ArticlePDF Available

For whom didn’t it click? A study of the non-use of seat belts in motor vehicle fatalities in New Zealand


Abstract and Figures

There is an increased risk of death or serious injury for occupants who did not wear a seat belt in a crash. In New Zealand, between 2006 and 2016, the non-use of seat belts accounted for 19-30% of the overall motor vehicle road deaths, and this figure shows no sign of decreasing. It is important to better understand the contextual factors associated with crashes where seat belts are not worn, so that more relevant and effective road safety interventions can be designed and implemented. The aim of this research was to determine the profiles for seat belt non-users who were killed in motor vehicle crashes in New Zealand between 2011 and 2015. An in-depth analysis of 200 fatalities where seat belts were not worn (186 crash cases) was carried out following a Safe System framework, using NZ Police reports. Following this, a Multiple Correspondence Analysis (MCA) developed five profiles of vehicle occupants who were killed in crashes where seat belts were not worn. While the stereotypical ‘young risky’ males were an important group, a range of other people and contexts emerged: ‘driving for work’; ‘elderly and retired’; ‘overseas passengers’; and ‘people driving in rural settings’. This has implications for tailored road safety interventions, as a variety of motivations and influences are likely to be at play, depending on the people involved.
Content may be subject to copyright.
Journal of the Australasian College of Road Safety – Volume 30, Issue 3, 2019
For whom didn’t it click? A study of the non-use of seat belts
in motor vehicle fatalities in New Zealand
Lily Hirsch1, Hamish Mackie1, Richard Scott1, John de Pont2, Simon Douglas3, and Dylan Thomsen3
1 Mackie Research, Auckland, New Zealand
2 TERNZ, Auckland, New Zealand
3 AA Research Foundation, Wellington, New Zealand
Corresponding Author: Lily Hirsch, PO Box 106525, Auckland City, Auckland, 1143, New Zealand,, +64 9 394 7040
Key Findings
Seat belts substantially reduce the likelihood of injury or death in a crash
In New Zealand, between 2006-2016, vehicle occupant fatalities where a seat belt was not worn accounted for 19-30%
of the total road fatalities
The research identied ve occupant proles for people who did not wear a seat belt and died on New Zealand’s roads
The development of proles can lead to better-targeted safety initiatives
There is an increased risk of death or serious injury for occupants who did not wear a seat belt in a crash. In New Zealand,
between 2006 and 2016, the non-use of seat belts accounted for 19-30% of the overall motor vehicle road deaths, and this
gure shows no sign of decreasing. It is important to better understand the contextual factors associated with crashes where
seat belts are not worn, so that more relevant and effective road safety interventions can be designed and implemented. The
aim of this research was to determine the proles for seat belt non-users who were killed in motor vehicle crashes in New
Zealand between 2011 and 2015. An in-depth analysis of 200 fatalities where seat belts were not worn (186 crash cases)
was carried out following a Safe System framework, using NZ Police reports. Following this, a Multiple Correspondence
Analysis (MCA) developed ve proles of vehicle occupants who were killed in crashes where seat belts were not worn.
While the stereotypical ‘young risky’ males were an important group, a range of other people and contexts emerged: ‘driving
for work’; ‘elderly and retired’; ‘overseas passengers’; and ‘people driving in rural settings’. This has implications for
tailored road safety interventions, as a variety of motivations and inuences are likely to be at play, depending on the people
Seat belt non-use, crash analysis, Safe System, prole development
It has been well documented that in a crash, occupants who
wear seat belts are less likely to experience serious injury
or fatal outcomes (Fildes et al., 2003; de Pont, 2016; Han,
2017). Seat belts protect vehicle occupants from crash forces
by retaining them in their seat during a crash, limiting their
movement, and managing the energy transmitted (World
Health Organisation, 2009; Road Safety Observatory, 2013).
For front seat drivers and passengers, seat belt use reduces
fatal and non-fatal crash injuries by between 40-60% (Høye,
2016; World Health Organisation, 2016). Likewise, for
rear seat passengers, seat belt use reduces fatality risk by
between 25-75% (World Health Organisation, 2016), and
also dramatically reduces fatality risk for front occupants
(Bose et al., 2013; Høye 2016).
In New Zealand, wearing a seat belt has been mandatory for
vehicle occupants since 1989. Surveys of vehicle occupants
generally show a high rate of compliance with these laws. In
2014, seat belt usage rates in the front seats were 97.1% and
92% for people seated in the rear. However, in 2016, front
seat usage rates dropped to 96.5% (Ministry of Transport,
2014; Ministry of Transport, 2016). These wearing rates may
not be representative of the entire New Zealand population,
however they are the most comprehensive rates available.
Between 2006 and 2016, fatalities where people were
not wearing a seat belt annually accounted for 19-30% of
the overall motor vehicle occupant road deaths. Over this
period, the proportion of these fatalities has uctuated but
in 2015 and 2016 seat belt non-use fatalities were at their
highest, accounting for 29-30% of all motor vehicle road
deaths (New Zealand Transport Agency, 2017). Note that
these gures were produced from a database query. The
number is likely to be an under-estimate of the true gures as
there are were several “unknown” entries under the ‘seat belt
wearing’ option.
Journal of the Australasian College of Road Safety – Volume 30, Issue 3, 2019
Many variables associated with the non-use of seat
belts, both in New Zealand and internationally are well
understood. For example: males are more likely to die in
crashes whilst not wearing a seat belt than women (Palamara
et al., 2009; Romano & Voas, 2011); drivers aged 75 and
older are most likely to wear a seat belt (Romano & Voas,
2011), whilst drivers in their late teens and early 20s are least
likely to wear a seat belt (Eluru & Bhat, 2007; Alver et al.,
2014); and seat belt usage can be understood as an equity
issue, with usage rates being lower among people with fewer
academic qualications (Begg & Langley, 2000; Demirer,
Durat & Haşimoğlu, 2012), and lower among people from
marginalised and minority ethnic backgrounds (Raftery &
Wundersitz, 2011; Shin et al., 1999).
Whilst there is some understanding of ‘why’ people do
not wear seat belts, mostly this information is understood
as individual variables only. For example: there is a link
between seat belt enforcement laws and wearing rates
(Shults et al., 2016; Bhat et al., 2012); for some people, the
discomfort of wearing, or the difculty of fastening a seat
belt may result in non-use particularly by those aged over
75 years, people who are obese, and people who experience
arthritis (Fong et al., 2016; Begg & Langley, 2012). Finally,
the inuence and attitudes of other people in the vehicle
and a person’s perceptions of the riskiness of a journey can
affect the ‘decision policy’ to wear or not to wear a seat belt
(Alattar et al., 2016).
