Written evidence from Richard Armitage, David Hurdle, Adrian Lord and Alex Sully, transport consultants (CYS0128)

A balanced approach to cycle safety

Summary

  1. The rapid rise in cycling over the last few years has raised the media profile of cycling as a mode of transport. Following the Times’ Cities fit for Cycling Campaign, other publications have also given prominence to reporting death and injuries to cyclists, particularly in London, along with much editorial about who is to blame and what can be done to improve safety. This debate is welcomed as it led directly to the ‘Get Britain Cycling Parliamentary Inquiry’ which identified some positive steps that government could adopt to promote more and safer cycling.  However it has involved some sensationalist reporting (in comparison to deaths of pedestrians, car drivers and passengers which often go unremarked) and perhaps led to a public perception that cycling is an inherently ‘hazardous’ mode of transport.
  2. In London, it has been shown that HGVs are disproportionately involved in fatal pedal cycle collisions, where 53 per cent of pedal cycle fatalities between 2008 and 2012 involved direct conflict with a HGV. Following an apparent trend of tipper lorry involvement in collisions resulting in a cyclist fatality, a review of the 2011 data for London was undertaken. It was found that seven of the nine large goods vehicles involved in a fatal cyclist collision that year were construction vehicles. There are inherent challenges in operating such vehicles in busy urban areas where pedestrians and cyclists are in close proximity. The outcomes of collisions between HGVs and unprotected pedestrians and cyclists, even at low speeds, are inevitably serious due to the nature of the injuries sustained, which are often multiple due to the initial impact and then being crushed under the wheels. The majority of KSI collisions involving cyclists and HGVs appear to occur due to cyclists positioning themselves and then being caught in the blind spot of a left-turning vehicle.  Some operators and local authorities have taken steps to improve road user education and to develop and install infrastructure and vehicle design improvements and detection systems to help address this issue. However, it should not be forgotten that the vast majority - for instance, 69% of KSIs in London 2010-12 - of collisions resulting in death and serious injury to pedestrians and cyclists involve cars and taxis[1].
  3. Most cycling, and therefore most accidents, occur in urban areas during daylight and in the summer months. However, the rate of accidents increases in the dark winter months when people who continue to cycle are proportionately at greater risk. Fewer accidents occur in rural areas (around 25%) but they tend to result in more serious injuries.
  4. The number of cycling injury accidents is likely to be significantly under-reported in official statistics, and while fatalities continue to reduce, there appears to be a recent upward trend in serious injury accidents. With the exception of motorcyclists, cyclists appear to be at greater risk on the roads than other road users.
  5. However, viewed in the context of the amount of cycle journeys undertaken, the risk of death and serious injury while cycling is relatively low. There is a separate issue of ‘perceived safety’ in which people feel at risk on the roads due to a culture of fairly aggressive driving behaviour in combination with a lack of any widespread network of safe cycle infrastructure in most towns, cities and rural areas. Many consider this to be a barrier to more cycling especially among women, children and elderly people.  Many individuals feel both unable and unwilling to participate in a traffic system that requires higher speed, assertive cycling in order to mix safely with other vehicles.
  6. Recommendation: The Department for Transport should undertake a thorough review of evidence from the UK and abroad to evaluate what design features of cycle infrastructure and general highway design particularly address the most common factors of injury accidents to cyclists, particularly at junctions where 75% of all accidents occur. This factor appears to be amplified in London, where 85% of pedal cycle collisions (resulting in 1 or more cycle casualty) occur at junctions, and 78% of all collisions occur at junctions. This will help traffic engineers and planners to understand the risks and benefits of different junction layouts for cyclists and to design out the kind of conflicts that commonly result in injury accidents.  It is especially important that government provides the Department for Transport with sufficient resources to commission this work and to promote the resulting guidance to ensure that it is widely understood and adopted. The Department for Transport should take the lead on this as it will help to ensure a more consistent, and therefore safer, national approach to infrastructure design. In this context, the Department should continue to support policies on encouraging more cycling, such as Active Travel, Health and Well-being and Travel Planning.
  7. Recommendation: A national programme of publicity should be developed to make cyclists and vehicle drivers aware of the ‘blind spots’ immediately around HGVs and LGVs. 
  8. Recommendation: PSV and HGV operators should sign up to schemes such as FORS that have clear objectives to improve cyclists' safety through driver education, use of improved vehicle design to improve visibility, vehicle technology and incorporating cycle safety awareness into driver CPC.
  9. Recommendation: The Government should promote and encourage local authorities to adopt a default urban speed limit of 20mph on all single carriageway roads of less than 7.5m width within urban areas, and all residential access roads.

