As legalization of medicinal and recreational cannabis continues to expand throughout the U.S and worldwide, so too does the need for an objective means of determining recent cannabis use and impairment, which cannot be established using currently available breath-based or blood-based testing methods. To address this critical unmet need for the benefit of law enforcement, employers, and others for whom determining recent cannabis use and impairment is a necessity, the goal of our study was to devise a comprehensive test based on pharmacological changes in Δ9-THC and other cannabinoids that occur with time in exhaled breath and blood, a test that improves the accuracy of breath-based testing by incorporating a confirmatory blood test to prevent false positive results. It was hypothesized that a two-sample testing strategy (collecting two samples separated by a known time interval) could be used to detect cannabinoids in breath during their distribution phases, which occur only during the first few hours after smoking. A confirmatory blood test could be used to compare relative levels of cannabinoids in breath. Higher relative levels of cannabinoids in breath compared to blood would be indicative of very recent use because it is not possible for higher relative levels of cannabinoids to redistribute back into the breath from the blood. Coupled with physical and subject self-assessments of impairment, it was believed that the results of the test could also be tied to recent use within the impairment window.
In the 44 subjects from whom both blood and breath samples were collected, all 44 (100%) tested positive for recent use within the first hour after smoking, which is the time of peak impairment, using the two-point breath and one-point blood test. Pre-smoking, 0% (0/34) of subjects tested positive for recent use, indicating no false positive test results. A positive test result using the two-point breath and one-point blood test indicates that a subject has used cannabis recently through inhalation, i.e., smoking or vaping, and that they are within the three-hour impairment window. Interestingly, approximately 68% of these subjects had detectable levels of Δ9-THC in their breath at baseline prior to smoking, in agreement with recent reports by Lynch et al.  and Olla et al. . This finding suggests that the mere presence of Δ9-THC in breath does not conclusively demonstrate recent use within the impairment window, which could prove to be a major short-coming of the commercial cannabis breathalyzers currently in development.
A potential limitation of the recent use breath test is the possibility for false negative test results. As already pointed out, evidence of recent cannabis use in breath declines rapidly beyond the first hour after smoking, so it is possible that breath samples collected during the second or third hour after cannabis inhalation may no longer exhibit sufficient evidence of recent use, even though the subject may still be impaired. A false negative result obviously favors the test subject, while a false positive result could have potentially serious consequences, including wrongful termination of employment and prosecution. With this test, it is possible to observe short Δ9-THC half-lives in breath and breath/blood Δ9-THC ratios ≥ 2 when concentrations of Δ9-THC are very low, at or near assay limits of detection. In order to avoid false positive test results due to assay variation at very low levels, it is important to ensure that concentrations fall within the validated range of the assay. It may be practical to set a 1.0-ng/mL Δ9-THC cutoff level to avoid false positives due to concen-trations near detection limits.
In our study, a wide variation was observed in pre-smoking Δ9-THC blood concentrations, which ranged from undetectable to as high as 80 ng/mL, in the absence of impairment, which may be attributable to tolerance and variable usage patterns. After smoking to the desired effect, however, all subjects became impaired, which was correlated with nystagmus. The described two-point breath test was able to accurately detect recent use within the impairment window in subjects with a background of frequent cannabis use. Because of interindividual variation in tolerance, frequency of use, and metabolism among cannabis users, it is possible to observe impairment beyond three hours post-smoking.
As we observed in our study, evidence of recent cannabis use in breath dissipates rapidly. It is thus critical that breath samples are collected as soon as possible following a workplace incident, for example, or once a need for testing arises, in order to prevent the loss of breath evidence. On-site breath collection using a device such as the one employed in this study can easily be performed without specialized training. Likewise, the blood sample can be collected on site using one of the commercially available capillary blood draw devices designed to be used without specialized training. For detecting recent cannabis use by inhalation, a two-point breath test is the most practical application. In the event that a test subject is positive for only a short D9-THC half-life in breath, the matching blood sample can optionally be used to confirm recent use if desired. A matching blood sample would always be collected to allow testing for other drugs that can induce impairment, including prescription drugs, and it can be additionally analyzed for cannabinoid content if needed.
While the focus of this test has been on detection of recent inhaled cannabis use, it should be emphasized that the two-point breath and one-point blood test is not limited to just cannabis. The same testing strategy we employed for cannabis in the present study may also prove to be useful for detecting recent use of other impairing drugs such as methamphetamine, phencyclidine, and cocaine that can be administered through vaporization. Exhaled breath testing has already been proven useful for detecting multiple drug types [15, 16], and this test allows the simultaneous testing for cannabis as well as other potentially impairing drugs in both breath and blood. Potential applications include sports medicine, enforcement of workplace drug policy, and law enforcement.
Another limitation of the test as currently designed is that it can detect recent use of cannabis only through inhalation within the impairment window. Further study is needed to detect recent use and impairment following oral consumption of cannabis products. Hypothetically, a similar strategy utilizing two blood samples could be deployed for detecting recent use of orally administered cannabis as well as other orally administered impairing drugs. It is well known that cannabinoid pharmacokinetics differ depending on the route of administration. Because D9-THC metabolites cannot be detected in breath, the blood may contain critical information pertaining to recent use, including concentrations of glucuronide metabolites and changes in the ratios of the D9-THC metabolites such as 11-hydroxy-D9-THC and 11-nor-9-carboxy-D9-THC to D9-THC and to each other, as previously reported [14, 17].
In conclusion, a new test for recent cannabis use and impairment, based on two-point breath sampling with or without a one-point confirmatory blood test, has been developed that can accurately detect whether a subject has used cannabis through inhalation within the three-hour impairment window, regardless of the potency of the cannabis strain smoked, with no false positive results. The test is based on multiple parameters, including cannabinoid half-lives, which confirm distribution phase kinetics, the presence of key cannabinoids that are observed only after smoking, and a blood test to determine the breath/blood D9-THC ratio, which confirms whether the test subject was within the impairment window post-smoking. It is our belief, based on our research, that this test provides an answer to the unmet need of determining recent cannabis use, which in turn protects the public from a safety standpoint by detecting inappropriate use, e.g., DUI, while at the same time protecting responsible cannabis users such as medicinal users from wrongful termination and prosecution. This test may finally help bring an end to cannabis discrimination.