I did *not* say peer review means the article is true. I said it means the background of the author should not matter much - i.e. the peer review ought to catch any naïve errors. It is my impression that the journal appears to be reasonably reputable in the field, e.g. the impact factor isn't bad at 1.87 (for whatever that's worth). It's worth noting Elvik is its current editor. I can not establish whether or not that was the case when he had that paper published. Looking at the other editors, they seem to include people from quite reputable organisations, including the UK's Transport Research Laboratory. Elvik himself works for reputable state transportation research organisation. He seems to be quite active in the field. He does have publications elsewhere - PubMed need not be representative of research in transport safety. Accident Analysis & Prevention is somewhere he publishes a lot though. That's not per se a bad thing. It could be that that journal is the focus of his field - particularly for its top contributors. I have seen other papers of his, unrelated to helmets, discussed and they seem quite sensible and appreciated by others.

As for the Cochrane review, see my other comment about which one to believe (if either): http://slashdot.org/comments.pl?sid=3157085&cid=41538937 . I do know what the Cochrane review is, and its reputation. It is usually the first source I look to, to see what the most credible work says on whatever subject. That said, the Elvik meta-study is certainly interesting as an inclusive meta-study, to compare with the exclusive Cochrane version.

I'll be honest, I don't know which one is correct. I don't have the expertise to say. However, generally it is better to accumulate bias & errors over more studies in the hope they are distributed evenly, and so should cancel out (this is what meta-studies are about, no?), than to try exclude studies subjectively and potentially introduce a systematic bias.

So I don't know whether it was better to include or exclude studies specifically. Lacking the ability to judge the specifics, my instinct is to give the more inclusive study more credence.

Well, at least you're consistent then. Wrong, I think, but consistent. :)

crash often enough I don't bother with bicycle helmets which are designed to disintegrate ablatively. I wear a skater's helmet, which I could bang with a hammer, &c., without damage.

Actually, the expanded polystyrene helmets are supposed to absorb force by crushing, from the inside. If they break, it means they failed. Motorcycle and skater style helmets are much better because the smooth, hard-shell helps spread the force across more of the polystyrene, thus allowing more of it to crush, and helps prevent localised shear failures. For anyone who feels they need good head-protection (and for some sports, that might be wise), I strongly recommend choosing a hard-shell helmet over the soft-shells - with as few ventilation ports as you can bear.

You are basically repeating my point: You can not generally correlate the extent of the damage with the efficacy of the device. As you point out, the skater's helmet suffers *less* damage (still in one piece) than the soft-shell! Yet it protects as well, indeed it should protect *better*! Also, note that you're not just using the damage as evidence, you're also going by the impact you *felt*.

Again, my point was about the logic of your argument. "My helmet broke, therefore it saved me" is simply bogus logic.

You continue to assert things that you do not have proof for. E.g. it is quite possible that some of the impacts you have may have been fatal without a helmet. You can not know that without repeating the exact same crash without a helmet - which I doubt you'll want to do ;) - however I'll grant it's quite possible.

However, you still can not conclude that had you taken those cycles without a helmet that it would have been fatal. Because you likely would have changed your behaviour had you not been wearing a helmet. You would have felt less protected and you likely would have taken fewer risks. You might never have gotten into that impact situation as a result. Further, there is strong evidence that says that people are bad estimating risks, and specifically that safety-devices can lead them into over-confidence. Wearing a helmet may have made you take more risks. Your head may have been protected somewhat, but other parts of your body have *0* protection from that helmet. Indeed, to the extent that a helmet makes you comfortable with more risks, that helmet is then actually *harmful* to the other parts of your body!

Finally, based on what you say about crashing bikes and high, possibly near-fatal impacts, I'm thinking that probably you're into BMX or mountain-biking, and doing high-risk jumps and/or downhills. In that case, yeah, wear the safety equipment you think you need. However, know that many cyclists don't do this stuff. They're just pottering around town to get from place to place. There isn't a 1-size fits all. Further, there is evidence that suggests that helmet compulsion depress rates of utility cycling (which constitutes the *mass* of cyclists, anywhere where cycling is not a minority activity).

