14 April 2009

Putting a Price on Bicycle Helmet Laws (You Can't Afford It)

One of our readers here at Copenhagenize is Piet de Jong, Professor of Actuarial Studies at Macquarie University in Sydney.

He dropped me a line last month when he published a mathematical health benefit model that puts a price on helmet laws and we're pleased to include it here. Evaluating the Health Benefit of Bicycle Helmet Laws.

His work suggests that Australia's all-ages bicycle helmet laws incur a health cost to the nation of more than a half a billion dollars [AUD] every year. In other words, there is now evidence that bicycle helmet laws have a direct, negative impact on Australian health costs. They amount to $301 million a year and a total cost of $519 million when combined with the non-health costs of the law.

The model's mathematical formula uses various inputs, including the exercise benefits of cycling, head injury percentages and reductions in cycling popularity due to helmet laws.

Professor de Jong has a spreadsheet with which his formula can be calculated using inputs from other regions. It's available online and if you know the stats for your country, you can do your own calculations.

He has further calculated that if America adopted nationwide helmet laws like those in Australia, the USA would suffer a health cost of $4.75 billion every year.

In Great Britain, nationwide all-age helmet laws would result in a health cost of $400 million per year and the Netherlands would be incredibly hard hit and suffer a $1.9 billion annual loss.

Here's the abstract for the paper:

A model is developed which permits the quantitative evaluation of the benefit of bicycle helmet laws. The efficacy of the law is evaluated in terms of the percentage drop in bicycling, the percentage increase in the cost of an accident when not wearing a helmet, and a quantity here called the "bicycling beta." The approach balances the health benefits of increased safety against the health costs due to decreased cycling.

Using estimates suggested in the literature of the health benefits of cycling, accident rates and reductions in cycling, suggest helmets laws are counterproductive in terms of net health. The model serves to focus the bicycle helmet law debate on overall health as function of key parameters: cycle use, accident rates, helmet protection rates, exercise and environmental benefits.

Empirical estimates using US data suggests the strictly health impact of a US wide helmet law would cost around \$5 billion per annum. In the UK and The Netherlands the net health costs are estimated to be \$0.4 and \$1.9 billion, respectively.

Here's the link to the paper.

Let's hope certain Danish politicians are listening.

Via: Prof. Piet de Jong and also www.cycle-helmets.com in Australia.


Martin said...

Hey, remember the old 'Your bike is hot' flyers you designed a while back? How about a leaflet, one page, that exposes why helmets aren't good for cycling. Succinct, with references to the science, etc etc. Then we can print them out ourselves and leave them on bikes around town. Sweden needs some info to counter the push by helmet manufacturers and 'safety' fascists.

Clive said...

Martin - Why would you want to do that?

This article outlines what is wrong with helmet laws, not with helmets themselves.

If people are already riding, with or without helmets, can't we just respect their personal decision?

The people to worry about feeling they have to wear a helmet are surely those who are NOT yet riding, so are unlikely to find one of your fliers attached to their non-existent bike.

Why bother someone who is happily riding with a helmet?

This shouldn't be about getting rid of helmets themselves (which are harmless), but about resisting any move towards compulsion (which is clearly shown here to be an awful idea).

Mikael said...

personally, i find it alarming that cyclists are told about the scientific studies that show that people wearing bicycle helmets have a higher of brain injury.

brain injury is usually caused by rotational injuries, which a bike helmet cannot protect the head from, but rather increases the risk.

then there's the higher risk of neck injuries.

oh, and then there's the studies that show cyclists wearing helmets have a greater chance of getting into an accident.

goodness me. harmless? au contraire.

Anonymous said...

Please link to the study showing that cyclists wearing helmets have a greater chance of getting into an accident – I'd like to read that. Even if it's true, it doesn't necessarily prove that helmet-wearing is the cause of those accidents. Maybe helmet-wearers get in more accidents because in addition to wearing a helmet they also happen to ride a lot more than non-helmet wearers. Maybe they ride more often in traffic or at higher speeds ... I'm with Clive on this one – if there's a problem with helmets it's with legislation, not the helmets themselves.

