Factors Affecting the Risk of Urban Road Traffic

Crashes: A Case-control Study Based on Wardha city

Police Data

Sarfaraz Ahmed1, Vikas Ravekar2, Bandhan Bandhu Majumdar3 SiddardhaKoramati4

Scientist, Flexible Pavement Division, CSIR-Central Road Research Institute, New Delhi-
1100251
Assistant Professor, Department of Civil Engineering, Bajaj Institute of Technology, Wardha-
4420012
Assistant Professor, Department of Civil Engineering, Birla Institute of Technology and Sci-
ence, Pilani, Hyderabad-500783
Research Scholar, Department of Civil Engineering, Birla Institute of Technology and Science,
Pilani, Hyderabad-500784

Abstract. Road safety is a topic of concern for both health and development.
Road traffic crashes are a disaster for humans. In terms of untimely deaths, inju-
ries, it entails high human suffering and monetary costs. Road traffic crashes are
controllable to corrective measures. It is, therefore, necessary to examine the
causes of road traffic crashes and recommend corrective measures at a poten-
tial segment, particularly to developing countries such as India where road traffic
crashes are prevalent and data on road crashes are scarce. The objective of this
study is to identify risk factors related to the vehicle and the environment, that
are associated with a road traffic crash on urban roads in India, based on police
data records (FIR). By analyzing this basic but important data, when other geo-
metric and land use information is missing, could help to determine the risk fac-
tors involved in road traffic crashes that are responsible for the fatal or non-fa-
tal crash. For this study, crash data for the year 2014-2019 was collected from
the Wardha City Police Department for seven road segments. A case-control
study is conducted where the case is: fatal road traffic crash; and the control was
the non-fatal crash. A case-control analysis is designed to assess whether expo-
sure is related to an outcome. The risk analysis was calculated by the odds ratio
(OR) estimate, with a 95 % confidence interval. Bicycles and pedestrians as vic-
tims, heavy motor vehicles as accused vehicle categories, and crossing type of
vehicle maneuver were found to be major risk factors for fatal crashes, as were a
few other statistically insignificant variables. Identifying factors related to the
risk of being involved in a roadway traffic crash on a specific road segment aids
us in recommending primary preventive measures and resolving this major threat
to human health, particularly in the Indian context.

Keywords: Road traffic crash, case-control study, risk factors, odds ratio.
2

1 Background

India has the highest number of traffic crashes among the 199 countries, followed by
China and the United States (IRF, 2019). According to the Ministry of Road Transport
and Highways of the Government of India (GOI), there were 4,67,044 incidents and
1,51,417 deaths in 2018, averaging 1,280 accidents and 415 deaths per day and approx-
imately 53 accidents and 17 deaths per hour. Road fatalities increased by 0.46 percent
in 2018 relative to the previous year, 2017, and the number of people killed increased
by 2.37 percent (2). This data suggests that in India, fatal road traffic crashes are in-
creasing significantly in comparison to non-fatal crashes, which must be addressed by
developing preventive actions. To improve road safety, GOI introduced Motor Vehicle
Amendment Bill 2019, amending the Motor Vehicle Act 1988. After its enactment, the
Motor Vehicle Amendment Act 2019 has focused on road safety and included provi-
sions such as penalties for traffic violations, improved coverage for hit-and-run cases,
cashless care during the golden hour, vehicle health and driving checks, and more. De-
spite the passage of this law, India's unregulated vehicle growth and diverse traffic con-
ditions may continue to be an issue (3). Road traffic crashes are complicated incidents
that are caused by a combination of human, mechanical, and environmental causes. The
case-control analysis is the most common method of research design used in epidemi-
ology to analyze factors and estimate injury risks (4). Controls are a category of non-
fatal crash-involved vehicles, while cases are a group of fatal crash-involved vehicles.
We analyzed crash data obtained from Wardha police department FIR records for seven
urban traffic corridors in Wardha city, India, for this study. Crash data was gathered
over six years, from 2014 to 2019, during which 30 people were killed and 174 were
injured. It is important to define risk factors that influence fatal crashes to administer
effective steps to reduce the fatality rate. In India, where accident reports from city
police stations are the only credible source of evidence, information about the road ge-
ometry and traffic characteristics for crash sites is usually not available. The crash re-
ports include information such as the date of the crash, the time of the crash, the iden-
tification of the accused and victim vehicles, the number of injuries and deaths, the road
type, type of road geometry, vehicle maneuver, the cause of the crash, and the type of
accident. The study's key goal was to determine risk factors associated with fatal inju-
ries caused by traffic crashes on the urban section that were linked to the driver, the
vehicle, or the environment in the absence of other primary data.

2 Literature Review

Several previous studies examined the factors that influence traffic collisions. Using
odds-ratio analysis, Híjar et al. (2000) identified pedestrian, truck, and environmental
risk factors associated with road accidents on the Mexico-Cuernavaca highways. They
discovered that young drivers, regular driving, work trips, alcohol use, weekday trips,
and inclement weather all play a role in fatal crashes on that route. Based on police
reports, Valent et al. (2002) used logistic regression to determine the main risk factors
for fatal crashes in Udine, Northeast Italy. Male drivers, older drivers, cyclists, time of
 3

the incident (between 1:00 and 5:00 a.m.), lack of seat belt use, etc., were all linked to
fatal crashes. In Hong Kong, Yau (2004) used stepwise logistic regression models to
identify factors that influence traffic accidents by vehicle type. District board, gender,
age of the vehicle, time of the accident, and street light conditions were found to be
dominant attributes correlated with seriousness in private vehicles; seat-belt use and
weekday incident were found to be dominant attributes in goods vehicles; and age of
the vehicle, weekday, and time of the accident was found to be dominant attributes in
motorcycles. Using odds-ratio analysis, Vorko-Jović et al. (2006) established
nighttime, junctions, and overspeeding as the key causative factors for fatal accidents
in Zagreb, Croatia. To identify the key causative factors specific to serious traumatic
brain injury caused by traffic accidents, Javouhey et al. (2006) performed a population-
based study based on five-year accident results (1996-2001) obtained from the Rhone
region of France's road trauma registry. To measure their impact on crash propensity,
odds ratios were determined for all associated factors due to the accident. The most
prevalent risk factors were found to be gender: man, elderly male (age more than 55
years), and un-helmeted motorcyclists, accompanied by an accused vehicle, location of
the incident, time of the collision, and so on. A cross-sectional analysis was undertaken
in Dubai by Al Marzooqi et al. (2010) to recognize traffic accident risk factors such as
high-speed limits and truck volume. Moskal et al. (2012) investigated the impact of risk
factors on injury injuries involving powered two-wheeler (PTW) riders in France. They
performed a case-control analysis considering the odds-ratio using ten years of police
data. Sex, helmet use, alcohol consumption, driving license, and transporting passen-
gers in PTWs and leisure trips were found to be the most important risk factors for
crashes.