Certainly, it is a fact that helmets are NOT a prerequisite for safe, mass utility cycling - just look at the Netherlands...

I hope you wear a helmet when you walk, when you get in the bath, when you.. etc. Otherwise, you're being silly.

Yes, there are many factors why the Netherlands has good safety. I say you should educate yourself as to what those factors are. A lack of safety-wear is perhaps not a causal factor in the Netherlands, I would agree.

However, I believe there is reason to think that the heavy emphasis on safety-wear in the Anglo-Saxon world acts to actually decrease cycling safety (in those parts that adopt mandatory laws - it has a proven depressing effect, NB). It has made little positive difference to injury. At the same time, the increasing cultural "push" that cyclists should wear such equipment has gone hand in hand with a decline in cycling (levelling off recently - there may even be small gains in some places, like London and Oregon - but they're still tiny relative to the decline in the last decade or three). I believe, though I can not prove it, that this cultural "push" has two effects:

a) It makes cycling seem dangerous.

Look at the street, you see pedestrians, buses, cars, cyclists, motorcyclists. With just *2* of these forms of transport, people are urged to don safety equipment. One is actually extremely dangerous. The other, in statistical terms it is not appreciably more dangerous than those where people don't think of wearing safety equipment. Depending on whether you measure by distance or by exposure time, cycling is just a little less or a little more dangerous than walking by the road (in the UK). Cyclists going around bedecked in safety-equipment - unlike many other day-day, common forms of transport, re-inforces this mistaken view.

b) It distracts from the *actual* issues that affect cyclist safety

Again, the Netherlands has significantly better safety - for all road users - than the UK and USA. It achieves this without cyclists wearing special safety equipment. Yet in many Anglo-Saxon countries, the discussion about cycling safety seems to focus heavily helmets, on how irresponsible cyclists are who don't wear them, on whether they should be mandatory. Helmets however have now been *proven* to not improve cyclist safety in the real-world (no matter what tests in labs with metal head forms might tell you - we simply havn't seen any improvement in cyclist safety in the real-world). Yet we keep talking about helmets, distracting the debate away from what has been proven to work: Separate cycle paths of sufficiently high quality that cyclists want to use them; 30 km/h or lower speed limits on any roads forming routes that do not have such separate paths.

Your attitude that you just do not care saddens me, but it explains a lot.

You may think its worthwhile having fewer idiots on the road - but many of them will go drive cars instead, plus you'll find the politicians care less when you need to ask them to do things to make the roads safer for you.

PS: falling off bicycles at 40 to 65 km/h is not going to be fun. I've done it from a motorbike in jeans and winter jacket and a *much* sturdier helmet than a bicycle helmet. No matter how good your helmet, other parts of your body may hurt - a *lot*.

I find it tragically amusing to be lectured on bicycle safety, by people who then often take greater risks than I will on downhills. Yet I'm the reckless one for not wearing a helmet.

13 to 20 cm - perfectly consistent with what I said: No higher than what a person can reach standing on their tipy-toes. Bear in mind that when people run, they will at times be elevated above their normal height, as their foot extends to launch them forward and to cushion their next footfall. Further, that height difference is several times smaller than the standard deviation of height in the population.

To claim this constitutes a significant difference energy wise between pedestrians/runners and cyclists is not credible. If it were, it would mean taller people should be more vulnerable to head injuries, and we would have fairly clear accident/hospital data to show this (I'm not aware of such). Further, tall pedestrians would then be in far *more* need of helmets than shorter utility cyclists.

Do you go around telling tall people they need to be extra careful about falling, and they should wear helmets?

It's cancelled out not in economic terms, but in injury terms. Overall, in terms of head, neck and face injury, there is no net benefit to helmets. Given that there is some benefit in terms of head injury alone, this means neck and face injuries are increased.

You could argue you would prefer a neck or face injury to a head injury though.

It's not in the abstract, but this is in the conclusion:

When the analysis is updated by adding four new studies,
the protective effects attributed to bicycle helmets are further
reduced. According to the new studies, no overall effect of bicy-
cle helmets could be found when injuries to head, face or neck
are considered as a whole.