How many cycling accidents are there?

  1. In Great Britain, the number of cyclist fatalities reported in 2011 was just 107, the second lowest figure for 62 years, although this increased to 118 in 2012; the figures for London were 16 (2011) and 14 (2012). Cyclists represent about 2%[2]mode share across the UK although this is very variable by location, time of day and even between neighbourhoods. In 1950, with far fewer motor vehicles on the roads, cyclists made up 34% of traffic, accounted for some 12 billion vehicle miles per annum, and 805 cyclists were killed on the roads[3].
  2. The rate of fatalities compared to overall distance travelled has decreased over time although there have been periodic  increases within the overall downward trend (see Figure 1).
  3. The number of cyclists seriously injured in 2012 was 3,222. Unlike fatalities, the number of serious injuries to cyclists has risen since 2004 following a long period of decline.  Over 15,500 cyclists were reported as slightly injured. 

Figure 1: Rate of Cyclist Fatalities per billion vehicle miles (Source: DfT, Reported Road Casualties Great Britain and National Road Traffic Survey)

 

Figure 2: Rate of cyclist serious injuries per billion vehicle miles (Source: DfT, Reported Road Casulaties Great Britain and National Road Traffic Survey)


  1. While the figures for fatalities are likely to be accurate, it is widely acknowledged that many road accidents in which cyclists (and other road users) sustain slight and even serious injuries are not reported to the police, and therefore do not get entered into the official government statistics. There is also inaccuracy by the police in assessing the extent of injuries at the scene of the accident. Hospital admission data reveals higher numbers of cyclists involved in traffic collisions, but this excludes treatment for injuries at A&E departments and GP surgeries. A study in 1996 suggested that the STATS19 data for cyclist injuries should be factored up by 2.35 for serious injuries and 5.73 for slight injuries to reflect the true numbers[4]. Injuries sustained on cycle tracks away from the highway, and on other rights of way such as bridleways with no lawful access for motor vehicles do not appear in traffic safety reports, so it is not possible to accurately compare the safety of on and off-road cycling from any official government statistics.
  2. The numbers involved are shocking, and almost anybody who rides a bike on a regular basis can recount some tale of a near miss or an accident in which they were personally involved. Serious injuries are often life-changing for individuals and their families, with permanent disability, loss of income and sometimes implications for the mental health and well-being of all those affected.

How much cycling is there?

  1. One factor that has to be taken into account when considering the accident data is that the amount of cycling has increased significantly since 2000, and particularly in the last five years. Monitoring of cycling across the UK is not particularly accurate, and official figures have been slow to capture any discernible increase. There are many reasons for this:
  1. There are however a number of indicators that illustrate how much cycling is going on, and the magnitude of the increase. Within England, the Cycling Demonstration Town project between 2006 and 2009 included very accurate and comprehensive monitoring[5] before and during the project. This found that there was :
  1. In London, there have been significant increases in cycling on local roads and on the strategic route network.  This is most noticeable on the bridges over the Thames in central London where cyclists make up over 60% of the traffic flow during the weekday peak travel times[6]. Figure 3 illustrates the increase in cycling on the main road network since 2000, while Figure 4 illustrates the longer term trend across the whole city and shows how cycling really started to ‘take-off’ in 2000 and hints at the differences between central and outer London.

Figure 3: Trends in Cycle Flows on the TLRN – annualised and periodic indices

Figure 4: Long term trends in cycling across strategic cordons and screenlines in London, 24 hour, weekdays, both directions

 

  1. Outside London, the numbers of cycle journeys are smaller, but the trend is still upwards. It is estimated that around 1,000 to 3,000 cycle trips per day occur within each city centre of the six metropolitan Passenger Transport Executives (PTEs) in England[7]: Metro (West Yorkshire), SYPTE (South Yorkshire), Nexus (Tyne and Wear), GMPTE (Greater Manchester), Merseytravel (Merseyside) and Centro (West Midlands).  These cities have typically seen a growth in cycling of between  75% and 100% over ten years, although as with London, there has been a more marked increase in recent years. Smaller compact cities such as Cambridge, York and Oxford have some of the highest levels of cycling, but large cities that have invested in cycling since the 1990s, such as the hilly city of Bristol, have above average levels of cycling.
  2. Across the UK as a whole, approximately 3.1 billion vehicle miles were covered by cyclists in the year 2012[8]. Around 8% of the population (3 million people) cycle 3 times a week or more, in total 34% of the population (20 million people) say they cycle once a year or more[9].
  3. Taken in the context of the number of people cycling and the number of trips being undertaken, the numbers of cyclists killed and seriously injured is less alarming. There are 8 million cycle trips for each cycling death and 27 million miles cycled for each death – equivalent to over 100 trips to the Moon and back or 1,000 times around the world.  Cycling safety has substantially improved in recent years, although at a slower rate than for most other modes. Cycling was 61% safer in 2012 than it was in 2002 (per mile travelled)[10].

