Urine, Saliva, Hair, Blood: What Each Drug Test Looks For

Different drug tests are not the same test performed on different samples — they look for different compounds over different time windows. Urine testing targets a metabolite; oral fluid and blood testing generally target the parent compound; hair testing targets what was incorporated into the growing shaft.

That distinction is why “how long does it stay detectable” has no single answer, and why the specimen your programme uses is one of the first things worth establishing. This post describes what each specimen measures. It does not give detection windows for individuals, because those depend on physiology, exposure history, and cutoffs.

Urine

The most common workplace specimen, and the one the rest of this site’s testing content assumes unless stated otherwise.

Analyte: THC-COOH, a metabolite produced after the body processes delta-9 THC. Not THC itself, and not cannabidiol.
What that implies: a urine test is evidence of metabolic processing, so it reflects exposure that has already been through the body rather than very recent exposure.
Structure: a two-stage process — an immunoassay screen with an administrative cutoff, then a mass-spectrometry confirmation at a lower cutoff. The stages are described in how a workplace urine screen works.
Additional checks: specimen validity testing for dilution, substitution, and adulteration, which can produce a result that is neither positive nor negative.
Why cannabinoids are distinctive here: THC and its metabolites are fat-soluble, which is why frequency of exposure affects urine findings differently from water-soluble substances.

Oral fluid

Increasingly used where recent use is the question, partly because collection is observed and therefore harder to interfere with.

Analyte: generally the parent compound, delta-9 THC, rather than the metabolite. Some programmes include metabolites as well.
What that implies: oral fluid results are weighted toward recent exposure. There is also a contamination pathway that urine does not have — residue in the mouth from something recently consumed, distinct from anything absorbed.
Cutoffs: far lower in absolute terms than urine cutoffs, because concentrations in oral fluid are lower. The numbers are not comparable between specimen types, and treating a urine cutoff as a reference point for an oral fluid test is a category error.
Confirmation: the same two-stage logic applies — screen, then mass spectrometry.

Blood

Used in clinical, forensic, and impairment contexts more often than in routine employment screening.

Analyte: typically delta-9 THC and sometimes its metabolites, measured directly.
What that implies: the narrowest window of the four, because blood concentrations change quickly. Blood is the specimen closest to “what is circulating now”.
Where it appears: post-incident testing, roadside follow-up in some jurisdictions, and medical settings.
A caution: blood concentration is not a measure of impairment for cannabinoids in the way it is treated for alcohol. That is a contested area and not one this site is equipped to adjudicate.

Hair

The longest window, and the one whose mechanism differs most from the others.

Analyte: compounds incorporated into the hair shaft as it grew. Laboratories differ in whether they target the parent compound, the metabolite, or both, and the choice is methodologically significant.
What that implies: hair reflects a period of growth rather than a moment. It is poor at establishing recent exposure and better at establishing a pattern.
The external contamination problem: hair can pick up substances from the environment. This is why methods include washing steps and why targeting a metabolite — which the environment does not deposit — is regarded as a stronger design than targeting the parent compound alone.
Practical variation: hair colour, treatment, and growth rate all affect results, which is one reason hair testing is used less in routine employment programmes than urine.

What the specimen does not change

Two things stay constant across all four, and they are the two that matter for a label-reading site.

None of them looks for cannabidiol. CBD is not the analyte in any standard cannabinoid panel and is not a precursor to the metabolite that urine testing targets. The question is always about THC content in what was consumed — which is a question about the product, and therefore about the THC rows on its lab report.

None of them identifies the source. A confirmation identifies a compound. It does not distinguish a lawfully purchased hemp product from any other route of exposure. This is the gap that catches people out, and it is the subject of does CBD show up on a drug test.

Why “how long does it last” is unanswerable here

Every published window is a population generalisation, and the variables that move it are exactly the ones a website does not know about you: frequency and quantity of exposure, body composition, metabolism, the specimen, the cutoff, and the laboratory’s method.

For cannabinoids specifically, fat solubility means cumulative exposure behaves differently from a single instance, which widens the spread further. Any site offering you a confident number is offering you a guess with the uncertainty removed, and the removal is the problem. We would rather say plainly that we do not know than give you a figure you might rely on.

Where the label work actually helps

You cannot change how a test works. You can establish what is in a product.

Find out which specimen your programme uses and what its cutoffs are. Ask the programme, not a retailer.
Read the written policy on hemp-derived products; many address them explicitly.
Get the batch-matched certificate for anything you use and read the THC rows — the walkthrough is here.
Read a non-detect together with its detection limit, as in why non-detect does not mean zero.
Discount the phrase THC-free as evidence, per what it can and cannot mean.
Treat delta-8 and other hemp-derived THC products as a different question entirely — they are forms of THC, not CBD, as delta-8 vs. CBD explains.
Escalate to the right professional. Testing programme for test questions, occupational health or a lawyer for employment consequences, pharmacist or doctor for anything about your own body or medication.

Eased is a consumer-information site about product labelling. Nothing here is medical or legal advice, and nothing here describes what CBD does or does not do in the body. Talk to a doctor or pharmacist before using any CBD product, especially if you take other medication, and check the current law where you live.