Can Oral Fluid (Saliva) tests detect cannabis impairment?
Updated November 2024
Watch a video version of this post here:
Read the text version of this post below:
As more and more places across the world embrace the legalization of cannabis, the challenge of reliably detecting cannabis impairment has become increasingly important. It’s tempting to use traditional chemical tests for this purpose. After all, that method has worked well for alcohol. However, tests for alcohol impairment —such as breathalyzers and blood tests—are rooted in well-established science that correlates blood alcohol content (BAC) with impairment levels. This is appropriate for alcohol because it is linearly metabolized and because the impairment experienced mirrors the amount of alcohol in the body. However, the same cannot be said for cannabis (or virtually any other drug). Thanks to dozens of rigorous scientific studies, there is now consensus in the scientific community that measuring delta-9 tetrahydrocannabinol (THC) levels in the body cannot determine impairment.
Said another way, measuring THC in the body cannot tell us if someone is currently impaired.
The Complexity of THC in the Human Body
This post includes a very short overview of how THC works in the body. If you’d like a much expanded exploration of this topic, see the following blog post: Human Metabolism of THC
Delta-9 THC, the primary psychoactive compound in cannabis, has long been the focus of traditional chemical drug tests (like saliva and blood tests). While alcohol impairment can be measured by quantifying blood alcohol concentration (BAC), applying the same approach to THC has been proven to be totally inaccurate due to key differences in how these substances behave in the body.
Alcohol is water-soluble and metabolized relatively predictably in a linear fashion, clearing from the bloodstream within hours. In contrast, THC is fat-soluble, meaning it is stored in fatty tissues and released gradually over time, leading to a more variable rate of metabolism and radically extending the window of detection. THC creates its psychoactive effects by binding to CB1 receptors in the brain shortly after consumption, causing the characteristic "high." However, after initial metabolism in the liver, THC is converted into metabolites that linger in the body, but these are not directly related to ongoing impairment. This is why the high from cannabis use lasts 90 minutes to 2 hours on average (though much longer for edible THC).
The biggest challenge in using THC levels as an indicator of impairment is the discrepancy between how long THC and its metabolites remain detectable in the body and the actual duration of psychoactive effects. For example, a person may test positive for THC days or even weeks after last using cannabis, long after any impairment has ceased. This contrasts sharply with alcohol, which is metabolized more quickly, and impairment closely correlates with its presence in the bloodstream.
Bottom line: The key challenge in detecting cannabis impairment is that THC can remain in the body long after its psychoactive effects have worn off, unlike alcohol, where presence of the substance and impairment closely align.
Oral Fluid Tests are Only Able to Determine Prior Use of Cannabis
Oral fluid drug tests, which detect the presence of drugs like THC in saliva, have been proposed as a tool for identifying marijuana-impaired drivers and workers. These tests are often favored for their non-invasive nature and ease of administration. However, a vast number of scientific studies have shown conclusively that quantifying the THC in the body of a marijuana user is not predictive of any experienced impairment. Oral fluid tests are no exception. THC can be detected in saliva for at least 48 hours after consumption, radically longer than the window of impairment. Therefore, these tests should then be limited in scope to determining if someone has recently used marijuana, not if they’re currently impaired.
This means that an oral fluid drug test commonly will produce a positive result in cannabis users, even if the individual is not currently under the influence of cannabis.
The following are just a few of the studies that have investigated this issue and found that THC levels in the body do not correlate to impairment:
ScienceDirect, 2024: This study found no consistent relationship between blood THC concentrations and actual driving impairment in individuals.
AAA Foundation for Traffic Safety, 2017: Research on Washington drivers showed that THC levels do not reliably predict driver impairment, with high THC levels sometimes observed in unimpaired drivers.
NHTSA, 2014: The study reported no significant correlation between THC levels and crash risk, highlighting the variability of impairment even with detectable THC concentrations.
Tennessee District Attorneys General Conference, 2020: This study concluded that THC concentration was not a reliable indicator of impairment when assessing drivers through Drug Recognition Expert evaluations.
Congressional Research Service, 2019: The report emphasized that the presence of THC does not equate to impairment, and current testing methods do not correlate well with driver performance.
NHTSA, 2017: This report to Congress found little evidence to suggest that blood THC levels could reliably determine the level of driving impairment.
NIJ, 2021: The study showed that both field sobriety tests and THC concentrations are poor indicators of marijuana intoxication and impairment.
Clinical Chemistry, 2013: This research highlighted the difficulty of correlating THC blood concentrations with impairment, given the body's variable response to cannabis.
Clinical Chemistry, 2014: The study found that THC levels in blood or oral fluid had little relationship to the degree of cognitive or motor impairment experienced.
PubMed, 2016: The findings suggested that the relationship between blood THC levels and impairment is weak and inconsistent across individuals.
