The Effect of Cannabidiol in Learning and Memory of Adults: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for The Effect of Cannabidiol in Learning and Memory of Adults (NCT06074172), a randomized phase 2 study. The primary outcome is Sum of Trials (I-V).

2. Clinical Question

3. Trial Design

The registry identifies this as a phase 2, randomized, triple-masked, crossover study. It was sponsored by Colorado State University and is listed as completed. The recorded start date is 2020-03-14 and the primary completion date is 2020-11-06. Total enrollment is listed as 57.

The crossover arrangement means that the participants described in the primary-outcome population received both study conditions at separate visits. There were two orders: placebo followed by cannabidiol, or cannabidiol followed by placebo. The registry descriptions specify a 1-week washout period between the visits. The reported primary-outcome population consists of 35 healthy adults who attended both visits and received both interventions.

The interventions were 246 mg press pills of cannabidiol and placebo. The registry uses the label triple masking but does not state which specific groups or roles were masked. Further details about sequence generation, concealment of allocation, or how masking was maintained are not reported in the registry record.

A crossover design can allow comparison of conditions within the same participants, which is useful because each person can serve as their own comparator. That general feature does not establish how the analysis handled the paired observations. The statistical model, treatment of period or sequence effects, and approach to possible carryover are not reported in the registry record. The registered study objectives also included interference during learning and possible demographic influences, but the posted primary result addressed the Sum of Trials (I-V) score.

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.

Overall Study

Analysis population for the primary outcome, as defined in the registry: The data reported is for 35 adult healthy human subjects who came in for visits 1 and 2 and were randomized to a treatment group (CBD or placebo). Since the 35 subjects came in twice, each of them received both interventions (CBD and placebo).

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: Enrollment was stopped after interim analyses were completed. 57 subjects were enrolled in the study, but 35 were included in the final analyses. Subjects were required to come for visits 1 and 2. A large dropout between visits 1 and 2 was due to the COVID pandemic which caused for subjects to be lost to follow up. The baseline characteristics are reported for the 35 repeat subjects which is representative of the participant population that results were reported for.

6. Primary Endpoint

The registry lists the primary outcome as Sum of Trials (I-V), measured over Each trial is 45 seconds for encoding and recall.

The primary endpoint, Sum of Trials (I-V), is a continuous score on a scale with a stated range from 0 to 75. Participants listened to a list of 15 words, called List A, and were asked to recall it across five trials, A1 through A5. After each trial, the words were repeated to them. The score for each trial was based on the number of words correctly recalled, and the primary score is the sum across the five trials.

The endpoint is intended to measure learning and memory performance across repeated encoding and recall opportunities. The registry describes each trial as 45 seconds for encoding and recall. Combining performance over the five trials gives a summary of recall across the task rather than relying on a single recall attempt. The registry states that a higher Sum of Trials score indicates greater performance on this memory test and a lower score indicates lower performance.

This is a task-specific measure. It records performance on the described word-list test, not every aspect of learning, memory, or daily functioning. The registry record does not report further validation details, such as how the score relates to other memory measures. The endpoint is therefore best interpreted according to its stated construction and scale, without extending it to outcomes that were not measured in this result.

7. Results Posted in the Registry

• Cannabidiol (Administered During Visits 1 or 2): mean 47.17 versus 49.09 on control; difference -1.92 (95% CI -7.05 to 3.21), standardized effect d = -0.18; z-test p = 0.463.

8. Statistical Interpretation

9. Statistical Methods Explained

The primary outcome is a continuous score, and the registry reports a mean for each treatment condition. A mean summarizes the central value of a set of scores by averaging them. For this endpoint, the score combines correct recalls over five trials, so the reported mean describes average performance on that constructed task score in the respective treatment condition.

The registry reports standard errors of 1.85 for both treatment means. In general, a standard error describes the sampling variability of an estimated statistic, such as a mean. It is not the same as the spread of individual participants’ scores: a standard deviation describes variation among observations, while a standard error concerns the precision of an estimate. The registry does not provide the observed standard deviations for the treatment groups. A pooled standard deviation of 10.94 appears with the comparison statistics, but the analysis method used to derive or apply it is not described in the registry record.

