Issue 15 / 23 April 2012

IT’S important to make a distinction between the efficacy of bicycle helmets (how well they protect the head) and the effectiveness of mandatory helmet legislation (designed to reduce community rates of head injury).

Although helmets can protect cyclists in some crashes involving impact to the head (most do not), there is very poor evidence that mandatory helmet legislation has achieved the desired reduction in head injury rates at a population level.

Bicycle helmets are designed to protect the head from a direct impact to the front, side or back of the head. They do not protect the lower face or neck, and must be worn correctly to be useful.

Children and non-regular riders are the least likely to wear helmets correctly, and are also the most likely to crash.

Because of their soft-shell design (to absorb the impact of a direct blow) helmets can actually increase the risk of angular acceleration (rapid twisting of the head) as a result of oblique impact and cause diffuse axonal injury. Motorcycle helmets are better designed to avoid this kind of brain injury.

Brain injury is the most serious head injury, obviously more important than superficial scrapes and grazes which “count” as a head injury in bicycle statistics.

A recent review of the evidence for helmet efficacy found that they were likely to prevent only about 10%–15% of head injuries. This surprisingly low figure is due to publication and temporal bias identified in an earlier analysis of the same studies reporting on bicycle helmet efficacy. It does not take into account possible increases in some kinds of brain injury because of the design of bicycle helmets.

Is a 10%–15% injury reduction a good result?

A significant problem with mandatory helmet legislation is that it stops people from cycling. When mandatory helmet legislation was introduced in Australia in 1990 it led to a 30%–40% reduction in the number of people cycling.

The same thing happened in New Zealand in 1994 when they introduced similar legislation. A recent survey found that about one in five adults (20% of both non-regular and regular cyclists) said they would cycle more if they didn’t have to wear a helmet.

Numerous studies have identified that the health benefits of cycling (through increased physical activity) significantly outweigh the injury risk. Therefore, there is a population health cost due to helmet legislation.

Mandatory helmet legislation could be justified if it achieved significant reductions in head injury rates. Time-series analyses of head injury rates show a significant decline in head injuries for all road users (motorists, pedestrians and cyclists) before the introduction of helmet legislation. This decline is attributed to the general improvement in road conditions that resulted from a series of road safety programs (eg, random breath testing, speed reduction, social marketing) through the 1980s. Bicycle helmet legislation came in on the tail of these improvements.

The slight reduction in cyclist head injury rates in the two years immediately after the legislation was introduced is due to the continued benefits of improvements to the road safety environment and, importantly, because there was a large drop in the number of people cycling and hence lower exposure to risk.

The Cochrane reviews which concluded that helmet legislation was effective in reducing head injuries did not consider exposure (how many people were cycling and how far) and cannot rule out that small reductions in head injury rates are attributable to fewer people cycling.

It is worth considering that not all types of cycling are equally dangerous. Mountain biking and road racing are far riskier than recreational cycling on a bike path.

Legislation is a blunt instrument to protect people from their own behaviour. We do not make it punishable by the police to smoke tobacco, eat junk food or not wear sunscreen.

Conversely helmet legislation restrains a fundamentally healthy behaviour. It has more negative health effects than positive.

Professor Chris Rissel is with the School of Public Health, University of Sydney. His research interests focus on obesity prevention and active transport with a particular emphasis on cycling advocacy.

