Mohan, Tiwari and Bhalla

ROAD SAFETY IN INDIA:

STATUS REPORT 2021

Transportation Research & Injury Prevention Centre

INDIAN INSTITUTE OF TECHNOLOGY, DELHI

1
 Mohan, Tiwari and Bhalla
Geetam Tiwari
Rahul Goel
Kavi Bhalla

Road Safety in India: Status Report 2021

Transportation Research & Injury Prevention Centre

Indian Institute of Technology Delhi

2
 Mohan, Tiwari and Bhalla
NOVEMBER 2021

Acknowledgement
Road Safety in India Status report was conceptualised by Prof Dinesh Mohan who led the
writing of these annual reports since 2015. He passed away in May 2021. This is the first
volume of this report since his passing.
Formatting and publication support Mahesh K. Gaur.
Design and layout by Pragati Singhal

Recommended citation
Tiwari, G. Goel, R. and Bhalla, K. (2022) Road Safety in India: Status Report 2021. New Delhi:
Transportation Research & Injury Prevention Programme, Indian Institute of Technology
Delhi. www.iitd.ac.in/-tripp.

Contents may be reproduced with attribution to the authors.

Cover page shows the percentage distribution of road deaths among different modes of
transport.

© Transportation Research & Injury Prevention Centre (TRIP Centre) Indian Institute of
Technology Delhi.

3
 Mohan, Tiwari and Bhalla
CONTENTS
Executive summary 8

Introduction 16

National road traffic injury fatality rate 16

Vehicle Population 17

Road traffic crash and injury data in India 20

Recording of crashes 20

Reporting of RTI crash data 21

RTI fatality estimates 22

Non-fatal injury estimates 23

Ranking in causes of death and population health 25

International comparison 26

Data used In this report 28

Injury and fatality data 28

Data from MoRTH reports 29

Summary 34

Analysis of data at national level 35

National fatality rates 35

Estimates of modal share of RTI fatalities in India 37

Fatality distribution by age and sex 38

State wise analysis 40

Summary 43

Urban Safety 44

4
 Mohan, Tiwari and Bhalla
City data 44

RTI details for selected cities 48

RTI victims and impacting vehicles 49

Road traffic fatalities by type of road user and time of crash 50

Road user risk analysis 50

Occupant risk per hundred thousand vehicles 51

Personal fatality risk per 10 million trips 52

Fatalities associated with each vehicle type accounting for exposure 53

Conclusions from detailed city studies 54

Summary 55

District level safety 56

Intercity highways 59

Introduction 59

Crash patterns 61

Other studies 64

Summary 65

International knowledge base 66

International knowledge base for control of road traffic injuries 66

Results of systematic reviews 67

Legislation and enforcement 67

Education campaigns and driver education 67

Vehicle factors 69

Environmental and infrastructure factors 69

5
 Mohan, Tiwari and Bhalla
Pre-hospital care 70

Summary 71

Way forward 72

Practice points 72

Pedestrian and bicyclist safety 72

Motorcyclist and motor vehicle safety 73

Road measures 73

Enforcement 73

Pre-hospital care, treatment and rehabilitation 73

Research agenda 73

Institutional arrangements 74

References 77

6
Mohan, Tiwari and Bhalla

7
 Mohan, Tiwari and Bhalla
EXECUTIVE SUMMARY

ROAD TRAFFIC CRASH AND INJURY DATA

• According to official statistics 151,113 persons were killed, which

corresponds to 11.6 deaths per 100,000 population.

• Of those who died, only 14% are females and the rest are males.
Internationally, females always have a minority share in road deaths.
However, in India, their share is among the lowest in the world. This
may be because of much lower exposure of females to traffic risk. Share
of women in total number of motor vehicle license holder is about 6
percent, while the rest of the license holders are men.

• Over the last decade (2009-2019) road traffic crashes have been 13th
largest contributor to health burden (deaths and disabilities) in India.
For the working age population (15-49 years), they are the sixth largest
contributor.

• The number of cars and motorised two-wheelers (MTW) registered in

2019 was 36.5 and 221.2 million respectively. The official registration
data over represent the number of vehicles in actual operation because
vehicles that go off the road due to age or other reasons are not removed
from the records. The actual number of personal vehicles on the road is
estimated to be 50%-60% of those mentioned in the records.

• Censuses and population-level sample surveys indicate that percentage

of households owning a car as well as those owning a MTW have more
than doubled in the past decade (2008-2017). Over this period, about
1.5 million cars and 10 million motorcycles were registered every year—
equivalent to 0.6% new households owning a car and 4% owning a
motorcycle every year. In 2017, seven percent households owned at
least one car and 45% owned at least one motorcycle. Cycle ownership,
on the other hand, has stabilised between 40-45 percent.
8
 Mohan, Tiwari and Bhalla
• There is evidence suggesting that number of road
deaths in India is under-reported, however, its extent
is not well understood. ‘Global Burden of Diseases,
Injuries, and Risk Factors Study’ estimated that in
2019, 211,975 deaths (95% confidence interval:
159,343 - 250,315) due to road injuries occurred in
India.
This estimate is 40% greater than government-
reported number of deaths. A National Burden
Estimates study, using Sample Registration System
(SRS) estimates of deaths by different causes,
reported 275,000 road deaths in 2017. This estimate
is 82% higher than the government-reported
number (150,785) for the corresponding year.
Figure 1 : Number and rate of road deaths and annual fuel consumption
in India from 1971 through 2019 (Source:NCRR 2015 and transport
research wing 2020. (Ref page 16)
• Police data should not be used for studying the
epidemiology of non-fatal road traffic injuries (RTI)
in the country. The official estimate of non-fatal RTI
in 2019 was 451,361 which probably underestimates
injuries requiring hospitalization by a factor of 5 and
all injuries by a factor of 20.

• Annual reports published by Ministry of Road

Transport and Highways (MORTH) and National
Crime Records Bureau (NCRB) have erroneously
reported district-level deaths for million-plus cities.
In some years, they have reported correctly for
the cities. Because of this inconsistency, the yearly
changes in road deaths in the cities are not reliable.

• Country income level cannot be taken as excuse for

inefficient data collection systems and it is possible
for countries like India to set up professionally
managed data collection systems that give a Figure 3 Car and motorcycle ownership in India (Ref page 19)
reasonably accurate estimate of RTI fatalities.

• The numbers and proportions of different road

users killed and injured as mentioned in Ministry of
Road Transport and Highways (MoRTH) reports are
erroneous and cannot be used for any analysis.

• Tables dealing with causes of road traffic crashes

should not be used for any analysis or policy making.

• This situation can only be improved by MoRTH with

a complete revamp of the data collection systems in
collaboration with the Ministry of Home Affairs and
establishment of a professional data and analysis
department.

Figure 3 Car and motorcycle ownership in India (Ref page 19)

9
 Mohan, Tiwari and Bhalla
ANALYSIS OF DATA AT NATIONAL LEVEL

• The total number of deaths in 2019 was 13 times • Tamil Nadu witnessed the greatest reduction in
greater than in 1971 with an average annual road death rate over the 5-year period, where the
compound growth rate (AACGR) of 5.4%, and the rate reduced from 25 to 15 per 100,000 persons. An
fatality rate in 2019 was 5.4 times greater than in in-depth understanding of how this was achieved
1971 with an AACGR of 3.9%. would be useful for developing effective road safety
policies across the country.
• The only way the decline of RTI fatalities can be
brought forward in time is to institute evidence- • In the states of Bihar, Uttar Pradesh, Jharkhand,
based India-specific road safety policies that are and Odisha, death rates have increased by 25% or
more effective. more. This is a worrying trend as these four states
contribute one in four road deaths in the country.
• The official estimates of share of pedestrian
among all road deaths are extremely low compared • The impact of Motor Vehicles (Amendment) Act
to independent researchers’ estimates (~15% vs that was passed in August 2019 cannot be evaluated
~35%), therefore, official estimates for all other without monthly data and a greater understanding
modes will also be wrong. The error in the official of the extent of its implementation across the states.
reports regarding types of road users killed probably Due to COVID restrictions in 2020 and 2021, it may
arises from a wrong coding of the victims’ status be few years before a robust analysis could be done
and the procedure needs to be reviewed carefully to understand its effectiveness.
and revised.
• Since road death rates in states and union
• It is not known why the involvement rate of territories do not seem to be influenced strongly
children (<18 years) and the elderly (>59 years) in by location in the country (culture) it suggests that
India is lower than that in the USA when a large state RTI fatality rates may be more influenced by
number of children walk, cycle and travel on infrastructure availability, vehicle modal shares,
overloaded vehicles to school in India. Reasons for road design, and enforcement.
these differences need further study. Higher level
of under-reporting of road deaths among older • Much more attention will have to be given to
adults, as indicated by independent population- street and highway designs and enforcement issues
level surveys, may explain this inconsistency for that have an influence on vulnerable road user
older adults. safety than current practice of focussing on motor
vehicles. This will require a great deal of research
• Among the 18 largest states contributing 96% and innovation as designs and policies currently
of country’s road deaths, during the 5-year period being promoted do not seem to be having the
from 2015 to 2019, road death rates have reduced desired effect in improving road safety.
in half the states, while they have increased in the
other half. Significant reductions occurred in Tamil
Nadu, Gujarat, Telangana, and West Bengal while
significant increase in Assam, Bihar, Jharkhand,
Odisha, Madhya Pradesh, and Uttar Pradesh.

10
 Mohan, Tiwari and Bhalla
URBAN SAFETY

• During the two-year period of 2018 and 2019, the

average fatality rate for all 53 cities with a population
of 1 million or more was 11.2 per 100,000 population
which is slightly lower than the national average of
11.6 per 100,000.

• The five cities with the highest death rates are

Allahabad, Vijayawada, Asansol, Kollam and Jaipur
with an average death rate of 22 per 100,000, which
is twice the national average. Cities with death
rates lower than half the national average are the
following (in ascending order of death rates)—
Kolkata, Greater Mumbai, Srinagar, Hyderabad,
Kannur, Pune and Ahmedabad.

• A detailed study of police reports of road deaths

was conducted in nine Tier-I and Tier-II studies for
the 2008-2011 period. The proportion of vulnerable
road users (pedestrians, bicyclists, and motorised
two-wheelers) among all road deaths in these
cities range between 84% and 93%. Car occupants
constituted between 2% and 7% of all road deaths,
and occupants of three-wheeled scooter taxis less
than 5 percent.

• An interesting feature emerging from this analysis

is the involvement of MTW as impacting vehicles for
pedestrian, bicyclist and MTW fatalities in cities. The
proportion of pedestrian fatalities associated with
MTW impacts ranges from 8 to 25 per cent of the
total.

• MTW and pedestrian deaths are relatively high at 8

PM to 11 PM when we would expect traffic volumes
to be low. Surveys done in Agra and Ludhiana
suggest that due to lower volumes vehicle velocities
can be higher at night, adequate street lighting is
not present, and there is very limited checking of
drivers under the influence of alcohol.

• Following countermeasures need to be given

priority in cities: safe pedestrians paths and crossing
facilities, speed control by traffic calming measures
like raised pedestrian crossings, change of road
texture, rumble strips and use of roundabouts.

11
 Mohan, Tiwari and Bhalla
DISTRICT - LEVEL SAFETY

• An analysis of road traffic deaths in six districts (including urban

and rural areas) of Chhattisgarh state from 2017 to 2019 shows that
the motorcyclists form most of road death victims with a share of 60
percent. This share of motorcycle in road deaths is much greater than
those reported in the past across multiple locations in India. This may
be because of rapidly increasing ownership of motorcycles in India that
is resulting in equally rapid changes in traffic injury patterns.

• Pedestrians and cyclists form the second largest group of road users
among road death victims, with a total share of 25% (21% and 4%,
respectively). In total the share of vulnerable road users (including
motorcyclists) is about 85 percent of all deaths.

• One in four motorcycle deaths resulted from single-vehicle crashes

i.e. skidding or hitting a fixed object on the road. Another 40% resulted
from crashes with trucks or tractors. Up to 75% of pedestrian and cyclist
deaths occurred in crashes with trucks/tractors or motorcycles.

• To contain the large burden of deaths due to motorcycle use, there

needs be strict enforcement of motorcycle helmets for riders as well as
passengers. The enforcement should ensure that helmets are of standard
quality, correct size and are strapped properly. The enforcement should
not be limited to cities or towns but should be extended to rural roads.

INTERCITY HIGHWAYS

• National Highways (including expressways) comprise only 2% of

the total length of roads in India but account for 36% of the fatalities.
Fatality rate per km of the road is the highest on NH with 0.67 deaths
per km annually and this fact should be the guiding factor in future
design considerations. Expressways had a length of only 1,000 km in
the country in 2014 but a high death rate of 1.8 per km per year. This
should be a cause for concern.

• A majority of those getting killed (68%) on highways in India comprise

pedestrians, cyclists and motorcyclists. Pedestrian and MTW proportions
are very high except on six-lane highways where the proportion of truck
victims is much higher. The high level of involvement of vulnerable
road users on highways is highly unexpected in many high-income
countries.

12
 Mohan, Tiwari and Bhalla
• Trucks and buses are involved in about 70 percent
of fatal crashes in both rural and urban areas. This is
again very different from western countries where
there are significant differences in rural and urban
crash patterns.

• On 4-lane divided roads head-on collisions

comprise 19% of the crashes. Divided 4-lane roads
are justified on the basis that these would eliminate
the occurrence of head-on collisions. The fact that
this is not occurring means that many vehicles are
going the wrong way on divided highways. This is
probably because tractor and other vehicles go the
wrong way when they exit from roadside businesses
and the cut in the median is too far away. Figure 1 : Number and rate of road deaths and annual fuel consumption
in India from 1971 through 2019 (Source:NCRR 2015 and transport
research wing 2020. (Ref page 16)
• Rear-end collisions (including collisions with
parked vehicles) are high on all types of highways
including 4-lane highways. This shows that even
though more space is available on wider roads rear-
end crashes do not reduce. This is probably due to
poor visibility of vehicles rather than road design
itself. Countermeasures would include making
vehicles more visible with the provision of reflectors
and roadside lighting wherever possible.

• Following countermeasures need to be

experimented with: physical segregation of slow
and fast traffic, provision of 2.5m paved shoulders
with physical separation devices like audible &
vibratory pavement markings, provision of frequent
and convenient under-passes (at the same level as
surrounding land) for pedestrians, bicycles and
other non-motorized transport, and traffic calming
Figure 3 Car and motorcycle ownership in India (Ref page 19)
in semi-urban and habited areas.

• Analysis of road deaths on an access-controlled

expressway shows that 22% of road death victims
are pedestrians. This share of pedestrian even on
access-controlled roads highlights that there may
be large population who access these highways to
either go across or to access public transportation.
In any case, provisions need to be made so that
either of these functions can be safely served by the
expressways.

• Safety would be enhanced mainly by separating

local and through traffic on different roads, or by
separating slow and fast traffic on the same road,
and by providing convenient and safe road crossing
facilities to vulnerable road users.
Figure 3 Car and motorcycle ownership in India (Ref page 19)

13
 Mohan, Tiwari and Bhalla
STATUS OF RESEARCH IN ROAD SAFETY INTERNATIONAL KNOWLEDGE BASE

• India despite having the distinction of being the • Imposing stricter penalties (in the form of higher
second most populous country contributed only fines or longer prison sentences) will not affect
0.7% published articles on road traffic injuries road-user behaviour significantly. In general, the
worldwide. deterrent effect of a law is determined in part by the
swiftness and visibility of the penalty for disobeying
• When normalized for population levels in 2011, the law, but a key factor is the perceived likelihood
India’s output appears poor in comparison with of being apprehended on the road and sanctioned.
both Brazil and China. The gap between India and
China has widened considerably in the past decade. • Driver or pedestrian education programmes by
themselves usually are insufficient to reduce crash
• The number of papers from China per-person per rates. The only effective way to get most motorists
US$ per-capita income are more than three times to use safety belts and motorcyclists to wear
greater than that from India in all areas. This means helmets is with good laws requiring their use and
that if we want to catch up with China in ten years strict enforcement.
with their present levels of productivity, we will
have to grow at more than 10 per cent per year. • Use of seatbelts and airbag-equipped cars can
reduce car-occupant fatalities by over 50%.
• A review of peer reviewed papers on road safety
published from India indicated that only about • Use of daytime running lights on cars shows a
one-third of them included statistical analysis and reduction in the number of multi-vehicle daytime
modelling. crashes by about 10–15%. Similar results have been
confirmed for the use of daytime running lights by
• Road traffic injury research output is still sub- motorcyclists.
critical in India and not enough original research
findings can be used for India specific policy making • Traffic-calming techniques, use of roundabouts,
for the future. and the provision of bicycle facilities in urban areas
provide significant safety benefits.

• A great deal of additional work needs to be done

on rural and urban road and infrastructure design
suitable for mixed traffic to make the environment
safer for vulnerable road users. This would require
special guidelines and standards for design of, (a)
roundabouts, (b) service lanes along all intercity
highways, and (c) traffic calming on urban roads
and highways passing through settlements.

14
 Mohan, Tiwari and Bhalla
WAY FORWARD

• Reserve adequate space for non-motorized modes • Research agenda

on all roads where they are present.
o Development of street designs and traffic-
• Notification and enforcement of mandatory use calming measures that suit mixed traffic
of helmet and daytime headlights by two-wheeler with a high proportion of motorcycles and
riders. non-motorized modes.
o Highway design with adequate and safe
• Traffic calming in urban areas and on rural highways facilities ffor slow traffic.
passing through towns and villages. o Pedestrian impact standards for buses &
trucks.
• Construction of service lanes along all 4-lane o Evaluation of policing techniques to
highways and expressways for use by low-speed minimize cost and maximize effectiveness.
and non-motorised traffic. o Eff
Effectiveness of pre-hospital care measures.
o Traffic calming measures for mixed traffic
• Removal of raised medians on intercity highways streams including high proportion of
and replacement with steel guard rails or wire rope motorised two-wheelers.
barriers.
• Establish National Board/Agency for Road Safety
• Modern knowledge regarding pre-hospital care
should be made widely available with training of • Establish a special central department for coding
specialists in trauma care in the hospital setting. and recording all fatal crash data. The data so
collected should be anonymised and made available
publicly for analysis.

• Establish safety departments within operating

agencies.

• Establish multidisciplinary safety research centres

at academic institutions.

15
 Mohan, Tiwari and Bhalla
INTRODUCTION

NATIONAL ROAD TRAFFIC INJURY FATALITY RATE

Figure 1 : Number and rate of road deaths and annual fuel consumption in India from 1971 through 2019 (Source:N-
CRR 2015 and transport research wing 2020.

According to official statistics 151,113 persons were The basis for these estimates is given in a later
killed and 451,361 injured in road traffic crashes section. Road traffic injuries (RTI) in India have been
in India in 2019 (MORTH, 2020). This corresponds increasing over the past twenty years though the
to a death rate of 11.5 per 100,000 population. rate of increase has been varying (Figure 1). Number
However, the number of injuries is probably an of road deaths increased rapidly from years 2004
underestimate, as not all such cases are reported through 2011 at a rate of 6.8% every year. Since 2012,
to the police (Mohan et al. 2009, Gururaj 2006). road deaths have been increasing at a much lower
The actual number of injuries requiring hospital rate of 0.8% every year. Figure 1 shows number of
visits may be 2,000,000-3,000,000. In GBD-2010, road deaths and number of deaths per 100,000
we estimated that there were 2.2 million injuries population. During the 8-year period from 2012 to
in India that warranted hospital admission, and 18 2019, the death rate has been stable at around 11.5
million injuries warranted an emergency room visit deaths per 100,000 persons.
(Bhalla et al. 2014).

