“Unlike alcohol, THC remains in the body long after its psychoactive effects have worn off. So, unlike commonly used alcohol sobriety tests, blood tests for cannabis that are currently widely used … cannot distinguish between recent or past use.”
— National Institute on Drug Abuse Director Nora Volkow[1]
The soundest approach to criminalizing driving under the influence of cannabis is the most common approach taken for both cannabis and prescription medications — effects-based laws. In most states, drivers are guilty of driving under the influence of drugs (DUID) if the state proves, in light of all of the evidence, that they were actually impaired at the time of arrest.
Unfortunately, not all states have taken this sensible approach to cannabis. A small subgroup of states have “zero tolerance” laws for cannabis, meaning individuals will automatically be guilty of DUID if any THC and/ or metabolites are found in their blood. Other states have “per se” laws, meaning drivers can be convicted solely for having a certain concentration of THC or metabolites in their blood. As the AAA Foundation for Traffic Safety concluded, per se laws are “arbitrary and unsupported by science.”[2] Metabolites can be detected weeks after impairment wears off, and THC can be present for days or even weeks after some regular consumers last used marijuana.[3]
DUID laws in medical cannabis states: A snapshot of the different approaches
Forty states and the District of Columbia have effective medical cannabis laws on their books. Twenty-seven of those states and Washington, D.C. have effects-based DUID laws for cannabis impairment, meaning individuals — whether or not they are patients — are guilty of driving under the influence of cannabis if the evidence, taken together, proves the drivers were impaired.[4] In these states, the mere presence of THC — absent some evidence of impaired performance, such as the results of a field sobriety test — will not automatically lead to a conviction.
Five other medical cannabis states —Delaware, Illinois, Oklahoma, Rhode Island, and Utah — generally have “zero tolerance” or “per se” laws criminalizing driving with any amount or a set amount of THC or cannabis metabolites in a person’s system. However, all of those states have an explicit or implicit exception for medical cannabis patients who are not impaired.[5]
Five additional states — Nevada, Montana, Ohio, Pennsylvania, and Washington — have per se DUID laws with no exception for patients.[6] In Pennsylvania, drivers are guilty of DUID if the level of THC in their blood exceeds one nanogram per milliliter (ng/mL). In Ohio and Nevada, the cut-off is two ng/mL, and in Montana and Washington, the limit is five ng/mL. These laws can result in patients being convicted for driving many hours, or perhaps days, after any intoxicating effects have worn off.[7] A sixth state — West Virginia — appears to apply a per se standard of three ng/mL in blood serum only to patients.
Colorado has a law that may also result in medical cannabis patients being convicted of DUID when they are not impaired. Colorado has a permissible inference of impairment if a driver’s blood THC concentration is five ng/mL or greater. This means that the jury or judge may infer that a driver drove under the influence of cannabis if that level of THC was in his or her system. However, the driver has a chance to rebut the evidence.
Prescription medications, over-the-counters, and driving while intoxicated
While many prescription and over-the-counter medicines cause impairment, to the best of MPP’s knowledge, no state has created a zero tolerance or per se DUID law for simply testing positive for a medicine that was prescribed to the driver. To avoid criminalizing sober drivers who have used medications in the past, states instead use an effects-based approach: The state can present evidence relevant to impairment — such as testimony or footage of erratic driving, roadside sobriety test results, observations from trained officers, and lab results — but a blood test showing past use will not, on its own, result in a conviction.
What does the science say?
The AAA Foundation for Traffic Safety evaluated data on THC-positive drivers and drug-free controls, along with the results of drug recognition expert evaluations, to see if the data supported a set threshold for a per se driving law for cannabis. It did not.[8] As AAA Director of Traffic Safety Advocacy and Research Jake Nelson explained, “There is no concentration of [THC] that allows us to reliably predict that someone is impaired behind the wheel in the way that we can with alcohol.”[9]
A limit of five nanograms per milliliter of whole blood, the AAA Foundation analysis found, would miss large numbers of cannabis-impaired drivers.[10] A lower limit ensnares sober drivers who used cannabis much earlier and potentially some people who had been exposed to second-hand cannabis smoke.
