I deal with matters related to traffic safety in the course of my work, so I can address this point with a certain level of professional credibility. The numbers you've quoted there are absolutely meaningless, and provide a clear illustration of why statistics can easily be twisted to convey a political message to support a political agenda.
If you look carefully at the methodology used by the NHTSA to estimate "alcohol-related accidents," you'll find that these statistics include many accidents in which alcohol may not (and in some case, absolutely DO NOT) have any causal influence at all. An "alcohol-related accident" is one in which one or more of the principals involved (driver, injured party, or fatality) is deemed to be under the influence of alcohol, regardless of whether the alcohol played any role in the accident -- and regardless of whether the person under the influence of alcohol was operating a motor vehicle.
Just consider these examples . . .
1. Suppose I go out with three friends of mine who all get drunk, and I drink nothing but club soda all night long because I have every intention of serving as the group's "designated driver" for the night. If I get in an accident on the way home (through my own fault, or otherwise) and one of my drunk passengers is injured or killed, the NHTSA considers this an "alcohol-related accident" even though none of the drunk people were driving the car.
2. If I'm driving home alone at night (totally sober) and I run over a drunken pedestrian who stumbled into the street in front of my car, the NHTSA considers this an "alcohol-related accident" even though the only person under the influence of alcohol was the pedestrian.
3. And here's the most nebulous case of all -- which is also the best argument against ever getting behind the wheel of a car with any alcohol in your system regardless of how alert and clear-minded you might be. If you are driving with a blood-alcohol level of around 0.06% and some incompetent (but sober) @sshole runs a red light and causes a serious accident, the NHTSA considers this an alcohol-related accident even if the sober driver was fully responsible for the accident. And the driver who blows the 0.06% on the breathalyzer in this case will likely be charged with driving under the influence, too.
I know these scenarios are all included in the statistics, and I would like to see a breakdown. But I'm quite sure that a solid majority of the included fatalities involved a drunk driver.
The problem with all activist groups is that after they succeed in their goals, their new goal is simply to survive... even if they have to artificially create a new reasons to.
http://www.duigulag.com/agenda.htm
Even the principal founder of MADD has left the organization, citing a lack of focus on the real public safety issue, that of getting high BAC chronic/alcoholic drunk drivers off the road.