Slashdot Mirror
Federal Study: Marijuana Use Doesn't Increase Auto Crash Rates
An anonymous reader writes: After the legalization of marijuana in multiple states around the U.S., many are worried about a corresponding uptick in car crashes as people drive while under the influence of pot. But according to a new federal study (PDF) commissioned by the National Highway Traffic Safety Administration, those fears seem unfounded. They report that after adjusting for other factors (people who tend to drive after using marijuana also tend to be more crash-prone in general), there was no statistically significant increase in crash rates by drivers who tested positive for the drug. It's still a bad idea to drive high, but driving drunk is far, far worse: "One substance was shown to have a major influence on crashes: alcohol. The study confirmed the enormous danger of drinking and driving, even after age and sex adjustment: drivers with a 0.05% blood-alcohol level were found to be twice as likely to be in a crash. For a person weighing 180 to 190 pounds, that could be a single can of beer, glass of wine, or shot of liquor. At 0.08% (two drinks), the likelihood is quadrupled, and at .20% (four drinks or more), the risk is higher by 23 times."
Studies more than 40 years old have always consistently shown this, including one I read as a young boy long ago, that showed professional race drivers after mild marijuana intoxication had IMPROVED lap times, though this edge dropped off at higher intoxication levels. Trying to point stuff like this out over the decades had jerkwads accusing me of the most awful things. Whatever I just don't care anymore. Marijuana being illegal while alcohol is not is insanity by definition, but most people are dumb animals and our world is run by sociopaths and there's nothing I can do about it.
The summary butchered the article (surprisingly). What it should have said was that after adjusting for other factors (namely age and sex) in the group that were in accidents, there was not a difference between using and not using Marijuana.
For example young men are more likely to be in an accident (Regardless of drug use), they are also more likely to smoke weed compared to other groups.
I suppose it really is to much to ask that people Read The Fucking Study...
In a nutshell, this study collected drug use data from 3095 drivers involved in crashes and 6190 matched control drivers. THC was detected in 234/3095 crash involved drivers, vs 379/6190 controls. That sample size is plenty. If you think otherwise, please explain why you think the studies' methodology is statistically underpowered.
The biggest caveat is probably that THC testing can be positive even if the drug use was days or weeks ago. I'm not aware of a test that, like BAC, can detect whether someone is high as balls right now. That makes the conclusions a bit weaker, but we can still conclude that people who frequently use marijuana are not riskier drivers than anyone else, and blood THC testing is not a measurement of impairment.
This study from 1993 (mentioned earlier probably) shows that this was already known to federal authorities, but was probably swept under the rug or willingly ignored by legislators for obvious reasons.
http://ntl.bts.gov/lib/25000/2...
You clearly can't read then.
They're not comparing individual people while intoxicated and while not intoxicated. They're comparing the accident rates of intoxicated people to sober people.
If the substance in question impairs driving, there will be an increased rate of accidents among the intoxicated population for said substance.
If there is no such increase in accidents while people are intoxicated on a substance, then the data indicates that said substance does not impair driving.
The data from this study shows:
Alcohol intoxication impairs driving, as shown by the greatly increased accident rate.
Marijuana intoxication does *not* impair driving, as shown by the *lack* of an increased accident rate. (In other words, the accident rate for those on weed is statistically the same as the accident rate of people who are completely *sober*.)