25 52 698 750 1.09 577 3,934 4,511 6.57
t6:9 6.9% 93.1% 100% 27.2% 12.8% 87.2% 100% 26.0%
t6:10 26–35 43 546 589 0.81 500 3,125 3,625 5.01
t6:11 7.3% 92.7% 100% 21.4% 13.8% 86.2% 100% 20.9%
t6:12 36–45 68 441 509 0.63 728 2,763 3,491 4.29
t6:13 13.4% 86.6% 100% 18.5% 20.9% 79.1% 100% 20.2%
t6:14 46–55 81 439 520 0.72 659 2,641 3,300 4.59
t6:15 15.6% 84.4% 100% 18.9% 20.0% 80.0% 100% 19.0%

F
t6:16 56 & over 36 304 340 0.31 238 1,375 1,613 1.46
t6:17 10.6% 89.4% 100% 12.4% 14.8% 85.2% 100% 9.3%
t6:18 Total 288 2,465 2,753 0.55 2,864 14,459 17,323 3.43

O
t6:19 10.5% 89.5% 100% 100.0% 16.5% 83.5% 100% 100.0%

O
429 drivers that died in a fatal collision and tested positive for alcohol in To place this in a broader context, motorcycle registrations in Canada 451
430 2012 (35.4%), was smaller than the percentage of fatally injured pedes- have increased from a low of 273,957 in 1999 to 661,452 in 2012 (141% 452

R
431 trians who tested positive. Similarly, the percentage of PV drivers in a increase). In 2012, there was approximately one motorcycle registered 453
432 fatal collision that tested positive for drugs in 2012 (41%) was smaller for every 47 motor vehicles registered (2.1%). Nonetheless, motorcy- 454
433 than the percentage of pedestrians that tested positive. The data for cy- clists still accounted for 7.5% of Canada's traffic fatalities in 2012. Ac- 455

P
434 clists were less stable, since comparatively fewer Canadian cyclists were cording to the CCMTA Vulnerable Road Users' Expert Working Group 456
435 fatally injured in any given year, and because testing rates for cyclists on Motorcycles (CCMTA, 2012) “the major contributing factors for mo- 457
436 (48.5%) tended to be lower than for motorcyclists (80.1%) and pedes- torcycle single-vehicle fatal collisions include speed, alcohol and drugs,
D 458
437 trians (58.1%). driver inexperience, inattention and distraction, animal action, and road 459
surface conditions” (page 7). 460
According to the CCMTA Vulnerable Road Users' Expert Working 461
E
438 4. Discussion Group on Pedestrians (2013), Canada is behind top performing coun- 462
tries when it comes to pedestrian safety and by implementing proven 463
T

439 4.1. Overall results measures, the current level of pedestrian fatalities could be reduced 464
by over 60% (CCMTA, 2013). This is particularly important considering 465
C

440 The results of this study suggest that while every year fewer Cana- the increasing popularity of bicycles, motorcycles, and walking as 466
441 dians overall died on the roadways from 1990 to 2012, occupants of means of transportation. Statistics Canada (2008, 2013) reported an in- 467
442 PVs appear to benefit the most from progress in road safety. Although crease of 18% in the percentage of bike share for work trips in Canada 468
E

443 the number of VRUs killed and seriously injured on Canada's roadways from 1.1% in 1996 to 1.3% in 2011. A comparison of results from the 469
444 each year seem to diminish, the percentage of deaths among these vic- 1996/1997 National Population Health Survey with those from the 470
R

445 tims as a percentage of all road user deaths did not show a similar de- 2005 Canadian Community Health Survey shows that the proportion 471
446 cline. Of concern, the proportion of fatally injured motorcyclists of people who spent at least six hours a week walking or bicycling as 472
447 significantly increased, and the proportion of each seriously injured a means of transportation rose from 19% to 24% (Gilmour, 2007). 473
R

448 VRU type also increased among all road user deaths. In contrast, the pro- The most reported injury types for VRUs in the CHIRPP were super- 474
449 portions of fatally and seriously injured PV occupants significantly ficial injuries, fractures, head injuries, and soft tissue injuries. Among 475
O

450 decreased. pedestrians and cyclists, the most common types of trauma were super- 476
ficial injuries (29% and 29.5%, respectively), and fractures represented 477
t7:1 Table 7 the injury type most commonly found among motorcyclists (36.4%). 478
C

t7:2 Average annual injury distribution (of most severe injury reported), among VRUs, CHIRPP, These results as indicators of the outcome of injuries for VRUs could 479
t7:3 1990–2010. Source: PHAC's CHIRPP surveillance system.
be considered in the context of another study with data from the emer- 480
N

t7:4 Nature of injury Pedestrians Cyclists Motorcyclists gency departments of four hospitals from Australia, where the percent- 481
t7:5 An. % An. % An. % age of minor injuries among pedestrians (59%) and bicyclists (41.7%) 482
freq freq freq were larger than the percentage of motorcyclists with minor injuries 483
U

