Written evidence from Dr Donal McNally (CYS0119)

Author

  1. Dr McNally is Associate Professor and Reader in Bioengineering and Head of the Structural Integrity, Dynamics and Bioengineering Research Group at the University of Nottingham.  He has been engaged in post-doctoral research in biomechanics since 1989 and specifically impact biomechanics since 2001.  He has published more than 160 articles in journals and conferences and has won a number of international and national research prizes including the International Society for the Study of the Lumbar Spine Award, the International Society of Biomechanics Clinical Biomechanics Award and the European Spine Society Acromed Prize for Spinal Research.
  2. Dr McNally is a keen cyclist who used to commute daily (100 miles per week) by bicycle until 2010 when he was hit by a car whilst cycling home.  Three vertebrae in his neck and one in his lower back were fractured in this collision; he also sustained a head injury resulting in a short period of unconsciousness, some cracked ribs and 4 months off work.  Not surprisingly, he has focussed his recent impact biomechanics research on the mechanisms of injury in bicycle collisions and their mitigation.

Is cycling dangerous?

  1. Cycling is the most dangerous common form of transport with the exception of motorcycling.  For an equivalent journey, such as a commute or trip to the shops, a cyclist is 40 times more likely to be killed or seriously injured than a car driver (Department for Transport, 2014a).  For more detail see Figure 1.

Figure 1: Comparison of injury rates for different road users – pedestrian and car safety is considerably better than cycle safety.  Data taken from Department for Transport data (Department for Transport, 2014b; Department for Transport, 2014d; Department for Transport, 2014f)

  1. In 2012 more than 2000 cyclists were killed or injured (Department for Transport, 2014b).  The Department for Transport places a value of prevention on such casualties of £1.3 billion (Department for Transport, 2014c).
  2. In 2012, 118 cyclists were killed in Great Britain.  To put this into context, the 2009-10 and 2010-11 H1N1 influenza pandemics together were responsible for 835 deaths in England (Mytton, Rutter, & Donaldson, 2012); over the same period 322 cyclists died on UK roads (Department for Transport, 2014b).
  3. Dramatic improvements in vehicle safety have not been reflected by large improvements in cycle safety.  Since 1979 the killed or seriously injured rate (number per billion miles) for cyclists has reduced by 44% (Department for Transport, 2014b) compared to 86% for car occupants (Department for Transport, 2014d).  The substantial reduction in vehicle occupant safety reflects the advances in vehicle safety engineering such as seatbelts, airbags, crumple zones and protected occupant space.  Such engineered improvements are of little relevance to cyclists.  Figure 2 demonstrates recent significant improvements in car and pedestrian safety compared to a worsening in cycle safety.

Figure 2: Improvement in road user safety since 2003 – pedestrian and car safety has improved substantially whilst cycle safety has become worse.  Data from the Department for Transport (Department for Transport, 2014b; Department for Transport, 2014d; Department for Transport, 2014f)

  1. Cyclists are much more vulnerable to serious injury than car occupants when a collision does occur.  In 2012 17% of cyclist casualties were killed or seriously injured (Department for Transport, 2014b) compared to only 8% of car occupants (Department for Transport, 2014d).
  2. The risk of being killed or seriously injured whilst cycling is approximately 1 per million miles (Department for Transport, 2014a).  Whilst this seems to be a small risk, it corresponds to a 5% chance of being killed or seriously injured whilst cycle commuting 10 miles each way for 10 years.  Many people would consider this to be an unacceptably high risk.

What can be done to prevent collisions involving cyclists?

  1. Make better use of current best practices.  There is a remarkable difference in casualty rates between local authorities. For example, in 2012, the casualty rate in Walsall was 96/million population, whilst in Portsmouth it was 832/million (Department for Transport, 2014e).  If the median casualty rate was reduced to the level of the current best performing local authority (Rochdale), casualty rates would drop by a factor of 3.
  2. Physical separation of cyclists from other road users.  Other European countries such as the Netherlands have moved further towards this solution unlike the UK with dedicated cycle lanes that are physically separated from other traffic.  On road cycle lanes do not physically separate bicycles from vehicles, hence in London from 2001-2006, 29% of fatalities occurred on on-road cycle lanes (Kaigan, Cuerden, & Wheeler, 2009).
  3. Evaluate carefully the benefits of shared bus and cycle lanes.  Whilst such shared lanes take cyclists out of stop-start traffic, it places them in direct conflict with buses that are both travelling at speed and which pull in to the kerb regularly.  Since the most frequent cause of cyclist deaths on urban roads is large vehicles turning left (Kaigan, Cuerden, & Wheeler, 2009), shared bus and cycle lanes may not be a safe haven they are supposed to be.  More research is required in this area.
  4. Better standards and planning guidelines for cycle lanes. It is my personal experience of cycle lanes in the UK that they are frequently poorly designed.  They are often narrow, end abruptly at locations of increased hazard such as junctions and places where the road narrows, and are obstructed by street furniture.