Amsterdamize said...

for crying out loud, Mikael, you know damn well that facts have a liberal bias!

Mikael said...

yes, marc, i'm a science-lovin' pinko commie fag.

anyway. here's a good place to start about rotational injuries.Minor impacts converted from linear forces to rotational.

"in addition to wearing a helmet they also happen to ride a lot more than non-helmet wearers"?!
Listen, most people in the world don't wear them. Only 50% in the US and among the 100 million daily cyclists in the EU, there are very few.

Stan - BCN said...

Hi Mikael,

I've been reading about this topic for a while now and I must say that I still can't make up my mind about it.
I guess it's because all of this is going a bit too far, we're not even asking ourselves the right question which would be : wearing a helmet can save lives? I'll say yes but if I were a politician willing to decrease bike accidents I'll be teaching people how to ride a bike properly.

I am not kidding, lots of people don't know how to ride a bike correctly and are scared when riding it and thus are potential dangers!

All in all riding a bike is almost like driving a motorbike in town and although you're not going as fast you're exposed to the same risks. The difference is that you get proper training in order to be allowed to drive a motorbike which isn't the case for bikes.

A final word on helmet laws: in the case of DK it is more a financial interest than a public health issue. If every cyclist in DK has to buy a helmet someone is going to become quite rich!

Although I don't share your position about helmet efficiency I support your action because this law is not the right solution.

Good luck!

Anonymous said...

Instead of opposing bicycle helmets you should promote to move these fu***** motorcycles away from the bicycle lanes. There is no rational reason in the world why they should use the bicycle lanes, but in Copenhagen they do. They stink, they are way too fast for a bicycle lane and they don't keep a proper distance from you when overtaking. Actually, they are the only reason why I would ever consider wearing a bicycle helmet.
I'm a foreigner in Copenhagen, so if there is a good reason for this, please explain it to me. In most other countries of the world motorcycles are driving on streets and it works perfectly fine. Why not here?

Anonymous said...

Huh? Motorcycles use the road togehter with the cars. Are you taking about mopeds?

Anonymous said...

Interesting, thanks for posting the link. Not wanting to pay 30$ to read the entire study, I understand from the abstract that helmets may reduce minor injury in minor accidents, but may increase major injury in more serious accidents. That seems feasible. So maybe it comes down to whether a person prefers to sustain an avoidable minor injury, in hopes of preventing deeper harm in the event of a major crash. Looking at it like that maybe it's better to Not wear the helmet. On the other hand, a person could also consider the likelihood of a major crash vs. a minor fall ... Lots to think about. For me, the only injury I've ever had on a bike was when I crashed on some ice, and both hands were reduced to hamburger on the asphalt ... So do I wear gloves now? Almost never .... But I do wear a hemlmet, usually. Still prefer to do it by choice, not by legislation though.

Anonymous said...

The abstract on rotational injuries notes that these injuries are a very small percentage of cycling injuries. In other words, the majority of bike head injuries are impact injuries, which helmets do protect against. Sounds like a good reason to wear a helmet.

Anonymous said...

There are no excuses ever, in no circumstances, not to wear a helmet. Even in bed, as you may fall off that too.

Adrienne Johnson said...

My oh my. Why does anyone else care about what I wear on my head? All this energy... if we put it into actually riding, we would all have pedaled around the world by now : )

@ Mikael- While I am quite sure that you are a pretty normal person, i am certainly willing to entertain the idea that you are, in fact, a "a science-lovin' pinko commie fag" but forgive you for it, anyway. I am just that kid of person.

Anonymous said...

Mopeds, motorcykles. Whatever. They don't belong on the bicycle lanes. Just calling it moped doesn't mean they would drive any slower and more carefully. Ban them.

road a said...

This is just pure bullshit. I've deleted my bookmarks to this page, and the copenhagencyclechic.

Over and out.

Mikael said...

yes, science can be scary.

Anonymous said...

I suggest you cut out this glib "we have science on our side" bullshit. This paper does not appear to be peer-reviewed (nor is your oft-cited cyclehelmets.org). Ergo, it ain't science of the quality that can be relied upon for public debate.

