IATSS Research 38 (2015) 83–91
Contents lists available at ScienceDirect
IATSS Research
Safety of young children on motorized two-wheelers around the world:
A review of the global epidemiological evidence
Kavi Bhalla a,⁎, Dinesh Mohan b
a
b
Johns Hopkins International Injury Research Unit, Johns Hopkins Bloomberg School of Public Health, 615 N Wolfe St, E8138, Baltimore, MD 21205, United States
Transportation Research & Injury Prevention Programme, Indian Institute of Technology Delhi, Room 815 Main Building, 7th Floor, Hauz Khas, New Delhi 110016, India
a r t i c l e
i n f o
Available online 20 November 2014
Keywords:
Children
Motorized two wheeler safety
a b s t r a c t
The safety of children younger than 10 years on motorized two-wheeled vehicles (MTWs) in low- and middleincome countries receives substantial attention from global road safety advocates. However, there is little empirical evidence available to describe the magnitude of the problem. Therefore, we constructed a population-level
database of road traffic injury statistics disaggregated by age (b 5, 5–9, 10+ years) and mode of transport. Our
database included mortality data from 44 countries and 5 Indian cities, and hospital admissions from 17 countries. The MTW fleet in these settings ranged from 2% to 70% of all registered vehicles. We find that children
under 5 years averaged 0.05% (SD 0.13%) of all road traffic deaths, and 5–9 year olds averaged 0.11% (SD
0.25%). Even in regions with high prevalence of MTWs, young children comprised at most 1.5% of all road traffic
deaths and 5.8% of all MTW deaths. Young children were a slightly larger proportion of all road traffic deaths in
countries where MTWs were more common. However, after adjusting for population age structure, this effect
was no longer evident. The percentage of child road traffic injuries that are due to MTWs increased with increasing MTW use, but at a much lower rate. Our findings suggest that children may be at lower risk from MTW
crashes than previously assumed, and certainly at a lower risk than as pedestrians. Further studies are needed
to explain the underlying mechanisms that regulate risk of road users.
© 2014 Production and hosting by Elsevier Ltd. on behalf of International Association of Traffic and Safety Sciences.
1. Introduction
There appears to be near unanimous consensus among global road
safety advocates that the safety of young children on motorcycles in
low- and middle-income countries is of grave concern. Thus, a recent
World Health Organization (WHO) report to the Ministry of Transport
of Vietnam advises, “WHO encourages adults not to transport children
on motorcycles unless absolutely necessary. If this is the case, then
both WHO and UNICEF promote the use of standardized, correctly fitted,
helmets for children as a harm reduction strategy.” [1] (Bold in original)
Similarly, a report from the WHO South East Asia Regional Office recommends, “Governments at the national, provincial and municipal levels
should consider policies … that would obviate the need for transporting
children on motorcycles” [2].
These policy statements are primarily referring to infants and children younger than 10 years (hereafter referred to as “young children”)
⁎ Corresponding author. Tel.:+1 954 849 8692 (Cell).
E-mail address: [email protected] (K. Bhalla).
Peer review under responsibility of International Association of Traffic and Safety Sciences.
who are often transported as passengers on motorized two-wheeled
vehicles (MTWs) in many parts of the world. Anecdotal mentions of
such use of MTWs have been reported from South, East, and Southeast
Asia [1,2] (including Malaysia [3] and Vietnam [4]), sub-Saharan Africa
(including Cameroon [5] and Uganda [6]), North Africa and the Middle
East (including Iran [7]), Latin America (including Brazil [8]), among
others. MTW riders are inherently vulnerable in crashes because, unlike
most motor vehicles, MTWs do not have a steel protective shell. Head
injuries are the most common cause of death in MTW crashes. Therefore, the promotion of helmet use has been a primary thrust of MTW
safety policy worldwide [9]. Unfortunately, providing protective head
gear for infants is difficult for several reasons, including the fact that
the size and shape of the human head evolve rapidly during the first
four years of life [10]. Although helmets exist for older children (aged
5–10) for different purposes, helmet use in this age group of MTW riders
is relatively rare in most of the world [2,11]. Therefore, it is natural that
road safety advocates worry about the vulnerability of young children in
MTW crashes.
However, despite the apparent importance of the issue and the
strong public positions taken by international agencies, there is surprisingly little empirical evidence supporting the urgency of addressing the
safety of children on MTWs. Public debates related to the health of
children usually inspire strong emotions and a tendency to view issues
as a matter of values that are so fundamental that they are beyond rational debate [12]. Thus, advocates for the safety of children on MTWs have
http://dx.doi.org/10.1016/j.iatssr.2014.11.001
0386-1112/© 2014 Production and hosting by Elsevier Ltd. on behalf of International Association of Traffic and Safety Sciences.
84
K. Bhalla, D. Mohan / IATSS Research 38 (2015) 83–91
Table 1
Global vehicle and motorized two-wheeler (MTW) fleet in 2010.
