Death related to road traffic

crashes

By:
Bola Boules Mousa

Supervision:
Dr. Ismail Youssef

Administration:
Faculty of Engineering, El-Minya University
Contents:

1. Abstract

2. Introduction

3. Methods

4. Results and Discussion

5. Conclusion

6. References
1. Abstract

Background:
Due to a lack of an effective registry system for road traffic deaths, some
international organizations like the World Health Organization provide the
estimated number of road traffic deaths. It was shown that there are differences in
the number of road traffic deaths between the WHO estimates and national reports
even in High-Income Countries. This study aimed to investigate the differences
between the national reports and World Health Organization estimates about road
traffic deaths.

Methods:
This study focuses on investigating the World Health Organization reports of
Global Status Report for Road Safety in the years 2009, 2013, 2015, and 2018 and
related articles about the estimates of deaths related to road traffic crashes and the
observed differences between the WHO estimates and national reports.

Results:
The findings showed that the observed differences between the WHO estimates
and national reports could be due to errors in the road traffic death registration
system, errors in the regression model that was used for estimation, proposed
variables for estimations, or all of them.

Conclusions:
The estimations of WHO about road traffic deaths can be useful for countries,
especially for those that don’t have a road traffic registry system or the registry
system does not meet the quality criteria. These estimates may not be sufficiently
robust if disregard for spatial differences, the epidemiological pattern of risk
factors among the countries, and the type of regression model that was used for
estimation.
2. Introduction
Road traffic injuries are one of the major public health problems in many
countries around the world, especially in low- and middle-income countries
(LMICs).1-3 One of the weaknesses of road safety management and planning for
controlling and decreasing road traffic injuries (RTIs) is related to the
unavailability of accurate data due to the lack of road traffic registry systems. The
lack of road traffic mortality registry systems leads to the inability to determine
the size and nature of the traffic safety problem. This condition makes it difficult
to access accurate information for setting policy goals as well as developing,
monitoring, and evaluating action plans.1

In many high-income countries (HICs), there are high-quality data registry

systems (DRSs), which enable countries to identify risk factors. Such registry
systems can be applied to implement the most appropriate and timely
interventions for the prevention of road traffic crashes and their related injuries. In
countries where road traffic registry systems do not exist or are of poor quality,
the important step in identifying the situation and risk factors, and also providing
appropriate solutions is impaired. In LMICs, it may not be possible to set up a
good quality registry system in the short term. Therefore, there is a need to use
estimation methods to gain information about the number of deaths to provide
management and to take appropriate interventions.

Regression models are statistical methods that are used to estimate road traffic
crashes and traffic-related deaths. These estimation methods are used not only by
countries but also by international organizations, such as the World Health
Organization (WHO), for providing the required information and implementing
international-level interventions. 1 WHO has provided the estimated number and
rate of deaths related to road traffic crashes (per 100000 population) in the Global
Status Report of Road Safety (GSRRS) in 2009, 2013, 2015, and 2018.

It has been shown that there are differences in the number of road traffic deaths
between the WHO estimates and national reports. These differences are seen
among countries, but the level of difference is considerable, especially in LMICs.
According to previous studies, there may be under-reporting of road traffic-related
deaths (RTDs) in these countries.8-10 On the other hand, the difference observed
 in some HICs with a high-quality injury surveillance system raises the question of
whether the WHO estimates are likely to be erroneous, and what are the sources
of these possible errors? There are some important issues in the differences
observed between WHO estimates of deaths related to road traffic crashes and
those of the national reports in some countries.1,5-7 To the best of our knowledge,
no study has investigated the possible errors in WHO estimates. This study aimed
to investigate the reasons for the differences observed between the national reports
and WHO estimates of road traffic-related deaths.

3. Methods
This study aimed to investigate the WHO reports of GSRRS in the years 2009,
2013, 2015, and 2018, and also to review related articles about the estimates of
deaths related to road traffic crashes and the differences observed between WHO
estimates and national reports. Every two to three years, WHO publishes road
safety reports, which provide information about road safety for countries around
the world. Road traffic deaths were estimated for countries by using the negative
binomial regression model. At first, a fitted model was using the data of 86
countries with death registry systems. The completeness criteria for death
registration in these countries was 80% or more. The fitted model was used for
estimating road traffic deaths for countries. The covariates that were used for
estimating road traffic deaths included gross domestic product (GDP), total
vehicle per 1000 persons, total road per 1000 hectares, national speed limits on
rural and urban roads, health system access, working population, percentage of
motorbikes, corruption index, national policies for walking and cycling, and the
total population. The differences between WHO estimates and national reports
were studied in GSRRS.1

To determine the reasons for the differences observed between WHO estimates
and national reports, all studies about road traffic death registry systems,
regression models for estimation of road traffic-related deaths, and risk factors
associated with road traffic-related deaths were searched. The following terms
were searched in the databases of PubMed, Web of Science, and Scopus:
estimation, road traffic deaths, under-reporting, national report, and the
combination of these keywords.
 4. Results and Discussion
In 2009, 2013, 2015, and 2018 GSRSS, it was shown that there are differences in the
number of road traffic-related deaths between the WHO estimates and national reports.
Most of the countries in which the differences were observed belonged to the LMICs
(Table 1). These differences could be due to errors in the road traffic death registration
systems, errors in the regression model that was used for estimation, proposed
variables for the estimations, or all of them.