t7:6 Superficial injury (bruise, abrasion) 216.0 29.0 111.7 29.5 19.5 17.0 (31%) and the percentage of critical injuries was larger for motorcyclists 484
t7:7 Fracture (incl. spine, excl. skull) 152.4 20.5 68.1 18.0 41.7 36.4 (2.9%) than for pedestrians (0%) and bicyclists (0.7%) (Meuleners, Lee, & 485
t7:8 Head injury 129.5 17.4 60.1 15.9 8.7 7.6 Haworth, 2006). Overall, there were no apparent changes in the most 486
t7:9 Minor head injury 65.1 8.7 27.4 7.2 2.1 1.8 serious injury types for VRUs over time. Analyses of both datasets, 487
t7:10 Concussion or intracranial injury 51.1 6.9 26.0 6.9 5.9 5.1
CHIRPP and the fatality database, show that head injuries were more 488
t7:11 Skull fracture 13.4 1.8 6.7 1.8 0.7 0.6
t7:12 Soft tissue injury 104.6 14.1 54.1 14.3 11.0 9.6 prevalent among pedestrians and cyclists than in motorcyclists. 489
t7:13 Laceration 38.6 5.2 32.0 8.4 6.4 5.6
t7:14 Multiple injuries 37.9 5.1 13.9 3.7 14.3 12.5 4.2. Gender and age of victims 490
t7:15 Sprain/strain or dislocation 30.9 4.1 22.3 5.9 7.5 6.6
t7:16 Othera 34.3 4.6 16.8 4.4 5.3 4.6
t7:17 Total 744.1 100 378.9 100 114.5 100 The data in this paper also corroborate previous research, showing 491
a
that males were overrepresented among all seriously and fatally injured 492
t7:18 The “other” category among all VRUs includes internal organ, crushing injury, muscle/
VRUs. This overrepresentation was more pronounced in the case of cy- 493
t7:19 tendon/nerve, dental injury, nature of injury not specified, eye injury (globe), burn/corro-
t7:20 sion, injury to blood vessel including hemorrhage, traumatic amputation, poisoning, for- clists and motorcyclists than in pedestrians. Over 80% of fatally and se- 494
t7:21 eign body, asphyxia, bite, drowning, systemic overexertion, and pulled elbow. riously injured cyclists and motorcyclists were male. Interestingly, 495

Please cite this article as: Vanlaar, W., et al., Fatal and serious injuries related to vulnerable road users in Canada, Journal of Safety Research (2016),
http://dx.doi.org/10.1016/j.jsr.2016.07.001
 8 W. Vanlaar et al. / Journal of Safety Research xxx (2016) xxx–xxx

F
O
O
R
Fig. 4. VRU fractures and head injuries percentages among each type of VRU, CHIRPP, 1990–2010, all ages. Source: PHAC's CHIRPP surveillance system.

P
496 females outnumbered males among the seriously injured pedestrians that a larger proportion of motorcyclist riders in Canada between the 511
497 aged 66 and older (54.3% vs. 45.7%). ages 16 and 24 years old ride well above the posted limit, pass other ve-
D 512
498 Child and teen cyclists were overrepresented among all seriously hicles when it is not safe to do so, often perform stunts on public roads, 513
499 and fatally injured cyclists, while 56 and older cyclists are overrepre- and weave in and out of traffic, than any other age group; see Vanlaar 514
500 sented among fatalities. In the case of children and teen cyclists expo- et al. (2009) for information about these data. 515
E
501 sure and underdeveloped skills (mental, physical, behavioral due to Elderly pedestrians were overrepresented among fatally injured pe- 516
502 logical development) may be possible factors. Pucher and Buehler destrians, followed by young pedestrians aged 16 to 25 years old. As 517
T

503 (2007) acknowledged that children and adolescents have the highest Canada's population ages, the need to ensure the safety of elderly 518
504 rates of cycling in almost every country. VRUs will need to receive greater priority (Robertson & Vanlaar, 519
C

505 Among motorcyclists, a major concern are the 27.2% of fatally in- 2008), particularly pedestrians, as walking becomes an important 520
506 jured motorcyclists and 26% of seriously injured motorcyclists that mode of transportation for the elderly. This result is in line with other 521
507 were between the ages 16 and 25 years old while this group represent- studies in several countries (Kaimila, Yamashina, Arai, &Tamashiro, 522
E

508 ed only approximately 14% of the Canadian population. One possible 2013; Jancey et al., 2013; SWOV, 2012, 2015), that found that elderly pe- 523
509 factor could be that this age group is more prone to risk taking behav- destrians are more likely to die than older adults who use other modes 524
R

510 iors. From the TIRF Road Safety Monitor data from 2008, it was found of transportation. These authors identified functional limitations, 525
R

t8:1 Table 8
t8:2 Number and percent of fatally injured VRUs using alcohol by victim type and gender, 1990–2012.
t8:3 Source: TIRF fatality database.
O

t8:4 Victim type Number of victimsa Victims tested Victims grouped by BAC (mg/dL)
(% of tested)
C

t8:5 (% of total) Negative 1–49 50–80 81–160 N160

t8:6 Pedestrians Female 3,255 1,623 1,149 51 23 100 300

N

t8:7 49.9% 70.8% 3.1% 1.4% 6.2% 18.5%

t8:8 Males 5,662 3,554 1,809 128 105 346 1166
t8:9 62.8% 50.9% 3.6% 3.0% 9.7% 32.8%
U

