Oct 7, 2012

What are the benefits of bicycle helmets?

A new study reports the rate of hospitalisations for cycling-related head injuries in NSW has fallen markedly and consistently since 1990. The authors say it's due to helmets and infrastructure.

Alan Davies — Editor of The Urbanist

Alan Davies

Editor of The Urbanist

Rates per 100,000 population of bicycle related head and arm hospitalisations. Source: Olivier et al, 2012

There’re two key arguments in the debate on the mandatory helmet law (MHL). On the one hand, it’s argued compulsory helmets deter cycling. On the other, it’s contended they reduce head injuries. Both claim a social benefit in lower health costs – more exercise vs fewer head injuries.

I’ve looked at the deterrent argument before (here and here). Now, a new paper analysing bicycle-related head injuries provides an opportunity to look at the claimed benefits of helmets (you can download the full paper here).

The paper is by three researchers from UNSW, led by Dr Jake Olivier from the School of Mathematics and Statistics. The authors examined hospital admissions data in NSW from 1990, the year before the mandatory helmet law was introduced, until 2010.

To isolate the impact of helmets, they compared hospitalisations for bicycle-related head injuries with those for bicycle-related arm injuries. The logic is that, while there’s no reason they should be the same, they should change over time at the same rate, unless an intervention is directed at one and not the other.

In 1990, just prior to the introduction of the MHL, head injury rates were higher than arm injury rates (see exhibit). However following enactment of the law, they fell 29% relative to the change in arm injuries.

That was a substantial and dramatic benefit. It’s equivalent to around 170 avoided hospitalisations in the first year alone. It’s compelling evidence of the benefits conferred by wearing a helmet.

The authors go further, though, and attribute the social benefit to the mandatory helmet law, because it drastically increased the rate of helmet wearing. Prior to the law, 10% of children and 25% of adults wore helmets, but this jumped to around 80% within two years.

We should expect the helmet law to provide a one-off, permanent lowering of the head injury rate. But what’s very interesting about the exhibit is the two variables didn’t run in parallel over the period from 1991 to 2010 but diverged sharply.

The number of arm injuries increased by 145% over the period while head injuries increased by only 20%. The latter figure is broadly in line with the increase in population, but it’s much, much lower than other indicators of the growth in cycling cited by the authors.

For example, a survey of NSW residents found the number who had cycled within the previous 12 months increased 50% between 2001 and 2010. Bicycle imports grew 145% between 2000 and 2009 and the number of cyclists counted in the Sydney CBD increased 156% between 2002 and 2010.

Thus head injury hospitalisations fell massively in ‘real’ terms over the 20 years from 1990 to 2010.

The exhibit also shows there was a distinct turning point in the trend at 2006. Thereafter the number of both arm and head injury hospitalisations declined, although the reduction in head injuries was even stronger.

The authors attribute the difference between the two variables up to 2006 as entirely due to the protective effect of wearing helmets. Dr Olivier says:

We found that the overall benefit of mandatory helmet legislation in lowering head injuries was larger than previously reported and has been maintained over the past two decades. Before the law commenced in 1991, bicycle-related head injury rates exceeded those of arm injuries. By 2006, head injuries were 46% lower than arm injuries.

On the other hand, they ascribe the post-2006 decline to the combined effect of the mandatory helmet law and the construction of safer cycling infrastructure.

Infrastructure matters because head injuries are strongly correlated with collisions between motorised vehicles and cyclists. These usually occur at higher speeds than other bicycle accidents.

While much cycling infrastructure is only a painted line on the road, it nevertheless increases the separation between bicycles and heavier, faster-moving vehicles.

Why head injuries fell continuously and sharply in ‘real’ terms over the entire 20 year period isn’t clear. It’s possibly helmet-related e.g. improved helmet fitting or fewer ‘risk-takers’ going helmetless.

It might possibly reflect the dramatic decline (mostly for reasons unrelated to helmets) in high school children cycling to school over the last 20 years. Or perhaps the rudimentary cycling infrastructure of the 1990s and early 2000s made a greater contribution to safety than it’s usually given credit for.

We can be pretty confident the big jump in rates of helmet wearing accounted for the initial spectacular reduction in head injuries. Their role in the subsequent equally remarkable further improvement is less clear.

That doesn’t mean however that mandating helmets was necessarily good policy. It’s possible rates of voluntary helmet wearing might’ve improved significantly without the law, perhaps driven by education, better helmet design, and rising concern about the dangers of cycling on roads.

