Updated in response to comments:
Summary:
========
As a cyclist with a keen interest in safety issues I read the Department for Transport's Road Safety
Research Report No 30, 2002 (RR30)
(
http://www.roads.dft.gov.uk/roadsafety/roadresearch/bicyclehelmets/index.htm) with some dismay.
My unhappiness stems not from the accuracy or otherwise of the report itself, although its
acceptance without comment of figures which have since been publicly corrected is not a good sign,
but from the fact that its stated limitations have clearly not been taken into account in the uses
to which its conclusions have already been put.
RR30 states in its Project Brief: "As part of its policy to improve the safety of pedal bicyclists
DfT promotes the use of bicycle helmets, particularly amongst children. However, there is a wealth
of published evidence both for and against promotion and compulsory use of bicycle helmets, and DfT
requires an independent objective critique of the most up-to-date evidence on the efficacy of
bicycle helmets. It is also important to have up-to-date information on legislative measures
internationally and their impact on bicycling activity levels and safety." (my italics)
The report then goes on to consider the epidemiology of bicycle accidents without properly
addressing the evidence against promotion or compulsion of bicycle helmets. Some of this evidence is
alluded to, and some is misquoted, but no attempt is made to pursue the real concerns raised by the
most widely studied introduction of compulsory helmet use, that in Australia. While it is surely
fair to claim that helmet promotion raises helmet use, this is not the question at issue. Having
narrowed the focus to exclude the main arguments against helmet compulsion and include the main
argument in its favour, the report unsurprisingly concludes that the evidence supports helmet
promotion. In short, by addressing only half of the stated need, the report reaches a biased and
probably invalid conclusion.
Discussion:
===========
This report focuses quite deliberately on the effect of cycle helmets once a crash has happened. It
excludes, equally deliberately, time-series and whole-population studies which indicate the overall
effect of helmet promotion and compulsion. The report is therefore quite clearly of no value at all
in setting public policy which, by definition, affects whole populations. It excludes the
considerations which might make it useful in that context, so it may at best be viewed as an
interesting sideline to the real debate: how to reduce casualty levels among cyclists.
One could take a cynical view and conclude that the report was commissioned deliberately to include
only that data which is supportive of helmet promotion, and exclude that which casts doubt on such a
policy. Certainly that would appear at face value to be the underlying agenda of the authors of the
report, who are declared helmet advocates. Having read that there are those in the DfT who have it
in mind to bring forward helmet compulsion legislation as soon as wearing levels are high enough to
justify it, the cynical view looks even more attractive. John Franklin's thorough and credible
rebuttal of the report is available here:
http://www.cyclenetwork.org.uk/papers/broader.pdf
In its conclusions RR30 references Unwin (1996) and his criteria for helmet compulsion:
1. There must be a high level of scientific evidence that bicycle helmets are effective in reducing
the rate of head injury to bicyclists.
2. The benefits to society and others of mandatory bicycle helmets must be convincingly
demonstrated, mandatory bicycle helmets cannot be justified simply to protect individual adult
bicyclists.
3. There must be widespread agreement, ideally by a large majority, that the potential benefits of
compulsory bicycle helmets outweigh the infringement of personal liberty and other disbenefits.
4. There must be good evidence to suggest that compulsory helmet wearing would not make the public
health benefits of increased levels of bicycling significantly harder to obtain.
RR30 concludes that criterion 1 is now met, but this conclusion is fatally undermined by, among
other evidence, the increased injury rates seen in Australia post compulsion. TRL report 286 raised
similar concerns regarding helmet promotion. This is a proper subject for further research.
Quality of the review
---------------------
The review has been undertaken by a group who acknowledge that they are in favour of helmet
promotion. This is not, in itself, a barrier to realistic appraisal of the arguments. There are,
however, some disturbing indications of possible bias in the report. For example, in Section 5 the
authors review the situation in Victoria, Australia, and make the following observations: "While the
increased rate of helmet wearing and reduced level of bicyclist casualties noted above is
impressive, it is worth noting that it is possible that some of these changes were influenced by
decreases in exposure. Following the introduction of the bicycle helmet law the estimated adult
bicycling exposure increased marginally..."