t8:10 All 8,917 5,177 2,958 179 128 446 1466

t8:11 58.1% 57.1% 3.5% 2.5% 8.6% 28.3%
t8:12 Motorcyclists Female 144 109 89 6 4 4 6
t8:13 75.7% 81.7% 5.5% 3.7% 3.7% 5.5%
t8:14 Males 3,188 2,561 1,619 180 75 288 399
t8:15 80.3% 63.2% 7.0% 2.9% 11.2% 15.6%
t8:16 All 3,332 2,670 1,708 186 79 292 405
t8:17 80.1% 64.0% 7.0% 3.0% 10.9% 15.2%
t8:18 Cyclists Female 271 122 111 4 0 2 5
t8:19 45.0% 91.0% 3.3% 0.0% 1.6% 4.1%
t8:20 Males 1,321 650 486 28 10 40 86
t8:21 49.2% 74.8% 4.3% 1.5% 6.2% 13.2%
t8:22 All 1,592 772 597 32 10 42 91
t8:23 48.5% 77.3% 4.1% 1.3% 5.4% 11.8%
t8:24 All VRUs 13,842 8,619 5,263 397 217 780 1,962
t8:25 62.3% 61.1% 4.6% 2.5% 9.0% 22.8%
a
t8:26 Excludes passengers of motorcycles and cycles.

Please cite this article as: Vanlaar, W., et al., Fatal and serious injuries related to vulnerable road users in Canada, Journal of Safety Research (2016),
http://dx.doi.org/10.1016/j.jsr.2016.07.001
 W. Vanlaar et al. / Journal of Safety Research xxx (2016) xxx–xxx 9

t9:1 Table 9
t9:2 Number and percent of fatally injured VRUs using alcohol and drugs by victim type and gender, 2000–2012. Source: TIRF fatality database.

t9:3 Victim type Number of victimsa Alcohol results Drugs results

t9:4 Victims tested (% of tested) Victims tested (% of tested)

t9:5 (% of total) Negative Positive (% of total) Negative Positive

t9:6 Pedestrians Female 1,565 823 584 239 392 235 157
t9:7 52.59% 70.96% 29.04% 25.05% 59.95% 40.05%
t9:8 Males 2,775 1,816 979 837 1,014 616 398
t9:9 65.44% 53.91% 46.09% 36.54% 60.75% 39.25%
t9:10 All 4,340 2,639 1,563 1,076 1,406 851 555
t9:11 60.81% 59.23% 40.77% 32.40% 60.53% 39.47%
t9:12 Motorcyclists Female 99 73 60 13 41 31 10
t9:13 73.74% 82.19% 17.81% 41.41% 75.61% 24.39%
t9:14 Males 1,826 1,505 1,045 460 942 648 294
t9:15 82.42% 69.44% 30.56% 51.59% 68.79% 31.21%

F
t9:16 All 1,925 1,578 1,105 473 983 679 304
t9:17 81.97% 70.03% 29.97% 51.06% 69.07% 30.93%

O
t9:18 Cyclists Female 115 63 59 4 39 26 13
t9:19 54.78% 93.65% 6.35% 33.91% 66.67% 33.33%
t9:20 Males 639 361 269 92 187 129 58
t9:21 56.49% 74.52% 25.48% 29.26% 68.98% 31.02%

O
t9:22 All 754 424 328 96 226 155 71
t9:23 56.23% 77.36% 22.64% 29.97% 68.58% 31.42%
t9:24 All VRUs 7,019 4,641 2,996 1,645 2,615 1,685 930

R
t9:25 66.12% 64.56% 35.44% 37.26% 64.44% 35.56%
a
t9:26 Excludes passengers of motorcycles and cycles.

P
526 greater physical and behavioral vulnerability, and lack of proper infra- posed by alcohol consumption, particularly in large quantities. Other 544
527 structure as risk factors for elderly pedestrians. possible interventions related to engineering countermeasures, en-
D 545
forcement, and transportation alternatives are suggested by Stewart 546
Q9
528 4.3. Alcohol and drug use among vulnerable road users (1995). 547
E
Real progress in reducing drinking and driving can be observed with 548
529 Alcohol and drug use among VRUs has been a relatively neglected motorcyclists based on the data. It may be postulated that this reduction 549
530 issue when compared with alcohol and drug use among drivers of is partly attributable to the types of road users being targeted by anti- 550
T

531 PVs. While the percentages of fatally injured pedestrians and motorcy- drunk driving campaigns, namely drivers of motorized vehicles. In com- 551
532 clists that test positive for alcohol have decreased, the numbers for pe- parison, the 26.1% of fatally injured motorcyclists who tested positive 552
C

533 destrians with respect to this indicator remain consistently high for alcohol in the period 1990–2012 in Canada is smaller than the 28% 553
534 (39.7% in 2012). This result is similar to the 41% of pedestrians killed with a BAC of 0.08 g/dL or higher in 2012 in the United States (NHTSA, 554
E

535 in fatal crashes in the United States that tested positive for alcohol 2015). With respect to statistics for bicyclists, these data are volatile 555
536 (NHTSA, 2014). Of note, a large proportion of fatally injured pedestrians and there were no significant trends over the 20-year time period stud- 556
537 who had been drinking consumed a significant quantity of alcohol. To il- ied. More data are needed before it can be concluded that fatally injured 557
R

538 lustrate, 86.2% of the fatally injured pedestrians between 2000 and 2012 bicyclists have an alcohol-involvement similar to, or different from, that 558
539 who tested positive for alcohol, registered a BAC above 80 mg/dL; and of pedestrians, motorcyclists, or drivers of PVs. 559
R