The way in which factors associated with the non-use
of seatbelts interrelate is less well understood. This is an
important gap in the research as the complexity of humans
means that the isolated study of one variable will result in a
full picture. Therefore, understanding this interrelationship
of variables will give a fuller picture of the ‘proles’ of
people who did not wear seat belts and who were killed in
road crashes. This clearer understanding of ‘who’ does not
wear seat belts can lead to better and more informed research
to establish ‘why’ particular user groups do not wear seat
In the New Zealand context, the fact that these potentially
preventable deaths are not decreasing is an issue worthy of
investigation. The aim of this research was to understand
common contextual factors associated with seat belt non-
use fatalities for people aged fteen years and over in New
Zealand, and in doing so develop proles of seat belt non-
user types. This may lead to the design and implementation
of more relevant and effective road safety interventions.
The goal for the analysis was to understand the context
relating to fatalities where seat belts had not been worn. To
achieve this, the method was divided into two parts: 1) a
crash analysis of seat belt non-use fatalities in New Zealand
using a Safe System framework; and 2) the development of
occupant proles through MCA.
In New Zealand, between 2011-2015 there were 290 crash
cases where at least one fatally injured vehicle occupant was
not wearing a seat belt (New Zealand Transport Agency,
2017). Data from New Zealand’s Crash Analysis System
(CAS) database in the form of Trafc Crash Reports (TCRs)
and Serious Crash Unit (SCU) reports produced by NZ
Police were retrieved. Trafc crash reports are completed
by police ofcers at the scene of all road crashes. They
record the details of where, when, how and why the crash
happened. For fatal crashes, the Serious Crash Unit conducts
an in-depth investigation of the crash case to ensure all
causative factors are identied. These reports include
witness statements, blood analyses, photographs, and details
of the condition of the road and vehicle. Although serious
injury cases are relevant to this eld of research, they were
excluded from this study because the detail provided in
serious injury crash reports (TCR reports only) was not
Empirical Analysis
Criteria were developed which excluded 76 crash cases. The
criteria were: crashes involving a bus, tractor, or vehicles
where seat belts are not required; cases where people
travelled out of the vehicle i.e. the tray of a ute; crashes not
occurring on a public road; and unrestrained, or incorrectly
restrained children aged under 15 years. Of the remaining
crash list, each fatality was assigned a randomly generated
number using the MS Excel RAND function. These were
then sorted from the smallest to largest number and the rst
200 fatalities (186 crashes) were analysed for this study.
The TCR and SCU reports were coded into 53 nominal
and 10 continuous variables by a single analyst following
a Safe System framework which acknowledged that fatal
crashes happen when a range of system failures occur
(Larsson & Tingvall, 2013; New Zealand Government
& National Road Safety Committee, 2016). Each fatality
case was examined using variables relating to the four Safe
System Pillars: Speed; Roads and Roadsides; Vehicles; and
Users (New Zealand Government, & National Road Safety
Committee, 2016). As the aim of the research was ultimately
to understand occupant behaviour in relation to seat belt use,
the user pillar was investigated in-depth, whereas the other
pillars were more supercially explored. To ensure coding
rigour, ten ‘test’ cases were initially coded by the analyst,
then separately by the rst author. There was a strong
level of agreement, which is understandable given that
the exercise mostly involved identifying data, rather than
subjective coding.
Statistical Analysis
Whilst the involvement of many individual factors in seat
belt non-use crashes are well known, the combination
and pattern in which these factors present are less well
understood. In the R statistics programme and FactomineR
add-in package (Husson et al., 2014; Das et al., 2018), a
Multiple Correspondence Analysis (MCA) was conducted
Journal of the Australasian College of Road Safety – Volume 30, Issue 3, 2019
on 21 of the variables coded from the 200 fatality cases
(Table 1). MCA is an extension of correspondence analysis
(CA) because of its applicability to explore the association
between a large set of categorical variables rather than
ordinal data. Through its proximity mapping, MCA helps to
reveal the main features from a multi-dimensional dataset
(Das et al., 2018), analyse the correlations between the
category variables, and develop new composite variables
which are combinations of the category variables and are
independent of each other. The MCA analysis was used
as pre-processor for a Euclidean cluster analysis which
identied groups of individuals close to each other in terms
of composite variables. The aim was to detect and represent
the underlying relationships between variables and thereby
identify clusters or ‘proles’ of individual fatality cases with
similar characteristics.
Most of the variables in the database were nominal
categorical variables, for example, the vehicle type can be
“car”, “truck”, “van” etc. Some variables such as victim age,
vehicle age, and km/h travelling over the speed limit were
continuous numerical variables and these were converted
into category variables as shown in Table 1.
Finally, a probability sampling method through the
generation of a random number applied to each fatality
case was conducted. The random numbers were sorted
from smallest to largest and the rst 10 cases were selected.
A manual ‘sensemaking’ check was conducted to validate
that each case was best suited to the cluster or ‘prole’
derived through the MCA. This process returned full
agreement and no further checking was conducted.
Table 1. Variables for the Multiple Correspondence Analysis
Variable Categories
Time Evening; Middle of day; Middle of night; Morning
Vehicle age (years) 1-7; 8-14; 15-21; 22+
Intended trip duration Long; Short; Unknown
Crash location Urban; rural
Journey purpose Driving after drinking (pub) driving after drinking (party); evading police; joy ride;
possible suicide; recreation; tourism; utility trip; driving for work; unknown
Previous driving offences Yes/ no/ unknown
Kilometers over the speed limit 0km/h; 10 km/h; 20 km/h; 30 km/h; 40 km/h; 50 km/h; 60 km/h; 70 km/h; 80 km/h;
90 km/h; unknown
Location in vehicle Driver; passenger
Drugs present Yes/ no/ unknown
Evidence victim was a habitual
seat belt non-user Yes/ no/ suspected
Heightened emotional state Yes/ no
Vehicle type 4WD/SUV; car; rental; ute; van; truck
WoF/ CoF1Yes; No
Victim age (bands) 16-25; 26-35; 36-45; 46-55; 56-65; 66-75; 76+
Victim gender Male; female
Technicians and trades workers; community and personal service workers; sales
workers; machinery operators and drivers; labourers; beneciary; retired; student;
unemployed; unknown
Victim ethnicity Pākehā2; Māori; Asian; Pasika
Driver’s licence type Disqualied/ suspended/ forbidden; expired; unlicenced; full; lerner; restricted;
Alcohol present Yes/ no
Evidence of fatigue Yes/ no/ unknown
Medical condition or event
leading to the crash Yes/ no/ unknown
1 A regular vehicle check in New Zealand to ensure that the vehicle meets specic safety standards.
Warrant of Fitness (WoF) or Certicate of Fitness (CoF).
