Casualties on cycle paths

Below is the substance of a letter I sent to John Grimshaw of Sustrans in June 1998. John had written to me to ask on what evidence I continued to state that cycle paths are often dangerous. He claimed that there had not been one reported (sic) serious accident on the Bristol to Bath cycle path in 15 years.

The letter is reproduced here not as an attack on Sustrans, but in a genuine wish that the matters it raises be considered and discussed more widely. I also received much encouragement from people who had already seen the letter to make it available more widely.

Dear John,

According to TRL, less than 3 per cent of casualties on cycle paths are reported. Our experience in Milton Keynes is that the figure is lower than this. Even one of last year's fatalities did not get recorded as a cycle accident. Reported crash statistics are almost irrelevant when it comes to assessing the safety of cycle paths.

Similarly, fewer than one in six cycle crashes involves a motor vehicle, yet this does not necessarily imply less serious injury. There was no other vehicle involved in two of the six crashes in which off-road cyclists have died in Milton Keynes over the past decade. The most common causes of cyclist injuries are poor surfaces and the inability to control a cycle properly. Providing for cyclists away from roads does not, therefore, address the great majority of crash circumstances. To the contrary, off-road paths generally have poorer surfaces, worse discipline, fewer traffic signs and markings, and numerous other factors which make control of a cycle more difficult and injury more likely.

In Milton Keynes over the past decade, there have been six deaths to off-road cyclists against just one (a child) on the alternative road network, where main roads largely have a 70mph speed limit and there are large roundabouts at all major junctions. Even when account is taken of relative distance cycled, the death rate for the cycle paths is significantly greater than for the roads. The same is the case for serious crashes. Even the official statistics, for all their shortcomings, have included 23 serious injuries on cycle paths over the decade, against 21 on main roads; the actual path total is much higher. Milton Keynes Hospital has treated some 2,000 cyclists over the decade for injuries relating to the cycle paths, and less than one-third that number for roads covering a larger geographical area.

The point about these figures is that I have not conjured them up to support an hypothesis, and certainly not to berate the efforts of Sustrans. They have been collated quite independently by hospital staff and the police, and are simply a reflection of the reality of cycle path safety in Milton Keynes. This reality has not changed significantly for nearly 20 years, nor is there any noticeable difference between the safety of older paths and those introduced recently.

Although I have previously been open to the argument that Milton Keynes is a 'special case' when it comes to cycle path safety, I no longer think this to be so. The main difference between Milton Keynes and other towns is that we have a longer history of cycle paths and more of them, and that research has been carried out over a long period to track trends, and this includes non-reported crashes. I know of no similar research in other towns. I do know, however, that in two of the closest towns to Milton Keynes – Aylesbury and Bedford – there have also been serious crashes on cycle facilities in recent years. And there have, of course, been a number of off-road fatalities in other parts of Britain too.

As I believe you know, I carry out work from time to time investigating cycle crashes throughout the UK and give expert evidence in litigation. As little as four years ago, around 90 per cent of this work (outside Milton Keynes) involved road crashes. Since that time the pattern has been transformed, and now around 90 per cent involve cycle facilities, many of them being Sustrans or similar paths. The number of road cases has remained roughly static; the change in relative proportions is accounted for entirely by the considerable growth in injuries on cycle facilities, probably tracking the growth in such provision. Most of the cases I deal with involve serious injury.

I do know, therefore, that there have been serious injuries on some of your routes. I also believe the incidence of cycle path casualties to be out of all proportion to the number (or use) of such facilities in the highway infrastructure of Britain. Indeed, a common factor to a great many cases is that upon investigation, by myself or others, one invariably finds that the injury we are investigating is just the tip of an iceberg. For example, before my involvement in one case, on one of your more popular paths, the solicitor had already discovered from the local health centre that they regularly treat many injuries to cyclists using the path on busy days. A number of these injuries have had long-term consequences.

