Pregnenolone and Marijuana Dependence: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Pregnenolone and Marijuana Dependence (NCT02439814), a randomized phase 2 study. The primary outcome is Subjective Rating of Marijuana Craving on 1-7 Likert Scale.

2. Clinical Question

3. Trial Design

The registry record describes a randomized, double-masked, parallel-group phase 2 trial comparing pregnenolone with placebo. Enrollment was 21. The arm summaries list 11 participants for pregnenolone and 10 for placebo. The study status is completed. The sponsor is the Medical University of South Carolina.

The recorded study start was 2015-06, and primary completion was 2016-04. The condition listed is marijuana dependence. The brief summary frames the work as a pilot study of an acute pregnenolone dose and cue-related craving in individuals with cannabis use disorder. The record does not provide the dose, dosing schedule, detailed visit procedures, or the criteria used to define the study population.

The parallel design compares the groups at the endpoint rather than describing repeated crossover treatment periods. Randomization indicates that assignment was intended to be allocated by chance; the method used to generate or conceal the allocation is not reported in the registry record. Double masking indicates that the trial is recorded as masked at two levels, but the identities of those masked and whether masking was assessed are not reported.

The arm descriptions contain no additional detail. The registry record lists one primary outcome measure and no secondary outcomes posted. It contains no analysis entries. The trial’s original sample-size assumptions and its planned statistical procedures are not reported in the registry record, so the posted result should be read as the available endpoint summary rather than as a complete account of the protocol or analysis plan.

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

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 Subjective Rating of Marijuana Craving on 1-7 Likert Scale, measured over 12:40 to 14:45 (post med administration, post Marijuana cue.

The primary endpoint is titled “Subjective Rating of Marijuana Craving on 1-7 Likert Scale.” It measures subjective craving on a scale where 7 means the craving is most severe. The description specifies change in craving from after medication administration to after an active marijuana cue. The recorded timeframe is 12:40 to 14:45, post medication administration and post marijuana cue.

A cue-related craving measure is relevant to the study’s stated question because the pilot was designed to examine craving in response to a marijuana cue after medication. It captures a participant’s subjective rating in that context. It is not, by itself, a direct measure of marijuana use, dependence over time, intoxication, or a biological mechanism. The registry record does not report additional endpoint measurement details, such as how the change score was calculated from the two ratings or how the scale was administered.

The outcome is continuous and summarized as a mean with a standard deviation. The posted mean values are 1.64 in the pregnenolone arm and 0.8 in the placebo arm. The record describes the endpoint as change, but does not specify the subtraction order for that change. Therefore, the numerical direction should not be translated into an increase or decrease in craving without that detail.

7. Results Posted in the Registry

• Pregnenolone: mean 1.64 versus 0.80 on control; difference 0.84 (95% CI -0.66 to 2.34), standardized effect d = 0.48; z-test p = 0.273.

8. Statistical Interpretation

9. Statistical Methods Explained

The registry record presents a continuous endpoint as a mean and standard deviation in each group, together with a mean difference, an interval, and an unadjusted p-value. The record does not describe the statistical test, model, covariate adjustment, or method used to construct the interval. The following explanations are general statistical interpretation, not a claim about the trial’s unreported analysis plan.

What does a group mean represent?

A mean is an average of the observed values in a group. For this outcome, each value represents a subjective rating of change in marijuana craving on the recorded Likert scale. The mean gives a compact summary of the group’s ratings, but it does not show how individual ratings were distributed or whether the mean represents most participants’ experiences. The scale’s bounds and direction are described in the registry record, but the order used to calculate change is not.

The reported standard deviation describes the spread of individual values around the group mean. It is not the uncertainty of the mean itself. A larger standard deviation indicates more variability among observations. The posted standard deviations are 1.71 for pregnenolone and 1.79 for placebo. Those summaries alone do not reveal the shape of the distributions, the presence of unusual values, or whether assumptions used in an analysis were checked.

What does the mean difference express?

A mean difference expresses the contrast between the group means in the endpoint’s original units. The reported value is 0.84 for pregnenolone relative to placebo. Unlike a standardized effect, this estimate remains on the scale used for craving change. The registry record does not specify the direction in which the individual change scores were calculated, so the contrast should not be paraphrased as a rise or fall in craving.

A mean difference is an estimate, not a guaranteed effect for every participant. It also does not account for every possible source of bias. Randomization can support a causal comparison when appropriately implemented and analyzed, but the allocation procedures and analysis details are not reported in the registry record. The posted summary cannot establish whether all randomized participants contributed data or whether any exclusions occurred.

How should an interval be understood?

In general, a confidence interval expresses uncertainty around an estimate under a specified method. A narrower interval indicates greater precision, while a wider one indicates less precision. The interval reported for this comparison spans values on both sides of zero. That feature signals uncertainty about the direction of the group difference. The registry record does not state the interval’s confidence level or construction method, so those features should not be inferred.

