If you could look into a crystal ball to see the future of drug-impaired driving on Ontario roads come October 17, you might see Colorado.

Since cannabis was legalized there four years ago, statistics on marijuana and driving compiled by the state found that more than 20 per cent of drivers involved in fatal accidents had some level of cannabis in their system.

And a survey this year of more than 15,000 cannabis users in the state found that nearly 70 per cent of them had driven under the influence of marijuana at least once in the past year. Forty per cent of them said they didn’t think it affected their ability to drive.

“It was a signal that reaffirmed our suspicion that a lot of marijuana users drive after using marijuana. And that’s problematic,” says Sam Cole, a safety communications manager at the Colorado Department of Transportation.

It’s a worrying trend across the U.S., where 12 of every 100 weekend nighttime drivers involved in crashes between 2013 and 2014 tested positive for THC, up from 8.6 in 2007, according to the National Highway Traffic Safety Administration. As of this year, recreational marijuana is legal in nine states.

No one knows for sure whether legalization means more people will drive high on Ontario roads, yet rates of drug-impaired driving charges — not necessarily all cannabis-related — have increased here recently, according to Statistics Canada. They are still a fraction of alcohol-impaired charges.

But with Colorado seeing throwback attitudes about driving while under the influence of drugs — not unlike when people thought it was OK to drink and drive — police enforcement has become key.

“Until people aren’t driving impaired, we need to have a robust law-enforcement program that cracks down on drugged driving, just the way we have with alcohol,” Cole says. “Back in the 1980s when MADD started, and alcohol fatalities were sky high, and we started telling people don’t drive drunk, they were like, ‘Oh, I’m fine when I’m drunk. I’m a safer driver. I drive drunk all the time.’

“Which is the same thing we’re hearing from marijuana users,” Cole says. “It’s the exact same trajectory.”

In Canada, some question whether law enforcement will have the necessary training to police drug impairment on our roads when recreational cannabis use becomes legal.

Some 833 officers have been certified as drug recognition experts as of October 1, having completed an international program that teaches police how to evaluate if a driver is drug-impaired. This rate — 2.3 for every 100,000 residents — falls short of the 6 per 100,000 recommended in a 2009 report on the DRE program in the U.S.

The training includes a 10-day course in a local classroom where officers learn how to conduct a 12-step physical evaluation that assesses markers such as a driver’s psychomotor abilities — similar to the standardized field sobriety test used roadside by police that requires drivers to stand on one leg or walk and turn — as well as their vital signs, including blood pressure and temperature.

Next, officers attend a three- or four-day program in Jacksonville, Fla., one of a few locations where there are enough subjects from a nearby drug-dependence clinic who can provide the “volume and the breadth of drug diversity that we require,” says RCMP Sgt. Ray Moos, who is charge of the training.

There, an officer has to correctly assess 12 individuals and determine which category of drug — not only cannabis but drugs in other categories such as stimulants, depressants or hallucinogens — the subject is on.

The officer’s conclusions are then confirmed with a urine sample from the subject.

“It is a challenging course,” says Moos. “The course really assesses how drugs affect human physiology. Officers have to be competent with knowing a lot of different drugs and what category of drugs those drugs would fall under, and how those drugs affect the body.”

About two of the roughly 20 to 24 students in a class don’t pass, he says.

Here in Canada, the evaluations — which take 45 minutes — would be conducted at a police station after an arrest, which will occur based on what an officer sees or smells in the vehicle, from the state of the individual, or because of the results of the standardized field sobriety test, according to Moos.

Officers may also factor in the results of a roadside saliva test that can detect THC. Neither the RCMP nor the OPP have decided if or how they will deploy the devices, which can determine if the drug is present in the saliva but do not determine impairment. In Colorado, the saliva devices were tested earlier this year and the results of the tests will be released later this fall.

Drivers who are arrested may also be required to give a blood sample in accordance with the new legislation, although the RCMP says the “frequency with which this demand is used will be dependant on the jurisdiction in which the offence takes place and the mechanisms in place to facilitate a blood draw.”

Moos says that could mean a visit to a hospital emergency department or another setting, such as a medical clinic that has been authorized to draw blood. A suspected drug-impaired driver’s failure to comply “will result in criminal charges which carry the same penalty as driving while impaired,” says an RCMP spokesperson.