You will understand that I cannot detail personal cases of injuries on Sustrans routes (the more serious will probably be sub-judice for 5 years or more), but I believe that Sustrans would be wise to treat safety much more seriously, and in particular to stop proclaiming the routes as inherently safe, which typically leads people to take less care and be more vulnerable.

You refer to European experience where, you suggest, casualty rates are 5 to 10 times less than in the UK. This does not match with the research of which I am aware, and I would be pleased to consider any evidence you can give me. I suspect that you are again only referring to reported casualties.

In Sweden the University of Lund deduced a relative risk of 3.4 for roadside cycle paths compared with using the road, but rising to 11.9 under particular circumstances. Cycle lanes were better, but still 10 to 15 per cent more dangerous than nothing at all.

Several German police forces are campaigning to repeal the laws that require cyclists to use cycle tracks in order to improve cycle safety. A West Berlin police study found that cyclists are 4 times more likely to have an accident on a cycle path than on the adjacent road, and the likelihood of serious injury or a fatality is similarly increased. I understand that there are up to 10 deaths a year on Berlin paths.

Research in Belgium and Austria has drawn attention to the poor safety record of cycle paths, as have some papers associated with the Dutch Masterplan. I have spoken to Dutch engineers who have told me that they continue to build paths only because of public insistence; from a safety point of view they would often do differently. More than 300 cyclist deaths occur each year in the Netherlands; most are associated with cycle paths.

You may be aware of the US and Canadian risk statistics presented at this year's Transportation Research Board Annual Meeting. The US research showed that Sustrans-type off-road paths are more than twice as dangerous per mile cycled as major roads, and that shared footways (often used on Sustrans routes) are up to 24 times more dangerous. The Canadian statistics from Toronto and Ottawa reveal an injury rate twice as high on cycle paths as on roads, with shared footways being up to 7 times more dangerous. I was interested to read on the Internet last week that La Route Verte has apparently conceded that its paths might lead to more casualties, and that it is to put more effort into developing cyclist skills.

I am at present looking further into Dutch and Danish casualty statistics. I think it true that the Netherlands and Denmark are safer places to cycle than the UK (although I suspect the ratios usually suggested are exaggerated), but the connection between this and cycle facilities is tenuous. I have heard of one study that suggested that the safest parts of the Netherlands are those with fewest facilities! In Denmark a before-and-after study of 105 new cycle paths with a total length of 64km showed that on average cyclist casualties increased by 48 per cent following the introduction of the paths (cycle use did not go up significantly). Furthermore, car drivers, moped riders and pedestrians also suffered more accidents due to the new paths, with an overall rise in casualties of 27%.

It seems to me that other factors are primarily responsible for the better Dutch and Danish casualty rates, and that these might be even better with fewer cycle facilities. The principal factor would seem to be the much greater number of cyclists (individual risk generally decreases with cyclist numbers, although this is not always true on cycle paths), which in turn leads to a climate where drivers take more care. Getting to that situation from a lower user base (as in the UK) may be more of a problem, but I have yet to find evidence that most cycle facilities help rather than hinder this process.

I know that many people in the cycling community have tended to give Sustrans the benefit of the doubt with regard to its promise that the National Cycle Network would lead to more and better cycling. The feedback I now get from individual cyclists and CCN groups is that many people are becoming disillusioned. In Milton Keynes, the Cycle Users' Group agreed to support the NCN (albeit not for funding) due to the promise that this would raise standards. This has not been the case, and your initial route is to an extremely low standard with appalling danger spots that should never have been condoned. The route includes many features that have a proven history of leading to serious injury. If this standard is typical of what you will accept, I do not see how you can expect the NCN to have a better safety record than that of the Milton Keynes Redways.

I know that other people are concerned that parts of the NCN are introducing new difficulties, even for experienced cyclists. Yet the promised growth in cycling seems to be mainly limited to car-borne leisure journeys, which is certainly not sustainable. Nationally there has been no increase in general cycling for the past few years despite a significant increase in various types of cycle facility.