Where do accidents occur on the roads and who is involved?

  1. The information in the following sections is reproduced from the TRL study Collisions involving cyclists on Britain’s roads: establishing the causes.
  2. Almost three-quarters of all cyclist KSIs in Great Britain are injured on urban roads, while almost half of cyclist fatalities occurred on rural roads. This indicates that, while the frequency of injuries is greater on urban roads, their severity tends to be greater on rural roads.  Casualty severity is found to increase with the posted speed limit.
  3. Almost two-thirds of cyclist KSIs were at or near junctions where the risk of conflict is greater.
  4. Based on census journey to work data, there is no clear correlation between the amount of cycle commuting and the number of injury accidents. Some areas with low levels of commuting have relatively high levels of accidents and vice versa. It is likely that geography plays a part, for example in Cumbria and West Yorkshire there are low levels of cycle commuting but high levels of accidents. There is little choice of route in these areas with few alternatives to the low-lying main roads.
  5. 82% of cyclist KSIs were male, although males were only slightly more vulnerable than females when exposure (number of kms cycled from the NTS) was taken into account (i.e. most cyclists are male).
  6. There is a particular risk to young people. Children account for almost a quarter of cyclist KSI casualties, with the majority being between 10 and 15 years old. Cyclists aged 10 to 15 years are more at risk of injury (per km cycled based on the NTS) than any other age group.
  7. Adult cyclists aged 16 to 29 years are more at risk of injury per KM cycled (based on the NTS) than any other adult group. KSIs have increased sharply for the 30-49 year age group since the year 2000. Collisions involving cyclists aged 50 years old or more tended to have more serious outcomes than the younger age groups.

Is weather and darkness the issue?

  1. 80% of cyclist KSIs were injured in fine conditions on dry roads. 78% of cyclist casualties were injured in daylight. Collisions at night/in the dark were more likely to result in a fatality. Rural roads present particular difficulties as not only are the speed limits generally higher but the roads are often unlit.
  2. In Britain, a high proportion of cyclist KSIs occurred in the summer, with a particularly strong peak for child cyclists between May and September.
  3. A higher proportion of adults were injured on their bicycles during the working week rather than at the weekend and there were peaks in the morning (6am to 9am) and late afternoon (3pm to 6pm).
  4. Children were injured predominately in the afternoon (42% between 3 and 6pm).
  5. The daily proportions of child cyclists KSIs were roughly equal on Monday through to Saturday with a lower proportion on Sunday.
  6. This suggests that weather and darkness are not the most significant factors influencing cyclists safety. The strongest case for wearing high visibility clothing would appear to be on unlit roads where there are higher speed limits so that drivers can see cyclists well in advance.

What are the main causation factors?

