Back
Report numberRA-2005-51
TitleDeterminants of injury severity in traffic accidents
Subtitle
AuthorsBram De Brabander
Published byPolicy Research Centre for Traffic Safety 2002-2006
Number of pages122
Date14/01/2005
ISBN
Document languageDutch
Partner(s)LUC (nu UHasselt)
Work packageOther: Enforcement and policy
Summary

This report examines the chance to get injured when persons are involved in a traffic accident. Chances to get killed, to get minor and major injuries are calculated, as well the chance to incur only material damage. These chances are calculated with a multinomial logistic regression model. The accident severity is also examined with a CHAID analysis.

 

Variables in the model included age groups, the location of the accident (by means of the type of street – highways were not included in the study), different variables to describe weather conditions (snow, storm, rain, temperature below/above 0°C), the timing of the accident (night, weekend, official holiday, school holiday) and the type of accident (differenct types of casualties involved, as well as a distinction of the other party involved).

 

Empirical data from 2000 and 2001 from Antwerp and surroundig districts were used to build the models. The data from 2002 was used to examine the reliability of the variables found in the first model. The results from the 2002 data show that the variables used in the 2000-2001 model are reliable.

 

Results show that a number of variables do not have a significant impact on the outcome of an accident: none of the weather conditions or variables relating to the timing of the accident were significant.

 

The logistic regression model and the Chaid analysis show that:

  • Gender has an important impact on the outcome of accidents: women get almost twice as much lightly injured compared to men, when involved in an accident. Women also get more than twice severely injuried when involved in an accident. Young women up to 17 are a specific vulneralbe group;
  • Children up to 10 years have a much lower chance to get killed in a road accident, compare to other age groups. They bear, however, a worrying higher chance to get severely injured, when involved in an accident: depending on the type of accident and the location of the accident, these children face a chance twice up to 30 times to get severely injured, compared to other age groups;
  • Drivers and occupants of 4-wheel motor vehicles face much smaller chances to get killed, no matter where the accident happens or the age of the driver or occupant. When involved in an accident with an obstacle, these persons have a 10 times higher chance to get killed and a 4 times higher chance to get severily injured, compared to an accident in which they are involved with another road user;
  • Pedestrians face higher chances to get severily injured when involved in an accident. However, the age of the pedestrian is important: persons older than 59 years have a 3 times higher chance to get killed, compared to pedestrians being 30-59 years old;
  • Accidents in which a pedestrian is involved with a 2-wheel vehicle, almost never has fatal consequences. These pedestrians face, however, a more than average chance to get severily injured in this type of accidents;
  • It seems that experience plays an important role for 2 wheel vehicle road users: when involved in an accident –so besides the fact that more experienced persons might be less involved in an accident- older persons are less frequently severily injured. This is true for the case in which they are involved with another 2 wheel vehicle and a 4 wheel motor vehicle;
  • Older drivers and occupants of 4 wheel motor vehicles have more chance to get injured when involved in accident: they have a 4 times higher chance to get killed, 2,5 times more chance to get severily injured and 1,35 times more chance to get minor injuries. As a 2 wheel driver they face a 4 times higher chance to get killed, twice as much chance to get severily injured and 15% more chance to get minor injuries.

 

Policy recommendations are the following:

  • The development of an action plan focusing on the elderly seems important. It is shown in this research that the elderly have a higher probability to sustain severe injuries or to get killed when involved in an accident. In order to assess whether the elderly are a sufficiently large target group, one can observe that the group of people older than 59 years, constitute 25,55% of the population in the area studied in this research (Nationaal Instituut voor de Statistiek, FOD Economie, KMO, Middenstand en Energie). They represent, however, only 5,6% of the parties involved in accidents (which could be because they are less mobile – persons of the of 65 or older do about half the number of trips in Flanders compared to adults (Zwerts and Nuyts, 2002)  - or more cautious). Unfortunately they stand for 10% of the persons injured in accidents and even 17% of the persons seriously injured or killed.
  • If an infrastructure strategy would be chosen, priority should be given to arterial roads. From the CHAID analysis it was concluded that every type of injury existed more in accidents which happened on arterial roads. The absolute number of accidents on arterial roads is higher as well: 38,6% of the severely injured or killed casualties happen on this type of street, whereas only 25% of all parties involved in accidents are on arterial roads.
  • If investments in main streets and collector roads are done, priority should be given to the protection of pedestrians: 26% of the severely injured and killed casualties on these types of roads are pedestrians which are involved in an accident with a 4 wheel motor vehicle. An accident in which a pedestrian is involved with a 2 wheel vehicle has a high probability for the pedestrian being seriously injured or killed. This type of accident has, however, a very low frequency. The same is true for 2 wheel vehicle or pedestrians being involved in an accident with a train or tram: a high probability of being severely injured or killed is the case – they constitute 3% of the severely injured or killed. Combined with the fact that this type of accident has a low frequency (only 0,12% of the persons involved in an accident), this type of accident shouldn’t be given a high priority.
  • If it is chosen to improve road safety in residential streets, priority should be given to pedestrians and 2 wheel vehicles which are involved in an accident with a 4 wheel motor vehicle. Although they constitute 9% of the casualties in accidents in residential streets, they make up 53% of the seriously injured and killed casualties.
  • If government wishes to develop strategies for road user groups, it is necessary to take into account the different age groups. One can think of specific ways to communicate with these age groups or prepare specific measures for these age groups. Separate strategies for pedestrians and 2 wheel vehicles are important. Strategies to avoid accidents between pedestrians and 4 wheel motor vehicles should be given priority. These pedestrians make 9% of the injured or killed casualties and even 20 % ofthe seriously injured or killed casualties. Secondly, priority should be given to 4 wheel vehicles which are involved in an accident with anoter 4 wheel vehicle: 47% of those involved in accidents are this type of casualty and they make up 35% of the seriously injured or killed casualties). A lower priority should be given to accidents between 4 wheel vehicles and obstacles: they make only 3% of the injured or killed casualties and constitute 7% of the seriously injured or killed.
  • A number of accident types lead to a high probability of serious injury or death, but have a low frequency. This way, they shouldn’t be a priority for decision makers. The accident types are the following: pedestrian-2 wheel vehicle, 2 wheel vehicle-train/tram, 2 wheel vehicle – obstacle, pedestrian – train/tram.
DownloadPDF icon RA-2005-51.pdf
Lijn

Mission

The Policy Research Centre for Traffic Safety carries out policy relevant scientific research under the authority of the Flemish Government. The Centre is the result of a

cooperation between Hasselt University, KU Leuven and VITO, the Flemish Institute for Technological Research.

Partners

Leuven vito