Table 1. The estimated and reported number of RTDS among some countries around the
world and Country Population in reports of 2013, 2015, and 2018.

GSRRS 2013 2015 2018

Countries NRNRTD1 ENRTD2 Country NRNRTD ENRTD Country NRNRTD ENRTD Country
with higher Population Population Population
estimation
than the
reported
number of
deaths
Iran 23249 25224 78,458,928 17 994 24 896 81,790,841 15932 16426 85,617,562

Viet Nam 11859 21651 90,267,739 9 845 22 419 92,191,398 8417 24970 94,914,390

Thailand 13365 26312 69,578,602 13 650 24 237 70,294,397 21745 22491 71,127,802

India 130037 231027 1,291,132,063 137 572 207 551 1,322,866,505 150785 2990191 1,369,003,306

Germany 3648 3830 81,680,591 3 339 3 540 82,073,226 581 599 82,896,696

Turkey 5253 8758 76,403,031 4786 6687 80,014,226 7300 9782 84,415,969

Pakistan 30131 26751 205,337,562 9 917 25 781 210,969,298 4448 27582 219,731,47917

Nigeria 5279 53339 174,726,123 6 450 35 641 183,995,785 5053 39802 198,387,623

China 70134 275983 1,363,240,000 62 945 261 367 1,379,860,000 58022 256180 1,402,760,000

Egypt 9608 10729 93,377,890 8 701 10 466 97,723,799 8211 9287 104,740,765

Bulgaria 775 776 7,431,167 601 601 7,309,253 708 730 7,309,253

Azerbaijan 1202 1202 9,614,478 1 256 943 9,863,480 759 845 10,152,522

Mexico 17301 16714 117,290,686 17 139 15 062 120,149,897 16039 16725 124,013,861

(1) National Reported Number of Road Traffic Deaths (NRNRTD).

(2) Estimated Number of Road Traffic Deaths by WHO (ENRTD).

Issues related to road traffic death registration systems
There are different types of road traffic death registration systems around the world. In
HICs, the injury surveillance system provides high-quality information about road
traffic-related injuries and deaths. However, in most LMICs, due to the lack of reliable
data registry systems, it is not possible to directly and accurately measure road traffic-
related injuries and deaths. Therefore, there are some other road traffic death
registration systems, which are used for national and international reports. For
instance, in countries like India and China, a reliable registration system is not
available. However, both countries have a sample registration system (SRS) that
provides an estimation of RTDs. In many African countries with no vital registration
systems, the information resources of cemeteries and the population and health care
networks provide the required information about RTDs. Many African cities have
cemeteries in which the cause of death is usually recorded for forensic reasons, while
the population and health care network information source reports the cause of death
in rural areas of Africa.

The death registration systems in LMICs often have local coverage that can only
provide reliable data for the covered regions. Therefore, the coverage of death
recording data may not be complete in these countries. Moreover, the quality of
registration of death causes may be inappropriate and thus lead to incorrect
information. Death registration systems typically use the International Classification of
Diseases (ICD)-10 framework for coding, which creates a more detailed set of codes
for recording deaths. However, detailed information may not be determined for
classification. In the national estimation of RTDs, the types of road users may be
undetermined. For instance, data may be recorded as a road user instead of
determining the pedestrians, drivers, occupants, and cyclists.

The findings of related studies show that there is under-reporting of RTDs. Results of
a study in Karachi, Pakistan showed that the under-reporting of RTDs in hospital
registry systems was about 20%, while the WHO estimated number of RTDs was
about 6 times higher than that of national reports.1 The number of RTDs in Iran’s
national report was lower than the estimated number of the WHO in the GSRRS
reported in 2009, 2013, 2015, and 2018. These estimations can be useful for countries,
especially for those that do not have road traffic registry systems or for countries in
which the registry system does not meet quality criteria such as completeness and
coverage. Countries in which the completeness of death registration data was 80% or
more were considered as countries meeting the quality criteria. According to the 2015
GSRRS, the number of RTDs was approximately 28% lower than the estimated
 number in Iran (17994 versus 24894). This difference in the 2013 report was about 8%
(23249 versus 25224), and in the 2009 report, it was approximately 11% (22918
versus 25491). The results of some studies indicate that there is under-reporting of
RTDs in Iran too. However, the amount of discrepancy reported in these studies differs
from that of the WHO estimate. The findings of related studies that have been
conducted in local areas show that there is 12-16% under-reporting of the registration
of road traffic deaths. These differences are also seen among other countries,
especially LMICs. For example, based on the 2015 GSRRS, this difference is about
66% in Vietnam, 3% in Germany, 22% in Egypt, 26% in Turkey, and 88% in China
(Table 1). The remarkable note is that some of these countries are categorized as
countries with high-quality registry systems for RTDs by the WHO, which were
entered to fit the regression model.

5. Conclusion
The estimations of WHO about road traffic deaths can be useful for countries,
especially for those that don’t have a road traffic registry system or the registry system
does not meet the quality criteria. These estimates may not be sufficiently robust if
disregard for spatial differences, the epidemiological pattern of risk factors among the
countries, and the type of regression model that was used for estimation.
6. References
1. Macrotrends
2. Worldometers
3. World Health Organization. Global status report on road safety 2018.
4. World Health Organization’s estimates of death related to road traffic
crashes and their discrepancy with other countries' national report
Alireza Razzaghi a ,b , Hamid Soori b ,* , Alireza Abadi c , d , Ardeshir Khosravi e
(LINK)