And of course the injury avoidance benefits have to be assessed against the costs. Some cyclists are convinced the cost to the community in foregone exercise due to the deterrent effect of helmets far outweighs the benefits from avoided head injuries.

(Visited 50 times, 1 visits today)


Leave a comment

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

30 thoughts on “What are the benefits of bicycle helmets?

  1. Sung-Che Lin

    Look at the graph closely, after 2006 there is a clear decline in head injury rates, while the arm injury rates is constantly increasing. One of the factor that should be considered in the graph is the increased amount of people cycling per year. Nonetheless,it is evident that helmet do protect the most important part of our body.

  2. Alan Davies

    Nik Dow #28:

    Yes Nik, as per my comment on 11 October at #23 above

  3. Nik Dow

    Alan, 30 was the number of people who did the helmet free but illegal ride from Coburg down to the launch. 60 was the number in the meeting room. We published a photo of the ride on our site.

  4. Philip Darbyshire

    Thanks Alan, very thoughtful ideas. I’m not a ‘fanatic’ on either side of this issue but I do seem to be coming down on the side of repealing MHL and making them a choice for people.

    Re your questions. 1) bus service is very good and easy as is walking. Lots of cars about and the city is ‘busy’ with cars, obviously especially at ‘rush hours’ and parking is hard so I can’t imagine people being keen to take their car. Good point about the various ‘socially productive’ alternatives.

    2) Glad to hear that the walkers get more exercise 🙂 makes me feel good.

    3) Interesting one. Purely from chats and observation, I’d agree that the people in Bolzano would NOT be happy if they suddenly had to wear helmets (and for no good reason) but I wonder if people here would not see the loss of compulsion as a kind of gain, especially those who might be consequently tempted to take to a bike? For those cyclists who like a helmet and would continue wearing it, then no problem there. As for how much cycling would increase here if MHL were repealed, we’re unlikely to find out, unless we take the plunge i suppose.

    This may be a bit ethereal but there really was something wonderful about a city that was so alive with people and where people cycling on a whole range of “low”, medium and high value trips was just so natural and just part of ‘who we are and what we do’. I could go all Heideggerian here 🙂 about the Bolzano way of ‘being with a bicycle’ but that’s maybe a bit much. I think that something profound happens when the fundamental ‘worldview’ or approach to bikes and cycling changes. When the change is from ‘this is fun’, this is handy’, ‘this is easy’, getting on my bike to go somewhere is something almost unreflective to ‘this is a danger or potential life-threatener that must be legally and nationally managed as a risk and safety issue’. I think we may lose something when this happens. But I could be very wrong.

  5. Alan Davies

    Philip Darbyshire #25:

    That’s an interesting way of looking at it. If it’s supposed helmets were made mandatory in Bolzano, here are some issues that occur to me:

    First, the size of the reaction would depend on the availability of suitable alternative means of transport. Could cyclists who didn’t like helmets drive, take a bus, or walk? Or would they have to suck it up? Perhaps some trips would no longer get made at all, but they’d have to mostly be low value trips – would whatever these persons did instead be any less socially productive than the foregone trip?

    Second, those who choose to walk instead would lose time, but they’d get more exercise (walking requires roughly double the effort for the same distance as cycling). With bike share schemes the overwhelming majority of users say they would otherwise have walked, used their own bicycle, or taken public transport. In the case of the London and DC bikeshare schemes, only 1% say they would otherwise have driven.

    Third, people are loss averse. Residents of Bolzano would feel the loss of helmet choice far more keenly than Australians would feel happy about having helmet choice. We can’t say that if cycling by adults dropped 10% in Bolzano that it’s likely it would correspondingly increase by a similar amount here in the event the law were repealed.

  6. Philip Darbyshire

    I’m a bit reluctant to weigh in here as a non-expert and very ‘former’ cyclist. So this is purely an observation but it was so dramatic that it is worth a mention.

    I have spent over 3 months this year in Bolzano, Northern Italy. I’m a ‘walker’ and what I noticed all over the city and surrounds was the number of cyclists. They ranged from the tiniest kindy kids out with parents to the nonnas and poppas. People were clearly cycling to work, to the shops and markets, and everywhere else. I lived near about 4 schools and students came to school in their hundred by bike. Yes, even teenage girls!! In Bolzano, they estimate 30% of journeys are made by bike.

    I walked a lot on walk/cycle trails and here, Bolzano has a wonderful infrastructure record with over 50 kms of dedicated bike/walk trails and paths. All great.