3 Methodology

The research methodology includes selecting a study area, collecting crash data, extract
information, and analyzing it. Figure 1 depicts the overall research methodology.

Step 1: Literature review

Step 2: Study area and Data collection

Step 3: Data description

Step 4: Identification of factors influencing fatal

crashes

Step 5: Discussion and conclusion

4

3.1 Study area characteristics and data collection

Wardha city, the study area, is a municipal body in Maharashtra's Wardha district, In-
dia. Wardha is well-linked by road to the rest of Maharashtra's cities. The city is crossed
by National Highway No.361 (Nagpur-Wardha-Yavatmal-Nanded-Latur-Tuljapur). It
is also crossed by the Nagpur-Aurangabad-Mumbai Express Highway. As a result,
heavy traffic is expected to flow through the area, and road traffic collisions seem to be
on the rise. Hence, Wardha City was selected as the study area to determine risk factors
specific to fatal crashes to help avoid potentially fatal crashes. Between 2014 and 2019,
police records of 204 crashes on seven crash-prone road segments were used to create
a road accident database (Wardha Police, 2021). The seven road segments are: Bajaj
Chowk to Ambedkar putla; Bajaj Chowk to Deoli Naka; Bajaj Chowk to Shivaji
Chowk; Shivaji Chowk to Dhuniwale Math; Ambedkar putla to Mahatma Gandhi putla;
Shivaji Chowk to Arvi Naka; and DuniWale math to Nalwadi Road. After ensuring that
no geometric changes (lane form, horizontal curvature, or vertical gradient) had oc-
curred during the research period, these seven road segments were chosen for the study.

3.2 Data description

There were two forms of crash severity included in the database: fatal and non-fatal.
From 2015 to 2019, fatal and non-fatal crashes accounted for 14.7 percent and 85.83
percent of total crashes for seven segments, respectively. According to preliminary
analysis, total, fatal, and non-fatal crashes were at their peak in 2016. From 2014 to
2016, the crash proportion rose, but decreased in 2018, before increasing again in 2019.

Season of crash
Seasonal data could be extracted based on the date of a crash's occurrence. Summer,
monsoon, autumn, winter, and spring are defined as the five seasons in this study. Ac-
cording to a descriptive investigation, the monsoon season (25.9%) is correlated with
the highest number of fatal crashes, followed by summer (25.5%), spring (18.6%), win-
ter (17.2%), and autumn (12.7 %) as shown in Figure 1.
 5

Autumn
Monsoon
Spring
Summer
17.2% 12.7%
Winter

25.9%
25.5%

18.7%

Fig. 1. The proportion of crashes that occurred during each season

Time of occurrence
The time of the crash event is determined by police reports. The 24-hour day is split
into two phases based on natural sunlight availability: daytime (6:00 AM to 5:59 PM-
when natural light is available) and nighttime (6:00 PM to 5:59 AM- when artificial
illumination is required for vehicular movement). Figure 2 shows that daytime crashes
account for 70.1 percent of all crashes and nighttime crashes account for 29.9% of all
crashes, respectively.

Day
Night

29.9%

70.1%

Fig. 2. Crash proportions as a function of time.

6

Accused vehicle details

There are many vehicle types in the database. Two-wheelers (bikes and scooters), three-
wheelers, four-wheelers, heavy motor vehicles (buses, trucks, tankers, etc.), others (cy-
clists, pedestrians, Hand Carts), and unknown vehicles are the six types of vehicles.
Two-wheelers, led by four-wheelers, and HMVs, were the most often involved vehicles
among all crashes during the six-year sample period (51.9 %, 13.7 %, and 12.3 %, re-
spectively) as shown in Figure 3.

Accused Four Wheeler

Accused HMV
Accused Others
Accused Two wheeler
11.3% 13.7% Accused unknown
2.4% Accused three wheeler

12.3%

8.4%

51.9%

Fig. 3. The proportion of accused vehicles that were involved in a crash.

Victim vehicle details

Two-wheelers, led by Others (pedestrians, bicycles, etc.) were the most common victim
vehicles, accounting for 35.3% and 34.3 % of all crashes, respectively as shown in
Figure 4. Both of these classes of road users are considered Vulnerable Road Users
(VRU), and they need special precautions to protect their lives.
 7

Victim Four wheeler

Victim HMV
Victim Others
Victim Two Wheeler
7.2% 10.7% Victim Three Wheeler

12.6%

35.3%

34.3%

Fig. 4. The proportion of Victim vehicles that were involved in a crash.

Road Type
The three types of roads are arterial roads, state highways, and major district routes.
Arterial roads accounted for 69.1% of all crashes, followed by state highways and major
district roads, which accounted for 22.5 percent and 8.4 percent of all crashes, respec-
tively as shown in Figure 5.

Arterial Road
Major District Road
State Highway

22.5%

8.4%

69.1%

Fig. 5. The proportion of crashes on various types of road

8

4 arm Road present

curved present
straight present

7.9%

54.9% 37.2%

Fig. 6. The proportion of crashes that occurred on different road geometry.

Road Geometry information

The four-arm lane, the curved road, and the straight road are the three forms of road
geometry. More than half of the crashes, 54.9 %, occurred on straight roads, 37.2 per-
cent occurred on curved roads, and 7.9 percent occurred on four-arm roads as shown in
Figure 6.