Vehicle design can mitigate cyclist injuries following a collision

  1. The dynamics of adult cyclist impacts with cars are very different from those of pedestrian impacts.  Initial contact is between the legs of the pedestrian and the front end of the car causing the pedestrian to rotate striking their heads either on the bonnet or windscreen (Dunmore, Brooks, Madeley, & McNally, 2006); the most severe injuries come from the impact with the car.  Cars are now designed with ‘soft’ front-ends, bonnets and windscreens to minimise pedestrian injury (see Paragraph 14). Cars hit the bicycle rather than the cyclist, the cyclist then slides up and over the car (often incurring only minor injuries) before being thrown into the air.  Cyclists can be thrown to heights in excess of 2m before hitting the ground.  It is the impact with the ground that results in the most severe injuries (McNally & Whitehead, 2013).
  2. Modern cars are designed with pedestrian safety in mind.  Pedestrian protection was first regulated within the EU in 2003 by Directive 2003/102/EC and this regulation has undergone a succession of refinements up to the current Regulation (EC) No 78/2009.  Similar testing is enshrined in the voluntary Euro NCAP tests.  These tests are relevant to pedestrian impacts only, collisions with cyclists are not adequately considered.  Since the introduction of the legislation in 2003, pedestrian injury rates (number per billion miles) have dropped by 25% (Department for Transport, 2014f), whist cyclist injury rates have increased by 19% (Department for Transport, 2014b) (see Figure 2).
  3. Mitigation of cyclist injuries requires considerable innovation in car design.  Recent advances in active pedestrian safety systems such as external windscreen pillar air-bags indicate that such innovations are possible without compromising aerodynamic efficiency of the vehicle.  The history of pedestrian safety systems suggests that the best way to drive such innovation and implementation is by the adoption of voluntary vehicle safety tests and regulation.

Cycle helmets mitigate cyclist injuries following a collision

  1. Currently, the only things that reduce injury severity once a collision has occurred are cycle helmets.  They do this by absorbing the energy of the impact before it reaches the head.  The energy that needs to be absorbed, in a typical cycle collision, is equivalent to being struck on the head by a baseball bat.  Cycle helmets are carefully designed and regulated to provide this level of protection. 
    When I was hit by a car, the collision and subsequent impact with the ground, would have left me dead or in a wheel chair had my helmet not absorbed a considerable amount of energy.  The evidence for the effectiveness of cycle helmets is both compelling and robust, coming from a range of different types of study.
  2. The most direct evidence comes from hospital admission studies, where the injuries and outcomes of cyclists are compared on the basis of helmet wearing.  These studies are difficult to perform and complex to interpret, however their findings are clear.  There was a Cochrane systematic review of these studies (Thompson, Rivara, & Thompson, 1999) which concluded that helmet wearing provides a 63- 88% reduction in the risk of head, brain and severe brain injury for all ages of cyclist.  These findings have been supported by other similar analyses (Elvik, 2013).  It should be noted that such admissions studies take into account all features of the collision including risk compensation behaviour by both the cyclist and the driver of any vehicle involved.
  3. Manufacturers and Regulators assess helmet performance in much the same way as the motor industry assess vehicle safety system performance, using standardised tests and a relevant injury threshold criterion.  In the case of cycle helmets, this normally involves a test where an instrumented head-form wearing the helmet is dropped onto one or more anvils.  These tests reproduce the impact speeds found in collisions, whilst the anvils are chosen to represent a flat road surface, kerb edge etc.
  4. Computer simulations (McNally & Rosenberg, 2013; McNally & Whitehead, 2013) have been used to bridge the gap between clinical statistics (as outlined in paragraph 17) and abstract physical tests (as described in paragraph 18).  Not only do such studies confirm the effectiveness of helmets in reducing the risk of injury, they are able to show the following refinements:
    1. Wearing a helmet prevents the non-life threatening head injuries, such as simple skull fractures and mild concussion, that would have been sustained had a helmet not been worn.
    2. Helmets are effective over the full range of cycle speeds up to 14 m/s (a 31 miles/hour race speed).
    3. Helmets are effective even in collisions with cars moving at speeds up to 40 miles/hour.
    4. Helmets are protective against neck injuries and rotational brain injuries.
  5. There is one widely cited paper (Walker, 20017) that appears to demonstrate that car drivers will pass cyclists closer if they are wearing a helmet.  Careful reanalysis of the original data has shown that there is no difference in terms of dangerous close passing, but that extremely wide overtaking of cyclists not wearing a helmet is more common (Olivier & Walter, 2013).  Further, even if driver behaviour towards cyclists in helmets was more dangerous, these increased dangers would be included in the evaluations made by the hospital admissions studies discussed in Paragraph 17.
  6. Mandatory cycle helmet wearing is enforced by law in Australia and New Zealand and in parts of Canada and the United States.  Such laws are highly controversial given their impact on an individual’s freedom to choose.  However, scientifically, there is no controversy; the evidence that these laws are effective in reducing head injury is very strong (Walter, Olivier, Churches, & Grzebieta, 2011).  Initially, there was some scientific debate because such studies are difficult to perform and complex to interpret since there are many uncontrolled factors which affect injury rate, such as other road safety improvements, changes in cycling popularity etc.  In my opinion, the effectiveness of cycle helmet laws has been demonstrated beyond doubt by careful control of such variables.
  7. Maintaining the balance between the freedom of the individual and the corresponding cost to society in general is the most important role of government.  My personal view as a cyclist, injury biomechanics expert and casualty is that the costs of helmet wearing are massively outweighed by the benefits to public health and finance not just those of the individual cyclist casualty.
  8. The health benefits of cycling for a population where diseases of inactivity, such as obesity and heart disease, are so prevalent is often cited by health professionals and cycling organisations as a reason to avoid highlighting the dangers of cycling with safety campaigns and helmet laws.  However, this argument is essentially conjecture; there is little or no evidence that people stop cycling as a result of these interventions, that those that stopped took no alternative exercise, and that a health burden was incurred as a result (Biegler & Johnson, 2013).  I believe that it is more ethical to do everything possible to make cycling safer, including presenting safety information in a meaningful context so that members of the public can make properly informed decisions, and actively promoting helmet wearing and recreational cycling away from the dangers of other road traffic.