Global region
Southeast Asia
South Asia
East Asia
Western Sub-Saharan Africa
Caribbean
Andean Latin America
Tropical Latin America
Southern Latin America
North Africa & Middle East
Eastern Sub-Saharan Africa
Central Latin America
Western Europe
Eastern Europe
Central Europe
High Income Asia Pacific
Australasia
Southern Sub-Saharan Africa
High Income North America
Central Sub-Saharan Africa
Central Asia
Oceania
Grand total
Population (1000s)
612,705
1,628,745
1,363,753
339,896
27,962
53,078
207,069
62,103
402,319
342,560
231,953
425,264
209,827
118,070
186,282
26,170
68,589
354,776
93,636
47,448
2041
6,804,247
Fleet
Ownership (per 1000 people)
Vehicles (1000s)
MTWs (1000s)
MTWs (% of veh)
Vehicles
Motorcycles
171,133
126,398
211,958
18,118
7425
5105
65,737
18,826
74,426
4925
47,985
271,574
65,604
50,547
110,877
19,288
11,564
280,345
791
11,017
552
1,574,198
129,224
88,869
104,412
6746
2674
1546
16,746
4716
17,306
921
6781
28,615
4845
3509
6906
773
430
8525
16
154
6
433,719
76
70
49
37
36
30
25
25
23
19
14
11
7
7
6
4
4
3
2
1
1
28
279
78
155
53
266
96
317
303
185
14
207
639
313
428
595
737
169
790
8
232
271
231
211
55
77
20
96
29
81
76
43
3
29
67
23
30
37
30
6
24
0
3
3
64
Countries are grouped into GBD-2010 regions [25].
Regions are sorted by proportion of vehicle fleet that is MTWs.
had a tendency to frame facts as “everybody knows” without supporting
empirical evidence.
The goal of our study is to review the empirical evidence on the
safety of young children on MTWs globally. In particular, we focus on
a cluster of questions that affect public policy:
• Are young children more vulnerable than adults in MTW crashes and
do children require additional safety considerations?
• Are MTW crashes an important risk factor for child health around the
world? Should public health professionals focus special attention on
the issue?
• Are child MTW passengers a large proportion of road traffic deaths
around the world? Does the road safety community need to focus
special attention on the issue?
• Do countries with more MTWs have higher child road traffic death
rates? Should transport planners steer societies away from developmental trajectories that involve large MTW fleets?
In this paper, we summarize the literature on what is known about
the biomechanical tolerance of young children to impact forces, and
briefly review the global burden of road traffic injuries among young
children. We develop a database of population-level MTW and road
traffic injury statistics and explore the epidemiological evidence related
with MTW injuries among young children in countries across global
regions.
1.1. Biomechanics of injuries in young children
Despite large advances in knowledge about human injury tolerance,
the vulnerability of children relative to adults remains poorly understood. Over the last century, substantial efforts have been directed at
studying the capability of the human body to withstand external forces
and accelerations. The body regions most vulnerable in MTW crashes
(head, torso, and extremities) are commonly involved in many other
types of injury events and have been a central focus of many biomechanical investigations. Studies have focused on analyzing real-world
injury incidents, conducting experiments on human volunteers,
animals, and cadavers, and developing and validating mechanical
and computational models. Among these, cadaver experiments have
been particularly important because they can be subjected to injury-
inducing forces and have proven especially useful for understanding
the response of adult hard tissue. However, child cadavers have been
rarely used for biomechanical research due to ethical and logistical
constraints [13].
Children are not miniature adults from a biomechanical standpoint.
Human body segments grow at different rates during childhood leading
to large differences between the relative geometric and inertial properties of children and adults [14]. For instance, the infant head is 25% of its
standing height but the adult head is only 11% [15]. The infant head
reaches 90% of the size of the adult head by the age of four [10]. Similarly, the mechanical properties of human tissue evolve considerably
during childhood [16–18]. Experiments on human tissue suggest that
the elastic and viscous components of the complex shear modulus (a
measure of how much the tissue distorts during impact) of frontal
brain tissue increases significantly with age affecting the mechanical response of the brain to impact forces [17]. Furthermore, the frontal bone
of the skull in infants consists of two halves that are connected by a
dense connective tissue (suture) that fuses by the age of six. Since
sutures have a much lower elastic modulus, the pediatric skull case
can undergo large shape changes without permanent injury [16].
Nevertheless, the empirical evidence on age-dependent mechanical
characteristics of skull and brain tissue is still weak, and it remains unclear if children are more or less vulnerable to head injuries than adults.
The biomechanics of the neck, thorax, and extremities of children is similarly poorly understood [13].
In practice, therefore, injury thresholds used in the design and development of protective equipment, such as helmets, are based on the
biomechanical tolerances of adults. The benchmark for estimating the
probability of head injuries is the head injury criteria (HIC), which
uses a time-averaged weighted integral of linear acceleration measured
at the center of gravity of the head [19]. Helmet test standards use
methods that are even simpler. The US design standard for motorcycle
helmets (FMVSS 218) requires that accelerations never exceed 400 g
and do not exceed 200 g for more than 4 ms during a guided free fall
of a helmeted head-form from a height of 1.8 m [20]. Other agencies require similar tests with peak accelerations limited to 275–300 g [20].
There have been substantial efforts to adapt these testing methods to
the needs of child helmets but there is relatively little consensus on
key test parameters, such as the mass of the test head-form and acceleration thresholds for injury [21].
K. Bhalla, D. Mohan / IATSS Research 38 (2015) 83–91
The free fall drop height in helmet tests (1.8 m) is similar to the vertical elevation above ground level of the head of a standing MTW rider.
This is because the trajectory of the head and body in an MTW crash can
be roughly approximated as a projectile that has been imparted a horizontal velocity. The laws of Newtonian mechanics dictate that the head
will hit the ground with a vertical velocity that is equal to that produced
by free fall (i.e. under acceleration due to gravity) from the original
height. Falls from a height of 2 m have been shown to cause serious
head injuries to children [22].
In summary, it remains unclear if children are more vulnerable to
injuries than adults in MTW crashes. However, crash forces generated in MTW crashes are expected to exceed injury thresholds for
both adults and children. Therefore, all MTW riders are vulnerable
in crashes.
