BMA's helmet stance questioned as USA safety authorities drop key helmet claim

In an editorial in the current British Medical Journal (BMJ – i.e. the magazine of the British Medical Association, BMA), co-authored with risk professor David Spiegelhalter, Goldacre openly questions the BMA’s support for laws that would ban people from cycling without helmets.

They point to a new study (also published in the BMJ) showing that Canada’s helmet laws have made no detectable difference to cyclists’ safety. This is significant, as the BMA's original decision to support helmet laws was based on a study which appeared to show that Ontario's helmet law had not reduced cycle use.  However it subsequently emerged that this law was not enforced, hence it had made no difference to helmet use either.  In short, an unenforced helmet law does nothing whatsoever. It therefore remains the case that enforced helmet laws significantly reduce cycle use, with all its health and other benefits.

Even with very optimistic assumptions as to the efficacy of helmets, relatively minor reductions in cycling on account of a helmet law are sufficient to cancel out, in population average terms, all head injury health benefits."

Prof Piet de Jong
Macquarie University (Australia)

Meanwhile, the USA’s National Highway Traffic Safety Administration (NHTSA) recently dropped the widely-parroted claim that helmets could prevent 85% of head injuries, following a successful challenge by the Washington Area Bicyclist Association.  WABA pointed to more recent research indicating that the effectiveness of helmets is far lower than the Seattle study had suggested.

At the same time, my colleague Chris Peck has been at the international Velo-City cycling conference this week. You need to go into slide-show mode to see his presentation properly. In it, he talks about true and false comparisons between cycle safety and other forms of transport, and the reasons why it’s not worth promoting helmets to try and avert the relatively low risks of cycling, given the much greater harm this would almost inevitably do to public health.

Two distinct questions: for the individual and for society

In their editorial, Goldacre and Spiegelhalter rightly identify two very distinct questions in the helmet debate:

1. At an individual level, “What is the effect of wearing a helmet?” and
2. At a societal level, “What is the effect of a public health policy that requires or promotes helmets?”

On the first question, CTC’s stance is entirely agnostic - in that respect we are neither 'pro-' nor 'anti-helmet'.  The evidence for the effectiveness of helmets is complex and contradictory, providing as much support for those who are deeply sceptical of helmets as for those who swear by them.  Whether or not it is worth wearing a helmet may depend on what type of cyclist you are, or what type of cycling you do.

The idea you have to dress up like a bloody spaceman to ride a bike is just completely potty."

Steven Norris
Ex-transport minister and former chair of the National Cycling Strategy Board

However, the idea that there are different types of cycling or cyclist is often overlooked in helmet research. For instance, the now discredited 85% statistic came from a study of cyclists hospitalised in Seattle, comparing the injuries of those with and without helmets. Early critiques of that study pointed out that the helmet-wearers were more likely to be white, affluent and to be cycling in parks, while the non-wearers were more likely to be from lower-income ethnic minority groups riding on busy streets. Another critique explained how the same methodology and data could be used to show that helmets were 77% effective at preventing injuries to parts of the body other than the head!

Interestingly, a recent Dutch report (see Google translation) seems to prove the same point, only in reverse.  It shows that over 13% of cyclists hospitalised in the Netherlands hospitals were wearing helmets at the time, even though less than 1% of Dutch cyclists wear helmets. Does this prove that helmets are actually really harmful? Of course not. The explanation is that almost all of the helmet-wearers were riding racing bikes or mountain-bikes, whereas most dutch cyclists ride sit-up-and-beg utility bikes. In other words, the difference isn't due to the helmet itself, but the type of cycling being done by those who chose to wear helmets (i.e. mountain-biking and racing).

There is also the famous study which found that drivers leave less space when overtaking cyclists with a helmet than those without. Another found that cyclists with helmets had 14% more injuries per mile travelled (see p28).  It is unclear whether this is because people who are more risk-prone are more likely to wear helmets, or whether the wearing of a helmet makes you more risk-prone.  Still, there is good evidence that people do ‘risk-compensate’ (i.e. act less cautiously) when wearing helmets.

In short, any 'case-control' research study which compares hospitalised cyclists with and without helmets - based on the assumption that the two groups are alike in every respect apart from the use of helmets - has to be regarded as fundamentally flawed. The same methodology has produced seriously misleading results on hormone replacement therapy, MMR vacines and other issues.

Yet the delegates to the BMA’s 2005 ‘Annual Representatives’ Meeting’ fell for the pro-helmet evidence. Or more accurately, just over half of them did, in a very close vote.  After the BMA’s Chairman and Board of Science had proposed the motion calling for helmet laws, two BMA representatives who are also CTC members spoke against it.  In a debate lasting just a few minutes, they came incredibly close to persuading the audience that the loss of cycling's health benefits would far outweigh any possible injury savings - the votes had to be counted.

Whilst every cyclists' death is obviously tragic, the total numbers are mercifully low enough that it's not worth trying to prevent them by telling people to wear helmets - because the resulting loss of life due to people NOT cycling would almost certainly far outweigh any possible reductions in fatal cycling injuries, however effective helmets might be.

Since then, the BMA has made mercifully little effort to persuade the Government to act on its misguided policy. Yet its hastily taken conference vote continues to be quoted as if this represented authoritative medical opinion.

That brings us to Goldacre and Spiegelhalter’s second question. Should governments and other public bodies promote or require helmet use? In CTC’s view, the answer is an unequivocal “No”.

CTC's main piece of evidence for this stance is a study by an Australian statistician named Piet de Jong.  He showed mathematically that it would take only a very small reduction in cycle use for helmet laws or promotional campaigns to cause more premature death (due to obesity, heart disease etc) than could possibly be prevented by cycle helmets. As de Jong says: "Even with very optimistic assumptions as to the efficacy of helmets, relatively minor reductions in cycling on account of a helmet law are sufficient to cancel out, in population average terms, all head injury health benefits.”

Using de Jong's mathematical model, CTC estimates that you reach the point at which helmets would inevitably shorten more lives than they could possibly save – even if they were 100% effective at preventing head injuries (which they obviously aren’t!) – if cycle use fell by as little as 2-3% (see Appendix A of our helmet evidence briefing).

In short, whilst every cyclists' death is obviously tragic, the total numbers are mercifully low enough that it's not worth trying to prevent them by telling people to wear helmets - because the resulting loss of life due to people NOT cycling would almost certainly far outweigh any possible reductions in fatal cycling injuries, however effective helmets might be.

As former Transport Minister Steven Norris said years ago: "The idea you have to dress up like a bloody spaceman to ride a bike is just completely potty".  Meanwhile London Mayor Boris Johnson is now seeking to "delycrafy" cycling in the capital.

It really is time for the BMA to get the message that the best way to maximise cycling's health benefits is to promote it as a safe, normal, stylish and enjoyable activity which anyone can do in whatever clothes they feel like wearing.

* * * *

For more information see:

* CTC briefing on helmets - summary and full briefing (677KB PDF)

* CTC overview of helmet evidence - summary and full briefing (422KB PDF)