OK, so nobody has really answered the original question, me included (see original answer Jan 25th)

The question title "Are cycle helmets effective?" and the question text which relates really to the data presented in the linked article are fundamentally two separate issues. I therefore revise my answer to the following:

Regarding the article and it's relation to the question it looks at data relating to the change in number of cyclist (i.e. reduction) when compulsory helmet wearing is introduced and upon further revision to the legislation (i.e. relaxing compulsion for adults on pavements) leading to an increased uptake of cycling.

Therefore does this article help?

Well assuming the data is reliable all we can take from this article is the fact that legislation for the compulsory wearing of cycle helmets reduces the number of people who participate in cycling.

Does this make cycle helmets effective? Not as a safety device, it has no bearing on whether in an individual collision the helmet would either increase, decrease or have no effect on the likely injury outcome. However a reduction in the number of cyclist will more than likely reduce the total number of accidents. Thus it is effective to reduce number of accidents, but not to affect the likelihood of injury, unless the wearing of helmets was proven to reduce injuries also (Double win!). Therefore we must address the more serious side to the argument.

Does wearing a cycling helmet make cycling safer?

• Helmets are designed to meet limited safety criteria, i.e. Falling from cycle onto onto flat surface or against kerbstone with your head at upto 15mph. They reduce the g-force of deccelaration to below that which would, likely, cause long term head injury (<300g). Given the limited range of collision types and forces applied the efficacy is also highly dependent on the correct and secure adjustment of the helmet's fit. The lightweight materials will also be subject to deterioration over time and if dropped accidentally should be replaced. Whether repacement is done in practice is debatable, thus the likely efficacy of any given helmet in an impact will be compromised against it's original testing standard. I have seen quoted but can't find the link where the reduction in force of an impact is akin to 30mph reduced to 27.5mph. I would imagine at higher combined speeds this margin will reduce to nearly zero.

• A helmet absorbs/dissipates energy by destroying it's own integrity, thus as alluded to above abuse of the helmet during its lifetime will reduce effucacy. Also during a collision with tarmac or kerb it will do it's job given the parameters are within its design limits, however I would propose that most fatal/serious accidents especially with motor vehicles are at energy levels well above the testing standards. It must also be assumed that there may be multiple impacts with the vehicle and/or road surface/furniture. Thus the secondary impacts will be with a helmet that has already suffered catastrophic failure of it's integrity.

  http://www.cyclehelmets.org/1081.html

  http://tinyurl.com/DFT-Cycle-Helmets

  • I have, in my previous answer, spoken of the impact of rotational injuries. It must be noted that these types of injuries are most commonly associated with longer term brain injury.
  • I also mentioned risk taking, this is double edged as a cyclist may take more risks or use higher speeds when wearing a helmet and other road users may take more risks also when the cyclist is wearing a helmet.

So to conclude my answer I would propose that the effectiveness of cycling helmets is entirely dependent on a multitude of risk factors and the conclusion is therefore multipart.

1. Low speed, low energy impacts against stationary blunt solid objects will be made safer by wearing a helmet. Thus I would say a child/person learning to ride, not yet riding on the open road or riding in a BMX park/dirt track/woodland who will likely be going at lower speeds would benefit from wearing a helmet. Even at higher speeds the helmet will still reduce energy so must be beneficial. The only caveat which cannot be predicted is the rotational injury but I would suggest the lower speeds would reduce the risk of this type of injury also. However lower speeds and energy will less likely cause the type of brain injury which helmets etc were hoped to prevent so again we must question if neccessary even in some of these circumstances.
2. Higher speed impacts on public highways with other road users will, likely, generate more forces and less predictable angles of deflection and multiple impacts. In these situations the efficacy of the helmet will be reduced possibly to zero or even negatively if rotational forces are applied.
3. Poorly fitted, old or abused helmets will already be compromised.
4. Legislation is only as effective as the efficacy of the helmet at reducing injury, but may reduce number of accidents as a total. This does not reduce the number per head of cyclist though.

As to the original questioners thought on gut feelings weirdly enough I wear a cycle helmet more often than not, mainly off-road. Ironically I did more stupid things as a youth without a helmet than I would dare do now with one, so I will be modifying my behaviour in line with my conclusions above.

Update (01/03/2010): Some testing on airbags for cyclists