How Long Does a Saliva Drug Test Detect Substances?

Saliva drug tests generally detect most substances for one to three days after use, though the window varies widely depending on the specific drug, how often you use it, and the cutoff concentration the test applies. Cannabis can show up for roughly 24 to 72 hours in occasional users but has been detected for over a week in heavy daily smokers. Cocaine and its metabolites typically clear oral fluid within about 12 to 24 hours, while amphetamines and methamphetamine can linger for several days. These ranges are rougher than most people expect, and the reasons for that messiness are worth understanding.

Cannabis in Oral Fluid

THC is one of the trickiest substances for saliva testing because it enters oral fluid through two separate routes. Like other drugs, THC circulates in the blood and diffuses into saliva. But when cannabis is smoked or vaped, THC also coats the lining of the mouth directly, creating a depot of drug that sticks around and slowly re-releases into the fluid collected during a test.¹ That oral contamination can produce extremely high THC concentrations right after smoking, which then decline steeply over the next few hours.

In a controlled study of heavy cannabis users who stopped cold turkey, THC was detected in oral fluid up to 78 hours (about three days) after their last use.² But in another study of frequent smokers admitted for detoxification, THC turned up in oral fluid for as long as eight days. Researchers noted a peculiar pattern: negative samples were interspersed among positive samples days after cessation, meaning the drug seemed to come and go rather than declining steadily.³ This erratic behavior likely reflects THC being slowly released from fatty tissue back into the bloodstream and then into saliva at variable rates.

One nuance that matters for cannabis testing is the metabolite THC-COOH. This breakdown product does not exist in cannabis smoke, so its presence in oral fluid confirms that THC was actually metabolized in the body rather than simply deposited in the mouth from passive exposure to smoke.⁴ Some labs look for THC-COOH specifically to distinguish a real positive from environmental contamination, though the federal oral fluid panel primarily targets parent THC.

Cocaine and Its Metabolites

Cocaine moves into oral fluid fast. In a controlled dosing study, cocaine appeared in saliva within about five to twenty minutes after administration and was eliminated relatively quickly, with a half-life of roughly one to four hours.⁵ That means parent cocaine itself fades from oral fluid within several hours. The metabolite benzoylecgonine (BE) sticks around longer, with a half-life of roughly three to fourteen hours in the same study.

An evaluation of rapid oral fluid test devices after a 25 mg intravenous cocaine dose found that the median last detection time for a BE test strip was about 12.5 hours.⁶ In real-world use, where doses are often higher and may be repeated, the detection window could stretch beyond a day. Smoking or snorting cocaine also creates a local depot effect in the mouth and nasal passages similar to what happens with smoked cannabis, temporarily boosting oral fluid concentrations well above what would be predicted from blood levels alone.¹

Amphetamines, Methamphetamine, and MDMA

Amphetamine-type stimulants are basic drugs that concentrate in saliva at levels similar to or higher than those in blood plasma.¹ Because saliva is naturally slightly acidic compared to blood, basic compounds tend to become trapped in it once they diffuse across, which inflates oral fluid concentrations and extends detection times.

In a study of patients admitted for detoxification after heavy stimulant use, amphetamine and methamphetamine were detected in oral fluid for up to eight days, which was actually longer than the detection window in urine at the cutoff values applied.⁷ For single-dose or moderate use, the window is shorter. A study of MDMA (ecstasy) found the drug detectable in oral fluid for at least five hours after a single 75 mg dose.⁸ Most practical guidance puts the typical window at one to three days for casual users, though heavy chronic use can push it considerably further.

Opioids

Opioids as a class are basic drugs, and many of them appear in oral fluid at concentrations comparable to or exceeding those in blood. One advantage saliva has over urine for opioid detection involves heroin. The heroin metabolite 6-monoacetylmorphine (6-MAM) is a specific marker of heroin use, and it shows up with much higher prevalence in oral fluid than in urine, where it degrades quickly.⁹ This makes saliva testing especially useful when the question is whether someone used heroin versus another opiate.

The federal SAMHSA oral fluid panel includes cutoffs for morphine, codeine, hydromorphone, hydrocodone, oxymorphone, oxycodone, and 6-MAM.¹⁰ General detection windows for most opioids in oral fluid range from about one to three days, though the exact duration depends on the specific compound, the dose, and whether the opioid was taken sublingually (like buprenorphine), which creates a local depot effect and elevates oral concentrations.

