Cannabis Edibles Packaging Imagery Experiment: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Cannabis Edibles Packaging Imagery Experiment (NCT06358144), a randomized na study. The primary outcome is Product Appeal Scoring.

2. Clinical Question

3. Trial Design

The Cannabis Edibles Packaging Imagery Experiment is a randomized, single-masked, parallel behavioral study. Participants were assigned to packaging imagery conditions or to a no-imagery control. The registry lists five arms: flavor imagery, nature imagery, medical imagery, animal imagery, and no imagery. The four imagery arms displayed packages with a theme-specific image; the control packages had no added imagery. Each arm used four product categories: gummies, chocolate, mints, and cookies.

The record reports enrollment of 1260 and status as completed. The listed study start was 2024-05-23 and primary completion was 2024-06-19. The timeframe for the primary outcome was after exposure to product packaging, up to 1 min. The record does not identify which party was masked, describe the randomization procedure, or give further details about how participants encountered or rated the packages.

The primary measure was a mean score for product appeal. Secondary outcomes listed in the record include appeal to children and relative harm perceptions. The record also lists 11 outcome measures, but does not report a full analysis plan, a detailed account of all outcome definitions, or the original sample-size justification. The reported arm results are descriptive means with standard errors. Comparisons with the control and their confidence intervals are available for the primary outcome, while the registry does not report a complete set of analyses for every listed outcome.

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: Approximately 50% of the participants reported using waterpipe within the past year; the remaining 50% of the sample consisted of not-past-year users and never users.

Overall Study

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.

6. Primary Endpoint

The registry lists the primary outcome as Product Appeal Scoring, measured over After exposure to product packaging - up to 1 min.

The primary endpoint, Product Appeal Scoring, measures a participant’s response to the question, “How appealing is this product to you?” The response scale runs from 0, “Not at all appealing,” to 10, “Very appealing.” The registry reports the mean score across products within each condition, with a standard error. Ratings were collected after exposure to the product packaging for up to 1 min.

This endpoint captures perceived appeal rather than purchasing behavior, actual product use, or a health outcome. As a continuous measure, it preserves differences across the rating scale instead of reducing responses to a yes-or-no category. General statistical context: comparing group means can describe whether ratings tended to be higher or lower under different packaging conditions, while uncertainty estimates help show how precisely those differences have been measured.

The endpoint is relevant to the experiment’s stated question because the study changes package imagery and then asks participants to rate product appeal. It does not by itself establish why a package received a particular rating, whether the rating would persist beyond brief exposure, or whether it would predict real-world choices. The registry record does not report validation information for this single-item measure or a threshold for interpreting a particular score as meaningfully different.

7. Results Posted in the Registry

• Flavor Imagery: mean 6.11 versus 5.54 on control; difference 0.57 (95% CI 0.08 to 1.06), standardized effect d = 0.21; z-test p = 0.021.
• Nature Imagery: mean 5.73 versus 5.54 on control; difference 0.19 (95% CI -0.30 to 0.68), standardized effect d = 0.07; z-test p = 0.443.
• Medical Imagery: mean 5.70 versus 5.54 on control; difference 0.16 (95% CI -0.34 to 0.66), standardized effect d = 0.06; z-test p = 0.530.
• Animal Imagery: mean 5.75 versus 5.54 on control; difference 0.21 (95% CI -0.28 to 0.70), standardized effect d = 0.07; z-test p = 0.396.

8. Secondary Efficacy Endpoints

The registry also posts results for secondary outcomes. For each outcome below, the endpoint type, the test applied to the posted summary statistics, the result and its interpretation are given in turn. Secondary endpoints are usually not adjusted for multiple testing, so they support rather than confirm the primary finding.

