Intended for healthcare professionals


Death and injury on roads

BMJ 2006; 333 doi: (Published 06 July 2006) Cite this as: BMJ 2006;333:53
  1. Shanthi Ameratunga, director, Injury Prevention Research Centre (s.ameratunga{at},
  2. Rod Jackson, professor of epidemiology,
  3. Robyn Norton, professor of public health
  1. School of Population Health, Faculty of Medical and Health Sciences, University of Auckland, Private Bag 92019, Auckland, New Zealand
  2. School of Population Health, Faculty of Medical and Health Sciences, University of Auckland, Private Bag 92019, Auckland, New Zealand
  3. University of Sydney, George Institute for International Health, PO Box M201, NSW 2050, Australia

    Lowering the road toll will take much more than altering road users' behaviour

    In 2002 a theme issue of the BMJ focused on the unacceptable, and largely neglected, global toll of road traffic crashes.1 The subsequent report by the World Health Organization highlighted that low and middle income countries bear the brunt of this burden, accounting for more than 85% of the deaths and 90% of disability adjusted life years lost from road crashes.2 In contrast, many high income countries (including the United Kingdom) were shown to have sharply reduced their rates of road crashes in recent decades, exemplifying what could be achieved.

    It is tempting to bask in the glory of such creditable achievements, but, as always, the devil is in the detail. A heterogeneous and unplanned collection of three papers appearing in this week's BMJ35 and one being published on line6 provides a timely reminder of the extent to which the potential to prevent road traffic injury remains a challenge, even in rich countries.

    Examining 20 years of data up to 2001, Edwards and colleagues report a 63% decline in the rates of child deaths from injury in England and Wales.6 Below the surface of this heartening trend, however, are steep social class gradients indicating that the gains in life and health have largely eluded children from the poorest families. The authors calculate that 600 fewer children would have died from injury in 2001-3 if all children in England and Wales could aspire to the low death rate of the highest income group. These disparities were especially marked for deaths among child pedestrians and children on pedal cycles, with children in the lowest socioeconomic group experiencing cause specific mortality more than 20 times that of children in the highest group.

    Similar disparities in child mortality are evident elsewhere, with important differences in the exposure to risk and the environmental risk factors found to underlie the apparent socioeconomic differentials. These include the speed and density of traffic in residential areas, access to safe play areas, and fenced driveways.7 8 Approaches to reduce these inequities must tackle economic and transport policy as well as interventions affecting the environment, vehicles, and road users, rather than relying solely on changing the behaviour of victims.

    The study by Walker and colleagues provides little room for complacency regarding our ability to influence human behaviour.4 Although robust legislation was enacted more than 20 years ago, this study found that one in six drivers in London did not wear seatbelts.4 A worrying group also flouted highly publicised new legislation banning the use of hand held mobile phones while driving. The lack of an observed change in compliance during the transition from the grace period to penalty phase of implementing the law suggests a disturbing lack of awareness or disregard of the new legislation by these drivers. While the report focused on the more risky behaviour of drivers of four wheel drive vehicles, the behaviour of other drivers showed plenty of room for improvement.

    Also in this issue, Nabi and colleagues draw attention to the strong association between sleepiness and the risk of road traffic crashes.5 French drivers reporting that they drove when sleepy once a month or more over 12 months had almost three times the risk of serious injury in the subsequent three years compared with those who did not drive when sleepy. While seat belts, speed, and alcohol consumption are emphasised in numerous road safety campaigns, sleepiness does not seem to have attained front line status. Yet strategies to prevent driving while acutely sleepy could prevent almost 20% of car crashes that cause serious injury.9 As noted by Nabi and colleagues, however, the main challenge may be to convince people to stop driving when they are sleepy rather than helping drivers to recognise their signs of sleepiness.

    Regardless of the context, policy makers and planners require one or more yardsticks by which to measure the performance and celebrate the achievements of programmes designed to drive down the death toll and serious injury on the roads. With competing priorities, not to do so is programmatic suicide.

    But, also in this week's BMJ, Gill and colleagues show the need to be circumspect when monitoring trends in non-fatal road traffic injury.3 First, their study confirms previous research suggesting that under-reporting of non-fatal injuries in police statistics is increasingly common, including injuries severe enough to warrant hospital admission.10 Secondly, for every death due to road traffic injury in England, there are about 20 admissions to hospital, indicating that the more reliable mortality statistics represent only the tip of the iceberg of road traffic injuries.

    Finally, if robust indicators of serious non-fatal injury are difficult to find in high income countries, the degree of under-reporting is probably considerably greater in low and middle income countries, which experience a disproportionate burden of crashes.2 11 12 We have almost certainly underestimated considerably the global epidemic of road traffic injury.


    • Competing interests None declared.

    • Research p 71, 73, 75


    1. 1.
    2. 2.
    3. 3.
    4. 4.
    5. 5.
    6. 6.
    7. 7.
    8. 8.
    9. 9.
    10. 10.
    11. 11.
    12. 12.
    View Abstract