I agree with the parent that distance driven should be in the equation. If there's a certain risk percentage associated with driving as a whole, then it's almos

I agree with the parent that distance driven should be in the equation. If there's a certain risk percentage associated with driving as a whole, then it's almost certainly somehow bound to distance driven. For example, the risk of a car crash should be near zero if the distance driven is also zero. But even in the safest of vehicles, driving 100KM, 1,000KM, or 10,000KM+ each are going to have higher rates of accidents assuming real-life road conditions. In the US, on average, people need to travel further to get to their destinations, and so on average I'd expect car accident and death statistics to likewise be higher. I'm very in favor of reducing urban sprawl, utilizing public transit, etc., but in present state the US has vast quantities of existing road networks which can't be consolidated or mitigated overnight, so the next best way to improve those statistics would be improving traffic controls, vehicles, safety features, etc.