In fact the measured (not estimated) level of adult cycling declined by nearly 30% in the year
following compulsion, and remained suppressed a decade later according to other studies. The rate of
head injuries increased by over 20% in the two years following compulsion. The total number of head
injuries reduced, but by less than the reduction in cycling overall. This suggests that cycle
helmets are, in this case, either dangerous or irrelevant to cyclist safety. Given these known
facts, the above is a serious misrepresentation of the Victoria experience.
TRL report 286 raised the possibility that a major effect of helmet promotion is to deter cycling.
It is noted that this requires further investigation. The need for such investigation, regarding the
fundamental issue of the probability of a crash happening in the first place, is clearly more
important than measures designed to mitigate the results once a crash has happened.
RR30 also states that:
- The pro-bicycle helmet group base their argument overwhelmingly on one major point: that there
is scientific evidence that, in the event of a fall, helmets substantially reduce head injury.
- The anti-helmet group base their argument on a wider range of issues including: compulsory
helmet wearing leads to a decline in bicycling, ‘risk compensation’ theory negates health
gains, scientific studies are defective, the overall road environment needs to be improved.
- The way in which the debate has been conducted is unhelpful to those wishing to make a
balanced judgement on the issue.
This misrepresents or misunderstands the nature of the debate. Those who are pro helmet are a
relatively small, vocal and homogeneous group, including many doctors. There is no significant
anti-helmet group, but the group which is anti /compulsion/ (and by extension has profound
reservations regarding /promotion/) is indeed diverse. It encompasses road safety professionals of
international renown such as John Franklin, Mayer Hillman and Robert Davis. It includes the UK's
oldest and largest national cycling organisation, the CTC. It includes Steve Norris, the
Government's "cycling supremo." With such a diverse group, the range of concerns is naturally wide.
But there is surprising uniformity on the issue of deterrent effect, because such studies as have
been published, few of which set out to prove such an effect, make this conclusion inescapable.
If "the way in which the debate is conducted is unhelpful to those wishing to make a balanced
judgement on the issue," it is probably because the two sides have evidently not agreed on the issue
in question. For the anti compulsion lobby the issue is reduction of cyclist injuries; for the pro
compulsion lobby it is reduction of the severity of those injuries which do happen. As long as the
means of achieving the latter is inimical to the former, agreement is impossible.
Then again, perhaps the issue is how you define balance. Many groups have come to make what they
perceive as a balanced judgement: balanced, that is, in that it weighs the evidence for helmets
against the balancing evidence agin them. If the result is inconclusive, surely that is indicative
more of the complexity of the issue than any failure in the debate. Yes, the research on bicycle
helmets stubbornly refuses to conform to the notions of those concerned primarily with injury
mitigation. Scientific method is awfully prone to dealing mortal blows to neat theories, and proving
that the world is more complex than we would like it to be.
Those of us who are against helmet promotion and compulsion are, by and large, well-informed. We are
also, for the most part, cyclists. It is in our own best interests to understand the issues, to
judge the risks and benefits. Far from being anti-helmet, many of us wear helmets as a matter of
course; few will allow their children to ride without helmets; and fewer still will challenge the
right of others to forego the notional protection of a helmet should they so wish. We are all aware
of the realities of risk compensation, a theory so blithely dismissed by the pro lobby, to the
extent its being placed in quotes by the authors of RR30. If risk compensation does not exist, what
then explains the observed rise in pedestrian, cyclist and rear-seat passenger fatalities following
compulsory seat belt legislation in the UK? Given the availability of safety cages, crumple zones,
ABS, airbags, high-performance tyres and road surfaces - surely driver fatalities in the UK should
be, to all intents and purposes, a thing of the past?
Factors improving cyclist safety
--------------------------------
There are three factors which are acknowledged to affect, above all other considerations, the safety
of cyclists on the roads. They are, in no particular order, driver behaviour, rider behaviour and
the number of cyclists using the roads.
The new policy, allowing only helmeted cyclists to be shown in official publications, will
inevitably reinforce the inaccurate public perception of road cycling as a dangerous activity.
Interestingly I believe it will also make it impossible to show Steve Norris on his bike, since he
is an avowed non helmet wearer.
Helmet promotion of any kind, however well-meaning, has two direct effects: it deters people from
cycling on the roads, and it by extension it encourages the use of more dangerous alternative
provision such as shared-use cycle paths. It should be unnecessary to point out that the safety of
cyclists on the road improves as the number of cyclists on the road increases, but this kind of data
(widely available in credible peer-reviewed research) has clearly been excluded from consideration
while drafting this policy. It should equally be unnecessary to remind you that shared-use provision
is associated with greater risk of injury for the cyclists, greater risk of conflict with
pedestrians, and undermines the legal position with respect to cycling on footways - a practice
which is undesirable for many reasons.