For a projected population of 1.31 Billion

16
 Mohan, Tiwari and Bhalla
VEHICLE POPULATION

Figure 2 : Cars and motorized two-(MTW) registered in India by year 1950-2016. (Source : Transport Research Wing,
2018).
Figure 2 shows the growth of personal motor
vehicles registered in India by year according to
official data (Transport Research Wing 2018). The
official registration data over represent the number
of vehicles in actual operation because vehicles that
go off the road due to age or other reasons are not
removed from the records. This is because personal
vehicle owners pay a lifetime tax when they buy a car
According to official statistics
and do not de-register their vehicle when they junk
151,113 persons were killed and
them. The actual number of personal vehicles on the
451,361 injured in road traffic
road is estimated to be 50%-55% of those registered
crashes in India in 2019. This is
in India (Expert Committee on Auto Fuel Policy 2002,
probably an underestimate, as
Goel et al. 2015, Mohan et al. 2014).
not all injuries are reported to
The number of cars and motorised two-wheelers
the police. The actual numbers
(MTW) registered in 2019 was 38.4 and 221.3 million
of injuries requiring hospital
respectively (TRW, 2021). If we assume that ~60%
visits may be 2,000,000-
of them were actually on the road, then the actual
3,000,000 persons.
number of cars and MTWs present on the roads would
be approximately 23 and 133 million respectively, and
total personal vehicle ownership (including cars and
MTW) estimated at ~12 per 100 persons in 2019. Since
the actual number of vehicles on the road is much less
than that officially registered in India, any RTI fatality
Note: Actual numbers on the road would
rates calculated per vehicle on the basis of official data
be considerably less, see text. will give unrealistically low estimates.
17
 Mohan, Tiwari and Bhalla
Table 1 : Personal vehicle ownership and official road traffic fatality rates per 100 population (Source : WHO, 2015)

*Vehicle ownership rate adjusted for number of actual vehicles on road. See text.

Table 1 shows the personal vehicle ownership

Since the actual number of
and official road traffic fatality rates per 100,000
vehicles on the road is much less
population for ten countries including India (WHO,
than that officially registered
2015). This table shows eight countries with much
in India, any RTI fatality rates
higher vehicle ownership rates than India but lower
calculated per vehicle on the
RTI fatality rates. These data show that it is not
basis of official data will give
necessary that increases in vehicle ownership rates
unrealistically low estimates.
always result in increases in RTI fatality rates.

1 https://morth.nic.in/sites/default/files/RTYB-2017-18-2018-19.pdf
18
 Mohan, Tiwari and Bhalla
Figure 3 Car and motorcycle ownership in India

Figure 3 presents trend of percentage household We use 2019 India population of 1.31 billion, in-use
owning cars and motorised two-wheelers over the number of cars and MTWs of 23 and 133 million,
25-year period from 1993 to 2017. This data includes respectively, and an average household size of
ownership levels reported by National Family Health 4.8. Using these numbers and assuming each
Surveys (NFHS) for 1993, 1998, 2005, and 2015 (IIPS, households owns only one MTW or one car, 2019
2021), those reported by Census in 2001 and 2011 estimate of household ownership of 8.4% for cars
(Chandramouli, 2012) and Longitudinal Ageing and 49% for MTW. These estimates are close to
Study of India (LASI) for 2017 (IIPS et al., 2020). population-survey estimates of 7.5% and 45% for
Both NFHS and LASI are population-representative 2017 as reported in Figure 3 above. Note that there
sample surveys. For motorcycles, from 2011 to is a gap of 2 years between the two estimates.
2017, there are three data points lying on a linear
trend. The sudden jump of car ownership from 2015
The time trend indicates a rapidly increasing
to 2017 could be an overestimate, as LASI survey
ownership of vehicles. For motorcycles, the rate of
for 2017 included only those households that have
growth in urban and rural areas is almost the same.
at least one member of age 45 years or older. This
However, car ownership trend shows it is increasing
condition could result in overestimation of car
at a much faster rate in urban areas than in rural
ownership. areas. Motorcycle and car ownership has doubled
in the decade from 2008 to 2017. This rapid rate of
Next, we compare the two sources of vehicle vehicle ownership is troubling from road safety’s
numbers— vehicle registration and household- perspective if policies continue to lag.
level ownership from surveys.

2 http://rchiips.org/nfhs/
3 https://www.iipsindia.ac.in/sites/default/files/LASI_India_Report_2020_compressed.pdf
19
 Mohan, Tiwari and Bhalla
ROAD TRAFFIC CRASH AND INJURY DATA IN INDIA

RECORDING OF CRASHES

As in most countries, traffic police are the source of offences under provisions of the Indian Penal Code
official government statistics related to road traffic and the Motor Vehicles Act of India 1988 (Ministry
injuries in India. The main sources of traffic crash of Road Transport and Highways 1988).
data at the national level are the annual reports
published by the National Crime Record Bureau Some of the relevant provisions are:
(Ministry of Home Affairs) titled Accidental deaths
and Suicides in India (NCRB, 2020), and the annual Indian Penal Code
publication of the Transport Research Wing of • Section 279. Rash driving or riding on a public way.
Ministry of Road Transport & Highways (MoRTH) • Section 304A. Causing death by negligence.
titled Road Accidents in India (MORTH, 2020). The • Section 336. Act endangering life or personal
basic information for both these reports comes safety of others.
from all the police stations in the country based on • Section 337. Causing hurt by act endangering life
the cases reported to them. A brief description of or personal safety of others.
the process through which statistics are compiled • Section 338. Causing grievous hurt by act
at the national level follows. endangering life or personal safety of others.

When the occurrence of a traffic crash is brought to Motor Vehicles Act

the notice of a police station (by anyone involved • Section 185. Driving by a drunken person or by a
in the crash; anyone who knows about the crash; person under the influence of drugs.
or a police officer who comes to know about the • Section 184. Driving dangerously.
crash) the information reported is recorded in a
First Information Report (FIR). The details recorded The above provisions determine how a police officer
in the FIR are as observed by the person reporting investigates the crash to assign blame to one of the
the crash. This sets in motion the process of ‘criminal participants in a crash (usually one of the drivers).
justice’ and the police take up investigation of the This is an important issue, as the ‘cause’ of the crash
case. After an FIR has been filed the contents of the has to be recorded as a ‘fault’ of a road user under
FIR cannot be changed except by a ruling from the one or more of the above provisions in most cases.
High Court or the Supreme Court of India. After the This procedure ensures that 80% or more of the
investigation is complete a case file is prepared which cases get attributed to ‘human error’ and there is no
records the details of the crash as determined by place for understanding crashes as a result of a host
the police department (which need not necessarily of factors including vehicle, road and infrastructure
tally with those in the FIR) and the ‘offending party’ design.
(as determined by the investigation) is charged with
20
 Mohan, Tiwari and Bhalla
REPORTING OF RTI CRASH DATA

Statistical tables that summarize key information

about road traffic injuries are reported by police
stations to their district’s Crime Records Bureau, from
where aggregated statistical tables flow upwards to Global Burden of Diseases,
the state’s crime records bureau, and the National Injuries, and Risk Factors Study,
Crime Records Bureau (NCRB), which publishes the estimated the rate of deaths
official statistics for the country (e.g. NCRB 2015). due to road traffic injuries to be
Police-based statistics under-report road traffic 218,876 in 2017.
deaths and injuries in many countries (Bhalla et
al. 2014, W.H.O. 2015, Rosman and Knuiman 1994,
Derriks and Mak 2007).

It has usually been assumed that in India while

many injury cases may be taken to private hospitals
and not get recorded by the police, most fatal RTI
cases get recorded for the following reasons:
• For serious injury cases and deaths on the spot,
or before arrival at a hospital, FIRs are filed with the
police especially if those involved want to pursue a
court case or claim insurance compensation.
• Under Section 165 of The Motor Vehicles Act 1988
(Ministry of Road Transport and Highways, 1988), all
State Governments have been authorised to set up A National Burden Estimates
Motor Accident Claims Tribunals for adjudicating study estimates RTI deaths in
upon claims for compensation in respect of road 2017 in India to be 275,000.
traffic crashes involving death, bodily injury or
property damage. Claims can be made by the person
who has sustained the injury, by the owner of the
damaged property, and by legal representatives of
the deceased. Victims or their legal representatives
in the case of hit-and-run cases can also make
claims. For this reason lawyers look out for such
cases in hospitals or police stations and promise
legal help to make the claim.
• When a RTI victim is admitted to a government
hospital and declared as a RTI case, the patients’
details are recorded as a ‘Medico Legal Case’ by a
police officer stationed at the hospital. If the victim
dies in the hospital, irrespective of the length of
stay in the hospital, the body is released only after Overall, this would imply that
a mandatory autopsy and the relevant details it is possible that the actual
are provided to a police officer seconded by the number of RTI deaths in India
relevant police station. may be more than 40% higher
• Section 146 of the Indian Motor Vehicles Act 1988 than the official estimate.
(Ministry of Road Transport and Highways, 1988)
requires that all motor vehicles (except those owned
by the Central or State Governments) operating in
a public space must be insured against third party
risks.

21
 Mohan, Tiwari and Bhalla
RTI FATALITY ESTIMATES

The extent of under-reporting of road traffic deaths in India is not well

understood. For instance, a record linkage study in Bengaluru covering
23 hospitals found that police data only missed 5% of road traffic
deaths (Gururaj, 2006). Two recent studies have estimated national
road traffic deaths using data from the Sample Registration System,
Registrar general of India. Dandona et al. (2020) as part of the Global
Burden of Diseases, Injuries, and Risk Factors Study (GBD), estimated
the rate of deaths due to road injuries in each state of India from 1990
to 2017 based on several verbal autopsy data sources.

They estimate that there were 211,975 (95% confidence interval: 159,343
- 250,315) deaths in India in 2019 (IHME, 2020). The mid-point estimate
is 40% higher than government-reported number of 151,113 deaths
for the same year. Interestingly, the two estimates are much closer than
they were in 2016, when GBD estimate was 68% higher. Another study
by Menon et al. (2019) reports National Burden Estimates to provide
transparent and understandable disease burdens at the national levels
and estimates RTI deaths in 2017 in India to be 275,000. The latter figure
is 86% higher than the Transport Research Wing estimate of 147,913 for
2017 (MORTH, 2020).

It is possible that most of the critical and immediately fatal cases

get recorded in crowded urban areas of India and those who die in
government hospitals enter the official statistics, however, some fatal
cases in rural areas and those involved in single vehicle crashes may not
get reported. It is likely that the fatality statistic for urban areas in India
may be underestimated by say 10%-20%. According to the MoRTH, 61%
of the RTI fatalities occur in rural areas and it is possible that a larger
number of cases go unreported on rural roads. In a review of European
and Japanese RTI data linkage, Lai et al. (2006) report that total RTI
victims dying within 30 days of the crash are about 30% greater than
those dying on the first day.

If we assume that a significant proportion of fatalities that occur many

days after the crash in rural areas are missed (that would reduce the
number by less than 30% of the total deaths) and a smaller proportion
of deaths on the spot or on the way to the hospital are missed, then we
can expect under-reporting to be around 50% of rural deaths. Overall,
this would imply that the under-reporting of fatalities in India may be
less than 50%. However, this issue cannot be resolved satisfactorily until
such time as the recording of traffic crashes is done in a manner open
to public scrutiny and mechanisms are established to audit the quality
of official statistics of road traffic deaths on a regular basis.

22
 Mohan, Tiwari and Bhalla
To understand the level of under-reporting by
different age groups, Figure 4 presents ratios of
number of road deaths reported by GBD to the
number reported by MoRTH for the five age groups
for year 2019.

A ratio of one indicates that the two sources of data

are in complete agreement with each other. The
disagreement between GBD and MoRTH reported
is the greatest in the age groups 45 years or older.

For 45-60 years, GBD-estimated road deaths are

about twice as many as MoRTH-reported number.
For 60+ age group, GBD-estimates number is more
than five times as high. The reasons for this variation
in the level of under-reporting by age is not yet clear.

Since proportion of all deaths in 60+ age group is

small, high-level of underestimation for this age
group does not translate to equally high levels of
overall under-reporting.

Figure 4 : Age - Specific comparison of GBD and MORTH

reported road deaths.

Ratio of 1 indicates the two sources of data report equal number of

deaths.

NON-FATAL INJURY ESTIMATES

While there is uncertainty among experts about the with decreasing injury severity, 33-38% for severe
level of under-reporting of road traffic deaths, all injuries and 15% for minor injuries (Amoros et al.
experts agree that police reports are a poor source 2008, Amoros, Martin, and Laumon 2006). Studies
of information for non-fatal injury statistics in India. from India also indicate similar trends. A study done
Police databases typically report a small fraction of in Bangalore (now Bengaluru) shows that while
the non-fatal road traffic injuries that occur in most the number of traffic crash deaths recorded by the
countries, including most developed countries police may be reasonably reliable, the total number
(Derriks and Mak 2007, International Traffic Safety of injuries is grossly underestimated (Gururaj, 2001).
Data and Analysis Group 2011). According to a recent According to the study, the ratio of injured people
IRTAD (2014) report police records alone are usually reporting to hospitals to that killed was 18:1. It is
inadequate to carry out analysis on the nature important to note that even this ratio would be
and consequences of serious injuries because the an underestimate as among those injured many
reported number is underestimated. A report from others would have taken treatment at home or from
France also states that under-reporting is inversely private medical practitioners.
and strongly associated with injury severity: there is
a clear gradient of decreasing probability of being
police-reported

23
 Mohan, Tiwari and Bhalla
Another detailed study done in rural northern
A conservative estimate can be
India recorded all traffic-related injuries and deaths
made that the ratios between
through bi-weekly home visits to all households
deaths, injuries requiring
in 9 villages for a year and showed that the ratio
hospital treatment, and minor
between critical, serious and minor injuries was
injuries in India are likely to be
1:29:69 (Varghese and Mohan 1991).
about 1:15:50.
International experience is somewhat similar. In
2013 in the US police-reported motor vehicle traffic
crashes included 30,057 persons killed, 1,591,000
The official estimate of non- injured, and 4,066,000 damage only crashes giving
fatal RTI in 2019 was 451,361 a ratio of 1:53:135 respectively (National Centre for
which probably underestimates Statistics and Analysis 2015). Other studies report
injuries requiring hospitalisation ratios between deaths:serious-injuries:minor-
by a factor of 5 and all injuries injuries as 1:13:102 (Martinez 1996) and 1:14:80
by a factor of 20. (Evans 1991). A more recent report states that in
Netherlands the ratio of the estimated number of
fatalities and hospitalised persons for the year 2000
As non-fatal injury data in India was 15.7 (Derriks and Mak 2007).
are unreliable and the biases
implicit in which cases get Using the epidemiological evidence from India and
recorded not known, police data other countries where better records are available,
should not be used for studying a conservative estimate can be made that the
the epidemiology of road traffic ratios between deaths, injuries requiring hospital
injuries in the country. treatment, and minor injuries in India are likely to
be about 1:15:50.

If the estimate of road traffic fatalities in India (official)

Over the last two and a half in the year 2019 is taken as 151,113, then the estimate
decades the burden of road of serious injuries requiring hospitalization would
traffic injuries in India has be 2,267,000 annually, and that for minor injuries
increased while that due to 7,555,000. The official estimate of non-fatal RTI in
many infectious diseases has 2019 was 451,361 which probably underestimates
declined. In 1990, road traffic injuries requiring hospitalisation by a factor of 4 and
injuries were the 16th leading all injuries by a factor of 20.
cause of health loss, in 2016 it
was ranked 10th. As non-fatal injury data in India are unreliable and
the biases implicit in which cases get recorded
not known, police data should not be used for
Lower national income levels studying the epidemiology of road traffic injuries
cannot be taken as excuse in the country. Any statistical analysis using injury
for inefficient data collection data would be unreliable and this which would
systems and it is possible for render indices such as accident severity (number of
countries like India to set up persons killed per 100 accidents) meaningless.
professionally managed data For these reasons, only fatality data have been used
collection systems that give a for analysis in this report as police data should not
reasonably accurate estimate be used for studying the epidemiology of non-fatal
of RTI fatalities. road traffic injuries in India.

24
 Mohan, Tiwari and Bhalla
Figure 5 : Ranking of the cause of health burden (DALYs) in India in 2009 and 2019 for all ages (Source : Institute for
Health Metrics and Evaluation (IHME, 2018)

RANKING IN CAUSES OF DEATH AND POPULATION HEALTH

Figure 5 shows the diseases ranked by their contribution to overall

health burden for all age and sex groups combined for years 2009 and
2019. (Institute for Health Metrics and Evaluation (IHME). 2018). The
burden reported here is expressed as Disability-adjusted Life Years
(DALYs) which includes burden due to deaths as well as disabilities.

Road traffic injuries have been the 13th leading cause of premature
death in India over the last decade. The burden due to injuries exceeds
the number of those who succumb to many diseases like malaria
and HIV that are acknowledged to be important health issues for the
country.

25
 Mohan, Tiwari and Bhalla
INTERNATIONAL COMPARISON

Figure 6. RTI fatality rate per 100,000 persons reported by different countries vs per capita income. (Source: WHO,
2018).

The 2018 WHO Global Status Report on Road Safety some with higher income levels have also have
provides two sets of road traffic death statistics for higher levels of under-reporting. This suggests that
every country (WHO, 2018). These are the official lower national income levels cannot be taken as
government statistics (usually based on police data) an excuse for inefficient data collection systems,
reported by each country to WHO, and estimates and it is possible for countries like India to set up
produced by WHO through statistical analysis of professionally managed data collection systems
national health data (including vital registration). that give a reasonably accurate estimate of RTI
Figure 6 shows the official RTI fatality rates for fatalities.
different countries plotted against per capita Both the official country data and WHO estimates
income of the countries and Figure 7 shows the (Figures 5 and 6) show that there are countries with
rates for the same countries as estimated by the incomes similar to India that have RTI fatality rates
WHO (WHO, 2018). These figures show that for 43% lower than India. Again, demonstrating that lack of
of the countries the WHO estimates are 1.5 times finances does not necessarily mean that a society
greater and for 26% more than 3 times greater than has to have absence of safety on the roads. At the
the official rates reported by the countries. same time, many countries much richer than India
The ratio of WHO estimate and the official rate for have much higher fatality rates. Therefore, we
different countries is shown in Figure 8. The ratio for cannot depend on growth in national income alone
India is 2.0 as the official reported rate is 11.4 deaths to promote road safety. It will be necessary to put
per 100,000 persons and the WHO estimate 22.6. in place evidence based national safety policies to
These data indicate that some countries with similar ensure improvements in traffic safety.
incomes have lower levels of under-reporting and

26
 Mohan, Tiwari and Bhalla
Figure 7. RTI fatality rate per 100,000 persons as estimated by WHO for different countries vs per capita income.
(Source: WHO, 2018).

Figure 8. Ratio of WHO estimates and official RTI fatality rate per 100,000 persons for different countries vs per capita
income (Source: WHO, 2018)

27
 Mohan, Tiwari and Bhalla
DATA USED IN THIS REPORT

Injury and fatality data

Table 2. Examples of commonly used indicators of the road traffic injury problem. Source: Mohan, D. et al., 2006.

Table 2 shows the different indicators generally In such a situation, the fatality statistics should
used for assessing RTI issues (Mohan et al. 2006). be adequate for predicting trends and relative
Out of all these indicators we only use the number comparisons between different risk factors. Fatalities
of fatalities and fatalities per 100,000 population per 100,000 population are used for all comparisons
for most of our analyses. Only fatality statistics from because the population statistics are expected to
MoRTH reports are used for analysis. We assume be reliable and the index is a good indicator of the
that though the Indian fatality statistics may suffer health burden on the population.
from some underestimation there may not be a
systematic bias in recording of fatalities of specific
road users.
28
 Mohan, Tiwari and Bhalla
Fatalities per population can also be used as proxy under-reporting for pedestrian and bicycle injuries,
for risk of death per trip as international experience night-time crashes, hit and run cases, and crashes
suggests that the average number of trips per on rural roads (Abay 2015, Amoros, Martin, and
person remains relatively stable over time, incomes Laumon 2006, Rosman and Knuiman 1994, Derriks
and place (Knoflacher 2007). Knoflacher further and Mak 2007).
states that average trip rates in cities around the
world vary from 2.8 to 3.8. That total trip rates do Fatalities per 10,000 vehicles and fatalities per
not vary much and generally remain between 3 vehicle-kilometre have not been used in this report
and 4 trips per person per day has been supported except for a few specific comparisons. The official
by many studies around the world (Giuliano and number for number of vehicles in India and cities
Narayan 2003, Hupkes 1982, Santos et al. 2011, are all overestimates (as explained in an earlier
Transport for London 2011, Zegras 2010). section), and therefore, cannot be used for any
calculations. In addition, the indicator fatalities per
Non-fatal injury data are not used at all in this report 10,000 vehicles should not be used for comparison
as they are not likely to give any useful insights. if the modal shares differ from place to place
Injury and accident statistics suffer from a very (Mohan and Tiwari 2000). The number of fatalities
high margin of underestimation as discussed in an per 10,000 vehicles always decreases as the number
earlier section. In addition, international experience of vehicles per capita increase in a society even
suggests that injury and non-fatal crash data can when no specific safety measures have been put in
suffer from many other biases such as relative place (Adams 1987).