2 New Zealanders of European descent.
Journal of the Australasian College of Road Safety – Volume 30, Issue 3, 2019
Empirical Results
User factors
The empirical analysis identied that fatally injured seat belt
non-users were predominantly male, representing 75% of
victims (Figure 1). For both males and females, those aged
15-24 were more strongly associated with seat belt non-use
fatalities than people aged 25 years or over. For women, the
age group 15-19 was overrepresented in fatality cases with
13 cases, or 26.5% of all female cases in the study. In 12 of
these cases the deceased was a passenger in a vehicle and in
10 of these cases the driver was a young male (average age
21.5 years).
Figure 1. Age and gender profile of seat belt non-use victims
3020100 0 10 20 30
A summary of how the key variables coded under the
User Pillar, were associated with fatalities and crashes is
presented in Table 2.
Of those fatality cases where alcohol was involved (n=107
fatalities, n=95 crashes), in 95% of the crash cases the
driver’s blood alcohol content was over the legal driving
limit of 50mg per 100ml. In 38% (n=36) of alcohol-involved
crash cases, the driver’s blood alcohol was more than 200mg
per 100ml. Alcohol-involved fatalities were typied with the
journey purpose being driving home from a party or the pub
(n=64), and utility trips (n=26).
Through interviews and witness statements, the Police
reports identied that in 4 cases the victim usually wore a
seat belt but had not worn it on that occasion. In 31 cases
the fatally injured victims were described as habitual
non-wearers of seat belts and 9 victims were described as
part-time non-users of seat belts. Some witness statements
elaborated on the reasons for the habitual or part-time
non-use which included: frequent stops; short trip duration;
difcult to fasten; more people in the vehicle than seat belts;
physical discomfort; others were not wearing them.
Time of day
Two thirds (n=122) of the crashes occurred during dusk or
after dark, with the modal time occurring between 11pm
and 2am (24.7%, n=46). These late-night crashes were more
associated with multiple fatality outcomes. This pattern is
counter to normal travel patterns which have a peak demand
in the morning and afternoon. Only 4.3% (n=8) of the
crashes happened during the regular commuting hours of
8-9am and 5-6pm.
Roads and Roadsides and Speed Environment
A summary of the location of crashes, and the surface
condition of the road at the time of the crash is presented in
Table 3.
In New Zealand, speed limits are default 50 km/h in urban
areas and 100 km/h on rural or open roads unless stated
otherwise. Therefore, it is logical that these speed limits
were represented in 88% of crash cases. Vehicles in areas
with a posted speed limit of 100 km/h were involved in 137
crash cases and 150 fatalities. Fewer cases were reported in
50 km/h zones, with 27 crash cases and 28 fatalities.
Vehicle factors
A summary of vehicle factors recorded from the crash
reports is presented in Table 4.
Statistical Results
The MCA analysis revealed ve proles of people who
did not wear seat belts and who were fatally injured in
crashes: ‘young and risky’; ‘driving for work’; ‘elderly and
retired’; ‘overseas passengers’; and ‘people driving in rural
settings’. Every one of the 200 victims was ascribed to one
and only one prole. Because the proles show the best t
of the occupant groups, they are not equally populated. The
ve proles have been retrospectively named based on the
pattern of variables that they represent.
Young and risky
This prole comprised 28% (n=56) of the study’s sample.
Within this prole, 46 victims were male, 39 were aged
between 15-25 years, and 14 were aged between 26 and
40 years. People whose ethnic background was Māori
or Pasika represented 35 fatalities, with the remaining
being Pākehā. In 18 of the fatalities the driver was either
unlicensed, had their licence suspended, or held an illegal
licence, and in 24 cases the driver was reported to have had
previous driving offences. Vehicles associated with these
crashes were predominated by older vehicles of more than
14 years of age (n=46).
The behavioural characteristics of members of this prole
leading up to the crash were associated with inherently
risky behaviours. These included: high speeds – in 41 cases,
vehicle speed prior to the crash was more than 20 km/h
over the speed limit; alcohol involvement - for 32 fatalities,
alcohol readings were more than 100mg per 100ml of
Journal of the Australasian College of Road Safety – Volume 30, Issue 3, 2019
3 In fatal crashes, there can sometimes be a delay of hours or days before the victims are found, or before blood is taken for
testing, so in some cases, the degree of alcohol-involvement may be uncertain.
Female 48 49
Male 138 151
Māori 63 71
Pākehā 106 111
Pasika 9 9
Other 8 9
Journey purpose
Driving home from a party 34 38
Driving home from the pub 24 26
Driving for work 18 18
Utility trip (to work, shops, school) 70 77
Recreation/ tourists 8 8
Joy ride/ evading police 18 19
Possible suicide 7 7
Unknown 7 7
Intended journey duration
Short 53 55
Long 125 137
Unknown 8 8
Speed above the limit
10-25 km/h over limit 18 19
25-40 km/h over limit 21 22
40+ km/h over limit 20 20
Alcohol involvement3
Yes 95 107
No 87 88
Unknown 4 5
Evidence of illegal drugs (i.e. THC, methamphetamine, ketamine), or overdose of
prescription medication
Yes 52 53
No 127 137
Unknown 710
Evidence of fatigue (of victim)
Yes 67 73
No 108 110
Unsure 11 17
Driver’s emotional state compromised (i.e. clear evidence of anger or being upset)
Yes 33 33
Medical conditions or event attributed to the crash (i.e. heart attack, stroke, seizure, panic
Yes 29 29
Evidence of habitual seat belt non-use
Yes 40 40
Table 2. User Pillar empirical resuls for dichotomous and polychotomous variables by fatality and crash cases
Journal of the Australasian College of Road Safety – Volume 30, Issue 3, 2019
4 For the purposes of this research, the denitions of
‘urban’ and ‘rural’ were based on images from the crash
location. An urban area was classied as having a high
density of buildings, and urban motorways were also
included. A rural area included farmland, forest, and/or a
low density of buildings. Speed was not used to identify a
rural versus urban location as the measurement is too crude
(complicating factors can include urban motorways and
temporary speed restrictions on rural roads).
Road surface condition
Dry 132 139
Wet 50 54
Icy 3 6
Unknown 1 1
Rural roads4152 165
Urban roads 34 35
Mid-block 169 183
Intersection 17 17
Table 3. Roads and Roadsides Pillar empirical results
for dichotomous and polychotomous variables by
fatality and crash cases
blood alcohol; drug involvement - in 18 cases THC and/ or
methamphetamine were identied in the victim’s system;
and a risky journey purpose – such as 18 cases of ‘driving
home from a party or the pub’, and 19 cases of ‘evading
police’ or ‘joy ride’. In addition, in 23 cases there was
evidence of unbalanced emotional state including suicidal
tendencies and anger.
