In the linked study, 75% of the fatalities happen at night. 32% of pedestrians struck are drunk, and 13% of striking drivers are drunk. There is mention of drug

In the linked study, 75% of the fatalities happen at night. 32% of pedestrians struck are drunk, and 13% of striking drivers are drunk. There is mention of drug treatment as a prevention method, but no specific statistics for non-alcohol intoxication. It's also notable that fatality rates are only worse for some minorities-- those typically in the lower socioeconomic brackets. And the difference isn't nearly as dramatic as the difference in homicide victims in general. Yes, we need smaller cars, but it's worth noting that the common case isn't someone being run down walking to the train on their morning commute, as the Doerner quote might insinuate without further research.