So first the case against helmets – from this video in the Guardian.
OK there is obviously a widespread debate on the use of bike helmets – I’ve read around and there are as many, possibly more, campaigning websites arguing against helmet use as those arguing in favour. Something like the impact on overall health and safety of bike helmets is a complex issue with many variables, so it’s hard to have an absolutely slam-dunk scientific conclusion either way – as with many issues, like cancer and tobacco notably, which was argued over for decades.
I also acknowledge I’ve been a bit too dismissive of the case against helmets and have been a little too sweeping in my judgements at times – so I apologize for that.
I am now familiar with most of the arguments on either side. We can have strong and passionate opinions – I do – and so do many who disagree with me. I think we can still have a reasonable debate – and the key issue is that we support our case with credible evidence and allow for some uncertainty in a complex case like this.
So now to the video.
To be fair the reporter does cite a doctor and a witness emphasizing the severity of injuries that can happen without helmets, and also says he personally wears helmets.
I think there isn’t much disagreement that there is a strong correlation between helmet legislation (requiring bikers to wear helmets) and a significant reduction in the number of deaths and severe head injuries on the road – where legislation has been implemented. There are a great many studies showing this correlation – I’ll confine myself to the studies cited below.
The main questions posed by those opposed to helmet legislation are these:
1 Does that reduction in deaths come at the expense of a significant reduction in ridership, thus negating any benefits by actually producing larger numbers of unhealthy adults and therefore, in the end, higher numbers of deaths from other causes (namely major diseases like heart disease and cancer)?
2 Does helmet legislation create a false sense of security for cyclists and thus produces more risky behaviour on the roads?
And these are indeed the core arguments of the video.
Dr Harry Rutter stresses that helmet legislation reduces bike use and thereby negatively impacts overall health since it reduces the number of people taking that form of healthy exercise – so therefore the net effect of helmet legislation (requiring people to wear helmets) is negative. The doctor sets up a direct conflict: the health benefits of cycling vastly outweigh the risks.
The problem here is that the video, beyond making the verbal assertions that cycling reduces ridership, does not actually cite any evidence in support of this claim.
So the three sources I’m going to cite are
1) the Cochrane Collaboration,
2) a major systematic review of studies by the Canadian Paediatric Society:
3) and more individually, this report from the universities of Ottawa and Manitoba in Canada:
I’ll begin with the Cochrane Collaboration, which is the international collaboration between governments, to provide evidence-based systematic reports on health issues – and is widely regarded as one of the most reliable and neutral international references. It is rather strictly and austerely focused on evidence-based medicine. It doesn’t consider the wider context of scientific plausibility and simply narrowly reads from approved studies – and it has a high benchmark for studies it considers to be of an adequate methodological level. So sometimes you don’t get the whole picture but you do get stringently solid and cautious conclusions.
Cochrane basically says that there is evidence that helmets protect from death and severe injury but concludes that there are no qualitatively good enough studies to say whether ridership is increased, reduced or stays the same:
“Bicycle helmet legislation appears to be effective in increasing helmet use and decreasing head injury rates in the populations for which it is implemented. However, there are very few high quality evaluative studies that measure these outcomes, and none that reported data on possible declines in bicycle use.”
The Canadian Paediatric Society, which refers to a whole range of issues related to helmets takes a slightly less stringent approach and concludes thusly: “There is also ample research indicating that legislation reduces risk of bicycle-related head injury.”
This study also reviews various studies on ridership – some of which saw a decrease and some of which saw no change – and concludes that there the evidence is a “wash” as the Americans say – ie that there’s no strong evidence either way. They argue that other factors (ie not to do with helmets) affecting ridership have probably not been accounted for in the studies.
Significantly, it points out a key fact about a study in Australia that is frequently cited by anti-helmet websites – the study did show a decrease in ridership – but the critics leave out an important and crucial point – that ridership among children was falling BEFORE the legislation was brought in and the decline in ridership among adults returned to pre-legislation level after two years, although adolescents remained lower. And for example, this study is contradicted by the Canadian study from Ottawa cited above that found no decline after legislation: “Helmet legislation is not associated with changes in ridership.”
So the evidence is mixed and does not point conclusively in either direction.
So the assertions by the video reporter and Dr. Rutter are not supported by the current weight of scientific evidence – and therefore their key argument, that reduced ridership actually means an overall negative affect, does not hold. And as to his suggested conflict between health and safety – there is no necessary conflict – why not promote riding AND helmet use and get the most of both worlds? ie a healthy population exposed to fewer severe accidents. There is no logical conflict there at all.
There’s also a question – raised by one of the studies – but so far unanswered: let us suppose for the sake of argument that helmet legislation did reduce ridership. In that case do we definitely conclude that people will exercise less or might people turn to other forms of exercise? We don’t know.