As for who wrote. Well, I too am usually a little more sceptical than usual of arguments made by people working outside their field. Some such people can be quite brilliant though. An academic in our computer science dept is an economist by training, has written history books on socialism, but also has an extremely deep understanding of several fields of computer science, and to which he's contributed (and he generally knows more than most people about most subjects - no joking). So you can't always dismiss economists out of hand! :)

In this case it's an article which has been peer-reviewed, so the expertise background of the author shouldn't matter much. Further, the author appears to have been working in the field of transport safety for a long time. Finally, meta-studies are meant to be systematic, and so follow a well-established formula. There's supposed to be less scope for getting it wrong (though, earlier meta-studies introduced significant selection bias by excluding results, this meta-study argues).

Height, don't be ridiculous. A cyclist will never be higher than they would be if standing on their tipy-toes. Indeed, they should be just a little lower than that, otherwise they have a bike that's far too big for them, or perhaps they're riding a penny-farthing (which, last I checked, are not used much anymore). What rubbish..

Sigh. So your argument is based on an anecdote where you got injured, where it was your shoulder that broke the fall - not your head.

Here's a better argument, based on actual data drawn from many millions of bicycle trips per year - and hence which has actual statistical significance: The Netherlands, which has very low helmet usage rate (next to nil), has low rates of death and injury amongst cyclists. It has much lower rates than countries with high rate of helmet useage, such as the UK.

So these are clear, undisputable facts, backed by a mountain of evidence (unlike your anecdote):

1. Helmets are not a pre-requisite for safe cycling.

2. The approach common in Anglo-Saxon countries, of trying to improve cyclists safety primarily through helmets & hi-viz, has failed to have any significant effect.

If you like cycling, if you believe more people should cycle, if you believe cyclists deserve a safer environment, then you should examine what it is about the Netherlands that is achieving such safety for so many cyclists. It quite clearly is not helmets. Indeed, there is strong evidence that helmet compulsion (through legal means, though I would suggest through cultural norms would have the same effect) devastates rates of cycling, and so greatly harms cycling - fewer people cycling makes it politically even more difficult to change the environment to be cycling friendly for the remaining cyclists.

Stop killing cycling by scaring people off it, by continuing to promote safety strategies that have failed. Start agitating for what is *has* worked: dutch style cycling. Yes, it takes time to implement, but you have to start somewhere - like the dutch did in the late 1970s (their bicycle paths didn't always exist, and didn't spring up overnight).

Most utility cyclists cycle at less than 25 km/h. My wife rarely gets over 20 km/h - even downhill. These are speeds that joggers can do. It is bizarre to argue that such cyclists should wear helmets, unless you also argue joggers should too.

Cyclists who want to go fast, particularly on inhospitable terrain. If they want to wear a helmet, sure. But slow, about town utility cyclists? Stop being ridiculous. Quit trying to kill off every-day cycling.

Correlation != causation.

How many of those cyclists also had alcohol in their system? (For both the USA and the Netherlands, it's a significant percentage, IIRC). Next, how many cyclist don't wear helmets (and you have to actually survey the cyclists that are on the road - not do phone surveys of people affiliated to cycling organisations)? If, say, 95% don't wear helmets, then guess what it means if 95% of deaths weren't wearing helmets.

Re your statistics, I am dubious of the quality of figures sourced from websites that have an agenda. Much better to look at peer-reviewed, systematic meta-studies. See my other comment (and replies) for discussions on some of the ones available:

http://slashdot.org/comments.pl?sid=3157085&cid=41523419

Yes, I'm well aware of how helmets work. And yet, visual inspection of the damage to a helmet still does not allow you to determine how much force, if any meaningful level, was absorbed before reaching the head. If you want to know that, you need to measure the force at the head. Also, the human skull is much, much stronger than polystyrene. A cracked or dented soft shell helmet by no means equates with a cracked open skull without the helmet.

Basically, I am not disputing a cycling helmet absorbs a significant amount of force in certain impact situations. I am saying the logic of "my helmet was cracked, therefore my head was saved" is, of itself, fallacious.