Driving for work
This category comprised 10% (n=20) of the total sample
used in this study, 19 of whom were male. They were
typied by their journey purpose which was driving a
vehicle for work. Trucks and vans were the predominant
vehicles, and the majority (n=18) of drivers were travelling
within the speed limit and had their full license (n=16).
Elderly and retired
A total of 6% (n=12) of the sample used in this study formed
this category. Two were aged between 66 and 75 years and
ten were aged 76 years or over. All occupants were retired,
none were speeding, eleven had a full license, and ten were
Pākehā. Medical conditions which were acknowledged in
the SCU reports as likely contributing factors to the crash
were identied in eleven cases. These included seizures,
strokes, and suicidal tendencies.
Overseas passengers
This was a reasonably small group, but an important group
when considering the safety outcomes of tourists in New
Zealand. The group comprised 4.5% (n=7) of the study’s
sample and consisted of people who were visiting New
Zealand. Six of the group were female and four were of
Asian descent. All members of this cohort were passengers
in vehicles where a long journey had been planned and many
were asleep across the rear seats when the crash occurred.
People driving in rural settings
This large group comprising 52.5% (n=105) of the sample
all crashed in rural settings. Most (n=83) had been planning
a long trip and all vehicles were light vehicles such as
passenger sedans (n=56) and 4-wheel drives, vans, or
utes (n=49). The presence of drugs and alcohol was 10%
higher in this cohort than the overall sample, with 70 cases
involving alcohol, and 29 cases involving illegal, or abused
prescription drugs.
Vehicle Age
14 or under 56 58
15 or over 130 142
Vehicle Type
Passenger sedan 114 121
4x4/ van/ SUV 62 69
Truck 10 10
Current Warrant or
Certificate of Fitness?
Yes 143 153
No 43 47
Did the vehicle roll?
Yes 84 90
No 102 110
Vehicle safety systems
Front airbags present 65 69
Side airbags present 14 14
Seat belt reminder
present 17 17
Table 4. Vehicle Pillar empirical results for dichotomous
and polychotomous variables by fatality and crash cases
Journal of the Australasian College of Road Safety – Volume 30, Issue 3, 2019
This research furthered the understanding of seat belt non-
use crashes in the New Zealand context by identifying how
patterns of factors were associated with different crash types
and the formation of the ve proles. Whilst some authors
have previously identied the ‘young and risky’ category
(Begg & Langley, 2000; Shults et al., 2016), other seat belt
non-use proles have not previously been described.
With regards to individual crash factors, this research
reiterated ndings from the USA (Alattar et al., 2016;
McCartt & Northrup, 2004; Steinhardt & Watson, 2007)
and Australia (Raftery & Wundersitz, 2011; Steinhardt et
al., 2007) that most crashes occur in the evening and early
morning. In addition, crashes on rural roads were more
commonly associated with seat belt non-use fatalities than
urban roads. This may partly be due to the typically higher
speed environment and the decreased chance of survivability
in high-speed crashes (Bédard et al., 2002; Elvik 2012), but
also reects USA and Australian literature which suggests
that seat belt wearing rates may be lower in rural settings
(Knight, Harris, & Iverson, 2008; Raftery & Wundersitz,
2011; Steinhardt et al., 2007).
This study showed a signicant disparity between fatal
outcomes between men and women – with far fewer women
being represented. Whilst this research did not describe seat
belt usage rates, it did examine non-use outcomes. Evidence
that women are more likely to wear seat belts than men
has been demonstrated in New Zealand (Fergusson et al.,
2003), USA (Eluru & Bhat, 2007; Reagan et al., 2013), the
United Kingdom (Richards et al., 2008), and Turkey (Alver,
Demirel, & Mutlu, 2014). In addition, there is a common
theme that those not wearing seat belts in fatal crashes are
more likely to be male (Palamara et al., 2009; Raftery &
Wundersitz, 2011; Romano & Voas, 2011).
An association between age and seat belt use is a common
theme throughout the literature, with drivers in their late
teens and early 20s being least likely to wear seat belts
(Alver et al., 2014; Eluru et al., 2007; Romano et al.,
2011). This trend is compounded for young males (Alattar
et al., 2016; McCartt et al., 2004; Raftery & Wundersitz,
2011). These patterns were reiterated by this study, but
were associated for all vehicle occupants, not just drivers.
Women in the age group 15-19 were overrepresented in this
study’s fatality cases (n=13), although in 10 of these cases
the deceased was a passenger in a vehicle driven by a young
male who t the criteria for the ‘young and risky’ prole
(note, driver survivors were not included in the analysis).
The non-use of a seat belt in these crashes may in part be due
to peer pressure (Jaccard et al., 2005).
Māori were overrepresented in seat belt non-use fatalities
(35%), compared to their proportion of the New Zealand
population (15%). Conversely, Pākehā were under-
represented (54%) compared to their proportion of the
population (74%) (Statistics New Zealand, 2013). This
raises questions about underlying socioeconomic and
colonialisation issues. Indeed, the association between lower
seat belt-wearing rates and people from disadvantaged ethnic
backgrounds has been described for Indigenous Australians
(Raftery & Wundersitz, 2011) and in the USA for people
from African American and Hispanic backgrounds (Shin et
al., 1999; Shults et al., 2016).
The literature reinforces that for many people the use of
seat belts may be governed by numerous factors, known
as a ‘decision policy’ (Alattar et al., 2016). Evidence of
these factors were illustrated in this research, including: the
inuence of the behaviour and choices of others (Han, 2017;
McCartt & Northrup, 2004; Jaccard et al., 2005); perceptions
of the journey’s risk (Begg & Langley, 2000); and the
planned number, speed, and duration of trips (Reagan et
al., 2013; Alattar et al., 2016). Kawsnicka et al. discuss that
“habitual behaviours are likely to dominate when resources
are limited” (2016, p.287) and for part-time seat belt users
who do not have ingrained habitual behaviours regarding
seat belt use, wearing a seat belt may be more inuenced
by external factors than those who habitually use them. The
results of this study suggest that fatigue, which was present
in 36.5% of fatality cases may have been a contributing
factor to some victims’, particularly passengers, lack of
seat belt use. This was particularly evident for passengers
sleeping across the rear seats. In addition, it is likely that
alcohol consumption, which is a known limiter of cognitive
resources may have played a part in the decision of some
of the 53.5% of fatality cases to wear a seat belt prior to the
crash. Indeed, the high rate of alcohol involvement in non-
seat belt crashes is an international issue (Begg & Langley,
2000; Raftery et al., 2011; Romano et al., 2011; Bogstrand et
al., 2015; Shults et al., 2016).