Helmet promotion has an insidious effect on motorist behaviour. It portrays the cyclist as one who
is knowingly engaging in a dangerous activity, rather than one who is engaging in a safe and legal
activity which the behaviour of drivers may make dangerous. It is well known that the largest cause
of car vs. bicycle crashes is error by the driver of the car, and the most likely outcome is injury
or worse to the cyclist. It should therefore be a matter of pressing concern to redress this balance
by a mix of stronger penalties and better public information aimed at careless drivers. Instead we
have the recent Think! campaign proposing defensive riding by users of powered two wheelers as a
solution for inattentive car drivers. Helmet promotion comes out of the same bag: an unwillingness
to deal with the problem at source, so shift the focus onto the victim. It is not hard to see why
Governments are unwilling to deal with the issue of careless drivers; the way that motoring
organisations throw their collective toys out of the pram when speed enforcement is increased is
sufficient proof that they believe motorists' "right" to drive badly overrides the rights of other
road users - after all, the motorist is nice and safe in his steel cage, so anybody venturing onto
the roads without such protection is asking for trouble.
Nor is it surprising that drivers fail to make the connection between everyday acts of aggression
and discourtesy, and the risks these present to users of benign modes. There have been numerous
incidents where drivers have caused the death of cyclists through bad driving and have attracted
penalties so light as to amount to an endorsement of such behaviour. A driver who kills a cyclist
and is allowed to retain their driving licence, despite a history of driving offences, can hardly
fail to conclude that cyclists rate alongside dogs, badgers and foxes as road-kill: regrettable but
not something over which one should lose sleep.
Cyclist behaviour, too, is subtly influenced by helmet wearing. One of the reasons why time-series
and whole-population data are so important in assessing the effects of helmet promotion is that data
based solely on injury rates after crashes fails to take into account the question of whether the
wearing of helmets makes the crash more likely. There is no doubt that the most survivable crash is
the one which doesn't happen in the first place. The mechanisms for increased accident rates among
populations with high helmet-wearing rates are imperfectly understood, although there is strong
evidence to support risk compensation. The majority of helmet promotion material focuses on the
danger from cars, but cycle helmets are of very limited benefit in car v bike crashes, which almost
invariably exceed their design parameters. Helmets give protection primarily in single vehicle
accidents at speeds up to 12mph. This covers many of the situations where bikes are in use, but
absolutely not the one which is purported to be the greatest concern. A cyclist hit by a car at
speed is likely to suffer injuries to many parts of the body, and the impact speed is unlikely to be
within the ability of the helmet to give meaningful protection. The solution to this kind of injury
is to stop the car hitting the cyclist in the first place.
The most compelling data, that from Australia, demonstrates quite clearly that not only has cycling
declined significantly since compulsion was introduced, but accident and injury rates have
increased. This may well be due to reduced numbers of cyclists on the roads, but until the reasons
are fully understood any form of aggressive helmet promotion remains risky.
Government policy
-----------------
The Government's attitude to cycling appears almost schizophrenic at present. The talk is about
promoting cycle use, migrating journeys to benign modes and making the roads safe for all users. The
practice shows reasonable understanding of the bicycle as a leisure item, includes some promotion of
cycling as an alternative to the school run, and leaves the bulk of the dangers and inconveniences
suffered by transportational cyclists either untouched or worsened. It is as though the objectives
are set by cyclists and the policies by drivers frustrated by school traffic and having to wait a
few seconds before overtaking the occasional bike.
In short, then, the DfT appears to have taken a research report whose conclusions are readily
identified as flawed, extended it into an area which is beyond the scope of the report, and as a
result introduced a policy which is likely to increase cycle crashes in order to promote something
which is not proven to reduce injury when such crashes occur.
If the DfT is serious about cyclist safety and promoting cycle use they should do as Steven Norris
suggests and show ordinary cyclists going about their ordinary lives wearing ordinary clothes. A
publication which shows in the background a City gent on his Brompton, suited, with his briefcase on
the carrier and his hair unperturbed by his stately progress, will have more positive effects than
any unrealistic photo shoot with smiling happy helmeted families riding mountain bikes on
well-graded paths.
