FT.com | The Undercover Economist | Should we cyclists bother to wear helmets?

July 9, 2008

Should we cyclists bother to wear helmets?

Fascinating letter in yesterday’s Financial Times:

Sir, There has been debate about helmets and cyclists’ safety (‘Some healthy advice, Boris: please put a lid on it’, July 1, and Letters, July 3 and 4). As author of a number of papers on the topic, I may be able to inform the discussion.

Driving helmet use is zero. Walking helmet use is zero. None of this is controversial. The cycle helmet debate must stem from a belief that cycling is much more dangerous than walking or driving. Serious analysis of risk has been lacking. In the late 1980s, the Transport and Road Research Laboratory produced a report called Risk in Cycling. It was never published. I obtained a copy from a private library. Disturbed by the ill-informed basis of debate, I expanded on the TRRL work and published Assessing the Actual Risks Faced by Cyclists in 2002.

The main conclusions were:

* On-road cycling is a low-risk form of travel in all the countries assessed, including the UK. Cyclists face risks similar to drivers or pedestrians. For instance, even an active British cyclist faces lower lifetime risks than the average French or Belgian driver.

* If there is more cycling, it gets safer; if there is a drop in cycling, the risk per cyclist increases.

These findings have been repeated by a later study for the European Cycling Federation. Wearing a helmet for daily cycling is only “important” if such travel is actually dangerous. Since it is not actually dangerous, it cannot be important to wear a helmet. Boris Johnson ought to have made this clear in his article in a national newspaper a few weeks ago.

Be wary of gruesome anecdotes about smashed helmets. Helmet laws have been in force in Australia and New Zealand for more than 15 years. Although initial studies claimed modest improvements from these laws, the issue was largely settled in 2006 in a paper in the British Medical Journal. This showed that no discernible reduction in serious head injuries could be seen in the hospital admission data from those countries with sharply rising helmet use, although there was some reduction in scalp lacerations. Note that three published studies have observed an excess of deaths with rising helmet use (in the US, UK and Australia). Personally, I cannot understand why the issue is still debated at all.

The health benefits of cycling vastly exceed the risks of being killed in a crash, being comparable to giving up cigarette smoking. It is the sedentary lifestyle of car-dependency that is truly dangerous. Cycling to work dodges the jams and a gives you a workout all in one shot. Countries with cycle helmet laws, or strong helmet policies, have the highest levels of obesity. These countries also have the poorest safety records, due to the low level of cycle use.

In London, half of road deaths are pedestrians, just 6 per cent are cyclists. I cannot understand anyone involved in casualty treatment who insists cyclists should wear helmets, when they themselves do not wear a walking helmet.

Malcolm Wardlaw,
Edinburgh

Very interesting. The argument is not watertight, though, for a couple of reasons. One is the Peltzman effect: it could be that behelmeted cyclists take advantage of an increase in safety by taking more risks, such as cycling faster and running more red lights. If that was true (I am not aware of any research on the question) then helmets would not seem to make cyclists safer from the view of the statistician, but they would be giving cyclists the (privately) beneficial ability to get to their destinations more quickly.

Nor is it true that helmet use is justified only if cycling is more dangerous than walking or driving. That is not the question: the question is the marginal benefit of the helmet. If pedestrians don’t wear airbags, that is not because walking is safer than driving, but because airbags offer no benefits to pedestrians.

Interesting to note that half of London road deaths are of pedestrians and six per cent are of cyclists. Proves nothing by itself, but ten times as many “journey stages” in London were by foot as by cycle. (source: p6) That suggests, broadly, that the risks of a cycle journey are about the same as the risks of a pedestrian journey. Who knew?

I’ll keep wearing a helmet. I am

a) Risk averse and more importantly

b) Signalling to my wife that I pay attention to her opinion.

Update: I should have remembered to point out that the letter was one of many responses to my colleague Stefan Stern’s column here.

July 9th, 2008 in Everyday economics | Permalink

27 Responses to “Should we cyclists bother to wear helmets?”

Comments

I am a cyclist though mainly off-road. My (non-economic) opinion is - why not wear a helmet? It will offer some protection. It’s also more socially acceptable for a cyclist to wear one than a pedestrian or driver.