    BUT…..the cyclists were also out on the city’s roads with the cars and busses, and many of these streets are narrow, winding, ‘old alpine’ roads and streets with no dedicated bike lanes.

    What absolutely struck me though, was that few if any people wore a cycle helmet. The cyclists who did seemed to be the youngest children, the ‘serious’ cyclists/racers and the mountain bikers. For everyone else, they just got on their bikes and cycled as if it were the most natural thing in the world, and somehow they seemed to live to tell the tale.

    If I think of Bolzano bringing in MHL tomorrow, the idea that the number of cyclists would NOT drop drastically seems almost delusional. There is no question that people who want to wear a helmet should be able to if they wish, but I do wonder if our MHL may be one of the most glaring examples in Public Health of the negative consequences of the ‘Law of Unintended Consequences’.

    As I say, just an observation.

  7. Stephen Anderson

    Thanks for the response, Nik. Much appreciated.

  8. Alan Davies

    Nik Dow:

    Nik, what are you on about with “…if you make up your stats…”? I linked to Sue Abbott’s account to support my contention there were 30 helmetless riders!!! You can also read Kathy Francis’s account of the day (page 110), who cites the figure of 60 attendees. Your response is just deceptive spin.

  9. Nik Dow

    Yes Alan the pre-publicity was good, we had people standing up the back of the room for lack of seating, and if you make up your stats and opinions like you did the number of attendees then you should stop writing.

    The Age article got us about 100 new signups on the Freestyle Cyclists online petition. Each signup displayed has confirmed the email address by clicking on a link that we send them, to ensure the integrity of the petition.

    Stephen Anderson, the two most recent surveys that I am aware of include the Cycle Promotion Fund at (Aus wide survey) and (Sydney area survey).

    Jan Gehl, the architect of Melbourne’s revival as a livable City cited a study done by the Copenhagen municipality that estimated they would lose 50% of their cyclists if helmets were compulsory. The bigger the take-up of cycling, the more damaging helmet law is. That’s why as we expand cycling in Australia the law becomes more and more significant as a retardant.

    And to those who say we should only work on getting better infrastructure – why would you build stuff and maintain a law that puts 1 in 5 off using it? We need to address all the issues, including law, enforcement, attitudes, infrastructure – not just one. I personally do a lot of work on infrastructure, government funding, law enforcement and other advocacy work. Helmet law is part of the big picture.

  10. Chard Jeffrey

    A helmet saved my life. I fell hard, hit ground head first, helmet cracked, I was able to hobble m way home. What better statistic is that

  11. SBH

    See, ridestowork the problem I have is what appears to be the inability of some people to accept or read basic information – like I don’t give two slippers for whether its good or bad, I just don’t care. it’s an irrelevance. For my money the ‘evidence’ for the negative effects of helmets is about as honest and useful as the tosh put forward by people who deny the earth is heating up but I won’t waste my time on the argument. Moses and the prophets etc etc……….

    happy to argue for driver education, better paths, showers at work ride to school day, less parking, congestion taxes but not this one. It’s a dead duck.

  12. RidesToWork

    SBH: the BHRF’s page, how helmet promotion and laws affect cycle use – and CRAG’s page, Brief Summary of Surveys Showing a Decline in Cycling due to MHL, provide some info on the effect of helmet laws on cycle use.

    Austin M: I think prob1975 is saying that cycle use fell more than either head or arm injuries, so the risk of injury per cyclist actually increased. As I said in my previous post, census data on cycling to work in NSW: 1986: 1.09%; 2006: 0.84%.

    If census data are indicative of the amount of cycling, then helmet laws have probably made cycling less safe, the most likely factors explanations being risk compensation, reduced safety in numbers and possibly, as argued by overseas researchers, that helmet laws tend to discourage the safest cyclists

  13. SBH

    I can find similar stuff to helmet freedom to justify opposition to seat belts, speed limits, motorcycle helmets, vaccination, fluoridation, and to support homeopathy, magnets the spirit world and a lot else. I can find stuff that using the same argument promotes and denigrates all manner of things.

    In the end all these lobby groups are just lobby groups – they’re political. Good on ’em for having a view. I don’t like my helmet but I wear it. I don’t like pants but I wear them too.

    The point is – is this really the biggest issue? does it really represent such an attack on our freedoms? Light a candle and stop yelling at the dark.

  14. Stephen Anderson

    Spot on Alan. Thanks. I’m getting back in the car. After reading all this cyling, is way too dangerous.