Data from MoRTH reports

The latest report on RTI in India, Road accidents If the pedestrian fatality proportions are so low in
in India – 2019 (Transport Research Wing 2020), these official reports, then it stands to reason that
includes many tables giving details of crashes as proportions and numbers for all other road users
reported to the Transport Research Wing (Ministry of will also be wrong. More data will be presented to
Road Transport & Highways, Government of India). strengthen this argument in subsequent sections
Much of the details provided in the official report for of this report. The numbers and proportions of
RTI in India could not be used in the present analysis different road users killed and injured as mentioned
as the data do not appear to be reliable. A summary in MoRTH reports are erroneous and cannot be used
of the reasons why data from various tables in the for any statistical analysis.
report could not be used is given in Table 3. Although it is clear that NCRB and MoRTH reports do
Work done by independent researchers using not provide valid statistical tabulations on types of
police reports (same sources as used by MoRTH) road-users killed and other successfully generated
from different cities and highway locations show reasonable estimates by inspecting detailed police
very different results as shown in Table 4 (Mani and reports. Such case files are paper-based and usually
Tagat 2013, Delhi Traffic Police 2014, Tiwari, Mohan, available at the police station with jurisdiction over
and Gupta 2000, Tiwari 2015). In the nationally the location where the crash occurred or at the
representative mortality survey of 1.1 million homes district’s crime records bureau office. Researchers
Hsiao, M. et al. (2013) reported that pedestrians and who are able to acquire requisite permissions need
motorcyclists constituted 37 and 20 per cent of to undertake a tedious process of working with
total RTI fatalities respectively. A more recent study multiple police stations to acquire copies of all
(Dandona, et al, 2020) reports that pedestrians police reports and extracting relevant information.
accounted for 76,729 (35·1%) of all deaths due Clearly this cannot be done over a large region as
to road injuries, and motorcyclists accounted for researchers have track changes over time without
67,524 (30·9%). These data make it clear that the the use of substantial resources. Nevertheless,
proportion of pedestrian fatalities in India cannot collecting such data even for a small region or a
be as low as 15 or 9 per cent. In all probability, the short period of time can provide valuable insights
pedestrian fatalities may comprise around 30-35 to researchers and policy makers interested in
percent of all fatalities. addressing local road safety issues.

29
 Mohan, Tiwari and Bhalla
Table 3. Summary of reasons why data from some tables in the Road accidents in India – 2018 (Transport Research Wing,
2019) report could not be used in the present analysis.

30
 Mohan, Tiwari and Bhalla
Table 3 continued. Summary of reasons why data from some tables in the Road accidents in India – 2016 (Transport
Research Wing 2017) report could not be used in the present analysis.

31
 Mohan, Tiwari and Bhalla
Table 3 continued. Summary of reasons why data from some tables in the Road accidents in India – 2016 (Transport
Research Wing 2017) report could not be used in the present analysis.

Table 4. Modal share of road traffic fatalities in Mumbai, Delhi and four rural highway locations in India.

32
 Mohan, Tiwari and Bhalla
Much of the details provided The data regarding cause of crashes and persons
in the official report for RTI in responsible for crashes as reported in the NCRB and
India could not be used in the MoRTH reports is also not reliable. As mentioned
present analysis as the data do earlier it is the IPC codes that decide how a police
not appear to be reliable. officer assigns blame to one of the participants
in a crash (usually one of the drivers). This is an
important issue, as the ‘cause’ of the accident has to
be recorded as a ‘fault’ of a driver under one or more
The data regarding type of road of the 4 or 5 provisions.
user killed, cause of crashes and
persons responsible for crashes This procedure ensures that 80% or more of the
as reported in the NCRB and cases get attributed to ‘human error’ and there is no
MoRTH reports are not reliable. place for understanding crashes as a result of a host
of factors including vehicle, road and infrastructure
design. For example, the MoRTH report (Annexure
36) attributes ‘Drunken driving/consumption
of alcohol and drugs’ as contributing to 4,188
If one of the risk factors is
deaths which amounts to only 3% of the total.
underestimated by a large
Independent studies estimate alcohol and drugs
margin then the estimates for
as a contributing factor in more than 20-30 percent
all the other ‘causes’ or other
of the crashes India (Arora, Chanana, and Tejpal
factors becomes unreliable.
2013, Das et al. 2012, Esser et al. 2015, Gururaj 2006,
Therefore, tables dealing with
Mishra, Banerji, and Mohan 1984). If one of the risk
various causes of road traffic
factors is underestimated by a large margin then the
crashes should not be used for
estimates for all the other ‘causes’ or other factors
any analysis or policymaking.
becomes unreliable. Therefore, tables dealing with
various causes of road traffic crashes should not be
used for any analysis or policymaking.

The summary of data usability in Table 3 suggests

that at present MoRTH reports can only be relied
upon to provide information like date, place, location
and time of fatal crashes. This situation can be
improved by MoRTH only with a complete revamp
of the data collection systems in collaboration with
the Ministry of Home Affairs and the establishment
of a professional data and analysis department
(National Transport Development Policy Committee
2014a).

33
 Mohan, Tiwari and Bhalla
SUMMARY

• According to official statistics 151,113 persons • Over the last decade (2009-2019) road traffic
were killed and 451,361 injured in road traffic crashes have been 13th largest contributor to health
crashes in India in 2019. However, this is probably burden (deaths and disabilities) in India. Among
an underestimate for injuries, as not all injuries are working age population (15-49 years), they are the
reported to the police. sixth largest contributor to health burden.

• The number of cars and motorised two-wheelers • Lower national income levels cannot be taken
(MTW) registered in 2019 was 36.5 and 221.2 million as excuse for inefficient data collection systems
respectively. The official registration data over- and it is possible for countries like India to set up
represent the number of vehicles in actual operation professionally managed data collection systems
because vehicles that go off the road due to age or that give a reasonably accurate estimate of RTI
other reasons are not removed from the records. fatalities.
The actual number of personal vehicles on the road
is estimated to be 50%-60% of those mentioned in • Lack of finances does not necessarily mean that a
the records. society has to have absence of safety on the roads.
We cannot depend on growth in national income
• In 2019, household vehicle ownership is at least alone to promote road safety. It will be necessary to
7.4% for cars and 45% for MTW. This ownership has put in place evidence based national safety policies
grown rapidly from the 2011 ownership levels of to ensure improvements in traffic safety.
4.7% and 21%, respectively.
• The numbers and proportions of different road
• The extent of under-reporting of road traffic users killed and injured as mentioned in the NCRB
deaths in India is not well understood. Global and MoRTH reports are erroneous and cannot be
Burden of Diseases, Injuries, and Risk Factors Study used for any analysis.
(GBD) estimated that in 2019, 211,975 deaths Tables dealing with causes of road traffic crashes
(95% confidence interval: 159,343 - 250,315) due should not be used for any analysis or policy making.
to road injuries occurred in India. This estimate is This situation can only be improved by MoRTH with
40% greater than government-reported number of a complete revamp of the data collection systems in
deaths. A National Burden Estimates study estimates collaboration with the Ministry of Home Affairs and
road deaths in 2017 in India to be 275,000. This establishment of a professional data and analysis
estimate is 86% higher than the MoRTH number department.
(147,913) for the corresponding year. Police data
should not be used for studying the epidemiology
of non-fatal road traffic injuries in the country. The
official estimate of non-fatal RTI in 2019 was 451,361
which probably underestimates injuries requiring
hospitalization by a factor of 5 and all injuries by a
factor of 20.

34
 Mohan, Tiwari and Bhalla
ANALYSIS OF DATA AT NATIONAL LEVEL

NATIONAL FATALITY RATES

Figure 9 shows the official estimates for total number of RTI fatalities The total number of deaths in
and fatalities per 100,000 persons in India from 1971 to 2019 (Transport 2018 was 10 times greater than
Research Wing 2019). The total number of deaths in 2019 was 13 times in 1970 with an average annual
greater than in 1971 with an average annual compound growth rate compound growth rate (AACGR)
(AACGR) of 6%, and the fatality rate in 2019 was 5.4 times greater than of 6%, and the fatality rate in
in 1971 with an AACGR of 4%. Over this 50-year period, road fatalities 2018 was 4.3 times greater than
have grown at a varying rate. There have been periods when road fatality in 1970 with an AACGR of 4%.
trend flattened or when absolute number of fatalities reduced, or the
period when fatalities grew at a fast rate. It is known that motor vehicle
crash rates have a tendency of decreasing along with a downturn in the
national economy (International Traffic Safety Data and Analysis Group
2015):
The only way the decline of
RTI fatalities can be brought
“ Economic downturns are associated with less growth in traffic or a forward at time is to institute
decline in traffic volumes. They are associated with a disproportionate evidence based India specific
reduction in the exposure of high-risk groups in traffic; in particular road safety policies that are
unemployment tends to be higher among young people than more effective.
people in other age groups. Reductions in disposable income may
be associated with more cautious road user behaviour, such as less
drinking and driving, lower speed to save fuel, fewer holiday trips. “

To investigate the link between economy and road traffic deaths, The Indian official estimates
Figure 8 presents annual consumption of diesel and petrol as reported of pedestrian fatalities are
by Ministry of Petroleum and Natural Gas (MoPNG). Both diesel and extremely low compared to
petrol consumption follow a long-term trend similar to that of number independent researchers’
of road deaths. However, diesel consumption has a much stronger estimates (~15% vs ~35%),
correlation with its short-term changes coinciding with similar changes therefore, official estimates
in road deaths. Diesel being used by goods traffic is a strong indicator for all other modes will also be
of economy. wrong.

35
 Mohan, Tiwari and Bhalla
Figure 9. Total number of RTI fatalities and fatalities per 100,000 persons in India (Source: Transport Research Wing
2019).

The slowdown in the growth rate of fatalities since society’s concerns for safety, and predicts an earlier
2012 coincides with the similar slowdown in diesel date of 2030 for the start of decline in RTI fatalities
consumption during this period. Similarly, the in India. If we assume the average growth rate of
flattening of number of fatalities in early 2000s 6% per year declines to nil by 2030, then we can
coincides with the flattening of diesel consumption expect about 200,000 fatalities in 2030 before we
during that period. This shows that road safety in see a reduction in fatalities. The above models use
India is strongly linked to the economy. In other the experience of high-income countries (HIC)
words, if the economy grows at a high rate in the over the past decades in calculating relationships
near future, road deaths are also likely to grow at between vehicle ownership levels and risk of death
the same rate. per vehicle. Therefore, the models presuppose the
Two modelling exercises have attempted to predict onset of decline at specific per-capita income levels
the time period over which we might expect fatality if the past road safety policies of HICs are followed
rates to decline in different countries (Koornstra in the future in countries like India. Based on an
2007, Kopits and Cropper 2005). Kopits and Cropper analysis of RTI fatality trends in Europe and the USA,
use the past experience of 88 countries to model the Brüde and Elvik (2015) suggest that:
dependence of total number of fatalities on fatality ‘A country does not at any time have an “optimal” or
rates per unit vehicle, vehicles per unit population “acceptable” number of traffic fatalities. In countries
and per capita income of the society. Thus, based on with a growing number of traffic fatalities, one
projections of future income growth, they predict cannot count on this trend to turn by itself; active
that fatalities in India will continue to rise until 2042 policy interventions are needed to turn the trend’.
before reaching a total of about 198,000 deaths and If this is true, then the only way the decline of
then begin to decline. Koornstra uses a cyclically RTI fatalities can be brought forward at time is to
modulated risk decay function model, which in a institute evidence based India specific road safety
way incorporates the cyclically varying nature of a policies that are more effective.
36
 Mohan, Tiwari and Bhalla
ESTIMATES OF MODAL SHARE OF RTI FATALITIES IN INDIA

Figure 10. Estimates of the share of different road user fatalities in India (Source: Transport Research Wing 2019, Hsiao,
M. et al. 2013, GBD: Institute for Health Metrics and Evaluation (IHME), IIT Delhi estimate: authors of the present report).

Figure 10 shows estimates of the share of different Since Hsiao et al. and Dandona et al. have estimated
road user fatalities as reported by MoRTH (Transport the fatalities from verbal autopsies with a statistically
Research Wing 2019), estimates made by Hsiao, M. representative sample of households in India (a part
et al. (2013), IIT Delhi, and Dandona et al. (2020). of the Sample Registration System of the Registrar
Hsiao et al. estimates are based on a nationally General of India), it is likely that their numbers are
representative mortality survey of 1.1 million closer to the truth. The IIT Delhi estimate is made
homes in India which reported 122,000 RTI deaths, from detailed analysis of police reports from
IIT Delhi estimate is based an analysis of police various parts of the country, and therefore, may be
records obtained from 8 cities(Delhi Traffic Police, considered as being based on official data, though
2014, Mani, A. and Tagat, A., 2013, Mohan, D. et from a smaller sample in the country.
al., 2013) and a number of locations on rural roads
around the country (Gururaj, G. et al., 2014, Tiwari, Since these latter estimates for pedestrian fatalities
G., 2015, Tiwari, G. et al., 2000, and Dandona et al. are similar, it is quite certain that these estimates are
(2020) estimate is based on several verbal autopsy more reliable than those in MoRTH reports. The error
data sources. in the official reports probably arises from wrong
coding of the victims’ status and the procedure
The MoRTH estimates suggest that pedestrian needs to be reviewed carefully and revised. The
fatalities constitute only 15% of total RTI fatalities Indian official estimates of pedestrian fatalities
in the country. The Hsiao et al. (2013), IIT Delhi are extremely low compared to independent
and Dandona et al. (2020) estimates for share researchers’ estimates (~15% vs ~35%), therefore,
of pedestrian fatalities are 37%, 33% and 35% official estimates for all other modes will also be
respectively. This is a very large gap between the wrong. For the time being we will have to use
official and researchers’ estimates. research estimates for modal share of road traffic
fatalities and not the official number.
37
 Mohan, Tiwari and Bhalla
FATALITY DISTRIBUTION BY AGE AND SEX

Figure 11. RTI fatality distribution and population distribution by age in India
and USA. (Source: Transport Research Wing 2019 and National Centre for
Statistics and Analysis 2015).

Figure 11 shows the RTI fatalities and population It is not known why the involvement rate of children
distribution by age in India and USA (National Center (<18 years) and the elderly (>59 years) in India is
for Statistics and Analysis 2015, Central Bureau of lower than that in the USA when a large number of
Health Intelligence 2019, Transport Research Wing children walk, cycle and travel in overloaded vehicles
2019). In India, the proportion of fatalities for the to school in India. It is possible that the exposure
age group 18-59 is greater than their representation rate of the elderly in India is less than for those in the
in the population and less for the age groups 0-18 USA and this may explain their lower involvement.
years (1:5 of the population) and >59 years (1:1.4 However, reasons for these differences need further
of the population). In the USA, children <15 years study. As the health status of the Indian population
have a much lower representation in RTI fatalities improves the age structure would become more
as compared to their ratio in the population (1:5.1) similar to that in the USA, and this would require a
but all the other age groups have a slightly higher greater focus on policies for ensuring safety of older
representation. persons on the roads.
38
 Mohan, Tiwari and Bhalla
Figure 12. Death per 100,000 population by age and sex

Of those who died, only 14% are females and the rest are males. Globally, It is not known why the
females often have a involvement rate of children
4 minority share in road deaths. However, in India, their share is among (<15 years) and the elderly (>59
the lowest in the world. This may be because of much lower exposure of years) in India is lower than that
females to traffic risk. Share of women in total number of motor vehicle in the USA when a large number
license holders in India is about 6 percent, while the rest of the 5 license of children walk, cycle and
holders are men. travel on overloaded vehicles to
school in India.
Another possible reason is lower participation rate of women in formal
employment in India compared to men (World Bank 2015a), and this
gender gap is one of the highest in the world. Figure 12 presents
number of deaths per 100,000 population by age groups and sex.

There is a wide gap in the death rate of females and males. For both sex
groups, death rates are the highest from 18 to 45 years. In India the ratio of female : male
However, as discussed above, it is likely that the low death rate of 45 fatalities in 2016 was 1:6.1 and
years or older is because of greater under-reporting of deaths for these the ratio in the USA in 2013 was
age groups (Dandona et al., 2020). According to GBD6, death rate of 1:2.4. One of the reasons why
70+ age group in India is more than twice the death rate of 15-49 years. the female fatality ratio in India
Similarly, Million Death Study for year 2005 also reported that death is lower than that in the USA
rates increased with age (Hsiao et al., 2013). Globally, road death rates could be a lower participation
often increase with age. Therefore, the U-shape of death rates in India rate in formal employment in
may be an underestimate for 45 years or older age groups. India

4 EU https://ec.europa.eu/transport/road_safety/sites/default/files/pdf/statistics/dacota/bfs2018_gender.pdf
US https://www.iihs.org/topics/fatality-statistics/detail/males-and-females
5 Licensehttps://morth.nic.in/sites/default/files/RTYB-2017-18-2018-19.pdf
6 https://vizhub.healthdata.org/gbd-compare/india

39
 Mohan, Tiwari and Bhalla
STATE WISE ANALYSIS

Figure 13. Total number of fatalities per 100,000 population by state and territory
in 2014/15 and 2018/19 for the 18 largest states and all India.

There are 36 states and union territories in India. Up to half of the

country’s road deaths are contributed by the following six states—
Uttar Pradesh, Maharashtra, Madhya Pradesh, Karnataka, Rajasthan,
and Tamil Nadu. Another 25% are contributed by the following five
states—Andhra Pradesh, Gujarat, Bihar, Telangana, and West Bengal.
To compare road death statistics over 5-year period, we used average of
2014 and 2015 (referred to as 2014/15) as the base and average of 2018
and 2019 (2018/19) as the comparator. Use of two years gives a more
stable estimate of rates and moderates the effect of an outlying year.
Among those above-mentioned states, the greatest increase in number
of deaths from 2014/15 to 2018/19 occurred in the states of Uttar
Pradesh, Madhya Pradesh, and Bihar, by an average of 30 percent. The
greatest reduction of 26% occured in the state of Tamil Nadu followed by
an average of 5% reduction in Andhra Pradesh, Gujarat, and Telangana. 7 The deaths are reported by Ministry
of Road Transport and Highways and
Overall, road deaths in India increased by 6% over this period. Next, population estimates are reported by
we present deaths per 100,000 population for the 18 largest states that the Spatial Data Repository of The DHS
represent 96% of India’s population. (7) Program.
40
 Mohan, Tiwari and Bhalla
Figure 14. Percentage change in death rates over the 5-year period from 2014/15
to 2018/19 for all states and union territories of India.

These death rates for the years 2014/15 and 2018/19 are presented in
Figure 13 along with the national averages, in the descending order
of 2018/19 death rate. Telangana and Haryana are the states with the
highest death rate (18.5 per 100,000), which is about 60% greater than
the national average (11.6 per 100,000). On the other extreme, West
Bengal and Bihar have the lowest rate (5.9 per 100,000), which is about
half the national average. In general, all the southern states have higher
death rate than the national average.
Figure 14 shows the percentage change in death rate from 2014/15 to
2018/19 and the death rate for 2018/19. The figure includes all the 36
states and union territories. The states above X-axis are those where
death rates have increased over the 5-year period, and the states below
are those where rates have decreased. In the following four states,
death rates increased by 25% or more—Bihar, Uttar Pradesh, Jharkhand
and Odisha. Increase in road deaths in these states has a significant
implication as these four states contribute more than 25% of national
road deaths. Five years ago, all these states had the lowest death rates.
Other large states with a significant increase in death rates are Assam,
Madhya Pradesh and Chhattisgarh. Note that many of the states which
witnessed increase in death rates are in north and east of India.
Most significant reduction occurred in Tamil Nadu where death rate
reduced by 30 percent. This state had the highest death rate five years
ago and contributed 10% of all road deaths. Over the 5-year period
from 2014/15 to 2018/19, its death rate has reduced from 21 to 15 per
100,000. There are now 8 more states that have higher death rates than
this state.
41
 Mohan, Tiwari and Bhalla
Death rates also reduced in West Bengal, Gujarat, Andhra Pradesh and Up to 50% of the country’s road
Rajasthan. It is interesting that rates decreased in West Bengal even deaths are contributed by the
though it already had one of the lowest rates among the large states following six states— Uttar
in India. Pradesh, Maharashtra, Madhya
Since there is no reliable information available regarding use of Pradesh, Karnataka, Rajasthan,
safety equipment (like helmets and seatbelts), enforcement of speed and Tamil Nadu.
regulations and implementation of safer road design features in
different states it is impossible to assign any scientific reasons for these
changes over time. However, it is surprising that the number of fatalities In the following four states,
in Bihar in 2018 was greater by 24% as compared to that in 2015 even death rates increased by 25%
though alcohol was banned in the state from 1 April 2016. Similarly, it or more—Bihar, Uttar Pradesh,
is not possible to find out why the fatality rates have decreased in West Jharkhand and Odisha. Increase
Bengal, Andhra Pradesh, and Tamil Nadu as no data are available on in road deaths in these states
what safety policies were responsible for these changes. has a significant implication
as these four states contribute
The states in India seem to be going through a transition. Though more than 25% of national road
there are exceptions, many states that had low levels of death rates deaths.
five years ago have recently witnessed significant rise in death rates.
While others that had high levels of death rate are becoming safer. One Most significant reduction of
likely explanation is that many states that had low levels of death rates 30% occurred in Tamil
were also among the poorest in India (e.g. Uttar Pradesh, Bihar and Nadu. This state had the highest
Jharkhand). In these states, vehicle ownership may be increasing at a death rate five years ago and
much greater rate while road safety policies are not in place to control contributed 10% of all road
traffic injuries. On the other hand, the states that had high levels of deaths. Over the 5-year period
death rates may have put in place enforcement and infrastructure from 2014/15 to 2018/19, its
measures for improving safety. The reduction in death rates may be an death rate has reduced from 21
outcome of those measures. to 15 per 100,000.