Some gaps in that report anyway, unless I am missing the point.

A car crash is much more likely to be reported (blame, insurance…), whereas a pedestrian or cyclist falling over will probably not. If I fell off my bike and hurt my arm, I would just get back on. If I was in a car crash and did the same, I would go to a hospital.

A cyclist falling off a bike at, say, 20 mph, is more dangerous than a pedestrian tripping up at 3 mph.

If you are a cyclist in an accident that will most likely kill you, then perhaps a helmet is only a small factor. But in an accident that is less likely to kill you, the helmet will probably offer more protection. (ie head hitting floor = pain and blood).
Posted by: David White | July 9th, 2008 at 8:35 am | Report this comment
I am 55 and have been cycling all my life. I cycled 18 miles to school and as a young man cycled from Colchester to Ipswich for a while daily on the(then relatively new) A12 dual carriageway. That is impossible today because of the increased size of vehicles (esp HGVs), volume of traffic and I would say, decrease in respect for other road users.

If you cycle regularly, it is inevitable that sooner or later you will be cut up, knocked off your bike, forced onto a pavement or into a ditch/hedge or simply fall off avoiding an incident. When you do, you find the classic tumble involves hitting the ground with your elbow, then shoulder, then head. The head is to some extent catapulted onto the ground by the moment of force around the neck.

My friends would not say I am averse to risk - on the contrary I am a also a qualified pilot and motorcyclist. My advice - if you want to avoid unnecessary pain and injury - WEAR A HELMET.
Posted by: Andrew Phillips | July 9th, 2008 at 9:27 am | Report this comment
If like many people you are commuting by bike to work every day you might see the helmet as more than just an unnecessary accessory. I have had a couple of accidents on the busy and when wet very slippery & greasy London roads. On one occasion I dived head first and without my helmet I would have seen the inside of a hospital. Having landed on my precious helmet from all angles it was mostly my knees suffering and not my head and face.

You might now think I am not careful on the streets or go too fast, but there is nothing you can do when a dog runs onto the street and you do not feel like running it over, a car does not see you as it rushes out of a side street or you just went around a corner that little bit too fast.

Personally I would not recommend cycling without a helmet as it certainly does make a difference and makes me feel safer and brave enough to face the challenge every day.
Posted by: Mona Steininger | July 9th, 2008 at 9:39 am | Report this comment
Clearly FT columnists should keep their helmets on at all times: you just never know what is going to be thrown at you next.

Here is where this recent flurry started:

http://www.ft.com/cms/s/0/f5295332-46c2-11dd-876a-0000779fd2ac.html
Posted by: Stefan Stern | July 9th, 2008 at 10:47 am | Report this comment
Frankly, the answer is obvious. Every cyclist should wear a helmet but there should not be a helmet law.
It appalls me that alleged analysts such as Mr Wardlaw can’t understand even the most basic statistics and, as ever, twists the numbers to suit his argument.
My own reason for wearing a helmet while cycling and not while walking is that, as David White points out, 20 mph is faster than 3 mph. I would add that getting mown down on the pavement is rare while on a bike, commonplace. More than that, I have a great faith in my ability to walk, far less so in my cycling abilities.
Posted by: Nick Walmsley | July 9th, 2008 at 12:12 pm | Report this comment
Surely the question is - if one is involved in an accident whilst cycling will a helmet lessen serious damge to one’s head? It cannot make the risks greater and nor can it make no difference; the answer must be ‘yes’.

I am a London cyclist and always wear a helmet and so far my theory has not been tested by me. Statistics provide no protection.
Posted by: Tim Simon | July 9th, 2008 at 2:07 pm | Report this comment
I’m a US cyclist and a statistician.

I wear a helmet because it provides protection in some cases (banging your head against a low hanging branch, for example), because my wife likes it, because I made the kids wear them, and because my peer group (the Evanston Bicycle Club) requires them.

