How Much Weed Is 5 Nanograms Per Milliliter?

There is no single amount of cannabis that reliably produces exactly 5 nanograms of THC per milliliter of blood, because the number depends on how you consumed it, how often you use cannabis, your body composition, and how much time has passed since your last dose. A single puff of a moderate-potency joint can send blood THC levels far above 5 ng/mL within seconds, only for the number to crash back below it within a couple of hours. Meanwhile, a daily user who has not touched cannabis for a day or more can still exceed 5 ng/mL without having consumed anything recently. The figure matters because it is the legal per se limit for driving under the influence in several U.S. states, yet it tells a surprisingly incomplete story about impairment.

What Happens to THC in Your Blood After Smoking

When you inhale cannabis smoke or vapor, THC enters the bloodstream through the lungs almost immediately. Blood concentrations spike within the first few minutes and then fall steeply. In a controlled study where chronic frequent users smoked a single cannabis cigarette, the median peak THC level in plasma hit about 28 ng/mL half an hour after the session, while those allowed to smoke freely reached a median peak of roughly 84 ng/mL in plasma.¹ Another controlled smoking study reported a median whole-blood peak of about 50 ng/mL just 15 minutes after the first puff.²

So even a modest smoking session can produce blood THC levels that are ten or twenty times the 5 ng/mL legal threshold at their highest point. But that peak is fleeting. THC redistributes into fatty tissues and the brain within minutes, and blood levels drop steeply. One study found that the peak concentration fell by about 74% within the first half hour after smoking, and by roughly 90% within an hour and a half.³ In practical terms, a single moderate hit of cannabis could push your blood THC well past 5 ng/mL for an hour or two, then leave you below the threshold even though you still feel some effects.

Whole Blood, Plasma, and Serum Are Not the Same Number

One of the most confusing aspects of the “5 ng/mL” question is that it depends on which fluid the lab actually tests. THC binds heavily to proteins in plasma and does not distribute well into red blood cells. As a result, THC concentration in plasma runs roughly double the concentration in whole blood for the same sample. The median whole-blood-to-plasma ratio for THC is about 0.39, meaning whole blood reads around 40% of the plasma value.⁴ A separate study confirmed this with a multiplying factor of about 1.6 from whole blood to plasma.⁵

This distinction matters for anyone trying to understand their own test result. If a state law sets the limit at 5 ng/mL in whole blood, a plasma reading of 5 ng/mL would actually correspond to a whole-blood level of only about 2 ng/mL, well under the limit. Some state statutes specify “whole blood” while others say “blood” without clarifying. If you are reading research papers, most pharmacokinetic data reports plasma values because plasma is easier to work with in the lab. When a study says peak THC hit 76 ng/mL, that probably means plasma, and the corresponding whole-blood figure would be closer to 30–40 ng/mL. Always check which matrix is being reported before comparing numbers across studies or to a legal limit.

How Edibles Compare

If smoking produces a fast, sky-high spike that crashes quickly, edibles produce the opposite pattern: a slow rise, a modest peak, and a long tail. In a controlled oral dosing study, participants who ate brownies containing 25 mg of THC reached an average peak of only about 3.5 ng/mL in whole blood, and those who ate 50 mg peaked at roughly the same level, around 3.3 ng/mL.⁶ That is remarkably low compared to the peaks seen after smoking. The 10 mg dose barely registered at about 1 ng/mL in whole blood.

These numbers surprise a lot of people. A standard recreational edible in many legal states is 10 mg per serving, and even eating two or three servings would be unlikely to push whole-blood THC above 5 ng/mL based on this data. But there are important caveats. Absorption from the gut varies wildly from person to person. Some people absorb THC from food much more efficiently than others, and a high-fat meal can increase absorption. The study also noted that THC from edibles remained detectable in blood for up to 22 hours, even though levels were low. So while edibles tend to keep you under the per se limit by a wider margin, the window during which any THC is detectable stretches much longer.

Why Daily Users Are a Special Case

Regular cannabis users accumulate THC in body fat over time. Because THC is highly fat-soluble, it slowly leaches back into the bloodstream long after the last use, keeping blood levels measurably above zero for days. In a study of nearly 200 frequent cannabis users who stopped using, about 5% still had blood THC above 5 ng/mL at baseline after a period of short-term abstinence, and roughly 24% exceeded 2 ng/mL. Even with zero-tolerance cutoffs, 43% of participants still tested positive.⁷

For a daily user, then, the question “how much weed is 5 ng/mL” might have a genuinely uncomfortable answer: zero. You might not have smoked anything in the past day and still blow past the legal limit. This is one of the fundamental criticisms of per se THC limits. Unlike alcohol, which clears the body at a relatively predictable rate, THC hangs around in chronic users at levels that can exceed legal driving limits with no recent intoxication whatsoever.