Vehicle maneuver
Several vehicle maneuvers have been registered by the police to the database. Crossing,
diverging, merging, heading straight, overtaking, U-Turn, weaving, and driving on the
wrong side of the road was among the data sets. Crossing paths and going straight are
similarly responsible for collisions (26.4 percent and 26.4 percent, respectively) as
shown in Figure 7, which is surprising given that traveling straight is believed to be a
lower risk factor than crossing, and was followed by wrong-side driving, which is also
very common in Indian cities.
 9

Crossing present
Diverging present
Straight present
Merging present
Overtaking present
U-turn present
26.4% Weaving present
23.5% Wrong side driving present

9.5% 6.9%

0.9%
1.9%
4.5%
26.4%

Fig. 7. The proportion of crashes for various vehicle maneuvers

Cause of crash
The police report detailed the causes of the crash, which involved overspeeding, using
a mobile phone while driving, vehicle skidding, and negligence, among other things.
As a result, the variables were classified into four groups based on their characteristics.
Negligence, overspeeding, mobile phone use, drunken driving, and signal jumping are
all aspects of human causes. Similarly, vehicle defects and vehicle overturn are classi-
fied as vehicle faults, while road repairs, road distresses, road runoff, and skidding are
classified as road conditions, and wrong-side driving/merging/crossing/U-turn is clas-
sified as directional related. As is widely believed, human error was a major factor in a
large number of collisions, accounting for 67.1 percent of all crashes. Directional re-
lated causes accounted for 24.1 percent of the total, while road conditions and vehicle
faults only accounted for 5.4 percent and 3.4 percent, respectively as shown in Figure
8.
10

Driver factor
Direction factor
Road factor
Vehicle factor
5.4% 3.4%

24.1%

67.1%

Fig. 8. The Proportion of Crash Causes

Head on present
Hit & run present
Hit from side
Rear end collision
6.4% Hit with parked vehicle
18.2% 27.9%

13.7%
33.8%

Fig. 9. Accident Type Proportion

Type of accident
Head-on collisions, rear-end collisions, hit-and-runs, side-impact collisions, and colli-
sions with parked vehicles are all mentioned in the data. With 33.8 percent, the hit from
the side had the biggest share, confirming prior findings on crossing route maneuver-
related collisions. Other categories of crashes, such as head-on collisions, rear-end col-
lisions, and hit-and-run, accounted for 27.9%, 18.2 %, and 13.7 cents of all crashes,
respectively as shown in Figure 9.
 11

3.3 Case-control study

The following vehicle, person, and environmental variables were investigated, accord-
ing to traffic police reports: (a)season (summer, winter, monsoon, fall, spring); (b) the
day of the accident (day or night); (c) type of accused vehicle; (d) type of victim vehicle;
(e) road geometry (straight road, curved route, junction); (f) road type (arterial, state
highway, Major District Road); (h) cause of the crash condition (driver, route, vehicle);
(g) vehicle maneuver (crossing, merging, diverging, etc); (i)Cause of the crash (Head-
on, Rear-end collision, hit and run, Hit from the side, hit with parked vehicles ). The
two outcome groups have been compared:1. Fatal-persons who died 2. Non-Fatal- in-
jured persons. Since the studied event is rare, the approximate odds ratios for each var-
iable provide a good prediction of the potential chances of being involved in a fatal ac-
cident. Case-control studies are used to determine whether a single exposure is linked
to a specific outcome (Alexander et al., 2015). Case-control studies look at whether a
single exposure is unequally spread in cases and controls, meaning that the exposure is
a risk factor for the outcome under examination. Identifying case and control, or con-
sequence and cause, is a crucial step in a case-control analysis. Following that, the pro-
portions of cases and controls that show the exposure of concern must be identified. If
the exposed cases and controls are a and b, and the unexposed cases and controls are c
and d, then the exposure odds ratio (OR) can be calculated using Equation-1 (12).

𝑂𝑑𝑑𝑠 𝑜𝑓 𝑒𝑥𝑝𝑜𝑠𝑢𝑟𝑒 𝑎𝑚𝑜𝑛𝑔 𝑐𝑎𝑠𝑒𝑠 𝑎⁄

𝑐
𝑂𝑅 = 𝑂𝑑𝑑𝑠 𝑜𝑓 𝑒𝑥𝑝𝑜𝑠𝑢𝑟𝑒 𝑎𝑚𝑜𝑛𝑔 𝐶𝑜𝑛𝑡𝑟𝑜𝑙𝑠 = 𝑏 (1)
⁄𝑑

The upper and lower confidence intervals can be calculated at 5% significance using Equation
2,3 to approximate the significance of OR.

𝑈𝑝𝑝𝑒𝑟 95% 𝐶𝐼 = 𝑒^(ln(𝑂𝑅) + 1.96√((1 ⁄ 𝑎 + 1 ⁄ 𝑏 + 1 ⁄⁄ 𝑑) )

(2)

𝐿𝑜𝑤𝑒𝑟 95% 𝐶𝐼 = 𝑒^(ln(𝑂𝑅) − 1.96√((1 ⁄ 𝑎 + 1 ⁄ 𝑏 + 1 ⁄⁄ 𝑑) )

(3)

4 Results and Discussions

Based on the theoretical framework outlined in the methodology section, the case-con-
trol analysis is carried out. It compares how much each group is exposed to a risk factor
to see whether there is an association between the risk factor and the fatal outcome.
When each component is present and absent, the fatal and non-fatal crash frequency is
compared, and a case-control analysis is conducted. Exposure is a risk factor if the OR
value is greater than one, indicating that the existence of that variable may result in a
12

high fatality rate. For example, when Accused HMVs are involved, the risk of being
involved in fatal rather than non-fatal outcomes are higher (odds ratio OR=3.35(>1);
and associated CI being in the range of 1.29–8.69(both upper and lower limit >1) can
be defined as a significant risk factor (Sig. RF)
When other victim category vehicles were involved in crashes (pedestrians, bicy-
cles, and hand carts), the chance of death is higher (OR=2.91; 95 percent CI, 1.32–
6.42). The Crossing maneuver can result in more deaths than injuries (OR=3.46; 95
percent CI, 1.55–7.69). Monsoon has an OR of 1.26 (>1), but the related CI is in the
range of 0.5399 to 2.9697 (lower limit <1), indicating that it is a minor risk factor (Insig.
RF). Spring has an OR of 0.85(1) and a lower CI limit of less than 1 (0.3044 to 2.3991),
rendering it a non-risk factor (NRF). All other factors are similarly defined and inter-
preted to classify risk factors associated with fatal crash outcomes. The case-control
study's results are presented in Tables 1–6.

