How Long CBD Stays in Your System: What the Data Shows

Half-life, in plain terms

Half-life is the working unit for this whole question. It names the time a measured concentration in blood takes to fall by half. One period removes half. The next removes half of what remains. That arithmetic is why clearance gets discussed in multiples of the half-life instead of a single cut-off day [1].

The 2018 systematic review by Millar collected the published human trials on cannabidiol pharmacokinetics [1]. What came out of that collection was a set of estimates, not one consensus figure [1]. The included trials differed in route, in dosing frequency, and in what they sampled, so the review reports a span rather than a number [1].

Here is the span. After a single administration, some routes produce a half-life of a few hours [1]. Oral cannabidiol given repeatedly sits at roughly two to five days [1]. Same molecule, same review, two very different orders of magnitude.

The width of that range is not a bookkeeping problem. Millar and colleagues present it as a finding in its own right [1]. Half-life and concentration estimates differ substantially by route of administration, and the review names that as the reason against averaging everything into one tidy value [1].

Which brings us to persistence, the thing most people actually mean when they ask. Because clearance plays out over multiples of the half-life, the review's figures translate into measurable presence lasting days and, at the long end of the range, into weeks [1]. Use pattern decides which end of that span applies [1].

So the honest answer has two halves: a range, and the conditions attached to it. A single figure would be easier to print. It would also be a figure the published human data does not support [1]. Since 2014 we have preferred the range with a source over the number without one.

None of this depends on the brand of oil in your cupboard. It depends on how cannabidiol was given, how often, and to whom, which is exactly how the review sorts its estimates [1]. Route first, then dosing pattern.

Where cannabidiol goes after absorption

Cannabidiol is lipophilic, meaning it dissolves better in fat than in water. Huestis, writing in 2007 on human cannabinoid pharmacokinetics, describes how cannabinoids of this kind partition into fatty tissue rather than staying in the bloodstream [2].

Two things follow from that in the 2007 account. Repeated exposure builds a tissue reservoir, and the tissue gives the compound back to the blood slowly [2]. Accumulation first, then slow release, both described in general terms [2].

This is also why the shape of a concentration curve can mislead. Redistribution out of fatty tissue changes what the terminal phase of the curve actually represents [2]. The late, low readings are not simply the last of one administration leaving; they can be tissue returning what it stored [2].

Put the two papers side by side and the estimates cluster in a way that makes sense. The longer half-life values come from repeated oral dosing [1][2]. The shorter ones come from single administration [1][2]. Frequency of use is the variable doing most of the work in that split, and the two to five day figure in the review belongs to the repeated dosing group [1].

Duration of use pushes in the same direction. In the 2007 account, fatty-tissue accumulation builds with repeated exposure, so a longer history of use implies a larger reservoir and a longer slow-release phase [2].

Body composition enters here too. Cannabinoids partition into fatty tissue, so the size of that compartment works as a distribution factor, and it differs from one person to the next [2].

What neither paper supplies is arithmetic for an individual. How much accumulates in a given person: unquantified in both publications [1][2]. How long that person's tissue compartments take to empty: also unquantified in both [1][2]. The mechanism is on record. The per-person sum is not [1][2].

That gap is worth sitting with, because it is where most of the confident numbers online come from. Someone takes a mechanism described in general terms and turns it into a day count for a stranger. The published sources stop well short of that [1][2].

Where the evidence thins out

Not every variable in this topic rests on the same footing, and the 2018 review is fairly open about it [1]. Route of administration and dosing pattern carry the most weight, because those are the lines along which the review's estimates are grouped and reported [1].

Body composition and individual metabolism sit on weaker ground [1]. Between-subject variation was recorded across the included studies, and metabolic differences are named as a source of it, but neither factor was isolated experimentally [1]. Sample sizes were small, which leaves wide confidence intervals around the pooled parameters [1].

So the direction is documented and the magnitude is not [1]. A quicker metabolism, a larger fat compartment: the review records that such differences show up in measured concentrations and in the parameters derived from them, without putting a size on either [1].

There is a larger blank still. The review contains no point of absence for any named individual [1]. No included study was designed for that endpoint [1]. That is not an oversight to complain about; it reflects what the trials set out to measure [1].

Which means the practical question, whether cannabidiol will be gone from one specific person by Friday, has no published answer in these sources [1]. We would rather say that plainly than invent a figure. Swiss habit.

A batch report answers a different question, and answers it well. Independent analysis tells you the cannabinoid content of the bottle in your hand, verified batch by batch. Pharmacokinetics is about what happens after that, and it is measured in people, in trials, with the spread those trials report [1].

Two reference points stay solid across both papers. In humans, cannabidiol's half-life spans hours to several days depending on route and dosing pattern [1]. Lipophilic cannabinoids accumulate in fatty tissue and are released slowly under repeated exposure [2]. Anything more specific is where the research is still open.

Read any number online with that in mind. Ask which route it came from, whether the dosing was single or repeated, and how many participants stood behind it [1].

Fourteen points from the two papers

1. Half-life names the time a measured blood concentration takes to fall by half, and clearance runs over multiples of it rather than one cut-off [1].
2. The 2018 review by Millar collected published human trials on cannabidiol pharmacokinetics, and its output is a set of estimates rather than a consensus figure [1].
3. After a single administration, some routes give a half-life of a few hours [1].
4. Oral cannabidiol under repeated dosing is estimated at roughly two to five days [1].
5. The width of that range is presented as a finding, not as a reporting flaw [1].
6. Half-life and concentration estimates differ substantially by route, which the review gives as its reason against pooling them into a single value [1].
7. Measurable presence is framed as days into weeks, across multiples of the half-life, and dependent on use pattern [1].
8. Longer estimates cluster around repeated oral dosing, shorter ones around single administration [1][2].
9. Huestis, in 2007, describes lipophilic cannabinoids partitioning into fatty tissue, accumulating with repeated exposure and releasing slowly [2].
10. Redistribution from fatty tissue changes what the terminal phase of a concentration curve represents [2].
11. A longer history of use implies a larger tissue reservoir and a longer slow-release phase [2].
12. Fat compartment size acts as a distribution factor and differs between individuals [2].
13. Between-subject variation in measured concentrations and derived parameters is recorded, with metabolic differences named as a source, and small samples leave wide confidence intervals [1].
14. Body composition and individual metabolism were not isolated experimentally, so direction is documented while magnitude is not [1].
15. Neither publication quantifies how much accumulates in a given person, nor how long that person's tissue compartments take to empty [1][2].
16. The review holds no point of absence for a named individual, and no included study was designed for that endpoint [1].
17. Established across both papers: hours to several days in humans by route and dosing pattern [1], plus fatty-tissue accumulation with slow release for lipophilic cannabinoids under repeated exposure [2].
18. Anything more precise than those two statements, for one person on one date, is not in these sources [1][2].