  1. Most reported cyclist KSI casualties (83%) in 2005-07 involved a collision with another vehicle, usually a car or taxi (69%).
  2. The bicycle was generally hit by the front of the other vehicle. Over a quarter of fatal collisions involved the front of the vehicle hitting the back of the bike.
  3. Half of the cyclist fatalities involved a car or taxi and these collisions were divided almost equally between urban and rural locations.
  4. The main collision configurations involving a bicycle and car were the car turning right or left while the cyclist was going straight ahead and the cyclist making a right turn while the car was going straight ahead.
  5. When a cyclist was involved in a collision with a large goods vehicle, they were more likely to be killed (18% of fatal cycle accidents involved an HGV compared with 4% of serious accidents). Almost half of these fatal accidents in Britain occurred at an urban junction.
  6. When an HGV was involved, the main collision configuration was the HGV driver making a left turn while the cyclist was going ahead.
  7. Single cycle accidents (without a preceding collision with another vehicle) are less likely to be reported to the police. Nevertheless, 16% of cyclist KSI casualties for the period 2005-07 in the STATS19 database did not involve a collision with another vehicle. The majority of non-collision cycle accidents occurred away from junctions and a higher proportion of non-collision fatalities occurred in rural locations.
  8. Contributory factors are assigned by the police reporting officer and provide their view on the likely cause(s) of the collision. It should therefore be borne in mind that, as they are based on the subjective view of a police officer who was not present at the time of the collision, they may not necessarily represent the true cause(s). They should therefore be seen as giving an indication of likely causes.
  9. In over three-quarters of collisions where a child cyclist was seriously injured, the child’s behaviour was reported as the primary contributory factor for the collision. It is not clear whether this means children are more likely than adults to behave in ways that result in a collision or whether the police are simply more likely to attribute contributory factors to a child.
  10. Where contributory factors were assigned to the driver, ‘failed to look properly’ was by far the most commonly reported factor (56% of serious collisions), followed by ‘poor turn or manoeuvre’ (17%) and ‘careless, reckless, in a hurry’ (17%).
  11. Cyclist injury severity was greater when the following contributory factors were assigned to the driver: ‘impaired by alcohol’, ‘exceeding the speed limit’, ‘travelling too fast for conditions’ and ‘vehicle blind spot’ for HGVs.
  12. ‘Passing too close to the cyclist’ was judged to be a contributory factor in a quarter of accidents resulting in a serious injury involving HGVs, buses and coaches.
  13. The main contributory factors attributed to the cyclist included: ‘cyclist failed to look properly’ (43% of serious collisions) and ‘cyclist entering the road from the pavement’ (20% of serious collisions). ‘Cyclist entering the road from the pavement’ was most likely to be attributed to a child (34% of fatal and serious collisions involving children). ‘Loss of control’, ‘dark clothing being worn at night’ and ‘travelling too fast for conditions’ were more common factors in fatal collisions, being reported in twice as many fatal collisions as serious collisions.

Some of the implications of cycling collisions

  1. As well as the obvious personal distress and damage from accidents there is the cost to the NHS of treating casualties. All this is at a time when cycling is being positively encouraged through Active Travel and Health and Well-being policies and Travel Planning. The concern so often expressed in surveys and focus groups is that ‘there are not enough safe cycle routes to persuade me to cycle to work ....’. Thus cycling safety concerns and perceptions are directly hindering more sustainable travel.

What highway design features improve safety?

