- K Gilbert,
- M McCarthy
- Correspondence to: Dr McCarthy.
- Accepted 7 February 1994
Objective: To determine the characteristics of cyclists and vehicles involved in fatal cycling accidents.
Design: Analysis of data routinely collected by police for each accident from January 1985 to December 1992 and held in a national master file (Stats 19) by the Department of Transport.
Setting: Greater London, which comprises inner London (12 boroughs and the City of London) and outer London (20 boroughs).
Subjects: 178 cyclists who died (78 in inner London and 100 in outer London; age range 3-88).
Main outcome measures: Associations between characteristics of cyclists, type of vehicle involved, and place of accident.
Results: Motor vehicles were involved in 173 deaths. Heavy goods vehicles were involved in 75 deaths (30/100 (30%) in outer London and 45/78 (58%) in inner London); cars in 74 (54/100) (54%) in outer London and 20/78 (26%) in inner London); light goods vehicles in 12/178 (7%); and buses in 6/178 (3%). Thirty five of the people who died were children aged <=16. Female cyclists were especially at risk from heavy goods vehicles in inner London (22 deaths), while male cyclists were especially at risk from cars in outer London (50 deaths).
Discussion: Cyclists who died in urban areas are more likely to be adults than children. In inner London, in relation to their traffic volume, heavy goods vehicles are estimated to cause 30 times as many cyclists' deaths as cars and five times as many as buses. Until the factors leading to this excess risk are understood, a ban on heavy goods vehicles in urban areas should be considered.
Cycling is healthy and should be encouraged, but the dangers to cyclists need to be reduced
Most studies of cycling accidents and their prevention have described the victim rather than the cause of the accident
Almost all cyclists' deaths in London are due to collision with motor vehicles, especially heavy goods vehicles
Primary prevention of accidents involving cyclists could be achieved by reducing dangers from high risk vehicles
Cycling is environmentally clean and protects against coronary heart disease.1 Many people, however, are frightened of cycling in towns and cities because of motor vehicle traffic.2 In Britain in 1991 there were 242 deaths of cyclists, representing about 1 in 20 of all road deaths.
Targets for reductions in road casualties and road accident deaths have been set by the Department of Transport and the Department of Health respectively.3,4 So that realistic policies to reduce the number of deaths of cyclists can be formulated, information is needed on how these deaths occur.
Because deaths of cyclists are relatively rare, studies based on data from accident and emergency departments have included few or no such deaths.*RF 5-7* The findings on the circumstances of cyclists' deaths cannot be extrapolated readily to cover other accidents involving cyclists because different severities of injury are probably caused by different sorts of accident. Studies of cyclists' deaths in the United States,6,8,9 Australia,10,11 Sweden,12 and New Zealand13 have described the characteristics of cyclists and their injuries, but few have considered the type of motor vehicle involved. We used data for London in a preliminary investigation of the circumstances of deaths of urban cyclists in Britain.
In Britain road accidents resulting in injury must by law be reported to the police. For each reported case an accident report book is completed routinely by a police officer. Details from completed report books are coded by the police information processing unit for storage on a national master file of the Department of Transport known as Stats 19. Data for London are available to the public from the London Research Centre (81 Black Prince Road, London SE1 7SZ). Data in Stats 19 for accidents involving cyclists comprise the time, weather, age, and sex of the cyclist and a brief account of the circumstances; no details are stored about injuries sustained, use of a bicycle helmet or lamps, or alcohol concentrations of cyclists. Data on other people involved but not killed in the accidents are incomeplete.
We assembled data from Stats 19 on fatal accidents involving a cyclist in greater London from January 1985 to December 1992. We divided data into two categories: inner London (comprising 12 boroughs and the City of London), which has a high density of buildings and had a population of 2 265 815 in 1991; and outer London (comprising 20 boroughs), a more suburban area with a population of 4 021 611 in 1991. Vehicles involved in these fatal accidents were classified in a standardised abbreviated form - for example, “heavy goods vehicle” indicated a goods vehicle >1.524 tonnes unladen weight and “light goods vehicle” indicated a goods vehicle <=1.524 tonnes unladen weight.
Routine statistical information - for example, national and local statistics on accidents3 - is based on data held in the Stats 19 file. Random variation among police officers in recording accidents is reduced because the report books are used regularly, are clearly laid out, and only a few items have to be completed. A more important problem of systematic bias is that some road accidents, especially those involving cyclists, are not reported at all.6 Police reports are probably completed more fully for accidents involving a death than for less serious accidents.
We sought to determine rates of use of different types of road transport in London to provide denominators. In the most recent survey in London, in 1991, 60 000 (2%) households were interviewed about their use of transport.*RF 13a* Results are available from the London Research Centre and the Department of Transport.