PubMed, 2009: This study demonstrated that THC concentration was not a reliable predictor of impaired driving performance in participants.
ScienceDirect, 2021: The study confirmed a weak correlation between THC levels and impairment, noting that factors such as tolerance significantly impact an individual’s performance.
PubMed, 2022: Research concluded that the variability in individual response to THC made it difficult to correlate blood THC levels with actual impairment, calling into question the validity of THC-based impairment tests.
LWW Journal of Analytical Toxicology, 2014: This study found that THC could be detected in oral fluid for extended periods (up to 8 days) after consumption, with no reliable correlation between THC presence and impairment, highlighting the inadequacy of oral fluid THC tests as indicators of real-time cannabis intoxication.
The body of research on THC and impairment clearly and consistently demonstrates that there is little to no reliable correlation between THC concentrations in the body and actual impairment. These studies show that while THC may be present in blood, oral fluid, or other biological samples, its concentration does not provide a clear or consistent indication of an individual's functional impairment. THC levels can vary significantly depending on factors like metabolism, tolerance, and the time since consumption, which makes it an unreliable predictor of cognitive or motor skills necessary for activities like working or driving. Some studies even report cases of unimpaired individuals with high THC levels, and conversely, impaired individuals with low or undetectable THC concentrations. As a result, relying solely on THC levels for determining impairment is scientifically baseless and flawed.
This finding has significant implications for the use of oral fluid tests as a measure of cannabis impairment. Oral fluid tests, often used for roadside and workplace testing, measure past cannabis consumption but cannot accurately assess the current level of impairment. Since THC may remain detectable in oral fluid long after its psychoactive effects have subsided, such tests risk flagging unimpaired individuals as impaired. These limitations suggest that current biological tests for THC, including oral fluid tests, are not suitable for determining real-time impairment and raise concerns about their fairness and effectiveness in legal and workplace settings. Since this can have massive implications for employment and/or criminal charges, the use of oral fluid tests as impairment tests should be avoided.
This issue has been very thoroughly studied and the results are conclusive. Oral fluid testing for THC is deeply unfair to safe and responsible cannabis users when used as impairment tests. With one study finding that THC is detectable for 8 days following the last administration of cannabis, it’s exceptionally clear. Oral fluid tests are for detecting previous use only.
Accuracy Issues with Certain Oral Fluid THC Tests
In addition to the fundamental scientific issues with using oral fluid tests to assess impairment, there are also significant concerns about the accuracy of the leading devices used by law enforcement that further undermine their reliability. Note: tests and devices used by commercial organizations can have significantly better accuracy rates than those indicated below.
The devices used by law enforcement are notoriously prone to false positives for THC, which can arise from a variety of factors beyond actual cannabis use. Exposure to secondhand marijuana smoke, cross-contamination, certain medications, or even the consumption of specific foods can all trigger a false positive result. This lack of specificity means that even individuals who are not impaired and may not have recently consumed cannabis could be incorrectly flagged as driving under the influence.
Evaluations of popular devices like the Draeger DrugTest 5000 and Abbott SoToxa highlight these flaws:
Draeger DrugTest 5000 evaluation showed a 14.5% false positive rate for THC (and a whopping 65.5% false positive rate for opiates): Study
Abbott SoToxa evaluation showed an 11.2% false positive rate for THC: Study
Abbott SoToxa showed a nearly 24% false positive rate in a Michigan pilot study: Study
Despite these high error rates, these devices are still being actively being used in Canada and several U.S. states, including Minnesota and Indiana, with more states considering their implementation. The widespread use of these flawed tools poses a serious risk to drivers, who could face unjust penalties due to inaccurate test results. Fortunately, in many jurisdictions, drivers still have the right to refuse these tests, a critical safeguard given the devices' unreliability. As more states evaluate the use of oral fluid tests, it is crucial to weigh their technological shortcomings and the potential for false positives against their intended purpose of detecting impairment.
What’s the Gold Standard in Oral Fluid Testing?
Accurate oral fluid tests certainly exist. Those that are analyzed using mass spectrometry are virtual unimpeachable, assuming proper procedures were followed in collection and chain of custody. Gaize often recommends confirmatory oral fluid tests that have the ability to be lab confirmed using mass spectrometry.
Bottom line: do plenty of research on your preferred oral fluid testing system. They’re not all created equal and, in particular, significant accuracy differences may exist. Look for a solution that’s able to be lab confirmed using mass spectrometry.
Why are some Law Enforcement agencies deploying questionable oral fluid tests?
At this point, the limitations of the most common oral fluid testing systems used by law enforcement have been well established. Despite these significant limitations, oral fluid tests are still used in several states for roadside drug testing. The reasons for this are multifaceted. First, saliva tests are relatively easy to administer and can provide quick results on-site. Law enforcement agencies favor them because they can be performed without the need for medical personnel and give a relatively rapid result. Moreover, these tests are seen as a deterrent to driving under the influence, much like alcohol breath tests. Deploying oral fluid tests in this way is, however, deeply unethical, given the many problems we’ve already discussed.