This research provides a part of the wholistic understanding
of seat belt non-users in New Zealand. To that end,
only seat belt non-use fatality cases were examined and
therefore we were unable to draw comparisons between
the proles identied from this research and proles of
belted occupants who died in crashes. Further research
to allow for comparison of these crash types would be
benecial when drawing broader conclusions. In addition,
although developing an in-depth understanding of serious
injury cases would have been benecial to better inform
the proles, the analysis was limited by the available data.
Another methodological limitation was that only people
aged 15 years and over were examined, as the funding scope
excluded unrestrained or incorrectly restrained children.
With regards to the ndings, the prole ‘people driving in
rural settings’ contained just over half of the fatality cases
and the MCA was unable to meaningfully split it into smaller
categories. The individuals in this prole exhibited the
least homogenous behavioural attributes and it may be that
the MCA method was limited by the number of variables
entered (n=21). However, it might simply be that some crash
circumstances may not t neatly into particular categories.
Certainly, the patterns of factors in the other four proles
were strongly aligned. Finally, this research was designed to
understand ‘who’ died on New Zealand’s roads whilst not
wearing a seat belt, not ‘why’. Future research, particularly
through qualitative interviews with seat belt non-use crash
survivors, as well as non-crash-involved people who t the
proles from this research would be valuable.
Journal of the Australasian College of Road Safety – Volume 30, Issue 3, 2019
This research provides a deeper understanding into the
contexts behind fatal crashes where seat belts were not worn
in New Zealand. It shows that a broad range of people and
situations are represented in these crashes, and highlights
that for many victims, the non-use of a seat belt may be the
only risky aspect of their otherwise normal journey.
Compared with the generally high rate of seat belt wearing
in New Zealand, the number of fatalities for seat belt non-
users as a proportion of all vehicle occupant fatalities (at
least one quarter) is high. Merely getting vehicle occupants
to wear their seat belt may not reduce their likelihood of
crashing, but it should reduce their fatality rate substantially
(Høye, 2016). These ndings suggest that the issue of seat
belt non-use will not be solved by focusing on seat belts
alone, rather it is part of a broader Safe System issue.
The next step towards meaningful road safety initiatives
to improve seat belt compliance is to understand why the
proles identied in this research do not wear seat belts.
The data presented in this paper pertain only to people who
did not wear a seat belt and died. A fatal crash is a relatively
unusual driving outcome and it is therefore likely that there
is a broader cohort of people who may t the occupant
proles who are alive. There are a range of possibilities
about why people do not wear seat belts, and if the
mechanisms are more clearly dened for various contexts,
then road safety initiatives can be better targeted to address
these and have a greater likelihood of success. For some
proles, a general focus on risky driving is needed, or even
support from outside of the transport system. For others,
cultural norms and a focus on positive habits may be more
This work was nancially supported by the AA Research
Foundation. No conicts of interest have been identied.
Alattar, L., Yates, J. F., Eby, D. W., LeBlanc, D. J., & Molnar,
L. J. (2016). Understanding and reducing inconsistency in
seatbelt-use decisions: Findings from a cardinal decision
issue perspective. Risk Analysis, 36(1), 83-97. DOI:
Alver, Y., Demirel, M. C., & Mutlu, M. M. (2014). Interaction
between socio-demographic characteristics: Trafc rule
violations and trafc crash history for young drivers.
Accident Analysis & Prevention, 72, 95-104. DOI:
Alattar, L., Yates, J. F., Eby, D. W., Le Blanc, D. J., Molnar, L.
J. (2016). Understanding and reducing inconsistency in
seatbelt-use decisions: Findings from a cardinal decision
issue perspective. Risk Analysis. 36(1): 83-97. DOI:
Bédard, M., Guyatt, G. H., Stones, M. J., & Hirdes, J. P. (2002).
The independent contribution of driver, crash, and vehicle
characteristics to driver fatalities. Accident Analysis
& Prevention, 34(6), 717-727. DOI: 10.1016/S0001-
Begg, D. J., & Langley, J. D. (2000). Seat-belt use and related
behaviors among young adults. Journal of Safety Research,
31(4), 211-220. DOI: 10.1016/S0022-4375(00)00038-4
Bhat, G., Beck, L., Bergen, G., & Kresnow, M.-J. (2012).
Predictors of rear seat belt use among US adults. Journal of
Safety Research, 53, 103-106.
Bogstrand, S. T., Larsson, M., Holtan, A., Staff, T., Vindenes,
V., & Gjerde, H. (2015). Associations between driving
under the inuence of alcohol or drugs, speeding and
seatbelt use among fatally injured car drivers in Norway.
Accident Analysis & Prevention, 78, 14-19. DOI: 10.1016/j.
Bose, D., C. Arregui-Dalmases, D. Sanchez-Molina, J. Velazquez-
Ameijide and J. Crandall (2013). “Increased risk of driver
fatality due to unrestrained rear-seat passengers in severe
frontal crashes.” Accident Analysis & Prevention 53(1):
100-104. DOI: 10.1016/j.aap.2012.11.031
Chliaoutakis, J. E., Gnardellis, C., Drakou, I., Darviri, C., &
Sboukis, V. (2000). Modelling the factors related to the
seatbelt use by the young drivers of Athens. Accident
Analysis & Prevention, 32(6), 815-825. DOI: 10.1016/S001-
de Pont, J. (2016). Why do people die in road crashes?, Ministry
of Transport.
Eluru, N., & Bhat, C. R. (2007). A joint econometric analysis of
seat belt use and crash-related injury severity. Accident
Analysis & Prevention, 39(5), 1037-1049. DOI: 10.1016/j.
Das, S., Avelar, R., Dixon, K., & Sun, X. (2018). Investigation
on the wrong way driving crash patterns using multiple
correspondence analysis. Accident Analysis & Prevention,
111, 43-55. DOI: 10.1016/j.aap.2017.11.016
Demirer, A., Durat, M., Haşimoğlu, C. (2012). Investigation
of seat belt use among the drivers of different education
levels. Safety Science, 50(4), 1005-1008. DOI: 10.1016/j.
Elvik, R. (2012). Speed limits, enforcement, and health
consequences. Annual Review of Public Health, 33, 225-
238. DOI: 10.1146/annurev-publhealth-031811-124634
Fergusson, D., Swain-Campbell, N., & Horwood, J. (2003). Risky
driving behaviour in young people: Prevalence, personal
characteristics and trafc accidents. Australian and New
Zealand Journal of Public Health, 27(3), 337-342. DOI:
Fildes, B., M. Fitzharris, S. Koppel, P. Vulcan and C. Brooks
(2003). Benets of seat belt reminder systems. Annual
Proceedings/Association for the Advancement of Automotive
Medicine (Vol. 47).
