Yahoo Autos In a first, stoners outnumber drunks on American roads
MORE AT CARS.COM
Two new studies from the National Highway Traffic Safety Administration show that while drunken driving has decreased over the past decade, incidents of driving under the influence of marijuana or prescription drugs has increased.
NHTSA's 2014 Roadside Survey of Alcohol and Drug Use by Drivers, released on Feb. 6, found that the number of drivers with alcohol in their system has fallen by a third since 2007. But the number of drivers testing positive for drugs that could interfere with driving has increased to a nearly 1-in-4 ratio.
NHTSA collected data by setting up 300 voluntary roadside survey sites across the country, paying drivers for their time and assuring that the data collected would be anonymous. During nighttime weekend hours, 8.3 percent of drivers had alcohol in their system, and 1.5 percent had more than the legal limit. That's down 30 percent from a similar 2007 survey. But the portion of drivers with evidence of drugs in their system (both legal and illegal) climbed from 16.3 percent in 2007 to 20 percent in 2014. Drivers specifically with marijuana in their systems climbed from 8.6 percent of the group in 2007 to 12.6 percent in 2014.
A second study, which was conducted on more than 9,000 drivers in Virginia, analyzed the association between marijuana usage and crashes. It was a bit more inconclusive: Pot users are 25 percent more likely to be in a crash, NHTSA found, but they're also part of a crash-prone group — young men.
"Once we controlled for those demographic factors, we did not find a significantly higher crash risk among marijuana users as compared to those who did not have marijuana in their system," NHTSA spokesman Gordon Trowbridge told Cars.com.
Still, NHTSA warned the study couldn't control for factors like how much marijuana that users ingested, or how potent the dosages were.
You're also probably wondering how NHTSA got statistically random samples at its roadside survey sites, given a driver who had been drinking alcohol or getting high seems less likely to pull over for a voluntary test, regardless of the pay or assurances of anonymity.
"This is obviously a factor," Trowbridge conceded. "But with a long history of survey data with the same limitation, we can compare trends over time. We obviously can't know for sure how many impaired drivers bypass the survey sites, but we do know that once people enter the site, they are highly likely to participate, even if they are impaired."
Trowbridge went on to explain that researchers offered impaired drivers a ride home or a hotel room for the night without any legal trouble. NHTSA says that in four decades of surveying, no drunken drivers have ever driven away from a survey site.
The agency says it plans future studies with Washington state and the National Institute on Drug Abuse.
[Related video: Survey: among teenagers, e-cigarettes up, marijuana surprisingly down]
