AUTONEWS
A new way to study how cannabis use impacts safe driving
As marijuana legalization expands across the U.S., it is outpacing research on the impact of cannabis use behind the wheel. Researchers at the Virginia Tech Transportation Institute (VTTI) recently spent two years collecting real-world driving data from cannabis users to help fill in the gaps.
Using the patented naturalistic driving study methods, researchers collected 105,000 miles of driving data from 14,700 trips made by drivers ages 21 to 70 between 2021 and 2023. Of these, 9,000 miles involved cannabis use, 1,000 miles involved alcohol use, and 400 miles involved polysubstance use, or more than one drug. The remainder of the miles driven were considered substance free, creating a baseline for each participant.
"Because most studies have not followed a driver over time, that missing piece of how people adapt to driving under that influence is actually covered through this," said Kaitlyn Bedwell, study author and senior research associate at the institute. "Since the effect of cannabis and impairment is so individualized, this study shows use over time and provides an opportunity to look at that full individual window in depth."
The study and initial findings were published in a 2025 report, "Naturalistic Driving Study on Cannabis Use in Washington and Virginia." In addition to the naturalistic driving study data, participants provided 10,081 self-reported substance-use journal entries, which were combined with regular oral swabs to confirm substance use.
Early results from data analysis revealed:
-Time of day patterns: Cannabis-related trips spiked during typical lunch times and in the evening. Fridays were the most popular day of the week for imbibing.
-Behavioral changes: Some participants compensated for substance use by taking different routes, typically choosing to use more rural roads than sober trips.
Bedwell said the first-of-its-kind study's real value lies in its holistic approach, accounting for individual factors that affect driving under the influence of cannabis and providing both qualitative and quantitative measures that connect reported substance use with lab data. She said the framework can be replicated for future research and the sizable, novel database utilized by other researchers to further answer questions about cannabis use and driving performance.
"My takeaway is that I believe people want to know 'am I being safe? Am I keeping other people on the road safe?'" she said. "We've got to use research to provide that clarity."
Bedwell said data from sober trips can provide unique insights. By collecting data on participants when they are not using substances, this study effectively uses participants as their own baselines.
The challenge of measuring cannabis impairment...As of early 2026, 40 U.S. states have legalized either medical or recreational cannabis use. It is illegal to drive while impaired by any substance, but unlike alcohol, cannabis impairment is difficult to quantify.
"There is no equivalent to a breathalyzer for cannabis," Bedwell said. "There's no concrete way to actually say you are impaired while you are driving this vehicle right at this moment."
Alcohol impairment can be measured reliably by blood alcohol content (BAC), and there is a proven relationship between increasing BAC and decreased driving performance.
Tetrahydrocannabinol (THC)—the chemical in cannabis that causes the mind-altering effects—does not have a standard metric that distinguishes between impaired and unimpaired. THC is fat-soluble and can be detected in a person's system for up to 30 days, long after impairing effects have faded. Individual characteristics such as weight, metabolism, and even frequency of use can all alter how a person processes and is affected by cannabis. THC content in cannabis products also varies widely, ranging from 1% to 30%.
These factors make obtaining accurate crash data and policy development challenging. Bedwell said research in this area has also been slow to catch up and has often delivered conflicting results.
"We need more informed research in order to have informed policy," she said. "To this date, there is still no standard measure for how cannabis use affects impairment in a real-world environment."
VTTI's naturalistic driving research provided a method for overcoming some of these challenges by documenting an individual's behavior while both sober and under the influence over the course of multiple years. Pioneered by VTTI in the 1990s, the method involves installing unobtrusive cameras and sensors in participants' own vehicles to monitor them during daily routines. This allows researchers to have a wealth of information on driving behavior, vehicle information, and traffic scenarios, which can be used to study crash risk.
Bedwell said that the long-term naturalistic driving research study addresses the limitations of studies that use computer driving simulators or short-term studies that study drivers on a closed course.
"It allows for a precise assessment of driving performance in the real world, and that's what's going to capture the real complexity of driving," she said. "If you got this artificial simulation environment, it's got limitations because the participant doesn't face real consequences of crashing or the variability of other drivers."
Provided by Virginia Tech
Nenhum comentário:
Postar um comentário