I'd argue it's entirely fair. The fact that people in the US spend a lot more time in the car is part of the problem.

Fatalities per distance is fine for a technical comparison of road safety regulations.

Fatalities per population is more useful when comparing the safety of roads as they are used.