But as a statistician, the statistics (such as those cited by Wardlaw above) are pretty clear. There’s a minimal overall safety advantage to helmet wearing. If there’s a big effect, you’d be able to see it. If there’s only a small effect, it’s not important to worry about (relative to other things, like teaching cyclists to ride safely).

The “magic hat” culture is pretty deeply ingrained. When I had an accident 6 years ago, I continually kept hearing from medical personnel “it’s lucky you had your helmet on”. I patiently explained that since I hadn’t hit my head in any way, helmet wearing was of no consequence whatsoever.
Posted by: ZBicyclist | July 9th, 2008 at 2:23 pm | Report this comment
In response to ZBicyclist.

Part of my point was that statistics would only (in reality) cover the extreme scenarios for a cyclist - being killed or seriously injured - but not for minor accidents. That is - there is likely to be lots of data on cycle deaths, but very little on cyclists who fall off, get back on, even if they went home sick.

As I stated, a helmet is unlikely to save you getting run over by a lorry (statistic: there is no correlation for the survival rate of cyclists who wear or don’t wear helmets when run over by lorries) but they are likely to help when slipping on a wet road even if all they did was to stop a nasty graze, or concussion.

This gap in statistical data doesn’t lessen the effect that cycle helmets could have - in fact they could be skewed (this is an assumption I am making!) in favour of cycle deaths, because again cycle non-deaths are probably not recorded as often.

So in my view, the statistics are not clear.
Posted by: David White | July 9th, 2008 at 2:41 pm | Report this comment
Typical example of misuse of statistics. I can show you dents in my cycle helmet that would otherwise have been in my head. The details of this topic is not in the mean but in the tail - if you had an accident which involved a head trauma, would you rather be wearing a helmet or not? Stupid question, stupid debate.
Posted by: Tom McLaren | July 9th, 2008 at 3:41 pm | Report this comment
“Surely the question is - if one is involved in an accident whilst cycling will a helmet lessen serious damge to one’s head? It cannot make the risks greater and nor can it make no difference; the answer must be ‘yes’.”

I don’t think this is actually that clear. Just as cyclists may alter their behaviour when wearing helmets (as Tim points out) drivers also may alter their behaviour when they see cyclists wearing helmets. I can’t vouch for the validity of the results but this study is fascinating reading for any cyclist:

“Traffic psychologist Ian Walker at the University of Bath in England once conducted a similar experiment that vividly illustrates the problem with well-intentioned interventions that backfire.

Walker rode a bicycle equipped with a distance sensor, video camera and a computer. Over 15-to-20-minute periods, he rode with his helmet on, then with his helmet off. He rode some segments three feet from the curb and others closer to the edge of the road. With each iteration, he changed a single variable. In the interest of being rigorous, he even obtained a shoulder-length wig of curly black hair, so that some passing motorists would think he was a woman.

It is not known whether any drivers turned around for a look after they passed the bicyclist. If they did, they would have been puzzled: “It was slightly embarrassing, because I had a beard at the time,” Walker said. “I spent a couple of days going up and down the road wig on, and a couple of days going up and down the road, wig off.”

Walker was trying to figure out whether his interventions changed the way drivers passed his bike. He came to two conclusions: Cars gave him more leeway when drivers thought he was a woman with curly black hair. And they gave him less leeway — getting dangerously close — when he wore a helmet.

Walker thinks drivers are influenced by unconscious stereotypes — they may believe that female bicyclists are less steady, and that helmeted bikers are pros.”