Fildes, B., Stevenson, M., Hoque, S., & Hammid, A. (2016).
Restraint use in the Eastern Province of the Kingdom of
Saudi Arabia. Trafc Injury Prevention, 17(5), 488-494.
DOI: 10.1080/15389588.2015.1103849
Journal of the Australasian College of Road Safety – Volume 30, Issue 3, 2019
Fong, C. K., Keay, L., Coxon, K., Clarke, E., Brown, J. (2016).
Seat belt use and t among drivers aged 75 years and older
in their own vehicles. Trafc Injury Prevention, 17(2): 142-
150. DOI: 10.1080/15389588.2015.1052420
Hatamabadi, H., Vafaee, R., Haddadi, M., Abdalvand, A.,
Esnaashari, H., & Soori, H. (2012). Epidemiologic study
of road trafc injuries by road user type characteristics and
road environment in Iran: A community-based approach.
Trafc Injury Prevention, 13(1), 61-64.
Han, G.-M. (2017). Non-seatbelt use and associated factors among
passengers. International Journal of Injury Control and
Safety Promotion 24(2): 251-255
Høye, A. (2016). How would increasing seat belt use affect
the number of killed or seriously injured light vehicle
occupants? Accident Analysis & Prevention, 88, 175-186.
DOI: 10.1016/
Husson, F., Josse, J., Le, S., Mazet, J. FactoMineR: Multivariate
Exploratory Data Analysis and Data Mining with R.
R package version 1.25. http://CRAN.R
package=FactoMineR. Accessed June 3, 2019.
Jaccard, J., H. Blanton and T. Dodge (2005). Peer inuences
on risk behaviour: An analysis of the effects of a close
friend. Developmental Psychology 41(1): 135-147. DOI:
Knight, P. J., Harris, M. F., & Iverson, D. (2008). Early driving
experience and risk perception in young rural people.
Paper presented at the Australasian Road Safety Research,
Policing and Education Conference, Adelaide, Australia.
Kawsnicka, D., Dombrowski, S. U., White, M., & Sniehotta,
F. (2016) Theoretical explanations for maintenance of
behaviour change: A systematic review of behaviour
theories. Health Psychology Review 10(3): 277-296. DOI:10
Larsson, P., & Tingvall, C. (2013). The Safe System Approach – A
Road Safety Strategy Based on Human Factors Principles.
In D. Harris (Ed.), Engineering Psychology and Cognitive
Ergonomics. Applications and Services (Vol. 8020, pp. 19-
28): Springer Berlin Heidelberg.
McCartt, A. T., & Northrup, V. S. (2004). Factors related to
seat belt use among fatally injured teenage drivers.
Journal of Safety Research, 35(1), 29-38. DOI: 10.1016/j.
Ministry of Transport. (2014). Front seat safety belt use by
adults: Results of a national survey 2014. Wellington, New
Ministry of Transport. (2016). 2016 adult front seat occupants.
Retrieved from
New Zealand Government, & National Road Safety Committee.
(2016). Safer Journeys Action Plan 2016-2020. Retrieved
New Zealand Transport Agency (2017, March 27). Seat-belt
non-use fatality search 2011-2015. Retrieved from Crash
Analysis System database
Palamara, P., Oxley, J., Langford, J., Thompson, C., & Chapman,
A.-M. (2009). An investigation of the factors associated
with the non-use of a seat belt through the analysis of linked
Western Australian crash, death and hospitalisation data.
Paper presented at the Australasian Road Safety Research,
Policing and Education Conference, Sydney, Australia.
Raftery, S., & Wundersitz, L. (2011). No restraint? Understanding
differences in seat belt use between fatal crashes and
observational surveys. Journal of Safety Research, 31(4),
Reagan, I. J., McClafferty, J. A., Berlin, S. P., & Hankey, J. M.
(2013). Using naturalistic driving data to identify variables
associated with infrequent, occasional, and consistent seat
belt use. Accident Analysis & Prevention, 50, 600-607. DOI:
Richards, D. C., Hutchins, R., Cookson, R. E., Massie, P., &
Cuerden, R. W. (2008). Who doesn’t wear seat belts?
Paper presented at the International conference: Expert
Symposium on Accident Research, Hannover, Germany.
Road Safety Observatory (2013). Seat belts review. UK
Romano, E., & Voas, R. B. (2011). Drug and alcohol involvement
in four types of fatal crashes. Journal of Studies on
Alcohol and Drugs, 72(4), 567-576. DOI: 10.15288/
Routley, V., Ozanne-Smith, J., Li, D., Hu, X., Wang, P., &
Qin, Y. (2007). Pattern of seat belt wearing in Nanjing,
China. Injury Prevention, 13(6), 388-393. DOI: 10.1136/
Shin, D., Hong, L., & Waldron, I. (1999). Possible causes of
socioeconomic and ethnic differences in seat belt use among
high school students. Accident Analysis & Prevention,
31(5), 485-496. DOI: 10.1016/S0001-4575(99)00004-4
Shults, R. A., Haegerich, T. M., Bhat, G., & Zhang, X. (2016).
Teens and seat belt use: What makes them click? Journal of
Safety Research, 57, 19-25. DOI: 10.1016/j.jsr.2016.03.003
Statistics New Zealand. (2013). 2013 Census - Major ethnic
groups in New Zealand.
Steinhardt, D. A., & Watson, B. C. (2007). Night time seatbelt
non-use in serious crashes: A comparison of contributing
factors in rural and urban areas of the United States and
Queensland. Paper presented at the Australasian Road
Safety Research, Policing & Education Conference,
Melbourne, Australia.
Traynor, T. (2005). The impact of driver alcohol use on crash
severity: A crash specic analysis. Transportation Research
Part E: Logistics and Transportation Review, 41(5), 421-
437. DOI: 10.1016/j.tre.2005.03.005
World Health Organisation (2009). The need for seat-belts and
child restraints, World Health Organisation.
World Health Organisation. (2016). “Road Trafc Injuries.”
Retrieved from
... The TCR reports and the other crash-associated data sources described above were coded into 64 variables (49 polychotomous, 10 dichotomous, 5 open-ended) by a single analyst following a Safe System coding framework which, in its design, acknowledged that DSI crashes happen when a combination of system failures occur (Larsson & Tingvall, 2013). Each case was examined using variables relating to the four Safe System pillars: Speed; Roads and Roadsides; Vehicles; and Users (Hirsch et al., 2019;Mackie et al., 2017). The 'User' pillar was split into two to more equally represent drivers and pedestrians. ...