Appeal to Children Scoring

• Endpoint type: Continuous: a mean value per participant, compared as a difference in means.
• Test: Unadjusted two-sample z-test for the difference in means, using the posted standard errors.
• Result:
  • Flavor Imagery: mean 5.62 versus 4.73 on control; difference 0.89 (95% CI 0.43 to 1.35), standardized effect d = 0.34; z-test p = <0.001.
  • Nature Imagery: mean 4.94 versus 4.73 on control; difference 0.21 (95% CI -0.26 to 0.68), standardized effect d = 0.08; z-test p = 0.382.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the mean is statistically higher for Flavor Imagery than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Nature Imagery and the control. The interval gives the range of effects that remain compatible with the data.

The secondary outcomes listed by the registry concern appeal to children and relative harm perceptions. The posted appeal-to-children means vary across the imagery conditions, with the flavor and animal imagery groups having higher means than the no-imagery group and the nature imagery mean closer to control; the medical imagery mean is lower than control. The registry includes unadjusted comparison details for flavor and nature imagery for this outcome. Results for the listed relative harm-perception outcome are not reported in the registry record.

In broad terms, the secondary pattern suggests that packaging themes may prompt different judgments depending on what is being rated. That is distinct from the primary endpoint, which asks about general product appeal. The secondary measures do not simply repeat the primary result: ratings about appeal to children or perceived harm address different judgments and should not be treated as interchangeable with general appeal.

Secondary findings carry less weight for confirmatory conclusions when the registry does not report a full analysis plan, a complete set of outcome results, or adjustment for multiple comparisons. Multiple measures and contrasts create more opportunities for chance findings. Unadjusted results can still be informative as descriptive or exploratory evidence, but they do not provide the same basis for a single confirmatory claim as a prespecified, fully reported analysis would.

The available record does not state which secondary outcomes were designated in advance as confirmatory, how the several outcomes were prioritized, or whether a multiplicity strategy was planned. Consequently, the secondary pattern can help frame questions for further study but cannot establish that one packaging theme consistently affects all perceptions. It also cannot establish actual effects on children or on health; these are ratings, and the registry does not report behavioral or clinical outcomes.

Secondary outcomes registered for this trial, for reference (only those above were recalculated):

• Appeal to Children Scoring
• Harm Perceptions (Relative) Scoring

9. Statistical Interpretation

10. Statistical Methods Explained

The registry describes randomized allocation and a parallel design. In a parallel design, participants are assigned to a condition and the condition groups are compared. Randomization is intended, in general, to reduce systematic differences between groups at assignment; the registry record does not explain the randomization procedure or report checks of baseline balance. It describes the study as single-masked but does not identify who was masked or how masking was maintained.

What do the means represent?

The primary outcome is a continuous score, and the reported group value is the mean. A mean is the arithmetic average of observed scores within a group; it gives a compact summary of the group’s ratings but does not describe every participant’s response. The record gives the number of participants contributing to each arm result and reports a standard error beside each mean. A standard error describes uncertainty in an estimated mean, rather than the spread of individual ratings. General statistical context: a smaller standard error indicates a more precisely estimated mean, while the standard deviation describes variation among individual observations.

The primary comparisons are mean differences between each imagery condition and the no-imagery control. A positive reported difference means the imagery-arm mean was higher than the control mean; a negative difference would mean it was lower. For example, the flavor-imagery mean difference is 0.57, with the comparison oriented as imagery versus control. These are unadjusted comparisons, not adjusted estimates from a model described in the registry record. The record does not report covariate adjustment or a detailed statistical analysis plan.

How should the confidence intervals be read?

A confidence interval gives a range of estimates compatible with the observed data under the method’s assumptions. For a mean difference, an interval including zero is compatible with no mean difference as well as with differences in either direction. An interval that does not include zero is evidence against a zero difference under the stated interval method, but it does not by itself establish practical importance or resolve issues such as multiple comparisons. The intervals posted for the imagery-versus-control comparisons should be read alongside the corresponding mean differences and unadjusted p-values.