1.2. Global burden of MTW injuries among young children
Road traffic injuries are a large global public health problem for all
ages after the first year of life [23,24]. Overall, road traffic injuries
are the 8th leading cause of death and 10th leading cause of health
loss (measured in disability adjusted life years lost, DALYs), killing
1.3 million and injuring 78 million people annually [23,25]. Among
infants under 1 year, infectious diseases dominate global mortality
and morbidity and road traffic injuries are relatively rare even in settings where MTWs are prevalent. Thus, among infants aged 1 month
to 1 year, road traffic deaths rank relatively low — 19th leading cause of
death globally, 19th in Southeast Asia, and 26th in South Asia. Road traffic injuries emerge as a health issue among children 1–4 years old for
whom they are the 9th leading cause of death and 10th leading cause
of health loss [24]. They become more prominent among 5–9 year
olds, rising to the 5th leading cause of death with only 28% fewer deaths
than diarrheal disease, which is the leading cause of death [23]. Therefore, addressing road safety is important globally for the health of
young children after the first year of life.
While MTW crashes are a large proportion of road traffic crashes in all
ages, they constitute a substantially smaller proportion among children
(Fig. 1). Overall, MTW crashes resulted in an estimated 206,435 deaths
globally in 2010, 15% of all road traffic deaths [23]. Among young children,
MTW crashes resulted in an estimated 4514 deaths in children under 5
years, and 2616 deaths in children aged 5–9 years [23]. MTW deaths
amount to 6% and 8% of road traffic deaths in these age groups, which is
substantially lower than the proportion for all age groups. Notably, substantially more young children die as pedestrians (45,972 deaths among
0–9 years) than as MTW riders even in settings where MTW use is high.
Road Traffic Deaths, All ages
Therefore, while addressing MTW safety is important for road safety overall, MTW safety for young children is a comparatively smaller problem.
MTW use varies dramatically globally. Therefore, it is important to
consider the issue at the regional, national and city level. Estimated
MTW death rates are highest in Western sub-Saharan Africa (6.0 per
100,000 people) and South East Asia (5.7) [23]. These rates are about
twice the global average (3.0), and ten times the rate in Central Asia
(0.6), where MTW death rates are lowest [23]. Understanding the
impacts on the safety of young children requires characterizing the
age-distribution of MTW injuries in countries with varying amounts of
MTW use.
Mohan [26] conducted a review of epidemiological studies that reported the age distribution of MTW victims in regions where MTWs
are prevalent. Although the review found very few studies with such
data, a consistent finding was that young children were a relatively
small proportion of MTW victims. In a hospital in Chennai (India), 67%
of 2748 patients admitted between 1999 and 2005 with maxillofacial
injuries were MTW riders but only 3% were child MTW passengers
under 10 years [27]. Similarly, in Delhi (India), only 3% of MTW injuries
seen at a major hospital were children 0–14 years [28]. In Taipei
(Taiwan), in a study of 1160 MTW victims with craniofacial injuries,
only 2% were children aged 0–15 years [29]. In Northeast Thailand, children under 10 years comprised 2.1% to 3.8% of MTW injuries before and
after enforcement of a helmet law [30]. Finally, in a similar study in
Taiwan, children under 10 comprised 1.0% to 1.2% of fatalities before
and after a helmet law was implemented [31]. Although the reason for
the increase in percentage of children after the introduction of the helmet law in these two studies is unknown, it is clear that the percentages
are always small.
Unfortunately, existing global databases that report country-level
road traffic deaths do not allow cross-country comparisons of MTW
injuries among young children. The WHO Global Status Reports on
Road Safety (GSRRS) [32,33] present country-reported and modeled estimates of MTW deaths in all-countries but do not report deaths disaggregated by age. The 2010 Global Burden of Disease (GBD-2010) project
provided estimates of MTW injuries disaggregated by age in all regions.
However, the underlying mortality data used for GBD-2010 estimates
included relatively little information on age and transport-mode of
road injuries in the regions where MTWs are most common. As a result,
GBD-2010 estimates of childhood deaths and injuries on MTWs are
unlikely to be reliable for such investigation.
Therefore, we compiled statistics on MTW crashes from databases
that included reliable statistics on road traffic injuries disaggregated
by age and mode of transport and compared the incidence of child
MTW injuries across countries and cities. We discuss how risk to
Road Traffic Deaths, <5 years
Other
8%
85
Road Traffic Deaths, 5-9 years
Other
9%
Other
9%
Pedestrians
35%
Occupants
(3+ wheels)
36%
Pedestrians
47%
Occupants
(3+ wheels)
35%
Pedestrians
43%
Occupants
(3+ wheels)
32%
Bicyclists
6%
MTWs
15%
MTWs
8%
MTWs
6%
Bicyclists
3%
Bicyclists
8%
Source: GBD-201023
Fig. 1. Distribution of global road traffic deaths in 2010 by mode of transport among young children and people of all-ages.
86
K. Bhalla, D. Mohan / IATSS Research 38 (2015) 83–91
children varies across settings with varying amounts of MTWs in their
transport fleet.
2. Data sources and methods
2.1. Estimating the global MTW fleet
We obtained vehicle registration data from official statistics provided by country governments to WHO as reported in the 2013 WHO
GSRRS [33]. This included data from 161 of 185 countries. The years
for which vehicle registration data were available ranged from
2007–2011 (2010 or 2011 in 87% of these countries). For selected missing countries that are known to have large MTW fleets, we obtained statistics from alternate sources. These included Argentina [34], China [35],
Japan [36], and Uganda [37]. No estimates of registered vehicles and
registered MTWs were available from countries that together comprised 2.2% and 6.2% of the global population, respectively. For these
countries with missing data, we assumed regional vehicle and MTW
ownership levels.