  15. Austin M

    prob1975 I was a bit late on this one …. You seem to be saying in a similar way that because air bags have shown a significant reduction in head injuries against other injuries types, we should scrap airbags? … I mean its no secrete that serious injuries and motor vehicle accidents in general are on the increase and thus the conclusion must be made that people are demonstrating compensatory behavior and crashing more???.
    Doesn’t it seem more plausible that there is both more cyclists and more vehicles on the roads with more interactions/crashes but safety interventions like helmets/airbags have helped reduce the severity of crashes by better protecting users in the event of a crash.

    This debate smacks of the seatbelt debate by which some people thought they could better brace/move/avoid something if not constrained by a seatbelt, or jump from a car in a roll over etc. Unfortunately despite compelling evidence there will always be some who believe they know better as they grew up in the good old days where all this safety gear wasn’t needed. This is not just individuals but countries who don’t legislate safety measures and cultures that rally against safety improvements that they perceive as intrusion to their “freedom”. This is often to devastating effect on individuals and great effect on health systems (just look at the health burden in countries that have poor seatbelt compliance as an example). If an activity is worth doing it will still be worth doing with the right safety gear!

  16. Stephen Anderson

    Thanks Alan. Those data are for bicycle share schemes. Anything for all cyclists? I’ll keep looking.

    Stephen Anderson:

    I don’t think that’s right. Did you have a look at ‘The Facts’ section? AD

  17. Alan Davies

    Stephen Anderson #12,13:

    Pending Nick’s advice, you could take a look at

  18. Stephen Anderson

    Hi Nik, can you please point me in the direction of the surveys that state, as you say, “one in five people say helmets put them off riding”?

  19. Stephen Anderson

    Very interesting debate. Can anyone point me in the direction of the data that shows helmet use as a requirement is a deterrent to riding a bike? The only ‘barriers and obstacles’ I’ve read previously are (i) bad weather, (ii) fear of motor vehicles, (iii) gaps in cycling networks, (iv) infrastructure design shorcomings, (v) lack of end faciliies, (vi) disconnect with other transport modes, (vii) Other [slower, can’t carry cargo, less convenient]. These are from the (i) Vic Cycling Strategy, (i) Research into Barriers to
    Cycling in NSW, (iii) Barriers to Cycling (Deakin Uni). I’d be very interested to hear from someone that has seen some stats on this.

  20. Alan Davies

    michael r james #10:

    Michael, I think your jaundice is getting the better of you. I’ve always argued that infrastructure is the name of the game.

  21. michael r james

    But St Etienne, SBH’s point (which I second) is that the MHL issue is a gross distraction from the main game: SERIOUS CYCLING INFRASTRUCTURE.
    AD likes writing about anything but the infrastructure issue because there is easy pseudo-analytical data to hand on all the diversionary issues (almost all of which disappear if there is good infrastructure), and it stirs up pointless argument. I believe it has come to push out more important issues.
    I hope it is the last article on this topic for a long time–say at least 12 months.

  22. St Etienne

    SBH, I often hear the comment that MHL is a side issue and that we should be focussing on the important issues. This fails to recognise that a lot of critics of the law are already active campaigners for better infrastructure and promoting bicycles as a transport option.

  23. SBH

    and there are bigger threats to our life and freedom than being required to wear a helmet (like plumes of methane bubbling up from the sea floor in the arctic) so could we all please grow up and focus on something important?

  24. SBH

    The thing that continues to worry me in this ‘debate’ is the focus the anti helmets create on the issue of helmets.

    The MHL discussion appears enormously subjective and slave to our own ideas of whether helmets are a good or bad thing. Anything I’ve read by Chris Rissel fits this category. A similar and equally equivocal ‘debate’ rages in crikey about the price, value as an investment and cause of Australian housing stock.

    It would be better for us all to treat MHL as a side issue and focus on cycling infrastructure and driver education.

  25. St Etienne

    Given the release of this report I expect to see governments and cycling groups around the world voice their support for a similar all-ages helmet law. No doubt the already considerable number of everyday utility cyclists in countries such as Germany or Denmark will remain at their current levels; in fact, cycling will most likely flourish as couples travelling to the theatre in Berlin happily don their plastic hats while riding their inherently unstable 21-gear mountain bikes. Australia will be held up as the pinnacle of cycling culture as Dutch citizens swap their omafiets and civvies for sports bikes and hi-vis vests. The UNSW researchers will be jetted around the world to consult cycling organisations on how to achieve rapid growth in cycling participation through proper legislation. Yes, I can certainly see this all happening.

