Can CBD-Dominant Cannabis Oil Produce a Positive Urine THC Screen When Plasma THC Is Mostly Unquantifiable? A 43-Person Trial

What is the precise forensic question?

A urine cannabinoid screen often asks whether a THC-associated metabolite crosses a specified decision threshold. A forensic blood result may ask a different question: how much parent THC is measurable in a particular blood specimen at collection. The two tests are not competing readings of one substance. One can be positive while the other is unquantifiable because the targets, specimen compartments, laboratory methods, sampling schedules, and reporting limits differ.

This distinction becomes especially important when a product is described informally as “CBD-rich.” That phrase does not establish that the product contains zero THC. Here the active oil supplied 5.4–21.6 mg THC per day alongside 120–480 mg CBD. Thus, a THC-metabolite screen after repeated administration is biologically intelligible without calling the laboratory result a false positive or assuming THC-free exposure.

How were participants randomized and treated?

The trial randomized 43 distinct healthy adults, aged 18–55 years, into five parallel groups: four active strengths and a placebo. Each participant had reported at least two previous lifetime exposures to THC-containing cannabis, but recent cannabinoid use and several medications were excluded. The controlled residential setting permitted repeated sampling; it did not make the sample representative of chronic patients, professional drivers, or people involved in collisions.

Four groups received Spectrum Yellow oil made with cannabinoid-containing resin in medium-chain triglyceride oil. Its stated concentration was 20 mg/mL CBD and less than 1 mg/mL THC. Nominal total daily treatment arms were 120/5.4, 240/10.8, 360/16.2 and 480/21.6 mg CBD/THC, respectively. The highest daily THC content was therefore not trivial simply because CBD predominated. Placebo oil was formulated without cannabinoids and matched the administration process.

The detailed protocol administered divided doses approximately 12 hours apart after standardized meals on Days 1–6, followed by a final morning dose on Day 7. Thus the repeated-dose design was not one single oil exposure. Postdose collections continued after the final dose, including later outpatient sampling. The authors’ abstract describes dosing over seven days; the detailed timetable clarifies that the final day supplied only the morning dose. A forensic reconstruction should follow the actual timetable, not infer a second Day-7 dose that the Methods do not describe.

Independent people versus pharmacokinetic measurements

The randomized allocation was nine participants in the lowest active group, eight in the next group, nine in each of the two highest active groups, and eight in placebo. The reported pharmacokinetic analysis used 32 active recipients, eight per active arm; the eight placebo participants were not part of that cannabinoid-exposure PK population. The 43 randomized people, 32 PK participants, 45 reported adverse-event episodes, and numerous serial blood draws must not be counted as four competing measures of sample size.

Blood was drawn repeatedly before and after dosing. Tubes were promptly cooled; plasma was separated by centrifugation and stored frozen for analysis. The physical analytical material was therefore plasma, not an aliquot of unseparated legal venous whole blood. Urine specimens were collected on a different schedule. These compartments may have substantially different concentrations and cannot be converted by inventing a universal plasma-to-whole-blood multiplier.

What did the dose groups actually show?

The first table combines the study’s randomized allocation with original Day-7 plasma geometric-mean peak concentrations for CBD and its carboxylated metabolite. Those peaks are analytical plasma concentrations in ng/mL, not doses, not total-body drug amounts, and not direct forensic whole-blood concentrations. The values should not be used to predict an unmeasured person’s peak.

Interpretation: Cmax entries are geometric means from the active-arm PK table. The parent and metabolite are distinct molecular species. The table does not establish a positive urine screen for any named individual and should not be read as a cannabis-related impairment scale.

What laboratory methods analyzed CBD, THC and metabolites?

The investigators used a validated two-dimensional liquid-chromatography/tandem mass-spectrometry procedure, described as LC–LC–MS/MS, to quantify cannabinoids and selected metabolites in both plasma and urine. The plasma assay’s lower limits of quantification included 0.780 ng/mL for THC, CBD and 11-COOH-THC; 3.125 ng/mL for 11-OH-THC; and 1.560 ng/mL for each of 7-OH-CBD and 7-COOH-CBD. These are analyte-specific analytical thresholds, not impairment or legal per-se thresholds.

