THC Blood Levels Correlate With Impairment
The claim that blood THC predicts impairment like BAC does is flatly contradicted by pharmacology and government research.
This myth is policy dressed up as science. Blood THC tells you when someone used cannabis, not how impaired they are right now. THC clears blood fast but leaves the brain slower. Chronic users show detectable blood THC while completely sober. Every major government review says per se cutoffs have no scientific basis. But the laws exist anyway because admitting we lack a clean test is politically difficult.
The Claim
Alcohol has a clear per se threshold: 0.08% blood alcohol concentration (BAC) in most US states reliably predicts impairment. The logic has been borrowed wholesale for cannabis: pick a number for THC in blood — commonly 2 nanograms per milliliter (ng/mL) or 5 ng/mL — and anyone above it is legally too impaired to drive, without needing to prove actual impairment.
This reasoning is written into the per se DUI laws of Washington, Colorado, Nevada, Ohio, Pennsylvania, Illinois (for some categories), and others [1], plus countries including Germany and parts of Australia [2]. Police, prosecutors, and much public commentary still invoke a blood THC number as if it reliably predicts how stoned a driver is.
The appeal is obvious: it mimics a legal framework that works for alcohol, and it offers prosecutors a simple, biochemical case instead of relying on messy field sobriety tests.
What the Evidence Actually Shows
Three core pharmacological facts demolish the alcohol analogy.
1. Blood THC clears fast; impairment does not.
After smoking, blood THC spikes within minutes and crashes within 1–2 hours — often before peak psychomotor impairment occurs [1][2]. This is because blood concentrations reflect what is in the plasma, but the behavioral effects track brain THC, which redistributes and lags behind. By the time a driver is pulled over, processed, and given a blood draw 1–3 hours later, blood THC has dropped sharply while the driver may still be notably impaired. Edibles complicate this further: they produce lower peak blood THC but longer subjective and objective impairment [3].
2. Chronic users carry residual THC for weeks without impairment.
THC is highly lipophilic (fat-soluble) and redistributes slowly out of adipose tissue. Daily cannabis users routinely show blood THC above 2 ng/mL — and sometimes well above 5 ng/mL — even during periods of complete abstinence, on cognitive and driving tests showing no acute impairment [4][5]. Karschner et al. (2009) documented residual THC in chronic users abstinent for up to 30 days [4]. This is the single biggest flaw in per se THC laws: they criminalize a marijuana user for being a marijuana user, not for being impaired.
3. The correlation between blood THC and actual driving performance is weak.
The most authoritative review is the US National Highway Traffic Safety Administration's 2017 report to Congress. NHTSA concluded flatly: "a specific level of THC in the blood does not necessarily mean that an individual is impaired" and that available research "does not support the development of impairment thresholds based on blood THC concentrations" [6] Strong evidence. The Congressional Research Service reached the identical conclusion in 2019. A 2013 meta-analysis by Wolff et al., commissioned by the UK Department for Transport, examined crash culpability and likewise found that the relationship between blood THC and driving impairment is inconsistent enough that per se cutoffs cannot be scientifically justified the way BAC cutoffs can [7].
Meta-analyses of experimental driving studies (e.g., Rogeberg & Elvik, 2016) do find that recent cannabis use roughly doubles crash risk — a real effect, but smaller than the alcohol effect at 0.08% BAC, and crucially not well predicted by any specific blood THC number [8] Strong evidence. The scientific consensus on this point is not contentious. It is settled. Policy simply has not caught up.
Where This Myth Came From
The per se THC idea is almost entirely a legislative borrowing, not a scientific discovery.
In the early 2000s, as US states began legalizing medical and then recreational cannabis, legislatures faced a straightforward problem: how do we write a DUI statute? Alcohol provided an obvious template. BAC works as a per se standard because ethanol's pharmacokinetics are simple and roughly linear across most people: blood concentration correlates with brain concentration, which correlates with impairment, all on similar timescales. Legislators assumed — or were advised — that THC would behave similarly.
A handful of early proposals, notably from some European working groups around 2005–2007, floated 7–10 ng/mL serum cutoffs, loosely based on driving simulator studies conducted in occasional users [7]. These were small, underpowered studies. They were never intended to establish a universal threshold. They were adopted as one anyway, often at lower numbers (down to 2 ng/mL), partly because lower cutoffs were easier to defend as "strict" politically.
The pharmacology community objected almost immediately. By the time NHTSA formalized its verdict in 2017, per se laws were already on the books in multiple states. Repealing them has proven politically difficult — no legislator wants to vote to "weaken DUI laws," even if the weakness is in the law, not the enforcement.
The myth persists less because anyone defends it scientifically and more because the alternative — admitting that we do not yet have a clean roadside biochemical test for cannabis impairment — is politically uncomfortable.