In 2025, a study of regular cannabis consumers evaluated their performance on a driving simulator and found, “Many regular users of cannabis exceed zero tolerance and per se THC cutpoint concentrations days after their last use, risking legal consequences despite no evidence of impairment.”[11]
In addition to the approach being unscientific, research has shown that per se DUID laws do not make the roads safer. As implemented, these laws have “no discernible impact on traffic fatalities.”[12]
Similarly, there is no scientific basis for a zero tolerance approach. A rigorous study by the National Highway Traffic Safety Administration found that drivers who tested positive for THC had an identical crash risk to those testing negative once researchers controlled for demographics and alcohol use.[13]
Protecting patients from unfair DUID laws
Per se THC limits and zero tolerance DUID laws are a major concern for medical cannabis patients who often use cannabis more frequently than other cannabis consumers. There is no realistic way for patients to determine what their blood’s THC levels are, meaning they would have no way of knowing if it’s legal for them to drive. Even if they could somehow test their blood levels regularly, THC is fat-soluble and its levels can go up without a person having consumed cannabis. If a medical cannabis state’s per se or zero tolerance DUID law cannot be repealed, it should at least carve out a compassionate exception for patients, as was done in several states.
Last updated: January 2, 2026
[1] Dr. Nora Volkow, "A new roadmap for cannabis and cannabis policy research," NIDA, Nora's Blog, January 14, 2025, available at https://nida.nih.gov/about-nida/noras-blog/2025/01/new-roadmap-cannabis-cannabis-policy-research
[2] “Impaired Driving and Cannabis,” AAA Foundation for Traffic Safety.
[3] Janith Galhenage, Asiri Rodrigo, “Prolonged detection of cannabis in urine of a consumer in a Forensic hospital, Victoria, Australia, up to 102 days of abstinence — A case report,” Psychiatry Research Case Reports, Volume 2, Issue 1, 2023; Bergamaschi MM, et al' blood on per se drugged driving laws. Clin Chem. 2013 Mar;59(3):519-26.
[4] The 27 medical cannabis states with effects-based laws are: Alabama, Alaska, Arizona, Arkansas, California, Connecticut, Florida, Hawaii, Kentucky, Louisiana, Maine, Maryland, Massachusetts, Michigan, Minnesota, Mississippi, Missouri, Nebraska, New Hampshire, New Jersey, New Mexico, New York, North Dakota, Oregon, South Dakota, Texas, Vermont, and Virginia.
[5] In Delaware and Illinois, a patient’s conviction cannot be based solely on metabolites or cannabis components. In Rhode Island, a patient’s conviction cannot be based solely on metabolites. In Arizona, a patient’s conviction cannot be based solely on metabolites or components of cannabis in a concentration that is insufficient to cause impairment. In 2013, the Michigan Supreme Court ruled the state’s zero tolerance law did not apply to patients. Also, in both Michigan and Delaware, language from adult-use legalization may supersede zero tolerance cannabis laws for all drivers.
[6]In Nevada, the per se law is limited to felonies — cases where there was a serious injury or death.
[7]After a 15-hour period of abstinence, including a full night’s sleep, Westword’s medical cannabis critic’s THC levels were still 13.5 ng/mL. According to his physician, Mr. Breathes was in “no way incapacitated” at the time. (William Breathes, “THC blood test: Pot critic William Breathes nearly 3 times over proposed limit when sober,” Westword, April 18, 2011.)
[8] Barry Logan, Ph.D., et al., “An evaluation of data from drivers arrested for driving under the influence in relation to per se limits for cannabis,” AAA Foundation for Traffic Safety, May 2016.
[9] Jonah Bromwich, “How Much Is Too Much Marijuana to Drive? Lawmakers Wonder,” The New York Times, May 13, 2016.
[10]Barry Logan, Ph.D., et al., “An evaluation of data from drivers arrested for driving under the influence in relation to per se limits for cannabis,” AAA Foundation for Traffic Safety, May 2016.
[11] Robert L Fitzgerald, et al, Per Se Driving Under the Influence of Cannabis Statutes and Blood Delta-9-Tetrahydrocannabinol Concentrations following Short-Term Cannabis Abstinence, Clinical Chemistry, Volume 71, Issue 12, December 2025, Pages 1225–1233.
[12] M. Anderson and D. Rees, "Per Se Drugged Driving Laws and Traffic Fatalities," IZA Discussion Paper 7048, 2012, p. 5.
[13] R.P. Compton and A. Berning, “Drug and alcohol crash risk,” National Highway Traffic Safety Administration, Washington, D.C., Research Note DOT HS 812 117, Feb. 2015. (Meanwhile, drivers with a 0.05 BAC — which is below the legal limit — had more than double the crash risk.)