Effects of Marijuana on Symptoms of OCD: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Effects of Marijuana on Symptoms of OCD (NCT03274440), a randomized phase 1, phase 2 study. The primary outcome is Yale-Brown Obsessive-Compulsive Challenge Scale (YBOC-CS).

2. Clinical Question

The trial’s stated purpose was to test whether components of the marijuana plant, specifically THC and CBD, may reduce OCD symptoms. The record describes these compounds as acting on the brain’s endocannabinoid system, which has been hypothesized to play a role in OCD. That is the rationale recorded for studying the conditions, not evidence of an effect. The study is listed as a pilot, with phases 1 and 2, randomized allocation, triple masking, and a crossover model. The statistical question is therefore about short-term outcome patterns across conditions in this small study, rather than a definitive estimate of longer-term change. The posted results do not include the numerical primary outcome findings needed to assess that question.

3. Trial Design

The registry describes a randomized, triple-masked, crossover trial, listed in phases 1 and 2. It was conducted by the New York State Psychiatric Institute and is marked completed. The recorded start date is 2017-10-01, and the primary completion date is 2019-03-01. Enrollment is listed as 14. These registry entries describe the planned and administrative structure; they do not by themselves establish how many participants contributed usable data to each outcome analysis.

Three conditions are described: high THC/low CBD marijuana, low THC/high CBD marijuana, and placebo cannabis with no THC and no CBD. The record identifies cannabis as a drug intervention and placebo as another intervention type. Because the design is crossover, the conditions were intended to be compared within a crossover framework rather than as three wholly separate parallel groups. However, the sequence in which participants received conditions, the period length, any washout interval, and the handling of order or carryover effects are not reported in the registry record.

Triple masking is recorded, but the identities of the masked roles and the procedures used to preserve masking are not reported in the registry record. Likewise, the method used to generate the random assignment sequence and conceal allocation is not reported. These omissions limit what can be concluded about the details of implementation from the registration alone.

The primary completion date marks the registry’s reported completion milestone, not the date of publication or the timing of any later analysis. The registry status is completed. It lists two outcome measures and six analyses, but does not report how those analyses correspond to conditions, time points, or participant groups. The actual primary outcome estimates and inferential results are not reported in the registry record.

4. Participant Flow and Accounting

The registry reports how participants moved through each stage of the study. All counts below are exactly as posted; missing counts are not inferred from other stages.

Before assignment: Participants who met study criteria and chose to participate were assigned to receive each of the three study conditions (high THC/low CBD; low THC/high CBD; placebo with no THC or CBD) in random order. All sessions were separated by at least 1 calendar week.

Overall Study

Reasons for not completing this period:

5. Baseline Characteristics

The table reproduces the baseline characteristics posted in the registry, which describe the participants analysed at baseline and help show whether the randomized groups were comparable.

Population: Baseline characteristics for all study completers

6. Primary Endpoint

The registry lists the primary outcome as Yale-Brown Obsessive-Compulsive Challenge Scale (YBOC-CS), measured over Baseline was obtained at the beginning of each session. Then, change from baseline was measured following cannabis administration (minute 0) at minutes 20, 40, 60, 90, 120, and 180. This procedure was repeated for each of the 3 conditions..

The primary endpoint is the Yale-Brown Obsessive-Compulsive Challenge Scale (YBOC-CS), reported as a mean score on a scale. It is described as an adapted version of the Yale-Brown Obsessive-Compulsive Scale that measures symptom change over short time frames, since the last assessment, rather than over the past week. The total score ranges from 0, representing no symptoms, to 40, representing severe symptoms.

The registry describes baseline measurement at the beginning of each session. Change from baseline was then measured after cannabis administration at minute 0 and at minutes 20, 40, 60, 90, 120, and 180. The procedure was repeated for each of the three conditions. This repeated, short-interval measurement is relevant to the trial’s stated focus on acute symptom changes after administration. It should not be confused with evidence about longer-term symptom course.