Benzodiazepines

Benzodiazepines are an interesting case for oral fluid testing because saliva tends to carry the parent drug, whereas urine often contains only the metabolites. That difference means oral fluid can sometimes identify which specific benzodiazepine was taken, not just that one was present.⁹

Detection windows vary substantially across different benzodiazepines. In a study of patients admitted for detoxification after heavy, repeated benzodiazepine use, diazepam (Valium) was detected in oral fluid for up to seven days, clonazepam (Klonopin) for up to five days, and alprazolam (Xanax) for up to two and a half days. Researchers confirmed that these subjects were not taking additional doses during the observation period because their urine concentrations were steadily declining.¹¹ After a normal therapeutic dose, detection times would be considerably shorter.

Oxazepam, another common benzodiazepine, had a median detection time of about 67 hours (roughly three days) in oral fluid after an experimental single dose. The sleep medication zopiclone, which is not a benzodiazepine but works on the same receptor, was detectable for a median of about 59 hours in oral fluid, and it was actually detectable longer in saliva than in urine.¹²

Alcohol

Alcohol in saliva tracks blood alcohol concentration fairly closely. The concentration-time profiles in blood, breath, and saliva agree well within individuals, though saliva ethanol levels tend to run slightly above those in blood.¹³ This means saliva ethanol testing can estimate intoxication reasonably well, with one important caveat: if you have just taken a drink, residual alcohol in the mouth can produce artificially high readings that do not reflect your actual blood level. One study found a poor correlation between oral fluid and breath alcohol concentrations in subjects who had been drinking immediately before being tested.¹⁴

Because the body eliminates alcohol at a fairly predictable rate, saliva ethanol is really only useful for detecting current or very recent drinking, not use from the previous day. For most people, alcohol clears from saliva within 12 to 24 hours of the last drink, and usually much sooner.

Synthetic Cannabinoids

Synthetic cannabinoids (the chemicals in products once marketed as “Spice” or “K2”) pose a particular challenge for oral fluid testing because the compounds change rapidly as manufacturers modify their formulations to stay ahead of regulations. A study on JWH-018, one of the earlier synthetic cannabinoids, found peak concentrations in oral fluid about 20 minutes after smoking, with the compound still detectable at 12 hours after a single session in some subjects but not all.¹⁵

Newer third-generation synthetic cannabinoids (compounds like 5F-ADB and CUMYL-4CN-BINACA) can be detected in oral fluid using specialized extraction methods, though the analytical techniques required are more complex than standard workplace testing panels.¹⁶ Standard point-of-care tests are not designed to catch these substances, meaning most routine saliva tests would miss them entirely.

Why the Ranges Are So Wide

If the detection windows above seem frustratingly imprecise, that reflects the real science. Several factors push times longer or shorter for any given person:

Route of use: Smoking, vaping, or snorting a drug deposits it directly in the oral cavity, creating a local reservoir that persists independently of what is in the blood. This depot effect inflates oral fluid concentrations for hours after use and is the main reason smoked cannabis and snorted cocaine produce longer-than-expected detection times in saliva.
Saliva pH: Because most drugs of abuse are basic compounds, they concentrate in saliva (which is slightly acidic relative to blood) through ion trapping. But saliva pH varies between individuals and even within the same person throughout the day. Deliberate alteration of saliva flow rate and pH can produce roughly twofold changes in drug concentrations.¹⁷
Chronic vs. single use: A person who has used a drug heavily for weeks will take longer to clear it. The eight-day amphetamine and eight-day THC windows described above were both from heavy, repeated use in clinical populations, not from casual one-time exposure.
Food and stimulation: Eating, drinking, or using anything that changes saliva flow can dilute drug concentrations. This is why most collection protocols require the person to avoid eating or drinking for at least 10 minutes before the sample is taken.¹
Drug chemistry: How easily a drug crosses from blood into saliva depends on its lipid solubility, how strongly it binds to blood proteins, and its acid-base properties.¹⁸

How Saliva Compares to Urine

The most common comparison people want is saliva versus urine. In a broad overview, urine offers a detection window of roughly one to three days for most drugs, while saliva has traditionally been described as capturing very recent use in a window of about one to 36 hours.¹⁹ That 36-hour figure underestimates the saliva window for several drugs when you consider the controlled studies discussed above, but the general principle holds: saliva is best at detecting use within the past day or two, while urine can catch use a bit further back.