2.2. Estimating age-specific MTW crashes
We constructed a population-level database containing agedisaggregated MTW injury statistics using information from the following sources:
1. WHO Mortality Database: ICD-coded cause of death tabulations disaggregated by age were obtained from the WHO Mortality Database,
which reports the underlying causes of deaths reported by national
civil registration systems [38]. We classified injury deaths by road
user according to ICD-based definitions developed by the GBD2010 Injury Expert Group. We audited the data for coverage, completeness, and quality of cause-of-death coding using a method
developed previously [39]. The method for the quality audit primarily assesses the proportion of deaths that area assigned to
less-specific causes of death (such as unspecified road user; unspecified accidents, unspecified injury, and unspecified causes of
death) that are commonly used on death certificates. If the most
recent data did not pass audit, we used the most recent year for
which the quality threshold was met. For selected countries with
small populations, we aggregated data for multiple years. Finally,
we restricted analysis to countries that had more than 100 MTW
deaths.
2. IRTAD: We acquired data from the International Road Traffic Accident
Database [40] for the year 2005 and retained data for 17 countries
that reported fatalities disaggregated by age and had more than
100 MTW deaths in 2005. IRTAD reports road traffic deaths standardized to a 30-day definition. For several countries, where we also had
data from the WHO mortality database, we used IRTAD data because
it is expected to be of higher quality [39].
3. UNECE Data: We obtained road traffic deaths from UNECE Transport
Statistics [41] disaggregated by road user and age groups for 5
European countries that were not included in IRTAD. The UNECE reports deaths standardized to a 30-day definition.
4. Indian Cities: Age distribution of MTW deaths is not readily available
in most of urban South Asia [42]. We obtained data from traffic police
in five Indian cities for deaths caused by a road traffic crash. Traffic
police reports in India include road traffic deaths regardless of when
the death occurs after the crash — i.e. there is no time restriction
when the death occurs as an admitted ‘accident’ patient in a hospital.
In four of these cities, we assessed the text-based first information reports for the period 2007–2012 [43]. We identified MTW crashes
where the victim was identified as “boy” or “girl”, which we assumed
to be a child under 10.
5. Hospital Admissions: We extracted data on MTW victims from the
GBD-2010 Injury Expert Group's hospital data collection [44] for
the 17 countries where the database contained more than 100
MTW injury admissions.
6. Other Data Sources: In addition to the above, we included the following sources that provided MTW injuries disaggregated by age:
national death registration data from Iran [45], police data from
Malaysia [3,46], and mortuary data from a city (Ibadan) in Nigeria
[44].
We grouped ages into b 5 years, 5–9 years, and older, except for the
data from Indian cities where a single child age group 0–9 years was
used. The two childhood age groups were chosen because these groups
have different implications for helmet policy. In order to compute per
capita injury rates, we used population data from the UN population
division [47].
While the IRTAD and UNECE data exclude deaths that happen more
than 30 days after the crash, data from the WHO mortality database and
Indian cities have no time restriction. However, according to the 2013
WHO Global Status Report on Road Safety, only 3% of deaths occur
more than 30 days after a crash.
Throughout, by MTWs we mean two-wheeled motorized vehicles,
which include mopeds, scooters and motorcycles. In principle, this
also includes bicycles with electric motors (e-bikes) that are becoming
increasingly popular in many regions. While we expect that our MTW
injury data include e-bike passengers, it is likely that the vehicle fleet
statistics only include e-bikes if the local government requires registration of these vehicles. In China, for instance, there are about 100 million
e-bikes that are not included in the fleet statistics [48].
In all, the database contained mortality statistics for 44 countries and
5 Indian cities, and hospital admissions from 17 countries. The MTW
fleet in these countries ranged from 2% to 70% of all registered vehicles.
We classified the countries and cities into Low MTW Use, where MTWs
comprised less than 5% of the vehicle fleet (12 countries), Medium
MTW Use, 5–20% of vehicle fleet (21 countries), and High MTW Use,
N20% of vehicle fleet (11 countries and 5 Indian cities).
3. Results
We estimate that there are approximately 434 million MTWs comprising 28% of the vehicle fleet in the world (see Table 1). Approximately 74% of the global MTW fleet is in East, South, and Southeast Asia. The
proportion of vehicles that is MTWs varies dramatically by region,
reaching 70% or more of the vehicle fleet in South and Southeast Asia.
In 8 of 21 global regions, MTWs comprise more than one-fourth of the
vehicle fleet. In addition to the aforementioned regions, these include
the Caribbean, Western Sub-Saharan Africa, and Andean, Tropical and
Southern Latin America. These 8 regions together have 63% of the global
population implying that most of the global population lives in regions
where MTWs comprise a substantial proportion of the vehicle fleet.
Per capita MTW ownership levels are high in some regions where
MTWs comprise a relatively small portion of the vehicle fleet. For
instance, although MTWs only comprise 11% of the vehicle fleet in
Western Europe, ownership levels in the region (67 MTWs per 1000
people) are higher than that in South Asia [55].
MTW deaths among young children (under 10) comprised at most
1.5% of all road traffic deaths in the countries in the database
(Fig. 2). India-Amritsar (1.50%), Paraguay (0.99%), and India-New
Delhi (0.87%) had the highest percentages. On average, 0.16% (SD
0.29%) of all road traffic deaths were young children on MTWs in these
countries. Children under 5 years averaged 0.05% (SD 0.13%), and 5–9
years averaged 0.11% (SD 0.25%) of all road traffic deaths. As expected,
settings with High MTW Use had a higher percentage of child MTW
deaths. However, even in these settings, the average percentage of
child MTW deaths was only 0.41% (SD 0.41%), with children under 5
years averaging 0.19% (SD 0.20%) and 5–9 year olds 0.28% (SD 0.39%).
