Paper given by Thomas Krag at the Velo-city conference in Copenhagen, August 1989

Cycling is generally considered being unsafe compared to other means of transport.

The current knowledge on the safety of cycling often does not take into account the number of cyclists, thus leading to wrong conclusions. Also, the representation of cyclists in the statistics is very sparse, having only about 7% of the total number of injured.

While having a much greater risk than the motorist of being light and medium injured by traffic accidents, the risk of being killed as a cyclist is however, when the time spent in the traffic is used as a measure, less than the risk for a car user.

Several methods have been developed to calculate the benefits of the motor traffic. For the cycle traffic similar methods have not been used, leading to general misunderstanding, that cycling is neither safe or beneficial for the traffic economy, compared to other means of transport.

Development of methods calculating the benefits of the cycle traffic, under here the increased mobility of the non-motorised groups, the improved urban environment and the increased health by cycling, are necessary.

By taking into consideration the general risk of living, including the increased risk of certain diseases induced by inactivity, it can be calculated, that cycling is a far safer means of transport than trains, buses and cars.

Thus, alone by setting up the relevant figures for the improvement of health gained by cycling, conclusions opposite to the traditional ones on safety are gained.

This should not serve as an argument for not making the traffic conditions safer for cyclists, but as an argument against not using the bicycle or campaigning for the bicycle as long as the conditions are not better than at present.

Cycle helmets is often proposed as a way to increase safety for cyclists. Being very unattractive to the general attitude of cycling as a nice and normal behavior in Denmark, the helmets should be introduced with the greatest care. Aggressive campaigns for helmets, making the cyclists beleive that cycling is very dangerous, will have the adverse effect of intended, making fewer use the bicycle and thereby increasing the overall risk of living. Attempts to make the helmet compulsory must be most strongly dissuaded by the same argument.

The scope of this paper is to draw attention to some fundamental facts and considerations about the safety of cycling.

Safety, in the headline denoted as an Achilles' Heel for Cycling, could also be denoted a two-egged sword.

Thus, safety for the existing cyclists often plays the most important role when decisions of building cycle facilities, e.g. cycle lanes, are taken.

But safety, in its traditional conception, is as often used as argument for not building cycle lanes, because investigations has disclosured, or at least postulated, a decrease in safety when such facilities have been build.

Also, so-called solutions, probably hoped to increase the safety for cyclists, makes the cyclists' way less passable. This might reduce the number of accidents in a specific site, but as long as this reduction is due to a smaller number of cyclists, the rest of them choosing another route or another means of transport, it must be said to be, at the best, theoretical of nature.

Finally, when only acting towards the existing cyclists, wrong conclusions are made. Several times it has been found, that a new cycletrack or cycleroute did increase or re-direct the cycletraffic substantially (ref 1,2). However, as long as the number of accidents is playing the most important role, the potential for cycling will not be taken into account, and no solutions will be made along the stretches, where the conditions are experienced to be far too dangerous by all but the most indurated cyclists.

Cycling has a number of positive impacts to the user as well as to the society, e.g.

• it is healthy
• it improves mobility
• it is cheap
• it is non-polluting.

Among the negative impacts are

• safety of the user and
• safety of pedestrians

far the most important.

We are living in societies, where figures are given an enormous importance. If you can substantiate something with hard figures, your probability of breaking through is increased dramatically.

Therefore, giving cycle traffic the support it deserves, methods for evaluation of all the positive aspects are imperative.

For the motor traffic such methods has been developed years ago and used extensively since. In Denmark, a publication (ref 11) is issued regularly by the Road Directorate. It contains theoretical prices of saved time for the motorists and for traffic accidents, and has been used to justify expensive roadbuilding projects giving increased average speed for the motorists. Everything is calculated into money, and one killed can in this way be equated to 15 years jof saved time.

Not one single figure in the publication, except of course the price of an accident, is involving cyclists.