4.1 Season
Based on the OR estimates, none of the seasons were statistically significant risk factors
causing fatal crashes from 2015 to 2019. At a 95% confidence interval, summer, mon-
soon, and winter were found to be marginal risk factors. This demonstrates that severe
weather has had an impact on the occurrence of road crashes. In general, Monsoon
(OR=1.26), Winter (OR=1.58), and Summer (OR=1.07) are linked to a higher risk of
fatal crashes than other seasons as shown in Table 1, implying that they are important
causes. In the winter, road accidents are more prevalent than in other months. This
might be due to India's foggy weather in December and January, which causes poor
road visibility(13). People drive a little slower in foggy conditions, but they also retain
a closer following distance to the vehicle in front of them. This, along with the narrower
field of view, raises the danger of a collision (14). The Indian monsoon season usually
lasts from July to September. Rain can have a variety of effects on driving. Rain will
decrease the output of headlamps, tail lights, brake lights, and other warning lights, as
well as the driver's ability to see clearly (15). Wet roads, from a technical standpoint,
will reduce the traction of a vehicle's tires on the ground, making driving conditions
riskier(16). Summer in India lasts from April to June. Drivers are affected by high tem-
peratures on both a psychological and physiological level. People get more irritated
with others, they become fatigued, lose their focus, and their reaction time slows as the
temperature rises (14). This might explain why the odds of a traffic collision are higher
in the summer. In India, the spring and autumn seasons are usually pleasant, which may
explain why they were determined to be non-risk factors (OR 1).

4.2 Time
Time is split into two categories: Day (6 am to 5:59 pm) and night (6 pm to 5:59 am).
As shown in Table 1, Night-time is observed as an insignificant risk factor, with OR
2.07. This finding may be attributed to higher speeds levels during night travel, which
raises the risk of a crash. The main contributory risk factors associated with night trav-
 13

els are low night vision combined with poor visual direction on roads(17). As a re-
sponse, special attention must be paid to make nighttime driving safer. The day time is
determined to be non-risky (OR <1). This might be because light and visual glare are
not an issue for the driver's eyesight during daylight driving.

Table 1. Fatal versus Non-fatal environmental risks

Risk
Attributes Outcomes Total OR CI Inference
variables
Fatal 5 0.3044 to
Season Spring NRF
Non-fatal 33 0.8545 2.3991
Fatal 8 0.4463 to
Summer 1.0744 Insig. RF
Non-fatal 44 2.5862
Fatal 9 0.5399 to
Monsoon 1.2662 Insig. RF
Non-fatal 44 2.9697
Fatal 1 0.0268 to
Autumn 0.2055 NRF
Non-fatal 25 1.5774
Fatal 7 0.6213 to
Winter 1.587 Insig. RF
Non-fatal 28 4.0534
Fatal 17 0.2250 to
Time Day 0.4982 NRF
Non-fatal 126 1.1032
Night Fatal 13 0.9065 to Insig. RF
2.0074
Non-fatal 48 4.4451

4.3 Type of Accused Vehicle

The distribution of incidents and OR Accused vehicles is seen in Table 2. The six types
of accused vehicles include two-wheelers, three-wheelers, four-wheelers, HMV, others,
and unknown. HMVs are found to be significant risk factors, with a greater risk of fatal
crashes (OR=3.35) than most of the accused vehicles. Because of their large size, mass,
and greater momentum at a particular speed, fatal crashes are often more likely when
smaller vehicles collide with heavy motorized vehicles (HMVs) such as buses and
trucks (18, 19). As a result, HMV movements must be given careful consideration.
Three-wheelers and unknown accused vehicle categories were observed as insignificant
risk factors associated with fatal crash outcomes. (OR =1.33 for Three wheelers, and
OR= 4.07 for unknown respectively). Despite their reputation for being extremely un-
stable, three-wheelers appear to have a negligible risk of causing a fatal collision. This
is most likely since these vehicles are powered by scooter engines and so have modest
average speeds, and because their floor levels are low, a fall from them does not typi-
cally result in fatal injuries (20). In situations when the hitting vehicle is unknown, such
as in hit-and-run accidents, the victim may not receive prompt medical attention. Due
to their diminished physical power, as they age, older drivers are more likely to cause
serious injuries (3). As a result, three-wheeler movements and unknown vehicles must
be closely monitored. Non-risk variables linked with fatal collision outcomes included
two-wheelers, four-wheelers, and other vehicle types. (OR=0.48 for two-wheelers,
OR= 0.43 for four-wheelers, and OR=0.75 for others.). Because two-wheelers and other
vehicles (bicycles, for example) are so small, their impact on fatal collisions could be
minimal. The severity of the offense increases as the accused vehicle goes from a two-
wheeler to a truck (21). Multi-vehicle two-wheeler collisions were caused by younger
14

riders, improper vehicle speed, increased traffic congestion, and road design difficulties
(22). In the city, road designs should discourage excessive speeds. This might be ac-
complished by including safe and well-designed speed-limiting devices into the road
design (20).

4.4 Type of victim Vehicle

Other type of vehicles, such as cyclists, cyclists, and hand carts (slow-moving vehicles),
were major risk factors (OR=2.91) linked to a higher incidence of fatal collisions as
shown in Table 2. Pedestrians and bicyclists are the most vulnerable road users, accord-
ing to the findings, owing to their limited numbers and lack of security (Rifaat et al.
2011). Surprisingly, HMV was found to be a minor risk factor which might be due to
error while recording the data. From 2015 to 2019, Victim Two Wheelers, Three
Wheelers, and Four Wheelers were shown to be non-risk factors based on OR estima-
tions. When compared to other motorized vehicle users such as two, three, and four-
wheelers, bicycles and hand carts (slow-moving vehicles) are less protected during a
crash, which might be resulting in more severity.

Table 2. Fatal versus Non-fatal type of vehicle risks

Risk
Attributes Outcomes Total OR CI Inference
variables
Fatal 11 0.0913 to
Accused Vehicle 2W 0.4814 NRF
Non-fatal 95 1.8111
Fatal 5 0.2163 to
3W 1.3333 Insig. RF
Non-fatal 18 1.0718
Fatal 2 0.0980 to
4W 0.4379 NRF
Non-fatal 28 1.9573
Fatal 8 1.2969 to
HMV 3.3583 Sig. RF
Non-fatal 17 8.6962
Fatal 2 0.1641 to
Others 0.7571 NRF
Non-fatal 15 3.4936
Fatal 2 0.8311 to
Unknown 4.0714 Insig. RF
Non-fatal 3 45.4028
Fatal 7 0.2025 to
Victim Vehicle 2W 0.498 NRF
Non-fatal 66 1.2247
Fatal 1 0.0465 to
3W 0.3655 NRF
Non-fatal 15 2.8753
Fatal 0 0.0066 to
4W 0.1111 NRF
Non-fatal 22 1.8816
Fatal 5 0.5033 to
HMV 1.4571 Insig. RF
Non-fatal 21 4.2186
Fatal 16 1.3236 to
Others 2.9155 Sig. RF
Non-fatal 54 6.4220

