A person’s THC levels after using cannabis does not accurately predict impairment, according to the findings of a federally funded study.

The research comes courtesy of scientists from RTI International, with funding from the National Institute of Justice (NIJ). The purpose of the study was to investigate whether the biofluid levels of THC in a person’s system correlate with performance in field sobriety and cognitive tests, which are typically used to determine alcohol impairment in drivers.

After assessing 20 participants who either ate or vaporized marijuana as part of the tests, the researchers concluded that while all who received doses of 5mg of THC or more were impaired to some degree, “THC levels in biofluids were not reliable indicators of marijuana intoxication.”

To understand the relationship between THC levels and impairment test performance, the researchers collected blood, urine and oral fluid from the participants for analysis throughout the study period. In some of the tests administered to participants, such as standing on one leg or walking and turning, the researchers found no indication of compromised motor functions following cannabis consumption.

“Results from the toxicology tests showed that the levels of all three targeted cannabis components (THC, cannabidiol, and cannabinol) in blood, urine, and oral fluid did not correlate with cognitive or psychomotor impairment measures for oral or vaporized cannabis administration,” NIJ wrote.

“Many of their study participants had significantly decreased cognitive and psychomotor functioning even when their blood, urine, and oral fluid contained low levels of THC,” the federal agency added. “The researchers also observed that standardized field sobriety tests commonly used to detect driving under the influence of drugs or alcohol were not effective in detecting marijuana intoxication.”

So while THC may impair a person, its concentration does not reliably indicate just how impaired a person is. The results also indicate that in low doses, some people’s performance in standard field sobriety and cognitive tests are not negatively impacted at all.

This calls into question the “per se” marijuana DUI laws in force in six states in the US, which penalize drivers found with more than a certain level of THC in their system, along the same lines as for alcohol.

“Despite a handful of states imposing per se THC thresholds as part of their traffic safety laws, there exists no science demonstrating that these arbitrary limits are reliable predictors of either recent cannabis exposure or impairment,” said NORML Deputy Director Paul Armentano.

A 2019 congressional report into marijuana-impaired driving similarly concluded that “studies have been unable to consistently correlate levels of marijuana consumption, or THC in a person’s body, and levels of impairment.”

Another study from earlier this year found that states which permit recreational marijuana had seen no apparent increase in motor vehicle crash incidents involving elevated THC levels following legalization. That said, while determining the relationship between THC levels and impairment is more complex and uncertain than with alcohol, few would recommend driving under the influence of cannabis.