540 67.6% registered at least double (over 160 mg/dL). Such high levels of Drug use among VRUs is another area of road safety research that 560
541 intoxication certainly put these pedestrians at an increased risk of has been largely neglected. While testing rates are generally low, be- 561
542 crash involvement. Educational programs can be designed to meet the tween 30% and 40% of fatally injured VRUs (among those who were 562
O

543 needs of these pedestrians and to inform VRUs about the dangers tested) tested positive for drugs. These results suggest that drugs may 563
C

t10:1 Table 10
t10:2 Logistic regression models for the likelihood of being tested for alcohol (and drug) and the likelihood of a positive test after a fatal crash, 2000–2012. Adjusted odds ratios (OR) controlling
for road user type, sex and agea. Source: TIRF fatality database.
N

t10:3

t10:4 Alcohol Drugs

U

t10:5 Variables/categories Tested vs/not tested Positive test vs/negative test Tested vs/not tested Positive test vs/negative test
(among those who were (among those who were
tested) tested)

t10:6 Adj. OR p-Value Adj. OR p-Value Adj. OR p-Value Adj. OR p-Value

t10:7 Victim type Cycle Reference Reference

t10:8 Pedestrian 1.25 b0.001 2.97 b0.001 1.16 0.129 1.33 0.078
t10:9 Motorcycle 2.19 b0.001 1.02 0.911 1.81 b0.001 0.81 0.203
t10:10 Pass. veh. 3.08 0.015 1.99 0.001 2.02 b0.001 1.19 0.243
t10:11 Sex Female Reference Reference
t10:12 Male 1.33 b0.001 2.56 b0.001 1.35 b0.001 1.18 0.002
t10:13 Age 16–25 2.95 b0.001 4.53 b0.001 2.20 b0.001 1.34 b0.001
t10:14 26–35 2.70 b0.001 5.59 b0.001 2.04 b0.001 1.51 b0.001
t10:15 36–45 2.84 b0.001 4.43 b0.001 1.78 b0.001 1.43 b0.001
t10:16 46–55 2.26 b0.001 2.98 b0.001 1.68 b0.001 1.16 0.034
t10:17 56 & over Reference Reference
a
t10:18 Age category 0–15 was excluded.

Please cite this article as: Vanlaar, W., et al., Fatal and serious injuries related to vulnerable road users in Canada, Journal of Safety Research (2016),
http://dx.doi.org/10.1016/j.jsr.2016.07.001
 10 W. Vanlaar et al. / Journal of Safety Research xxx (2016) xxx–xxx

t11:1 Table 11 Dhondt, Macharis, Terryn, Van Malderen, & Putman, 2013; Teschke 614
Q14
t11:2 Regression models for trends in VRU and PV fatal victims testing positive for alcohol, et al., 2013). 615
t11:3 1990–2012 and drugs 2000–2012. Source: TIRF fatality database.

t11:4 Variables Coeff. S.E. p-Value

5. Conclusion 616
t11:5 Pedestrian alcohol positive trend −0.261 0.114 0.033
t11:6 Motorcyclist alcohol positive trend −1.177 0.163 b0.001 The safety of VRUs is an issue that is only increasing in importance. 617
t11:7 Cyclist alcohol positive trend 0.036 0.254 0.888
The promotion of walking and cycling as forms of exercise, growing 618
t11:8 Pass. veh. alcohol positive trend −0.341 0.095 0.002
t11:9 Pedestrian drug positive trend 0.29 0.24 0.26 popularity of motorcycling (especially among middle-aged riders with 619
t11:10 Motorcyclist drug positive trend 1.05 0.59 0.1 more disposable income to buy heavy, large motorcycles), overall pop- 620
t11:11 Cyclist drug positive trend 1.61 0.83 0.08 ulation increases, as well as predicted increases in the number of elderly 621
t11:12 Pass. veh. drug positive trend 0.34 0.21 0.12
road users means that actions must be taken to ensure that proper mea- 622
sures are in place to protect these road users. 623
Advances have been made in recent years, most notably with respect 624
564 be a contributing factor in serious VRU crashes, however, this needs to to drinking and driving among motorcyclists and the total number of 625

F
565 be further studied. annual pedestrian deaths, but safety gaps continue to exist for VRUs as 626
a whole. Comprehensive data like those used in this study can help to 627

O
inform prevention and mitigation initiatives. The website for the Cana- 628
566 4.4. Strengths and limitations dian Road Safety Strategy 2025 (http://crss-2025.ccmta.ca/en/) pro- 629
vides 91 initiatives that could be adopted or adapted to address VRU 630

O
567 The TIRF databases contain the population of fatalities and injuries, road safety challenges. 631
568 rather than a sample. As such, both these databases contain very reliable Although our study presents some important analyses, more re- 632
569 data of high quality, representative of Canada. These databases have

R
search is needed to better understand and prevent fatal and serious in- 633
570 been maintained for several decades and have been used, for example, juries related to VRUs in Canada. Future research could, for example, 634
571 “to assist policy development, public awareness and monitoring of incorporate better exposure data as these become available, and look 635