4.5 Road type and geometry

According to the findings, as shown in Table 3, state highway (SH) roads have a higher
(OR=1.961) but the insignificant association with fatal crashes than major district roads
(MDR) and arterial roads in Wardha city. Due to urban sprawl, state highways often
become part of the urban road network. They link the city with towns and districts, as
 15

well as National Highways. Vehicles drive at high speeds as a result of the reduced
traffic and increased number of lanes, making it more likely of a fatal crash. To avoid
fatal crashes, highway segments that travel through concentrated land use must be pri-
oritized(23). As the number of intersections increases, the speed of traffic on arterial
road segments decreases, resulting in fewer fatal collisions (24). This might be one of
the reasons why arterial roads are reported to have a lower risk of fatal crashes than
non-fatal crashes. The low risk of fatal crashes on MDR can also be ascribed to the fact
that they are the roads that connect production and market locations. As a consequence,
the number of intersections may grow, while traffic speeds drop, resulting in fewer fatal
incidents. As compared to straight and 4 arm junctions, curved roads (OR=1.58) are
found to be an insignificant risk factor. Drivers who are traveling at high speeds and
are unaware of the magnitude of a turn can lose control of their vehicles, leave the lane,
and slam into a barrier or tree, or another vehicle. Drivers will have a more dynamic
driving situation on curves, requiring them to track more objects and spread their focus.
In addition, constant curves on the roadway will inevitably trigger vehicle instability
and safety concerns (25). In general, roads in traffic networks are not always straight,
and traffic phenomena on a curved road differ from traffic phenomena on a straight
road because vehicles traveling on a curved road are impacted by centripetal force. This
results in an erratic traffic flow that is more difficult than on straight roads (26).

Table 3. Fatal versus Non-Fatal Road related risks

Risk varia-
Attributes Outcomes Total OR CI Inference
bles
Fatal 19 0.3274 to
Road type Arterial Road 0.7362 NRF
Non-fatal 122 1.6556
Fatal 1 0.0435 to
MDR 0.3405 NRF
Non-fatal 16 2.6684
Fatal 10 0.8249 to
SH 1.9167
Non-fatal 36 4.4534 Insig. RF
Road Ge- 4 arm Road Fatal 2 0.1759 to
0.8163 NRF
ometry present Non-fatal 14 3.7891
Fatal 14 0.7235 to
curved 1.5806 Insig. RF
Non-fatal 62 3.4534
straight Fatal 14 0.3119 to NRF
0.6786
Non-fatal 98 1.4763

4.6 Vehicle maneuver

Concerning the factor vehicle maneuver, crossing paths were observed to be associated
with a 3.46 times higher probability of being involved in fatal crashes than other ma-
neuvers as shown in Table 4. Crossing path collisions are a form of a traffic collision
in which one moving vehicle cuts across the path of another as they were coming from
either lateral or opposite directions, resulting in a collision at or near an intersection.
Vehicle-vehicle crossing path collisions are common, but so are vehicle-pedestrian and
vehicle-cyclist collisions. When there are no physical crossing facilities at intersections,
drivers prefer to travel faster and pay less attention to crossing pedestrians(27). This is
can be attributed to unmarked roadways, where people cross the road at different speeds
16

and directions, resulting in a high risk of road injuries. U-turns were shown to be an in-
significant risk factor for fatal collisions, with a 1.31 times higher chance of being in-
volved. U-turns can be found at both junction and midblock (i.e. crossing U-turn inter-
sections and median U-turn respectively). The left turn (right turn in India), or angle,
the crash is one of the most dangerous forms of junction collisions (28). As a result,
distinct phases for right-turn movements (i.e., protected right-turn phases) are given to
reduce the likelihood of such crashes, particularly in heavy right-turn traffic. As a result,
at junctions, higher cycle durations and delays are recorded. As a response, optimal
designs are necessary to improve traffic efficiency while also ensuring safety such as
the prohibition of right-turn operations at the main junction in part or all. Diverging,
merging, overtaking, weaving, and driving on the wrong side of the road were all found
to have an OR less than one, indicating that they are not risk factors at study segments
in Wardha city.

4.7 Cause of crash

In general, the human factor is the most important factor that contributes to traffic col-
lisions followed by road conditions and vehicle defects(29). As shown in Table 5 Hu-
man factors such as negligence, overspeeding, drunk driving, and so on were found to
cause fatal crashes 1.72 times more often (OR=1.72) than vehicle or road condition-
related causes in this study. As a result, careful consideration must be paid to enforce-
ment to ensure safe driving. With an OR of less than one, the direction, road condition,
and vehicle fault were shown to be non-risk variables. It may seem paradoxical, but
degraded pavements (one-way portions, four-legged/signalized junctions) may be ben-
eficial to safety. Drivers may be more careful and drive at lower speeds on poorly main-
tained pavements, which might explain this (30). Commercial vehicles, as opposed to
non-commercial cars and motorbikes, were involved in more road crashes due to me-
chanical failure (31). As a result, commercial vehicle maintenance should be given ex-
tra attention.

4.8 Accident Type

Type of accident is classified into four types as head-on, Hit & run, Hit from the side,
Rear-end collision. Head-on collisions (OR=1.61) and side-impact collisions
(OR=1.89) were found to be risk factors for fatal crashes at greater chances than hit-
and-run and rear-end collisions. Head-on crashes cause the most severe injury, while
collisions with parked vehicles/objects have the least impact(32). Side impact collisions
are linked to both failures to yield and disobeying traffic signals (33). Provisions can
be made for opposing and side-by-side traffic streams to reduce head-on collisions and
side-impact collisions. The variables including hit and run, rear-end collision, and hit
with the parked vehicle all had an OR less than one, indicating that they are minor risk
factors. According to the Ministry of Road Transport and Highways (MoRTH), hit-
and-run incidents accounted for over 14% of all accidents in India, up from 11.6 percent
the year before (MoRTH, 2017) (34). As a result, hit-and-run cases should be prioritized
so that their number does not arise in the future, becoming a significant risk factor for
fatal collisions. Researchers have long known that the danger of a fatal rear-end acci-
dent is more than twice as high in the dark as it is during the day. As a result, cars
 17

should be equipped with rear lights, rear marker plates, and retro-reflective tapes to
make the vehicle's rear visible to oncoming traffic. To avoid being hit by a parked car,
reflective warning signs must be used to make them visible (17).