In my view, this is because alongside the promotion of facilities has also been the promotion of fear. The message has gone out that roads are just too dangerous for cycling, and that cycling is not possible without special (especially Sustrans-type) routes. The fear of traffic that this has generated has had an effect much more devastating than traffic itself. It has scared thousands of people who might otherwise cycle from doing so. And the more separate facilities that are provided, the more we get stuck in the vicious circle of increased fear, declining skill and narrower horizons.

One manifestation of this is the tremendous growth over just a few years in the number of people who now cycle on pavements. I do not go along with the excuse that this is simply a consequence of traffic. Most of the pavement cycling I see is alongside roads with sufficiently little traffic to pose no realistic difficulty to anyone. My perception is that the growth of pavement cycling has more closely mirrored the growth in cycle facilities (especially cycle/pedestrian paths) than traffic growth.

At the same time the special routes are not delivering the safe havens that have been promised. I fear that the myth of cycling danger on-road is being turned into a reality where there are segregated cycle facilities. Poorly designed infrastructure and declining skill levels are resulting in bad cycling habits that put people more at risk. For example, wrong-side riding on roads was almost a unique American problem until recently, but I see it more and more now in Britain, particularly near cycle paths.

Sustrans has often cited the fact that Dutch cyclists sometimes leave the ferry at Harwich and find traffic so difficult to deal with that they go back home! Interestingly, this problem is not experienced by cyclists arriving from France, Spain or the USA. Proficiency in using roads on a regular basis is essential to maximise safety, and to maximise one's cycling horizons. I would not like to see Britain on the slope down to Dutch levels of cycling competence.

I know that I and others of similar feeling are sometimes slighted by Sustrans as laggards against progress, although no-one has cared to ask for my position on these matters. In fact I have much admiration for Sustran's enthusiasm, and I have personally enjoyed a number of your routes. I see them, however, as essentially 'challenge' routes for relatively experienced cyclists, by and large unsuitable for people who have not yet acquired the skills, and do not have the self-discipline, to deal with the less predictable hazards that are often present. I certainly do not see them as 'safe' routes, nor as any kind of stepping stone by which to acquire the skills and confidence that people so desperately need in order to cycle more widely and more often. I do think that Sustrans, through its publicity, is guilty of scaring many people from cycling in places where they might readily gain competence.

Over the years I have met a considerable number of people who have been hurt whilst using cycle facilities and I know both the long-term suffering that this can lead to and also the disillusionment with cycling that typically results. On these grounds alone it would be wrong for me not to give the benefit of my experience in drawing attention to the many hazards I see in current cycle facility design. One of my greatest fears, which recent events suggest to be well-founded, is that Sustrans will accept the lowest standards and compromises with safety in order to complete the NCN by its target dates.

I have noticed a considerable maturing of the views of cycle campaigners in recent years based, no doubt, on experience in the real world. With this has come increased recognition of the limitations of cycle facilities and of the need to integrate rather than segregate cyclists in traffic. I was delighted only a few weeks ago to learn that one CCN group, whom I had given up as a lost cause in terms of cycling policy, has by its own admission made a U-turn away from segregation and in favour of a minimum facilities approach. I sense that Sustrans is becoming more and more isolated in the direction it is taking, and that this will lead to increased friction at both national and local levels. There is already a feeling in many places that Sustrans too often runs rough-shod over the views of local cyclists. Is it really in the best interests of any of us for this to continue?

I am copying this letter to the CTC, and I will circulate it within CCN and to other parties, as I feel it important for these matters to be considered as widely as possible. We all need to learn from experience and to agree a common strategy for cycling that will lead more people to cycle competently and safely, wherever they wish to travel, and in a way that is sustainable and commands respect as an equal to others on the road. It is my view that present strategies are doing none of these things.

As ever, if you wish to continue the dialogue or provide alternative evidence that should be considered, I would be more than pleased to hear from you.

With regards,

John Franklin

Cycling Digest Infrastructure Index