A workplace drug testing comparison of paired urine and oral fluid samples made this concrete. Substances were detected in about 3.7% of urine samples but only 0.5% of oral fluid samples. In 49 cases, a substance was detected in urine alone; in only one case was something detected in oral fluid alone.²⁰ This does not mean saliva testing is bad. It means the two matrices catch different windows of use, and most of those extra urine-only positives reflected drug use that had fallen outside the saliva detection window by the time of testing.

One area where saliva actually outperforms urine is in identifying specific drugs rather than just their metabolic byproducts. Oral fluid tends to carry parent drugs, while urine predominantly contains metabolites. This is why saliva can distinguish heroin from other opiates (via 6-MAM) and can identify which benzodiazepine was taken, rather than just showing a generic positive for the class.

Can You Beat a Saliva Test

Products marketed as oral fluid adulterants do exist, and researchers have tested them. Two commercial products, sold under brand names for defeating saliva drug tests, were found to work primarily by rinsing residual drug compounds out of the mouth rather than by destroying the drugs or changing salivary pH. An ordinary mouthwash performed a similar function.²¹ Common foods, beverages, and cosmetics tested did not produce false positives when the test was conducted 30 minutes after consumption.

For cannabis specifically, even a water rinse produced a statistically significant reduction in THC concentrations in oral fluid. But in the same study, peak THC concentrations 15 minutes after smoking were still more than three times the cutoff used by Spanish police for roadside testing, regardless of which mouthwash was used.²² Rinsing your mouth knocks concentrations down, but it does not eliminate detection if the test is administered soon after use.

How Accurate Are the Devices

Not all oral fluid screening devices are equal, and their accuracy varies by drug. In a roadside evaluation of four commercial devices used by police officers, overall sensitivity (the ability to correctly flag a positive) ranged widely. For cocaine, the best-performing device caught 97% of positives, while the worst caught 81%. For THC, the gap was even wider: the top device caught 92% of positives, but the lowest-performing device caught only 38%.²³ The researchers concluded that laboratory confirmation of any on-site screening result is always necessary because of the significant number of false positives and missed positives.

A separate evaluation of three rapid devices at substance abuse clinics found similarly uneven results. Methamphetamine and opiate detection were reasonably reliable across devices, but THC sensitivity was abysmal in all three: one device caught 22% of THC-positive cases, and two others caught none at all.²⁴ THC is the hardest drug to detect reliably with rapid oral fluid screens, which is ironic given that cannabis is among the most commonly tested substances.

The Collection Device Matters Too

Even the physical device used to collect the saliva sample can influence results. An in vitro comparison of several commercial collection devices found that the volume of oral fluid actually collected ranged from about 0.8 to 1.9 mL across devices, and the percentage of that volume that could be recovered for analysis ranged from 18% to 83%. Recovery rates for specific drugs varied even more: amphetamine recovery ranged from 16% to 59% depending on the device, while cocaine recovery ranged from 61% to 97%.²⁵ A device that retains most of the drug on its absorbent pad rather than releasing it into the buffer solution for analysis will naturally produce lower measured concentrations and shorter apparent detection windows.

Federal Cutoffs and What They Mean

Whether a test comes back positive depends not just on whether the drug is present but on whether its concentration exceeds the cutoff threshold the lab applies. The federal SAMHSA oral fluid guidelines set screening cutoffs for THC at 4 ng/mL, cocaine and opioids at 30 ng/mL, amphetamines at 50 ng/mL, and 6-MAM (heroin marker) at 4 ng/mL.¹⁰ These thresholds were updated in 2015, and the new cutoffs had real consequences for detection rates. In an addiction-medicine population, a simulation found that the updated cocaine and BE cutoffs confirmed only about 65% of previously positive samples, and for morphine and codeine the confirmation rate was about 44%.²⁶ In other words, where the line is drawn on the concentration scale can change your detection window by a day or more.