What Actually Works (and What Doesn't)
There is no cannabis equivalent of a breathalyzer that converts a single number into a legal presumption of impairment. Oral-fluid THC devices exist and can detect recent use, but they suffer from the same fundamental problem: presence and concentration do not equal impairment.
The approaches with the best supporting evidence are behavioral:
Standardized Field Sobriety Tests (SFSTs) and Drug Recognition Expert (DRE) assessment. When administered by trained personnel, these detect cannabis impairment with moderate accuracy [9]. They are not perfect and have their own validity debates, but they measure something closer to what actually matters: can this person safely operate a vehicle right now.
Combined-substance detection. Most cannabis-involved crashes also involve alcohol or other drugs. Looking for polysubstance use remains the highest-yield enforcement target.
For individuals: time-based abstinence. The practical guidance from impairment research is straightforward: do not drive for at least 3–4 hours after inhaling cannabis, longer (6–8+ hours) after consuming edibles, and remember that tolerance may mask subjective dizziness or euphoria without fully protecting driving-relevant skills like reaction time and attention [3]. This is not perfect but is more honest than any blood cutoff.
Legal risk for patients. If you are a medical cannabis patient in a per se THC state, you should know that you can be charged with DUI while completely unimpaired, based on residual blood THC alone, even if field sobriety tests are normal. This is a real legal exposure that the science does not justify but the law does not care about. Consult a local defense attorney about the specific per se threshold in your jurisdiction and the presence of any residual-THC defense.
Bottom Line
Blood THC is a marker of recent (or sometimes not-so-recent) cannabis exposure. It is not a measure of impairment. The alcohol analogy fails for clear pharmacokinetic reasons: THC clears blood before the brain, chronic users retain blood THC for weeks without impairment, and the correlation between any specific blood level and driving performance is weak.
Every major government-commissioned review of this question over the last two decades — NHTSA, the Congressional Research Service, and the UK Department for Transport — has reached the same verdict: per se THC cutoffs are not scientifically justified.
If you hear someone — a police officer, a prosecutor, a journalist, a well-meaning friend — claim that a specific nanogram number means a specific level of impairment, they are wrong. The honest answer is that we do not yet have a good biochemical proxy for cannabis impairment, and the laws written as if we do are bad laws. See Cannabis and Driving Impairment Laws for state-by-state details and Cannabis Driving Impairment Laws in Colorado, Washington, and other jurisdictions for how per se thresholds are applied in practice.
Sources
- Government National Highway Traffic Safety Administration (2017). Marijuana-Impaired Driving: A Report to Congress. US Department of Transportation, DOT HS 812 440.
- Government Congressional Research Service (2019). Marijuana Use and Highway Safety. CRS Report R45719.
- Peer-reviewed Huestis MA. (2007). Human cannabinoid pharmacokinetics. Chemistry & Biodiversity, 4(8), 1770-1804.
- Peer-reviewed Grotenhermen F. (2003). Pharmacokinetics and pharmacodynamics of cannabinoids. Clinical Pharmacokinetics, 42(4), 327-360.
- Peer-reviewed Karschner EL, Schwilke EW, Lowe RH, et al. (2009). Do Delta9-tetrahydrocannabinol concentrations indicate recent use in chronic cannabis users? Addiction, 104(12), 2041-2048.
- Peer-reviewed Bergamaschi MM, Karschner EL, Goodwin RS, et al. (2013). Impact of prolonged cannabinoid excretion in chronic daily cannabis smokers' blood on per se drugged driving laws. Clinical Chemistry, 59(3), 519-526.
- Peer-reviewed Rogeberg O, Elvik R. (2016). The effects of cannabis intoxication on motor vehicle collision revisited and revised. Addiction, 111(8), 1348-1359.
- Peer-reviewed Hartman RL, Richman JE, Hayes CE, Huestis MA. (2016). Drug Recognition Expert (DRE) examination characteristics of cannabis impairment. Accident Analysis & Prevention, 92, 219-229.
- Peer-reviewed McCartney D, Arkell TR, Irwin C, McGregor IS. (2021). Determining the magnitude and duration of acute Δ9-tetrahydrocannabinol (Δ9-THC)-induced driving and cognitive impairment: A systematic and meta-analytic review. Neuroscience & Biobehavioral Reviews, 126, 175-193.
How this page was made
Generation history
Drafting assistance and fact-check automation are used, with a human operator spot-checking on a weekly basis. See how articles are made.
Related
- Cannabis and Driving Impairment Laws — Legal frameworks, testing methods, and the gap between law and impairment science.
- Cannabis Driving Impairment Laws in Colorado — Colorado law, enforcement, penalties, and the limits of THC blood testing for impairment.
- Cannabis Driving Impairment Laws in Washington — Washington's 5 ng/mL per se limit, search warrant rules, and the science-law gap.