The numerical urine concentration work was not a creatinine-normalized serial-abstinence algorithm. The urine samples were not hydrolyzed and their reported concentrations were not corrected for creatinine. The parallel immunochemical urine screen, addressed below, is a different method from a compound-specific mass-spectrometric concentration. A report must identify which one generated the result under discussion.

Why did CBD’s carboxylated metabolite dominate the plasma profiles?

CBD undergoes metabolism to 7-OH-CBD and then to 7-COOH-CBD. In this study, 7-COOH-CBD plasma exposure on Day 7 was much greater than exposure to parent CBD: the authors reported ratios of roughly 23.9, 36.8, 45.6 and 44.7 across the four active arms. The estimated accumulation ratio of this metabolite’s AUC was approximately 3.8–5.4 across groups. These are specific results of this dosing and assay schedule, not fixed metabolic ratios for every product or patient.

Day-7 parent CBD exposure was broadly dose proportional within the studied doses: reported AUC0–t slope 1.03, with a 95% interval of 0.70–1.36, and Cmax slope 0.92, interval 0.53–1.31. The width of those intervals should remain visible when describing proportionality. The researchers interpreted predose measurements as consistent with a CBD steady state by Day 7, but that does not mean every metabolite was at the same steady state or that an individual’s actual dose can be read from one concentration.

Most plasma THC readings were below quantification—not proof of no THC exposure

In the lowest active arm, all measured parent-THC plasma concentrations on Days 1 and 7 were below the study’s 0.780-ng/mL quantification limit. In the three higher active arms, the majority were similarly below it. Occasional values were measurable, but no participant had sufficient quantifiable parent-THC time points on a given study day to calculate reliable parent-THC pharmacokinetic parameters. The resulting THC PK table therefore reports many entries as below quantification rather than a defensible THC Cmax or AUC for the group.

Below quantification is not chemical absence. It means the validated procedure did not establish a numerical concentration at or above its quantification threshold in that aliquot. The person had nonetheless been administered measurable nominal THC doses. Conversely, the dosing record does not show that their plasma value would satisfy a different laboratory’s detection threshold, much less a law’s whole-blood threshold.

The carboxylated THC metabolite, 11-COOH-THC, was more readily quantified in plasma and exhibited greater exposure with higher daily THC doses. This contrast supports the concept that parent-drug and metabolite trajectories answer different questions. It does not establish that a metabolite concentration alone quantifies contemporaneous psychomotor impairment.

How long did the urine THC-metabolite screen remain positive?

The study used a separate rapid urine screen targeting 11-COOH-THC, described with a 50-ng/mL screening threshold. The screen is not the same measurement as the plasma LC–LC–MS/MS parent-THC result. Figure 3 tracked the percentage of active-arm participants with positive screens after the final treatment. At least one person in each active group had a positive result through 72 hours; at 144 hours, one participant in group B and two in group D still screened positive.

These are observed results after repeated ingestion of oil that actually contained THC. It would be scientifically misleading to describe them as positive urine tests caused by THC-free CBD alone. It would also be misleading to call every positive a confirmed drug identity or to conclude that an individual was impaired several days after the last dose. The reported observation concerns a THC-metabolite-targeted screen.

Repeated positives across time points were collected from the same study participants. One person remaining positive at 144 hours is not an independent replication by a new person at 72 hours. The reported data do not deliver a validated probability of urine positivity for all CBD-rich oils, all consumer populations, or every laboratory threshold.

What do the most important positive and null findings establish?

Every row has a different denominator and endpoint. The safety count of 45 is a count of events, not 45 patients; the PK sample of 32 is not the population for every subjective-effects analysis; and a urine screen is not a plasma analyte concentration.

Can a positive urine screen distinguish recent administration from residual excretion?

Not on this evidence alone. The study documents positive screens after known repeated administration, but it does not independently validate an individual postcessation decision rule. A particular case would require the original collection chronology, the target analyte and cutoff, the screening method, confirmatory results if any, and the actual chain of custody. The possibility of persistent metabolite excretion cannot specify whether an unobserved additional dose occurred in someone else.

When consecutive urine results are compared in a real case, hydration and creatinine information may also matter. This study did not supply a creatinine-normalized new-use rule, and its non-hydrolyzed quantitative urine assay should not be silently compared with a laboratory method measuring total deconjugated THC-COOH. The exact testing procedure controls the interpretation.

Does an unquantifiable plasma THC result contradict a positive urine metabolite result?

