What Is Considered a High THC Level in Blood?

There is no single number that universally defines a “high” THC level in blood, because the answer depends on who is asking and why. Legal thresholds for driving impairment typically sit between 1 and 5 nanograms per milliliter (ng/mL) of whole blood, depending on the jurisdiction. But peak blood THC after smoking can easily exceed 100 ng/mL in a casual user and climb past 300 ng/mL in someone using concentrated products. What makes this question genuinely complicated is that THC in blood behaves nothing like alcohol in blood, and the science behind setting meaningful cutoffs is far less settled than most people realize.

Legal Thresholds Around the World

Many jurisdictions have adopted what are called per se limits for THC in blood, meaning any driver whose blood THC exceeds a specific concentration is automatically considered impaired under the law, regardless of how they actually performed behind the wheel. In the United States, states that have set numeric thresholds most commonly use 5 ng/mL of whole blood (Colorado and Washington are well-known examples). Several European countries set the bar much lower, at 1 ng/mL. Other places, including some U.S. states and parts of Europe, use a zero-tolerance approach: any detectable amount of THC in your blood is enough for a charge.

These numbers were not derived the same way alcohol limits were. Research in Italy found that when per se limits roughly equivalent to various blood-alcohol concentrations were applied to cannabis, the number of drug-driving cases dropped by about 79%, suggesting the thresholds screened out a large proportion of people who had used cannabis recently. ¹ That kind of drastic reduction raises legitimate questions about whether the cutoffs are meaningfully tied to crash risk or simply set so high that almost nobody exceeds them.

Why Blood THC Does Not Work Like Blood Alcohol

The reason alcohol-style per se limits are so controversial for cannabis is that blood THC concentration does a poor job of predicting whether someone is actually impaired. A recent review in the addiction literature put it bluntly: there are no empirically supported thresholds for blood or oral fluids that reliably indicate cannabis impairment. ² With alcohol, there is a relatively straightforward relationship between how much is in your blood and how impaired you are. With THC, that relationship is weak, inconsistent, and confounded by tolerance, body composition, and timing.

One core problem is that THC concentrations in blood spike almost immediately after inhalation, often peaking within five to fifteen minutes, then plummet rapidly even while the cognitive effects are still building. A driving-simulator study found that the highest risk of car crashes came about four hours after smoking, well past the point when blood THC had already dropped from its peak. ³ In other words, the moment your blood THC reads highest is not the moment you are most dangerous on the road. This is essentially backwards from how alcohol works, and it is the reason that researchers who have spent decades searching for a cannabis equivalent to the 0.08 g/dL alcohol limit have called that goal “a mirage.” ⁴

That said, the picture is not entirely random. A study of apprehended drivers in Norway found that those judged impaired by clinical examination had a median blood THC of 2.5 ng/mL, compared to 1.9 ng/mL for those judged not impaired. Drivers above 3 ng/mL had a meaningfully increased risk of being judged impaired. ⁵ So there is some signal in the noise, especially at higher concentrations. The problem is that the overlap between impaired and unimpaired people is enormous, which makes any single cutoff a blunt instrument.

Typical Blood THC Concentrations by Consumption Method

How you consume cannabis has a dramatic effect on how high your blood THC climbs. Smoking a joint or using a pipe delivers THC to the lungs, where it crosses into the bloodstream almost instantly. Peak blood concentrations in occasional users who smoke typically land somewhere in the range of 50 to 150 ng/mL within minutes, though there is wide individual variation. In a driving study comparing occasional and daily users, occasional users averaged about 6.4 ng/mL in whole blood after smoking, while daily users averaged around 36.4 ng/mL, reflecting the much higher baseline and accumulation in frequent users. ⁶

Concentrates push blood levels much higher. A naturalistic study found that people using cannabis concentrates (dabs, wax, shatter) reached average blood THC levels around 321 ng/mL, compared to about 140 ng/mL in flower users. ⁷ The same study noted something surprising: despite blood levels more than twice as high, concentrate users did not show significantly greater impairment on cognitive and psychomotor tests than flower users. This is one of the clearest illustrations of why raw THC numbers in blood can be misleading as a measure of how intoxicated someone actually is. Among inhalation methods, dabbing produced the highest plasma THC levels immediately after use, significantly higher than smoking a joint. ⁸ By one hour later, though, the differences between methods had largely disappeared.