The mean difference is presented as cannabidiol minus placebo. The estimated difference is -1.92, consistent with the posted means. A difference scale keeps the comparison in the original score units, making the estimate interpretable as a change in the average Sum of Trials score. It remains an estimate rather than a known fixed quantity; the interval and p-value convey aspects of its uncertainty under the analysis used.

The reported interval extends from -7.048 to 3.208. In general, a confidence interval is a range constructed by a statistical procedure to express uncertainty about an estimated quantity. Its interpretation depends on the procedure and assumptions used. The confidence level and construction method are not reported in the registry record, so the interval should be described using its posted endpoints without assigning an unreported confidence level.

The registry also reports an unadjusted p-value of 0.463. In general, a p-value evaluates how compatible the observed data are with a null hypothesis under a specified statistical model. It does not measure the size or practical importance of an effect, and it does not establish that the null hypothesis is true. The record does not identify the test statistic, model, degrees of freedom, or handling of the crossover pairing for this comparison. Without those details, the p-value cannot be independently reconstructed from the registry record.

A crossover study has a distinct analysis structure because observations from different treatment periods come from the same participants. In general, accounting for within-participant association can affect estimates of uncertainty compared with treating all observations as unrelated. Visit period and treatment order can also matter in crossover analyses. The registry describes the treatment sequences and washout interval, but does not report whether or how the posted analysis addressed these features. No particular model should be assumed from the design label alone.

The sample-size illustration uses an observed-effect basis and stated planning assumptions. In general, power is the chance that a specified analysis would meet its statistical criterion if a chosen effect were the true effect and the model assumptions held. A planning calculation uses assumptions about the target effect, variability, allocation, and test procedure. It is not an observed property of the completed trial. The original trial’s sample-size justification and design assumptions are not reported in the registry record.

The reported comparison is unadjusted. In general, adjustment for multiple comparisons may be considered when several hypotheses are tested, because repeated testing can increase the chance of at least one apparently notable result under a global null. The record lists five outcome measures and one analysis, but does not describe a multiplicity strategy or identify the full relationship between those outcomes and the analysis. No correction should be presumed.

Missing-data handling and analysis-population definitions are also important parts of statistical interpretation. The primary-outcome description says the data are for 35 healthy adults who attended both visits and received both interventions. It does not report how absent or incomplete measurements were handled, whether any participants were excluded from the analysis, or whether an intention-to-treat or other named population was used. Those details are not reported in the registry record.

These general concepts help explain the kinds of issues relevant to a continuous crossover comparison. They do not establish which specific methods the investigators used. The registry’s reported numerical estimates should be distinguished from methodological details that the record does not state.

10. Confidence Intervals and Precision

The estimated mean difference for cannabidiol versus placebo was -1.92 points on the Sum of Trials scale. The reported interval extends from -7.048 to 3.208. The interval includes zero and values on either side of zero, so the posted estimate does not pin down the direction of the difference. Its endpoints also indicate that the data are compatible with a range of mean differences rather than only values close to the point estimate.

In general, an interval expresses uncertainty around an estimate in a way that a point estimate alone cannot. A narrower interval would indicate greater precision under the same method and assumptions; a wider interval indicates less precision. The interval is not especially decisive about whether the mean score is lower or higher under cannabidiol because it crosses zero and extends in both directions. The confidence level and construction method are not reported in the registry record, so no particular coverage interpretation can be assigned from the posted information.

The unadjusted p-value was 0.463. It is consistent with the interval’s inclusion of zero, but neither statistic should be used alone to make a broader claim about learning or memory. The p-value does not quantify the probability that the treatments are equivalent, and the interval does not show the range of individual participant responses. Both concern uncertainty in an estimated group-level comparison under an analysis procedure.

The reported standard errors for the two means are 1.85 in each condition. The treatment means and interval should be interpreted in the context of the crossover design and the analysis details that are not reported. In particular, the record does not describe how within-participant dependence was represented in the interval calculation. That limits how fully its precision can be evaluated from the registry record alone.