References

Department for Transport. (2014a, January 13). RAS30070: Relative risk of different forms of transport, Great Britain: 2011. Retrieved from Department for Transport statistics: https://www.gov.uk/government/uploads/system/uploads/attachment_data/file/10201/ras30070.xls

Department for Transport. (2014b, January 13). RAS30065: Reported pedal cycle casualties by age, Great Britain, 1979 - 2012. Retrieved from Department for Transport statistics: https://www.gov.uk/government/uploads/system/uploads/attachment_data/file/244626/ras30065.xls

Department for Transport. (2014c, January 13). RAS60001: Average value of prevention1 per reported casualty and per reported road accident: GB 2012. Retrieved from Department for Transport statistics: https://www.gov.uk/government/uploads/system/uploads/attachment_data/file/243679/ras60001.xls

Department for Transport. (2014d, January 13). RAS30067: Reported car user casualties, Great Britain, 1979 - 2012. Retrieved from Department for Transport statistics: https://www.gov.uk/government/uploads/system/uploads/attachment_data/file/244628/ras30067.xls

Department for Transport. (2014e, January 13). RAS30045: Reported casualty rate per million population by region, local authority and road user type, England, 2012. Retrieved from Department for Transport statistics: https://www.gov.uk/government/uploads/system/uploads/attachment_data/file/239776/ras30045.xls

Department for Transport. (2014f, January 13). RAS30064: Reported pedestrian casualties by age, Great Britain 1979 - 2012. Retrieved from Department for Transport statistics: https://www.gov.uk/government/uploads/system/uploads/attachment_data/file/244625/ras30064.xls

Dunmore, M. C., Brooks, R., Madeley, N. J., & McNally, D. S. (2006). The effect of leg fracture level and vehicle front-end geometry on pedestrian knee injury and response. Proceedings of the Institution of Mechanical Engineers Part H: Journal of Engineering in Medicine, 220, 857-869.

McNally, D. S., & Rosenberg, N. M. (2013). MADYMO simulation of children in cycle accidents: A novel approach in risk assessment. Accident Analysis and Prevention, 59, 469– 478.

McNally, D. S., & Whitehead, S. (2013). A computational simulation study of the influence of helmet wearing on head injury risk in adult cyclists. Accident Analysis and Prvention, 60, 15-23.

Mytton, O. T., Rutter, P. D., & Donaldson, L. J. (2012). Influenza A(H1N1)pdm09 in England, 2009 to 2011: a greater burden of severe illness in the year after the pandemic than in the pandemic year. Euro Surveillence, 17(14), pii 20139. Retrieved from http://www.eurosurveillance.org/ViewArticle.aspx?ArticleId=20139

Olivier, J., & Walter, S. R. (2013). Bicycle Helmet Wearing Is Not Associated with Close Motor Vehicle Passing: A Re-Analysis of Walker, 2007. PLoS ONE , 8(9). doi:10.1371/journal.pone.0075424

Thompson, D. C., Rivara, F., & Thompson, R. (1999). Helmets for preventing head and facial injuries in bicyclists. Cochrane Database of Systematic Reviews (4). doi:10.1002/14651858.CD001855

Walker, I. (20017). Drivers overtaking bicyclists: Objective data on the effects of riding position, helmet use, vehicle type and apparent gender. Accident Analysis & Prevention, 39, 417–425.

Walter, S. R., Olivier, J., Churches, T., & Grzebieta, R. (2011). The impact of compulsory cycle helmet legislation on cyclist head injuries in New South Wales, Australia. Accident Analysis and Prevention, 43, 2064– 2071.

January 2014

              Page 2 of 6