Markian Hawryluk
The Bulletin

States such as Oregon that have legalized recreational marijuana are seeing an increase in motor vehicle accidents in comparison to neighboring states that have not legalized, according to two studies released this month by the Insurance Institute for Highway Safety.

While the studies couldn’t prove the accidents were a direct result of legalized marijuana, a growing amount of evidence has led officials to conclude the relationship exists. The studies underscore concerns in Oregon that the state has struggled to adapt to the challenges of identifying marijuana- impaired drivers, leaving citizens at greater risk for traffic accidents and fatalities.

“When we waded into this three years ago, our first calls were to Washington and Colorado, to say, ‘What did you do well? … How can we avoid some of the pitfalls?” said Dan Estes, who heads the impaired-­driving program for the Oregon Department of Transportation. “Data was one that we were not able to avoid. The biggest problem was we weren’t measuring the impact of marijuana on traffic safety before we legalized it, so we don’t have anything to compare it to.”

One of the studies looked at police-reported crashes before and after sales of recreational marijuana began in Colorado, Oregon and Washington, finding that accident rates were 5.2 percent higher in the three legalization states than in neighboring states that hadn’t legalized.

The second study, conducted by the Highway Loss Data Institute, found collision insurance claims in Colorado, Nevada, Oregon and Washington rose 6 percent higher than in Idaho, Montana, Utah and Wyoming. The researchers controlled for differences in driver rating, the mix of urban and rural driving, unemployment, weather and seasonality.

“This is a correlation study not a causation study, but we designed the study in a way to try to eliminate as many conflicting factors as we can,” said David Harkey, president of the highway safety institute. “I think the bottom line for this is we’re still seeing this higher crash risk that is occurring in states that made this move to legalize the recreational use of marijuana.”

Harkey said other studies have taken different approaches to try to gauge the effect of legalization on traffic accidents, with results ranging from no effect to a 30 percent increase.

“We feel that maybe we’re a little conservative in what the estimates are, but there’s no doubt we’re seeing an increase,” he said. “It’s not so much about the magnitude of the number at one point in time, as the direction.”

The studies found smaller effects in Oregon than in the other states, and the results for Oregon were not statistically significant on their own. That could be, in part, because Oregon was one of the later states to legalize, and less data was available.

Perception of use

Harkey said the rise in marijuana-impaired driving could reflect differences in how the public perceives marijuana and alcohol risk. Roadside surveys conducted in Washington, for example, found cannabis use in drivers increased from 15 percent before legalization to 19 percent six months later, and to 21 percent a year later. Moreover, while alcohol impairment tends to happen more in the evening hours, marijuana impairment is more common during the day. While almost no drivers with children in the car had alcohol in their system, 14 percent of those driving children had marijuana in their system.

“We have some work to do to help the public understand the seriousness of impairment with marijuana,” Harkey said.

Surveys in Oregon have found similar results.

“Three out of four Oregon adults agree that driving under the influence of marijuana increases crash risk, which is less than for alcohol,” said Julia Dilley, senior research scientist with the Oregon Health Authority.

The state has asked about marijuana use and driving in two surveys in 2015 and 2016, finding that 21 percent to 34 percent of adults have driven within three hours of using marijuana. That includes more than half of frequent marijuana users, defined as using 20 or more times in the past month.

“We don’t have a prelegalization measure,” Dilley said, “so we don’t know if that’s gotten worse or not.”

The Oregon Healthy Teen Survey had a similar question in 2015, finding that 48 percent of 11th-graders who use cannabis and had their driver’s licenses had driven within that three-hour window. The measure was higher for boys than for girls. Dilley cautioned, however, that the three-hour window for driving is a better measure for some types of cannabis than others.

“Edibles can take quite a long time to affect a person,” she said. “If a person is using edibles, they probably shouldn’t use that in their mind as any kind of a guide.”

The agency has tried to track marijuana use in fatal car crashes and the numbers of marijuana DUIs.

“One of the challenges for us from the measuring perspective is that other folks are working to improve those measures, as well,” Dilley said. “So in the process of improving that, it confuses what the trend looks like. We’re trying to be careful and not overinterpret anything.”

Officials can’t be sure whether marijuana-linked fatalities or DUIs are increasing or if the state is just getting better at identifying them.

Blood or urine

Identifying cannabis-impaired drivers has unique challenges. Unlike with alcohol, where scientific research has shown a direct relation between blood alcohol content and impairment, no such standard exists for cannabis.

While Washington and Colorado have adopted a quantitative threshold for the active component of cannabis in a person’s blood, Oregon has resisted implementing a similar standard. Oregon doesn’t routinely collect blood to test cannabis levels, but relies on urine tests instead.

“Urine is great for proving use and presence in the body, but it doesn’t do anything to prove impairment,” said Estes, of ODOT. “It stays around so long, it doesn’t give any kind of accurate snapshot as to what was happening when that person was arrested for impaired driving.”

Urine can test positive for marijuana up to 30 days after use. But Oregon legislators have been reluctant to allow blood collection as part of the implied consent for drivers.

“It’s a stretch for them to see that if you sign for your driver’s license, you will allow someone to take your blood,” Estes said. “It’s a very personal thing. They see it as invasive. Some states have gone to a blood standard, but Oregon, I don’t think they’re there yet.”

The AAA Foundation for Traffic Safety recommends a standard including behavioral and psychological tests to gauge impairment along with a positive blood test. But the foundation’s research, published in 2016, suggests neither is a strong predictor of impairment.