Restricting attention to only MTW deaths (Fig. 3a), young children
comprised less than 6.0% (mean 0.74%, SD 1.13%) in all countries.
K. Bhalla, D. Mohan / IATSS Research 38 (2015) 83–91
87
Fig. 2. Percentage of road traffic injury deaths that are young children on MTWs.
Egypt (5.8%), India-Amritsar (3.8%), and India-New Delhi (2.8%) had the
highest percentages. In general, this percentage was higher in settings
with higher MTW use. However, even in the High MTW Use settings,
the percentage of MTW deaths that were young children averaged
only 1.52% (SD 1.50%). Among children under 5 years, this percentage
averaged at 0.82% (SD 0.79%) and among 5–9 year olds, averaged at
0.95% (SD 1.16%).
In contrast, the percentage of all road traffic deaths that are young
children is much higher (Fig. 3b). This percentage ranges from 1.0% to
12.1% and has an average of 2.81%, SD 1.94%. With three exceptions
(Latvia, Poland, and India-Amritsar), the percentage of all road traffic
deaths that are young children was always higher than the percentage
of MTW deaths that are young children. In two-thirds of these countries,
it was more than 5 times higher. Among children under 5 years, this
percentage averaged 1.53% (SD 1.15%), about six times higher than
the corresponding percentage of MTW deaths in this age group. Similarly, among 5–9 year old children, this percentage averaged 1.4% (SD
1.09%), more than twice the corresponding percentage of MTW deaths
in this age group. The percentage of road traffic deaths that are young
children is slightly higher in settings with more MTW use, increasing
from an average of 2.66% (SD 1.44%) in Low MTW Use settings and
1.90% (SD 0.8%) in Medium MTW Use settings, to 4.08% (SD 2.43%) in
High MTW Use settings.
The population age-structure varies substantially across the countries in the database. While young children (under 10 years) are only
8% of the population in Germany and Japan, they are over 30% in
Nigeria. Therefore, we adjusted for the age structure by estimating the
child to population rate ratio — i.e. the ratio of injury rates among children to injury rates among the general population (Fig. 4). The child-topopulation rate ratio for all road traffic deaths averaged 0.20 (SD 0.08).
In other words, the average rate of road traffic crashes among children is
one-fifth of the rate for the general population. In comparison, the rate
ratio of MTW deaths was only 0.05 (SD 0.07). Further, the rate ratios for
all road traffic deaths did not show a trend with increasing use of MTW.
The child-to-population rate ratio for road traffic deaths varied from
0.23 (SD 0.08) in Low MTW Use countries, 0.18 (SD 0.03) in Medium
MTW Use countries, to 0.21 (SD 0.11) in High MTW Use countries.
Thus far we have presented results related with fatalities. If we consider non-fatal hospital admissions (Fig. 5), the percentage of non-fatal
MTW injuries that are young children is substantially higher than the
corresponding percentage for fatalities. On average, young children
comprised 1.89% (SD 1.61%) of non-fatal hospital admissions for MTW
crashes, compared with 0.74% of fatalities. This was also true for both
age sub-groups. Children under 5 years comprised 0.64% of non-fatal
hospital admissions for MTW crashes, compared with 0.26% of fatalities.
Similarly, children 5–9 years old comprised 1.25% of non-fatal admissions, compared with 0.51% of fatalities. We had both non-fatal admissions statistics and fatality statistics for 9 countries. In each of these
countries, the percentage of non-fatal MTW admissions that are young
children was higher than the corresponding percentage for fatalities.
Notably, in Argentina, where the difference was largest, 5.0% of nonfatal MTW admissions were young children compared with 0.4% of
fatalities.
Fig. 6 illustrates the relationship between the percentage of road
traffic deaths among young children that are MTW riders and the extent
of MTW use in each country. Also shown is a 45-degree line, which corresponds to the two growing at the same rate. In settings where MTW
are a small proportion of the vehicle fleet, there was substantial scatter
in the percentage of child road traffic deaths that were MTW riders.
MTW comprised less than 10% of the vehicle fleet in four of the five
countries that were above the 45-degree line (UK, Latvia, France, and
Spain). India-Amritsar was an outlier with an unusually high percentage
of deaths (80%). In general, the percent of child road deaths that were on
MTWs grew linearly in countries with higher MTW use but at a much
slower rate than the growth in MTWs.
4. Discussion
Our analysis shows that young children rarely die in MTW crashes
even in settings where MTWs are a common mode of transportation.
88
K. Bhalla, D. Mohan / IATSS Research 38 (2015) 83–91
(a)
(b)
Fig. 3. Proportion of MTW deaths (a), and road traffic deaths (b), that are young children.
Reasons for using MTWs vary across the globe. Our database included several countries where MTWs are commonly used to transport children, notably including Brazil, Iran, five cities in India,
Thailand, and Uganda. However, even in these settings, young
children comprised at most 1.5% of all road deaths and 5.8% of all
MTW deaths. These findings are broadly consistent with several
previous studies that have presented MTW injuries disaggregated
by age [27–31].
K. Bhalla, D. Mohan / IATSS Research 38 (2015) 83–91
89
Fig. 4. Comparison of child-to-population rate ratios for MTW fatalities and road traffic fatalities.