11. Planning a Similar Trial: Sample Size

This is a planning illustration, not a re-evaluation of the original trial. It asks: if a new study wanted to detect a difference like the one observed, or a smaller one, how large would it need to be? It is deliberately not “observed power” of the finished trial, which is not a meaningful quantity.

The planning figures are participants per arm, not total events. They are planning illustrations based on the observed effect, rather than the trial’s original design assumptions. The listed assumptions are a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. The effect used for each illustration is the stated fraction of the observed effect. The trial’s own sample-size justification and original design assumptions are not reported in the registry record.

The 80% power column gives the number of participants per arm that the planning calculation indicates for an 80% chance of meeting the specified statistical criterion if the assumed effect and model are correct. The 90% power column gives the corresponding planning count for a 90% chance. These are conditional planning quantities, not guarantees about a future study’s results.

The rows use 100%, 75%, and 50% of the observed effect. For planning for 100% of the observed effect, the figures are 511 participants per arm for 80% power and 683 participants per arm for 90% power. For planning for 75% of the observed effect, they are 907 and 1,214 participants per arm, respectively. For planning for 50% of the observed effect, they are 2,041 and 2,732 participants per arm, respectively. Smaller assumed effects require more participants to reach the same target power, all else held to the listed assumptions.

The sample-size curve illustrates how the planning requirement changes with the effect assumption; it should be read as a planning aid, not as a description of the completed trial’s achieved power. This is not observed power. The calculations inherit uncertainty from using an effect estimated in a small reported primary-outcome population, and the registry does not report the original trial design calculation.

12. How the Sample Size Is Derived

Power is the probability that a study of a given size rejects the null hypothesis when a specific true effect exists. It depends on four things: the effect to detect, the variability of the outcome, the significance level (here a two-sided α of 0.05, so z_1−α/2 = 1.960) and the desired power (z_1−β = 0.842 for 80% and 1.282 for 90%).

Formula

n is the number of participants in each arm, Δ is the difference in means to detect and σ is the common standard deviation.

Worked calculation for this trial

1. Observed difference Δ = -1.920; pooled standard deviation σ = 10.945.
2. n = 2 × (1.96 + 0.842)² × 10.945² / -1.920² = 510.1, rounded up to 511 participants per arm for 80% power.

The smaller-effect rows in the table above repeat the same calculation with the effect multiplied by 0.75 and 0.5 (for ratios, applied to the logarithm of the ratio).

R code

alpha <- 0.05
z_a   <- qnorm(1 - alpha / 2)          # two-sided
delta <- -1.92                      # observed difference in means
sigma <- 10.9447475987343                     # pooled standard deviation
n_per_arm <- function(power, frac = 1) {
  ceiling(round(2 * (z_a + qnorm(power))^2 * sigma^2 / (delta * frac)^2, 6))
}
for (f in c(1, 0.75, 0.5))
  cat(sprintf("%3.0f%% of observed effect: n per arm = %s (80%% power), %s (90%% power)\n", 100 * f,
              format(n_per_arm(0.80, f), big.mark = ","), format(n_per_arm(0.90, f), big.mark = ",")))

# Check against R's built-in function (t-based, so very slightly larger)
cat("power.t.test:", round(power.t.test(delta = delta, sd = sigma, power = 0.80)$n, 1), "\n")

# Simulation: exact two-sample t-test using simulated means and variances
set.seed(6074172)
n    <- n_per_arm(0.80)
reps <- 20000
m_diff <- rnorm(reps, delta, sigma * sqrt(2 / n))             # difference in sample means
s2     <- sigma^2 * rchisq(reps, 2 * (n - 1)) / (2 * (n - 1))   # pooled sample variance
tstat  <- m_diff / sqrt(s2 * 2 / n)
cat(sprintf("Simulated power at n = %s per arm: %.3f (%d replicates)\n", format(n, big.mark = ","), mean(abs(tstat) > qt(0.975, 2 * (n - 1))), reps))

Output from running the code

100% of observed effect: n per arm = 511 (80% power), 683 (90% power)
 75% of observed effect: n per arm = 907 (80% power), 1,214 (90% power)
 50% of observed effect: n per arm = 2,041 (80% power), 2,732 (90% power)
power.t.test: 511.1 
Simulated power at n = 511 per arm: 0.799 (20000 replicates)

The simulation draws many sets of group means and variances at the planned sample size and counts how often the two-sample t-test rejects the null. It should land close to the target power; at very small required sample sizes it can fall below it, because the normal approximation is rough there.

