Does sprawl kill?  Looks like it.  This study found that people who live in sprawling counties—places with low population densities and poorly connected street grids, and with rigid segregation between stores, businesses and residences—are more likely to die in a car crash. 

Apparently, living in the sort of place where you can’t get anywhere without a car makes you drive more.  And people who drive more tend to crash more.  (When you put it that way, social science seems pretty simple, no?)

But, of course, the question remains:  how much?  How much more accident risk do residents of sprawling places really face? The answer is surprisingly straightforward: according to the model developed by the study’s authors, residents of Washington’s King County (relatively compact and urban) should face a 20 percent lower risk of getting in a car crash than the people who live in neighboring Snohomish County (relatively sprawling).

That’s the theory at least.  And in this case, the theory matches up pretty closely with reality.  Over the last few decades, the age-adjusted traffic fatality rate in King County has averaged about 20 percent lower than the rate in Snohomish.  (See table E-8, here.)

Or, said another way: if King County sprawled like Snohomish, about 800 additional King County residents would have died in car crashes since 1980.  And since the National Safety Council estimates that each traffic fatality is associated with 54 non-fatal injuries, King County residents also avoided more than 40,000 injuries, just because of how its urban and suburban areas are laid out.

Now, just imagine how many fatalities could have been avoided if King County looked like greater Vancouver, BC.