Concentrates Push the Numbers Much Higher

Modern cannabis products span a huge potency range. Flower typically contains somewhere between 15% and 30% THC, but concentrates like wax, shatter, and live resin can exceed 70% or more. Blood THC levels after concentrate use reflect that difference dramatically. One study cited in a review of cannabis concentrate research found that blood THC immediately after concentrate use averaged about 321 ng/mL, compared to a mean of 9 ng/mL in a study of THC-positive drivers who had used typical cannabis products.⁸ That is more than 60 times the 5 ng/mL legal threshold, reached instantly.

Even though those levels drop rapidly, the sheer magnitude of the peak means it takes longer to fall below 5 ng/mL after using a concentrate than after smoking flower. If you are thinking about the legal limit in practical terms, the type and potency of the product matters enormously. A single dab of high-potency concentrate is a fundamentally different pharmacological event than a single puff from a mid-grade joint, even though both are “using weed.”

Does 5 ng/mL Actually Mean You Are Impaired?

This is the question that makes the per se limit so contentious. With alcohol, blood concentration correlates reasonably well with impairment across most people. With THC, that relationship is much weaker. A review of driving skills research concluded that blood THC concentrations in the range of 2 to 5 ng/mL are associated with substantial driving impairment, particularly in occasional smokers.⁹ But the picture gets muddier with regular users, who develop tolerance and can function at THC levels that would meaningfully impair someone who smokes once a month.

A simulated driving study illustrated the disconnect vividly. Half an hour after vaporizing cannabis, all participants had plasma THC levels above per se legal limits, but 46% of them showed no measurable driving impairment. Three and a half hours later, the situation reversed: most participants had blood THC below 1 ng/mL, yet 57% showed significant impairment at that point.¹⁰ In other words, the legal limit caught people who were not impaired and missed people who were. THC blood level and impairment are related, but they run on different clocks.

A meta-analysis comparing cannabis and alcohol impairment found that cannabis effects vary more between individuals than alcohol effects do, because of differences in tolerance, smoking technique, and THC absorption.¹¹ Researchers who have examined the relationship from an epidemiological angle suggest that a serum THC concentration of 7 to 10 ng/mL correlates with impairment comparable to a blood alcohol level of 0.05%, and that serum concentrations below 10 ng/mL are not clearly associated with elevated crash risk.¹² Because serum and whole blood differ by roughly that 1.6 factor, 7 to 10 ng/mL in serum works out to about 4 to 6 ng/mL in whole blood, which is right in the neighborhood of the 5 ng/mL legal limit. So the limit is not arbitrary, but it is a blunt instrument applied to a highly variable biological process.

Crash Risk at 5 ng/mL

An updated meta-regression of culpability studies estimated that crash culpability risk begins to increase noticeably at blood THC levels of around 1.5 to 3 ng/mL, roughly doubles at 5 ng/mL, and potentially quadruples at 10 ng/mL.¹³ A prospective study looking at individual drivers found that those with THC at or above 5 ng/mL had an adjusted odds ratio of 1.74 for causing a crash, though that particular estimate did not reach statistical significance due to the small number of cases.¹⁴ For comparison, an alcohol level of 0.08% is associated with a crash risk multiplier closer to five to ten times baseline. Cannabis at 5 ng/mL appears to carry real but more modest added risk.

The Blood Draw Timing Problem

In practice, there is usually a significant delay between when a driver is pulled over and when blood is actually drawn. That delay matters because THC drops so quickly. One study found that THC concentrations previously associated with impairment during driving fell to a median below 5 ng/mL by about three and a half hours after dosing, and below 2 ng/mL by about five hours. Only one participant still had THC at 5 ng/mL or higher after three and a half hours. The researchers concluded that forensic blood THC concentrations can end up far below per se cutoffs even though the driver greatly exceeded them while actually behind the wheel, and that back-calculating the driving-time concentration is not scientifically reliable.³

In Colorado, an analysis of blood draw delays in vehicular homicide and assault cases concluded that many tests are unlikely to confirm that cannabis-impaired drivers had THC levels above the legal limit by the time blood is actually collected.¹⁵ This creates a perverse situation where a genuinely impaired driver whose blood is drawn an hour or two after the stop may test below the legal limit, while a frequent user who has not smoked in 24 hours might still test above it. The 5 ng/mL number, in isolation, cannot distinguish between these two cases.