Accurate information on the exposure of cyclists to other road traffic is lacking, but some assessment can be made from data in the London Traffic Monitoring Report.14 This report includes the results of traffic counts across three cordons - central, inner, and boundary cordons. The central cordon, which surrounds an area within a radius of 3 km from Aldwych, is monitored annually; but the inner cordon, which corresponds roughly to the boundary of the former London County Council, and the outer cordon, which corresponds roughly to the boundary of the former Greater London Council, are monitored every three years. The inner cordon separates approximately the inner and outer London boroughs. Traffic counts are taken for six minutes one to four times an hour between 0700 and 1900 at any point at which a road crosses a cordon. Twenty four hour counts are taken on a sample of roads. The volume of traffic and the type of vehicle are recorded.
Characteristics of deaths
In all, 178 deaths of cyclists were recorded by the police in greater London from 1985 to 1992; 78 occurred in inner London and 100 in outer London. Data on age were not available for two of the male deaths in inner London. Most (134) of the deaths were of males; 28 (36%) of the deaths in inner London were of females compared with 16 (16%) in outer London. The distribution of deaths by age in inner London (range 8-80 years) was unimodal, dominated by adults aged 20-39, but in outer London (range 3- 88) it was bimodal, the largest proportion being of children and adolescents and of older people (table I). Thirty five of the cyclists who died were aged <=16, 101 were men aged >16, and 42 were women aged >16.
Almost all (173) of the deaths involved a motor vehicle (table II). Four cyclists, however, died without a vehicle or pedestrian being involved and one died after a collision with a pedestrian. The main motor vehicles involved were heavy goods vehicles and cars. Heavy goods vehicles were involved in 75 of the deaths (30 (30%) of the deaths in outer London but and 45 (58%) in inner London). Cars were involved in 74 deaths (54 (54%) of the deaths in outer London but only 20 (26%) in inner London). Few deaths were associated with light goods vehicles (12), public service vehicles (buses and coaches) (six), or other motor vehicles including motorcycles.
Twenty two (28%) of the cyclists who died in inner London were women who were in collision with a heavy goods vehicle, compared with only eight (8%) in outer London. Fifty (50%) of the deaths in outer London were of male cyclists in accidents involving a car, compared with 17 (22%) in inner London.
According to the surveys for 1986-90 reported in the London Traffic Monitoring Report 1992,14 bicycles account for on average 0.7% of all road traffic at the boundary cordon, 1.1% at the inner cordon, and 2.2% at the central cordon. In greater London medium and heavy goods vehicles together (classified as heavy goods vehicles by us) account for about 7% of traffic, light goods vehicles 10%, buses and coaches 2%, and cars 75%. We found, however, that from 1985 to 1992 these vehicles were involved in 42%, 7%, 3%, and 42%, respectively, of cyclists' deaths. In inner London medium and heavy goods vehicles account for about 6% of traffic, light goods vehicles 11%, buses and coaches 2%, and cars 72%. These vehicles were involved in 58%, 8%, 4%, and 26%, respectively, of cyclists' deaths.
Stats 19 provides only a limited description of the manoeuvres that contributed to the accidents, but it is possible, tentatively, to group some of the accidents together according to the manoeuvre. In 30 fatal accidents the vehicle turned left across the cyclist's path, and in all but one of these accidents the vehicle was a heavy goods vehicle. In a further four deaths both the vehicle and the cyclist were apparently turning left together.
Sixteen collisions occurred when the cyclist was on the nearside of a vehicle that was going straight ahead, 22 cyclists were hit from behind, and eight were said to have swerved into a vehicle's path. Of the 35 children aged <=16 who died, 14 were struck by a vehicle after cycling off the pavement.
Four cyclists died after being hit by vehicle drivers opening their doors, and two were killed while cycling across zebra crossings. Alcohol intoxication in the driver was recorded as the main cause of the accident in only one of the 178 accidents.
Variations with time
The annual number of deaths fell from 25 to 18 between 1985 and 1992. An exception to this trend, however, occurred in 1989, a year which showed a sharp increase in inner London; this increase might be related to more cyclists being on the roads because of the dry summer and strikes on the underground. Each year, and for 1985-92 as a whole, peaks in the number of deaths occurred in the summer and autumn. In outer London, accidents were evenly distributed over the week, but in inner London few fatal accidents occurred at weekends. Daily variation in the numbers of deaths was bimodal, with peaks between 0600 and 1000 (the main peak in inner London) and between 1200 and 1800 (the main peak in outer London).
Most fatal accidents occurred in dry conditions (151) and most occurred in daylight (half an hour before sunrise to half an hour after sunset) (137).
