New Study Shows Cannabis Blood Limits Used in DUI Laws May Be Punishing Unimpaired Drivers

A new study published in Clinical Chemistry is raising serious questions about how the United States currently determines cannabis-related driving impairment. The research suggests that many regular marijuana users can exceed legal THC blood limits even after days of not consuming cannabis, all while showing no measurable driving impairment. With several states relying on strict per se THC limits or zero-tolerance laws, the findings highlight a growing gap between scientific evidence and legal policy.

What the Study Looked At

Researchers, led by Dr. Thomas D. Marcotte from the University of California, San Diego, wanted to explore whether the THC blood levels used by law enforcement actually correspond to real-world driving impairment. They recruited 190 regular cannabis users and instructed them to abstain from cannabis for at least 48 hours. After this abstention period, the team measured each participant’s blood THC level and assessed their driving using a driving simulator.

The approach was meant to mirror a common real-world situation: individuals who regularly use cannabis but have not used it recently. Because THC stays in the body far longer than substances like alcohol, the question is whether the lingering THC in a frequent user’s system truly reflects impairment.

The Legal Limits at the Center of the Issue

In the U.S., cannabis DUI laws vary widely:

• 6 states use fixed per se THC limits (typically 2 or 5 ng/mL)
• 12 states have zero-tolerance policies, meaning any detectable level can be grounds for a DUI

Under these laws, the presence of THC is treated as direct evidence of impairment — regardless of behavior, sobriety tests, or driving performance.

The problem? Scientific research hasn’t been able to tie specific THC concentrations to measurable driving impairment, especially in regular users who develop tolerance.

What the Researchers Found

The data from the study paints a clear picture of how unreliable these legal thresholds may be:

• 43% of participants still exceeded zero-tolerance thresholds after 48 hours without cannabis
• 24% had blood THC concentrations at or above 2 ng/mL
• 5.3% had THC at or above 5 ng/mL
• One participant even reached a baseline of 16.2 ng/mL after two full days of no consumption

Despite these elevated levels, the crucial point is that their driving performance did not differ from participants with much lower THC concentrations. On the driving simulator, participants who were “over the limit” did no worse than those who were “under the limit.”

In other words, the laws suggest they were impaired — but the data shows they weren’t.

The team also noted that after a controlled cannabis smoking session in a related part of the study, THC levels only increased by a median of 0.5 ng/mL after six hours. While this change was statistically significant, it remained small compared to the high baseline levels that regular users can maintain for days.

Why THC Blood Tests Don’t Work Like Alcohol Tests

Alcohol and cannabis behave very differently in the body. Alcohol levels rise and fall predictably, and impairment correlates closely with blood alcohol concentration (BAC). THC, however:

• Builds up in fat tissues in regular users
• Can remain detectable for days or weeks
• Does not correlate consistently with impairment
• Peaks rapidly after smoking but drops sharply, even as impairment continues
• Influences regular and occasional users differently due to tolerance

For regular users, THC concentration is a poor predictor of cognitive or motor impairment. Some people can test above 5 ng/mL with no impairment, while others who rarely use cannabis may show impairment at much lower levels.

This discrepancy makes THC blood limits a very blunt instrument — one that, according to this study, may be catching sober, safe drivers.

What This Means for DUI Laws

The findings strongly support a shift away from per se THC laws, which operate on the assumption that THC concentration alone can indicate impairment. The study adds to a growing body of work showing this assumption is scientifically flawed.

The researchers emphasize that assessing cannabis impairment requires a more holistic approach, combining:

• Behavioral observations in the field
• Psychomotor assessments
• Toxicology results
• Potential new technological tools

They also highlight the need for collaboration between scientists and law enforcement to develop fairer, evidence-based standards.

The Broader Scientific Context

This isn’t the first study to show THC levels don’t reliably predict impairment. Recent research has found:

• Regular users can have detectable THC with no increased crash risk
• Blood, saliva, and breath THC tests are often poor indicators of impairment
• Occasional users show more consistent correlations between THC levels and impairment, but even then the variability is high

Large meta-analyses have also concluded that THC biomarkers (blood, saliva, breath) are not adequate stand-alone indicators of driving ability.

Additionally, a 2021 study involving 191 participants found no significant correlation between THC concentration and critical driving metrics such as lane position or car-following ability. This reinforces the idea that cannabis impairment is a complex, multifaceted issue, not one that can be captured by a simple number.

Why This Matters for Society

Cannabis use is rising across the U.S., and more states are legalizing recreational or medical marijuana. Under current laws, millions of regular users could be at risk of:

• Wrongful DUI charges
• Criminal records showing impairment where none existed
• Job consequences tied to driving restrictions
• Legal penalties not based on actual driving behavior

Since regular users can exceed THC limits days after use, many could be driving legally, safely, and unimpaired — yet still be subject to arrest and felony charges.

For policymakers, this study is a wake-up call: current THC driving laws are outdated and unsupported by scientific evidence.

Moving Forward: What Needs to Change

Based on the findings, several policy improvements seem necessary:

• Shift from THC concentration–based laws to performance-based assessments
• Train officers in better observational methods specific to cannabis
• Research and validate technological tools (e.g., cognitive tests, reaction time evaluations, eye-movement assessments)
• Update legal standards to reflect regular vs occasional use, since tolerance plays a major role
• Encourage partnerships between law enforcement, toxicologists, neuroscientists, and policymakers

The goal isn’t to weaken impaired driving laws, but to make them more accurate, more fair, and better aligned with real impairment, improving safety for everyone.

Final Thoughts

This study underscores an important truth: cannabis impairment is not as simple as measuring THC in blood. As legalization expands, lawmakers must ensure that DUI laws keep pace with scientific understanding. Otherwise, we risk a system where regular but unimpaired cannabis users are treated as criminals, while truly impaired drivers may slip through gaps in outdated testing methods.

Science is making the issue clearer, and this new research offers one of the strongest cases yet for rethinking how cannabis impairment should be defined and measured on the road.

Research Reference

Per Se Driving Under the Influence of Cannabis Statutes and Blood Delta-9-Tetrahydrocannabinol Concentrations Following Short-Term Cannabis Abstinence
https://doi.org/10.1093/clinchem/hvaf121