P
572 alcohol-impaired driving in Canada” (Beasly et al., 2011, page 4). Never- into different aspects (e.g., behavioral, infrastructure) that could be as- 636
573 theless, it can be argued that the lack of detailed information about the sociated with these injuries. 637
574 severity of injuries such as an Abbreviated Injury Score (AIS) is a limita- D
575 tion. However, the categorization of injuries in these databases is consis-
Uncited references 638
Q15
576 tent with databases typically used in this type of research and adheres
577 to strict guidelines. Although the presence of drugs in victims of fatal
E
Arregui-Dalmases et al., 2010 639
578 collisions is reported here, future studies may report on the different
Basset et al., 2008 640
579 types and quantity levels of drugs, which were not consistently avail-
Bohmer and Proust, 2005 641
T

580 able for this study. Nevertheless, since 2010, over 90% of fatal victims
Campbell and Wittgens, 2004 642
581 were tested for drugs in some Canadian jurisdictions.
Canadian Council of Motor Transport Administrators (CCMTA), 643
C

582 On the other hand, the CHIRPP database is not population based.
Derry-Damsere et al., 2010 644
583 Strictly speaking this may bias the results to a certain extent in that
Eilert-Petersson and Schelp, 1997 645
584 the injury information collected is only a sample of the hundreds of Ca-
E

Hinds et al., 2007 646

585 nadian hospitals that provide emergency room treatment of injuries.
Insurance Institute for Highway Safety, 2011 647
586 Moreover, the injury information collected was from persons who pre-
Linn et al., 1998 648
R

587 sented with injuries from within each hospital's catchment area, most
Macarthur and Pless, 1999b 649
588 often located in Canada's metropolitan areas. As such, residents of
Mikocka-Walus et al., 2010 650
589 smaller centers and rural and remote areas may be underrepresented
R

Peng and Bongard, 1999 651

590 in the CHIRPP database. Older teens and adults may be underrepresent-
Pucher et al., 2011 652
591 ed because of the high proportion of pediatric hospitals in the program.
Rivara et al., 1997 653
O

592 Fatal injuries were also underrepresented in CHIRPP because many vic-
Royal and Miller-Steiger, 2008 654
593 tims who die at the scene were not transported to the hospital. CHIRPP
Statistics Canada, 2009 655
594 forms are completed voluntarily by presenting patients or a proxy re-
C

Transport Canada and Canadian Council of Motor Transport Admi 656

595 spondent (most often a parent) and therefore selection bias exists to
596 an extent as well. Persons bypassing the emergency department regis-
Acknowledgment 657
N

597 tration process because of severe injuries needing immediate treatment

598 may not always be captured either. Nevertheless, it warrants mention-
599 ing that research has demonstrated the validity of the CHIRPP database This research has been made possible with funding from the Traffic 658
Q16
U

600 (cf. Methods section for references). Injury Research Foundation (TIRF) and the Public Health Agency of 659
601 Lack of readily available exposure data for VRUs is a limitation for the Canada (PHAC). The data from TIRF's Fatality Database and Serious Inju- 660
602
Q10 interpretation and comparison of results. Vanparijs et al. (2015) con- ry Database used for the analyses in this research have been collected 661
603 ducted a literature review on exposure measurement in bicycle safety with financial support from the CCMTA, Transport Canada and State 662
604 analysis and found only 2% of the studies from their search collected Farm. 663
605 some form of exposure. In this study we used registered motorcycles
606 as a proxy exposure measure but unfortunately other data such as dis- References 664
607 tance traveled per mode of transportation, do not exist in Canada.
608
Q11 Teschke et al. (2013) explored this lack of data in Canada, presented Arregui-Dalmases, C., Lopez-Valdes, F., & Segui-Gomez, M. (2010). Pedestrian injuries in 665
eight European countries: An analysis of hospital discharge data. Accid. Anal. Prev., 666
609 some exposure data from British Columbia and emphasized the impor- 42(4), 1164–1171. 667
610 tance to implement a national travel survey to allow injury rate calcula- Basset, D., Pucher, J., Buehler, R., Thompson, D., & Crouter, S. (2008). Walking, cycling, and 668
611 tions for all modes of travel. National travel surveys have proven useful obesity rates in Europe, North America, and Australia. J. Phys. Act. Health, 5, 795–814. 669
Beasley, E. E., Beirness, D. J., & Porath-Waller, A. J. (2011). A comparison of drug- and 670
612 providing exposure data (Beck, Dellinger, & O′Neil, 2007; International alcohol-involved motor vehicle driver fatalities. Ottawa, ON: Canadian Centre on Sub- 671
Q13613
Q12 Traffic Safety Data and Analysis Group, 2011; McAndrews, 2011; stance Abuse (CCSA). 672

Please cite this article as: Vanlaar, W., et al., Fatal and serious injuries related to vulnerable road users in Canada, Journal of Safety Research (2016),
http://dx.doi.org/10.1016/j.jsr.2016.07.001
 W. Vanlaar et al. / Journal of Safety Research xxx (2016) xxx–xxx 11

