I am guessing you got this from here: http://www.cyclecraft.co.uk/digest/helmet_research.html which quotes an earlier 1988 paper but not his 'inconclusive' later paper which, to make things even more confusing, doesn't reference his own earlier paper in the relevant section. Weird.The most surprising finding is that the bicycle-related fatality rate is positively and significantly correlated with increased helmet use"
I can't find an online copy of the 1988 paper or the Journal of Product Liability. It would seem a strange place to publish an important and difficult statistical analysis. Yet I see quotes to it all over the place treating it as gospel.
Well I'm certainly confused. Can those that quote it please lead us through the timeline and the consistency or otherwise in Rodger's view?