An interval is not a range containing all individual responses, nor does it describe the distribution of participant-level ratings. It concerns uncertainty in the estimated group contrast under the method used. Because the analysis method is not reported, the interval is best interpreted descriptively as the uncertainty range recorded for the estimated mean difference.

What does the p-value contribute?

In general, a p-value measures how unusual data at least as incompatible with a null hypothesis would be under that hypothesis and a specified statistical model. The unadjusted p-value is 0.273. It does not give the probability that the null hypothesis is true, and it does not measure the probability that the result will replicate. It also does not measure the clinical importance of the estimate.

The label “unadjusted” indicates that the value is reported without a multiplicity adjustment in the listed comparison. The registry record does not state whether another analysis used adjustment or whether a broader analysis plan existed. With one primary outcome measure and no secondary outcomes posted, the record contains limited information for evaluating multiplicity across outcomes, but it does not establish the full set of analyses that may have been considered.

Why does variability matter?

General statistical reasoning links precision to both the amount of information and variability in the measurements. More variable ratings make a group contrast harder to estimate precisely. In other settings, extra dispersion relative to a model’s assumptions is called over-dispersion; it can make standard errors too small if the model does not account for it. The endpoint is summarized with means and standard deviations, but the registry record does not report a model or evidence of over-dispersion. It would therefore be inappropriate to claim that over-dispersion was present or that a particular correction was used.

The comparison also includes a standardized effect estimate of 0.480. In general, a standardized effect expresses a difference relative to a measure of variability, which can aid comparisons across scales. It is less directly interpretable in the original craving units than the mean difference. The registry record does not explain how this standardized value was constructed, so it should be treated as a supplementary summary rather than as a replacement for the raw-scale contrast.

The standard errors listed for the two group means are 0.566 and 0.516. A standard error describes uncertainty in an estimated quantity, rather than spread among individual participants. The registry record does not explain how these standard errors were calculated or which inferential procedure used them. Their presence does not resolve the uncertainty about the analysis method or the interpretation of the interval.

10. Confidence Intervals and Precision

The reported mean difference is 0.84. The interval around it extends from a value below zero to a value above zero. Its span therefore includes both a negative contrast and a positive contrast, so the posted estimate does not precisely settle which direction the group difference takes. The registry record does not report the confidence level or method used to construct this interval.

Precision concerns how tightly the data identify an estimate, not whether the estimate is important or whether an intervention should be used. Here, the interval is broad in relation to the point estimate, and the group standard deviations are 1.71 and 1.79. Together, these summaries indicate that the estimated contrast has considerable uncertainty. The endpoint ratings vary among participants, while the arm summaries represent relatively limited information.

The unadjusted p-value of 0.273 is consistent with substantial uncertainty under the null model used for that calculation, although the registry record does not identify the model or test. A p-value does not replace the interval: the interval communicates a range of estimates compatible with the reported method, while the p-value summarizes a different aspect of the evidence.

The record does not provide a threshold for a meaningful difference on the craving scale. It is therefore not possible to determine from the registry record whether any particular value within the interval would be considered important. The interval is informative about uncertainty in the estimated comparison, but not about a clinical decision threshold or durability beyond the recorded timeframe.

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 sample-size figures are planning illustrations, not the number of participants observed in the trial. They are expressed as participants per arm, rather than as total event counts. The assumptions listed for this illustration are a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation. The effect to detect is set to a stated fraction of the observed effect. The registry record’s original design assumptions are not reported, and these figures should not be read as the trial’s original recruitment target.

In plain language, the 80% power column describes the number of participants per arm the illustration estimates would be needed for an 80% chance of detecting the specified effect under its assumptions. The 90% power column describes the corresponding planning count for a 90% chance. These are prospective planning quantities; they do not describe the chance that this completed trial found a true effect.

For planning at 100% of the observed effect, the illustration lists 69 participants per arm for 80% power and 92 per arm for 90% power. At 75% of the observed effect, the figures are 121 and 162 per arm. At 50% of the observed effect, they are 273 and 365 per arm. The pattern reflects a general principle: smaller effects are harder to distinguish from random variation and therefore require more participants, assuming the other planning conditions remain unchanged.

These figures are not observed power. Observed power would not clarify the uncertainty in the result; the estimate and its interval are more informative about the precision of the observed comparison. The planning illustration also depends on assumptions that may not match a future study’s design, outcome distribution, or analysis. It should be understood as a conditional planning aid based on the observed effect, not as a definitive recruitment recommendation.

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 = 0.840; pooled standard deviation σ = 1.748.
2. n = 2 × (1.96 + 0.842)² × 1.748² / 0.840² = 68.0, rounded up to 69 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 <- 0.84                      # observed difference in means
sigma <- 1.74835110288162                     # 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(2439814)
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 = 69 (80% power), 92 (90% power)
 75% of observed effect: n per arm = 121 (80% power), 162 (90% power)
 50% of observed effect: n per arm = 273 (80% power), 365 (90% power)
power.t.test: 69 
Simulated power at n = 69 per arm: 0.807 (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 record lists one primary outcome measure and no secondary outcomes posted. It also records no analysis entries. This means the record does not provide a broader set of posted outcome analyses against which to assess how many comparisons were considered. It does not establish that no other analyses were conducted.