Posted 23 April 2012

18 thoughts on “Chris Rissel: Wrong-headed laws

  1. Linda Ward says:

    Jake Olivier has already pointed out that Prof Rissel has misrepresented Elvik’s results as showing that helmets “prevent only about 10-15% of head injuries”, when the actual reduction was much higher, 42%.
    Prof Rissel’s claim that “This surprisingly low figure is due to publication and temporal bias identified in an earlier analysis of the same studies reporting on bicycle helmet efficacy” is yet another misrepresentation of Elvik’s paper.  Elvik in fact stated that Attewell et al. were “correct in their conjecture that the possible presence of publication bias did not greatly influence summary measures of effect”.  Elvik also said “Do bicycle helmets reduce the risk of injury to the head, face or neck? With respect to head injury, the answer is clearly yes, and the re-analysis of the meta-analysis reported by Attewell et al. (2001) in this paper has not changed this answer.”.
    Elvik’s results showed that helmet use reduced the risk of a brain and fatal (head/brain) injuries by about 60%.
    In response to requests/s to provide references to support the anti-helmet/law claims made in, the top 3 items on Prof Rissel’s list were the anti-helmet web sites, (website of the, as noted by RT, “strangely titled” Bicycle Helmet Research Foundation), and
    Prof Rissel’s ‘interpretation’ of the Elvik study is ‘remarkably similar’ to the misrepresentation at

  2. University of New South Wales says:

    Professor Rissel states “A recent review of the evidence for helmet efficacy found that they were likely to prevent only about 10%–15% of head injuries.” This is a misquote from the paper cited (Elvik, 2011). The closest result to this figure is the summary odds ratio from a random effects model after adjusting for publications when HEAD, FACE and NECK injuries are combined (OR=0.85, 95% CI: 0.74, 0.98). The equivalent odds ratio for head injuries in this paper is 0.58 (95% CI: 0.45, 0.75) or a reduction in the odds of 42%. 

    The Elvik paper has gone through at least two full-length corrections since originally published. The latest can be found here. Through these corrections, the odds ratio quoted by Professor Rissel changed to 0.82 (95% CI: 0.72, 0.93) to finally 0.67 (95% CI: 0.56, 0.82). I find it odd that face and neck injuries have been included with head injury considering bicycle helmets are not designed to protect these body parts and their inclusion inevitably pulls the odds ratio towards a null effect. But, when they are combined, there is an estimated 33% decrease in the odds of these injuries when wearing helmet versus not. 

    More information regarding the changes made in the quoted Elvik paper can be found here.

  3. Department of Health Victoria Clinicians Health Channel says:

    I strongly disagree with  Professor Rissel’s message that we should not have compulsory bicicle helmet laws because the reduction in head injury is minimal.  Check our editorial in the MJA  and also see Accident analysis and prevention.

     I am a neurosurgeon at the Alfred Hospital, Melbourne.  We see and treat head injury patients frequently, some of whom are bicyclists. 

    I dispute his claim that diffuse brain injury (DAI) is worsened by wearing a helmet. Helmets are likely to lessen the severity of head injury after bicycle collisions.  A high speed deceleration diffuse axonal  injury may not be lessened by the helmet but the trauma to the scalp and skull is lessened and the extraaxial bleeding, brain contusions and haemorrhages such as epidural and subdural haematomas which often accompany these injuries (even in lower velocity impacts) will be lessened by wearing a helmet.  These are often life-threatening injuries. Even if head injury following bicycle collison is mild, the patient  may suffer  ongoing consequences resulting in  chronic disability in the near term and/ or long term.  Individual head injury can have an enormous and often lifelong impact on the individual and their family (and society).  We should do ALL we can to lessen this impact. 

    Australia prides itself on having an advanced heath system.  We should not be advocating inferior standards of health prevention.


  4. RT says:

    The links lead to what is clearly an advocacy website for removal of compulsory bicycle helmet laws, somewhat strangely titled “The Bicyle Helmet Research Foundation.” I am always suspicious of advocacy websites with titles like this. Honestly I don’t expect this from the MJA or associated publiations. There is plenty of research that also indicates that bicycle helmets save lives and I won’t let my daughter ride without one.

  5. Sue says:

    Bad science is bad science, even when done with good motivation. It appears that the survey cited provides a low level of evidence – certainly not stong enough to base public policy on. Why not ask current children whether they ride or don’t ride because of helmets? Perhaps we might find that, having grown up with them, it’s a non-issue. Comparison with countries with very different riding cultures also lacks validity, as we are not translating the terrain, traffic culture and national culture to our own.