Full-text available
In 2016 in New Zealand, pedestrians accounted for 7.6% (n=25) of all road fatalities and 6.6% (n=257) of serious injuries (Ministry of Transport, 2017). The aim of this research was to analyse a sample of pedestrian deaths and serious injury (DSI) cases to understand the contribution of Safe System gaps in serious harm outcomes. A sample of 100 pedestrian fatality and 200 serious injury crash reports from 2013-2017 were analysed to identify the contribution of the four Safe System pillars (roads and roadsides, vehicle, speed environment, user) in each crash case. The research identified common crash scenarios and highlighted the need for improvements in speed management, safer vehicles, safety campaigns, and infrastructure design. In addition, the research identified latent high-order sociotechnical system factors that obstruct the mechanisms to effectively address these Safe System issues and which ultimately perpetuate the occurrence of pedestrian DSIs.
Full-text available
Problem: Motor vehicle crashes kill more adolescents in the United States than any other cause, and often the teen is not wearing a seat belt. Methods: Using data from the 2011 Youth Risk Behavior Surveys from 38 states, we examined teens' self-reported seat belt use while riding as a passenger and identified individual characteristics and environmental factors associated with always wearing a seat belt. Results: Only 51% of high school students living in 38 states reported always wearing a seat belt when riding as a passenger; prevalence varied from 32% in South Dakota to 65% in Delaware. Seat belt use was 11 percentage points lower in states with secondary enforcement seat belt laws compared to states with primary enforcement laws. Racial/ethnic minorities, teens living in states with secondary enforcement seat belt laws, and those engaged in substance use were least likely to always wear their seat belts. The likelihood of always being belted declined steadily as the number of substance use behaviors increased. Discussion: Seat belt use among teens in the United States remains unacceptably low. Results suggest that environmental influences can compound individual risk factors, contributing to even lower seat belt use among some subgroups. Practical applications: This study provides the most comprehensive state-level estimates to date of seat belt use among U.S. teens. This information can be useful when considering policy options to increase seat belt use and for targeting injury prevention interventions to high-risk teens. States can best increase teen seat belt use by making evidence-informed decisions about state policy options and prevention strategies.
Full-text available
Background: Behaviour change interventions are effective in supporting individuals in achieving temporary behaviour change. Behaviour change maintenance, however, is rarely attained. The aim of this review was to identify and synthesise current theoretical explanations for behaviour change maintenance to inform future research and practice. Methods: Potentially relevant theories were identified through systematic searches of electronic databases (Ovid MEDLINE, Embase, PsycINFO). In addition, an existing database of 80 theories was searched, and 25 theory experts were consulted. Theories were included if they formulated hypotheses about behaviour change maintenance. Included theories were synthesised thematically to ascertain overarching explanations for behaviour change maintenance. Initial theoretical themes were cross-validated. Findings: One hundred and seventeen behaviour theories were identified of which 100 met the inclusion criteria. Five overarching, interconnected themes representing theoretical explanations for behaviour change maintenance emerged. Theoretical explanations of behaviour change maintenance focus on the differential nature and role of motives, self-regulation, resources (psychological and physical), habits, and environmental and social influences from initiation to maintenance. Discussion: There are distinct patterns of theoretical explanations for behaviour change and for behaviour change maintenance. The findings from this review can guide the development and evaluation of interventions promoting maintenance of health behaviours and help in the development of an integrated theory of behaviour change maintenance.
Full-text available
Objectives: This study set out to examine seat belt and child restraint use in the Dammam Municipality of the Kingdom of Saudi Arabia, based on the premise that an increase in seat belt use would significantly reduce personal injury in traffic crashes. It was expected that local data would help identify intervention strategies necessary to improve seat belt use in the region. Methods: The research involved 2 methodologies. First, 1,389 face-to-face interviews were conducted with male and female adults in regional shopping plazas regarding their own and their children's restraint use in their vehicles and reasons for these attitudes and beliefs. Second, 2 on-road observation studies of adult and child restraint use were conducted by trained observers. Occupants of approximately 5,000 passenger vehicles were observed while stopped at representative signalized traffic intersections. Results: The findings showed front seat belt use rates of between 43 and 47% for drivers and 26 to 30% for front seat passengers; rear seat belt use rates were lower. While there seemed to be some knowledge about the purpose and reasons for restraining both adults and children in suitable restraints, this failed to be confirmed in the on-road observations. Conclusions: Reasons for these rates and findings are discussed fully, and recommendations for improving seat belt use in the Dammam Municipality are included.
Wrong way driving (WWD) has been a constant traffic safety problem in certain types of roads. Although these crashes are not large in numbers, the outcomes are usually fatalities or severe injuries. Past studies on WWD crashes used either descriptive statistics or logistic regression to determine the impact of key contributing factors. In conventional statistics, failure to control the impact of all contributing variables on the probability of WWD crashes generates bias due to the rareness of these types of crashes. Distribution free methods, such as multiple correspondence analysis (MCA), overcome this issue, as there is no need of prior assumptions. This study used five years (2010-2014) of WWD crashes in Louisiana to determine the key associations between the contribution factors by using MCA. The findings showed that MCA helps in presenting a proximity map of the variable categories in a low dimensional plane. The outcomes of this study are sixteen significant clusters that include variable categories like determined several key factors like different locality types, roadways at dark with no lighting at night, roadways with no physical separations, roadways with higher posted speed, roadways with inadequate signage and markings, and older drivers. This study contains safety recommendations on targeted countermeasures to avoid different associated scenarios in WWD crashes. The findings will be helpful to the authorities to implement appropriate countermeasures.
Seatbelt use is the most effective way to save lives and reduce severe injuries. However, the percentage of non-seatbelt use is still high among drivers and passengers. Although the factors related to non-seatbelt use among drivers have been widely studied, the factors associated with non-seatbelt use among passengers have not been well documented. In addition, recent surveys showed that the driver's attitude has a significant impact on the passenger's seatbelt use. However, the lower response rate and less accurate of self-reported seatbelt use in survey studies, especially among participants who had a high level perception of penalty for non-seatbelt use. Therefore, we examined the association between passenger's seatbelt use and driver's seatbelt use with a statewide injury surveillance system. 36,012 passengers who were involved in motor vehicle crashes (MVC) in 2004-2013 were included in this study. Our results showed that if a driver wore a seatbelt, 92.6% of his/her passengers also wore seatbelts while if a driver did not wear a seatbelt, only 19.1% of his/her passengers wore seatbelts. Compared to the passenger whose driver wore a seatbelt, the passenger had a significantly higher probability of non-seatbelt use (odds ratio = 46.7; 95% confidence intervals, 42.7-51.1) if his/her driver did not wear a seatbelt. The driver has the greatest influence on the passenger's seatbelt use. The findings will provide important information for future public health practices to increase seatbelt use at the highest possible rate for passengers, such as educational interventions for drivers and seatbelt reminders use.