  26. Alan Davies

    Nik Dow #4:

    There were about 60 30 participants without helmets at the launch of the anti MHL Freestyle Cyclists on Saturday, notwithstanding good publicity in The Age and elsewhere and appealing speakers. I don’t think “some cyclists” is an unreasonable description.

  27. Nik Dow

    Alan, it’s not just “some cyclists” who are concerned that helmet laws are contributing to rising health costs and premature deaths. It’s also Professor Piet de Jong,
    Actuarial Studies Macquarie University and Professer Chris Rissel, University of Sydney school of public health, to name a couple of senior academics who have published on this.

    At the launch of Freestyle Cyclists on Saturday, Prof Rissel told us that the entire NSW health budget would soon be equal to the cost of treating type II diabetes in that state. Diabetes alone is estimated to cost Australia $10.3bn (see )

    Prof de Jong has conservatively estimated the health cost of helmet law in Australia at $0.5bn. A recent UK study has estimated that their health budget could save £17bn/year by raising cycling levels to equal Copenhagen’s

    Pro-compulsion arguments rely on the untenable claim that the law does not deter cycling. This is despite numerous surveys in which around one in five people say helmets put them off riding. This is despite the fact that two out of three fines issued to cyclists in Victoria are helmet fines –

    “Some cyclists” indeed. Alan, you hide your pro-compulsion bias under a veneer of objectivity, it’s wearing thin.

  28. RidesToWork

    The interesting thing about the claimed increase in cycling is that there has been a 23% decrease in the percentage cycling to work in NSW:
    1986: 1.09%; 2006: 0.84%

    Walker’s surveys (25 sites in Sydney on cycling to work),, Fig 9
    October 1990: 3798; October 1994: 2479
    April 1990: 4405; April 1996 2269

    Recent Sydney data:
    The 2008/09 Sydney Household Travel Survey (p26) shows there were 106,000 bike trips on average weekdays in 2008/09, compared to 101,000 bike trips in 2001/02 ( This small increase does not appear to be enough to compensate for the drop in cycling suggested by Michael Walker’s counts. The counts “at 3 primary locations that feed into cycleways within the Sydney CBD”, quoted in Olivier’s paper, suggest that special circumstances (including higher population densities and additional spending to encourage cycling in the CBD) may have led to increased cycling in this area that is not representative of the state as a whole.

    Australia-wide travel surveys
    A comparison of surveys in 1985/86 and 2011, showed that Australia’s rate of per capita cycling participation in Australia was 22.3% below the rate of population growth. Comparing populations aged 9+, there is 37.5% less cycling per capita than four years before the helmet law:

    In summary, the percentage cycling to work in 2006 was 23% less than in 1986, head injuries have remained static, as have helmet wearing rates (since 1991), but arm injuries have increased dramatically.

    The injury data I’ve seen suggest that cyclists who fall off their bikes suffer arm injuries, those in car/bike collisions suffer head and leg injuries. So the data appear consistent with an increase in cyclists falling off their bikes, possibly related to an increase in off-road recreational cycling, compared to transport cycling.

    One recently-published paper tried to investigate the differences in attitude of helmeted and other cyclists, saying: “With all the limitations that have to be placed on a cross sectional study such as this, the results indicate that at least part of the reason why helmet laws do not appear to be beneficial is that they disproportionately discourage the safest cyclists.”

    Perhaps we’ll know more when the 2011 census data on cycling to work are released.

  29. Last name First name

    Parker alan

    Hi Alan,

    To put the downturn both types of bicycle injuries since 2006 Into perspective i have attached my latest paper, as a PDF files,see figure 11 and then start at the beginning the

    Alan Parker OAM

    [email protected]
    Tel 613 5984 3578

  30. pjrob1957

    I am not sure what the graph is asking me to conclude.
    It appears that there was no real reduction in head injuries while there was a gain in arm injuries.
    Other research tell us that there was a 30%-40% decrease in bike use.
    This altogether does suggest what has has been concluded many times that helmets, like other interventions, caused risk-compensatory behaviour.
    If the helmet does reduce head injury (by no means should we conclude this means brain injury!) but also increases the likelihood of an accident in the first place then what exactly has been gained?

Leave a comment

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

Share this article with a friend

Just fill out the fields below and we'll send your friend a link to this article along with a message from you.

Your details

Your friend's details