http://www.washingtonpost.com/wp-dyn/content/article/2008/06/08/AR2008060801816.html
Posted by: Anon Cyclist | July 9th, 2008 at 7:23 pm | Report this comment
Can’t this be resolved by a simple cost-benefit consideration? Here in the U.S., the cost of a decent helmet is roughly $50. (Plus the implied costs of carrying it around with you and looking like a dork.) Let’s say your chances of involvement in an accident that is fatal if you’re not wearing a helmet are 1 in 1,000. If you value your life at $50,000, the relevant expected value would be $50. Assuming you value your life at more than a decent year’s salary, you should wear a helmet, right?
Posted by: Alex | July 9th, 2008 at 9:14 pm | Report this comment
If you bike regularly at high speeds, would a helmet be worthwhile? It was pointed out to me that eventually you or someone riding with you may have a tire blowout.
Posted by: aaron | July 9th, 2008 at 9:35 pm | Report this comment
Let’s look at what are the incentives for wearing a cycling helmet. Perhaps this is a stretch, but I for one am intrinsically motivated to wear a helmet when I ride. I am not sponsored by Giro, so I receive no financial gain. I tend to think that I wear a helmet in part because I want to set an example in my community (where I see parents and there children riding sans helmets) that wearing one not only limits risk, but is socially acceptable. Or perhaps it’s because of the fact that as a hyperactive child I had to wear one (and a harness) whenever I ate chocolate and now view it as a sort of security blanket.

The real question is: Which is safer, riding on streets with vehicle traffic, or on paths? I vote for the former…and always with a helmet.
Posted by: Tim | July 9th, 2008 at 11:20 pm | Report this comment
Some of you are missing the points in that letter.

The important part is summed up well by ZBCyclist

“There’s a minimal overall safety advantage to helmet wearing. If there’s a big effect, you’d be able to see it. If there’s only a small effect, it’s not important to worry about (relative to other things, like teaching cyclists to ride safely).”

And it is not at all clear that there can’t be negative effects from wearing a helmet. Head rotation being one are which needs more investigation.

Cyclists shouldn’t bother to wear helmets, but it is individual choice. There certainly should not be any mandatory law or public pressure to wear helmets.
Posted by: Neil | July 10th, 2008 at 10:10 am | Report this comment
“I want to set an example in my community” - but what example is that if the perception of it’s safety benefits are so overinflated compared to reality?

Helmets don’t contribute much to safety and concentrating on “did they have a helmet” is couterproductive. What about training? What about good highway design? What about increasing cycle numbers? All these are much, much more effective at improving safety for cyclists.
Posted by: Neil | July 10th, 2008 at 10:16 am | Report this comment
One final comment - why do some of you persist in the belief that a helmet will save your life and the faster/more vehicles involved the more important it is to wear a helmet?

If you are biking at high speeds you are likely beyond the design parameters of the helmet for speed (I believe 12mph) and no other vehicles involved. Remember that at 24mph you have 4 times the energy than at 12mph.

If another vehicle is involved, the mass of a car is many times more that bike+rider and therefore again will most likely far exceed the cycle helmet parameters.
Posted by: Neil | July 10th, 2008 at 10:22 am | Report this comment
I have heard the 12mph impact speed limit argument before. At face value, it seems reasonable: if you are travelling on a bike at 20 mph and are in collision with a motor vehicle travelling at 30 mph, then your helmet may not do much good. However, I might point out that the highest grade motorcycle helmet has a design impact speed of 16.8 mph. Given that a motorcycle helmet is far more enclosed (and offers more protection) than a bicycle helmet, you can still argue that a motorcyle helmet is of no use when in collision with a motor vehicle. However, it is now generally accepted that motorcycle helmets do reduce head injuries.

Having said that, I’m not convinced that bicycle helmets are beneficial. While case studies seem to show a significant reduction in head injuries, whole population studies in countries that have introduced compulsory helmet wearing do not show that the RISK of head injury is reduced.
Posted by: SteveR | July 10th, 2008 at 1:35 pm | Report this comment
It is some of my fellow readers perception that the safety benefits of wearing a helmet are overinflated and is certainly their right not to wear one. But I gather that my reality is different than theirs, i.e. the context of my cycling environment. Plus, when it comes to taking safety measures that are in my control, I too am risk adverse and will boldly rock my helmet. In fact, I may wear it in the shower tomorrow…because I hear it’s a good way to clean it. Cost equals foregone opportunities Stay vertical!
Posted by: Tim | July 10th, 2008 at 3:57 pm | Report this comment
What amazes me in this debate is the arrogance of the ignorants: if you wear a helmet, you are saying that you are smarter than tens of millions of Dutch and Danish people: how arrogant of Brits and Americans! The questions one should ask are:
1. Who is an expert in cycling and what are they doing?
2. Does it work, i.e. where are the mortality rates higher?
You Anglo-Saxons need to understand, that there are people that do things much better than you guys: try to imitate them and you will find that you will live longer.
Posted by: Andrea Casalotti | July 10th, 2008 at 11:13 pm | Report this comment
There are two sorts of studies here: those which document the differences in risk between helmeted and unhelemted cyclists, and seek to infer what would happer if the latter started wearing; and those which measure what *actually* happens when people *do* wear helmets.

