I'm guessing... (and this is just a guess!)
Helmets give the wearer a sense of security (whether that sense is false or not I'll leave to others) and for some it may have the effect of causing the rider to be *less* aware of his/her surroundings. This may result in more accidents with helmet wearers, I suppose.
I'm not sure I agree with that, even though I don't wear helmets for cycling. Just trying to figure out why a set of stats might generate such a result.