The states in India seem to be going through a transition. Though

there are exceptions, many states that had low levels of death rates
five years ago have recently witnessed significant rise in death
rates. While others that had high levels of death rate are becoming
safer. One likely explanation is that many states that had low levels
of death rates were also among the poorest in India (e.g. Uttar
Pradesh, Bihar and Jharkhand). In these states, vehicle ownership
may be increasing at a much greater rate while road safety policies
are not in place to control traffic injuries. On the other hand, the
states that had high levels of death rates may have put enforcement
and infrastructure measures in place for improving safety. The
reduction in death rates may be an outcome of those measures.

42
 Mohan, Tiwari and Bhalla
SUMMARY

• The total number of deaths in 2019 was 13 times • In Bihar, Uttar Pradesh, Jharkhand and Odisha,
greater than in 1971 with an average annual deaths per 100,000 increased by more than 25
compound growth rate (AACGR) of 6%, and the percent over the 5-year period from 2014/15 to
fatality rate in 2019 was 4.33 times greater than 2018/19. These states combined also contribute
in 1971 with an AACGR of 3.9%. one in four deaths in the country.

• The only way the decline of RTI fatalities can be • In Tamil Nadu, death rate reduced by 30 percent,
brought forward in time is to institute evidence which is the largest reduction among all the
based India-specific road safety policies that are states in India. Interestingly, Tamil Nadu had the
more effective. highest death rate in the country 5 years ago
and now there are eight more states that have
• The Indian official estimates of pedestrian greater death rate than this state.
fatalities are extremely low compared to
independent researchers’ estimates (~15% vs • Data suggests that the road deaths have
~35%), therefore, official estimates for all other increased significantly in those states that had
modes will also be wrong. low death rates 5 years ago (e.g. Bihar, Uttar
Pradesh, Jharkhand). On the other hand, deaths
• The error in the official reports regarding types have reduced in the states that had high death
of road users killed probably arises from a wrong rates earlier (Tamil Nadu, Telangana, Punjab,
coding of the victims’ status and the procedure Andhra Pradesh). The states that had low death
needs to reviewed carefully and revised. rates are also among the poorest in the country
and may be witnessing a much greater rate of
• It is not known why the involvement rate of increase in vehicle ownership. This may also
children (<18 years) and the elderly (>59 years) explain why number of deaths nationally have
in India is lower than that in the USA when a stabilised.
large number of children walk, cycle and travel
on overloaded vehicles to school in India. • Much more attention will have to be given to
Reasons for these differences need further street and highway designs and enforcement
study. Though higher level of underreporting by issues that have an influence on vulnerable road
police of deaths among older adults may partly user safety than current practice of focussing
explain this. on motor vehicles as has been the practice up
to now. This will require a great deal of research
• Telangana and Haryana have the highest death and innovation as designs and policies currently
rate in the country (18.5 per 100,000), and West being promoted do not seem to be having the
Bengal and Bihar have the lowest rate (5.9 desired effect in improving road safety.
per 100,000). This, there is a large variation in
the levels of road safety within the country. In
general, all the southern states in India have
greater death rate than national average.

43
 Mohan, Tiwari and Bhalla
URBAN SAFETY

CITY DATA

According to the MoRTH report, 49715 (33%) fatalities took place in Over the 5-year period from
urban areas and 101398 (67%) in rural areas in 2019 (Transport Research 2014/15 to 2018/19, deaths per
Wing, 2020). These data suggest that the urban RTI fatality share is 100,000 increased by 25% in 13
about the same as the estimated urban population share (34%) in 2018 out of 50 million-plus cities, and
(8). The recenttrend shows that every year the percentage of all road decreased by 25% in 11 cities.
deaths that occurred in rural areas has been increasing. In 2015, 39% of
all road deaths occurred in urban areas compared to 33% in 2019.

Within urban areas, details of fatalities and vehicles registered are

reported only for cities with populations greater than one million. The
latest report for 2019 includes details for 50 million-plus cities recording
a total of 17,202 fatalities (35% of all urban road deaths). In this chapter,
we only use total fatality data for cities from the MoRTH report (other
data are not reliable) and detailed analysis based on data reported in The five cities with the highest
published research reports. death rates are Allahabad,
Raipur, Jodhpur, Agra and
Data for 50 cities are included (population greater than 1 million) in the Jabalpur, with an average
MoRTH report published in 2019. Figure 15 shows deaths per 100,000 fatality rate of 34.3 per 100,000.
population for the 50 cities averaged over years 2018 and 2019. Data
for cities that did not have populations greater than 1 million in 2011
are not available. During the two-year period of 2018 and 2019, the
average fatality rate for all 50 million-plus cities combined was 14.5 per
100,000 persons which is 25% higher than the national average of 11.6
per 100,000 (9).

It is not possible to explain the

causes of these increases and
8 Rural population (% of total population) – India. The World Bank. decreases in the city fatality
https://data.worldbank.org/indicator/SP.RUR.TOTL.ZS?locations=IN rates as they do not have any
9 We used 2011 Census population of the million-plus cities and used district-level correlation with the size of the
population estimates reported by The DHS Program to estimate city populations
cities or their location in India.
from 2014 through 2019.
44
 Mohan, Tiwari and Bhalla

Figure 15. RTI fatality rate per 100,000 persons in million plus cities in India,
average of 2018 and 2019 (Source: Transport Research Wing 2020).

Among the 10 largest cities of India, Jaipur has the highest death rate
of 20.2 per 100,000 followed by Chennai with a rate of 13.4 per 100,000.
Rest of the eight cities have an average rate of 8 per 100,000. The five
cities with the highest death rates are Allahabad, Raipur, Jodhpur, Agra
and Jabalpur, with an average fatality rate of 34.3 per 100,000. Among
the ten cities with the highest death rates, eight are from the northern
states of Uttar Pradesh, Madhya Pradesh, and Rajasthan. The five cities
with the lowest death rates are Kolkata, Greater Mumbai, Srinagar,
Hyderabad, and Kannur.

45
 Mohan, Tiwari and Bhalla
Figure 16. Percent change in road death rates over the 5-year period from
2014/15 to 2018/19 in million-plus cities of India. X-axis shows the average
death rate over 2018/2019. Death rates increased in cities above X-axis (Source:
Transport Research Wing 2020)

Next, we discuss the changes in death rates over the 5-year period from
2014/15 to 2018/19. The percentage changes and average death rates
for 2018/19 are presented in Figure 16. Dhanbad and Thrissur cities
have the highest growth over this period with 219% and 99% increase
in death rates. In both these cities, a step change occurred from low
to high numbers, and there is a possibility that their death numbers
may have been misreported. These two cities have not been shown in
Figure 16. In half the cities, death rates increased and in the other half
they either reduced or remained the same.
46
 Mohan, Tiwari and Bhalla
These data show that compared to 2014/15, death
rates changed significantly in 2018/19 as follows:

• Increased by more than 25% in 13 cities:

Dhanbad, Thrissur, Jabalpur, Meerut, Jodhpur,
Raipur, Gwalior, Asansol, Ghaziabad, Kanpur,
Lucknow, Kota, and Chennai.

• Decreased by more than 25% in 11 cities:

Coimbatore, Patna, Pune, Tiruchirappalli,
Kolkata, Hyderabad, Greater Mumbai, Srinagar,
Vadodara, Nagpur, and Chandigarh.

A large majority of cities where deaths rates

increased significantly are in the northern states of
the country, and a large majority where road deaths
have reduced are in the western and southern
states. The increase in death rate in Chennai over
the same period was 25%, whereas the state of
Tamil Nadu recorded a decrease of 30%. It is curious
why the fatalities in Chennai did not decrease when
they showed a significant decrease in the state. It is
not possible to explain the causes of these increases
and decreases in the city fatality rates as they do not
have any correlation with the size of the cities or
their location in India.

It is not possible to explain the differences in city

fatality rates per hundred thousand persons as we
do not have details of the implementation of safety
policies in any of these cities. It is interesting to note
that none of the high rate cities include cities with
populations greater than three million, whereas the
low rate cities include five with population greater
than five million.
Since a vast majority of the victims in the cities
are vulnerable road users (see next section), one
possible cause of low death rates in low rate cities
(populations greater than 5 million) could be
reduction of vehicle speeds due to congestion. The
probability of pedestrian death is estimated at less
than 10% at impact speeds of 30 km/h and greater
than 80% at 50 km/h, and the relationship increase
in fatalities and increase in impact velocities is
governed by a power of four (Leaf and Preusser
1999, Koornstra 2007).

47
 Mohan, Tiwari and Bhalla
RTI DETAILS FOR SELECTED CITIES

Table 5. Proportion of road traffic fatalities by road user

type in nine Indian cities (Source: see text)

Table 5 shows the proportion of road traffic fatalities The proportions of pedestrian
by road user type in nine Indian cities. These cities fatalities reported in detailed
vary in population from 280 thousand to twenty research reports are very much
million. The data for Delhi was obtained from the greater than those reported
Delhi Police and for all other cities by analysing First by the MoRTH. The MoRTH
Information Reports (FIR) obtained for a period of estimates for RTI modal shares
three years from all the police stations in each city suffer from erroneous coding
(Mohan, Tiwari, and Mukherjee 2013). and should not be used.

The proportion of vulnerable road user (pedestrians,

bicyclists and motorised two-wheelers) deaths
in the nine cities range between 84% and 93%,
car occupant fatalities between 2% and 7%, and
occupants of three-wheeled scooter taxis (TSTs) less Largest proportion of fatalities
than 5% per cent, except in Vishakhapatnam where for all road user categories
the proportion for the latter is 8%. (especially vulnerable road
users) are associated with
The total of vulnerable road user deaths remains impacts with buses and trucks
relatively stable across cities of different sizes and and then cars.
the proportion of pedestrian deaths appears to be
higher in cities with larger population.

An interesting feature
emerging from this analysis
is the involvement of MTW
as impacting vehicles for
pedestrian, bicyclist and MTW
fatalities in cities.

48
 Mohan, Tiwari and Bhalla
Figure 17. Fatal RTI road user category and impacting
vehicles / objects in Vishakhapatnam and Bhopal (Numbers
in bars represent number of cases; TST: three-wheeled
scooter taxis).

RTI victims and impacting vehicles

Figure 17 shows the data for the distribution of road

traffic fatalities by road user category versus the
respective impacting vehicles/objects for two of
the nine cities, Vishakhapatnam and Bhopal. These
two cities are representative of the patterns in all
the cities studied and have been selected as the
fatality rates per 100,000 persons are different with
Vishakhapatnam at 24 and Bhopal at 14 in 2011.

In both the cities the largest proportion of fatalities

for all road user categories (especially vulnerable
road users) are associated with impacts with buses
and trucks and then cars. This is true for the other
cities also. The most interesting feature emerging
from this analysis is the involvement of MTW as
impacting vehicles for pedestrian, bicyclist and
MTW fatalities in cities.

The proportion of pedestrian fatalities associated

with MTW impacts ranges from 8 to 25 per cent
of the total. The highest proportion was observed
in Bhopal. The involvement of MTWs as impacting
vehicles in VRU fatalities may be due to the fact that
pedestrians and bicyclists do not have adequate
facilities on the arterial roads of these cities and that
they have to share the road space (the curb side
lane) with MTW riders.

49
 Mohan, Tiwari and Bhalla
Figure 19. Fatalities by road user category and time of day in Ludhiana and Agra

Road traffic fatalities by type of road user and time of crash

Figure 19 shows the fatalities by road user category and time of day in
Agra and Ludhiana. These two cities have been selected as they have
different fatality rates and as their traffic characteristics were studied in
greater details in these two cities. Pedestrian and bicycle fatalities have
high rates earlier in the morning. This may be because this class of road
users start for work earlier than those using motorised transport and
vehicle speeds may be higher at this time. The total fatality rate remains
somewhat similar between the hours of 10:00 and 18:00 and a strong
bimodal distribution is not observed. This could be because school and
work timings are reasonably staggered. Schools start around 08:00 in
the morning and close at 14:00 and some of them have a second shift.
Private offices open between 08:00-09:00, government offices between
09:00-10:00 and shops around 11:00. Most shops stay open up to 21:00
and restaurants up to 23:00.

The data also show that MTW and pedestrian deaths are relatively
high at 20:00-23:00 when we would expect traffic volumes to be low.
The details of risk factors for high rate of vulnerable road user fatalities
at night are not available for all cities but surveys done in Agra and
Ludhiana suggest that due to lower volumes vehicle velocities can be
higher at night, adequate street lighting is not present, and there is
very limited checking of drivers under the influence of alcohol (Malhan,
A., 2014). The situation would be similar in the other four cities except
in Vadodara where there is prohibition of alcohol use by law.

ROAD USER RISK ANALYSIS

Risk of fatality has been calculated using different indices to understand

the role of different motor vehicles, personal risk per trip by different
modes and the risk different vehicles present to society.

50
 Mohan, Tiwari and Bhalla
Figure 18. Motor vehicle occupant fatalities per 100,000 vehicles.

Occupant risk per hundred thousand vehicles

Figure 18 shows the number of motor vehicle occupant fatalities per

100,000 vehicles for four cities where the vehicle data were relatively
reliable. This has been obtained by dividing the total number of
occupant fatalities for each vehicle type estimated for 2011 divided by
the number of vehicles of that type estimated for the city (corrected
for overestimates). These data show that occupant fatalities per vehicle
decrease in the following order – TST:MTW:Car. Occupant fatality rates
for MTW and TST occupants are 2-3 and 3-5 times higher than that for
cars respectively.

The high rates per vehicle for TSTs would also be because they carry a
much larger number of passengers in the day as compared to MTWs and
cars. The MTW fatality rate is not more than 5 times the fatality rate for
cars in any of the four cities. For Europe and USA this ratio is reported to
be in the range of 10-20 (Peden, M. et al., 2004). We do not have detailed
data to explain with certainty why this risk ratio for MTW riders should
be lower in Indian cities where the helmet law is not being enforced.
The possible reason could be that the majority of motorcycles sold are
of low power (<150 cc), the riders are not motorcycling enthusiasts but
regular commuters, and also the effect of safety in numbers (Bhalla, K.
and Mohan, D., 2015).

51
 Mohan, Tiwari and Bhalla
Figure 20. Occupant fatality rates per 10 million trips.

Personal fatality risk per 10 million trips Owing to lower volumes, vehicle
velocities can be higher at night,
The personal fatality risk has been calculated by adequate street lighting is not
dividing the vehicle specific occupant fatality rate by present, and there is very limited
estimates of average number of occupants carried checking of drivers under the
by each vehicle per day. The numbers assumed are influence of alcohol. This may
(based on 3 trips per day for MTW and cars with be the cause of high crash rates
occupancy of 1.3 and 2.3 per trip respectively): MTW at night.
– 4, TST – 60, Car – 7 (Mohan and Roy 2003, Wilbur
Smith Associates 2008, Chanchani and Rajkotia
2012). The results of these calculations are shown
in Figure 20. It is clear that given the present trip
lengths for each vehicle type, the MTW rider is 3-6
times more at risk than a car occupant.

The MTW fatality rates per trip in Agra and A MTW rider is 3-6 times more at
Vishakhapatnam are much higher than the other risk than a car occupant and it is
three cities. The reasons for this are not known at very important to focus on their
present. At a personal level, risk per trip seems to safety (helmet use and daytime
be lowest for TST occupants in all the cities for the running lights).
assumed occupancy rates and number of trips per
day.

52
 Mohan, Tiwari and Bhalla
Figure 21. All fatalities associated with each vehicle category per 100,000 vehicle
km (estimated).

Fatalities associated with each vehicle type accounting for exposure

Figure 21 shows all the fatalities that each vehicle type is associated
with per 100,000-vehicle km per day. The following values have been
assumed for distances travelled per day.

• Car: 50 km
• TST: 150 km
• MTW: 25 km

This includes occupant fatalities and those of road users other than the
vehicle occupant. For example, if a motorcycle hits a pedestrian and the
pedestrian dies, then the pedestrian death will also be associated with
the motorcycle. This index gives a rough idea of the total number of
fatalities that is expected for each vehicle type given the present traffic
conditions and mode shares. These figures indicate that the relative
low rate for TSTs as compared to cars is due to the higher exposure
of TSTs per day. These indices appear to indicate that per km of travel
TSTs, MTWs and cars are very roughly equally harmful for society under
present conditions. Out of these three vehicles motorcycle riders bear
the highest risk and it is very important to focus on their safety (helmet
use and daytime running lights). TSTs need improvement for safety of
occupants as well as the VRUs it impacts.

53
 Mohan, Tiwari and Bhalla
Conclusions from detailed city studies

The total number of vulnerable road user deaths in the six medium sized
cities range between 84% and 93%, car occupant fatalities between 2%
and 4%, and TST occupants less than 5%, except in Vishakhapatnam
where the proportion for the latter is 8%. These total proportions are
similar to those in the megacities Mumbai and Delhi. Helmet use by
MTW riders was not enforced in any of the smaller cities though the use
is mandated by the Motor Vehicles Act 1988 of India. The high rate of
MTW fatalities can be reduced significantly if the existing mandatory
helmet laws are enforced in all the cities and laws introduced for
compulsory daytime running lights for MTW.

The largest proportion of fatalities for all road user categories (especially
vulnerable road users) is associated with impacts with buses and trucks
and then cars in Vishakhapatnam and Bhopal. This is true for the other
four cities also. The most interesting feature emerging from this analysis
is involvement of MTW as impacting vehicles for pedestrian, bicyclist,
and MTW fatalities in all the six cities.

The proportion of pedestrian fatalities associated with MTW impacts

ranges from 8 to 25 percent of the total. The involvement of MTWs
as impacting vehicles in VRU fatalities maybe due to the fact that
pedestrians and bicyclists do not have adequate facilities on the
arterial roads of these cities and they have to share the road space (the
curb side lane) with MTW riders. Provision of separate and adequate
pedestrian and bicycle lanes in all cities is a prerequisite for RTI control.

MTW and pedestrian deaths are relatively high at 20:00-23:00 when

we would expect traffic volumes to be low. Surveys done in Agra and
Ludhiana suggest that due to lower volumes vehicle velocities can be
higher at night, adequate street lighting is not present, and there is very
limited checking of drivers under the influence of alcohol. This suggests
that traffic calming methods, better street lighting and alcohol control
would be necessary to control RTI during night time.

Involvement of young children in fatal crashes appears to be low and

the reasons for this are not clear and need to be studied. Relative
risk of occupants of MTW is the highest but not as high as in high-
income countries. However, the estimated risk to society posed by
cars as estimated from total involvement in fatal crashes seems to be
greater than that posed by motorcycles and three-wheeled scooter
taxis. Further research is necessary to determine the veracity of these
findings.