A p-value describes how unusual data at least as incompatible with a specified null hypothesis would be under that hypothesis and the analysis assumptions. It is not the probability that the null hypothesis is true, nor does it measure the size or importance of a difference. The registry gives unadjusted p-values for the primary comparisons. Because the record does not describe a multiplicity adjustment, those values do not account for the fact that several imagery conditions are compared with the control.

Why can over-dispersion matter?

General statistical context: over-dispersion occurs when observed variability is greater than a model assumes. If a model understates variability, standard errors can be too small and confidence intervals too narrow, making evidence appear more precise than it is. The primary endpoint is a continuous rating, and the posted comparisons include pooled standard deviations and standardized effect estimates. The registry record does not identify a model that assumes a particular dispersion structure or report an over-dispersion assessment, so no claim about over-dispersion in this experiment is warranted.

The comparison details include pooled standard deviations and standardized effect estimates alongside the mean differences. A standardized effect expresses a mean difference relative to a measure of variability, which can assist comparisons across scales. It does not replace the original-scale difference: readers can interpret the mean difference directly in points on the 0-to-10 appeal scale. The reported standardized estimates are 0.206 for flavor imagery, 0.069 for nature imagery, 0.056 for medical imagery, and 0.074 for animal imagery.

The registry provides results for all five primary-outcome arms and unadjusted comparisons against the control. The record does not report analysis-population definitions, handling of missing scores, a missing-data method, or whether all randomized participants contributed to each outcome. These omissions limit assessment of how the analyzed observations relate to the randomized groups. The statistical description should therefore remain tied to the posted summaries rather than assume an intention-to-treat or other analysis population.

11. Confidence Intervals and Precision

Precision is reflected in the uncertainty around an estimate, not just in the estimate’s point value. For the primary outcome, the posted mean differences are accompanied by confidence intervals. Flavor imagery has a difference of 0.57, with a 95% confidence interval from 0.085 to 1.055. The interval lies above zero, but it spans values ranging from a small positive difference to a larger positive difference on the appeal scale.

For nature imagery, the mean difference is 0.19 and the 95% confidence interval is -0.295 to 0.675. For medical imagery, the difference is 0.16 and the interval is -0.340 to 0.660. For animal imagery, the difference is 0.21 and the interval is -0.275 to 0.695. Each of these intervals includes zero and also permits differences in either direction. Their widths indicate uncertainty about the size and direction of the corresponding mean differences.

Confidence intervals do not indicate the range of individual participant ratings. They describe uncertainty about a group contrast under the statistical method. The registry does not report confidence intervals for the arm means themselves, nor does it specify a minimum difference that would be practically important. Thus, precision can be discussed in terms of the ranges around the estimated differences, but not as proof that an effect is meaningful or unimportant.

The comparisons are unadjusted, and the record does not report a multiplicity procedure. That matters when considering the confidence intervals and p-values collectively: the intervals describe each contrast, while the record does not indicate an adjustment for examining multiple contrasts. Precision also cannot resolve limitations in the population description, missing-data reporting, or generalizability beyond brief package exposure.

12. 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 sample-size figures are planning illustrations expressed as participants per arm, not total events. They use a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. For each comparison, the effect to detect is set to the stated fraction of the observed effect. The registry’s original design assumptions and sample-size justification are not reported in the registry record. These calculations are not observed power.

The 80% power column gives the illustrative number of participants per arm under which the planning method would have an 80% chance of detecting the specified effect if that effect and the other assumptions held. The 90% power column uses a higher probability of detection under the same general planning framework. Power is a design property calculated for an assumed effect and analysis; it is not a measure of how likely a reported result is to be true.

Rows planning for 100%, 75%, or 50% of the observed effect show how the illustrative participant requirement changes when the target effect is set to different fractions of the observed estimate. A smaller target effect is harder to distinguish from random variation, so it requires more participants per arm under these assumptions. Estimates based on the observed effects should be treated cautiously: an observed effect from a completed study is itself uncertain, and the registry record does not establish that the same effect would recur in another study.

