From what I've read from the links in this forum, my impression of the stats comparing rural cycling versus city cycling is that they are in a sense misleading. If my memory serves me right, it's something like 40% vs 60%, more or less...
I wonder if a more revealing statistic would be number of motorists encountered. If we were able to account for that factor, I believe we would find that rural cycling is multitudes more dangerous than urban cycling. As in a city, an hour ride may have thousands of "car encounters" whereas in a rural area, on hour cycle my only reach a dozen "car encounters". I would bet that reassessing all collisions in relation to number of "car encounters" would give us an entirely different outlook on rural cycling.
The statistical approach of any question is always debatable, I think it's important to explore the different options for assessing risk.
Have any of you encountered this type of assessment?