54
 Mohan, Tiwari and Bhalla
SUMMARY

• During the past 5 years (2014/15 to 2018/19), • It is not possible to explain the causes of these
the proportion of total road deaths in India that increases and decreases in the city fatality rates
occurred within urban areas has been decreasing as they do not have any correlation with the size
every year—from 39% in 2015 to 33% in 2019. of the cities or their location in India.
The possible reasons can be widening of national
highways and development of new expressways • The proportion of vulnerable road user
leading to greater number of fatalities in rural (pedestrians, bicyclists and motorised two-
areas. In urban areas, increasing congestion wheelers) deaths in the nine cities range
could be resulting in reducing incidence of road between 84% and 93%, car occupant fatalities
deaths. Traffic enforcement such as motorcycle between 2% and 7%, and occupants of three-
helmets and seatbelt are also limited to urban wheeled scooter taxis (TSTs) less than 5% per
areas. cent, except in Vishakhapatnam where the
proportion for the latter is 8%.
• Government data reports road death statistics
for 50 cities that have a population of 1 million • An interesting feature emerging from this
or greater. In 2019, 35% of all urban deaths analysis is the involvement of MTW as impacting
occurred in 50 million-plus cities. The average vehicles for pedestrian, bicyclist and MTW
fatality rate for these cities combined was 14.5 fatalities in cities. The proportion of pedestrian
per 100,000 persons which is 25% greater than fatalities associated with MTW impacts ranges
the national average of 11.6 per 100,000. from 8 to 25 per cent of the total.

• During 2018/19, five cities with the highest • MTW and pedestrian deaths are relatively high
death rates are Allahabad, Raipur, Jodhpur, at 20:00-23:00 when we would expect traffic
Agra, and Jabalpur with an average rate of 34.3 volumes to be low. Surveys done in Agra and
per 100,000. Among the 10 largest cities in Ludhiana suggest that due to lower volumes
India, Jaipur has the highest death rate of 20.2 vehicle velocities can be higher at night,
per 100,000 and Kolkata has the lowest rate of adequate street lighting is not present, and
1.9 per 100,000. there is very limited checking of drivers under
the influence of alcohol.
• The regional pattern in the changes in road
deaths is also reflected in the cities. There are 13 • Occupant fatalities per vehicle decrease in the
cities in which death rates have increased by 25 following order – TST:MTW:Car.
percent, out of which 9 are in northern states.
There are 11 cities in which death rates have • Following countermeasures need to be given
reduced by 25% and majority of those are in the priority in cities: Safe pedestrians paths and
southern states. crossing facilities, speed control by traffic
calming measures like raised pedestrian
crossings, change of road texture, rumble strips
and use of roundabouts.

55
 Mohan, Tiwari and Bhalla
DISTRICT LEVEL SAFETY

INTRODUCTION

We present analysis of road deaths for the six districts of Chhattisgarh

state. Unlike data for cities that represent only urban areas and for
highways that represent only rural areas, districts consist of both rural
and urban areas, and are representative of road death statistics at
the state level. The six districts are Balod, Bemetara, Durg, Gariaband,
Kondagaon, and Raipur. These districts represent central and southern
part of the state. We collected FIRs of road crashes for these districts
for the years 2017 to 2019 and includes a total of 2544 deaths. The
death rates across these six districts range from 11.1 to 20.3 deaths per
100,000, with an average of 15.7 across all districts. In general, these
death rates are greater than the country average. Of the total death
victims, 13% are females and the rest are males. Average age of victims
is 37 years, with 30% of them younger than 25 years and 75% younger
than 45 years.

56
 Mohan, Tiwari and Bhalla
Table 6. Proportion of road death victims across districts in Chhattisgarh (2017-2019)

Table 6 shows for each district and all districts combined the percentage
of road death victims. The data shows that motorcyclists are the largest
group of road death victims with a share of 56 to 68 percent. Pedestrians
form the second largest group with 12 to 25 percent of the road death
victims. Cyclists contribute less than 5% and car occupants are another
3% to 12% of victims. The three vulnerable road users—pedestrians,
bicyclists and motorcyclists— have a combined share of 85 percent.
Among the motorcycle victims, 73% were drivers and 27% were pillion
riders. Among the car occupants who died, 33% were drivers and 67%
were passengers.

Table 7. Proportion of striking vehicles involved in crashes where other road

user died across districts in Chhattisgarh (2017-2019)

Table 7 presents distribution of striking vehicles involved in crashes in

which another road user was killed. With an average share of 37% across
the districts, trucks/tractors are the most frequent striking vehicle. The
next largest category is ‘no other vehicle’, with an average share of 27
percent. This is a significantly large share given that in these crashes a
road user was killed with no other road user involved. This may result
from vehicles skidding, tripping or colliding into a fixed object such a
tree or median. Truck, tractor, and no-other-vehicle are involved in two-
thirds of fatal cases. Motorised two-wheelers have a larger contribution
as striking vehicles than cars (16% and 10%, respectively).
57
 Mohan, Tiwari and Bhalla
Table 8: Distribution of striking vehicles in the fatal crashes of different road users

Table 8 presents proportion of striking vehicles for different road users

who died in road crashes for all six districts combined. More than 90%
pedestrians and cyclists died in crashes with three vehicle types— truck/
tractors, MTW or cars. Among crashes in which pedestrians died, MTW
and truck/tractor are equally likely to be involved as striking vehicles.
Almost half the cyclists died in crashes with truck/tractor. Crashes with
no other vehicle involved have significant contributions in the deaths
of motor vehicle users. Up to 60% of all truck/tractor deaths, 42% of car
occupant deaths, and 26% of MTW deaths are in single-vehicle crashes.
Truck/tractors are also significant contributor in the deaths of MTW and
car occupants. Up to half of car occupants died in crashes with trucks/
tractors.

Figure 22. Time of day distribution of road deaths across six districts of Chhattisgarh

Figure 22 presents time-of-day distribution of road deaths. Greater

number of crashes occurred during night than during the day. The
crashes peak between 7 PM to 9 PM. No such peak is present during
the morning, indicating that night-time crashes are likely because of
poor visibility and greater proportion of drunk driving.
58
 Mohan, Tiwari and Bhalla
INTERCITY HIGHWAYS

INTRODUCTION

Government of India has launched a major Recent studies show 3.08

programme to expand and improve highways in crashes/km/year on six-lane
India since 2000. Seventy thousand kilometres NH-1, followed by 2.54 crashes/
of National Highways (NH) are maintained by km/year on four-lane NH-24
the National Highway Authority (NHAI). Through bypass, and 0.72 crashes/km/
the National Highway Development Programme year on two-lane NH-8.
(NHDP), NHAI is upgrading nearly 49,000 km of NH.
Twenty-four thousand km of highways have been
upgraded. Upgradation includes increasing the
number of lanes (e.g. from four to six), converting
undivided roads to divided highways, and adding A majority (68%) of those
paved shoulders to 2 lane roads. The major getting killed on highways in
motivation behind highway up gradation has been India comprise vulnerable road
improving inter-city and interstate connectivity users and this fact should be the
through capacity enhancement as well as improving guiding factor in future design
highway safety. considerations.

59
 Mohan, Tiwari and Bhalla
Figure 23. Proportion of RTI fatalities on different categories of roads and the
proportion of road length for each category (Source: Transport Research Wing,
2019)

Traffic crashes on Indian Highways

Figure 23 shows the proportion of RTI fatalities on Trucks and buses are involved
different categories of roads and the proportion of in about 70% of fatal crashes
road length for each category (Transport Research in both rural and urban areas.
Wing, 2019). Fatality rate per km of road is the This is again very different from
highest on National Highways with 47.3 deaths western countries where there
per 100 km annually (Figure 24). The relatively high are significant differences in
death rate on NH could be because they carry a rural and urban crash patterns.
significant proportion of passenger and freight
traffic. However, since details of vehicle km travelled
on various categories of highways are not available,
it is not possible to make a comparison based on Safety would be enhanced
exposure rates. mainly by separating local and
Recent research studies have reported fatal crash through traffic on different
rates (fatalities per km) for three NH (NH- 1, NH- roads, or by separating slow and
8and NH 2) as 3.08 crashes/km/year on six-lane NH- fast traffic on the same road,
1, followed by 2.54 crashes/km/year on four-lane and by providing convenient
NH-24 bypass, and 0.72 crashes/km/year on two- and safe road crossing facilities
lane NH-8 (Naqvi and Tiwari 2015). to vulnerable road users.

60
 Mohan, Tiwari and Bhalla
Figure 24. Fatalities per 100 km on different categories of roads in India in 2018
(Source: Transport Research Wing, 2019).

CRASH PATTERNS

A detailed study of 35 selected locations on highways reported traffic

crash patterns using two different methods to collect road crash data
(Tiwari, Mohan, and Gupta 2000):

1. Analysis of road accident First Information Reports (FIRs) for a period

of one year from the police stations in the area.

2. Analysis of data collected by specially trained informers for a period

of three months for a 50-km section of the highway at each location.
The informers were instructed to travel over the section every day
and collect information on accidents occurring on that stretch.

The two methods of data collection gave the following insights:

a. The data available from the police records misses out many minor
injury and single vehicle accidents.

b. The data collected by the informers missed many fatal accidents

involving pedestrians and bicyclists. This is probably because the
vehicles involved in these cases are often able to drive away because
they do not suffer much damage. As such there is no evidence left at
the crash scene and the informer may miss the case when she travels
on the stretch of the highway after a day.
61
 Mohan, Tiwari and Bhalla
Table 9. Modal share of road traffic fatalities in Mumbai, Delhi and four rural highway locations in India.

Table 10. Proportion of impacting vehicle type in fatal crashes on selected highway locations.

A more recent study investigated police reports of fatal crashes on

selected locations on 2 lane NH8, 4lane NH24, and 6lane NH1 (Tiwari
2015). The results for modal shares of those killed on these locations
are given in Table 9. In the 1998 study of highways the proportions of
motor vehicle occupants and vulnerable road users were 32 and 68 per
cent respectively, whereas the numbers for urban areas were 5%-10%
vehicle occupants and the rest were vulnerable road users. Though the
motor vehicle fatalities are higher on highways than in urban areas,
as would be expected, the differences are not as high as in western
countries.

A majority (68%) of those getting killed on highways in India comprise

vulnerable road users and this fact should be the guiding factor in
future design considerations. Data from three highway segments from
2009-2013 show a similar pattern. Pedestrian and MTW proportions are
very high except on the six-lane highway where the proportion of truck
victims is higher.

Table 10 shows the involvement of different impacting vehicles in fatal

crashes on highways. This shows that as far as vehicle involvement is
concerned the patterns are similar in urban and rural area. Trucks and
buses are involved in about 70% of fatal crashes in both rural and urban
areas. This is again very different from western countries where there
are significant differences in rural and urban crash patterns.

The above aggregate data indicate that crash patterns on rural and
urban roads are more similar than would be expected based on
western experience. This is probably because at many locations there
is high-density habitation along the highways, and this may result in
the use of many sections of the highway as urban arterial arterials.
Therefore, safety would be enhanced mainly by separating local and
through traffic on different roads, or by separating slow and fast traffic
on the same road, and by providing convenient and safe road crossing
facilities to vulnerable road users.
62
 Mohan, Tiwari and Bhalla
Table 8 shows the distribution of crash types by type of highway and Crash patterns on rural and
type of crash (Tiwari, Mohan, and Gupta 2000). The statistics for single urban roads are more similar
lane may not be representative because of the small sample size. It is than would be expected based
interesting to note that there are no major differences in the proportion on western experience. This
of overturn accidents in 2-lane and 4-lane roads. Similarly, there are no is probably because at many
major differences in the proportion of head-on collisions on different locations there is high-density
types of 2-lane roads. habitation along the highways,
and this may result in the use of
Divided 4-lane roads are justified on the basis that these would many sections of the highway
eliminate the occurrence of head-on collisions. The fact that head-on as urban arterial arterialsz
collisions are common on divided roads means that many vehicles are Crash patterns on rural and
going the wrong way on divided highways. This is probably because urban roads are more similar
tractor and other vehicles travel the wrong way when they exit from than would be expected based
roadside businesses and the cut in the median is too far away. on western experience. This
This issue needs to be considered seriously and guidelines need to be is probably because at many
developed for the placement of cuts in the median or for providing locations there is high-density
under/overpasses for vehicles at convenient locations. habitation along the highways,
and this may result in the use of
Table 7 and 8 describe the types of crashes that occurred on different many sections of the highway
types of highways in 1997-2000 and in the last five years (2010-2014). as urban arterial arterialsz
The types of crashes that occur on hill roads, where run-off crashes
dominate, are clearly different from those that occur on other types
of highways. Rear-end collisions (including collisions with parked
vehicles) are high on all types of highways including 4-lane highways.
This shows that even though more space is available on wider roads
rear-end crashes do not reduce.
Head-on collisions are common
This is probably due to poor visibility of vehicles rather than road on divided roads means that
design itself. Countermeasures would include making vehicles more many vehicles are going the
visible with the provision of reflectors and roadside lighting wherever wrong way on divided highways.
possible. Impacts with pedestrians and bicycles have a high rate on all This is probably because tractor
roads including 4-lane and six-lane divided highways. and other vehicles travel the
wrong way when they exit from
The proportion is lower on 2-lane highways with wider (2.5m) paved roadside businesses and the cut
shoulders. For these types of crashes to be reduced the following in the median is too far away.
countermeasures need to be experimented with: physical segregation
of slow and fast traffic, provision of 2.5m paved shoulders with physical
separation devices like audible & vibratory pavement markings,
provision of frequent and convenient under-passes (at the same level
as surrounding land) for pedestrians, bicycles and other non-motorised
transport, and traffic calming in semi-urban and habited areas.
Rear-end collisions (including
Collisions with fixed objects are low only on 4-lane divided highways. collisions with parked vehicles)
Provision of adequate run-off area without impediments and design of are high on all types of highways
appropriate medians are obviously very important on highways. including 4- lane highways.

63
 Mohan, Tiwari and Bhalla
OTHER STUDIES

Saija and Patel (2002) and Shrinivas (2004) analysed road traffic crash
data obtained from the police records for the state of Gujarat and Tamil
Nadu respectively at a macro level but considered national highway
data in combination with other roads. Kumar, Venkatramayya, and
Kashinath (2004) have done a study of crashes on the Dindigul-Palani
section of NH 209 and report that about 50% of the crashes involved
buses and 25% of the victims were pedestrians and that two stretches
of the highway had a higher number of crashes than other sections. A
study of crashes on NH-8 passing through Valsad District found that
crashes were increasing at a rate of 3.9% annually, rear end crashes
comprised 40% and that heavy vehicles were involved in the largest
number of cases (Saija and Patel 2002).

These studies inform us that highways have some stretches that can
be identified as being associated with a higher number of crashes
than other locations; heavy vehicles are involved in a larger number
of crashes than lighter vehicles and vulnerable road users comprise a
significant proportion of those killed on national highways. However,
none of these studies provide information on speeds, modal shares and
highway design and their association with road traffic fatalities.

Shaheem, Mohammed, and Rajeevan (2006) have published two

detailed studies on road traffic crashes on the Aluva-Cherthala and
Pallichal-Kaliyikkavila sections NH- 47 in Kerala. For the Pallichal-
Kaliyikkavila section the authors evaluate the impact of four-laning of
38.5 km of the highway on road traffic crashes. They also report that
heavy vehicles had a high involvement and pedestrians and cyclists
were 28% of the victims. The most important finding of this study is that
the fatality rate based on the volume capacity ratio is more than three
times higher on the four-lane section compared to two lane sections.
The fatality rate based on population density of the associated regions
was higher on the four-lane section compared to two lane sections and
conversion of two-lane to four-lane resulted in increase in the fatality
rate from 41-51 % on the high crash rate sections.

In summary, it is clear that crash rates on intercity roads are high and
not reducing. The construction of 4 lane divided highways (without
access control) does not seem to have reduced fatality rates and
vulnerable road users still account for a number of crashes. The mix of
slow and fast-moving vehicles on highways creates serious problems as
speed differentials can account for significant increases in crash rates.
The High incidence of fatal rear-end crashes indicates a problem of
lack of visibility and conspicuity of parked vehicles. There is clearly a
strong case for redesign of intercity roads with separation of slow and
fast modes. The needs of road users on local short distance trips will
have to be accounted for to reduce the probability of head-on crashes
due to them going the wrong way on divided highways by provision of
safe road crossings at convenient distances. Solutions for many of these
issues are not readily available and research studies are necessary for
evolution of new designs.
64
 Mohan, Tiwari and Bhalla
SUMMARY

• National Highways comprise only 2% of the • Rear end collisions (including collisions with
total length of roads in India but account for parked vehicles) are high on all types of highways
36% of the fatalities. Fatality rate per km of the including 4-lane highways. This shows that even
road is the highest on NH with 0.67 deaths per though more space is available on wider roads
km annually and this fact should be the guiding rear-end crashes do not reduce. This is probably
factor in future design considerations. due to poor visibility of vehicles rather than road
design itself. Countermeasures would include
• A majority (68%) of those getting killed on making vehicles more visible with the provision
highways in India comprise vulnerable road of reflectors and roadside lighting wherever
users. possible.

• Data from three highway segments from 2009- • Following countermeasures need to be
2013 show a similar pattern. Pedestrian and experimented with: physical segregation of
MTW proportions are very high except on six- slow and fast traffic, provision of 2.5m paved
lane highways where the proportion of truck shoulders with physical separation devices
victims is much higher. like audible & vibratory pavement markings,
provision of frequent and convenient under-
• Trucks and buses are involved in about 70 passes (at the same level as surrounding land) for
percent of fatal crashes in both rural and urban pedestrians, bicycles and other non- motorized
areas. This is again very different from western transport, and traffic calming in semi-urban and
countries where there are significant differences habited areas.
in rural and urban crash patterns.
• Safety would be enhanced mainly by separating
• On 4-lane divided roads head-on collisions local and through traffic on different roads, or
comprise 19% of the crashes. Divided 4-lane by separating slow and fast traffic on the same
roads are justified on the basis that these would road, and by providing convenient and safe road
eliminate the occurrence of head-on collisions. crossing facilities to vulnerable road users.
The fact this is not occurring means that many
vehicles are going the wrong way on divided
highways. This is probably because tractor and
other vehicles go the wrong way when they
exit from roadside businesses and the cut in the
median is too far away.

65
 Mohan, Tiwari and Bhalla
INTERNATIONAL KNOWLEDGE BASE

INTERNATIONAL KNOWLEDGE BASE FOR CONTROL OF ROAD TRAFFIC INJURIES

International road safety research has involved a large number of very

well trained professionals from a variety of disciplines over the past
four decades. Some very innovative work has resulted in a theoretical
understanding of road traffic crashes as a part of a complex interaction
of sociological, psychological, physical and technological phenomena.
The results could be exchanged and solutions transferred from one high-
income country to another because the conditions in these countries
were roughly similar. This understanding of injuries and crashes has
helped high-income countries design safer vehicles, roads and traffic
management systems. A similar effort at research, development and
innovation is needed in India and similar countries. A much larger
group of committed professionals needs to be involved in this work for
new ideas to emerge.

International cooperation in the area of road safety should focus on

exchange of scientific principles, experiences of successes and failures,
and in scientific training of a large number of professionals in India. The
scientific principles of road safety can be exchanged for the benefit
of everyone. However, the priorities in road safety policies cannot be
global in nature because of the differing patterns of traffic and crash
patterns around the world. We analyse below the risk factors and the
availability of known road safety countermeasures in the context of
concerns specific to India.

66
 Mohan, Tiwari and Bhalla
RESULTS OF SYSTEMATIC REVIEWS

Legislation and enforcement

Most attempts at enforcing road-traffic legislation periodically will not

have any lasting effects, either on road-user behaviour or on accidents.
Imposing stricter penalties (in the form of higher fines or longer prison
sentences) will not affect road-user behaviour, and imposing stricter
penalties will reduce the level of enforcement (Bjornskau and Elvik
1992).

Increased normal, stationary speed enforcement is in most cases cost-

effective. Automatic speed enforcement seems to be even more efficient.
However, there is no evidence to prove that mobile traffic enforcement
for speed control with patrol cars is cost-effective (Carlsson 1997).

The only effective way to get most motorists to use safety belts is with
good laws requiring their use and sustained enforcement. When laws
are in place, education and/or advertising can be used to inform the
public about the laws and their enforcement (O’Neill 2001).

In general, the deterrent effect of a law is determined in part by the

swiftness and visibility of the penalty for disobeying the law, but a key
factor is the perceived likelihood of being detected and sanctioned.
Laws against drinking and driving are effective when combined with
active enforcement and the support of the community (Sweedler et al.
2004, Elder et al. 2004, Koornstra 2007).

Policing methods and enforcement techniques have to be optimized

for India to be effective at much lower expenditure levels. There are no
systematic studies evaluating different techniques followed around the
world. Research needs to be done on the effectiveness of professional
driver education, driver licensing methods, and control of problem
drivers in Indian settings.

