Literature Review:
Environmental Correlates with Injury
Motor vehicle collisions
Literature Review:
Environmental Correlates with Injury
Overview
According to Bunn et al. (2003), ten million people are permanently injured annually while a
further one million die from motor vehicle incidents. These figures are too significant to ignore, thus
making the study of incident correlations and preventative measures instrumental in an increasingly
urbanized world. For this overview, it is important to provide context for crash frequency and severity.
Traffic volumes are the main determinant of crash frequency, while traffic speed denotes crash severity
(Ewing and Dumbaugh, 2009). Ewing and Dumbaugh (2009) go on to state how the built environment
helps mediate traffic volume and speed, therefore affecting the number of crashes.
Road and Intersection Design
In terms of the design of roads themselves, wider roads are more accident prone than narrower
roads, where lanes over eleven feet wide have a noticeable impact on the number of crashes (Ewing and
Dumbaugh, 2009). Furthermore, downtown areas with wide travel lanes have, proportionally, the
largest amount of crashes (ibid.). Besides sheer width, the number of through lanes, the number of left
turn lanes (either exclusive or offset), the presence of a median, right turn lanes, and the angle of each
intersecting roadway (Guo, Wang, Abdel‐Aty, 2010) also influence injury crash rates. Additionally, Guo
et al. (2010) claims intersections should not be treated independently from their corridor in regards to
safety improvements and analysis. For intersections, traffic volumes, geometric design, traffic control,
and traffic operation characteristics are correlated with the number of crashes (Madanu, Zongzhi,
Abbas, 2010). In terms of an intersection’s geometry, crashes are more frequent at four‐ways, as
opposed to three‐ways (Ewing and Dumbaugh, 2009). Guo et al. (2010) adds intersection size, traffic
conditions by turning movement and coordination of signal phases in having significant impacts on the
safety of intersections. Surprisingly, intersections with signals are among the most dangerous arterial
sections because of complex traffic, conflicting movements, and frequently changing signals (Guo et al.
2010).
Not surprisingly, alcohol availability is strongly associated with increased alcohol‐related vehicle
crashes (Scribner, n.d.). Scribner’s (n.d.) study includes restaurants, bars, liquor stores, and minimarkets
as alcohol distribution locations. The density of these sites is believed to influence community
behaviour and alcohol consumption, therefore increased density correlates with injury crash density,
excluding bars (ibid.). Aside from alcohol, on‐street parking is another factor in the increasing
complexity of multi‐vehicle interactions, thus contributing to the number of crashes (Ewing and
Dumbaugh, 2009). Driveways and lanes also increase the probability of crashes by making vehicle
interactions more complicated (ibid.).
Preventative Measures
In general, short, uninterrupted roadway segments keep vehicle speeds low, thus reducing crash
frequency and severity (Ewing and Dumbaugh, 2009). Crash prevention measures can be placed into
two main categories, which are traffic calming measures and access management measures. According
to Bunn et al. (2003, p. 200), traffic calming refers to “the combination of mainly physical measures that
reduce negative effects of motor vehicle use, alter driver behaviour, and improve conditions for nonmorotised
street users.” These include, for example, speed bumps, medians, and roundabouts (Ewing
and Dumbaugh, 2009). On the other hand, access management is “the control of the location, spacing,
and operation of driveways, median openings, and street connections to a main roadway” (ibid., p. 358).
Area‐wide traffic calming can potentially reduce the number of injuries by reducing vehicle speed (Bunn
et al., 2003). In a study carried out by Morrison et al. (2003) traffic calming schemes, on average,
reduced accidents by fifteen percent. Other effective traffic calming schemes are rumble strips, as
opposed to raised road surfaces, before crossroads (Morrison et al., 2003). Additionally, road humps
may initially reduce accidents, but tend to increase them in other areas (ibid.). Outside from traffic
calming and access management measures, public lighting, guardrails, and crash cushions actually
increase the number of accidents while decreasing their severity (ibid.). Lighting’s effects include
increased speed and decreased concentration (Jorgensen, 2002). Although, normally, increased traffic
volume is thought to increase the number of accidents, examples from the United Kingdom say the
opposite: increased traffic volume was accompanied by a fall in serious and fatal accidents (Morrison et
al., 2003). The INSURANCE Institute for Highway Safety (2001) in the United States conducted a study
on the effects of red light camera enforcement on intersection crashes. The study, done in Oxnard,
California, found that after the cameras were implemented on 11 of the community’s 125 intersections
with traffic signals, crashes resulting in injuries decreased by 29 (ibid.). Interestingly, crashes resulting in
injuries declined in similar proportions across all 125 intersections, reinforcing the idea that perceived
surveillance is effective in reducing traffic violations (ibid.).