Children were a larger proportion of road traffic deaths in countries
where MTWs were more common. However, countries with more
MTWs tend to also have a larger proportion of children in their population. Therefore, we adjusted for population age structure by estimating
child-to-population rate ratios. These rate ratios for MTW deaths did not
show an increase with increasing MTW use.
It is surprising that children do not comprise a larger proportion of
road traffic and MTW fatalities. The reasons for this finding deserve to
Fig. 5. Proportion of non-fatal MTW injury deaths that are among young children.
90
K. Bhalla, D. Mohan / IATSS Research 38 (2015) 83–91
Child MTW deaths, % of all child road deaths
80%
60%
40%
-d
45
x
y=
e,
lin
ee
r
eg
20%
y = 0.2919x
0%
0%
20%
40%
60%
80%
Registered MTWs, % of all registered vehicles
Fig. 6. Relationship between registered MTWs and child MTW deaths.
be systematically investigated. First, it is possible that the exposure of
young children to MTWs may be lower than people perceive. Humans
may have a higher tendency to remember unusually risky events and
hence overestimate their prevalence. In our literature review, most
references to the common use of MTWs to transport children were
anecdotal [1–8] and we found no objective measurements of the prevalence of children on MTWs. Observation studies of MTWs are conducted
frequently, usually for measuring prevalence of helmet use [49–51].
Such studies could be easily extended to provide estimates of the prevalence of young children on MTWs.
Second, it is possible that MTWs carrying child passengers have
lower risk of crash involvement. It is possible that MTW drivers with
children on board may drive at lower speeds than when alone and
other drivers may become more careful in the vicinity of MTWs with
children. An accompanying paper in this issue (Why do Three-Wheelers
Carrying School Children Suffer Very Low Fatal Crashes?) documents the
fact that three-wheeled scooter taxis with child occupants had very
low crash rates.
Third, several studies have documented a “safety in numbers” effect
for bicyclists, where the risk of crashes to individual bicyclists is lower in
settings where bicycling is more common [52,53]. It is reasonable to expect that such an effect would apply also to larger two-wheelers, such as
mopeds and low powered MTWs that are common in many parts of the
world. In our analysis, child MTW deaths were disproportionately high
in settings with relatively few MTWs but disproportionately low in settings with high prevalence of MTWs (Fig. 4), suggesting the possibility
of a safety in numbers effect for MTWs. Simulator studies show that
car drivers detect MTWs quicker in settings with a high prevalence of
MTWs compared with settings where MTWs are relatively rare [54].
Such studies build on a large body of work that shows that humans
are more likely to miss low-prevalence targets. Similarly, research suggests that car drivers who have owned a motorcycle are less likely to be
involved in crashes with motorcycles, again suggesting that MTW riders
may be safer in settings with high MTW use [55]. Although these studies
are not specific to children, they suggest lower risks for MTW riders of
all age groups.
Finally, our analysis suggests that young children may be more commonly injured than killed in MTW crashes, although the percentages are
small. This may be due to the fact that MTWs with child passengers may
be driven at lower speeds than those without children. On the other
hand, it is also possible that children are more likely to be hospitalized
in relatively minor crashes where the accompanying adult did not
require admission. We had relatively few countries from High MTW
Use settings where such a comparison was possible, and systematic
investigations with more data would likely provide insights.
It is important to note the limitations of the data that we have
analyzed. While national death registration systems in Latin America
provide reliable information on age and mode of transport for national
road traffic deaths, such data were sparse in other regions where
MTW are prevalent and commonly used to transport young children.
We had comparatively little information from South and Southeast
Asia, and none from East Asia. Systematically collecting data from
these settings would help build confidence in our findings as well as
help explain their underlying mechanisms.
Although we have shown that MTWs are not a major threat to young
children, road safety is nevertheless an important issue for children
after the first year of life [25,33]. Globally and regionally, about threefourths of all road traffic deaths among young children are either pedestrians or vehicle occupants. It is interesting to note that bicycle use is
being promoted globally for health and environmental reasons but carrying children on bicycles does not evoke the same policy debates as
carrying them on MTWs. Child pedestrian fatalities were also substantially higher in the locations studied. An argument could be made that
if all child pedestrians and bicyclists wore helmets, more lives would
be saved than MTWs only.
The WHO World Reports on Road Traffic Injury Prevention [9] and
Child Injury Prevention [56] provide a wide range of strategies ranging
from increasing adult supervision of children on streets, to engineering
approaches, such as safer infrastructure around schools and playgrounds. These recommendations are grounded in a comprehensive
public health approach of systematically identifying risk factors and
appropriately prioritizing interventions. Children hold a special place
in most societies, making it particularly important that we are guided
by evidence when planning for their health and safety.
References
[1] World Health Organization Regional Office for South-East Asia, Recommendations
of the Expert Group on Preventing Motorcycle Injuries in Children, World Health
Organization SEARO, New Delhi, 2010.
[2] World Health Organization Country Office for Vietnam, The Case for Motorcycle
Helmet Wearing in Children: A Submission to the Ministry of Transport, World
Health Organization, 2009.
[3] M.M. Abdul Manan, A. Várhelyi, Motorcycle fatalities in Malaysia, IATSS Res. 36 (1)
(2012) 30–39.
[4] A. Pervin, Viet Nam's mandatory motorcycle helmet law and its impact on children,
Bull. World Health Organ. 87 (5) (2009) 369–373, http://dx.doi.org/10.2471/BLT.08.
057109.
[5] Sub-Saharan Africa Transport Policy Program, Understanding the Emerging Role
of Motorcycles in African Cities, World Bank, 2011. (Available at: http://www.
worldbank.org/afr/ssatp).