The table and curve are therefore useful for understanding the relationship between assumed effect size and planning requirements, not for reconstructing the trial’s original design or asserting its achieved power. The registry does not report the assumptions used when the original study was planned.

13. 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 = 0.570; pooled standard deviation σ = 2.760.
2. n = 2 × (1.96 + 0.842)² × 2.760² / 0.570² = 368.0, rounded up to 369 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 <- 0.57                      # observed difference in means
sigma <- 2.75996457141824                     # 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(6358144)
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 = 369 (80% power), 493 (90% power)
 75% of observed effect: n per arm = 655 (80% power), 876 (90% power)
 50% of observed effect: n per arm = 1,473 (80% power), 1,971 (90% power)
power.t.test: 369 
Simulated power at n = 369 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.

14. Multiplicity, Missing Data and Analysis Populations

The primary results compare each of four imagery themes with a shared no-imagery control. The registry reports unadjusted p-values for these comparisons and does not report a multiplicity adjustment. When several hypotheses are tested, the chance of at least one apparently unusual result can increase. General statistical context: methods such as family-wise error control or false-discovery-rate procedures can address different multiplicity goals, but the record does not state that either kind of procedure was used. The p-values should therefore be understood as unadjusted rather than as values corrected for the set of comparisons.

The record lists secondary outcomes in addition to the primary endpoint, but does not provide a complete account of the analysis hierarchy or which results were intended to be confirmatory. It reports comparison details for only some secondary contrasts, and relative harm-perception results are not reported in the registry record. Without a prespecified hierarchy and complete reporting, selective emphasis among outcomes and comparisons cannot be assessed from this record.

Missing-data handling is also not reported in the registry record. The record does not state whether scores were missing, how missing scores were handled, or whether the reported analyses included all randomized participants. The analysis-population definitions are not reported in the registry record. It is therefore not possible to determine from the available information whether the reported estimates follow an intention-to-treat approach or another population definition.

The registry identifies the study as randomized, but does not describe the randomization method, the analysis model, covariate adjustment, or a full statistical plan. These gaps do not invalidate the reported descriptive means, but they limit how fully the comparisons can be independently evaluated. The posted results are best interpreted as the reported group summaries and unadjusted contrasts, with no assumption that unreported procedures were conducted.

15. 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 reports zero serious adverse events and zero other adverse events in each arm. The at-risk counts reported for these safety summaries are 254 in the no-imagery control, 244 in flavor imagery, 245 in nature imagery, 252 in medical imagery, and 265 in animal imagery. Deaths are reported as 0 in every arm.

These figures describe adverse events recorded in the registry for this behavioral packaging experiment. They do not show that no unrecorded or unmeasured events occurred. The record does not describe the adverse-event collection method, observation period, or criteria for determining an event, so the safety figures should be read within the limits of that reporting. They are not evidence about the safety of cannabis products themselves, because the study intervention was packaging imagery rather than product consumption.

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

The results describe how participants rated cannabis edible packaging after brief exposure. On the primary appeal measure, the flavor-imagery group had a higher mean score than the no-imagery control, with a reported mean difference of 0.57 and a confidence interval that excludes zero. The other imagery comparisons had intervals that include zero. The data therefore provide evidence of a difference in the flavor comparison as reported, but do not establish that the other themes changed mean appeal.

That interpretation is limited to the measured outcome. A score on a 0-to-10 appeal scale is not a measure of purchasing, product use, or health impact. The record does not report whether ratings predicted later behavior or whether exposure to the packages had effects beyond the assessment that followed exposure for up to 1 min. It also does not report a threshold for deciding how large a score difference would matter in practice.

The trial does not establish that all consumers would respond in the same way. Participant eligibility, demographics, recruitment, and other population characteristics are not reported in the registry record. The imagery was organized around specified themes and four product categories, so the findings should not automatically be generalized to other package designs, products, or settings.