In this particular case, it relies on the selection of a whole slew of parameters, many of which are uncertain, and can be selected to suit personal biases.

The cycle helmet legislation debate is important and worth having. Muddying the waters with this rubbish doesn't help anyone.

Ryan said...

Someone I know who drives a taxi over here, takes people to the rehab centre for people with brain injuries. Next to none were caused by riding a bike. Most by driving a car or being in a car that was in an accident. Perhaps car drivers should be forced to wear a helmet?

I have NO problem if you choose to wear one while riding, however you should NOT BE FORCED to wear one.

In the Ontario right now, the police are cracking down on those who are not wearing seat belts while giving warnings to those who are speeding.
Sorry to say, but if the speed limits were lowered and enforced, seat belts would not be needed anymore. But of course this is North America where if you can't get there A.S.A.P. it's not worth anything.

LC_UK said...

I enjoy reading both this blog and copenhagencyclechic, I have got out more in my bike because of it, but mostly because I realised it's perfectly normal, cooler and classier to cycle with normal clothes and not all the lycra stuff ;) using my bike to go to work, to go shopping, to meet friends for coffee etc. I do wear a helmet, buses here are nuts! at the end of every bike ride I thank my lucky star I am still in one piece. I wouldn't want a nanny-state telling the population 'do this, do that', but personally I like wearing a helmet, it makes me feel more secure and it doesn't destroy my 'look' or make me looks frumpy either.

As much as I enjoy this blog this forcefulness on the argument 'helmet do more damage than good' is starting to be tiring. I have been to Copenhagen, a wonderful city, a perfect city that has understood the joy and benefits to swing the population over to the push bike than the car, but anywhere else cities are 40years behind! Our roads are a lot more unsafe than Copenhagen, so please do understand that outside Copenhagen our helmets actually save our lives.

I still stand that people should have a choice if to wear one or not. But science and stats about prons and cons, health, injuries and deaths should be left out of these arguments ;)

cycle love x

Anonymous said...

Very interesting articles and string of comments. I was interested in the peer-review criticism, so I looked for and found this: "Most information published on www.cyclehelmets.org has been subjected to multi-disciplinary peer review."

I'm going to continue to research this. If nothing else, it sounds like we need better helmet standards. I teach bike safety, and I want to be telling people to do something that actually protects them, not something that will just make them feel safer (that could lead to them riding less defensively, exposing them to higher risk). But at the end of the day, life is inherently risky and I'd rather go down riding.

Safe miles!

Anonymous said...

Suk: http://www.bt.dk/article/20090418/nyheder/90418023/

Anonymous said...

In regards to "Most information published on www.cyclehelmets.org has been subjected to multi-disciplinary peer review." -- that's just bullshit. It's true enough that some of the studies they discuss have been subject to peer review, but others have not. In any case, their meta analysis has not been peer reviewed.

For science n00bs, here's a quick way to check whether the study you're reading about is reputable -- visit Pubmed (http://www.ncbi.nlm.nih.gov/pubmed/), and search for the paper. Pubmed lists almost every medical/biomedical paper out there that has been published in a legitimate journal. You may have heard of the famous top-tier medical journals like BMJ, Lancet, NEJM, and so on which are usually a safe bet. But there are literally hundreds of other journals out there which publish high quality research as well.

I'm not claiming that peer-review is perfect, but it is a bare minimum means of quality control. It is possible to craft a scientific manuscript which is completely misleading, but will fool 99% of the public. Giving experts a chance to critique these papers before they reach the public can help prevent this.

Just a cyclist said...

Good idea... I think I'll do just that for cited pro-helmet studies...

Kiwehtin said...

To respond to "Anonymous" (post at 6:23 EDT, just above "Just a cyclist"):

I can't help wondering what the basis is for your assertion that the peer review process used by cyclehelmets.org is "just bullshit". You state that some of the sources used are peer reviewed and some are not.