Education campaigns and driver education

Road-safety campaigns often aim to improve road-user behaviour by

increasing knowledge and by changing attitudes. There is no clearly
proved relationship between knowledge and attitudes on the one
hand and behaviour on the other (O’Neill 2001, OECD 1986). Most
highway safety educational programmes do not work. They do not
reduce motor-vehicle crash deaths and injuries (Robertson et al. 1974,
Robertson 1980, 1983). Only a few programmes have ever been shown
to work, and contrary to the view that education cannot do any harm,
some programs have been shown to make matters worse (Robertson
1980, Sandels 1975).

67
 Mohan, Tiwari and Bhalla
Driver or pedestrian education programmes by Imposing stricter penalties
themselves usually are insufficient to reduce crash (in the form of higher fines or
rates (Elvik and Vaa 2004). They may increase longer prison sentences) will
knowledge, and even induce some behaviour not affect road-user behaviour
change, but this does not seem to result in a significantly.
reduction in crash rates (Duperrex, Roberts, and
Bunn 2003, Roberts and Kwan 2003). There is,
however, no reason to waste money on general
campaigns. Campaigns should be used to put
important questions on the agenda, and campaigns
aimed at changing road-user behaviour should be
focused on clearly defined behaviours and should In general, the deterrent effect
by preference fortify other measures such as new of a law is determined in part by
legislation and/or police enforcement. the swiftness and visibility of the
penalty for disobeying the law,
The effects of campaigns using tangible incentives but a key factor is the perceived
(rewards) to promote safety-belt usage have likelihood of being detected and
been evaluated by means of a meta-analytical sanctioned.
approach. The results (weighted mean effect)
show a mean short-term increase in use rates of
12.0 percentage points; the mean long-term effect
was 9.6 percentage points (Hagenzieker, Bijleveld,
and Davidse 1997). Research first from Australia,
later from many European countries, then from
Canadian provinces, and finally from some US Use of seatbelts and airbag-
states clearly shows that the only effective way equipped cars can reduce car-
to get most motorists to use safety belts is with occupant fatalities by over 50%
good laws requiring their use. Studies show that (provided car user is seat belted)
driver education may be necessary for beginners to
learn the elementary skills for obtaining a license,
but compulsory training in schools leads to early
licensing.

There is no evidence that driver education in

schools result in a reduction in road-crash rates. On
the other hand, they may lead to increased road- Driver or pedestrian education
crash rates (Williams and O’Neill 1974, Vernick et al. programmes by themselves
1999, Mayhew and Simpson 1996). While there may usually are insufficient to reduce
be a need to train professional drivers in the use crash rates
of heavy vehicles, there is no evidence that formal
driver education should be compulsory in schools
and colleges.

Compulsory helmet use reduces bicycle-related

head and facial injuries for bicyclists of all ages
Use of daytime running lights
involved in all types of crashes, including those
on cars shows reduction in
involving motor vehicles (Thompson, Rivara,
the number of multi-vehicle
and Thompson 2003). Similar results have been
daytime crashes by about 10–
confirmed for motorcyclists (Mohan et al. 1984,
15%. Similar results have been
McKnight and McKnight 1995, National Highway
confirmed for the use of daytime
Traffic Safety 1996, American College of 2001,
running lights by motorcyclists.
Bledsoe et al. 2002, Brandt et al. 2002, Liu et al. 2003)

68
 Mohan, Tiwari and Bhalla
Vehicle factors

Vehicles conforming to EU or USA crashworthiness standards provide

significant safety benefits to occupants and the effectiveness of the
following measures have been evaluated.

Use of seatbelts and airbag-equipped cars can reduce car-occupant

fatalities by over 50% (provided the car user is seat belted). It is
estimated that air-bag deployment reduced mortality by 63%, while
lap–shoulder-belt use reduced mortality by 72%, and combined air-
bag and seatbelt use reduced mortality by more than 80% (Kent, Viano,
and Crandall 2005, Crinion, Foldvary, and Lane 1975, Parkin, MacKay,
and Framton 1993).

High-mounted rear brake lights reduce the incidence of rear-end

crashes (ETSC 1993).
A meta-analysis of 17 studies that have evaluated the effects on traffic
safety of using daytime running lights on cars shows that their use
reduces the number of multi-vehicle daytime crashes by about 10–
15% for (Elvik 1993). Similar results have been confirmed for the use of
daytime running lights by motorcyclists (Radin Umar, Mackay, and Hills
1996, Radin Umar 2006, Yuan 2000).
Improvements in vehicle crashworthiness and restraint use have
contributed to a major reduction in occupant fatality rates and are
estimated to be more than 40% in most reviews (Koornstra 2007, Elvik
and Vaa 2004, Noland 2003).
However, not enough work has been done to make vehicles safer in
impacts with vulnerable road users or on vehicles specific to Indian
conditions.

Environmental and infrastructure factors

The road environment and infrastructure must be adapted to the

limitations of the road user (Van Vliet and Schermers 2000).
Traffic-calming techniques, use of roundabouts, and provision of bicycle
facilities in urban areas provide significant safety benefits and limited-
access highways with appropriate shoulder and median designs provide
significant safety benefits on long-distance through roads (Elvik 1995,
2001, Hyden and Varhelyi 2000).Though improvements in road design
seem to have some beneficial effects on crash rates, increases in speed
and exposure can offset some of these benefits (Noland 2003, O’Neill
and Kyrychenko 2006).

Road designs that control speeds seem to be the most effective

crash control measure (Aarts and van Schagen 2006). A great deal
of additional work needs to be done on rural and urban road and
infrastructure design suitable for mixed traffic to make the environment
safer for vulnerable road users. This would require special guidelines
and standards for design of, (a) roundabouts, (b) service lanes along all
intercity highways, and (c) traffic calming on urban roads and highways
passing through settlements.

69
 Mohan, Tiwari and Bhalla
Pre-hospital care

Cochrane Reviews have concluded that (Bunn et al. 2001, Sethi et al.
2004, Kwan, Bunn, and Roberts 2004b, a):

• There is no evidence from randomized controlled trials to support

the use of early or large-volume intravenous fluid administration
in uncontrolled haemorrhage. There is uncertainty about the
effectiveness of fluid resuscitation in patients with bleeding.

• The effect of pre-hospital spinal immobilization on mortality,

neurological injury, spinal stability, and adverse effects in trauma
patients therefore remains uncertain. Because airway obstruction is
a major cause of preventable death in trauma patients, and spinal
immobilization – particularly of the cervical spine – can contribute
to airway compromise, the possibility that immobilization may
increase mortality and morbidity cannot be excluded,

• In the absence of evidence of the effectiveness of advanced life

support training for ambulance crews, a strong argument could
be made that it should not be promoted outside the context
of a properly concealed and otherwise rigorously conducted
randomized controlled trial.

• A recent study by Lerner and Moscati shows that no scientific

evidence is available for supporting the concept of the ‘golden
hour’ (Lerner and Moscati 2001). While it is desirable that we save
time, it is equally important that ambulances do not endanger the
life of others while doing so, and do not waste scarce resources in
promoting systems of dubious benefit (Becker et al. 2003).

• Since the evidence shows that advanced pre-hospital interventions

do not necessarily improve outcomes, pre-hospital care should
focus primarily on transporting victims safely to a hospital facility
where they can receive definitive medical care.

• Before we import expensive pre-hospital care systems from

high income countries, it is necessary that their effectiveness be
established.

70
 Mohan, Tiwari and Bhalla
SUMMARY

• Imposing stricter penalties (in the form of high-

er fines or longer prison sentences) will not af-
fect road-user behaviour significantly. In gener-
al, the deterrent effect of a law is determined in
part by the swiftness and visibility of the pen-
alty for disobeying the law, but a key factor is
the perceived likelihood of being detected and
sanctioned.

• Driver or pedestrian education programmes by

themselves usually are insufficient to reduce
crash rates. Only effective way to get most mo-
torists to use safety belts and motorcyclists to
wear helmets is with good laws requiring their
use and enforcement.

• Use of seatbelts and airbag-equipped cars can

reduce car-occupant fatalities by over 50% (pro-
vided the car user is seat belted).

• Use of daytime running lights on cars shows re-

duction in the number of multi- vehicle daytime
crashes by about 10–15%. Similar results have
been confirmed for the use of daytime running
lights by motorcyclists.

• Traffic-calming techniques, use of roundabouts,

and provision of bicycle facilities in urban areas
provide significant safety benefits.

• A great deal of additional work needs to be

done on rural and urban road and infrastruc-
ture design suitable for mixed traffic to make
the environment safer for vulnerable road users.
This would require special guidelines and stand-
ards for design of, (a) roundabouts, (b) service
lanes along all intercity highways, and (c) traffic
calming on urban roads and highways passing
through settlements.

71
 Mohan, Tiwari and Bhalla
WAY FORWARD

PRACTICE POINTS

Some of the policy options are outlined below.

Pedestrian and bicyclist safety

1. Reserve adequate space for non-motorized

modes on all roads where they are present.

2. Free left turns must be banned at all signalized

junctions. This will give a safe time for pedestrians
and bicyclists to cross the road.

3. Speed control in urban areas: maximum speed

limits of 40-50 km/h on arterial roads need to be
enforced by road design and police monitoring.
Maximum speeds of 30 km/h in residential areas
need to be enforced by judicious use of speed-
breakers and mini roundabouts.

4. Increasing the conspicuousness of bicycles by

fixing reflectors on all sides and wheels and painting
them yellow, white or orange.
72
 Mohan, Tiwari and Bhalla
Motorcyclist and motor vehicle safety Pre-hospital care, treatment and rehabilitation

1. Notification and enforcement of mandatory use 1. Modern knowledge regarding pre-hospital care
of helmet and daytime headlights by two- wheeler should be made widely available with training of
riders. specialists in trauma care in the hospital setting.

2. Enforcement of seatbelt use laws countrywide. 2. Pre-hospital care programmes should be

rationalized on evidence-based policies so that
3. Restricting front-seat travel in cars by children scarce resources are not wasted.
and the use of child seats has potential for reducing
injuries to child occupants. Research agenda

4. Introduction of alcohol locks. 1. Development of street designs and traffic-

calming measures that suit mixed traffic with a
Road measures high proportion of motorcycles and non-motorized
modes.
1. Traffic calming in urban areas and on rural
highways passing through towns and villages. 2. Highway design with adequate and safe facilities
for slow traffic.
2. Improvement of existing traffic circles by bringing
them in accordance with modern roundabout 3. Pedestrian impact standards for buses and trucks.
practices and substituting existing signalized
junctions with roundabouts. 4. Evaluation of policing techniques to minimize
cost and maximize effectiveness.
3. Provision of segregated bicycle lanes and
disabled-friendly pedestrian paths. 5. Effectiveness of pre-hospital care measures.

4. Mandatory road safety audit for all road building 6. Traffic calming measures for mixed traffic streams
and improvement projects. including high proportion of motorised two-
wheelers.
5. Construction of service lanes along all 4-lane
highways and expressways for use by low- speed
and non-motorised traffic.

6. Removal of raised medians on intercity highways

and replacement with steel guard rails or wire rope
barriers.

Enforcement

1. The most important enforcement issue in India

is speed control. Without this it will be difficult to
lower crash rates as a majority of the victims are
vulnerable road users.

2. The second most important measure to be taken

seriously is driving under the influence of alcohol.
30%–40% of fatal crashes in India may have alcohol
involvement.

3. Enforcement of seatbelt and helmet use.

73
 Mohan, Tiwari and Bhalla
INSTITUTIONAL ARRANGEMENTS

International experience suggests that unless a The Committee also recommended that the Board
country establishes an independent national road be given power to not only set standards but also
traffic safety agency it is almost impossible to monitor their adoption and implementation. For
promote safety in a comprehensive and scientific this purpose, the Board would empanel auditors
manner. This was stated powerfully in a report to do spot checks and audits of highways under
Reducing Traffic Injury: A Global Challenge almost design, construction or operation to ensure that
22 years ago (Trinca et al. 1988): safety standards are adhered to. If standards are
not adhered to, the Board would have powers to
“Each country should create (where one does not issue suitable directions with regard to corrective
exist) a separate traffic safety agency with sufficient measures. The Board would have similar powers
executive power and funding to enable meaningful to ensure that mechanically propelled vehicles
choices between strategy and program options. conform to safety standards set by the Board. In
Such an agency would ideally report directly to the addition, the Board would have powers to seek
main legislative/political forum or to the head of information and reports and access records and
government.” documents. Where the standards set or directions
issued by the Board have not been adhered to the
The following suggestions made by the National Board should have the power to levy penalties.
Transport Development Committee (National
Transport Development Policy Committee 2014b) The Committee recommended that a minimum
should be considered for implementation. of one per cent of the total proceeds of the cess
on diesel and petrol should be available to the
Establish National Board/Agency for Road Safety Road Safety Fund of Centre and the States as road
safety is a matter of concern not only on national
This Board must be: highways but also on the state roads, village roads
and railway level crossings. Also, at least 50 per cent
(a) Independent of the respective operational of the amount retained by the Government of India
agencies to avoid conflict of interest. by way of the share of the national highways and
the Railways should be allocated to accident- prone
(b) The CEO of the Board should be of a rank of urban conglomerations and States in addition to
Secretary to the Government of India and report their entitlement. Assistance to the States from
directly to the Minister of the concerned ministry. the National Road Safety Fund should be released
to support road safety activities provided that the
(c) The Board should be staffed by professionals States enter into agreements with the Government
who have career opportunities and working of India in respect of these activities and faithfully
conditions similar to professionals working in IITs/ implement the agreements.
CSIR laboratories.

(d) The Board should have an adequate funding

mechanism based on the turnover of that sector.

(e) The terms of reference can incorporate the

recommendations similar to those included in the
reports submitted by the Committee on Roads
Safety and Traffic Management (Committee 2007).

74
 Mohan, Tiwari and Bhalla
Manpower requirements

International experience suggests that the proposed 1. This Centre would be responsible for coding and
National Road Safety and Traffic Management recording details of all fatal traffic crashes based
Board at maturity would need at least 250-350 on case files of each crash. The State Crime Record
professionals to man the eleven departments Bureau of each state would have to send copies
envisioned in the report of the Committee. Almost of completed fatal road accident case files every
all of these professionals would have to be at week to the national centre. The Centre will have to
the post-graduate level in the different areas of staffed by specially trained data coders to transfer
expertise needed for road safety. This is essential relevant details from the case files to a fatal accident
for the following reasons: (a) the agency would recording data base. The data so collected should
need to have in- house technical expertise to keep be anonymised and made available publicly for
abreast of scientific and technical advancements in analysis.
road safety knowledge internationally. (b) Since the
Board will have the responsibility of establishing 2. Centres of excellence have to be established
safety standards, it is essential that its staff have at selected IITs/NITs which can contribute to
domain expertise for the same. (c) The Board will continuous data analysis at regional and national
be sponsoring research in various areas of road level.
safety. For establishment of research priorities and
monitoring of projects the Board would need to The national safety agency can then use these data
have professionals whose expertise is similar to for statistical analysis for different policy making
those working in academic and research institutions. purposes. International experience suggests that
such departments need to employ about 50-100
National data base and statistical analysis systems statistical and epidemiology experts who design
surveys, data collection methods, perform statistical
At present very little epidemiological information is analyses and publish reports. It is equally important
available in India for deaths and injuries associated that all such data be available in the public domain
with transport. For evolution of evidence based so that independent researchers outside the official
safety policies and strategies based on the systems agency can also perform independent analyses and
approach, it is necessary to set up reliable data studies. The functions of these departments could
collection and analysis procedures for traffic include:
accidents in consonance with international
practices at different levels. This needs a special • Collating relevant data from existing surveillance
input for establishing special agencies in all sectors systems: Census Bureau, National Sample Survey
of transport. Organization, National Crime Record Bureau,
Central Bureau of Health Intelligence, etc.
At present the road traffic crash data as reported
by the MoRTH is not detailed enough or reliable for • Sample surveys for specially identified problems
epidemiological analysis and policy making. This
can only be done if data are reported and recorded • Sample surveillance systems in identified
in systematic manner by a specialised central hospitals
agency. The first step in this direction would be for
the Ministry of Home Affairs to establish a special • Establishment of multidisciplinary accident
central department for coding and recording all investigation units in academic and research
fatal crash data in a systematic manner: institutions

1. This Centre would be responsible for coding and • Coordinating with relevant ministries and
recording details of all fatal traffic crashes based departments at the central, state and city level
on case files of each crash. The State Crime Record for collating data collected by the respective
Bureau of each state would have to send copies agencies
of completed fatal road accident case files every
week to the national centre. The Centre will have to
staffed by specially trained data coders to

75
 Mohan, Tiwari and Bhalla
Establish safety departments within operating agencies

MoRTH should have an internal safety department (at different levels)

for ensuring day to day compliance with safety standards, studying
effectiveness of existing policies and standards, conducting safety
audits, collecting relevant data, and liaison with the National Safety
Agency, etc. These departments must employ 30-60 professional with
expertise in the relevant area of safety, with 30-40 per cent of the staff
on deputation from the field.

Agencies operating under the Ministry (e.g. National Highway Authority

of India) should also establish their own departments of safety with
domain specialists. The functions of these departments would include
field audits, before and after studies, data collection from the field, and
liaison with the relevant ministry and the national safety agency.

Establish multidisciplinary safety research centres at academic

institutions

The national safety agencies in each of the transport ministries should

establish multidisciplinary safety research centres in independent
academic and research institutions. These centres would ideally
include three or more disciplines of research, and for each area of work
should be at pursued in three or more centres. This would promote
competition among centres and is likely to result in more innovation.
Safety research involves the following disciplines: relevant engineering
sciences, statistics and epidemiology, trauma and medical care,
sociology, psychology, jurisprudence, and computer science. For these
centres to be productive, each centre should have a minimum of 8-10
professionals. It is also possible that one academic institution has more
than one of these safety research centres. It is recommended that 15
such centres be established by 2020 and another 15 by 2025.

The funding for each of these centres should include:

• Endowment for three or more professorial chairs

• Endowment grant for at least two postgraduate scholarships per

endowed chair

• Establishment funds for critical laboratories

• Funds for supporting visiting professionals

• Support for surveys, software, travel

For these centres to function effectively the minimum grant per centre
per year would be in the range of Rs. 30-40 million annually including
endowment funds. The national safety agency should establish
procedures for issuing calls for proposals and for evaluating the same
under open completion. A procedure should also be established for an
academic peer evaluation of each centre every two years.

76
 Mohan, Tiwari and Bhalla
REFERENCES
Aarts, Letty, and Ingrid van Schagen. 2006. “Driving speed and the risk of road crashes: A review.” Accident Analysis & Prevention
38 (2):215-224.

Abay, Kibrom A. 2015. “Investigating the nature and impact of reporting bias in road crash data.” Transportation Research Part
A: Policy and Practice 71 (0):31-45. doi: http://dx.doi.org/10.1016/j.tra.2014.11.002.

Adams, John. 1987. “Smeed’s Law: some further thoughts.” Traffic Engineering and Control 28 (2):70-73.

American College of, Surgeons. 2001. “Statement in support of motorcycle helmet laws [ST-35].” Bulletin of the American College
of Surgeons 86 (2).

Amoros, E., J. L. Martin, S. Lafont, and B. Laumon. 2008. “Actual incidences of road casualties, and their injury severity, modelled
from police and hospital data, France.” European Journal of Public Health 18 (4):360-365.

Amoros, Emmanuelle, Jean Louis Martin, and Bernard Laumon. 2006. “Under-reporting of road crash casualties in France.”
Accident Analysis & Prevention 38 (4):627-635.

Arora, P., A. Chanana, and H. R. Tejpal. 2013. “Estimation of blood alcohol concentration in deaths due to roadside accidents.”
Journal of Forensic and Legal Medicine 20 (4):300-304. doi: DOI 10.1016/j.jflm.2012.12.003.

Becker, L. R., E. Zaloshnja, N. Levick, Guohua Li, and Ted R. Miller. 2003. “Relative risk of injury and death in ambulances and other
emergency vehicles.” Accident Analysis & Prevention 35 (6):941-948.

Bhalla, Kavi, Marc Shotten, Aaron Cohen, Michael Brauer, Saeid Shahraz, Richard Burnett, Katherine Leach-Kemon, Greg
Freedman, and Christopher J. L. Murray. 2014. Transport for health : the global burden of disease from motorized road transport.
Washington, DC: World Bank Group.

Bjornskau, Torkel, and Rune Elvik. 1992. “Can road traffic law enforcement permanently reduce the number of accidents?”
Accident Analysis & Prevention 24 (5):507-520.