Table 4. Fatal versus Non-Fatal vehicle movement-related risks

Attributes Risk variables Outcomes Total OR CI Inference
Vehicle Fatal 15 1.5562 to
Crossing 3.4615 Sig. RF
manoeuvre Non-fatal 39 7.6998
Fatal 0 0.0105 to
Diverging 0.1815 NRF
Non-fatal 14 3.1236
Fatal 1 0.0862 to
merging NRF
Non-fatal 8 0.7155 5.9372
Fatal 6 0.2527 to
straight NRF
Non-fatal 48 0.6563 1.7043
Fatal 0 0.0862 to
overtaking 0.6211 NRF
Non-fatal 4 5.9372
Fatal 0 0.0530 to
U turn 1.1311 Insig. RF
Non-fatal 2 24.142
Fatal 2 0.1444 to
Weaving NRF
Non-fatal 17 0.6597 3.0142
Wrong side Fatal 6 0.3010 to
0.7857 NRF
driving present Non-fatal 42 2.0513

Table 5. Fatal versus Non-Fatal Cause-related risks

Attributes Risk varia- Out- Total OR CI Inference
bles comes
Cause of Human factor Fatal 23 1.7293 0.7017 to Insig. RF
crash Non-fatal 114 4.2618
Direction Fatal 6 0.7616 0.2920 to NRF
Non-fatal 43 1.9864
Road condi- Fatal 0 0.2331 0.0134 to NRF
tion Non-fatal 11 4.0606
Vehicle fault Fatal 1 0.9655 0.1121 to NRF
Non-fatal 6 8.3172

Table 6. Fatal versus Non-Fatal accident type related risks

Attributes Risk variables Outcomes Total OR CI Inference
Type of ac- Fatal 11 1.611 0.7128 to Insig. RF
cident Head on Non-fatal 46 3.6411
Hit & run 0.9615 0.3083 to
Fatal 4 2.9989 NRF

Non-fatal 24
Hit from side Fatal 14 1.8932 0.8633 to
Non-fatal 55 4.1517 Insig. RF
Rear end collision Fatal 1 0.0058 0.0008 to
Non-fatal 36 0.0443 NRF
Hit with parked Fatal 0 0.1961 0.0114 to NRF
vehicle Non-fatal 13 3.3874
18

As a result of the observations, each road section may be classified and prioritized as
very high risk, high risk, and moderate risk by comparing results of the case-control
study variable and frequency of a crash variable in each segment

Table 7. Risk factors across segments

Risk factor segment Comparison with Case-control study output
Segment Name & No. Description Significant risk Insignificant risk factor
factors
1.Bajaj Chowk to Very High-Risk Crossing, Ac- Hit from the side, Human factor,
Ambedkar putla Segment cused -HMV monsoon
2. DuniWale math to Nal-
Crossing Hit from the side, Human factor
wadi Road
3. Bajaj Chowk to Deoli- High-Risk Seg- Human factor, Hit from the side,
Victim others
Naka ments monsoon, Curved, SH
4. Ambedkar putla to Ma- Hit from side, monsoon, Human
Crossing
hatma Gandhi putla factor
5.Shivaji Chowk to Dhu- Hit from side, monsoon, summer,

niwale Math Three-wheeler, Human factor
6.Shivaji Chowk to Arvi- Moderate Risk Head on Collision, monsoon, Hu-

Naka Segment man factor, night time
7.Bajaj Chowk to Shivaji
 Head on Collision, Summer,
Chowk

As indicated in table 7, segment 1 contains two major risk variables that might result in
fatal collisions and is thus classified as a Very High-Risk Segment. Similarly, segments
2-4 have only one substantial risk factor that might result in a fatal collision and are
thus classified as High-Risk Segments. Segments 5-7 have no significant risk concerns
but do have minor risk variables that might result in a fatal crash, thus they are classified
as Moderate-Risk Segments. As a result, the highest priority can be assigned to each
section to prevent future fatal collisions as shown in Table 7.
4.9 Very High-Risk Segments
The road connecting Bajaj Chowk and Ambedkar Putla is found to be a very high-risk
region, with two major risk factors influencing crashes in this segment: crossing path
maneuver and Accused HMV, as well as three other insignificant risk factors. As a
consequence, this segment should be given the utmost priority to avoid any fatal colli-
sions

4.10 High-Risk Factors.

The path that connects DuniWale math to Nalwadi Road has been identified as a high-
risk area, with one major risk factor affecting crashes in this segment: crossing path
maneuver, as well as two minor risk factors such as hit from the side type of crash and
human factor as a cause of the crash. The road from Bajaj Chowk to Deoli- Naka has
been identified as a high-risk segment, with one major risk factor, other victim vehicles
(pedestrians and bicycles), and six minor risk factors, including human factor as a cause
of the crash, hit from the side type of crash, monsoon season, curved road geometry,
crossing path maneuver, and state highway road impacting crashes in this segment. The
 19

Ambedkar putla to Mahatma Gandhi putla section is also found to be a high-risk seg-
ment, with one major risk factor: crossing path maneuver and three minor risk factors
as causes of the crash: hit from the side, monsoon season, and human factor. As a result,
these segments should be given the second-highest priority in order to avoid any fatal
crashes.

4.11 Moderate Risk Factors

The road segments Shivaji Chowk to Dhuniwale Math, Shivaji Chowk to Arvi Naka,
and Bajaj Chowk to Shivaji Chowk are affected by several minor risk factors such as
head-on collisions, side-impact collisions, human factors, monsoon, and summer sea-
son. From Shivaji Chowk to Arvi Naka, three-wheelers were discovered to be a minor
risk factor, so special attention should be paid to this segment to avoid three-wheeler-
related crashes.