Oral Fluid Roadside Tests and Their Limitations

Because drawing blood takes time and a clinical setting, several jurisdictions have moved toward oral fluid (saliva) testing at the roadside. The appeal is obvious: you get a result in minutes, no hospital trip required. But oral fluid THC concentrations do not track blood concentrations in a straightforward way. A systematic review found that oral fluid testing at commonly used cutoff values is less sensitive and less specific for detecting people whose blood THC is above 5 ng/mL. The result is a large number of false positives if oral fluid is used as a proxy for “illegal” blood THC in randomly stopped drivers.¹⁶

The disconnect exists partly because THC coats the inside of the mouth during smoking, producing very high oral fluid concentrations that have nothing to do with what is in the bloodstream. That contamination fades, but on a different schedule than blood levels. So roadside saliva tests can flag someone who smoked hours ago and is no longer impaired, while also missing someone who ate an edible and has measurable blood THC but very little in their saliva.

Can Exercise or Stress Raise Your Blood THC?

Because THC is stored in fat, anything that breaks down fat can release stored THC back into circulation. A study of regular cannabis users found that moderate exercise increased plasma THC levels, suggesting that working out can push THC out of fat stores and into the blood.¹⁷ Animal research has also shown that food deprivation and stress hormones enhance the release of THC from fat back into the bloodstream.¹⁸

In practice, these re-released amounts are typically small and unlikely to push a casual user over 5 ng/mL on their own. But for a heavy daily user who already has residual THC circulating near the threshold, a hard workout or a stressful event could plausibly nudge the number upward. It is another illustration of why a snapshot blood reading, taken at one moment in time, is a poor proxy for what a person consumed or how impaired they are.

Secondhand Smoke and Passive Exposure

People sometimes worry about testing positive from being in the same room as someone who is smoking. A controlled study placed non-smokers in an unventilated room with people actively smoking cannabis. The non-smokers’ blood THC peaked at an average of 3.2 ng/mL immediately after the exposure session, and THC remained detectable for one to three hours afterward.¹⁹ None of those non-smokers reached 5 ng/mL, but 3.2 ng/mL is close enough to raise eyebrows, and the study used an extreme scenario with no ventilation. In any normally ventilated space, passive exposure is unlikely to produce detectable blood THC. Under the most extreme hotbox conditions, though, it can approach the neighborhood of legal limits without quite hitting them.

Breath-Based THC Detection

Breath testing for THC is an active area of research, with several companies developing devices meant to work like alcohol breathalyzers. Early data shows a correlation between THC in exhaled breath and THC in blood, though the relationship is not tight enough yet to map a breath reading to a specific blood concentration. One study found a correlation coefficient of about 0.7 between breath and blood plasma THC over a two-hour post-smoking window, which is encouraging but far from the near-perfect correlation seen with alcohol breathalyzers.²⁰ Another study found that while higher breath THC concentrations tend to track higher blood concentrations, individual measurements of THC in breath cannot be reliably linked to impairment.²¹ Breath-based devices may eventually serve as a useful screening tool for recent use, but they are unlikely to replace blood testing as the legal standard anytime soon, and the deeper problem of THC levels not mapping cleanly onto impairment would persist regardless of the collection method.

Sample Handling and Forensic Accuracy

Even after blood is drawn, the reliability of the number depends on how the sample is handled. THC degrades in stored blood over time, and the rate of degradation varies with temperature, whether preservatives were added, and whether the tube was exposed to light. A review of the forensic literature found that storage conditions are a critical and under-appreciated factor in cannabinoid stability across blood, plasma, urine, and oral fluid samples.²² If a blood sample sits at room temperature for hours before processing, the THC concentration the lab eventually reports could be meaningfully lower than what was in the blood at the time of the draw. In legal proceedings where the difference between 4.8 and 5.2 ng/mL determines a conviction, that kind of analytical uncertainty matters. Defense attorneys in DUI cannabis cases increasingly raise sample handling as a challenge, and the science backs up the concern that the number on the lab report might not precisely reflect what was in the driver’s bloodstream at the time of the stop.