Eating cannabis produces a completely different curve. Blood THC levels after oral consumption are much lower than after smoking the same dose, because the digestive system absorbs THC slowly and the liver converts much of it to metabolites before it ever reaches the general circulation. ⁹ One hallmark of oral cannabis use is that the ratio of the active metabolite 11-OH-THC to parent THC in blood is much higher than after smoking, because of that first-pass metabolism in the liver. ¹⁰ Forensic toxicologists sometimes use this ratio to help distinguish whether someone smoked or ate cannabis.

Whole Blood Versus Plasma

A detail that trips up a lot of people, including some professionals, is that THC concentrations differ substantially depending on whether the lab measures whole blood or plasma. THC tends to concentrate in the watery plasma fraction rather than in red blood cells, so plasma readings run significantly higher than whole blood readings from the same sample. Research has found that the whole-blood-to-plasma ratio for THC is roughly 0.39, meaning that a whole blood measurement is only about 40% of the plasma value. ¹¹ A separate study calculated a plasma-to-blood conversion factor of about 1.6, meaning you can roughly multiply a whole blood THC value by 1.6 to estimate the plasma concentration. ¹²

This matters because a “5 ng/mL” limit means something very different depending on the specimen type. If a jurisdiction sets its legal limit at 5 ng/mL in whole blood, that is roughly equivalent to about 8 ng/mL in plasma. If you are reading lab results or comparing numbers across studies, always check which matrix was tested. Many research studies report plasma values, while most legal per se limits are defined in whole blood. Confusing the two can make a result look twice as alarming, or half as significant, as it actually is.

Chronic Users and the Residual THC Problem

One of the most important complications in interpreting blood THC is that frequent cannabis users can have detectable, and sometimes legally significant, THC levels in their blood long after their last use. THC is highly fat-soluble. It gets stored in body fat and slowly leaches back into the bloodstream over days or even weeks.

In a controlled study of chronic daily smokers who entered a monitored facility and abstained completely, 27 of 30 participants were THC-positive on admission, with concentrations as high as 6.3 ng/mL. THC decreased gradually, but two out of five participants still had detectable THC at 0.3 ng/mL after 30 days of confirmed abstinence. About 5% of participants remained above 1.0 ng/mL for 12 consecutive days without using any cannabis. ¹³ A similar study measuring plasma found that 16 of 28 chronic frequent smokers still had THC at or above 2 µg/L (roughly equivalent to about 1.3 ng/mL in whole blood) a full 48 hours after their last use, and three of five participants tested positive at 30 days. ¹⁴

A systematic review synthesizing these and other studies concluded that blood THC above 2 ng/mL does not necessarily indicate recent cannabis use in frequent users. Some participants had blood THC above 5 ng/mL after just one day of abstinence, and a few maintained levels above 2 ng/mL for nearly a week. ¹⁵ This creates an obvious problem for per se driving laws: a daily user could be arrested, charged, and convicted based on residual THC in their blood even though they haven’t used cannabis in days and are not impaired in any meaningful sense.

Does Tolerance Change the Impairment Picture?

You might expect that someone who uses cannabis daily would develop tolerance and be less impaired at any given blood level than an occasional user. The research on this is more nuanced than you’d think. In the driving simulator study mentioned earlier, daily users had mean whole-blood THC more than five times higher than occasional users after smoking. Yet on a standard measure of driving impairment (how much a vehicle weaves within its lane), daily users showed a change that was not statistically distinguishable from non-users, while occasional users showed a significant increase in weaving. ⁶ That result looks like tolerance at work.

But another study challenged the idea that tolerance offers reliable protection for cognitive tasks. Looking at neurocognitive performance, researchers found that the acute effects of cannabis were similar across users regardless of their cannabis use history, suggesting an absence of tolerance for certain cognitive functions. ¹⁶ The takeaway is that tolerance may protect some driving-specific motor skills while leaving other cognitive processes just as vulnerable. A daily user with 30 ng/mL in their blood might hold a lane as well as a sober person but still have compromised reaction time or decision-making. This inconsistency is another reason a single blood number cannot reliably tell you whether someone is safe to drive.

