Jdsk wrote:A very rough cut:
Problems to be managed in studies:
the level of risk varies by cyclist behaviour that may change with helmet usage
Drivers behaviour may change by cyclists wearing helmets
Head injury is related in part to cyclist's behaviour
Helmet legislation when enforced can discourage cycling
For the first, behaviour will also change in different ways with helmets. You may have one group wearing because they're risk averse and another wearing because they embrace risk and want more of it.
Second may also vary with other factors, quite possibly related to the fist, so if A Driver comes across someone giving an air of knowing exactly what they're at (e.g. a TT rider in All The Gear) they may feel they can pass close in "safety", while a wobbly pensioner would be treated quite differently.
For the fourth, helmet promotion can discourage cycling too.
Jdsk wrote:Assertions about what is already known from studies:
wearing a helmet increases the accident rate
increases the frequency and number of helmet impacts compared to a bare head
helmet impacts generally last longer in duration than for bare head
Drivers behaviour may change by cyclists wearing helmets
Head injury is related in part to cyclist's behaviour
Helmet legislation when enforced can discourage cycling
Many differences have been reported when comparing the behaviour of injured cyclists for wearers and non-wearers.
The first of these is problematical because there's chicken <-> egg feedback in both directions: it's not a simple causal relationship. So I might take more risks because I'm wearing a helmet, but on the other I might be wearing a helmet specifically because I've chosen to take more risks (e.g., I want to do that descent trail quicker, I want to take that great corner without having to brake first, etc.).
It's also the case that risk compensation is in relation to perceived risk, so if your "increased risk" as a risk-averse type is getting on a bike at all it could be that you're actually accruing considerable health benefits by doing something you think is dangerous, but actually isn't.
And it's very much the case that different people have very different attitudes to risk and are affected differently by their perception of it, and it's difficult to account for the variation in a study aiming to see how effective helmets are. So on the last of those points, many differences in behaviour can be found in any group you select in all sorts of ways, not just whether they wear a helmet or not.
Given how hard it is to untangle the above, and how small the overall effects seem to be, I do wish people would put that sort of effort in to stuff which they know actually works...
Pete.