Overview
Crime is often associated with violence (Casteel and Peek‐Asa, 2000), which is classified as a
major health issue (Mair and Mair, 2003). According to Brink (1994, p. 38), interpersonal violence can
be defined as "intentionally inflicted bodily harm, injury, or death. Place, which is defined as "a very
small area reserved for a narrow range of functions, often controlled by a single owner, and separated
from the surrounding area...", is often a target for violence (Mair and Mair, 2003). According to Nelson,
Bromley, and Thomas (2001), violence, instead of largely taking place in rural areas, is concentrated in
urban ones. Environmental criminologists Brantham and Brantingham (1993, 1999) state an area's
likelihood for crime is based on various ecological features. The following two sections will explore
exactly what current literature asserts they are. The first section will explain features unrelated to
alcohol serving establishments, while the second will focus on them.
Everything Else
Although the following are not exactly aspects of the built environment, they do contribute to it.
Cusimano, Marshall, rinner, Jiang, and Chipman (2010) have found that areas of social deprivation were
strongly correlated with higher densities of serious injury. The researchers delineated these areas using
unemployment rates, the presence of homeless shelters, and social housing (ibid.). Furthermore, areas
with higher density of mixed‐use planning, such as areas zoned for commercial and residential purposes,
have less violence (Browning et al. 2010).
Lighting, visibility, and cameras have a negative correlation to robberies, which, under the broad
umbrella of crime, can be applied to interpersonal violence (Casteel and Peek‐Asa, 2000). Mair and Mair
(2003) supply an interpretation of Jane Jacobs' The Death and Life of Great American Cities. The three
factors influencing violence in public areas are houses facing the street, continuous sidewalks, and
clearly defined public and private areas (ibid.). The overarching theme of the previous two authors'
assertions is the idea that surveillance, or even perceived surveillance, decreases an individual's
probability of committing a violent act (Cozens, Saville, Hiller, 2005). Continuing on this train of thought,
Greunewald, Freisthler, Remer, LaScale, and Treno (2006) state areas, often times around the edges of
neighbourhoods or on the boundaries between them, with abandoned housing or greater retail activity
see higher rates of interpersonal violence. Mair and Mair (2003) also note that, in Los Angeles, gang
violence occurred more often on or near accessible roads which allowed for the perpetrators to escape
easily by car. Nelson et al. (2001) found that locations of violent crimes included nightclubs, public
houses, shops, car parks, cafés, and bus stops, and fast food outlets adjacent to alcohol serving
establishments.
Alcohol Serving Establishments
Violence is typically concentrated in city centre areas where the greatest number of pubs are
located (Brink, 1998). Alcohol sales are attributed to higher rates of violence, likely because they tend
to attract young makes, which are a more violence‐prone group (Greunewald et al. 2006). However, the
authors of this paper assert the impact of bar densities on violence generally depend on the general
demographics of the area, such as socio‐economic status (ibid.). Bingemann‐Handsen and Brink (2004)
corroborate this evidence in their study, which found the most common locations for violence within
the city were in the street near pubs and in pubs. It is important to mention these researchers did not
use the same injury scale used in British Columbia, it still provides significant support for examining
alcohol serving establishments as a correlate to injury because they looked at four different time periods
during the 1980s and 1990s in Aarhus, Denmark (ibid.). Another study done in New Jearsy by Gorman,
Speer, Greunewald, and Labouvie (2001) found density of alcohol availability had the strongest
correlation with violence. However, they also say “the foci of the routine‐activities theory are places,
rather than the effects of alcohol consumption per se. According to the theory, violent crime is more
likely to occur in places that bring together a motivated offender, a suitable target, and the absence of
effective guardians” (ibid., 635). They back this statement up by mentioning that bars in wealthier areas
did not increase the rate of interpersonal violence in their vicinity (ibid.).