[6] M. Nakitto, R. Lett, Motorised transport for kindergarten children in Uganda, Inj. Prev.
16 (Supplement 1) (2011) A46, http://dx.doi.org/10.1136/ip.2010.029215.168.
[7] S. Saadat, H. Soori, Epidemiology of traffic injuries and motor vehicles utilization in
the capital of Iran: a population based study, BMC Public Health 11 (2011) 488,
http://dx.doi.org/10.1186/1471-2458-11-488.
[8] A.L. Cavalcanti, C.R. Barros de Alencar, Injuries to the head and face in 0–4-year-old
child victims of fatal external causes in Campina Grande, PB, Brazil, Turk. J. Pediatr.
52 (6) (2010) 612–617.
[9] World Health Organization, World Report on Road Traffic Injury Prevention, World
Health Organization, Geneva, 2004.
[10] K.B. Arbogast, S.S. Margulies, M. Patlak, H. Fenner, D. Thomas (Eds.), Review of Pediatric Head and Neck Injury: Implications for Helmet Standards, Snell Memorial
Foundation and The Children's Hospital of Philadelphia, 2003.
[11] World Health Organization, Helmets: A Road Safety Manual for Decision Makers and
Practitioners, World Health Organization, Geneva, 2006.
[12] G.B. Melton, Treating Children Like People: A Framework for Research and Advocacy, J. Clin. Child Adolesc. Psychol. 34 (4) (2005) 646–657, http://dx.doi.org/10.1207/
s15374424jccp3404_7.
[13] C. Jenny, K.P. Hymel, N. Rangarajan, Biomechanics of Pediatric Head Injury, Springer
New York, New York, NY, 2014, pp. 435–453, http://dx.doi.org/10.1007/978-161779-403-2_9.
[14] R.K. Jense, Body segment mass, radius and radius of gyration proportions of children,
J. Biomech. 19 (1986) 359–368.
[15] A. Dimeglio, Growth in pediatric orthopaedics, J. Pediatr. Orthop. 21 (4) (2001)
549–555.
[16] B. Coats, S.S. Margulies, Material properties of human infant skull and suture at high
rates, J. Neurotrauma 23 (8) (2006) 1222–1232, http://dx.doi.org/10.1089/neu.
2006.23.1222.
[17] Experimental Injury Biomechanics of the Pediatric Head and Brain, Springer New York,
New York, NY, 2012, pp. 157–189, http://dx.doi.org/10.1007/978-1-4614-4154-0_4.
K. Bhalla, D. Mohan / IATSS Research 38 (2015) 83–91
[18] K.L. Thibault, S.S. Margulies, Age-dependent material properties of the porcine
cerebrum: effect on pediatric inertial head injury criteria, J. Biomech. 31 (12)
(1998) 1119–1126, http://dx.doi.org/10.1016/S0021-9290(98)00122-5.
[19] N. Yoganandan, F.A. Pintar, Biomechanics of temporo-parietal skull fracture, Clin.
Biomech. 19 (3) (2004) 225–239, http://dx.doi.org/10.1016/j.clinbiomech.2003.12.
014.
[20] M.J. Vander Vorst, P.C. Chan, Biofidelity of Motorcycle Helmet Criteria, Proceedings
of the 33rd International Workshop on Injury Biomechanics Research, Washington
DC, 2005.
[21] Becker E. Snell's response to CSA child helmet standard, 2013, Snell Foundation.
Available at: http://www.smf.org/docs/articles/csa. Accessed July 2014.
[22] D. Mohan, B.M. Bowman, R.G. Snyder, D.R. Foust, A biomechanical analysis of head
impact injuries to children, ASME J. Biomech. Eng. (1979) 250–260.
[23] R. Lozano, M. Naghavi, K. Foreman, et al., Global and regional mortality from 235
causes of death for 20 age groups in 1990 and 2010: a systematic analysis for the
Global Burden of Disease Study 2010, Lancet 380 (9859) (2012) 2095–2128,
http://dx.doi.org/10.1016/S0140-6736(12)61728-0.
[24] C.J.L. Murray, T. Vos, R. Lozano, et al., Disability-adjusted life years (DALYs) for 291
diseases and injuries in 21 regions, 1990–2010: a systematic analysis for the Global
Burden of Disease Study 2010, Lancet 380 (9859) (2012) 2197–2223, http://dx.doi.
org/10.1016/S0140-6736(12)61689-4.
[25] K. Bhalla, M. Shotten, A. Cohen, et al., Transport for Health: the Global Burden of Disease Due to Injuries and Air Pollution From Motorized Road Transport, World Bank
Global Road Safety Facility, and Institute for Health Metrics and Evaluation, Washington, DC, 2014. (Available at: http://www.healthmetricsandevaluation.org/gbd/
publications/policy-report/transport-health-global-burden-disease-motorizedroad-transport).
[26] D. Mohan, Safety of Children as Motorcycle Passengers, Proceedings of the Eastern
Asia Society of Transportation Studies, 2013, p. 9.
[27] K. Subhashraj, N. Nandakumar, C. Ravindran, Review of maxillofacial injuries in
Chennai, India: a study of 2748 cases, Br. J. Oral Maxillofac. Surg. 45 (8) (2007)
637–639, http://dx.doi.org/10.1016/j.bjoms.2007.03.012.
[28] B.K. Mishra, A.K. Banerji, D. Mohan, Two-wheeler injuries in Delhi, India: a study of
crash victims hospitalized in a neuro-surgery ward, Accid. Anal. Prev. 16 (5) (1984)
407–416.