13. Multiplicity, Missing Data and Analysis Populations

The registry lists five outcome measures and one analysis. It reports the Sum of Trials (I-V) as the primary outcome, while the number of secondary outcomes posted is zero. The registry’s study objectives also mention interference during learning and demographic factors, but the record does not post secondary results for those objectives. Details for the other listed outcome measures are not reported in the registry record.

The reported p-value is unadjusted. The registry record does not state whether a multiplicity plan was specified, which outcomes or contrasts were included in any testing family, or whether any adjustment was used elsewhere. In general, multiplicity matters when several statistical tests are interpreted together, because repeated testing can raise the chance of observing a nominally unusual result even if no underlying effects exist. Without a stated plan, the role of the unadjusted p-value in a broader testing strategy cannot be assessed.

The record does not report the statistical analysis plan, the particular model or test for the primary comparison, or whether treatment order and visit period were included. It also does not describe the handling of within-participant dependence despite the crossover design. These omissions make it difficult to evaluate the analysis against the design from the registry record alone.

The primary-outcome description identifies 35 healthy adults who attended both visits and received both interventions. Total enrollment is listed as 57. Reasons for the difference between those descriptions are not reported in the registry record. The record does not say whether all randomized participants were included in any analysis, how missing or incomplete outcome data were handled, or whether participants with partial data contributed to any results. Named analysis populations, such as intention-to-treat or per-protocol populations, are not reported.

These are reporting limitations, not evidence that a particular analysis or data-handling approach was or was not used. The posted result can be described, while the unreported choices remain uncertain.

14. 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.

Most frequent non-serious adverse event terms

The serious adverse event figures are reported separately for the two treatment conditions. For placebo, 0 serious adverse events were reported among 35 participants at risk. For cannabidiol, 0 serious adverse events were reported among 35 participants at risk. The record also reports 0 deaths in each condition.

These counts describe the events recorded in the registry, not a complete characterization of safety. A zero count means no serious adverse event was reported for the listed group in this record; it does not demonstrate that an event is impossible or rule out uncommon events. The primary-outcome population is described as crossover participants who received both interventions, so the treatment-group counts should not be read as counts of separate treatment-only cohorts.

The registry reports other adverse events affecting 1 participant under placebo and 4 under cannabidiol, with 35 at risk in each condition. Those are reported separately from serious adverse events and should not be conflated with them. Further details about event types, timing, severity for non-serious events, or relatedness are not reported in the registry record. These figures are descriptive and do not establish a causal difference between conditions.

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

The posted result is an estimate of the average difference in a particular word-list learning and recall score under cannabidiol and placebo. The cannabidiol mean was 47.17, compared with 49.09 under placebo. The estimated mean difference was -1.92, and the reported interval ranged from -7.048 to 3.208. The unadjusted p-value was 0.463. The interval crosses zero, so the result does not precisely establish whether the average score was lower or higher under cannabidiol in this comparison.

This result is about the Sum of Trials (I-V) score, which combines correct recalls across five trials. It does not directly measure every dimension of memory, learning, or functioning. Nor does a group mean describe the response of each participant. The registry’s stated scale ranges from 0 to 75, and higher values indicate greater performance on this task.

The result does not establish that cannabidiol improves or worsens memory, and it does not establish equivalence between cannabidiol and placebo. A lack of a clearly determined direction in this estimate is not proof that the conditions have identical effects. The uncertainty in the interval, the reported sample of 35 for this outcome, and the lack of analysis detail all matter when interpreting the comparison.