The secondary ratings address distinct perceptions, and their reported pattern is not a substitute for the primary outcome or a behavioral endpoint. In addition, the record does not report multiplicity adjustment, missing-data handling, or a full analysis plan. The results are best read as evidence about ratings in the experiment, with uncertainty and reporting limits, rather than as a broad claim about consumer behavior or the effects of cannabis packaging in the world.

17. Why This Trial Matters Statistically

This experiment illustrates a useful distinction in clinical and behavioral statistics: randomization can support comparisons between study conditions, but the meaning of a result depends on what was measured. Here, the primary endpoint is an immediate rating of appeal after brief exposure, not a later action or health outcome. Keeping that distinction clear prevents a measured difference in perception from being overstated as a difference in behavior.

The study also offers a practical example of interpreting several group comparisons against a common control. The point estimates and confidence intervals communicate both the direction of each contrast and its uncertainty. A p-value alone would not show the size of the difference or how precisely it was estimated. Reading these quantities together helps separate statistical evidence from judgments about practical importance.

Multiplicity is relevant because several imagery themes and secondary perceptions are considered. An unadjusted result can be informative, but it should not be treated as if only one comparison had been examined. The registry’s incomplete reporting of multiplicity plans, missing-data methods, and secondary results demonstrates why transparent analysis plans and complete outcome reporting matter when interpreting experiments with multiple questions.

Finally, the sample-size illustrations show how planning requirements depend on the effect a study aims to detect. Smaller target effects require more participants under the stated assumptions. Such calculations are planning aids, not evidence that a completed study had adequate power or that its observed effect will recur. Together, these statistical considerations help readers judge what the posted results answer, what remains uncertain, and what further evidence would be needed.

18. 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 primary measure is a single appeal question asked after exposure lasting up to 1 min. A brief self-reported rating may not represent stable preferences or real-world behavior. The registry does not report validation evidence for this item, a threshold for practical importance, or repeated assessments over time.

The stimuli covered four product types and four imagery themes, alongside packages without added imagery. The record does not report how the specific package designs were created, whether participants saw every package in a standardized order, or whether other package features were held constant. Those details would help assess whether responses can be attributed specifically to imagery themes rather than to other design differences.

Generalizability is difficult to judge because eligibility criteria, recruitment methods, and participant characteristics are not reported in the registry record. The record identifies Cannabis Use as a condition but does not give enough information to describe the sample’s background or how it relates to consumers in other settings.

For statistical interpretation, the registry reports unadjusted primary comparisons but does not report a multiplicity strategy, missing-data procedures, analysis-population definitions, or the full original analysis plan. Relative harm-perception results are not reported in the registry record, despite being listed as a secondary outcome. These reporting gaps prevent a complete assessment of all planned outcomes and analyses. Finally, because the study measures perceptions of packaging rather than product use or health outcomes, it cannot answer questions about actual purchasing, consumption, or clinical effects.

19. Related Tutorials

20. Related Statistical Calculators

21. Sources

• ClinicalTrials.gov: Cannabis Edibles Packaging Imagery Experiment — NCT06358144.
• Registry record last updated 2025-06-08; status last verified 2025-03.
• Results first posted 2025-06-08.
• 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.

22. Record Summary

The registry record describes a completed randomized, single-masked, parallel behavioral experiment with enrollment of 1260. It compares four cannabis edible packaging imagery themes—flavor, nature, medical, and animal—with a no-imagery control. The primary outcome is mean product-appeal rating on a 0-to-10 scale after exposure to packaging for up to 1 min.

The flavor-imagery arm had a mean appeal score of 6.11, compared with 5.54 in the control; the posted mean difference was 0.57, with a 95% confidence interval from 0.085 to 1.055 and an unadjusted p-value of 0.021. The other imagery comparisons had confidence intervals that included zero. Secondary outcomes concern appeal to children and relative harm perceptions, but the registry does not report results for relative harm perceptions. Multiplicity handling, missing-data methods, and analysis-population definitions are not reported in the registry record.