Has new auto safety technology made a significant dent in auto accidents and fatalities?

It’s hard to compare auto accident statistics and take into account how many more cars are on the road, how many miles are being driven, etc. So has new safety technology, like backup cameras and crash-avoidance systems, made a statistical dent in car accidents and fatalities? Has it been offset by things like driver distraction? Has a comprehensive study of this been done, especially one that shows which particular technologies are most effective at reducing accidents and fatalities?

Yes - http://www.census.gov/compendia/statab/2012/tables/12s1103.pdf

Accident and death rates are down and safer cars are part of it but also better driving habits, safer roads, etc.

There are many factors and it is difficult to isolate the effects of a single one from the others.

Imagine this, made-up hypothetical:

People used to drive at 50 MPH with regular, (non ABS) brakes and the accident rate was X.

ABS brakes were introduced and after some time the accident rate is found to be unchanged at X.

Does this mean ABS is useless? No, it means people feel safer driving faster so they are now traveling faster than they used to. It means now you can go 60 MPH with the same safety factor you had years ago with 50 MPH.

Accident rates have gone down but even if they didn’t we are driving today faster and in better comfort than in decades past and that is an improvement in itself.
If 80 years ago you tried driving at today’s speeds you were certain to end up dead.

The Census data at Telemark’s link tell part of the tale: death rates (per mile, per driver, per vehicle, per capita) have all been trending down over time, despite the relatively constant absolute number of crashes per year. This points to improvements in technology that bolster the odds of surviving a crash (e.g. crumple zones, multiple airbags, side impact beams, etc.), given that a crash is going to happen. So the answer to that question is yes, technology has reduced fatality rates.

Telemark’s link shows a fairly constant absolute number of crashes per year, but we need to prorate that against miles driven. Here ya go: the long-term trend is a reduction in crashes per 100M miles traveled, from a high of ~180 (in 1989) to less than 100 in 2009.

My understanding has been that many of the expected gains (lower crash rate and fatality rate) have been offset by changes in driver behavior. See risk compensation: the idea is that people are confortable with a particular level of perceived risk in certain activities, and if you take steps that result in perceptions of safety, they will alter their behavior to obtain greater benefits, up to the point that they perceive the same amount of risk that they started with. Skydiving is the classic example: equipment is far more reliable now than it was 30+ years ago, and as a consequence, skydivers feel the same amount of risk doing crazy shit they would never have dreamed about back then. Cars are the same way. Researchers have demonstrated, for example, that equipping cars with ABS changes how the drivers drive.. Toss in everything else:

-high-center brake lights
-electronic stability control
-all-wheel drive
-the aforementioned crash safety technology

and people these days tend to drive faster, more recklessly, and in more adverse conditions than they used to. It’s impossible to say what the crash rate would be if driver behavior were unchanged over the decades, but it seems a safe bet to say it would be even lower than it is now.

True but this is one more variable which has changed over time and which can be acted upon and influenced. Safety campaigns, public service anouncements, etc. have changed the culture to safer driving habits. Just look at the change in attitude towards drinking and driving. And how they drive in some other countries.

Humans are quite bad at evaluating real risk and our perception is very much cultural. It has taken much repetition to make people “feel” that tobacco is bad for you, that safety belts are good, etc.

OP, what time frame are you talking about?

  1. Another explanation for fewer deaths is better medical technology.
  2. I am unaware of any changes in driver behavior for the better. The two changes I can think of in the last couple decades would go in the other direction–talking on the cell phone and texting while driving.

I’m talking about now as compared to the last few decades. Point 1 is very good- I hadn’t thought of that. I wonder how much that has affected the statistics.