Jennifer Knudsen, traffic safety research prosecutor for Colorado, whose job is to help district attorneys throughout the state prosecute DUI cases for drugs and alcohol, says cannabis legalization has opened up a “whole new world.”

The DAs there not only have to be lawyers, she says, but medical experts, educating jurors on how THC impairs drivers because many of them don’t believe it does.

“Think about someone who is 25 and just graduated from law school and they may have never done a trial. And all of a sudden they have this case where they have to do the science and the law stuff,” she says. “It’s a huge challenge.”

The laws in Colorado around drug-impaired driving aren’t the same as in Canada. For instance, there’s no legal limit of THC concentration in the blood as there is here, where it’s five nanograms of THC per millilitre of blood. Instead, Colorado has a “reasonable inference” law, which says that a judge has the discretion to instruct a jury that a defendant is incapable of driving safely with that level of THC in their blood. It’s not used all the time.

And in many cases, blood isn’t taken, says Knudsen. Testing is expensive and about 30 per cent of drivers refuse the request, willing instead to take administrative penalties such as a licence suspension. As well, there aren’t experts available in far-flung parts of the state to interpret the results of the blood analysis for a jury.

There is also the issue of just how quickly THC leaves the blood.

Knudsen says studies have shown that blood concentration can drop 60 per cent just 15 minutes after someone takes a puff of marijuana. And for some people, taking more than the recommended dose of 10 milligrams of an edible may not result in blood concentrations above five nanograms per millilitre. Results can also differ, she says, depending on whether someone is an occasional user or uses all the time and typically has a baseline concentration in their blood.

“When you’re talking about numbers, and drugs that aren’t alcohol, it’s very difficult,” she says.

Typically, the more someone drinks, the higher the concentration of alcohol in their blood. But THC concentrations fall so rapidly in the blood it is difficult to correlate them to impairment.

In cases involving cannabis, “you could have someone with 23 nanograms and that person is not guilty,” says Knudsen. “And you could have someone who has two nanograms and is guilty. It’s hard.”

But Knudsen says drug recognition expert evaluations are rare and the state, similar to Washington where cannabis has also been legalized, has one of the lowest rates of evaluations in the world.

She says it’s because many of the officers who have the training are in supervisory roles or not available because they’re not on duty. Or she says, there may be one drug recognition expert covering a large geographic area who can’t make it to a site in time to make the evaluation worthwhile.

“There’s a lot of hurdles,” says Knudsen. District attorneys who prosecute the cases rely on what officers observe — what they see and smell in the car, and on the behaviour of the driver at the window, as well as in the roadside sobriety tests.

“We have to put all of that together. And that really is the crux of the case,” says Knudsen. “If we have a DRE evaluation, then that is really the icing on the cake.”

Officers trained as drug recognition experts in Canada (as of Oct. 1, 2018):

N.W.T.: 4 (RCMP), 8.9 per 100,000 pop.

P.E.I.: 13 (6 RCMP, 7 municipal), 8.5 per 100,000 pop.

Sask.: 74 (36 RCMP, 38 municipal), 6.3 per 100,000 pop.

Nova Scotia: 61 (29 RCMP, 32 municipal), 6.4 per 100,000 pop.

Nfld./Labrador: 27 (15 RCMP, 12 municipal), 5.1per 100,000 pop.

New Brunswick: 28 (13 RCMP, 15 municipal), 3.7per 100,000 pop.

Manitoba: 42 (17 RCMP, 25 municipal), 3.1 per 100,000 pop.

Yukon: 1 (RCMP), 2.6 per 100,000 pop.

Alberta: 102 (42 RCMP, 60 municipal), 2.4 per 100,000 pop.

B.C.: 111 (75 RCMP, 36 municipal), 2.3 per 100,000 pop.

Ontario: 240 (5 RCMP, 78 provincial, 157 municipal), 1.7 per 100,000 pop.

Quebec: 130 (0 RCMP, 53 provincial, 77 municipal), 1.5 per 100,000 pop.

Nunavut: 0

Total: 833, 2.3 per 100,000 pop.