  1. It is difficult to get a completely accurate picture of the rate of accidents that occur away from the highway or even on ‘cycle facilities’ such as cycle lanes or tracks, or whether junctions are fitted with safety features such as advanced stop lines. The presence or absence of such features is not routinely recorded by police for collisions within the highway, and accidents away from the highway do not appear in police STATS19 data at all.
  2. Despite widespread investment in ‘cycle facilities’, there have been very few studies of the ‘before and after’ impact on safety except at sites where measures have been introduced to tackle a specific accident problem.  The ‘coding’ used for describing causation factors may also be misleading, for example ‘cyclist entering the road from a pavement’ may refer to a cycle track, a pedestrian crossing or a footway, and the cyclist may or may not have had legal right of way.
  3. Evidence from TfL, Sustrans and Cycling England demonstrates that the provision of cycle routes leads to an increase in cycling, for example, there has been almost a 200% increase in use of London’s Cycle Superhighway routes[11].  The number of cycle injury accidents has not increased by 200% so they have had a positive impact on safety. Increased use on this scale over a relatively short period does change the way in which certain features along a route operate, and this may lead to an increase in accidents at certain locations. For example, where an on-carriageway cycle route is replaced with a separate off-road cycle track, there are likely to be some accidents at side road crossings, no matter how well designed, as this is a new potential conflict point that didn’t previously exist. Cycle routes are provided to encourage active low carbon travel, to reduce cyclists journey times or to encourage healthy recreation, so improved safety is not always the purpose of new infrastructure and the safety objective in the design may be just to achieve a neutral impact rather than an improvement.
  4. Recommendation: The Department for Transport should undertake a thorough review of evidence from the UK and abroad to evaluate what design features of cycle infrastructure and general highway design particularly address the most common factors of injury accidents to cyclists, particularly at junctions where 75% of all accidents occur. This will help traffic engineers and planners to understand the risks and benefits of different junction layouts for cyclists and to design out the kind of conflicts that commonly result in injury accidents. It is especially important that government provides the Department for Transport with sufficient resources to commission this work and to promote the resulting guidance to ensure that it is widely understood and adopted. The evidence and subsequent guidances hould allay common fears expressed about cycling safety and support policies on Active Travel and Health and Well-being, and Travel Planning.
  5. Many accidents arise from poor judgement and decision-making by road users. Many drivers will have been trained at a time when cycle facilities were uncommon on UK roads and will have had no formal instruction about what the various signs and markings mean. For example, drivers commonly think that cyclists using a nearside on-road cycle lane are committing an offence of ‘undertaking’ if the cycle is travelling quickly next to a slow moving traffic lane. Similarly, many road users (drivers and cyclists) think that cyclists should always give way to motor vehicles. Cyclists may not be aware of blind spots, not only on HGVs but also those on buses, vans and other larger vehicles. There is some evidence to suggest that cyclists have misinterpreted HGV drivers’ intentions when they have to pull over to the offside in order to turn left, and that this is one of the reasons why cyclists may end up in the nearside blind spot of a left turning vehicle.
  6. Recommendation: Thorough training of all new drivers and CPC training for professional drivers should include some specific experience of on-road cycling so that they have a direct knowledge of what it feels like to cycle in traffic. Cycle training schemes such as Bikeability should include the experience of sitting in an HGV cab so that cyclists can fully appreciate the extent of blind spots.
  7. One initiative that is definitely proven to be successful in reducing the number of accidents and severity of injuries is lower vehicle speeds. Pedestrians and cyclists have hugely improved rates of survival at lower impact speeds below 30mph because they sustain fewer and less traumatic injuries, and almost all would survive an impact at 20mph. One of the first studies of pedestrian injury and car impact speed found that at 20mph there was a 2.5% chance of being fatally injured, compared to a 20% chance at 30mph[12]. Lower speeds also offer drivers sufficient time and opportunity to avoid collisions in the first place. Reducing the speed limit would therefore be particularly helpful in addressing injury accidents that occur when a pedestrian or cyclist makes an ‘unexpected’ manoeuvre into the path of a vehicle, a common factor on busy urban streets.  Existing 20mph areas have been effective in reducing injury accidents among children (who form 25% of cyclist KSIs), with a 67% decrease12 in injury accidents.
  8. Recommendation: The Government should promote and encourage local authorities to adopt a default urban speed limit of 20mph on all narrow single carriageway roads of less than 7.5m within urban areas, and all residential access roads.
  9. There is a particular and specific risk associated with left-turning HGVs. It is a known risk and while accidents are rare, those that do occur almost invariably lead to serious injuries or death. Due to the nature of our towns and cities it would be impractical to simply ban HGVs from central areas throughout the day, as this would lead to more danger at other times and potential noise disturbance in neighbourhoods in evenings, weekends or overnight.
  10. Recommendation: A national programme of publicity should be developed to make cyclists and car drivers aware of the ‘blind spots’ immediately around HGVs.  PSV and HGV operators should sign up to schemes such as FORS that have clear objectives to improve cyclists safety through driver education and vehicle technology and incorporate cycle safety awareness into driver CPC courses.

About the Authors

Richard Armitage, David Hurdle, Adrian Lord and Alex Sully are consultants with a long history of working on projects that aim to promote cycling and other modes of sustainable transport. They regularly work on these matters with national government departments, local authorities and professional institutes.

 

              Page 10

[1] TRL (2009), Collisions involving pedal cyclists on Britain’s roads: establishing the causes, TRL Report PPR445, TRL Ltd, Wokingham

[2] DfT, National Travel Survey 2012, Table NTS0301

[3] DfT. Reported Road Casualties Great Britian, Table RAS40007

[4] Simpson H F (1996). Comparison of hospital and police casualty data: a national study. TRL Report 173: Wokingham: TRL Limited.

[5] Cycling England 2010 Cycling Demonstration Towns Monitoring Project Report 2006-2009

[6] TfL, Travel in London report 5, Transport for London, 2013

[7] Cycling in the City Regions, Sustrans/PTEG 2011

[8] DfT, Table TRA0401, Road traffic statistics, 2013

[9] CTC Cycling Statistics, from CTC website accessed 02-01-2014

[10] CTC Cycling Statistics, from CTC website accessed 02-01-2014

[11] Barclays Cycle Superhighways. Evaluation of Pilot Routes 3 and 7, July 2011, Transport for London.

[12] Road Safety Briefing, 20mph Zones and Limits, RoSPA, Nov 2012