In general, a continuous score can capture graded differences rather than reducing outcome status to a binary category. A mean summarizes the average score in a group or condition, while a standard deviation describes how dispersed individual scores are around that mean. The record names the mean and standard deviation as the outcome summary measures, but does not report the primary outcome values, condition contrasts, or uncertainty estimates. Consequently, the endpoint’s measurement schedule and scale can be described, but the magnitude and precision of the observed changes cannot be assessed from the registry record.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The registry identifies the primary outcome as a continuous measure summarized with a mean and standard deviation. In general, a mean is the arithmetic average of the observed scores, and the standard deviation describes how much individual scores vary around that average. The mean communicates the central level of the outcome; the standard deviation provides information about dispersion. Neither summary alone shows the uncertainty in a comparison between conditions. The registry record does not report the outcome values or the estimated contrasts.

What does a confidence interval communicate?

In general, a confidence interval expresses uncertainty around an estimate under the assumptions of the statistical method used. A narrower interval indicates greater precision than a wider interval, while an interval that includes values on both sides of a null contrast may be compatible with several interpretations. A confidence interval is not a guarantee that the true value lies within its endpoints. No confidence interval for the primary outcome is reported in the registry record, so precision cannot be evaluated for the trial’s condition comparisons.

How does a crossover design affect analysis?

In general, a crossover design assigns participants to receive more than one condition, allowing comparisons to account for within-person measurements. Scores from the same participant tend to be related, so an analysis should account for that dependence rather than treating all observations as independent. Analysis may also need to address treatment sequence, period, and possible carryover, depending on the design and outcome. The registry labels this study a crossover but does not report the condition sequences, period structure, washout procedures, or statistical handling of these features. No specific crossover model can therefore be attributed to the trial from the registry record.

The outcome schedule includes baseline assessment at the beginning of each session and repeated measurements after administration, at the listed minutes, for each condition. Repeated measurement creates multiple observations per participant. In general, analysis methods for such data account for correlation among observations from the same person and may model change over time or compare conditions at defined times. The registry does not state whether the primary analysis used a repeated-measures model, a time-specific contrast, a summary across times, or another method.

Why can over-dispersion matter?

In general, over-dispersion means that observed variation is greater than a model assumes. It can lead to uncertainty estimates that are too small if ignored. This concern is especially associated with outcomes whose variability is modeled through a distributional assumption, such as counts; the primary endpoint is described as a continuous score. The registry record does not report a model or a dispersion assessment, so there is no basis to say whether over-dispersion was considered or relevant to the trial’s analysis.

What can sample-size illustrations tell readers?

In general, a sample-size calculation describes how many participants might be needed under stated assumptions to detect a specified effect with a chosen significance level and power. The planning illustrations associated with this record are described as using an observed effect, a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation. They use a stated fraction of the observed effect as the target effect. For non-time-to-event outcomes, the described quantities are participants per arm for 80% or 90% power; time-to-event calculations, if applicable, would express total events. These are planning illustrations, not observed power and not proof that the trial was adequately powered. The numerical illustration outputs are not reported in the registry record.

Because the illustrations are based on an observed effect, their interpretation depends on the stability of that estimate. In general, effects estimated in a small pilot study may be imprecise, and a planning exercise based on them should not be mistaken for a prospective justification. The registry does not report the trial’s own sample-size justification, target power, or assumptions used to choose enrollment. It is therefore not possible to establish whether the enrollment target was derived from a formal power calculation.

The registry also lists six analyses, but does not describe the statistical tests, model terms, contrast definitions, treatment of missing observations, or adjustments for multiple testing. General statistical principles can explain why these choices matter, but cannot identify which method the investigators used. The posted outcome estimates and an analysis plan are needed to evaluate the methods actually applied.

10. Multiplicity, Missing Data and Analysis Populations

The primary outcome was assessed repeatedly after administration in each of three conditions, and the registry lists a secondary anxiety measure, two outcome measures in total, and six analyses. This creates the possibility of multiple statistical comparisons. In general, testing many contrasts or time points can increase the chance of finding at least one apparently unusual result by chance. Prespecifying a primary contrast and applying an appropriate multiplicity strategy can help readers interpret a collection of tests. The registry record does not report which comparisons were prespecified or whether any adjustment was used.

