Steady rider wrote:So what are you going to do at this conference?
We know its a difficult topic, so examining one aspect in detail may lead to recommending more research or recommending warnings be included in some publications.
That it's difficult should be enough for the warning. But if you've got lots of inter-related factors then the whole point is you can't "examine one aspect in detail". That way lies 85% cuts in head injuries...
Steady rider wrote:The accident rate could be assessed by;
A) comparing wearers to non-wearers in some form
B) comparing injury stats following a higher proportion wearing helmets
C) considering the physics and applied science to wearing a helmet
D) risk compensation considerations by cyclists or drivers or guardians
E) The issue of extra impacts plus any effects on head rotation and linear accelerations could be considered.
A) suffers as already noted. To do it right you need to be sure that the wearers are identical in all respects to the non-wearers except for their helmets, and even if you could do that impossible thing you can't have a properly randomised sample. If you're testing a drug you won't alter your behaviour much according to if you might have a placebo, but if you knew you might be wearing a fake helmet you'd be rather likely to do things as if you were sure you were. And so on. Read the critiques of You-Know-Who... it's just as hard proving they're a problem as proving they're a benefit.
B) is better, but we've been there already and it's pretty broad brush. We get to find out mass effects but we can't detect ecological fallacies (those who benefit as individuals even though the overall population is worse off, or vice versa), so we can't say if a random subject is better or worse off. We also have quite a lot of noise (reporting standards vary, for example), and because only serious injuries are reported to any degree of usefulness they're the only things we learn about (shame really, since helmets aren't designed for these but for minor stuff, and minor stuff's the most likely outcome of the loss of balance you want to look at).
C) loses the psychological aspects. Go and read up on seatbelts.
D) devilishly hard to pin down, like a lot of psychological things. At least in part because it doesn't work predictably on a given subject. Take, for example, two archetypal helmet wearing groups, the risk averse, wearing because they're particularly worried, and the risk seekers, wearing so they can do riskier things. If you just identify someone by "is wearing a helmet" they could be in either.
E) They could be, though at the end of the day you''ll have to decide how many (probably relatively common) abrasions and headaches are worth a (probably relatively rare) twisted neck, and even if you can do that then the lack of decent data about how many abrasions and headaches you're actually getting rather throws a blanket over that one. And so on.
Steady rider wrote:From this approach, fine tuned research may seem possible to try and quantify or establish effects and if suitable input came from a one day conference, recommendations may be possible. Universities may be able to build on this basis. If on the other hand the issue was still unresolved and the potential higher accident rate not fully explained, then there would be a weak basis for insisting on wearing helmets from a safety point of view. Some people may gain in having a better understanding of the pros and cons of helmets, depending on the information provided and input from those attending.
There already
is a weak basis for helmets from a safety point of view. The problem is not the (lack of) evidence, it's that it's a political matter decided by people with no idea but who think they have a very good one underpinned by their "common sense", with wide support of a "common sense" empowered public. From an evidential point of view all you
should need is what you already have: for serious injuries around zero plus or minus error bars across populations. That the effects are so small are why they are, to quote that editorial again, "too modest to capture".
Pete.
Often seen riding a bike around Dundee...