Even more misrepresentative is the traditional statistics of traffic. Usually, the cycle traffic is not given one single remark, and the figures of cars, buses, trains, ferries and planes are adding up to 100%, as if the cycle traffic did not exist at all (ref 3,4, among many others). When it comes to the accidents, however, the cyclists turn up with disquieting figures.

On this basis it is not surprising, that the cycle traffic is treated as a problem rather than an endowment to the society.

Before going into how to evaluate, in figures, the positive aspects of cycling, some remarks to the traditional conception of safety should be given.

It is split up into two, namely

• the actual safety, based on the number of accidents
• the experienced safety, based on the extent, to which the users feel, they are safe.

Generally, having no figures, the experienced safety will be a good guideline when judging about the actual safety. Findings have shown, however, that when choosing about two different solutions, the one that is felt as the most safe sometimes gives rise to more accidents than the less safe alternative. The interpretation is, that the road users react to their feeling of unsafeness by driving more carefully. It should be underlined, that this is generally not true, and that planning for the feeling of unsafety is tantamount to planning against all but the most skilled cyclists, leaving children and elderly without the mobility, that the cycle can offer them.

Evaluation of the feeling of safety can be easily achieved by asking a number of road users, cyclists as well as non-cyclists, about what they think of a given road or crossing. Though this can serve as a good guideline, it is only seldomly used, and also regarded as an un-scientific way of gaining knowledge of the safety.

Much more popular among traffic-planners are evaluations of the actual safety.

As definition of the risk, as the reciprocal of the actual safety, serves the following formulas.

For a crossing or similar:

For a stretch:

For general comparisons:

The formulas works with cyclists as well as other road users. Often the risk of fatalities, severe injuries and light injuries are calculated separately.

A recently issued plan of action for increased traffic safety in Denmark (ref 5) is based on calculations on the effect of various measures. The number of fatalities is multiplied with 34 and then added to the number of other injuries to get one single figure for comparison reasons.

When evaluating the safety for cyclists, a number of problems arises in spite of the seemingly simplicity of the formulas.

One problem is the number of accidents. The police should report any accident to the central statistical authority in Denmark. They also do so, to some extent, but regarding cyclists only the number of fatalities is a reliable figure. The number of severe accidents is, say, relatively reliable, though under-represented, while the number of light injuries reported by the police amounts less than one tenth of what actually happens.

The knowledge about the actual number of accidents is achieved from investigations at hospitals casualty departments (ref 6, 7). It has been found, that the total number of accidents reported with cyclists is only 7% of the real number (ref 7).

The figures achieved from the official statistics are, in spite of this fundamental falseness, used extensively in almost all investigations concerning safety of cyclists.

Another problem is the denominator in the formulas, the number of cyclists. The only reliable way to achieve this figure is to count the cyclists. This is also done, sometimes, but very, very often investigations are carried through without evaluating the actual traffic, which is instead considered as being at a sustained level.

The faults resulting from this are obvious: The changes, which leads to increased traffic, are automatically evaluated as less safe, than they actually are. Similarly, a too good safety measure is achieved at the facilities, that are of such inferior quality, that the cyclists avoid them.

While trains and buses are quite safe, the relevant means of transport to compare with is the car. To compare the risk, one has to take the only reliable figure from the accident statistics, the number of killed persons. This leads to (ref 8, 9, 10):

Fatal accidents in Danish Traffic, 1988

As is seen, the risk of getting killed is scarcely double when riding a bicycle compared to driving a car. When taking into account the time spent on/in each means of transport, usually 2-5 times as long per km on a bike as in a car, the car turns out to be a more dangerous place to sit than a bike saddle.

This table is by no means encouraging. It shows, that the risk of being injured is 50 times as high when biking as when driving a car. As a mean, 136,000 km by bike gives rise to one casualty. Figures can of course be produced, when taking into account the time or the number of jorneys, which to some extent levels out the differences. It should also be mentioned, that there is probably some under-representation of the motorist injuries as well. But there is still an immense number of casualties of medium and low severeness with bicycles, compared to the car.

The answer of the headline-question thus is yes, cycling is unsafe, when regarding safety as a question of the number of casualties in traffic and comparing to other means of transport.