The term “six analyses” does not reveal how many tests addressed the primary outcome, whether analyses were divided by time point, or whether the count includes secondary outcomes. It should not be used to infer the number of hypotheses or the method of multiplicity control. A detailed statistical analysis plan and reported contrast definitions are not reported in the registry record.

Missing-data procedures are also not reported. The registry does not state how incomplete sessions, missed time points, withdrawals, or unusable outcome measurements were handled. It does not report the number of participants contributing data at each assessment. In general, missing-data assumptions can affect estimates when outcome availability is related to participant characteristics or prior measurements. Without the analysis population, missingness pattern, and method, the extent of this issue cannot be assessed.

The record does not identify whether results were analyzed by randomized assignment, by received condition, or using another analysis population. It also does not report whether all randomized participants were included in the primary analysis. The enrollment figure alone cannot resolve these questions. These omissions limit assessment of both the robustness and the scope of any eventual treatment-condition comparison.

11. Safety Results

The registry reports serious adverse events as affected participants divided by participants at risk.

Deaths

Other (non-serious) adverse events

Participants with at least one non-serious event at or above the registry's 0% reporting threshold.

The registry’s safety entries list three groups: High THC/Low CBD Cannabis, High CBD/Low THC Cannabis, and Placebo Cannabis. Each group lists 12 participants at risk. For every group, the record lists 0 serious adverse events, 0 other adverse events, and 0 deaths. No serious adverse-event terms are listed.

These are the safety counts recorded for the listed groups. They do not establish that no symptoms or events occurred outside the registry’s reported categories, nor do they establish that every participant enrolled contributed to each safety denominator. The overall enrollment is 14, whereas each safety group lists 12 at risk; the record does not explain the difference. Since this is a crossover trial, the group entries should not be added together to infer a count of unique participants.

In general, zero events in a registry entry means that no events were recorded in that entry, not that an event is impossible or that a risk has been ruled out. Small numbers of participants also limit how precisely uncommon events can be characterized. The registry record does not provide additional detail on adverse-event collection methods, severity categories beyond the listed serious and other counts, or exposure duration. These counts should therefore be read as a limited description of the posted safety record, not as a comprehensive estimate of safety.

12. What the Result Does, and Does Not, Mean

The trial record is informative about the question being studied, the crossover structure, the three cannabis conditions, the primary symptom scale, and the timing of assessments. It also records the number enrolled and the safety counts for each listed condition. Those details help define the study’s scope. They do not establish what happened to the primary outcome, because numerical YBOC-CS results and condition comparisons are not reported in the registry record.

The stated scientific rationale is that THC and CBD act on the endocannabinoid system, which has been hypothesized to play a role in OCD. A hypothesis provides a reason to study an intervention, not a result. Without the observed score changes, comparison estimates, and uncertainty measures, the record cannot establish whether one condition had higher or lower scores than another, how large any difference was, or how precisely it was estimated.

The endpoint concerns short-term symptom change following administration at repeated time points. It is not a measure of long-term symptom course, and the record does not describe longer-term follow-up. The trial’s enrollment of 14 and pilot designation are relevant to its scale, but do not alone determine whether an estimate is reliable or whether findings apply beyond the participants studied.

In general, a result’s statistical significance and its clinical interpretation are distinct questions. A p-value does not communicate the full size or precision of an effect, and a difference in a scale score requires context to judge its importance. No outcome estimate or test result is reported, so neither statistical nor clinical importance can be assessed. The record also does not report whether either cannabinoid condition differed from placebo in the primary outcome.

The completed status and recorded safety counts should not be read as evidence of symptom benefit or as proof of absence of risk. They report trial status and recorded event counts, respectively. The available registry information supports a careful account of the design and endpoint, but leaves the central efficacy comparison unresolved.