However, their legal use raises serious questions about fairness and accuracy. Given that oral fluid tests do not reliably indicate impairment, individuals who use cannabis legally, whether medicinally or recreationally, may face consequences that do not align with their actual ability to operate vehicles or perform other tasks safely.
As with other chemical tests, saliva tests disproportionately impact marginalized communities. Saliva tests are more likely to be used against people of color and those from low-income communities. This is because these communities are more likely to be targeted by law enforcement. More specifically, over-policed populations will certainly be targeted by these tests. Given their absurdly high false positive rate, these populations will experience an even greater disparity of negative policing outcomes.
Here’s a recent study by the ACLU highlighting the issue of policing of minority populations for cannabis use: https://www.aclu.org/press-releases/new-aclu-report-despite-marijuana-legalization-black-people-still-almost-four-times
Alternatives to Oral Fluid Tests for Law Enforcement
The glaring limitations of oral fluid tests in roadside situations demands a shift towards more accurate, reliable methods of assessing cannabis impairment. Fundamentally, what is needed is a testing technology that fairly and accurately detects cannabis impairment as it’s being experienced, not past use. Impairment detection technology is the right solution because it evaluates actual performance, rather than relying on outdated biological measures like THC concentrations.
Physiological and cognitive testing is at the forefront of this movement, offering real-time, functional assessments of a person’s ability to perform tasks safely. Clinical research is emerging that shows that these tools offer clear and deeply accurate means of assessing a person for current impairment. For example, a 2020 study published in Frontiers in Psychiatry demonstrated the potential of digital cognitive tests, such as those measuring reaction time, attention, and memory, as tools to detect impairment. These tests provide a direct evaluation of current ability, addressing the critical issue of real-time impairment detection rather than relying on indirect markers like THC levels.
In particular, eye tests that include pupil reaction to light and gaze characteristics have long been used by human Drug Recognition Expert police officers to assess non-alcohol impaired driving. These tests have withstood court scrutiny and scientific inquiry for four decades and have emerged as highly predictive of cannabis impairment. Gaize has utilized these law enforcement tests to create a fully automated impairment detection device using cutting edge eye-tracking technology. This tool has the ability to measure subtle changes in eye movements and pupil behavior that is conclusively linked to cannabis impairment.
Gaize offers an objective, science-backed alternative to traditional chemical testing methods. Unlike oral fluid / saliva tests that cannot be correlated with impairment, Gaize focuses on measuring the mental state of the individual, ensuring that cannabis users are evaluated based on their actual ability to function, not the presence of THC in their system. This approach is not only more accurate but also far fairer, ensuring that those who are unimpaired are not unjustly penalized for past use of a legal substance. As we move towards smarter, more effective tools, it’s clear that technologies like Gaize represent the future of cannabis impairment detection, setting a new standard for both safety and fairness.
Conclusion
The scientific consensus is clear: oral fluid tests cannot detect cannabis impairment. While these tests can, in many cases, detect previous cannabis use, they are not a valid indicator of whether a person is impaired at the time of testing. Factors such as individual tolerance, the method of consumption, and the time since consumption all contribute to this disconnect between THC levels in saliva and actual impairment.
In pilot deployments with law enforcement agencies in the US and Canada, oral fluid drug tests have also been shown to produce unacceptably high false positive rates. This means that even drivers who have not previously used cannabis may be arrested for impaired driving simply due to poor quality tests. Further, states doing oral fluid drug testing are aware of these shortcomings and are using these tests anyway. This is an outrageous decision, particularly as more and more damning evidence has been released. As one example, Minnesota is now launching an oral fluid pilot program using the same equipment that has produced terrible results in other states. In fact, that state’s top law enforcement officials have knowingly and improperly characterized oral fluid tests as impairment tests.
Oral fluid tests for THC can be great solution when used in workplace contexts to establish "non-negative" results for THC, and then do a lab confirmation with mass spectrometry. Even a lab verified results though have no bearing on impairment at the time of testing. These tests are then best used in combination with impairment detection technology, like Gaize, to find out if impairment is being currently experienced.
As cannabis legalization continues to expand, the need for more accurate, reliable methods to assess impairment grows. Tools like Gaize’s Impairment Detection Technology hold the key to addressing this challenge. Treating oral fluid tests as impairment tests risks punishing safe and responsible cannabis users who may test positive for THC long after any impairment has subsided. There is absolutely no scientific data to support the use of saliva tests for detecting acute marijuana impairment, and the fact that states are deploying these tests for that purpose is both unfair, and antithetical to aims of the public in legalizing access to marijuana.
Impairment detection technology is deployed today and has been clinically proven to accurately detect cannabis impairment. Visit our Science page or contact us for more information.