Given the 98 individual reviewed publications they index at http://www.cyclehelmets.org/1160.php, that leaves a range between two already peer-reviewed out of 98 and two articles not already peer-reviewed. A cursory skim through the titles they index shows the proportion is much closer to the latter: mainly a handful of media articles, a few research reports and conference presentations (for which only the preliminary abstracts are reviewed), and briefs to government. The vast majority are from peer reviewed journals. You can do the precise count if you wish: I am very confident my impression will stand. To provide a clear understanding of the post publication debate that occurs in the literature, they also list responses and counter-responses to published papers (http://www.cyclehelmets.org/1146.html#letters).

That said, given that they select serious published sources almost exclusively, and (a very few) general media articles to the extent they provide reliable, clear information such as about helmet standards and testing, each publication on the site goes through a multi-disciplinary review process by members of the editorial board (http://www.cyclehelmets.org/1121.html), who have a range of specialisations from different disciplines. They do this, according to their policy statement (http://www.cyclehelmets.org/1004.html), because they "seek to ensure that not only is the information as accurate as practicable, but also that related issues and consequences are also taken into account. Despite best endeavours, errors and other shortcomings will occur, and BHRF welcomes feedback to improve the accuracy of this resource.".

The evaluations they make stick to purely scientific measures, such as the validity of claims made by authors based on the data provided, the strengths and weaknesses of the methodologies used and so on, and other relevant evidence from fields of study that the authors may or may not have taken into account when putting forward their claims. In this way, the secondary peer review provided by the BHRF editorial board catches many problems that go unnoticed in the primary pre-publication review process. That is because journal reviewers are typically individuals with a narrow area of specialisation similar to that of the authors and the purview of a given publication does not normally require interdisciplinary review of the papers if publishes. That comes later, after publication; in this case, by the BHRF editorial board members with their wide range of expertise in different areas.

But back to the original paper sent to Mikael. This is a working paper, applying the same kind of research the author (http://www.businessandeconomics.mq.edu.au/contact_the_faculty/staff/alphabetical_list_of_staff/piet_de_jong) does in his other published research on statistical analyses of risk for insurance-related studies, but this time to the question of bicycle helmets. It is, as the author notes, a working paper, not a final published version, and if you look through his publications page (http://www.businessandeconomics.mq.edu.au/contact_the_faculty/staff/alphabetical_list_of_staff/piet_de_jong/publications), you will see that it appears to be a later version of a talk he gave in 2007, at which he would have received feedback including criticisms if any from his audience of professional colleagues. Given that the paper is currently a working paper, it is likely that some similar version will be submitted for publication soon and may appear in a couple of years after the peer review process.

Sorry to be so long-winded, but the workings of scientific analysis are much more involved and complicated than the simplistic conceptions of those too willing to swing verbal scimitars at anything they disagree with.

Kiwehtin said...

Somehow the web address for Piet de Jong's web page was truncated in my previous posting. Here's the correct URL:


Kiwehtin said...

It seems the blogger text engine doesn't like long URLs because thee two URLs still came across truncated. I've separated the URLs into short lines that I hope it will be able to parse this time. To get to them you may have to copy and paste each line onto the end of the previous line before actually going to the link:


For the publications page: click on the link on his main page or just paste this at the end of the URL given above:


JT said...

Look at this interesting ad:


They had to draw the riders in a really weird position in order to visually justify helmets. You may fly and fall like that in 1 of... many many accidents.

As I said there, we could have used the same ad with an opposite message.

Anonymous said...


Thank you for providing a thoughtful response. I'm sorry if I seemed curt, but I stand by my remarks.

Before I elaborate, I want to make clear my personal opinion that there *may* be a basis against mandating cycle helmets in terms of overall societal cost, and it is something well worth investigating. Please don't attempt any straw man business, wherein my barbs are written off as "oh, you just say that because you disagree with us!". My problem is with the means of argument, not the premise itself.

Of the 98 articles you mention, I've gleaned that roughly 10% are not from peer-reviewed journals, which is not insignificant. The important issue here is their review, not the actual articles themselves (since few visitors to BHRF are likely to read those anyway). To state that "The evaluations they make stick to purely scientific measures..." is extraordinarily naive. Again, I'm sorry to be curt, but there is just no such animal. It is impossible to summarize someone else's research without introducing some bias, whether it be in the language used, or the statistics chosen to summarize the data. This is why any review article submitted to a legitimate journal is subject to peer review, just as original research is.