The previos mentioned investigations from hospitals casualty departments (ref 6, from 1986) gives the following figures regarding the number of casualties divided up on the counterparts in the accidents:

What is surprising in relation to the bicycle is the large proportion (72%) of injures with no counterpart in the traffic. Also, the bicycle is involved in almost one quarter of all casualties with pedestrians. This gives some support to the often rised critisism from pedestrians towards cyclists, though the severeness of the accidents with bicycles generally is far below what happens when cars and pedestrians coincide. Fatal accidents between cyclists and pedestrians is thus less than one per year, but a total of 141 pedestrians was killed in 1987 (ref 16).

One can conclude, that cycling is not dangerous to other road users, and that the extensive use of the car is responsible for ppractically all the fatal accidents as well as for the major proportion of the severe injuries.

These conclusions are banal and should be well known, but is often forgot in the discussion of traffic safety.

Now lets turn to the positive aspects, where the healthiness, the improved mobility, the cheapness and the non-polluting qualities of the bicycle was mentioned.

What regards the latter ones, they could rather easily be calculated into economic terms. Mobiblity for young and elderly cyclists could, for instance, be compared with the price of alternative motorized transportation. The same comparison could be used concerning the cheapness, and probably large annually savings could be calculated.

What regards pollution, this is an obvious risk factor as well. The only problem is, that nobody can tell what the risk of living in a polluted city is. However, governments in a number of countries have decided to do something about it by setting up low limits for the pollution from cars, these limits demanding special equipment (catalytic converters) to be built into the cars. Exactly the same result - less pollution - could be achieved by a change of the modal split. A catalytic converter may cost, say, 10,000 danish crowns and reduce pollution from a car to about the half level. The value of changing the modal split would thus, in pollution economic terms only, be equal to an investment of 20,000 crowns per car, that is fully replaced by bikes. Far more sophisticated calculations are of course necessary to establish the right bicycle-traffic economic figures, but a rough idea of how it should be done is now, I hope, given.

I will quickly return to the question of safety, hoping for apologises for speaking about economy. And the previous mentioned quality of the bicycle, increase in health, is an example of something, that can be calculated into risk- or safety terms.

Epidemiological studies have shown, that physical exercise, by lowering the risk of coronary heart disease (heart attack), as a mean did increase the length of life by a group of persons by 1-2 or even more hours per hour of physical exercise during their lifetime (ref 13). Cycling is an example of such physical exercise. Quite often cycling is the alternative to physial inactivity. 42% of the population regularly take exercise, and 29% regularly use the bicycle (ref 17), indicating that this is the case.

A rough calculation on the health gained by cycling in Denmark can thus be made as follows. Take the number of kilometres annually travelled by bicycle. This figure in Denmark is 5.6 billion. Given a mean speed of 15 km/hour this will require a time of 373 million hours equal to 43,000 years. Even if we take the lowest figure (1 extra hour of life per hour of cycling) these 43,000 years equals 600 full lifetimes (70 years each).

81 cyclists were killed in Denmark in 1988. This will at maximum represent 50 lost lifetimes, taking the ages of the killed cyclists into account.

A very simple calculation now can be made concerning the effect of cycling in Denmark. It is:

It is seen, that the health aspect is more than ten times as important than the safety aspect, and that there is an enormous net positive effect of cycling.

Of course also the non-fatal injuries could be taken into account. This will, lead to comparisons of economical nature, but taking the factor 34 used elsewhere and the number of injured cyclists of about 2.000, it will only about double the figure of lost lifetimes, still leaving a clearly positive net-effect of cycling.

A number of other objections can of course be made. But the figures clearly indicates, that there is a net positive effect of cycling, also when the number of accidents and the lost lives are taken into account.

Cycling is mentioned as a possible physical exercise to increase the health in the Danish government programme on prevention of illness (ref 15), but encouraging cycling as a means of transport has not come up as an idea in this subtext.

Considerable discussion is going on concerning the use of bicycle-helmets. An investigation made the 17th-23rd of October 1988 (ref 18) gave the following answers:

Answers to question:
"Do you think use of cycle-helmets should be compulsory?"

