THURSDAY, Feb. 9 (HealthDay News) -- Getting behind the wheel within three hours after using marijuana nearly doubles a driver's risk of having an accident, a large new research review finds.

The risk is especially high for fatal crashes, and the risk is only a little less than that of people who drive drunk, Canadian researchers say.

"On the whole, alcohol increases the risk of a crash at a higher level than cannabis [marijuana]," said lead researcher Mark Asbridge, an associate professor in the community health and epidemiology department at Dalhousie University, in Halifax.

But marijuana makes it harder to judge distance and drivers often tailgate and swerve from lane to lane, which cuts down their reaction time and leads to crashes, he explained.

Although the extent of the problem isn't known, some studies have found that 5 percent of people report driving after using marijuana; and for those under age 25, as many as 20 percent, Asbridge said.

Studies on the effect of driving under the influence of marijuana have had mixed results, he said.

"There were some studies finding that cannabis actually had a negative association with crash risk, so people were actually safer using cannabis driving than when they weren't, but these were poorly designed studies," Asbridge said.

"So our study gives some clarity to the issue in showing a doubling of the risk in the very best studies that are out there and adds some level of justification to existing policies that restrict drug-impaired driving," he said.

The report was published in the Feb. 10 online edition of the BMJ.

To see how marijuana affected driving, Asbridge's team reviewed nine studies that included more than 49,000 people. This process -- called a meta-analysis -- looks for patterns across studies.

The researchers found that those driving under the influence of marijuana were nearly twice as likely to have a car crash as those who were not under the influence.

Studies outside the review have shown that drivers aged 35 and younger are more likely to have car accidents after using marijuana, the authors noted.

"These findings reaffirm many of our accepted understandings regarding acute cannabis intoxication and psychomotor performance," said Paul Armentano, deputy director of NORML (the National Organization for the Reform of Marijuana Laws). "That is why operating a motor vehicle while acutely impaired by cannabis is presently a criminal offense in all 50 states."

This risk appears to be greatest in less-experienced cannabis users, younger drivers, and among those who combine the use of cannabis and alcohol, Armentano pointed out.

"That said, it should further be noted that cannabis-induced changes in performance are typically subtle, short-lived and less dramatic in more experienced cannabis consumers, who appear to develop tolerance to some of the drug's behavioral effects," he added.

"Further, this overall elevated risk is far less than the elevated risk of accidents associated with the consumption of alcohol, including its use in legal quantities," Armentano said.

While some suggest that drivers should be tested for marijuana, so far, no effective test exists that can be done at a traffic stop to accurately pinpoint when a driver used the drug.

"This evidence makes a case for introducing policies to reduce cannabis-impaired driving," said Wayne Hall, from the University of Queensland Centre for Clinical Research in Brisbane, Australia. He wrote an accompanying editorial for the journal.

He said that roadside drug testing, such as that used for alcohol, may be a useful approach.

But there are no handy devices, such as a breathalyzer, to tell if someone has recently used marijuana, Asbridge noted.

"The challenge is defining a level that equates with impairment. A number of countries have already introduced roadside testing by deciding that any detectable evidence of recent use constitutes impaired driving. However, we do not know how effective testing has been because the policy has not been evaluated," Hall said.

Another expert outlined the problems with such tests.

"Because THC, the active ingredient in marijuana, can be detected several weeks after use of marijuana, it is hard to determine with certainty if a driver testing positive for marijuana is indeed impaired by the substance at the time of testing," said Dr. Guohua Li, a professor of epidemiology at Columbia University in New York City.

So more research is needed. "This issue is especially urgent and important in light of the ongoing epidemic of drugged driving and increased permissibility and availability of marijuana worldwide," Li said.

While recognition that driving under the influence of marijuana is a problem is a first step in finding ways to curb it, Jan Withers, national president of Mothers Against Drunk Driving (MADD), stated that "drunk driving remains the primary threat to our families on the road."

However, she added, "this study underscores the importance of the work that MADD is doing to support people who have been victimized by drugged driving and recognize law-enforcement's efforts to pioneer effective strategies to stop drugged driving. Notably, it shows the increased danger posed by those drivers using both alcohol and drugs."

More information

For more about drugged driving, visit the U.S. National Institute on Drug Abuse.

SOURCES: Mark Asbridge, Ph.D., associate professor, department of community health and epidemiology, Dalhousie University, Halifax, Canada; Wayne Hall, Ph.D., University of Queensland Centre for Clinical Research, Brisbane, Australia; Jan Withers, national president, Mothers Against Drunk Driving; Guohua Li, M.D., professor of epidemiology, Columbia University, New York City; Paul Armentano, deputy director, NORML Foundation; Feb. 10, 2012, BMJ, online

Copyright © 2012 HealthDay. All rights reserved.