The peer review claimed to be undertaken by the BHRF does not meet the most basic criteria. I'll list some of them.

1. No author details for individual pages. No contact details, no affiliations, no acknowledgements of help received, nor declaration of any conflicts or funding sources.
2. No contact details for members of editorial board (given that some don't have affiliations that would otherwise allow mail to be forwarded to them). Some affiliations appear to be out of date.
3. No editor-in-chief. No explanation of review process, or organizational structure of editorial board.
4. Dubious qualifications. See below.

Here's an example of some sloppy thinking that would never have gotten past any competent peer review process. Check out the second figure of (http://cyclehelmets.org/1079.html), which is used to suggest that the more helmets worn, the more deaths occur (whether directly or indirectly). I've re-plotted that data, and done some stats (linear regression). There is *no* evidence of such a relationship in that dataset. The correlation coefficient is r=0.70 (respectable), but the data is so sparse that the slope is in the range of 0.42 ± 0.44 (95% confidence interval). That is, there could be a positive, slight negative, or no relationship. The sample is way too small to draw any conclusion. All sorts of potential for cherry picking data (countries). Non-linear curve fits without justification is all kinds of shifty. Seriously, this is just awful reasoning. Really, really terrible. If you presented this at an epidemiology conference, you'd be laughed at.

On the "working paper" business -- that's fine for circulating among a small group of peers, but no-one in their right mind would cite such a work, much less use it to advise the government. A few questions after a presentation is not a replacement for peer review. We can look at it again when and if it is properly published in a peer-reviewed journal. Until then, it cannot be relied upon for making policy decisions.

David (Anonymous 6:23 EDT)

Just a cyclist said...

You're right David. Using 8 just datapoints you cannot draw conclusions with any confidence.
So I guess that was not the work of any statistician, which I know they do have on their board.
One strength in the BHRF website though, is their critical analysis of the pro-helmet papers.

As for this "working paper", it's quite interesting to play a little with the spreadsheet based on the articles parameters, linked to in this post.

Even if you calculate a 100% effectiveness and 100% wearing rate, this law would yield a net negative societal cost according to the parameters.

But much more interesting yet is that the balance is substantially more negative if one sets the pre-law wearing rate high and the post law even higher (for instance 50% pre law and 80% post law).

Why is this interesting? Well because one common pro-law argument is that, at already high voluntary wearing rates, this legislation would result in fewer negative effects (and probably mitigate any negative political effects).

Anonymous said...

Sure, but tweak the reduction in cycling as a result of helmet laws down to 8% (which is not outlandish, especially in the long term), and you're back to neutral in terms of health cost. The range of feasible parameters available mean that this model is way too close to the tipping point to draw any strong conclusions.


Just a cyclist said...

But all the other hemlet benefit parameters in the spreadsheet are more exaggerated than what may be found even in the most positive papers on otheir benefits. Still it is difficult to make it worthwhile, according to the spreadsheet. Which makes it even more interesting.

Anonymous said...

Just thought you might be interested in the enthusiasm the New Zealand police show for enforcing helmet laws:

Mikael said...

yes, i've seen it. police bullying people who choose to pursue a life-extending, healthy form of transport is hardly the way to encourage cycling.

Anonymous said...

Sure, some parameters might be optimistic, but again, they're not outlandish. I agree that this model, for the range of available parameters, makes it difficult to discern any worthwhile gain. But what I'm *trying* to point out is that this paper is not suitable for informing public policy. Wait til it's reviewed, published, and there is some back and fourth from other investigators, to allow some sort of consensus to form.

I can't emphasise enough how important this is. Amateur analysis of unrepresentative segments of the scientific literature can cause very real damage (I'm alluding to the vaccines cause autism movement here).

