The difference is "grasped, but when quoting a paper such as Wardlaw in the BMJ, you can really only quote their work. Hence the reason milage is quoted.
The real problem is defining what is "fair" when comparing two groups. Cycling may be more dangerous if you consider time rather than distance, but is it the correct comparison?
Lets compare an person going to the shops on a "sit up and beg" and a racing cyclist going to the same set of shops from teh same starting point.
The distance is the same, so is the "exposure" if we use this as the measure.
If we then look at the time the racing cyclists will complete the distance in half of the time the shopper does, and therefore has less "exposure"
Does that make cycling slowly more dangerous?
It all depends on what you feel to be the correct one.
As with the one petrolhead group proving just how dangerous cyclists are to pedestrians.
If you take the number of pedestrians killed each year by motorists and by cyclists then the numbers demonstrate that vehicles kill many more than cyclists. However this group then corrected the figures for distance and quoted deaths per mile driven / cycled
They "proved" unequivocally that cyclists kill many more people per mile than vehicles and are a bigger danger to pedestrians than motor vehicles.!
