I see what you're trying to say here, but the study gets its mileage data from the US Department of Transportation Federal Highway Administration’s (FHWA) 2002 highway statistics, so it's an estimate of the total number of miles driven by each category of vehicle. I think the bigger problem with using this study to say that motorcycles are worse than cars is that the "3.77x more likely to kill a child per mile" is based on 4 deaths caused by motorcycles that year. We're dealing with numbers so small that one accident caused or prevented could swing the "probabilities" wildly in different directions.

Here's a link to the full study if you're interested. You're right that it doesn't seem to cover injuries though.