Oh No! I've been sucked in!
In the following post *ALL* I would like to discuss is injury chances and helmet use by cyclists - please no distractions along the lines of making pedestrians wear helmets or helmets bad as they reduce cycling numbers arguments!
My question - all of these studies (and I'm now relying on my memory from when I last found the site in the original post and other discussions) are from various places in the world.
Countries which have large volumes of cyclists are - from what I've gleaned - often quite different in terms of cycling facilities, where people cycle, how fast people cycle, and how other road users behave towards cyclists. I'm not even going to go into causality here (or correlation).
Now, one of the main reasons I wear a helmet is that if I do get knocked off (sideways, not over the bars) or blown off or fall off or whatever, the chances of me landing headfirst on the raised kerb is fairly good, I reckon - given where I cycle (urban traffic, for the most part).
However, my mental image of cycling in most other countries where cycling is more common (and, I hazard a guess, helmets less so - say, in those lovely cycle lanes along the highways in whichever Scandinavian contry Origamist was writing up about a while back); it seems to me the paths for cycling are more likely not to have a raised kerb but just to fade into grass or similar. In that case, I would again hazard a guess that if a fall does happen, instead of a head first smash, it's likely to be a shoulder/arm/etc first smash, and in that case a helmet is not going to help or hinder much, surely?
Maybe this is also an urban vs rural question.
My final point is - having had some statistical training, most of these studies are probably not that useful. Statistics depends on large numbers of identical experiments, and accidents are most definitely not all identical. You can test a helmet in a lab, fine, but as for effectiveness in real life...?