Bledsoe, G. H., S. M. Schexnayder, M. J. Carey, W. N. Dobbins, W. D. Gibson, J. W. Hindman, T. Collins, B. H. Wallace, J. B. Cone, and
T. J. Ferrer. 2002. “The negative impact of the repeal of the Arkansas motorcycle helmet law.” J Trauma 53 (6):1078-1086.

Bliss, Tony, and Jeanne Breen. 2009. Implementing the recommendations of the World Report on Road Traffic Injury Prevention.
Wahington D.C.: The World

Bank Global Road Safety Facility.

Borse, NN, and Adnan Ali Hyder. 2009. “Call for more research on injury from the developing world: Results of a bibliometric
analysis.”

Brandt, M. M., K. S. Ahrns, C. A. Corpron, G. A. Franklin, and W. L. Wahl. 2002. “Hospital cost is reduced by motorcycle helmet use.”
J Trauma 53 (3):469-471.

Brüde, Ulf, and Rune Elvik. 2015. “The turning point in the number of traffic fatalities: Two hypotheses about changes in under-
lying trends.” Accident Analysis & Prevention 74:60-68. doi: http://dx.doi.org/10.1016/j.aap.2014.10.004.

Bunn, F., I. Kwan, I. Roberts, and R. Wentz. 2001. Effectiveness Of Pre-Hospital Trauma Care. On behalf of the WHO Pre-hospital
Care Steering Committee.

Cochrane Reviews.

Carlsson, G. 1997. “Cost-Effectiveness of information, campaigns and enforcement and the costs and benefits of speed changes.”
In European Seminar On Cost-Effectiveness Of Road Safety Work And Measures. Luxembourg.

Central Bureau of Health Intelligence. (2019). National health profile 2019, 14th issue. New Delhi: Ministry of Health & Family
Welfare. www.cbhidghs.gov.in.

Chanchani, Radha, and Fagun Rajkotia. 2012. A study of the autorickshaw sector in Bangalore city. Bangalore: Centre for Infra-
structure and Sustainable Transportation & Urban Planning (CiSTUP), Indian Institute of Science.

77
 Mohan, Tiwari and Bhalla
CHANDRAMOULI, C. 2012. “Houses, Household Amenities and Assets Data 2001 - 2011 - Visualizing Through Maps.” New Delhi,
India. https://censusindia.gov.in/2011-Common/NSDI/Houses_Household.pdf.

Committee. 2007. “Report of the Committee on Road Safety and Traffic Management.” Committee on Infrastructure, Planning
Commission, Government of India, New Delhi, Last Modified 20080/07/22/.

Crinion, J. D., L. A. Foldvary, and J. C. Lane. 1975. “The effect on casualties of a compulsory seat belt wearing law in South Austral-
ia.” Accident Analysis & Prevention 7 (2):81-89.

Dandona, R., Kumar, G. A., Gururaj, G., James, S., Chakma, J. K., Thakur, J. S., . . . Dandona, L. 2020. Mortality due to road injuries in
the states of India: the Global Burden of Disease Study 1990&#x2013;2017. The Lancet Public Health, 5(2), e86-e98. doi: 10.1016/
S2468-2667(19)30246-4.

Dandona, Rakhi, G Anil Kumar, Gopalkrishna Gururaj, Spencer James, Joy K Chakma, J S Thakur, Amar Srivastava, Girikumar
Kumaresh, Scott D Glenn, and Gaurav Gupta. 2020. “Mortality Due to Road Injuries in the States of India: The Global Burden of
Disease Study 1990–2017.” The Lancet Public Health 5 (2): e86–98.

Das, Ashis, Hallvard Gjerde, Saji S. Gopalan, and Per T. Normann. 2012. “Alcohol, Drugs, and Road Traffic Crashes in India: A
Systematic Review.” Traffic Injury Prevention 13 (6):544-553. doi: 10.1080/15389588.2012.663518.

Delhi Traffic Police. 2014. Road Accidents in Delhi - 2013. New Delhi: Accident Research Cell, Delhi Police.

Derriks, Harry M. , and Peter M. Mak. 2007. Underreporting of road traffic casualties. edited by Harry M. Derriks and Peter M. Mak.
Paris: International Transport Forum.

Dhanoa, K. K. 2017. “Traffic safety assessment of small cities: the cases of Patiala and Rajpur.” M.Tech., Civil Engineering, Indian
Institute of Techology Delhi.

Duperrex, O., I. Roberts, and F. Bunn. 2003. Safety education of pedestrians for injury prevention (Cochrane Review). Oxford:
Update Software.

Elder, R. W., R. A. Shults, D. A. Sleet, J. L. Nichols, R. S. Thompson, and W. Rajab. 2004. “Effectiveness of mass media campaigns for
reducing drinking and driving and alcohol-involved crashes: a systematic review.” Am J Prev Med 27 (1):57-65.

Elvik, R., and Truls Vaa. 2004. The handbook of road safety measures. Amsterdam: Elsevier. Reprint, IN FILE.

Elvik, Rune. 1993. “The effects on accidents of compulsory use of daytime running lights for cars in Norway.” Accident Analysis
& Prevention 25 (4):383-398.

Elvik, Rune. 1995. “The safety value of guardrails and crash cushions: a meta-analysis of evidence from evaluation studies.”
Accident Analysis & Prevention 27 (4):523-549.

Elvik, Rune. 2001. “Area-wide urban traffic calming schemes: a meta-analysis of safety effects.” Accident Analysis & Prevention
33 (3):327-336.

Esser, Marissa B., Shirin Wadhwaniya, Shivam Gupta, Shailaja Tetali, Gopalkrishna Gururaj, Kent A. Stevens, and Adnan A. Hyder.
2015. “Characteristics associated with alcohol consumption among emergency department patients presenting with road
traffic injuries in Hyderabad, India.” Injury. doi: http://dx.doi.org/10.1016/j.injury.2015.07.022.

ETSC. 1993. Reducing Traffic Injuries Through Vehicle Safety Improvements. Brussels: European Transport Safety Council.

Evans, L. 1991. Traffic Safety and the Driver New York: Van Nostrand Reinhold. Reprint, IN FILE.

Expert Committee on Auto Fuel Policy. 2002. Urban road traffic and air pollution in major cities: Volume 1. New Delhi: Government
of India.

Giuliano, Genevieve, and Dhiraj Narayan. 2003. “Another Look at Travel Patterns and Urban Form: The US and Great Britain.”
Urban Studies 40 (11):2295-2312. doi: 10.1080/0042098032000123303.

Goel, Rahul, Sarath K. Guttikunda, Dinesh Mohan, and Geetam Tiwari. 2015. “Benchmarking vehicle and passenger travel
characteristics in Delhi for on-road emissions analysis.” Travel Behaviour and Society 2 (2):88-101. doi: http://dx.doi.org/10.1016/j.
tbs.2014.10.001.

Gururaj, G. 2006. Road traffic injury prevention in India. Bangalore: NIMHANS.

78
 Mohan, Tiwari and Bhalla
Hagenzieker, M. P., F. D. Bijleveld, and R. J. Davidse. 1997. “Effects of incentive programs to stimulate safety belt use: a meta-
analysis.” Accident Analysis and Prevention 29 (6):759-777.

Hsiao, M., A. Malhotra, J. S. Thakur, J. K. Sheth, A. B. Nathens, N. Dhingra, and P. Jha. 2013. “Road traffic injury mortality and
its mechanisms in India: nationally representative mortality survey of 1.1 million homes.” BMJ Open 3 (8):1-9. doi: 10.1136/
bmjopen-2013-002621.

Hupkes, Geurt. 1982. “The law of constant travel time and trip-rates.” Futures 14 (1):38-46. doi: http://dx.doi.org/10.1016/0016-
3287(82)90070-2.

Hyden, Christer, and Andras Varhelyi. 2000. “The effects on safety, time consumption and environment of large scale use of
roundabouts in an urban area: a case study.” Accident Analysis & Prevention 32 (1):11-23.

IHME. 2020. Global Burden of Disease Collaborative Network. Global Burden of Disease Study 2019 (GBD 2019) Results. Seattle,
United States: Institute for Health Metrics and Evaluation (IHME), 2020. Available from http://ghdx.healthdata.org/gbd-results-
tool. Accessed 2021-01-11

IIHS. 2021. “Fatality Facts 2019 Males and Females.” Insurance Institute for Highway Safety. 2021. https://www.iihs.org/topics/
fatality-statistics/detail/males-and-females.

IIPS. 2021. “National Family Health Survey, India.” 2021. http://rchiips.org/nfhs/about.shtml.

Indian Council of Medical Research, Public Health Foundation of India, and Institute for Health Metrics and Evaluation. GBD India
Compare Data Visualization. New Delhi: ICMR, PHFI, and IHME; 2017. Available from http://vizhub.healthdata.org/gbd-compare/
india (Accessed [INSERT DATE])

Indian Council of Medical Research, Public Health

Foundation of India, & Institute for Health Metrics and Evaluation (IHME). 2017. India: Health of the Nation’s States — The
India State-Level Disease Burden Initiative. New Delhi: Public Health Foundation of Indiahttps://phfi.org/wp-content/
uploads/2018/05/2017-India-State-Level-Disease-Burden-Initiative-Full-Report.pdf

Institute for Health Metrics and Evaluation (IHME). 2018. India Profile. Seattle, WA: IHME, University of Washington. Available
from http://www.healthdata.org/India. Accessed 20 April 2020.

International Institute for Population Sciences (IIPS), National Programme for Health Care of Elderly (NPHCE), MoHFW, Harvard T.
H. Chan School of Public Health (HSPH) and the University of Southern California (USC) 2020. Longitudinal Ageing Study in India
(LASI) Wave 1, 2017-18, India Report, International Institute for Population Sciences, Mumbai.

International Traffic Safety Data and Analysis Group. 2011. Reporting on serious road traffic casualties. Paris: International
Transport Forum/OECD.

International Traffic Safety Data and Analysis Group. 2015. Why does road safety improve when economic times are hard? Paris:
OECD/ITF.

IRTAD. 2014. IRTAD 2014 Annual Report. Paris: OECD/International Transport Forum.
Kent, Richard, David C. Viano, and Jeff Crandall. 2005. “The Field Performance of Frontal Air Bags: A Review of the Literature.”
Traffic Injury Prevention 6 (1):1-23.

Knoflacher, Hermann. 2007. “From myth to science.” Seminar 579:40-44.

Koornstra, Matthijs. 2007. “Prediction of traffic fatalities and prospects for mobility becoming sustainable-safe.” Sadhna -
Academy Proceedings in Engineering Sciences 32 (4):365-396.

Kopits, Elizabeth, and Maureen Cropper. 2005. “Traffic fatalities and economic growth.” Accident Analysis & Prevention 37
(1):169-178.

Kumar, R. P., V. Venkatramayya, and T. Kashinath. 2004. “Macro level study of road accidents on Dindigul-Palani section of NH-
209.” Indian Highways 32 (11):31-38.

Kwan, I., F. Bunn, and I. Roberts. 2004a. Spinal immobilisation for trauma patients (Cochrane Review) on behalf of the WHO Pre-
Hospital Trauma Care Steering Committee. Chichester, UK: John Wiley & Sons, Ltd.

79
 Mohan, Tiwari and Bhalla
Kwan, I., F. Bunn, and I. Roberts. 2004b. Timing and volume of fluid administration for patients with bleeding (Cochrane Review).
On behalf of the WHO Pre-Hospital Trauma Care Steering Committee. Chichester, UK: John Wiley & Sons, Ltd.
Laiou, Alexandra, Katerina Folla, George Yannis, Robert Bauer, Klaus Machata, Christian Brandstaetter, Pete Thomas, and Alan

Kirk. 2016. “856 Comparative Analysis of Road Accidents by Gender in Europe.” Injury Prevention 22 (Suppl 2): A305 LP-A305.
https://doi.org/10.1136/injuryprev-2016-042156.856.

Leaf, W. A., and D. F. Preusser. 1999. Literature Review on Vehicle Travel Speeds and Pedestrian Injuries. Washington, D.C.: National
Highway Traffic Safety Administration,U.S. Department of Transportation.

Lerner, E. Brooke, and Ronald M. Moscati. 2001. “The golden hour: Scientific fact or medical “urban legend”?” Academic
Emergency Medicine 8 (7):758-760.

Liu, B., R. Ivers, R. Norton, S. Blows, and S. K. Lo. 2003. Helmets for preventing injury in motorcycle riders. Cochrane Database of
Systematic Reviews 2003.

Mani, Akshay, and Anirudh Tagat. 2013. Safety assessment of auto-rickshaws in Mumbai. Mumbai: Embarq India.

Martinez, R. 1996. “Traffic safety as a health issue.” In Traffic Safety, Communication and Health, edited by H. von Holst, A. Nygren
and R. Thord. Stockholm: Temaplan AB.

Mayhew, D. R., and H. M. Simpson. 1996. Effectiveness and role of driver education and training in a graduated licensing system.
Ottawa, Ontario: Traffic Injury Research Foundation.

McKnight, A. James, and A. Scott McKnight. 1995. “The effects of motorcycle helmets upon seeing and hearing.” Accident
Analysis & Prevention 27 (4):493-501.

Menon, G. R., Singh, L., Sharma, P., Yadav, P., Sharma, S., Kalaskar, S., . . . Kulothungan, V. (2019). National Burden Estimates of
healthy life lost in India, 2017: an analysis using direct mortality data and indirect disability data. The Lancet Global Health, 7(12),
e1675-e1684

Ministry of Road Transport and Highways. 1988. Motor Vehicles Act, 1988. New Delhi: Ministry of Road Transport and Highways,
Government of India.

Mishra, B. K., A. K. Banerji, and Dinesh Mohan. 1984. “Two-wheeler injuries in Delhi, India: A study of crash victims hospitalized
in a neuro-surgery ward.” Accident Analysis & Prevention 16 (5-6):407-416.

Mohan, D., and Dunu Roy. 2003. “Operating on three wheels.” Economic & Political Weekly XXXVIII (03):177-180. Mohan, D., and
G. Tiwari. 2000. “Road Safety in Less Motorised Countries - Relevance of International Vehicle and Highway Safety Standards.” In
Proceedings International Conference on Vehicle Safety, 155-166. London: Institution of Mechanical Engineers.

Mohan, D., G. Tiwari, and K. Mukherjee. 2013. A study on community design for traffic safety. In Final report for IATSS. New Delhi:
Transportation Research and Injury Prevention Programme, Indian Institute of Technology Delhi.

Mohan, Dinesh, Geetam Tiwari, M. Khayesi, and F. M. Nafukho. 2006. Road traffic injury prevention training manual. Geneva:
World Health Organization. Reprint, IN FILE.

Mohan, Dinesh, K. P. Kothiyal, B. K. Misra, and A. K. Banerji. 1984. “Helmet and Head Injury Study of Crash Involved Motorcyclists
in Delhi.” In Proceedings 1984 International Conference on the Biomechanics of Impacts, 65-77. Bron, France: IRCOBI.

Mohan, Dinesh, O. Tsimhoni, Michael Sivak, and Michael J. Flannagan. 2009. Road safety in India: Challenges and opportunities.
Ann Arbor, MI: University of Michigan Transportation Research Institute.

Mohan, Dinesh, Rahul Goel, Sarath Guttikunda, and Geetam Tiwari. 2014. Assessment of motor vehicle use characteristics in
three Indian cities. Lyngby, Denmark: UNEP Risø Centre on Energy.

MoRTH. 2015. Annual Report 2014-2015. New Delhi: Ministry of Road Transport & Highways.

MORTH. 2020. “Road Accidents in India 2019.” New Delhi. https://morth.nic.in/sites/default/files/RA_Uploading.pdf.

Naqvi, H. M., and G. Tiwari. 2015. “Assessing risk factors for pedestrian fatal crashes on National Highways in India.” EASTS, Cebu,
Philippines.

80
 Mohan, Tiwari and Bhalla
National Center for Statistics and Analysis. 2015. Traffic Safety Facts 2013. Wahington D.C.: National Highway Traffic Safety
Administration.

National Highway Traffic Safety, Administration. 1996. Do Motorcycle Helmets Interfere With the Vision and Hearing of Riders?
Washington, D.C.: U.S. Department of Transportation.

National Transport Development Policy Committee. 2014a. India Transport Report: Moving India to 2032. 3 vols. New Delhi:
Rutledge.

National Transport Development Policy Committee. 2014b. “Safety.” In India Transport Report: Moving India to 2032, 574-616.
New Delhi: Routledge.

NCRB. 2015. Accidental Deaths and Suicides in India 2014. New Delhi: National Crime Records Bureau, Ministry of Home Affairs.

Noland, Robert B. 2003. “Traffic fatalities and injuries: the effect of changes in infrastructure and other trends.” Accident Analysis
& Prevention 35 (4):599-612.

OECD. 1986. OECD road safety research: a synthesis. Paris: OECD.

OECD/ITF. 2015. “Why Does Road Safety Improve When Economic Times Are Hard?” https://www.itf-oecd.org/sites/default/files/
docs/15irtadeconomictimes.pdf.

Office of The Registrar General & Census Commissioner. 2015. C-14 five year age group data by residence and sex 2011. New
Delhi: Ministry of Home Affairs, Government of India.

O’Neill, B. 2001. “Role of Advocacy, Education, and Training in Reducing Motor Vehicle Crash Losses.” In Proceedings from WHO
meeting to Develop a 5-Year Strategy on Road Traffic Injury Prevention. Geneva: WHO.

O’Neill, B., and Sergey Kyrychenko. 2006. Use and Misuse of Motor Vehicle Crash Death Rates in Assessing Highway Safety Per-
formance. Arlington, VA: Insurance Institute for Highway Safety.

Parkin, S., G. M. MacKay, and R. I. Framton. 1993. “Effectiveness and Limitations of Current Seat Belts in Europe.” Chronic Diseases
in Canada 14 (4 Supplement):S53-S59.

Peden, Margie, Richard Scurfield, David Sleet, Dinesh Mohan, Adnan A. Hyder, Eva Jarawan, and C. Mathers. 2004. World report
on road traffic injury prevention. Geneva World Health Organization Reprint, NOT IN FILE.

PPAC. 2013. “All India Study on Sectoral Demand of Diesel and Petrol, Petroleum Planning and Analysis Cell.” New Delhi, India.
https://www.ppac.gov.in/WriteReadData/Reports/201411110329450069740AllIndiaStudyonSectoralDemandofDiesel.pdf.

Radin Umar, R. 2006. “Motorcycle safety programmes in Malaysia: how effective are they?” International Journal of Injury Control
and Safety Promotion 13 (2):71-79.

Radin Umar, R. S., Murray G. Mackay, and Brian L. Hills. 1996. “Modelling of conspicuity-related motorcycle accidents in Seremban
and Shah Alam, Malaysia.” Accident Analysis & Prevention 28 (3):325-332.

Roberts, Ian, and Irene Kwan. 2003. School based driver education for the prevention of traffic crashes (Cochrane Review).
Oxford: Update Software.

Robertson, L. S. 1980. “Crash involvement of teenaged drivers when driver education is eliminated from high school.” American
Journal of Public Health 70 (6):599-603. Robertson, L. S. 1983. Injuries: Causes, Control Strategies and Public Policy. Lexington,
MA: Lexington Books. Reprint, IN FILE.

Robertson, L. S., A. B. Kelley, B. O’Neill, C. W. Wixom, R. S. Eiswirth, and W. Haddon, Jr. 1974. “A controlled study of the effect of
television messages on safety belt use.” American Journal of Public Health 64 (11):1071-1080.

Rosman, Diana L., and Matthew W. Knuiman. 1994. “A comparison of hospital and police road injury data.” Accident Analysis &
Prevention 26 (2):215-222.

Saija, K. K., and C. D. Patel. 2002. “Micro level study of accidents on NH-8 passing through Valsad District.” Indian Highways 30
(8):43-51.

Sandels, S. 1975. Children in traffic. Surrey: Elek Books Ltd. Reprint, IN FILE.

81
 Mohan, Tiwari and Bhalla
Santos, A. , N. McGuckin, H.Y. Nakamoto, D. and Gray, and S. Liss. 2011. Summary of travel trends: 2009 national household
travel survey. Washington D.C.: U.S. Department of Transportation, Federal Highway Administration.

Sethi, D., I. Kwan, A. M. Kelly, I. Roberts, and F. Bunn. 2004. Advanced trauma life support training for ambulance crews (Cochrane
Review). On behalf of the WHO Pre-Hospital Trauma Care Steering Committee. Chichester, UK: John Wiley & Sons, Ltd.