Different employers and different countries set their own thresholds. Roadside drug testing in Australia uses different cutoffs than workplace testing in the United States, which in turn differs from the European DRUID project standards. When someone asks “how long will this stay in my saliva,” the honest answer depends partly on which test, run by which lab, at which cutoff, using which collection device.

When Oral Fluid Cannot Estimate Blood Levels

A common misconception is that because a drug shows up in saliva, you can work backward to estimate how much is in the blood. For most drugs, there is a general correlation between oral fluid and blood concentrations, but the variation is so large that oral fluid levels should not be used to estimate blood levels for any individual person.²⁷ THC and lorazepam showed especially poor correlations between oral fluid and blood. A systematic review found that oral fluid THC had good sensitivity (about 71%) and specificity (about 98%) for simply detecting the presence of THC in blood, but performed less well when trying to identify people above a specific blood concentration threshold like 5 ng/mL.²⁸ This distinction matters for impaired driving laws, where legal limits are often defined in terms of blood concentration rather than oral fluid concentration.

1
PubMed Central. Drug testing in oral fluid
2
Forensic Science International. Residual cannabis levels in blood, urine and oral fluid following heavy cannabis use
3
PubMed. Detection Time for THC in Oral Fluid After Frequent Cannabis Smoking
4
PubMed Central. Oral fluid cannabinoid concentrations following controlled smoked cannabis in chronic frequent and occasional smokers
5
PubMed Central. Pharmacokinetics of Cocaine and Metabolites in Human Oral Fluid and Correlation with Plasma Concentrations following Controlled Administration
6
PubMed. Cocaine and benzoylecgonine oral fluid on-site screening and confirmation
7
PubMed. Extended Detection of Amphetamine and Methamphetamine in Oral Fluid
8
Forensic Science International. Validation of an ELISA-based screening assay for the detection of amphetamine, MDMA and MDA in blood and oral fluid
9
PubMed. Screening for drugs of abuse: which matrix, oral fluid or urine?
10
Toxicologie Analytique et Clinique. SAMHSA drug panel screening in oral fluid: Development of screening method at 8 seconds per sample using LDTD-MS/MS
11
PubMed. Detection Times of Diazepam, Clonazepam, and Alprazolam in Oral Fluid Collected From Patients Admitted to Detoxification, After High and Repeated Drug Intake
12
Journal of Analytical Toxicology. Detection Time of Oxazepam and Zopiclone in Urine and Oral Fluid after Experimental Oral Dosing
13
Clinical Chemistry. Pharmacokinetics of ethanol in saliva: comparison with blood and breath alcohol profiles, subjective feelings of intoxication, and diminished performance
14
PubMed Central. Poor correlation between alcohol concentration in oral fluid and breath in subjects consuming beverages immediately before testing
15
Journal of Analytical Toxicology. Synthetic Cannabinoids in Oral Fluid
16
Journal of Analytical Toxicology. Determination of Third-Generation Synthetic Cannabinoids in Oral Fluids
17
PubMed. Drug concentration in saliva
18
PubMed. Perspiration versus saliva–basic aspects concerning their use in roadside drug testing
19
PubMed. An overview of the use of urine, hair, sweat and saliva to detect drug use
20
PubMed Central. Comparison of Urine and Oral Fluid for Workplace Drug Testing
21
Forensic Science International. Oral fluid drug tests: Effects of adulterants and foodstuffs
22
PubMed. Assessment of different mouthwashes on cannabis oral fluid concentrations
23
Forensic Science International. Evaluation of four oral fluid devices (DDS®, Drugtest 5000®, Drugwipe 5+® and RapidSTAT®) for on-site monitoring drugged driving in comparison with UHPLC–MS/MS analysis
24
Forensic Science International. Evaluation of three rapid oral fluid test devices on the screening of multiple drugs of abuse including ketamine
25
Forensic Science International. Oral fluid collection: The neglected variable in oral fluid testing
26
Clinical Chemistry. The New Substance Abuse and Mental Health Services Administration Oral Fluid Cutoffs for Cocaine and Heroin-Related Analytes Applied to an Addiction Medicine Setting
27
PubMed. Comparison of drug concentrations between whole blood and oral fluid
28
Accident Analysis & Prevention. Correlation between oral fluid and blood THC concentration: A systematic review and discussion of policy implications