The inescapable conlusion is that coercing people into wearing helmets, does not turn them into the kind of cyclist who wore a helmet anyway. The head injury rates in countries like Australia and New Zealand remained exactly the same after helmet laws led to doubling in use rates in a single year.

Luckily, there has never been any credible evidence that cycling is either unusually dangerous or unusually productive of head injuries. So: on your bikes, and I for oen don’t give a toss whether you wear a foam hat or not.
Posted by: Guy Chapman | July 11th, 2008 at 9:01 am | Report this comment
Malcolm Wardlaw is demonstrably right and the evidence is extremely clear; the risks of cycling are very low, and helmets are not effective in reducing those risks. The question should be why are some people are so determined to make cyclists wear helmets, despite the evidence?

The only demonstrable effect of cycle helmet laws and promotion is to reduce the number of cyclists. Since there is no benefit from such measures, the resulting reduction in healthy exercise by a large number of people means that there is a massive public health cost.

Another question which might usefully be asked is why is no-one in the media reporting the failure of helmets to protect, even though helmet proponents have made huge, and completely disproved, claims for their efficacy?
Posted by: Richard Burton | July 11th, 2008 at 6:07 pm | Report this comment
Most people seem to be assuming that the only way to define an accident is either as fatal or non-fatal. Tom McClaren makes a good point about the dents in his helmet, and I’m glad somebody did. “Some reduction in scalp lacerations” from wearing helmets is good enough for me, because what happens when somebody opens their car door and you ride into it or you inadvertantly hit a pigeon? You’re highly unlikely to die, but splitting your head open or suffering from concussion are surely plausible outcomes.

I had a bit of a chuckle at Andrea’s example of the Dutch and the Danish. Last summer I was in Amsterdam and riding on the back of a Dutch girl’s bicycle (we were both pretty drunk) and she took a corner too sharply, went over the handle-bars (catapulting me off in the process), grazed her chin and hit her head. So not exactly the best example for us Brits!
Posted by: Andrew | July 11th, 2008 at 9:59 pm | Report this comment
[…] having more fun than the lady on her cell phone in the car."Bike folds into backpack.Helmet debate at the Financial Times. Man rides bike. Thugs try to take bike. Man refuses to hand bike over. Man […]
Posted by: LED Headlights for your bicycle - Cult of the Bicycle | July 11th, 2008 at 11:26 pm | Report this comment
I as a motorcyclist look at this debate and think that the anti helmet wearers are nothing more than ignorant baffoons. If they do not want to be told what to do then simply admit that and dont try to be clever with ignorant comments and twisting statistics about safety.

For your information, motorcycle helmets are only tested at between 9-16mph impacts. So that makes the bicycle helmets 12mph impact capabilities rather impressive.
Motorcycle helmets are not expected in the majority of cases to have perpendicular impacts, if they always did they wouldnt be much use, they have glancing blows and abrasive impacts the majority of the time. No-on questions the effectiveness of motorcycle helmets. Broken necks and internal organ damage occurs in higher speed accidents cyclists will be no different.
You seem to know better than motorcycle helmet developers as all of your impacts are going to be perpendicular listening to your comments.

All figures chosen are selective by these idiots.

In America, in some States helmets were imposed on motorcycle riders then removed, the crash rates increased and deaths increased between 40-55%. Therefore the safety gear and more casual riding theory is blown out of the water, but of course bicyclists are different??