  6. Scott Watson says:

    I’m pretty unsure what the right answer is here. I always thought mandatory cycling laws were good, but speaking with a health researcher yesterday, I began to think may they’re not good (on the basis that they might be reducing cycling and increasing inactivity, which may lead to inactivity related health problems could be a far greater magnitude than the head injury related problems. This new perspective made me read a little further in this article. Having read it, I remain largely unconvinced. One thing that bothers me in particular is that the author misstated the figure related to helmets being a disincentive. He says around 1 in 5 (20%) say helmets are a disincentive, when the number in the link is 16.5%, and the reason is 10th on a list of 18 reasons. Furthermore the number is based on self-reporting (which is arguably an unreliable measure). I’m not saying I know the answer to this debate. I’d really like to see more trials and more evidence. Perhaps mandatory laws could be adjusted for cycle paths/shared paths, and the rate of cycling and accidents monitored. Not a perfect study, but a start. Similarly, in a non-car priority area (like most of the City of Melbourne) perhaps helmets could be made strongly encouraged rather than mandatory for a trial period (say, monitored over two years with an option to reinstate mandatory laws at any time if things don’t go well).

  7. Gerrit says:

    In the netherlands it is not required to wear a helmet when riding a bike. Injuries and deaths are a lot lower as in Australia. Cycling is much more common. Kids ride to school, sports on there own. People would ride the bike less if they had to carry their helmet with them all the time. Where do you keep it when shopping? So you would take the car instead. Making roads much more unsafe. Cyclist are killed by cars not by themselves falling. Take the car out of the equasion and you have safe roads for yourselve and your children.

  8. cycledohc says:

    I am always interested in the viewpoints of others. There is certainly a mixed bag here. I am an active racing cyclist having raced for over 20 years. I have crashed (mainly while racing) many many times and never hit my head. Having said that I have seen fatal bicycle injuries both with and without helmets usually as the result of collision with a car. I would always race with a helmet on but am more than happy when in Europe to ride without one. However the European motorist is usually much more understanding of the cyclist especially in Holland where so many people ride. I did see where one European country dramatically reduced bicycle related injuries not by making cyclist wear helmets but by a public campaign educating car drivers.

    If given a choice I would definitely wear the helmet for a morning commute in traffic, 60km/hr with the cars down a hill, but not to go for a training ride out beyond the traffic.

    I would have to fully agree with the author and the article!

  9. Sue Ieraci says:

    It will be interesting to see whether helmet aversion disappears in the next generation, with children having brought up to consider them “normal” (much like sun protection). Children and occasional riders might have the greatest number of accidents, but do they have the most serious ones? The only fatal bicycling head injury I have seen was in a racing cyclist – before helmet days. What we need to see are the data about high-speed and high-trauma crashes. Lastly, most of southern Europe is no paragon of public health – smoking laws and playground design are a long way behind ours.

  10. Jaime says:

    Not to be forgotten is our c-spine including our airways remaining patent that requires protection. They marginally go unprotected and are quite vulnerable to pivotal forces on impact. As an example, motocross now has compulsory neck braces for all competitors. In time, this may filter to recreational cycling (ideally optional and not by law). I have lost two friends with top line helmets die from c-spine breaks leading to limited brain activity. In saying this, I have also witnessed skull fracture incidents,(situational to a racing background). Each as life changing as the other with devastating outcomes in my opinion. I too have travelled overseas and relished in the journey of riding through farming country in Europe without my helmet, it did feel strange, but I did feel quite free and safe in doing this, however horses for courses and I say helmets are needed in most disciplines. To me, its the law these days that is mostly feared. Stats are only recorded if someone is injured, but what about those who walk away uninjured? Maybe lucky, but that would be my guess.

  11. Liz says:

    Firstly, I’m a helmet wearing cyclist, about 5000 km each year. Not mentioned is the neck injury potential of the helmet with a lycra cover I was wearing in 1993 when a large obstacle fell onto the road in front of me and I went over the bars, not sliding on the gravel and injuring a cervical disc, one scaphoid and having amnesia as a head injury.
    Secondly, living in a community of Sikhs, it is not possible to wear a turban under a bicycle helmet and so the Sikh males are all in cars.