5. Conclusions

In urban India, road safety is a major concern, and this is especially true for both major
and minor cities like Wardha. Although most road segment safety assessments are
based on primary data and statistical analysis and modeling methods, only a few studies
have been conducted using secondary data, such as police data, which provides infor-
mation into the different characteristics involved in an accident. In India, primary data
such as traffic volume, speed, and other geometric measurements are seldom found in
databases, making manual collection difficult in normal times and impossible during
active pandemics. This study suggested a basic methodology for establishing a link
between fatal crashes and driver, road, and environmental variables using odds ratio in
the case-control study. The research-based on seven road segments in Wardha, India,
and identified several major and minor risk factors. It should be noted that since the
data collection was small (204 cases), certain risk factors were discovered to be risk
factors but were considered to be statistically insignificant. Larger data sets can yield
better results. As a measure, risk factors that are insignificant but have a high likelihood
of causing a fatal crash should be deemed risk factors so that they do not become major
risk factors in the future.
Several studies have recommended some measures like HMV be prohibited from
entering congested and high-crash-prone areas at peak hours in cities and conduct edu-
cation and training campaigns primarily aimed at educating heavy-vehicle drivers about
the safety of vulnerable road users may be a successful tactic (18). Providing pedestrian
travel infrastructures such as wide shoulders and footpaths with adequate lighting, es-
pecially in urban areas, and encouraging pedestrians to wear retroreflective clothing
will help to alleviate this problem to some degree (34). Kadali & Vedagiri (2016) em-
phasized the significance of a street barrier or median to distinguish vehicular traffic
and facilitate healthy pedestrian road crossing. To regulate human factors, such as
driver behavior, efforts should be made to establish and enforce appropriate traffic laws
and road safety regulations, as well as to raise public consciousness about traffic safety
(29). Adverse weather conditions during the monsoon and summer have been described
20

as a risk factor that could result in a serious traffic accident injury. During inclement
weather, drivers are urged to travel less and not exceed the posted speed limit. Addi-
tionally, drivers should inspect their tires daily to guarantee that they are in good con-
dition(36). Medians can be constructed to divide conflicting traffic to avoid head-
on collisions and crashes caused by improper passing and lane changes. There have
been suggestions for design features in braking that make halting or stopping for the
striking vehicle smoother, in the struck vehicle's strength and energy-absorbing ability
to shield the occupants from intrusion, in the design of interior surfaces of either vehicle
and in the visibility of either vehicle to give drivers better notice to prevent a side impact
collision(33).In light of the findings of this study, Wardha city authorities should con-
duct a thorough safety audit of their roadways to recognize and implement counter-
measures in crash-prone areas. Overall, it can be concluded that concentrating on
the risk factors is beneficial. In urban Indian environments, appropriate strategies can
be implemented to mitigate the severity of accident injuries and encourage a healthy
road ecosystem.
Before concluding, the authors would like to point out that this report only included
police-record-derived data with very small data and did not include geometric or traffic
data, which are both important contributors to fatal crashes. As a result, defining risk
factors, such as traffic and geometric variables, would be a potential focus of this re-
search. Furthermore, the findings are city-specific for the case study. However, the
research method could be used as a preliminary step for cities with similar characteris-
tics but it cannot be translated as it is.

References

1. India, I. Https://Irfnet.Ch/2020/12/04/Irf-India-Expresses-Concern-at-Increased-Number-
of-Fatal-Road-Accidents-in-the-Country/. https://irfnet.ch/2020/12/04/irf-india-expresses-
concern-at-increased-number-of-fatal-road-accidents-in-the-country/. Accessed Feb. 3,
2021.
2. MoRTH. Road Accidents in India-2018. New Delhi, 2019.
3. Panicker, A. K., and G. Ramadurai. Injury Severity Prediction Model for Two-Wheeler
Crashes at Mid-Block Road Sections. Taylor & Francis, 2020.
4. Moskal, A., J. L. Martin, and B. Laumon. Risk Factors for Injury Accidents among Moped
and Motorcycle Riders. Accident Analysis and Prevention, Vol. 49, 2012, pp. 5–11.
https://doi.org/10.1016/j.aap.2010.08.021.
5. Híjar, M., C. Carrillo, M. Flores, R. Anaya, and V. Lopez. Risk Factors in Highway Traffic
Accidents: A Case Control Study. Accident Analysis and Prevention, Vol. 32, No. 5, 2000,
pp. 703–709. https://doi.org/10.1016/S0001-4575(99)00116-5.
6. Valent, F., F. Schiava, C. Savonitto, T. Gallo, S. Brusaferro, and F. Barbone. Risk Factors
for Fatal Road Traffic Accidents in Udine, Italy. Accident Analysis and Prevention, Vol. 34,
No. 1, 2002, pp. 71–84. https://doi.org/10.1016/S0001-4575(00)00104-4.
7. Yau, K. K. W. Risk Factors Affecting the Severity of Single Vehicle Traffic Accidents in
Hong Kong. Accident Analysis and Prevention, Vol. 36, No. 3, 2004, pp. 333–340.
https://doi.org/10.1016/S0001-4575(03)00012-5.
 21