Oral Fluid Testing as a Roadside Alternative

Because drawing blood on the roadside is impractical, many law enforcement agencies use oral fluid (saliva) tests as a screening tool. THC shows up in saliva partly through oral contamination from smoking and partly through transfer from the blood. During the elimination phase, after the initial contamination from smoke has cleared, THC concentrations in oral fluid tend to be in a similar range as blood concentrations. ¹⁷

A systematic review of the correlation between oral fluid THC and blood THC found that saliva tests are quite good at telling whether any THC is in the blood at all, with about 71% sensitivity and nearly 98% specificity. But saliva becomes much less reliable when the question is whether blood THC exceeds a specific per se limit like 5 ng/mL. ¹⁸ Part of the trouble is the sheer variability between individuals. Using data from a large U.S. roadside survey, researchers found that oral fluid THC concentration explained only about 29% of the variation in blood THC concentration. ¹⁹ That means roughly 71% of the variation in blood levels is unexplained by what shows up in spit. For detecting any recent use, saliva works reasonably well. For pinpointing a blood level, it is too imprecise to be treated as an equivalent.

When Lab Handling Changes the Numbers

Even when blood is drawn properly, what happens to the sample afterward can change the THC concentration the lab ultimately reports. THC is chemically unstable in stored blood, and concentrations can drift depending on temperature, container type, and how long the sample sits before analysis. In blood stored at refrigerator temperature (4°C), THC remained stable for about 12 weeks, but the metabolite THCCOOH-glucuronide began degrading after just 4 weeks. Freezing at −20°C extended THC stability to at least 12 weeks. ²⁰ Separate research has shown that cannabinoids tend to decrease in concentration over time when stored in untreated glass vials, and the recommendation from forensic scientists is that samples should be analyzed as soon as possible to get the most accurate reading. ²¹

If you are ever in a situation where your blood THC result matters legally, the chain of custody and storage conditions of the sample are worth scrutinizing. A sample that sat at room temperature for weeks before testing may report a lower THC concentration than what was actually in your blood at the time of the draw. In some cases, that could work in your favor; in others, if the original level was borderline, degradation might push it below a detection threshold and obscure what really happened.

Emergency Department Visits and Acute Toxicity

For clinical purposes, blood THC levels rarely drive treatment decisions the way blood alcohol does. Emergency physicians treat cannabis intoxication based on symptoms, not on a specific number. A review of emergency department presentations in Switzerland found that among people who had used cannabis alone, the most common symptoms were nausea and vomiting (26%), palpitations (25%), anxiety (23%), and chest pain (15%). Most of these cases were classified as minor, and 83% of patients were sent home from the emergency department. Severe complications, including psychosis (7%), coma (6%), and seizures (5%), did occur but were uncommon, and there were no fatalities. ²²

Emergency room physicians generally do not order quantitative blood THC levels because the result would not change what they do. The treatment is supportive: a calm environment, anti-nausea medication if needed, monitoring for the rare patient who becomes severely agitated or psychotic. By the time a quantitative blood result comes back from the lab, the patient has usually already recovered. Quantitative THC testing is far more useful in forensic, legal, and research settings than in clinical medicine.

Postmortem Blood THC Interpretation

Interpreting THC levels in blood from deceased individuals adds yet another layer of complexity. After death, THC redistributes within the body as cell membranes break down and blood pools in different areas. A study of postmortem cases found that central blood (drawn from the heart or major vessels) had a median THC concentration about 1.5 times higher than peripheral blood (drawn from a limb vein), suggesting modest redistribution toward the central compartment after death. ²³

A more recent meta-analysis confirmed that postmortem blood tends to show significantly higher THC concentrations than antemortem blood would. It also revealed an unusual pattern: in living people, the inactive metabolite THCCOOH is typically present at higher concentrations than THC itself, but in some postmortem cases, the reverse is true, with parent THC exceeding its metabolite. ²⁴ Forensic pathologists have to account for these shifts when interpreting postmortem toxicology results. A blood THC level of 10 ng/mL in a deceased person does not mean the same thing it would in a living driver, and directly comparing the two would be a mistake.

Sex Differences and Body Composition

Body fat plays a meaningful role in how THC distributes and how long it sticks around, and there appear to be sex-related differences as well. Animal research in mice found that males accumulated substantially more THC in adipose (fat) tissue than females, with peak fat tissue concentrations in males more than three times higher than in females. Males also reached their peak fat storage much later, at six hours compared to one hour in females. ²⁵ Translating mouse studies directly to humans requires caution, but the underlying biology is consistent with what pharmacologists already suspect: people with more body fat may store more THC and release it more slowly, contributing to the prolonged detection window seen in some heavy users. This could partly explain why two people who smoke the same amount end up with different blood levels the next day.