The expected effects of increasing seat belt use on the number of killed or seriously injured (KSI) light vehicle occupants have been estimated for three scenarios of increased seat belt use in Norway, taking into account current seat belt use, the effects of seat belts and differences in crash risk between belted and unbelted drivers. The effects of seat belts on fatality and injury risk were investigated in a meta-analysis that is based on 24 studies from 2000 or later. The results indicate that seat belts reduce both fatal and non-fatal injuries by 60% among front seat occupants and by 44% among rear seat occupants. Both results are statistically significant. Seat belt use among rear seat occupants was additionally found to about halve fatality risk among belted front seat occupants in a meta-analysis that is based on six studies. Based on an analysis of seat belt wearing rates among crash involved and non-crash involved drivers in Norway it is estimated that unbelted drivers have 8.3 times the fatal crash risk and 5.2 times the serious injury crash risk of belted drivers. The large differences in crash risk are likely to be due to other risk factors that are common among unbelted drivers such as drunk driving and speeding. Without taking into account differences in crash risk between belted and unbelted drivers, the estimated effects of increasing seat belt use are likely to be biased. When differences in crash risk are taken into account, it is estimated that the annual numbers of KSI front seat occupants in light vehicles in Norway could be reduced by 11.3% if all vehicles had seat belt reminders (assumed seat belt wearing rate 98.9%), by 17.5% if all light vehicles had seat belt interlocks (assumed seat belt wearing rate 99.7%) and by 19.9% if all front seat occupants of light vehicles were belted. Currently 96.6% of all (non-crash involved) front seat occupants are belted. The effect on KSI per percentage increase of seat belt use increases with increasing initial levels of seat belt use. Had all rear seat occupants been belted, the number of KSI front seat occupants could additionally be reduced by about 0.6%. The reduction of the number of KSI rear seat occupants would be about the same in terms of numbers of prevented KSI.
This paper aims to describe seat belt wearing patterns and quality of seat belt fit among drivers aged 75 years and older. A secondary aim is to explore associations between body shape, comfort and seat belt use patterns. This is an observation and survey study of a cohort of 380 drivers aged 75 years and over. During home visits, photographs were taken of the drivers in their vehicles for later analysis of belt fit and a short survey was also administered to collect demographic data and information about seat belt use and comfort. Seat belt fit, and use of belt and seat accessories were analysed from the photographs. Data from 367 participants with photographs were analysed. While 97% reported using a seat belt and 90% reported their seat belt to be comfortable, 21% reported repositioning their seat belt to improve comfort. Good sash and lap belt fit were achieved in 53% and 59% of participants respectively but only 35% achieved overall good fit. Both poor sash and lap belt fit were observed in 23% of participants. Drivers who were in the obese category had over twice the odds (95% CI 1.2-4.1) of having a poor lap belt fit than those in the normal BMI range, and drivers who were overweight had 1.8 times the odds (95% CI 1.1-2.9) of having poor lap belt fit. Older females also had twice the odds (95%CI 1.3-3.5) of poor lap belt fit than older males, regardless of BMI. Sash belt fit did not vary significantly by BMI, stature or gender. However older drivers who reported they had not made any adjustments to the D-Ring height had 1.7 times the odds of having poor sash belt fit than those who made adjustments (1.2-2.9). Females were 7.3 times more likely to report comfort problems than males (95% CI 3.2, 16.3) but there was no association between reported comfort and BMI, or seat belt fit. Drivers who reported comfort problems had 6 times the odds (3.2-13.6) of also reporting active re-positioning of the belt. The results suggest older drivers face challenges achieving comfortable and correct seat belt fit. This may have a negative impact on crash protection. Belt fit problems appear to be associated with body shape, particularly high BMI and gender. There is a need for further investigation of comfort accessories, in the interim older drivers and occupants should be encouraged to use features such as D-ring adjusters to improve sash belt fit.
This article has two aims. The first is to present results that partly explain why some automobile drivers choose to use their seatbelts only part time, thereby exposing themselves to unnecessary risk. The second is to offer and illustrate the "cardinal decision issue perspective"(1) as a tool for guiding research and development efforts that focus on complex real-life decision behaviors that can entail wide varieties of risk, including but not limited to inconsistent seatbelt use. Each of 24 young male participants drove an instrumented vehicle equipped to record continuously seatbelt use as well as other driving data. After all trips were finished, each participant completed an interview designed to reconstruct how he made randomly selected seatbelt-use decisions under specified conditions. The interview also examined whether and how drivers established "decision policies" regarding seatbelt use. Such policies were good predictors of inconsistent seatbelt use. Drivers who had previously adopted policies calling for consistent seatbelt use were significantly more likely than others to actually drive belted. Meta-decisions about seatbelt policy adoption appeared to rest on factors such as whether the driver had ever been asked to consider selecting a policy. Whether a driver made an ad hoc, on-the-spot seatbelt-use decision was associated with a perceived need to make such a decision. Finally, participants with full-time policies were especially likely to deploy their seatbelts by default, without recognizing the need to decide about belt use on a trip-by-trip basis. We end with recommendations for reducing inconsistencies in seatbelt use in actual practice. © 2015 Society for Risk Analysis.
Seat belt use reduces the risk of injuries and fatalities among motor vehicle occupants in a crash, but belt use in rear seating positions is consistently lower than front seating positions. Knowledge is limited concerning factors associated with seat belt use among adult rear seat passengers. Data from the 2012 ConsumerStyles survey were used to calculate weighted percentages of self-reported rear seat belt use by demographic characteristics and type of rear seat belt use enforcement. Multivariable regression was used to calculate prevalence ratios for rear seat belt use, adjusting for person-, household- and geographic-level demographic variables as well as for type of seat belt law in place in the state. Rear seat belt use varied by age, race, geographic region, metropolitan status, and type of enforcement. Multivariable regression showed that respondents living in states with primary (Adjusted Prevalence Ratio (APR): 1.23) and secondary (APR: 1.11) rear seat belt use enforcement laws were significantly more likely to report always wearing a seat belt in the rear seat compared with those living in a state with no rear seat belt use enforcement law. Several factors were associated with self-reported seat belt use in rear seating positions. Evidence suggests that primary enforcement covering all seating positions is an effective intervention that can be employed to increase seat belt use and in turn prevent motor vehicle injuries to rear-seated occupants. Copyright © 2015. Published by Elsevier Ltd.