The challenge
Every year 30,000 people are killed or seriously injured on UK roads. To improve road safety, we need to understand the factors which contribute to accidents, but research by Dr Jonathan Rolison uncovered serious flaws in the way data was collected and reported.
He identified three major limitations which undermine the reliability of accident data: unreliable recollection when filling out accident reports back at the police station; biased judgement, including age and gender stereotypes, and subjective interpretation of the contributory factors listed on standard police accident reports.
What we did
Dr Rolison’s study investigated the main causes of road accidents by drawing on the expert views of police officers and official road accident records for all one and two-car accidents in the UK between 2005 and 2012.
Participants in the study were given details of six hypothetical accidents, involving both male and female drivers of varying ages, and were asked to identify what caused each one, and whether there were any other contributing factors. Their responses were then compared to the factors recorded in official accident reports.
Dr Rolison found discrepancies between the two, as he explained: “While police officers frequently identified driver distraction, and in particular mobile phone use, for the hypothetical scenarios, this factor was rarely reported in the accident records.
“Either mobile phone use is an under-reported factor in road accidents or police officers, and the general public, have exaggerated beliefs about the dangers of using mobile phones while driving.
“Given the wealth of evidence which shows driver distraction raises the risk of accidents – potentially by as much as 50% - we think mobile phone use is probably an under-reported factor in accidents.”
Similarly, the study concluded driving while under the influence of drink or drugs and poor vision could have contributed to far more accidents than previously thought.
Dr Rolison found the time delay between an accident and a police officer filling out the accident report was also an issue. At the scene, an officer has many duties, from looking after injured passengers, collecting witness statements, and managing traffic to potentially arresting a driver. So the accident report, including a description of the factors judged to have contributed to the collision, is often completed at the police station some time later.
This is problematic, as Dr Rolison found if police officers couldn’t remember details from the scene, they tended to rely on preconceived stereotypical ideas – drugs and alcohol, excessive speed, inexperience and distraction are thought to be more likely contributory factors for young drivers, while medical conditions and poor eyesight are often associated with accidents involving older drivers.
Finally, he found differences in the way forms were filled out – so classifying mobile phone use as driver impairment, rather than driver distraction - also led to inconsistent reporting or under-reporting of contributory factors.
To overcome the issues he had uncovered, Dr Rolison devised a mobile app which can be used at the scene to record details including the circumstances of the accident, the vehicle, casualties and contributory factors. It is linked to google maps to identify the precise location of an accident.
The contributory factors module, which was devised with the help of serving officers, was designed so the sequence of events leading up to an accident can be recorded – providing a much more detailed picture, than has ever been possible before.
The app reduces the time police officers spend recording information and the time it takes the Department of Transport to analyse the data, which they use to decide on road safety policy.
Data input at the scene removes reliance on memory and in turn reduces the likelihood of bias.
What we changed