The group looked at results from nearly a thousand drivers arrested for impaired driving and evaluated by drug recognition experts. Among drug-free drivers, 56 percent were able to pass a walk-and-turn test, 67 percent a one-leg-stand test, and 49 percent a finger-to-nose test, all standard field sobriety tests. Among those who had used cannabis, 6 percent, 24 percent and 5 percent of test subjects passed the tests.

However, only the finger-to-nose test had a higher failure rate among drivers with higher cannabis concentrations in their blood. Drivers who were above the 5 nanograms per milliliter threshold adopted in Colorado and Washington did not have higher failure rates on walk-and-turn or one-leg-stand tests than those below.

When the researchers looked at results from the nearly 4,800 drivers who were arrested for cannabis DUIs, the median blood concentration was 4 ng/mL. That means more than half of the drivers deemed impaired by police officers had blood concentrations below the threshold for impairment adopted in those two states. That said, experts agree that urine tests are an even worse measure.

“Blood is a much better proxy,” said Lindsay Arnold, a researcher with the AAA foundation. “There’s some research on it showing it’s a decent indicator, just not as good as it is for alcohol. But some people who are impaired by cannabis are not picked up by what we call a standard field sobriety test.”

Proving impairment

In Oregon, police run drivers suspected of cannabis impairment through a series of physical and psychological tests, such as asking someone to walk a straight line or recite the alphabet backwards. Many of those tests are designed to see if drivers can pay attention to commands or perform multiple tasks at once. If officers conclude drivers are impaired, they bring them in for Breathalyzer tests. If that test shows a blood alcohol content above a .08 percent threshold, no further tests are done. But if the results don’t match the level of impairment observed by the officers, they can bring in a drug recognition expert.

“It doesn’t have to be zero, it could be .05, but they’re a soup-sandwich,” said Kent Vander Kamp, a patrol sergeant and drug recognition expert with the Deschutes County Sheriff’s Office. “At that point, they’re saying, ‘Something’s not right. We need the drug recognition experts.’”

Drug recognition experts are police who have been specially trained to identify impairment from a variety of substances. The evaluation often lasts for hours, as they work through a 12-step process of tests, questions and examinations. The officer then writes an opinion and collects toxicology samples for evaluation. Statewide, toxicology reports back up these expert conclusions more than 90 percent of the time.

The number of drug recognition expert evaluations in Oregon has risen steadily since legalization. Deschutes County had 67 total evaluations in 2015, rising to 103 evaluations in 2017. Statewide, evaluations have risen from 1,508 in 2015 to 1,823 in 2017. And the percentage of those evaluations finding cannabis impairment is increasing, as well.

“We’ve seen that number go up significantly,” Vander Kamp said. “I would say cannabis is the leading reason for those increased callouts. But I’m not just talking about the flower part. I think the edibles have had a lot to do with it.”

That also means that demand on Deschutes County’s 11 drug recognition experts is higher than ever.

“In fact, a lot of calls go unanswered,” he said.

Vander Kamp said toxicology results from the Oregon State Police crime lab often take up to nine months to be completed, which can slow down the prosecution of impaired drivers. According to the Oregon State Police, of the 1,823 evaluations completed in 2017, 928 are waiting on toxicology results.

Capt. Timothy Fox, a public information officer with the state police, said the lab’s processing capacity is insufficient to manage the work coming through the door.

“We have fantastic scientists, and they have consistently leveraged improvements in technology and process to reduce time spent per submission, but there are limits to what can be achieved through efficiencies alone,” Fox said. “Our people remain our most essential system component, and we don’t have enough of them. It’s that simple.”

Estes said the lab needs more funding to reduce the backlog and, ideally, the money to purchase tests to do more blood analysis.

“We’ll get blood when we can, when we have a warrant, urine if we can’t to at least show that marijuana was consumed at some point. But it really comes down to proving impairment on the roadside,” he said. “We’ve been kind of taking it in the shorts in the courtroom because it’s hard to do that with marijuana.”

While jurors are much more familiar with how a drunk person looks or acts, marijuana is more of a cognitive impairment, he says.

“That can be more difficult to detect, and it’s also more difficult to explain,” Estes said. “Prosecutors love that .08; juries like that .08; it’s the silver bullet.”

In the absence of better testing methods, experts say states are going to have to get more serious about public awareness campaigns and enforcement of impaired driving laws.

In the 1990s, efforts to improve seat belt use relied on education campaigns alone and couldn’t do much to move the needle. It’s only when public information campaigns were tied to law enforcement efforts that compliance increased.

“That’s how Click it or Ticket was born in North Carolina and eventually became the national model,” Harkey said. “But the reality is with impairment, we still lose more than 10,000 lives annually in this country as a result of impairment with alcohol, and so we haven’t had the same success.”

Estes said he would like to see a shift in marijuana messaging in Oregon, particularly to kids. He says campaigns that urge teens to wait until 21 to use cannabis are misguided.

“When was the last time you ever heard a public service message saying, ‘Hey kids, wait until you’re 21 to start smoking tobacco?’” he asked. “Absolutely not. Tobacco is bad for you. Just don’t do it.”

But it may takes years to get public awareness on cannabis-­impaired driving to the levels around drunken driving, and to make a dent in rising accident rates.

“We’ve been in this 50 years with alcohol, and we’ve have 50 years of science and testing and social messaging,” Estes said. “Marijuana is not new as a substance, but it’s new a socially acceptable substance. It’s going to take some time.”

— Reporter: 541-633-2162,