An aside: If a scientific consensus is formed that helmet laws are likely to have a substantial net negative societal cost, this mightn't be enough to motivate law makers. What do you think is more costly to their careers? A handful of young people with head injuries, where the impact could have possibly been lessened by a helmet? Or, thousands of future 50 year olds dying prematurely because of lifestyle illnesses? Echoes of the Trolley problem (http://en.wikipedia.org/wiki/Trolley_problem). Unfortunately, I suspect they'd act to prevent the former, rather than the latter.


Adrienne Johnson said...

OK, boys. Here is a question-

Why do governments create legislation that limits healthy activities but subsidizes unhealthy ones?

Example- Helmet laws that have been shown internationally to limit the growth of bicycle riding as a non-recreational activity vs. subsidization of oil to artificially drop the cost of driving (an 50 year auto bailout) and corn subsidies that prevent the growing of multiple food types thus limiting our food diversity and increasing the need to create unhealthy items like HFCS and 'biofuel' to deal with the surplus?

Helmet legislation goes beyond the possible issues of efficacy. This should also be a siren to people of the many subtle ways that the everyday person is manipulated into conforming to a societal norm by people we never meet. There are many out there who insist on pushing their fear and need for control onto others. If they can push their agenda on to you, they are legitimized in their personal fear. Anyone who gets their panties in a twist over my helmet use is really upset that they can not impose what they feel is their control over the universe on me.

It may sound flaky and new age, but if others truly cared about my safety, they would legislate the factors that truly endanger me- not being hit by a car is a great deal more effective than a helmet could ever be. Not allowing vehicles to travel over 15 MPH in any city (my street is 30 MPH despite the fact that I live in a row of family dense apartment buildings), not allowing right turns on red lights, median strips that would not allow for midblock u-turns, limiting cross town traffic through residential areas, enacting one way traffic around schools ...... helmet laws are less complicated and are thus, an easy way to pass the buck.

The above statements have not been peer reviewed and may not be my own. They should be read with caution as true contemplation of some of the points I make may make you think differently : )

Just a cyclist said...

Helmetheads who may happen to look condescendingly at fellow bareheaded cyclists (it happens at times) probably seldom take any interest in the safety of the other cyclist but rather take offense merely by seeing those "daredevils" that have not adopted this "all important safety measure". I guess they should also be looking with dismay at people who lock up their belongings with just a single lock instead of triple locks.
One guess is that this have to do with them feeling insecure or inferior to someone with less fears. Some sort of envy perhaps?

Christopher said...

Well in South Africa its law but ownership of a helmet is not economically possible for the majority of cyclists (R700 for a Bell - heck most commuters don't earn that in a week). The law is seldom, if at all enforced here.

Personally I push the safety of helmet use locally and am critical of the fortunate few who disregard the use of helmets.

But then again we enjoy one of highest death rates on the road. Motorists and informal public transport (multi seater Taxis) have zero respect for cyclists and not wearing your helmet and hi-viz jacket 24/7 is an early ticket to that big single track in the sky.

Mikael said...

Christopher, you're assuming, quite wrongly, that helmets are a golden, magical life-saving piece of equipment, which they are sadly not.

They are designed to protect a head from non life threatening injuries in solo accidents under 20 km/h. The tests they go through simulate a pedestrian falling on their head.

Use your energy on campaigning for designated infrastructure instead of scaring people off of bicycles.

stephen said...

Hi All
Those who oppose the mandatory helmet laws and the discrimination and persecution against cyclists may find this usefull.

Libertarian policy is that victim-less crimes should not be punished and the LDP website specifically lists removing bicycle helmet laws on their website at ldp.org.au. check the Policy section under Victimless Crimes.

Goto the LDP (Liberal Democratic Party) policy section titled victimless crimes.

I have met with them at their open monthly meeting here in SA, they are interested to help, and they need more members so if anyone is interested in the
libertarian ideals it’s free and is online at ldp.org.au
I don’t agree with every policy they have but this is an opportunity to get some political representation which no other party seems prepared to offer.

If Cyclists vote for and Support the LDP even if just for this one reason hopefully the majors will take notice and we will see a change to the helmet laws sooner rather then later.

Regards Steve.

Henry said...

Great post! Can someone work the formula to show efficiencies/savings from separated bike lanes and the usage they encourage?