Shaheem, S., K. M. S. Mohammed, and Rajeevan. 2006. “Evaluation of cost effectiveness of improvements of accident prone
locations on NH-47 in Kerala state.” Indian Highways 34 (8):35-46.

Shrinivas, P. L. L. 2004. “Studies undertaken to identify critical causes of accidents in the highways of Tamil Nadu.” Indian
Highways 31 (8):11-22.

Sweedler, B., M. Biecheler, H. Laurell, G. Kroj, M. Lerner, M. Mathijssen, D. Mayhew, and R. Tunbridge. 2004. “Worldwide trends in
alcohol and drug impaired driving.” Traffic Injury Prevention 5 (3):175-184.

Thompson, D. C., F. P. Rivara, and R. Thompson. 2003. Helmets for preventing head and facial injuries in bicyclists. Oxford: Update
Software.

Tiwari, G. 2015. Highway safety in India. New Delhi: Indian Institute of Technology Delhi.

Tiwari, G., D. Mohan, and D. P. Gupta. 2000. Evaluation of Capacity Augmentation Projects of National Highways and State
Highways. New Delhi: Ministry of Surface Transport, GOI.

Transport for London. 2011. Travel in London, supplementary report: London travel demand survey (LTDS). London: Transport for
London. Transport Research Wing. 2014. Road Transport Yearbook 2011-12. New Delhi: Ministry of Road Transport & Highways,
Government of India.

Transport Research Wing. 2019. Road accidents in India – 2018. New Delhi: Ministry of Road Transport & Highways, Government
of India.

Trinca, G. W., I. R. Johnston, B. J. Campbell, F. A. Haight, P. R. Knight, G. M. MacKay, A. J. McLean, and E. Petrucelli. 1988. Reducing
Traffic Injury - A Global Challenge. Melbourne: Royal Australasian College of Surgeons. Reprint, IN FILE.

TRW. 2021. “Road Transport Year Book (2017 - 2018 & 2018 - 2019).” New Delhi, India. https://morth.nic.in/sites/default/files/
RTYB-2017-18-2018-19.pdf.

Van Vliet, P., and G. Schermers. 2000. Sustainable Safety: A New Approach for Road Safety in the Netherlands. Rotterdam: Traffic
Research Centre, Ministry of Transport.

Varghese, M., and D. Mohan. 1991. “Transportation Injuries in Rural Haryana, North India.” In Proceedings International Conference
on Traffic Safety, 326-329. New Delhi: Macmillan India Ltd.

Vernick, J.S, L. Guohua, S. Ogaitis, E. J. MacKenzie, S. P. Baker, and A. C. Gielen. 1999. “Effects of high school driver education on
motor vehicle crashes, violations, and licensure.” American Journal of Preventive Medicine 16:40-46.

W.H.O. 2015. Global status report on road safety 2015. Geneva: World Health Organization, .

Wilbur Smith Associates. 2008. Study on traffic and transportation policies and strategies in urban areas in India - Final report.
New Delhi: Ministry of Urban Development, Government of India.

Williams, Allan F., and Brian O’Neill. 1974. “On-the-road driving records of licensed race drivers.” Accident Analysis & Prevention
6 (3-4):263-270.

World Bank. 2015a. Labor force participation rate, female (% of female population ages 15+) (modeled ILO estimate). Washington
D.C.: The World Bank Group.

World Bank. 2015b. Urban population (% of total). Washington D.C.: The World Bank Group.

Yuan, Wu. 2000. “The effectiveness of the ‘ride-bright’ legislation for motorcycles in Singapore.” Accident Analysis & Prevention
32 (4):559-563.

Zegras, Christopher. 2010. “The Built Environment and Motor Vehicle Ownership and Use: Evidence from Santiago de Chile.”
Urban Studies 47 (8):1793-1817. doi: 10.1177/0042098009356125.

82
 Mohan, Tiwari and Bhalla
Appendix 1. List of research studies on road safety in India in 2016

Aarthy, D. K., S. Vandanaa, M. Varshini, and K. Tijitha. 2016. “Automatic Identification of Traffic Violations and Theft Avoidance.”. doi:10.1109/
ICONSTEM.2016.7560926.

Adnan, M. and M. S. Ali. 2016. “An Effective Methodology for Road Accident Data Collection in Developing Countries.” In International Business:
Concepts, Methodologies, Tools, and Applications, 462-474. doi:10.4018/978-1-4666-9814- 7.ch022. https://www.scopus.com/inward/record.
uri?eid=2-s2.0- 84969962384&doi=10.4018%2f978-1-4666-9814- 7.ch022&partnerID=40&md5=240e78c8c78d05bd7e9b720e805c0721.

Alam, K. and A. Mahal. 2016. “The Economic Burden of Road Traffic Injuries on Households in South Asia.” PLoS ONE 11 (10). doi:10.1371/jour-
nal.pone.0164362.

Anne Frank Joe, A., S. Celin, R. Thomas, and B. Vishwanath. 2016. “A Prototype Airbag Safety Device to Prevent Accidental Injuries for Bike Rid-
ers.” International Journal of Pharmacy and Technology 8 (2): 13501-13505.

Babu, A., A. Rattan, P. Ranjan, M. Singhal, A. Gupta, S. Kumar, B. Mishra, and S. Sagar. 2016. “Are Falls More Common than Road Traffic Accidents
in Pediatric Trauma? Experience from a Level 1 Trauma Centre in New Delhi, India.” Chinese Journal of Traumatology - English Edition 19 (2):
75-78. doi:10.1016/j.cjtee.2015.10.004.

Benjula Anbu Malar, M. B. and H. Yukesh. 2016. “Pace Control in Motor-Cycle Vehicle at Special Zones using IOT.” International Journal of Phar-
macy and Technology 8 (4): 25412- 25418.

Bhandari, R., B. Raman, and V. Padmanabhan. 2016. “Poster: Improving Road Safety through Smart-Sensing.”. doi:10.1145/2938559.2948797.

Bhoi, S., P. R. Mishra, K. D. Soni, U. Baitha, and T. P. Sinha. 2016. “Epidemiology of Traumatic Cardiac Arrest in Patients Presenting to Emergency
Department at a Level 1 Trauma Center.” Indian Journal of Critical Care Medicine 20 (8): 469-472. doi:10.4103/0972- 5229.188198.

Bhowate, S., N. Sheikh, and S. Asawa. 2016. “Patterns of Cranio-Cerebral Injuries in Fatal Head Trauma.” Indian Journal of Forensic Medicine and
Toxicology 10 (2): 17-22. doi:10.5958/0973-9130.2016.00054.2.

Bollapragada, R., S. Poduval, C. Bingi S, and B.

Brahmbhatt. 2016. “Solving Traffic Problems in the State of Kerala, India: Forecasting, Regression and Simulation Models.” Vikalpa 41 (4): 325-
343. doi:10.1177/0256090916675532.

Chandrasekharan, A., A. J. Nanavati, S. Prabhakar, and S. Prabhakar. 2016. “Factors Impacting Mortality in the Pre-Hospital Period After Road
Traffic Accidents in Urban India.” Trauma Monthly 21 (3). doi:10.5812/traumamon.22456.

Choudhury, R. and N. Singh. 2016. “Medico-Legal Injury Patterns Associated with Geriatric Patients in a Rural Medical Institute of India.” Medico-
Legal Update 16 (2): 131-135. doi:10.5958/0974-1283.2016.00074.8.

Dash, A., J. N. Senapati, B. C. Raulo, P. K. Brahma, and M. C. Sahu. 2016. “Prevalence of Trauma Cases in a Tertiary Care Teaching Hospital.”
International Journal of Pharmaceutical Sciences Review and Research 36 (1): 153-157.

Garg, A., C. Behera, S. Chopra, and D. N. Bhardwaj. 2016. “Mortality among Homeless Women Who Remain Unclaimed After Death: An Insight.”
National Medical Journal of India 29 (4): 207-208.

Garg, A., C. Behera, S. Chopra, R. Swain, and S. K. Gupta. 2016. “A Study of Unclaimed Deaths in Children at an Indian Tertiary Hospital.”Journal
of South India Medicolegal Association 8 (2): 81-84.

Gong, Y., W. Zhang, Z. Zhang, and Y. Li. 2016. Research and Implementation of Traffic Sign Recognition System. Lecture Notes in Electrical
Engineering. Vol. 348. doi:10.1007/978- 81-322-2580-5_50.

Grimm, M. and C. Treibich. 2016. “Why do some Motorbike Riders Wear a Helmet and Others Don’t? Evidence from Delhi, India.” Transportation
Research Part A: Policy and Practice 88: 318-336. doi:10.1016/j.tra.2016.04.014.

Jain, A., G. Ahuja, Anuranjana, and D. Mehrotra. 2016. “Data Mining Approach to Analyse the Road Accidents in India.”. doi:10.1109/
ICRITO.2016.7784948.

Joshi, J., A. Singh, L. G. Moitra, and M. J. Deka. 2016. “DASITS: Driver Assistance System in Intelligent Transport System.”. doi:10.1109/WAINA.2016.80.

Kadali, B. R. and P. Vedagiri. 2016. “Proactive Pedestrian Safety Evaluation at Unprotected Mid-Block Crosswalk Locations Under Mixed Traffic Conditions.”
Safety Science 89: 94- 105. doi:10.1016/j.ssci.2016.05.014.

Kamal, V., D. Agrawal, and R. Pandey. 2016. “Epidemiology, Clinical Characteristics and Outcomes of Traumatic Brain Injury: Evidences from Integrated
Level 1 Trauma Center in India.”Journal of Neurosciences in Rural Practice7 (4): 515-525. doi:10.4103/0976- 3147.188637.

83
 Mohan, Tiwari and Bhalla
Karuppanagounder, K. and A. V. Vijayan. 2016. “Motorcycle Helmet use in Calicut, India: User Behaviors, Attitudes, and Perceptions.”Traffic Injury Prevention
17 (3): 292-296. doi:10.1080/15389588.2015.1055736.

Kaur, H., R. R. Singh, and M. Singh. 2016. “A Study of Injury Pattern among Road Traffic Accident Victims Admitted in a Medical College in Amritsar.” Indian
Journal of Public Health Research and Development 7 (4): 59-63. doi:10.5958/0976-5506.2016.00190.X.

Kumar, S. and D. Toshniwal. 2016. “A Novel Framework to Analyze Road Accident Time Series Data.” Journal of Big Data 3 (1). doi:10.1186/s40537-016-
0044-5.

———. 2016. “Analysis of Hourly Road Accident Counts using Hierarchical Clustering and Cophenetic Correlation Coefficient (CPCC).” Journal of Big Data
3 (1). doi:10.1186/s40537- 016-0046-3.

Maha Vishnu, V. C. and M. Rajalakshmi. 2016. “Bio-Motion Visual Analysis for Minimizing the Death Rate of Human Life in Accidents using Road Side Video
Surveillance.” Biomedical Research (India) 2016 (Special Issue 2): S257-S266.

Mohanavalli, S., E. Suma, G. Senthamarai, and G. S. Vijayabala. 2016. “Ocular Injuries in Association with Middle Third Facial Injuries in Developing
Countries: A Prospective Study.” World Journal of Dentistry 7 (3): 135-140. doi:10.5005/jp-journals-10015-1382.

Naveen Kumar, C., M. Parida, and S. S. Jain. 2016. “Recognising Risk Factors Associated with Crash Frequency on Rural Four Lane Highways.”.

P., R. K., N. G., R. S. P., S. C. P., and L. Krishna Prasad. 2016. “An Unusual Anterior Dislocation of Fractured Mandibular Condyle Leading to Psuedo-Ankylosis
in a 8 Yr Old child—A Distinct Case Report.”International Journal of Surgery Case Reports26: 34-37. doi:10.1016/j.ijscr.2016.07.011.

Patil, D. and R. Chaitanya. 2016. “Study of Pattern of Skull Fractures in Head Injury Cases among Road Traffic Accident Victims.” Medico-Legal Update 16
(2): 205-207. doi:10.5958/0974-1283.2016.00090.6.

Pradeep Kumar, M. V. and S. G. Arakere. 2016. “Study on unidentified/unclaimed Dead Bodies in Indian Forensic Settings.” Journal of South
India Medicolegal Association 8 (2): 99-103.

Prinja, S., J. Jagnoor, A. S. Chauhan, S. Aggarwal, H. Nguyen, and R. Ivers. 2016. “Article: Economic Burden of Hospitalization due to Injuries in
North India: A Cohort Study.” International Journal of Environmental Research and Public Health 13 (7). doi:10.3390/ijerph13070673.

Raju, K. and Gunnaiah. 2016. “Fatal Abdominal Injuries in Blunt Trauma - A Retrospective Study at District Hospital, Tumkur.” Indian Journal of
Forensic Medicine and Toxicology 10 (2): 308-311. doi:10.5958/0973-9130.2016.00119.5.

Ram, T. and K. Chand. 2016. “Effect of Drivers’ Risk Perception and Perception of Driving Tasks on Road Safety Attitude.” Transportation Research
Part F: Traffic Psychology and Behaviour 42: 162-176. doi:10.1016/j.trf.2016.07.012.

Ramanan, S. V., K. Radhakrishna, A. Waghmare, T. Raj, S. P. Nathan, S. M. Sreerama, and S. Sampath. 2016. “Dense Annotation of Free-Text Critical
Care Discharge Summaries from an Indian Hospital and Associated Performance of a Clinical NLP Annotator.” Journal of Medical Systems 40
(8). doi:10.1007/s10916-016-0541-2.

Rani, P. S., P. Subhashree, and N. S. Devi. 2016.“Computer Vision Based Gaze Tracking for Accident Prevention.”. doi:10.1109/STARTUP.2016.7583976.

Rankavat, S. and G. Tiwari. 2016. “Pedestrians Risk Perception of Traffic Crash and Built Environment Features - Delhi, India.” Safety Science 87:
1-7. doi:10.1016/j.ssci.2016.03.009.

Rashid, S., B. Kaur, and O. P. Aggarwal. 2016. “Interpretation of Injuries and Causes of Death among Victims of Fatal Road Traffic Accidents in
Mullana.” Journal of Punjab Academy of Forensic Medicine and Toxicology 16 (1): 16-19.

Reddy, A., J. Tejas, and R. Balaraman. 2016. “Strategic Analysis of Injuries and Causes of Death in Fatal Two Wheeled Vehicle Accidents-an
Autopsy Oriented Study in Southern India.” Medico-Legal Update 16 (1): 107-113. doi:10.5958/0974-1283.2016.00024.4.

Sandhu, H. A. S., G. Singh, M. S. Sisodia, and R. Chauhan. 2016. “Identification of Black Spots on Highway with Kernel Density Estimation
Method.” Journal of the Indian Society of Remote Sensing 44 (3): 457-464. doi:10.1007/s12524-015-0500-2.

Satapathy, M. C., D. Dash, S. S. Mishra, S. R. Tripathy, P. C. Nath, and S. P. Jena. 2016. “Spectrum and Outcome of Traumatic Brain Injury in Children
<15 Years: A Tertiary Level Experience in India.” International Journal of Critical Illness and Injury Science 6 (1): 16-20. doi:10.4103/2229-
5151.177359.

Sathish, P. and D. Bharathi. 2016. Automatic Road Sign Detection and Recognition Based on SIFT Feature Matching Algorithm. Advances in
Intelligent Systems and Computing. Vol. 398. doi:10.1007/978-81-322-2674-1_39.

Satish, K. V., S. S. Pujar, A. Ganpule, Chetan, and K. S. Sukesh. 2016. “A Cross Sectional Study of Risky Attitude and Behaviors among Young
Vehicle Users.” Medico-Legal Update 16 (2): 125-130. doi:10.5958/0974-1283.2016.00073.6.

Selvaraj, T. and K. Rajavelu. 2016. “Trends of Suicidal Death in and Around Madurai City during the Period between January 2015 to December
2015-a Retrospective Study.” Medico- Legal Update 16 (2): 120-124. doi:10.5958/0974-1283.2016.00072.4.

84
 Mohan, Tiwari and Bhalla
Selvaraj, T. and S. Sadasivam. 2016. “An Epidemiological Retrospective Study of Fatal Head Injury Related to Road Traffic Accident Victims in
Medicolegal Autopsies in and Around Madurai City.”Indian Journal of Forensic Medicine and Toxicology10 (2): 235-240. doi:10.5958/0973-
9130.2016.00102.X.

Selvasofia, A. S. D. and P. G. Arulraj. 2016. “Accident and Traffic Analysis using GIS.” Biomedical Research (India) 2016 (Special Issue 2): S103-S106.

Sharma, S. and D. Shah. 2016. “Real-Time Automatic Obstacle Detection and Alert System for Driver Assistance on Indian Roads.” Indonesian
Journal of Electrical Engineering and Computer Science 1 (3): 635-646. doi:10.11591/ijeecs.v1.i3.pp635-646.

Siddiqui, S. M., S. Sagar, M. C. Misra, A. Gupta, M. Crandall, and M. Swaroop. 2016. “Patterns of Injury among Motorized Two-Wheeler Pillion
Riders in New Delhi, India.” Journal of Surgical Research 205 (1): 142-146. doi:10.1016/j.jss.2016.06.033.

Singh, A. V. and J. S. Bhasin. 2016. “A Variable Speed Limit (VSL) Based Model for Advanced Traffic Management through VANETs.”. doi:10.1109/
WAINA.2016.93.

Singh, D. P., N. Kumar, M. Gupta, and M. Kumar. 2016. “Head Injury Pattern in Fatal Road Traffic Accidents.” Medico-Legal Update 16 (1): 128-132.
doi:10.5958/0974- 1283.2016.00028.1.

Singh, G., S. N. Sachdeva, and M. Pal. 2016. “M5 Model Tree Based Predictive Modeling of Road Accidents on Non-Urban Sections of Highways
in India.” Accident Analysis and Prevention 96: 108-117. doi:10.1016/j.aap.2016.08.004.

Subba Reddy, K., P. Sukanya, and M. Abdul Khalid. 2016. “Spectrum of Skull Fractures in Traumatic Brain Injury – A Cross Sectional Study.” Indian
Journal of Forensic Medicine and Toxicology 10 (2): 197-199. doi:10.5958/0973-9130.2016.00093.1.

Swain, R., S. Pooniya, A. Yadav, and S. K. Gupta. 2016. “Mortality due to Non-Existence of Child Restraint System in India.” Trauma (United
Kingdom) 18 (3): 221-223. doi:10.1177/1460408615606925.

Thenmozhi, T. and R. M. Somasundaram. 2016. Towards Modelling a Trusted and Secured Centralised Reputation System for VANET’s. Advances
in Intelligent Systems and Computing. Vol. 398. doi:10.1007/978-81-322-2674-1_64.

Udugu, K., V. R. Saddala, and S. Lingan. 2016. “Active and Passive Safety: An Overview on Establishing Safety Assessment Standards in India.”
SAE Technical Papers 2016-February (February). doi:10.4271/2016-28-0244.

Urie, Y., N. R. Velaga, and A. Maji. 2016. “Cross-Sectional Study of Road Accidents and Related Law Enforcement Efficiency for 10 Countries: A
Gap Coherence Analysis.” Traffic Injury Prevention 17 (7): 686-691. doi:10.1080/15389588.2016.1146823.

Valantina, G. M. and S. Jayashri. 2016. “Mesh Routers Based Routing for Saving Human Life in Vehicular Adhoc Network.” Biomedical Research
(India) 2016 (Special Issue 2): S210- S216.

Vasudevan, V., P. Singh, and S. Basu. 2016. “Importance of Awareness in Improving Performance of Emergency Medical Services (EMS) Systems
in Enhancing Traffic Safety: A Lesson from India.” Traffic Injury Prevention 17 (7): 699-704. doi:10.1080/15389588.2016.1163689.

Vayalamkuzhi, P. and V. Amirthalingam. 2016. “Development of Comprehensive Crash Models for Four-Lane Divided Highways in Heterogeneous
Traffic Condition.”. doi:10.1016/j.trpro.2016.11.117.

———. 2016. “Influence of Geometric Design Characteristics on Safety Under Heterogeneous Traffic Flow.” Journal of Traffic and Transportation
Engineering (English Edition) 3 (6): 559- 570. doi:10.1016/j.jtte.2016.05.006.

Vijay Kumar, A. G. and S. Javvadi. 2016. “Facial Bone Fractures in Road Traffic Accident: A Post Mortem Study.” Medico-Legal Update 16 (2): 217-
219. doi:10.5958/0974- 1283.2016.00093.1.

85
 Mohan, Tiwari and Bhalla
Transportation Research and Injury Prevention Centre
Indian Institute of Technology Delhi

http://tripp.iitd.ac.in

86