  12. Martyn Kirk says:

    This opinion piece seems to lack decent supporting data and criticizes a recent Cochrane review. There is nothing new in this opinion piece that should affect public policy. I don’t think downplaying the importance of regulatory regimens like bicycle helmets is a good way of improving public health, which relies on complementary approaches. There are many other examples of regulatory measures that do work for public health where legislation is necessary and effective, such as seatbelts and fluoridation of water supplies. Many population health measures curtail individual freedoms, which is a fundamental tension of public health. I ride a bike regularly and am happy to wear a helmet at all times. I have seen many occasions where the helmet did the job it was designed for.

  13. John Stokes says:

    As a regular keen cyclist I would feel very uncomfortable without a helmet. Having seen the results of several accidents when cycling with others including some doctors who have avoided serious head injury because of the helmet fracturing and not their skulls I just can’t understand why anyone would suggest you don’t use one. None of my cycling friends would venture out on a road bike without a helmet.

  14. shaun says:

    “Children and non-regular riders are the least likely to wear helmets correctly, and are also the most likely to crash.”

    So perhaps children and non-regular riders need to be wearing helmets correctly because they are the most likely to crash.

  15. Michael King says:

    People seriously claim they would ride more if they didn’t have to wear a helmet? That’s bizarre! I can’t see how anyone would find wearing a helmet so onerous that it stops them riding. Maybe I’m just odd, but I find my helmet quick and easy to put on, light and cool to wear, and no problem at all to store. What am I missing?

  16. Stuart Paterson says:

    Completely agree. Many if not most serious accidents involving bicycles are due to them being struck by cars. Surely legislation should be aimed at reducing the number of cars on the road rather than reducing the number of bikes? Forcing the victim to take (largely ineffective) action rather than the perpetrator is surely ethically wrong. As a cyclist and motorcyclist, I am happy to wear a helmet for the latter activity but seriously resent having to do so for the former. I believe there are more serious head injuries in car accidents than in those involving bicycles. Surely in the name of equity, car drivers should have to wear helmets too (effective full face motorcycle ones of course). Surely less cars, more bikes is a good thing for health, safety and the environment. Policy should be guided by this rather than the easiest or oil lobby driven options.

  17. daman langguth says:

    The conclusions of the author are interesting. There seem to be some clear lack of common sense. Having had accidents on my bike pre and post helmet era, I can safely say the cuts and abrasions are a lot less. Whilst this is level 5 evidence, so is the study asking people if they would ride more if they didn’t have to wear a helmet. That’s just silly and there would be no proof it would be true. From the number of people who ride without their helmets, it can’t be a major factor. Unless there was some evidence that cycling rates had specifically declined per population after helmets were made compulsory, then there is no way that one can say they do more harm than good. Helmets do inhibit free bike scheme use, but implying that this has impact on exercise rates overall is beyond a joke.

    I have looked after people who died when wearing helmets, but they are cheap. I think over reliance on medical data (often not real world) in this setting makes conclusions somewhat challenging. A review of poor papers will have a poor conclusion.

    PS Most of Australia is not covered with bike paths!

  18. Anonymous says:

    As a road cyclist of 25 or more years, raised when helmets were not mandatory, I have had the misfortune to crash on several occasions. In each case, they were due to uncontrolled sliding of the rear wheel, usually in the wet, leading to contact between my body, including my head, with the road. In each case, the helmet was cracked on one or both sides and subsequently replaced, but my head was not injured, despite broken bones and abrasions elsewhere.

    Although there is some minor inconvenience, I would not ride without one under any circumstances. What’s more, when I have been lucky enough to ride in France for a total of about 2 months, where helmets are not mandatory, I am not comfortable riding without one. In fact, we often joke that you can spot the foreigners by their helmets.

    If mandatory helmet legislation keeps people from our roads, keep it in place. Let them go swimming instead – it’s healthy too, and you don’t need to wear a helmet.

Leave a Reply

Your email address will not be published. Required fields are marked *