13. Why This Trial Matters Statistically

This trial is statistically notable for combining a crossover design with repeated short-term measurements. In general, a crossover can make within-person comparisons possible, while repeated assessments can characterize how an outcome changes over time. Both features can be useful for studying short-lived responses. They also make the analysis more dependent on clear specification of condition sequence, time, within-participant correlation, and the contrast of interest. Those details are not reported in the registry record.

The primary measure is a continuous symptom score, recorded repeatedly after each condition. This provides a richer description than a single binary outcome, but it also raises choices about whether to compare conditions at particular times, across a time profile, or using a prespecified summary. The record describes the measurement schedule without identifying which comparison served as the principal inferential test. That distinction matters because different summaries address different scientific questions.

The study is listed as a pilot and enrolled 14 participants. Pilot data can help characterize feasibility and generate estimates for later research, but general statistical reasoning cautions against treating a small study’s effect estimate as definitive. The registry’s sample-size illustrations use the observed effect as a planning basis and are described as planning exercises, not observed power. The trial’s own sample-size justification is not reported.

The record is also a useful reminder that a completed study and a detailed outcome definition do not necessarily provide enough information to interpret the central result. Without endpoint estimates, uncertainty intervals, analysis populations, and multiplicity procedures, readers cannot judge the size, precision, or robustness of the comparisons. The result is not that the trial answers or fails to answer its question; rather, the numerical findings needed to answer it are not reported in the registry record.

14. Limitations

• The posted summary statistics are the only data these results rest on; patient-level data, full protocols and statistical analysis plans are not available.
• Recalculated results are simple, unadjusted calculations from posted summaries. They do not replace the sponsor's prespecified analysis.

The main limitation for interpretation is that numerical results for the primary YBOC-CS outcome are not reported in the registry record. The record gives the scale, summary measure, and assessment schedule, but not the means, standard deviations, condition contrasts, confidence intervals, or p-values. The secondary outcome is named, but its numerical findings are likewise not reported.

Several design and analysis details are also absent. The crossover sequence, period structure, washout procedures, and handling of period or carryover effects are not reported in the registry record. The roles covered by triple masking and the method of randomization are not specified. The statistical model, planned contrasts, and multiplicity strategy are not reported, despite the listed six analyses and repeated assessments across conditions.

Participant flow and outcome availability are unclear. Enrollment is listed as 14, while the safety entries list 12 at risk in each condition. The registry does not explain that difference or report the number analyzed for each endpoint and assessment time. Missing-data methods and analysis populations are not reported, limiting assessment of whether incomplete observations could affect interpretation.

The record does not report the trial’s own sample-size justification or a prospective target effect. Any planning illustrations based on the observed effect should not be confused with observed power or with evidence that the design could reliably detect a meaningful difference. In general, estimates from a pilot trial can be uncertain, and the absence of reported outcome estimates prevents even that uncertainty from being characterized. These limitations concern what the registry record allows readers to evaluate; they do not establish that the unreported procedures or analyses were not conducted.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: Effects of Marijuana on Symptoms of OCD — NCT03274440.
• Registry record last updated 2020-11-18; status last verified 2020-10.
• Results first posted 2020-06-23.
• Data retrieved from ClinicalTrials.gov on 2026-11-09.
• Group sizes, means, event counts and sponsor-reported p-values are reproduced as posted. Effect estimates, confidence intervals and p-values not attributed to the sponsor are independently calculated from the posted summary statistics.

18. Record Summary

The registry describes a completed randomized, triple-masked crossover study of cannabis conditions and placebo for obsessive-compulsive disorder. It lists phases 1 and 2, enrollment of 14, and the New York State Psychiatric Institute as sponsor. The recorded start date is 2017-10-01, and the primary completion date is 2019-03-01.

The primary outcome is mean change on the YBOC-CS, with repeated measurements after administration in each condition. The secondary outcome is the Spielberger State-Trait Anxiety Scale: State Version. The record lists two outcome measures and six analyses. Numerical outcome results and the analysis methods are not reported in the registry record. Safety entries list 12 participants at risk in each of three groups, with 0 serious adverse events, 0 other adverse events, and 0 deaths in each group.