The trial’s stated objectives extend beyond this primary score to interference during learning and possible demographic influences. Those questions are not answered by the posted primary comparison. The registry record does not post secondary outcome results, and details about analyses of those objectives are not reported.

Accordingly, the most defensible interpretation is narrow: the record reports a modest negative point estimate for the mean score difference, alongside an interval that includes differences in both directions. The evidence shown by this comparison does not resolve the direction or size of an average difference with high precision.

16. Why This Trial Matters Statistically

This trial illustrates why an estimated effect should be read together with its uncertainty. The mean difference of -1.92 is a single point on the score scale, while the interval from -7.048 to 3.208 shows that the estimate is compatible with a range of values. A reader who focuses only on the point estimate could overstate certainty about direction; a reader who focuses only on the p-value could miss the size and span of plausible differences.

The crossover design is also statistically important. Participants received both cannabidiol and placebo in different orders, with a 1-week washout period. In general, this design can make within-person comparisons informative, but the analysis should reflect the repeated observations and may need to consider period or sequence features. The registry describes the design but not how those aspects entered the reported analysis. Distinguishing design from analysis is essential when judging how well the inference matches the data structure.

The endpoint gives a concrete example of how a clinical or behavioral question becomes a statistical outcome. The score is constructed from correct recalls over five trials and has a stated range from 0 to 75. The resulting comparison is specific to this task and should not be generalized automatically to other measures of memory.

Finally, the sample-size illustration demonstrates the sensitivity of planning to the assumed effect. Smaller assumed fractions of the observed effect correspond to larger planned numbers per arm. Since the effect itself is estimated with uncertainty and the trial’s original design assumptions are not reported, those figures are best treated as conditional planning illustrations. Together, these points show how design, outcome definition, effect estimation, precision, and reporting detail all shape statistical interpretation.

17. 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.

Several limitations specific to this trial affect interpretation. The primary-outcome description reports data for 35 healthy adults who attended both visits and received both interventions, whereas total enrollment is listed as 57. The reasons for this difference and the handling of participants without complete primary-outcome data are not reported in the registry record. This limits assessment of how the reported outcome population relates to all enrolled participants.

The crossover design means each participant experienced both conditions, but the analysis details are not reported. In particular, the record does not state how the analysis handled pairing, treatment order, visit effects, or possible carryover. A 1-week washout is described, but the registry does not provide information that would establish how any residual effects were evaluated.

The primary endpoint is a task-specific score based on repeated recall of one word list. It may be informative about performance on that task, but the result alone cannot characterize all forms of learning and memory. The stated objectives also included interference and demographic factors, while no secondary outcomes are posted in the record.

The registry does not report the original sample-size rationale, the full statistical analysis plan, missing-data procedures, or a multiplicity strategy. The reported interval’s confidence level and method are also not stated. These omissions do not change the posted numbers, but they constrain independent evaluation of the result and its precision. The sample-size planning illustration uses the observed effect and should not be confused with the original study design.

18. Related Tutorials

19. Related Statistical Calculators

20. Sources

• ClinicalTrials.gov: The Effect of Cannabidiol in Learning and Memory of Adults — NCT06074172.
• Registry record last updated 2025-03-24; status last verified 2025-03.
• Results first posted 2025-03-24.
• 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.

21. Record Summary

The registry record describes a completed, randomized, triple-masked phase 2 crossover study sponsored by Colorado State University. It lists enrollment of 57, with a start date of 2020-03-14 and primary completion date of 2020-11-06. Participants received a 246 mg cannabidiol press pill and a 246 mg placebo press pill at separate visits, with a 1-week washout period and treatment order varying by sequence.

The primary outcome was the Sum of Trials (I-V) score, a continuous measure of correct recall across five trials. The primary-outcome description reports 35 healthy adults who attended both visits and received both conditions. The mean was 49.09 under placebo and 47.17 under cannabidiol; the estimated mean difference was -1.92, with a reported interval from -7.048 to 3.208 and an unadjusted p-value of 0.463.

The registry reports 0 serious adverse events and 0 deaths in each condition. Important analysis details, including missing-data handling and the method for accounting for the crossover structure, are not reported in the registry record.