673 Beck, L. F., Dellinger, A. M., & O′Neil, M. E. (2007). Motor vehicle crash injury rates by National Highway Traffic Safety Administration (NHTSA) (2014). Traffic safety facts: 2012 759
674 mode of travel, United States: Using exposure-based methods to quantify differences. data. http://www-nrd.nhtsa.dot.gov/Pubs/811888.pdf (accessed May 2016) 760
675 American Journal of Epidemiology, 166, 212–218. National Highway Traffic Safety Administration (NHTSA) (2015). Traffic safety facts: 2013 761
676 Bohmer, J., & Proust, A. (2005). Adult bicycle handlebar injury. Am. J. Emerg. Med., 24(5), data. http://www-nrd.nhtsa.dot.gov/Pubs/812148.pdf (accessed November 2015) 762
677 624–625. Peng, R., & Bongard, F. (1999). Pedestrian versus motor vehicle accidents. J. Am. Coll. Surg., 763
678 Brown, S., Vanlaar, W., & Mayhew, D. (2013). The alcohol crash problem in Canada: 2010. 189(4), 343–348. 764
679 Ottawa, ON: Traffic Injury Research Foundation. Prepared for: Canadian Council of Pickett, W., Brison, R. J., Mackenzie, S. G., Garner, M., King, M. A., Greenberg, T. L., & 765
680 Motor Transport Administrators Standing Committee on Road Safety Research and Poli- Boyce, W. F. (2000). Youth injury data in the Canadian hospitals injury reporting 766
681 cies and Transport Canada. and prevention program: Do they represent the Canadian experience? Inj. Prev., 767
682 Brown, S., Vanlaar, W., & Robertson, R. (2015). Alcohol and drug-crash problem in Canada: 6, 9–15. 768
683 2012 report. Ottawa, ON: Traffic Injury Research Foundation (Prepared for: Canadian Pucher, J., Buehler, R., & Seinen, M. (2011). Bicycling renaissance in North America? An 769
684 Council of Motor Transport Administrators). update and re-appraisal of cycling trends and policies. Transp. Res. A, 45, 451–475. 770
685 Campbell, R., & Wittgens, M. (2004). The business case for active transportation: The eco- Pucher, J., & Buehler, R. (2007). At the frontiers of cycling: Policy innovations in 771
686 nomic benefits of walking and cycling. Better Environ. Sound Trans.(March 2004). the Netherlands, Denmark, and Germany. World Transp. Policy Pract., Volume 13 772
687 Canadian Council of Motor Transport Administrators (CCMTA) (2013,). Countermeasures (Number 3). 773
688 to improve pedestrian safety in Canada. CCMTA vulnerable road users' expert working Rivara, F., Thompson, D., & Thompson, S. (1997). Epidemiology of bicycle injuries and risk 774
689 group on pedestrians. factors for serious injury. Inj. Prev., 3, 110–114. 775
690 Canadian Council of Motor Transport Administrators (CCMTA) (2012,). Countermeasures Robertson, R., & Vanlaar, W. (2008). Elderly drivers: Future challenges? Accid. Anal. Prev., 776
691 777

F
to address motorcycle safety: Building a foundation to support Canada's road safety strat- 40(6), 182–186.
692 egy 2015. The Vulnerable Road Users' Expert Working Group on Motorcycles. Royal, D., & Miller-Steiger, D. (2008). National survey of bicyclist and pedestrian attitudes 778
693 Crain, J., McFaull, S., Thompson, W., Skinner, R., Do, M. T., Fréchette, M., & Mukhi, S. and behavior. National Highway Traffic Safety Administration. 779

O
694 (2016). The Canadian hospitals injury reporting and prevention program: A dynamic Siman-Tov, M., Jaffe, D., Israel Trauma Group, & Peleg, K. (2012). Bicycle injuries: A matter 780
695 and innovative injury surveillance system. Health promotion and chronic disease of mechanism and age. Accid. Anal. Prev., 44(1), 135–139. 781
696 prevention in Canada. Res. Policy Prac., 36(6) (in press). StataCorp (2013). Stata: Release 13. Statistical software. College Station, TX: StataCorp. LP. 782
697 Derry-Damsere, J., Ebel, B., Mock, C., Afukaar, F., & Donkor, P. (2010). Pedestrians' injury Statistics Canada (2008). Catalogue no. 97-561-x. Commuting patterns and places of work 783

O
698 patterns in Ghana. Accid. Anal. Prev., 42(4), 1080–1088. of Canadians, 2006 census. http://www12.statcan.ca/census-recensement/2006/as-sa/ 784
699 Dhondt, S., Macharis, C., Terryn, N., Van Malderen, F., & Putman, K. (2013). Health burden 97-561/pdf/97-561-XIE2006001.pdf 785
700 of road traffic accidents, an analysis of clinical data on disability and mortality expo- Statistics Canada. (2009). Canadian Community Health Survey (CCHS). Ottawa, Canada. 786

R
701 sure rates in Flanders and Brussels. Accid. Anal. Prev., 50, 659–666. http://dx.doi.org/ Statistics Canada (2013). Catalogue no. 99-012-X2011003. Commuting to work. National 787
702 10.1016/j.aap.2012.06.019. Household Survey (NHS), 2011. http://www12.statcan.gc.ca/nhs-enm/2011/as-sa/99- 788
703 Eichelberger, A. H., Cicchino, J. B., & McCartt, A. T. (2013). Profile of fatally injured pedes- 012-x/99-012-x2011003_1-eng.pdf 789
704 790