8. Vorko-Jović, A., J. Kern, and Z. Biloglav. Risk Factors in Urban Road Traffic Accidents.
Journal of Safety Research, Vol. 37, No. 1, 2006, pp. 93–98.
https://doi.org/10.1016/j.jsr.2005.08.009.
9. Javouhey, E., A. C. Guérin, and M. Chiron. Incidence and Risk Factors of Severe Traumatic
Brain Injury Resulting from Road Accidents: A Population-Based Study. Accident Analysis
and Prevention, Vol. 38, No. 2, 2006, pp. 225–233.
https://doi.org/10.1016/j.aap.2005.08.001.
10. Al Marzooqi, A. H., M. Badi, and A. El Jack. Road Traffic Accidents in Dubai, 2002-2008.
Asia-Pacific Journal of Public Health, Vol. 22, No. SUPPL. 3, 2010, pp. 2002–2008.
https://doi.org/10.1177/1010539510372834.
11. Police, W. Wardha Police. http://wardhapolice.gov.in/.
12. Alexander, L. K., B. Lopes, K. Ricchetti-Masterson, and K. B. Yeatts. Case-Control
Studies. In ERIC Notebook, Epidemiologic Research and Information Center, Seattle.
13. Singh, S. K. Road Traffic Accidents in India: Issues and Challenges. Transportation
Research Procedia, Vol. 25, 2017, pp. 4708–4719.
https://doi.org/10.1016/j.trpro.2017.05.484.
14. Frits Bijleveld & Tony Churchill. The Influence of Weather on Road Safety. Swov, Vol. 3,
No. October, 2009, pp. 1–6.
15. Sangkharat, K., J. E. Thornes, P. Wachiradilok, and F. D. Pope. Determination of the Impact
of Rainfall on Road Accidents in Thailand. Heliyon, Vol. 7, No. 2, 2021, p. e06061.
https://doi.org/10.1016/j.heliyon.2021.e06061.
16. Bernardin, F., R. Bremond, V. Ledoux, M. Pinto, S. Lemonnier, V. Cavallo, and M.
Colomb. Measuring the Effect of the Rainfall on the Windshield in Terms of Visual
Performance. Accident Analysis and Prevention, Vol. 63, 2014, pp. 83–88.
https://doi.org/10.1016/j.aap.2013.10.008.
17. Ackaah, W., B. A. Apuseyine, and F. K. Afukaar. Road Traffic Crashes at Night-Time:
Characteristics and Risk Factors. International Journal of Injury Control and Safety
Promotion, Vol. 27, No. 3, 2020, pp. 392–399.
https://doi.org/10.1080/17457300.2020.1785508.
18. Balakrishnan, S., S. Moridpour, and R. Tay. Sociodemographic Influences on Injury
Severity in Truck-Vulnerable Road User Crashes. ASCE-ASME Journal of Risk and
Uncertainty in Engineering Systems, Part A: Civil Engineering, Vol. 5, No. 4, 2019, p.
04019015. https://doi.org/10.1061/ajrua6.0001023.
19. Kim, J. K., S. Kim, G. F. Ulfarsson, and L. A. Porrello. Bicyclist Injury Severities in Bicycle-
Motor Vehicle Accidents. 2007.
20. Mohan, D., and P. S. Bawa. An Analysis of Road Traffic in Delhi , India. Accident Analysis
and Prevention, Vol. 17, No. 1, 1985, pp. 33–45.
21. Krishnan, M. J., S. Anjana, and M. V. L. R. Anjaneyulu. Development of Hierarchical Safety
Performance Functions for Urban Mid-Blocks. Procedia - Social and Behavioral Sciences,
Vol. 104, 2013, pp. 1078–1087. https://doi.org/10.1016/j.sbspro.2013.11.203.
22. Allen, T., S. Newstead, M. G. Lenné, R. McClure, P. Hillard, M. Symmons, and L. Day.
Contributing Factors to Motorcycle Injury Crashes in Victoria, Australia. Transportation
Research Part F: Traffic Psychology and Behaviour, Vol. 45, 2017, pp. 157–168.
https://doi.org/10.1016/j.trf.2016.11.003.
23. Naqvi, H. M., and G. Tiwari. Safety Performance Functions for Fatal Crashes on National
Highways in India. Transportation Research Procedia, Vol. 48, No. 2019, 2020, pp. 1185–
1189. https://doi.org/10.1016/j.trpro.2020.08.142.
24. Prajapati, P., and G. Tiwari. Evaluating Safety of Urban Arterial Roads of Medium Sized
Indian City. Proceedings of the Eastern Asia Society for …, Vol. 9, No. December, 2013.
22

25. Kang, X., M. Namgung, A. Fujiwara, W. Kim, and W. Wang. Analysis of Vehicle
Maneuverability and Driving Characteristics on a Curved Road Condition. KSCE Journal
of Civil Engineering, Vol. 23, No. 1, 2019, pp. 420–432. https://doi.org/10.1007/s12205-
018-1803-y.
26. Kaur, R., and S. Sharma. Analysis of Driver’s Characteristics on a Curved Road in a Lattice
Model. Physica A: Statistical Mechanics and its Applications, Vol. 471, 2017, pp. 59–67.
https://doi.org/10.1016/j.physa.2016.11.116.
27. Wu, P., X. Meng, L. Song, and W. Zuo. Crash Risk Evaluation and Crash Severity Pattern
Analysis for Different Types of Urban Junctions: Fault Tree Analysis and Association Rules
Approaches. 2019.
28. Al-Omari, A., N. Shatnawi, T. Khedaywi, and T. Miqdady. Prediction of Traffic Accidents
Hot Spots Using Fuzzy Logic and GIS. Applied Geomatics, Vol. 12, No. 2, 2020, pp. 149–
161. https://doi.org/10.1007/s12518-019-00290-7.
29. Touahmia, M. Identification of Risk Factors Influencing Road Traffic Accidents.
Engineering, Technology & Applied Science Research, Vol. 8, No. 1, 2018, pp. 2417–2421.
https://doi.org/10.48084/etasr.1615.
30. Intini, P., N. Berloco, G. Cavalluzzi, D. Lord, V. Ranieri, and P. Colonna. The Variability
of Urban Safety Performance Functions for Different Road Elements: An Italian Case Study.
European Transport Research Review, Vol. 13, No. 1, 2021.
https://doi.org/10.1186/s12544-021-00490-6.
31. Muhammad Hussain and Jing Shi. Modelling and Examining the Influence of Predictor
Variables on the Road Crashes in Functionally Classified Vehicles in Pakistan. International
Journal of Crashworthiness, 2021.
https://doi.org/https://doi.org/10.1080/13588265.2021.1909839.
32. Yannis, G., A. Kondyli, and N. Mitzalis. Effect of Lighting on Frequency and Severity of
Road Accidents. Proceedings of the Institution of Civil Engineers: Transport, Vol. 166, No.
5, 2013, pp. 271–281. https://doi.org/10.1680/tran.11.00047.
33. Chipman, M. L. Side Impact Crashes - Factors Affecting Incidence and Severity: Review of
the Literature. Traffic Injury Prevention, Vol. 5, No. 1, 2004, pp. 67–75.
https://doi.org/10.1080/15389580490269218.
34. Sivasankaran, S. K., and V. Balasubramanian. Investigation of Factors Contributing to
Pedestrian Hit-and-Run Crashes in India. Taylor & Francis, 2020.
35. Kadali, B. R., and P. Vedagiri. Proactive Pedestrian Safety Evaluation at Unprotected Mid-
Block Crosswalk Locations under Mixed Traffic Conditions. Safety Science, Vol. 89, 2016,
pp. 94–105. https://doi.org/10.1016/j.ssci.2016.05.014.
36. Mao, X., C. Yuan, J. Gan, and S. Zhang. Risk Factors Affecting Traffic Accidents at Urban
Risk Factors Affecting Traffic Accidents at Urban Weaving Sections: Evidence from China.
International Journal of Environmental Research and Public Health, Vol. 16, No. 9, 2019.
https://doi.org/10.3390/ijerph16091542.