A large proportion of the Danes thus rejects the bicycle helmet to be compulsory. Note, that the negativeness increases with the use of bicycle.

One remark should be made about helmets. By no doubt the good helmets reduces the consequences of accidents, though they do not prevent one single accident. But taking into account the positive aspects of cycling - without helmets - one should be very careful to make a strong campaign for the use of helmets.

Thus, if the helmets could reduce the number of fatal accidents by 30% - which must be at maximum, if everybody used them - about 27 fatal accidents would be saved, which is of course a good thing.

But if the campaign had the effect - which is very probable - to lower the amount of cycling in Denmark by just 10%, the negative effect on the health of the Danes would override the positive gained by the use of cycle-helmets.

1): Cykel/knallerttællinger før og efter anlæg af 10 cykelstier på hovedlandevejsstrækninger (Counts of cycles and mopeds before and after establishing of 10 cycletracks along major roads). Vejdirektoratet, Økonomisk statistisk afdeling, 1983.

2): Cykelruten i Århus (The experimental bike-route scheme in Aarhus). Vejdirektoratet, Sekretariatet for Sikkerhedsfremmende Vejforanstaltninger, Rådet for Trafiksikkerhedsforskning, Institut for Veje, Trafik og Byplan, oktober 1987.

3): Statistisk Tiårsoversigt (10-year statistical survey). Danmarks Statistik (issued annually).

4): Yearbook of Nordic Statistics. Nordic Council, Nordic Statistic Secretariat (issued annually).

5): Færdselssikkerhedspolitisk handlingsplan (Plan of action for traffic-safety in Denmark). Betænkning afgivet af Færdselssikkerhedskommissionen. Betænkning nr 1157, December 1988.

6): Personskader opstået ved trafikulykker behandlet på skadestuen, Odense Sygehus 1986 og 1987 (Casualties from traffic accidents, treated at the Casualty Department of Odense Sygehus). Ulykkesgruppen ved Odense Sygehus, 1988 og 1989.

7): Medical officer Poul Grøn, at a conference titled "Bikelanes - the Right Solution in Towns?" arranged by the Danish Ministry of Traffic, April 27th, 1987.

8): Samfærdsels og Turisme (Traffic and Tourism, updated statistics on traffic and accidents, issued about 20 times a year). Danmarks Statistik

9): Persontrafik i 1975, 1981 og 1986 (TU 86) (Passenger traffic in Denmark 1975, 1981 and 1986). Trafikministeriet, Vejdirektoratet DSB, Energiministeriet og Rådet for trafiksikkerhedsforskning, 1988.

10): Trafikindeks for bil- og cykeltrafik (Traffic index for car- and cycle-traffic). Vejdirektoratet, Økonomisk Statistisk afdeling (issued at least quarterly).

11): Trafikøkonomiske enhedspriser 1986 (Unit-prices of Traffic-economy), Vejdirektoratet, Økonomisk statistisk afdeling.

12) Fysisk Aktivitet og sundhed, konsensusrapport (Physical activity and health, consensus-report). Danish Medical Research Council and Danish Hospital Institute.

13) Ralph S. Paffenbarger, Jr: Contributions of epidemiology to exercise science and cardiovascular health. Medicine and science in sports and exercise 2005, 1988.

14) Sundhed og sygelighed i Danmark 1987 (Health and sickliness in Denmark, 1987). Dansk Institut for Klinisk Epidemiologi (DIKE), 1988.

15) Regeringens forebyggelsesprogram (The danish Government programme on prevention of illness). Sundhedsministeriet, Komiteen for Sundhedsoplysning.

16) Færdselsulykker 1987 (Traffic Accidents 1987). Danmarks Statistik, 1989.

17) Befolkningen og kulturen (an analyses of the activities of the population). Socialforskningsinstituttet, 1989.

18) DMA/Dansk Marketing Analyse, undersøgelse offentliggjort i Aarhuus Stiftstidende 20.10.1988.