P
trians and bicyclists in the United States with high blood alcohol concentrations. Pro- Statistics Canada (2014). Table 051-0001. Estimates of population, by age group and sex,
705 ceedings of the 20th International Conference on Alcohol, Drugs, and Traffic Safety. Canada, provinces and territories, annual. CANSIM (database) (accessed: June 25, 791
706 Brisbane, Australia (CD-ROM). 2015). 792
707 Eilert-Petersson, E., & Schelp, L. (1997). An epidemiological study of bicycle-related inju- Statistics Canada (2015). Table 405-0004. Road motor vehicles, registrations, annual (num- 793
708 ries. Accid. Anal. Prev., 29(3), 363–372. D ber), CANSIM (database) (accessed: June 25, 2015). 794
709 Gilmour, H. (2007). Physically active Canadians. Health Reports (Statistics Canada, Cata- Stewart, K. (1995). Report of the ICADTS working group on alcohol-involved pedestrians. 795
710 logue 82-003). Vol. 18, No. 3. (pp. 45–65). 13th International Conference on Alcohol, Drugs and Traffic Safety. Adelaide, Australia. 796
711 Hinds, J., Allen, G., & Morris, C. (2007). Trauma and motorcyclists; born to be wild, bound SWOV (2012). Vulnerable road users. SWOV fact sheet. Leidschendam: SWOV Institute for 797
E
712 to be injured? Injury. Int. J. Care Injured, 28, 1131-118. Road Safety Research. http://www.swov.nl/rapport/Factsheets/UK/FS_Vulnerable_ 798
713 International Traffic Safety Data and Analysis Group (2011). IRTAD road safety 2010. An- road_users.pdf accessed May 2016 799
714 nual report. OECD/ITF. SWOV (2015). The elderly in traffic. SWOV fact sheet. Leidschendam: SWOV Institute for 800
T

715 Insurance Institute for Highway Safety (2011). Pedestrians stand to benefit from new ve- Road Safety Research. https://www.swov.nl/rapport/Factsheets/UK/FS_Elderly.pdf 801
716 hicle technology and design changes aimed at preventing crashes. Status Rep., 46(3), accessed May 2016 802
717 1–3. Transport Canada (2011). Road Safety in Canada. TP 15145 E. Motor vehicle safety with sup- 803
C

718 Jancey, J., Cooper, L., Howat, P., Meuleners, L., Sleet, D., & Baldwin, G. (2013). Pedestrian port from the Public Health Agency of Canada. Cat. T46-54/1-2011E (ISBN 978-1-100- 804
719 and motorized mobility scooter safety of older people. Traffic Inj. Prev., 14(6), 18621-4). 805
720 647–653. http://dx.doi.org/10.1080/15389588.2012.749465. Transport Canada, & Canadian Council of Motor Transport Administrators (2015i). 2013 806
E

721 Kaimila, B., Yamashina, H., Arai, A., & Tamashiro, H. (2013). Road traffic crashes and fatal- Canadian motor vehicle traffic collision statistics. https://www.tc.gc.ca/media/ 807
722 ities in Japan 2000–2010 with special reference to the elderly road user. Traffic Inj. documents/roadsafety/cmvtcs2013_eng.pdf 808
723 Prev., 14(6), 777–781. http://dx.doi.org/10.1080/15389588.2013.774085. Vanlaar, W., Marcoux, K., & Robertson, R. (2009). The road safety monitor 2008: Motorcy- 809
R

724 Kang, J., Hagel, B., Emery, C. A., Senger, T., & Meeuwisse, W. (2013). Assessing the repre- clists. Ottawa, ON: The Traffic Injury Research Foundation. 810
725 sentativeness of Canadian Hospitals Injury Reporting and Prevention Programme Vanlaar, W., McAteer, H., Brown, S., Crain, J., McFaull, S., & Mainegra Hing, M. (2015). 811
726 (CHIRPP) sport and recreational injury data in Calgary, Canada. Int. J. Inj. Contr. Saf. Injuries related to off-road vehicles in Canada. Accid. Anal. Prev., 75, 264–271. 812
727 Promot., 20(1), 19–26. World Health Organization (2015a). Road traffic injuries. Media Centre: fact sheet no. 358. 813
R

728 Keall, M. D., & Newstead, S. (2012). Analysis of factors that increase motorcycle rider risk http://www.who.int/mediacentre/factsheets/fs358/en/ 814
729 compared to car driver risk. Accid. Anal. Prev., 49, 23–29. World Health Organization (2013). Global status report on road safety 2013: Supporting a 815
730 Kim, J., Kim, S., Ulfarsson, G., & Porrello, L. (2007). Bicyclist injury severities in bicycle– decade of action. 816
O

731 motor vehicle accidents. Accid. Anal. Prev., 39(2), 238–251.

