All contents copyright ©, Drive and Stay Alive, Inc., 2003 onwards, unless specified otherwise. All rights reserved.
IMPORTANT: click here to read the DISCLAIMER
Multi-Country Per Capita Fatality Data for 2003
Last updated on January 20, 2005 (re USA and Republic of Ireland) and on January 29 (Jamaica)
It must be remembered that there are three primary measures for comparing multi-national crash and fatality data: the deaths per 100,000 population or per capita rate, as shown here, deaths in relation to overall distance travelled (known in the USA as the VMT rate), and deaths in relation to the number of registered motor vehicles in the country. All three measures should be considered when comparing disparate countries but using just one of these methods is generally acceptable when comparing countries of similar status (e.g. "highly motorised countries" [HMCs], developed nations, third world countries, etc.).
As a result, some countries in the above table may appear to present bizarre results, either because -- like China, for example -- they have a very high death toll but it is offset by a huge population, or they simply have, say, a very low proportion of motor vehicles per head of population -- such as Brunei, that is currently at the head of the per capita table, or Ghana.
There is also the question of how, exactly, a traffic fatality is defined in any particular country. Some may only include deaths at the scene, whereas others will stipulate deaths within 24 hours, and some may allow a full week or even 30 days.
In some cases, therefore, the data for the number of deaths simply cannot be relied upon as being accurate. In Turkey, for example, the national press state that over 9,000 people are killed in road crashes each year, and yet each year data is published by that country giving a much lower body count. For that reason we have elected to position Turkey in the table to allow for an approximate per capita rate of 13.06 (based on the aforementioned 9,000 estimate) but have not shown the rate in the relevant column.
a. Applies only to the thirty member-countries of the Organisation for Economic Co-operation and Development, as listed in
the IRTAD 2002 table (which may be updated to 2003 at some stage). The results posted here have not been prepared on
behalf of, or from data supplied by IRTAD and must not be taken as being official IRTAD results
b. The per capita rate represents the number of deaths for each 100,000 members of the population
d. Population figures are actually July 2004 and sourced from the CIA World Factbook, except where shown otherwise
e. In the USA, the measure used is deaths per 100 million Vehicle Miles Traveled (VMT). Elsewhere it is deaths per one
billion vehicle kilometres and it is this latter measure that is used in the above table
f. The official figure for Turkey is given as 3,966, as shown, but the Turkish press estimate the number of fatalities at 9,000
(see comments section, above)
g. The figure for Brunei illustrates two things. Firstly, it is to be expected that countries with a lower ratio of vehicles to people
will probably have a lower per capita rate (though this is not always the case). And, secondly, in countries with a low
number of deaths -- such as here -- the annual rate can fluctuate wildly. For example, in Brunei in 2004 twenty people
were killed in road crashes, effectively putting the per capita rate up by 66%, from 3.29 to 5.48
h. The figure was also stated of 4.5 deaths per 10,000 registered motor vehicles for 2004 (down from 4.9 in 2003)
k. Between 2003-2004, the number killed jumped from 336 to 380, and the rate from 8.46 to 9.57
l. The figure for Jamaica in 2004 fell significantly to 351, which will give a per capita rate of 12.95
m. The number of deaths for 2004 is being entered solely as a aide-memoir to help in the preparation of a future table. These
figures do not for m part of any calculations or conclusions pertaining to 2003 data.
ab. When the number of deaths given, for 2003, is compared to the population figure shown in the CIA World Factbook
(admittedly a 2005 estimate) the rate would actually be 27.88 -- or the population should have been around 48.83 million
in order to get a rate of 25.3 -- so there is an apparent anomaly somewhere in the South Africa figures shown in the table.
ae 2003 figure updated when the 2004 figure was posted, on June 30, 2005.
af The 2004 per capita rate for the USA, based on an estimated population of 293 million, is 14.55
ag The overall number of fatalities was updated in line with NHTSA announcements made in August 2005. This results in
an increase in the per capita death rate from the 14.66 that was still shown on the FARS database as at August 2, 2005,
to 14.75, and the above table has been amended accordingly.
c. ECMT Press Release 12/10/04, except where marked otherwise
g. Brudirect January 18, 2005
h. Daily Express, East Malaysia; January 19/20, 2005 (though the detail is vague. The figure is described as fatal road
accidents yet is then described as being "63 less than the 6,286 fatalities recorded in 2003".) The 2003 figure was,
however, confirmed by Bernama on May 29, 2005.
k. CNN International -- On Irish roads, hurry up and slow down -- January 20, 2005
l. The Jamaica Observer, January 27, 2005 -- Pickersgill preaches highway safety...
n. The Cyprus Weekly, 1 February, 2005.
o. GhanaWeb, 31 January, 2005.
q. The World Bank, quoted in the Moscow News, 28 February, 2005. Also RIA Novosti, April 10, 2005.
r. 2002 IRTAD data -- Belgium
s. 2002 IRTAD data -- Italy
t. White Paper on Traffic Safety in Japan, 2004 -- abridged, English language version
u. 2002 IRTAD data -- Republic of Korea
v. Australian Transport Safety Bureau -- Dec. 2003 Bulletin
w. NHTSA / FARS
x. Transport Canada -- Collisions and Casualties, 1984-2003
y. Land Transport Safety Authority, NZ
z. WHO estimates for 2002 (250,007 deaths) extrapolated to 2003 as shown in Single Planet
aa Helsingin Sanomat - March 8, 2005
ab Government wants to improve road safety, efficiency by 2010 World Cup from The Star (Malaysia) May 20,2005
(see 'ab' in notes, re apparent inaccuracy)
ac Helsingin Sanomat - June 13, 2005
"...The rate of deaths in Bahrain's roads in 2003 and 2004 were just over 10 per 100,000 people...."
ae Department for Transport press release 2005/0075, 30 June 2005:
The number of people killed on Britain's roads in 2004 was the lowest figure since records began in 1926, figures
published by the Department for Transport today show. In 2004, 3,221 people died in road accidents, down 287 or 8%
on the figures from 2003 when 3,508 people were killed. This is despite an estimated increase in road traffic of 2 per cent
af US DOT press release 106-05, August 1, 2004