How can anyone calculate how many times a rider wearing a bright cyclist helmet will have been spotted by a motorist and not if they hadnt been wearing one? How can anyone calculate how many helmeted riders fell off grazed themselves lightly and bashed their head on the helmet area, were ok but shaken and then simply bought another hat and continued with their hobby without feeling the need to report it???

Motorcyclists only have to wear helmets by law, they could ride their bikes in their boxers if they wish, but they dont, they wear collar bone protectors, back protectors, knee protectors, armoured gloves, bike boots with plated soles, toe protectors, ankle and shin protectors and so on, why? Because it hurts when you fall off, you bash your toes or cut your feet once and boots all of a sudden look like a good idea instead of a pair of sexy designer boots.

Those of you who ride your bicycles a lot, wear safety gear that you are comfortable with, we see you in brighter clothes, spot your helmets first over brows of hills and if you fall off then you will be glad you had them on, gloves and helmets a minimum, I think the casual sunday mountain bikers learn like us, fall off once then safety over style.
We have a saying, it is always best to learn from other peoples accidents, not our own.
Ignore those ignorant people twisting facts, and note the anti helmet people before and after a mishap.
Posted by: Ian | July 13th, 2008 at 2:38 pm | Report this comment
[…] FT.com | Should we cyclists bother to wear helmets? Mark me in the helmets are a nuisance category. Depends what you are doing. Not same for Motorbikes, though. (tags: bicycle cycle cycling helmet bikes bike) […]
Posted by: dropsafe : links for 2008-07-16 | July 16th, 2008 at 9:32 am | Report this comment
I cycle through London twice a day and wear a helmet as I’m sure it would protect my head should I fall onto it. However I wonder whether the helmet might cause more damage to my neck if I was unlucky enough to hit the ground or kerb at a strange angle. I also mistrust the straps, as I could imagine myself choking if I got caught up in a nasty position. any ideas, anyone?
Posted by: Helen Carmichael | July 17th, 2008 at 12:33 pm | Report this comment
Helen, re neck damage, again these anti helmet people have been muddying the waters, with motorcyle helmets there is a danger because of the higher speeds involved, therefore this danger is allowed for in their design, air vent controls snap off on impact etc.
With cycle helmets this risk is extremely low and been overplayed by people who just do not want to be told they have to wear a helmet.
The helmet needs to be the correct type for your activity and needs to fit properly. As with motorcycle helmets they are designed for different activities.
The anti helmet people seem to try and infer that the speed the transport goes, is the speed of bodily impact, it rarely is, usually the body slows down before impact, so many motorcycle first body impacts are under 20mph, bicycle body impacts are probably in the average of 5mph or a little over the minimum speed to which a body falls.
To break your neck you would need to hook on something at speed, if the hook is strong enough to grab your helmet and support the weight enough to snap your neck then it would easily damage your skull if the helmet hadnt been there, manhole covers etc couldnt do it and there isnt enough leverage (overhang) on a normal cycle helmet to do that so I cant see how it could happen on a kerb, however a crushed skull is easy and many legs are broken over kerb stones, I would guess long hair tangling on branches during an accident would be more likely to snap a persons neck and would create a higher risk, I know of people who have had nasty accidents with tangled hair.
The straps have never been a problem on motorcycle helmets and they ARE tight.
Safety gear is good not bad, I still say the best thing about a helmet is visibility get a splash of yellow, green or other bright colour in it.

The two issues are completely different in my opinion.
If helmets should be compulsory - I have no idea.

Are cycle helmets safer than not wearing one?- absolutely.

The arguments against are mad, you would never hear someone in a factory arguing about the merits of safety spectacles - but they dont protect my feet!! Or that wont save me if a one tonne pallet falls out of the sky from 20 feet and lands on my head!! No kidding?

It is exactly the same as motorcycling, you hit your head off a lampost or car bodywork at 7mph the result will be the same-hospital, in towns bicycles are usually going faster then I am on my motorcycle.
Posted by: Ian | July 18th, 2008 at 8:42 am | Report this comment