732 Linn, S., Smith, D., & Sheps, S. (1998). Epidemiology of bicycle injury, head injury, and hel-
733 met use among children in British Columbia: A five year descriptive study. Inj. Prev., 4, Dr. Ward Vanlaar is the Vice President Research with the Traffic Injury Research Founda- 817
734 tion (TIRF). He has M.A. degrees in statistics and criminology and a Ph.D. in transportation 818
C

122–125.
735 Macarthur, C., & Pless, I. B. (1999a). Evaluation of the quality of an injury surveillance sys- science. Ward has published over 150 reports and peer-reviewed articles in leading 819
736 tem. Am. J. Epidemiol., 149(6), 586–592. journals. He served as a board member of the Canadian Association of Road Safety Profes- 820
737 Macarthur, C., & Pless, I. B. (1999b). Sensitivity and representativeness of a childhood in- sionals (CARSP) and is a past board member of the Association of Transportation Safety In- 821
N

738 jury surveillance system. Inj. Prev., 5, 214–216. formation Professionals (ATSIP). Dr. Vanlaar is an editorial board member of the Journal of 822
739 Mackenzie, S. G., & Pless, I. B. (1999). CHIRPP: Canada's principal injury surveillance pro- Safety Research and Accident Analysis and Prevention. He serves on the Committee on 823
740 gram. Inj. Prev., 5, 208–213. Safety Data, Analysis and Evaluation and the Committee on Alcohol, Other Drugs, and 824
825
U

741 Macpherson, A. K., White, H. L., Mongeon, S., Grant, V. J., Osmond, M., Lipskie, T., & Transportation of the Transportation Research Board (TRB) of the National Academies.
742 Mackay, M. J. (2008). Examining the sensitivity of an injury surveillance program 826
743 using population-based estimates. Inj. Prev., 14, 262–265. Marisela Mainegra Hing is a research scientist at TIRF. She has a degree in computer sci- 827
744 Mayhew, D. (2011). The alcohol crash problem in Canada: 2009. Ottawa, ON: Traffic Inju- ence, MSc. in applied mathematics and a Ph.D. in operations research. Before joining TIRF 828
745 ry Research Foundation. (Prepared for: Transport Canada and the Canadian Council of she was a Postdoctoral Fellow and a part-time teacher at the Telfer School of Management 829
746 Motor Transport Administrators Standing Committee on Road Safety Research and at the University of Ottawa and an Assistant Professor of Mathematics at the Central Uni- 830
747 Policies). versity of Las Villas, Cuba. She has also been associated as a researcher at the Twente Uni- 831
748 Meuleners, L. B., Lee, A. H., & Haworth, C. (2006). Emergency presentations by vulnerable versity in The Netherlands, Research Center for Cuban Internal Trade, Bayes Forecast SA in 832
749 road users: Implications for injury prevention. Inj. Prev., 12, 12–14. http://dx.doi.org/ Spain and CRG Consulting in Canada. 833
750 10.1136/ip.2005.010389. 834
751 Mikocka-Walus, A., Gabbe, B., & Cameron, P. (2010). Motorcycle-related major trauma: Steve Brown is a Research Associate at TIRF. He holds an honor Bachelor of Arts in mass 835
752 On-road versus off-road incidence and profile of cases. Emerg. Med. Australas., 22, communications from Carleton University. He has overseen data collection at TIRF since 836
753 470–476. 1991 and co-authors the alcohol-crash and drug-crash problem reports which also include 837
754 Monk, J., Buckley, R., & Dyer, D. (2009). Motorcycle-related trauma in Alberta: A sad and information about prescribed drugs. Steve has also worked on questionnaire design and 838
755 expensive story. Can. J. Surg., 52(6), E235–E239. helped coordinate data collection and analysis for TIRF's Road Safety Monitor, a publica- 839
756 Motorcycle and Moped Industry Council (2013). 2012 Motorcycle, scooter and all-terrain tion which highlights the current state of affairs and trends in public knowledge and atti- 840
757 vehicle annual industry statistics report. Motorcycle and Moped Industry Council and tudes on various road safety issues such as drinking and driving, drugs and driving, 841
758 Canadian Off-Highway Vehicle Distributors Council. speeding and vehicle safety features. 842
843

Please cite this article as: Vanlaar, W., et al., Fatal and serious injuries related to vulnerable road users in Canada, Journal of Safety Research (2016),
http://dx.doi.org/10.1016/j.jsr.2016.07.001
 12 W. Vanlaar et al. / Journal of Safety Research xxx (2016) xxx–xxx

844 Heather McAteer was a research assistant at TIRF. She has a Master of Philosophy and a Steven R. McFaull is a Senior Injury Epidemiologist with the Surveillance and Epidemiol- 858
845 Bachelor of Arts in Philosophy and English from York University. Before joining TIRF, ogy Division of the Public Health Agency of Canada. His background is in biomechanics and 859
846 Heather was a Teaching Assistant in Philosophy at Dalhousie University in Halifax, Nova statistics. His research interests include sports and recreational injuries, product-related 860
847 Scotia and a Research Assistant at York University. She is currently a research associate injuries and anthropometry. 861
848 at the Conference Board of Canada. 862
849
850 Jennifer Crain is a Research and Data Manager with the Public Health Agency of Canada's
851 Injury Section, where she's worked since 2010 on national injury surveillance initiatives,
852 primarily the Canadian Hospitals Injury Reporting and Prevention Program. As her prima-
853 ry area of interest is road safety, Jennifer has represented the Agency on a number of relat-
854 ed epidemiological studies and working groups; however, she has also studied a variety of
855 other injury topics including pediatric burns.
856

F
O
O
R
P
D
E
T
C
E
R
R
O
C
N
U

Please cite this article as: Vanlaar, W., et al., Fatal and serious injuries related to vulnerable road users in Canada, Journal of Safety Research (2016),
http://dx.doi.org/10.1016/j.jsr.2016.07.001