Cannabis DUI Testing — Per Se Laws and Why They Are Broken

Several states have enacted "per se" cannabis DUI laws, setting specific blood THC concentration limits above which a driver is automatically considered impaired. The AAA Foundation for Traffic Safety has called these laws "arbitrary and unsupported by science." The research is clear that blood THC does not reliably predict driving impairment — but the laws are enforced regardless.

Per Se Laws Are Scientifically Indefensible — But Legally Enforceable

A 2025 Clinical Chemistry study found that 5.3% of regular cannabis users exceeded the 5 ng/mL blood THC limit at baseline after 48+ hours of abstinence, with no difference in driving simulator performance. The AAA Foundation has concluded that per se THC limits miss 70% of actually impaired drivers while criminalizing sober frequent users. Know your state's law.

Per Se Blood THC Limits by State

StatePer Se LimitLegal Standard
Colorado5 ng/mL blood THCPermissible inference of impairment
Washington5 ng/mL blood THCPer se
Montana5 ng/mL blood THCPer se
Nevada2 ng/mL blood THCPer se
Ohio2 ng/mL blood THCPer se (plus separate metabolite limit)
Pennsylvania1 ng/mL blood THCPer se (zero-tolerance)
Approximately 10 statesZero toleranceAny detectable THC or metabolite

The zero-tolerance states (Arizona, Delaware, Georgia, Iowa, Illinois, Indiana, Michigan, Oklahoma, Rhode Island, Utah, Wisconsin, and others) criminalize any detectable amount of THC or its metabolites. In practice, this means a cannabis user can be convicted of DUI weeks after their last use, with no impairment evidence at all.

Why Blood THC Correlates Poorly With Impairment

The scientific problems with per se cannabis DUI laws are well-documented:

The AAA Foundation Analysis

The AAA Foundation for Traffic Safety reviewed per se THC laws and concluded they are "arbitrary and unsupported by science." The foundation found that a 5 ng/mL blood THC limit:

  • Misses 70% of cannabis-impaired drivers (false negatives)
  • Catches sober chronic users who have no current impairment (false positives)

The limit is arbitrary in the sense that there is no meaningful biological threshold below which impairment is absent and above which it is present. Cannabis impairment depends on dose, tolerance, time since use, route of administration, and individual factors — not a simple blood concentration.

The NHTSA Case-Control Study

The National Highway Traffic Safety Administration conducted a large case-control study comparing crash-involved drivers to non-crash-involved drivers. The finding: drivers who tested positive for THC had identical crash risk to those testing negative once controlling for demographics and alcohol. This is a dramatic contrast with alcohol, where BAC correlates strongly with crash risk.

The Chronic User Problem

Chronic cannabis users can have blood THC levels above per se limits for days after last use, with no meaningful impairment. Bergamaschi et al. (2013, PMID 23449702) found blood THC persisting up to 30 days in chronic daily smokers during monitored abstinence. A 2025 Clinical Chemistry study found 5.3% of regular users above 5 ng/mL at baseline after 48+ hours of abstinence, with no difference in driving simulator performance.

This means the per se laws, designed to identify impaired drivers, instead end up criminalizing sober people whose bodies are still clearing a substance that was legally consumed days earlier.

Standard Field Sobriety Test Limitations

The Standard Field Sobriety Tests (SFSTs) — Horizontal Gaze Nystagmus (HGN), Walk-and-Turn, and One-Leg Stand — were validated for alcohol impairment, not cannabis. Research on their accuracy for cannabis is thin and contradictory.

Commonwealth v. Gerhardt (Mass. SJC, 2017)

The Massachusetts Supreme Judicial Court ruled that SFSTs cannot be presumed scientifically valid for marijuana impairment. The court allowed SFST observations to be admitted as lay opinion about a person's "appearance or demeanor" but not as expert evidence of impairment.

HGN Does Not Work for Cannabis

The HGN test (tracking eye movement with a pen) shows "no discernible difference" before and after cannabis consumption in multiple studies. Yet it remains part of standard DUI protocol.

Baseline Failure Rates

One study found a 26% baseline failure rate on SFSTs among completely sober subjects. In other words, one in four sober people fails these tests, even with no drug or alcohol use. This raises serious questions about their evidentiary value for cannabis cases.

Drug Recognition Experts (DREs)

Drug Recognition Experts are police officers trained in a 12-step evaluation protocol to identify drug impairment beyond what SFSTs reveal. DRE accuracy for cannabis is contested:

  • An Arizona study found a 16.4% false positive rate — no drugs found in cases where DREs concluded impairment
  • Six U.S. states have refused to let DREs testify as experts because of reliability concerns
  • State v. Olenowski (NJ, 2023) limited DRE testimony to phrasing like "consistent with" rather than categorical impairment conclusions

What to Do If You Are Stopped

  • Know your state's law. Zero-tolerance vs. 5 ng/mL vs. impairment-based DUI all have different defense strategies.
  • Be cooperative but concise. You must provide identification and insurance. You are not required to answer questions about drug use.
  • Do not consent to searches unless legally required.
  • Understand implied consent — refusing a chemical test typically has separate administrative penalties.
  • Request a lawyer immediately if arrested.
  • Preserve evidence of your medical cannabis status if applicable — though this generally does not provide a defense to DUI charges even in medical cannabis states.
  • Consider expert testimony about per se law scientific limitations if your case goes to trial.

Related Reading