We then move on to the question of increased risk-taking and a familiar friend, Ian Walker of Bath University is cited in the video. Walker cites a lab study that indicated that people gambled more riskily in the lab if they wore a helmet as opposed to a cap.
But unless this theory is put to the test on the road with all its different conditions, it’s not reasonable to draw conclusions for actual behaviour on the roads from that. So to be fair Mr. Walker did go out on the road and tested another aspect of this – he measured whether cars drove closer to him if he was wearing no helmet or a wig and found that cars drove slightly more than three inches closer to him if he was wearing a helmet – thus concluding that other vehicles are more likely to take risk with you if you wore a helmet.
His controlled study group? A group of one. Himself (!) And while the fact that Ian Walker is an anti-helmet campaigner doesn’t necessarily disqualify him, I’m afraid the other more important fact – that his study is based on a sample of one does disqualify this study as reliable scientific evidence.
To draw conclusions on helmet use we need plenty of studies and plenty of systematic reviews. A single study on a single person doesn’t make the grade.
So back to the Canadian Paediatric Society that did review a number of studies on cycling and risky behaviour. And again it’s a wash: studies have come up with conflicting and various results:
“One ED-based study found no evidence of a relationship between use of safety equipment and reported bicycling behaviour (cycling fast, taking chances) or injury severity among children injured in a variety of activities, including bicycling. Another found that helmeted bicyclists experienced less severe nonhead and non-neck injuries. Injury outcome-based studies involving all age groups have found that helmeted bicyclists experienced more frequent and severe nonhead injuries compared with nonhelmeted bicyclists. However, one European study found no relationship between bicyclist commission of a traffic violation and helmet use. The issue of risk compensation remains unresolved.”
So the second leg of the argument – increased risk-taking – is also not supported by the overall balance of evidence. And there’s an additional point to be made here: what if helmets did increase risky behaviour? Unless it was substantially proven that serious nonhead injuries increased – as one study says, but others don’t – should we be concerned anyway, since the net result is a significant reduction in deaths and severe head injuries? If that holds, the point becomes moot.
But at any rate, to repeat, there is no balance of evidence so far that risky behaviour increases.
Incidentally, the reporter’s claim that “major studies and scientists in Canada, New Zealand and Australia, concluded that helmets do not improve overall safety”– is clearly false, since I’ve cited at least two major studies from Canada above, and at least one significant Australian report, concluding that there was a decline in injury, is cited by the Canadian review.
Some other points made in the video: cycling is “not as dangerous as people think” it says – referring to the UK. But the UK is one of the very safest countries in the world and has made major headway in reducing road deaths over the past fifteen years (as I discovered while making a documentary on road safety for the UN). So what about all the other countries where road deaths are higher – and sometimes much higher?
The example of the Netherlands is interesting – since they do indeed have low levels of cyclist deaths and low use of helmets. As the video acknowledges, the Netherlands has spent 40 years improving road safety – including major improvements to infrastructure – and this reminds me of the mantra of the experts I interviewed for the documentary: road safety depends on the three Es: Enforcement, Engineering and Education. ie infrastructure improvements to roads and vehicles; stricter law enforcement and public information campaigns to reduce risky behaviour.
I fully agree and it makes sense – those in favour of helmets, like me are of course in favour of all other aspects of improving road safety – there is no contradiction here at all. Yes, all aspects of road safety need to be improved – and if helmets lower deaths and severe head injuries, as the evidence shows, then helmets should be very much part of that overall picture.
So the Netherlands has done well in creating a much safer environment with lower deaths – but one wonders if that low level of cyclist deaths might be lower still of helmet use was increased:
“More than 800 cyclists per year sustain head and/or brain injury in a collision with a motor vehicle. In addition, more than 2500 cyclists per year suffer from head and/or brain injury after a crash or a fall not involving a motor vehicle (bicycle-only crash). In 86% of the cases the head injury of a cyclist is (also) brain injury. 2010 – 14.”
Later in the video, cyclist Chris Boardman argues we should be free to choose and we should not impose “my will on you.” Yes, as a default, one doesn’t want to impose on people – but obviously when there’s a public health implication, we do precisely that: by requiring, for instance, that parents vaccinate their children, or that car drivers wear seat belts and don’t talk on handheld mobile phones – laws imposed on them that most of us heartily agree with, since it protects them as well as others. If the evidence is strong that helmets reduce death and severe head injury – and the evidence shows a strong correlation – with no proven reduction in ridership and no proven increase in risky behaviour – there’s a strong case for legislation.
And finally, the video turns to this argument: “The greatest number of head injuries is among motorists – in the UK…so maybe they should wear helmets.”
Seriously guys? Of course the number of head injuries is higher because there’s vastly greater use of motor vehicles than bikes! That’s a schoolboy prank of an argument, by using an absolute number instead of a proportional rate.
And if we are talking comparative rates? Cyclists are fifteen times more likely to be killed in the UK than car drivers per miles by road: