Recreational and medical weed, now legal in a growing number of states, is putting an emphasis for law enforcement on pulling people over for driving-while-high.

But the problem is that marijuana affects drivers differently than alcohol, is more difficult to detect, and there is much less data documenting stoned driving compared to drunk driving.

For Holyoke Police Lieutenant Jim Albert, whose department just received an approximately $3,000 state grant to support patrols discouraging impaired driving of all kinds, marijuana is a particular focus this year.

“The law is changing and what law enforcement is concerned about is people will relax their appreciation for how impaired you can be when driving impaired when using marijuana,” he said.

Enforcement will increase due to the grant, though Albert said $3,000 wouldn’t go very far. At the same time, having a police presence on the roads often has a deterrent effect even when people are not pulled over.

“When you see a cop, you slow down. It’s human nature,” he said.

Police are being sent to training to become drug recognition experts, a process Albert described as challenging and a years-long endeavor. His department does not have anyone who is designated drug recognition experts, but if there was a stop that required one, the department could ask state police or another agency to assist through mutual aid, he said.

Meanwhile, the National Highway Traffic Safety Administration released a report to Congress on marijuana-impaired driving in July. Its findings outline the challenges facing law enforcement.

Drug tests exist for THC, the chemical in marijuana that gets users high, but those tests do not correlate well to how impaired a driver is, according to the report. Because the drug is fat-soluble, rather than water-soluble — like alcohol — the substance can cause impairment long after it is detectable in the blood, the report says.

The report indicates that, while there is a study being funded, developers of impairment tests for marijuana have not yet shown that everyone dosed on marijuana shows an observable amount of impairment in a controlled laboratory setting.

“There is evidence that marijuana use impairs psychomotor skills, divided attention, lane tracking, and cognitive functions,” the report reads. “However, its role in contributing to the occurrence of crashes remains less clear.”

Does that mean marijuana users have the all-clear to smoke and drive? No. Driving while stoned is illegal everywhere in the United States, and there is some evidence that driving while stoned makes your more likely to get in an accident.

One 2010 study in nine European Union countries, cited by the Traffic Safety Administration report, found that drivers positive for THC were one to three times more likely to get into an accident than those who had not tested positive. This is far less than the 20 to 200 times more likely a driver would be for a highly elevated blood alcohol level (above .12 BAC) and less than the five to 30 times more likely a driver with an elevated blood alcohol level (between .08 and .12) would be to get into an accident.

A big problem, as the report points out, is a lack of data.

Alcohol crash data is easy to come by. Both the state and federal websites show how many of the state’s crash deaths were caused by an alcohol-impaired driver (143 of 354 in 2014; 96 of 306 in 2015 — the most recent year listed). Marijuana data is much more elusive.

The Traffic Safety Administration collects drug-test data in its database, but does not have reports, as they do for alcohol, linking marijuana use to driver deaths.

The report recommends, among other things, more data be collected on whether impaired drivers were using marijuana.

Easthampton officer Kyle Gribi is one of the department’s two drug recognition experts. He said the training involves work at the academy and also travel to Maricopa County, Arizona – the county that the embattled and recently pardoned Joe Arpaio was sheriff.

In that county, inmates arrested while under the influence of different drugs are used as test subjects for drug recognition trainees, according to Gribi, who said he did not know whether the subjects gave consent to be included in the training exercises.

Officers must determine what drugs the subjects are on, and correctly identify eight of 12 cases to pass the course.

“I don’t know the process for how they decide what people to take; unfortunately there are a lot of people that are under the influence,” he said. “It is not stressful for them; it is more stressful for us. We are being tested.”

For marijuana, Gribi said it is challenging to enforce the marijuana law because of a lack of technology. Unlike alcohol, where a breathalyzer test can be used, no technology exists to accurately assess how impaired someone is after having used marijuana.

“We’re in new territory at this point,” he said. “To some degree it is the unknown. Five years ago marijuana was arrestable, and now it is different. Now we are enforcing it in a different way. Not everyone is caught up to it as they are with alcohol.”

In Holyoke, police have made few arrests for marijuana-impaired driving, according to Albert.

At the same time, he said the idea of impaired driving affects him personally. His children are out on the roads, too, and unsafe driving can be deadly.

“Lifestyle changes shouldn’t affect safety on the road,” he said. “We don’t want that to happen.”

Gribi said the goal is public safety.

“It is not so much about getting people in trouble,” he said. “It’s more we don’t want people driving cars in public that are impaired and potentially going to hurt people.”

Dave Eisenstadter can be reached at deisen@valleyadvocate.com.