No. A limited parent-drug measurement and a persistent metabolite screen can legitimately coexist. The blood was separated to plasma, the urine followed its own excretion kinetics, and the results employed different analytical techniques and decision limits. Treating them as equivalent would ignore both analytical specificity and pharmacology.

In a disputed toxicology report, the correct questions are whether the sample was whole blood or separated plasma, whether the result targeted THC or THC-COOH, whether the statement concerns a qualitative screen or an identified quantitative compound, and whether the actual values were above the validated method’s reporting level. Negative, detected, and below-quantification are not synonyms without the laboratory’s reporting conventions.

What can the subjective-effect data actually say about intoxication?

Participants reported feelings using an instrument covering liking, bad effects, dizziness-related sensations, tiredness, alertness, restlessness, distraction and other experiences. Across Days 1 and 7, the investigators did not find a consistent active-versus-placebo pattern for most subjective items. Selected pairwise ratings, including “restless” and “distracted,” differed at some doses or visits, but these contrasts did not form a stable dose-dependent impairment classification.

A subjective Drug Effects Questionnaire is not an on-road lane-control test, an eyewitness assessment, or a validated individual fitness-to-drive evaluation. Lack of a consistent subjective difference does not prove that every person’s driving was unimpaired. Likewise, a complaint of feeling distracted cannot by itself establish that the oil caused any particular crash or injury.

What did the adverse-event data add?

Twenty-one of the 43 randomized participants reported at least one treatment-emergent adverse event, totaling 45 events. Forty-one episodes were classified as mild, three as moderate, and one as severe; the severe episode was anxiety in the highest active arm. There were no reported serious adverse events or deaths. The two highest active groups each had six participants report an adverse event, versus three in placebo; the lower arms had four and two, respectively.

Four participants experienced events leading to treatment discontinuation, although three withdrew from the study and one continued follow-up after stopping treatment. Events included symptoms such as dizziness, presyncope, insomnia, abdominal pain and diarrhea. These observations qualify the word “tolerated.” They are important adverse effects in healthy volunteers, but the trial was not designed to decide whether a particular defendant, employee, or patient was safe to drive at an unobserved time.

What limitations affect litigation interpretation?

The trial was single-center, used small parallel dose groups, enrolled screened healthy adults, and involved a particular CBD-dominant formulation under standardized feeding conditions. The investigators acknowledged wide confidence intervals around some exposure parameters. There was no direct observation of driving, no crash endpoint, no validated individual impairment threshold and no comparison against actual forensic venous whole-blood THC specimens from the same events. These limitations matter when applying a controlled trial to a disputed real-world case.

The study was funded by the product’s manufacturer, Canopy Growth Corporation, which also supplied the product and participated in design and interpretation. The original disclosure describes commercial relationships for several authors. Funding and conflicts do not erase the measured findings, but they warrant careful scrutiny of protocol, original tables, analysis choices and independently available validation evidence.

Finally, the ordinary term “CBD oil” encompasses products with very different actual THC content and regulatory controls. This experiment cannot be used to assert that all CBD products trigger urine cannabinoid screens, that THC-free formulations do so, or that a particular commercial oil had the same composition as Spectrum Yellow. Product testing and case-specific specimen analysis would be necessary.

A practical evidence review for forensic scientists and attorneys

Start with the original analytical report: identify the exact target (parent THC, 11-OH-THC, 11-COOH-THC, CBD or a CBD metabolite), the collection matrix and time, the specimen preparation, the assay’s LLOQ or screen cutoff, and whether confirmation occurred. Only then compare the available time course, product composition, possible repeated use and documented clinical or functional observations. A population study does not supply missing observations about an individual.

For foundational distinctions, see the published same-site explanations of biological matrices, pharmacokinetics and elimination and confirmatory drug testing. These contextual guides are not substitute proof for the methods and records from the particular laboratory.

The primary controlled-trial source remains Peters et al., DOI 10.1093/jat/bkab026. Its defensible conclusion is narrower than either “CBD always causes urine THC positives” or “unquantifiable plasma THC means no cannabis-related exposure”: the same controlled CBD-rich, THC-containing regimen can generate largely unquantifiable plasma parent THC and persisting urinary THC-metabolite screen positives. Neither endpoint, alone, establishes person-specific impairment or event causation.