Other fatal accidents involving cyclists
In addition to the 178 accidents in which cyclists died, there were eight fatal accidents involving bicycles in which the cyclist did not die. Five pedestrians who were crossing the road died in collisions with cyclists. Two motorcyclists died after hitting cyclists, and one car driver had a cardiac arrest after being in collision with a cyclist.
Three issues arising from this analysis of deaths of cyclists in London require emphasis. Firstly, most deaths were of adults rather than children. Secondly, cyclists rather than drivers died in collisions between cyclists and vehicles. Thirdly, there was a particular danger, especially in inner London, from heavy goods vehicles.
Risk to adults
Most cyclists dying in London are adults. The London area transport survey of 1991 indicated that 45% of the cyclists in London who cycled at least once a week were children aged 5-16*RF 13a*; Stats 19 indicates that children aged <=15 accounted for 21% of all reported accidents involving cyclists in greater London15; and our study indicates that only 20% of cyclists who died were aged 3-18. The rate of accidents and deaths involving child cyclists may not, therefore, be higher than that involving adult cyclists.
Studies of cycling accidents have tended to focus on child rather than adult cyclists.*RF 7-9,11,16,17* In the United States cycling accidents were more common among children than adults until the 1990s,8,9,18 but researchers and policymakers may also perceive that children are most obvious users of bicycles whereas adults use motor vehicles. A study of deaths of cyclists in northern Sweden showed that 22% occurred in children but that the median age at death was 60.12 Although cycling accidents involving children remain important, more attention should be given to the circumstances and prevention of cycling accidents involving adults.
Strategies for cyclists
New policies are needed to meet the targets for a reduction in road accidents set by the Department of Transport and the Department of Health. Some public bodies - for example, the Royal Society for the Prevention of Accidents,19 the Child Accident Prevention Trust,20 and the Parliamentary Advisory Council for Transport Safety21 - propose policies that restrict cyclists rather than motor vehicles. But a policy of more protection - for example, through the compulsory use of bicycle helmets - does not address the issue of the source of danger in accidents in which the injuries are severe or fatal.22,23 In contrast, other bodies - for example, the Confederation of British Industry,24 the British Medical Association,1 and the Cyclists' Touring Club25 - suggest that safety would be increased through restricting motor vehicle traffic, with more traffic free environments and safe bicycle tracks.
Motor vehicles, especially cars, constitute a hazard to health because of the possibility of accidents, because of air pollution, and because they inhibit exercise. Priority should be given to the use of bicycles in urban design and regulation.
Dangers from heavy goods vehicles
Our study draws attention to the particular danger to cyclists from heavy goods vehicles. In Britain heavy goods vehicles are recorded as being involved in only 3% of all cycling accidents but being involved in 18% of deaths.3 In London the impact of heavy goods vehicles is greater: during the period under review such vehicles were involved in almost a third of cyclists' deaths in outer London and almost two thirds of those in inner London. From the estimates of vehicles use in London, the risk of heavy goods vehicles being involved in accidents in which cyclists die in inner London can be estimated at five times that of buses, 14 times that of light goods vehicles, and 30 times that of cars.
The limited information available on vehicle manoeuvres showed that left turns by heavy goods vehicles were particularly hazardous. Turns are inevitably more common in inner urban areas, but buses in inner London were involved in far fewer accidents in which cyclists died than heavy goods vehicles. The danger from heavy goods vehicles, especially when compared with the danger from buses, may reflect either aspects of their design or less competence and awareness on the part of the drivers. The drivers of buses and heavy goods vehicles may also differ in the way they take risks.
The Transport and Road Research Laboratory studied the involvement of heavy goods vehicles in road traffic accidents in 1976.26 It estimated that the number of deaths due to a cyclist being run over by the rear wheels of the vehicle could be reduced by the installation of guard rails. Although most heavy goods vehicles now have such rails, no published study has evaluated the effectiveness of these rails. Since October 1988 all lorries over 12 000 kg gross vehicle weight have had to have close proximity mirrors on the top of the nearside doors, angled down to reflect an image of the area around the front wheel. The effectiveness of these mirrors has not been evaluated.
The higher proportion of women cyclists who die in accidents involving heavy goods vehicles in inner London cannot be explained satisfactorily. A closer study is needed of the circumstances of collisions between buses and cyclists and between heavy goods vehicles and cyclists, with an examination of the characteristics of drivers as well as those of vehicles. The results should be reviewed and recommendations for prevention implemented. Until then, where orbital roads exist around towns and conurbations there are grounds for banning vehicles of over 1.524 tonnes unladen weight in urban areas.
We thank the London Accident Analysis Unit of the London Research Centre for supplying the data on which this study was based.