In the mean time I suggest that whenever any member of the Department's staff is considering the
issue of cycle helmets, they adopt instead a term which is gaining in currency among those who have
a more realistic impression of the science which underlies their design and use. The term is
"polystyrene foam deflector beanie."
Personal perspective:
=====================
As a year-round transportational cyclist (who wears a helmet, by the way) I also offer below some of
the things which would make my life safer - and which would address the concerns of those cyclists I
know who are hovering on the brink of using their bikes for daily transportation:
- An effective campaign against competitive and aggressive driving. The thing which is most
likely to lead to my death on the road is not the wearing or otherwise of a helmet but the
driver whose journey is so urgent that they cannot bring themselves to wait for a safe place
to overtake. For the most part, of course, their destination is the back of the traffic queue
ahead, but this does not deter them from taking my life in their hands and overtaking
dangerously. I encourage you to observe the behaviour of drivers towards cyclists at a
typical traffic calming pinch point. These pinch points are almost invariably not narrow
enough to actually reduce speed (this requires them to be too narrow for LGVs and buses), so
drivers are accustomed to negotiating them without slowing. When they see a cyclist on
approach to such an obstruction their reaction is quite often to try to pass before the
obstacle. Since very few drivers realise that cyclists can easily be travelling at speeds of
25mph and more on a level road this judgement is often flawed. The result is that the car
gets halfway past and cuts in to avoid the bollards. Perversely, those who have actually
given the cycle sufficient space when passing cause more danger as they must cut in more
sharply, and usually overshoot, often into the path of the bike. All of this, and the classic
"left hook" where the driver overtakes the cyclist and immediately turns left, would be
reduced if people drove more calmly -if the driver's first thought were not "how can I pass
this slow-moving object" (bikes are rarely seen as vehicles) but "what is the safe way of
negotiating this situation." Perversely, the focus on speeding as pretty much the only aspect
of driver behaviour challenged by Government campaigns has resulted on an even greater
determination to maintain the illusion of progress, at whatever cost to other road users.
Campaigners against speed enforcement have validated the view that the driver's transient
personal convenience is the overriding priority, and the driver may safely be left to judge
the extent of his impact on other road users. Since few drivers even consider cyclists when
making these judgements, the danger of this viewpoint is evident.
- A proper public information campaign on overtaking techniques, and including overtaking in the
Driving Test. Overtaking is a most dangerous manoeuvre and is routinely handled extremely
badly. The principle is simple: drop back, wait until there is a clear space ahead, check
mirrors, accelerate and pass, and when well past pull in. The reality is that drivers
overtaking cyclists will do one or more of the following: sit within a few inches of the
cyclist's back wheel, look at the cyclist not the road ahead, pull out without signalling or
checking mirrors, start to pass and only then look ahead to see that there's a car coming the
other way, move left to avoid the oncoming vehicle (squeezing the cyclist) and pull in when
the cyclist is out of their peripheral vision (typically when the car is about 1/3 of the way
past). And of course the four drivers are following on the bumper of the overtaking vehicle
will follow through despite the fact that by now they are on a blind bend with more traffic
coming the other way, each one moving out a few inches less than the one in front. Rules
138-143 and 188 of the Highway Code appear on the face of it to be Terra Incognita to the
average motorist. This must be remedied if transportational cycling is to become a more
pleasant activity.
- Widely available expert training for adult cyclists. I know the CTC are working to promote
adult cycle training, but we have been trying for over a year now to find any group locally
who will run an open cycle training course which my wife, a relatively new cyclist, can
attend. She rides to school with the children several times per week and would value some
expert tuition. One would imagine that this is precisely the kind of thing which the
Government would wish to support, but there is a near-complete absence of training. I would be
prepared to run a course myself if there were a nationally accredited instructors' training
system. There isn't. From this year I will be running National Cycling Proficiency training at
the school where I am a governor - there is not training available for me in advance; it's a
matter of "sitting with Nellie" and hoping that "Nellie," in this case fortunately a well
respected local cyclist, is suitably expert.
Guy
===
** WARNING ** This posting may contain traces of irony.
http://www.chapmancentral.com (BT ADSL and
dynamic DNS permitting)
NOTE: BT Openworld have now blocked port 25 (without notice), so old mail addresses may no longer
work. Apologies.
. Or you could ask for a public commitment that the flawed report won't be used