[29] M.C. Lee, W.-T. Chiu, L.T. Chang, S.C. Liu, S.H. Lin, Craniofacial injuries in unhelmeted
riders of motorbikes, Injury 26 (7) (1995) 467–470.
[30] M. Ichikawa, W. Chadbunchachai, E. Marui, Effect of the helmet act for motorcyclists
in Thailand, Accid. Anal. Prev. 35 (2) (2003) 183–189.
[31] W.-T. Chiu, C.Y. Kuo, C.C. Hung, M. Chen, The effect of the Taiwan motorcycle helmet
use law on head injuries, Am. J. Public Health 90 (5) (2000) 793–796.
[32] World Health Organization, Global Status Report on Road Safety — Time of Action,
2009. 1–300.
[33] World Health Organization, Global Status Report on Road Safety — Supporting a Decade of Action, World Health Organization, Geneva, 2013.
[34] Swiss Argentine Chamber of Commerce, Argentina — Motorcycle Industry:
Trendspotting, Switzerland Global Enterprise, 2013.
[35] Ministry of Public Security of the People's Republic of China. National vehicle population. Available at: http://www.mps.gov.cn/n16/n1282/n3553/3413423.html.
Accessed July 2014.
91
[36] Lillehammer, Trends in Motorcycles Fleet Worldwide, OECD and ITF, 2008. 1–26.
[37] International Road Federation, World Road Statistics, 2009.
[38] World Health Organization, WHO Mortality Database, WHO, 2014. (Available at:
http://www.who.int/healthinfo/mortality_data/).
[39] K. Bhalla, J.E. Harrison, S. Shahraz, L.A. Fingerhut, Global Burden of Disease Injury Expert Group. Availability and quality of cause-of-death data for estimating the global
burden of injuries, Bull. World Health Organ. 88 (11) (2010) 831C–838C, http://dx.
doi.org/10.2471/BLT.09.068809.
[40] IRTAD, Road Safety Annual Report, International Transport Forum, 2014. 1–526.
[41] UNECE Statistical DatabaseAvailable at: http://w3.unece.org/pxweb2014.
[42] A.A. Hyder, O.H. Amach, N. Garg, M.T. Labinjo, Estimating the burden of road traffic
injuries among children and adolescents in urban South Asia, Health policy 77 (2)
(2006) 129–139, http://dx.doi.org/10.1016/j.healthpol.2005.07.008.
[43] D. Mohan, G. Tiwari, S. Mukherjee, A study on community design for traffic safety,
IATSS Report, 2013.
[44] K. Bhalla, J.E. Harrison, S. Shahraz, et al., Burden of Road Injuries in Sub-Saharan
Africa, Department of Global Health and Population, Harvard School of Public
Health, Boston, MA, 2013.
[45] K. Bhalla, M. Naghavi, S. Shahraz, D. Bartels, C.J.L. Murray, Building national estimates of the burden of road traffic injuries in developing countries from all available
data sources: Iran, Inj. Prev. 15 (3) (2009) 150–156, http://dx.doi.org/10.1136/ip.
2008.020826.
[46] N. Mohamed, W.S. Voon, H.H. Hashim, I. Othman, An Overview of Road Traffic Injuries Among Children in Malaysia and Its Implication on Road Traffic Injury Prevention Strategy, MIROS Malaysian Institute of Road Safety Research, 2011.
[47] United Nations Population Division, World Population Prospects, 2010.
[48] D. Cusick, et al., Can e-bikes displace cars? Sci. Am. (2012) (Available at: http://
www.scientificamerican.com/article/can-ebikes-displace-cars/).
[49] A.M. Bachani, N.T. Tran, S. Sann, et al., Helmet use among motorcyclists in
Cambodia: a survey of use, knowledge, attitudes, and practices, Traffic Inj. Prev.
13 (Suppl. 1) (2012) 31–36, http://dx.doi.org/10.1080/15389588.2011.630763.
[50] A.M. Bachani, C. Branching, C. Ear, et al., Trends in prevalence, knowledge, attitudes,
and practices of helmet use in Cambodia: results from a two year study, Injury 44
(Suppl. 4) (2013) S31–S37, http://dx.doi.org/10.1016/S0020-1383(13)70210-9.
[51] F. Zamani-Alavijeh, M. Bazargan, A. Shafiei, S. Bazargan-Hejazi, The frequency and
predictors of helmet use among Iranian motorcyclists: a quantitative and qualitative
study, Accid. Anal. Prev. 43 (4) (2011) 1562–1569, http://dx.doi.org/10.1016/j.aap.
2011.03.016.
[52] P.L. Jacobsen, Safety in numbers: more walkers and bicyclists, safer walking and bicycling, Inj. Prev. 9 (3) (2003) 205–209, http://dx.doi.org/10.1136/ip.9.3.205.
[53] D.L. Robinson, Safety in numbers in Australia: more walkers and bicyclists, safer
walking and bicycling, Health Promot. J. Aust. 16 (1) (2005) 47–51.
[54] V. Beanland, M.G. Lenné, G. Underwood, Safety in numbers: target prevalence affects the detection of vehicles during simulated driving, Atten. Percept. Psychophys.
76 (3) (2014) 805–813, http://dx.doi.org/10.3758/s13414-013-0603-1.
[55] D. Magazzù, M. Comelli, A. Marinoni, Are car drivers holding a motorcycle licence less
responsible for motorcycle–car crash occurrence? A non-parametric approach, Accid.
Anal. Prev. 38 (2) (2006) 365–370, http://dx.doi.org/10.1016/j.aap.2005.10.007.
[56] World Health Organization, World Report on Child Injury Prevention, World Health
Organization, Geneva: Geneva, 2008. 1–232.