The comparison reports an unadjusted p-value of 0.273. No multiplicity adjustment is reported in the registry record. In general, when many outcomes, timepoints, subgroups, or analytic choices are examined, the chance of finding at least one apparently unusual result can increase. Adjustment methods or a prespecified hierarchy can address some multiplicity concerns, but the registry record does not report a statistical plan, analysis hierarchy, or adjustment procedure. It is not possible to determine from the record whether or how these issues were handled outside the posted result.

Missing-data handling is not reported in the registry record. The group summaries identify 11 participants for pregnenolone and 10 for placebo, but the record does not explain whether those values represent all randomized participants, all participants with endpoint data, or another analysis population. It does not state how incomplete ratings were treated, whether any participants were excluded, or whether sensitivity analyses were performed.

The analysis population is also not reported. The record does not say whether the comparison followed an intention-to-treat approach, included only participants who completed the endpoint, or used another definition. Randomized allocation is recorded, but the relationship between randomization and the analyzed group summaries is not described. These omissions limit assessment of potential bias from missing observations or exclusions.

Accordingly, the posted comparison can be described as a group-level result, but its robustness to missingness, analysis choices, or a larger family of comparisons cannot be evaluated from the registry record. Those details should be considered unreported rather than assumed to have been handled in a particular way.

14. Safety Results

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

Other (non-serious) adverse events

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

The registry record lists serious adverse event figures by arm. For pregnenolone, 11 participants were at risk and 0 were affected by a serious adverse event. For placebo, 10 were at risk and 0 were affected. These are the posted counts for the serious-event fields; they are not a detailed account of event timing, monitoring, or severity categories.

The record also lists 0 affected by other adverse events in each arm, with the same at-risk counts. No serious adverse event terms are listed. A death count is not reported in the registry record, so the absence of a death number should not be interpreted as a reported zero.

Adverse-event counts in a trial of this size offer limited information about uncommon events. The registry record does not provide further safety details, such as event definitions, collection procedures, duration of follow-up, or a narrative description of events. These figures should therefore be read as the posted counts, not as a complete assessment of safety or as evidence about risks outside the recorded trial.

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

The registry result describes a mean craving-change score of 1.64 in the pregnenolone group and 0.8 in the placebo group, with a reported mean difference of 0.84. The estimate is accompanied by an interval that crosses zero and an unadjusted p-value of 0.273. Statistically, the posted comparison is uncertain about the direction and size of the group contrast.

This result does not demonstrate that pregnenolone reduced craving, increased craving, or had no effect. The interval includes possibilities on both sides of zero, and the registry record does not define a threshold for a meaningful change. A non-small p-value does not prove equality or absence of an effect. Likewise, a positive point estimate alone does not establish a reliable or clinically important difference.

The result is specific to a subjective rating after medication and an active marijuana cue at the recorded timeframe. It does not establish effects on marijuana consumption, dependence over time, functioning, withdrawal, or other outcomes. No secondary outcomes are posted. The record also does not report follow-up beyond the stated endpoint timeframe or whether any difference persisted.

The study is identified as a pilot, and its arm summaries include 11 participants for pregnenolone and 10 for placebo. The imprecision of the comparison is consistent with the limited amount of information available. The registry record does not report the original sample-size justification or analysis plan, so the posted planning illustration should not be mistaken for the trial’s intended design.

Overall, the registry supports a narrow description of the observed endpoint summaries and their uncertainty. It does not support a definitive conclusion about treatment efficacy, the absence of an effect, or broader clinical benefit. Further claims would require evidence and details not reported in the registry record.

16. Why This Trial Matters Statistically

This trial is statistically instructive because it illustrates the difference between an observed point estimate and a precise conclusion. The mean difference is 0.84, but the interval spans values on either side of zero. A single estimate can appear to point in one direction while its uncertainty remains broad enough that the data do not establish that direction.

The endpoint also highlights the importance of measurement context. The outcome is a subjective craving rating after medication and an active cue, not a direct count of marijuana use or a longer-term clinical outcome. A statistical comparison can only answer the question represented by the measure and timeframe; interpretation should not expand beyond them.

The posted group standard deviations show that participant ratings vary. In general, variability reduces precision unless balanced by more information or an analysis suited to the data. The planning illustration makes the related design principle visible: detecting a smaller effect generally requires more participants under otherwise fixed assumptions. Those planning values are not evidence that the observed comparison has a particular probability of success.

Finally, the record demonstrates why transparent analysis details matter. The outcome summaries and unadjusted p-value are available, but the statistical test, missing-data approach, analysis population, interval confidence level, and original design assumptions are not reported. Without those details, readers can interpret the result descriptively but cannot fully assess the analysis choices or evaluate how robust the estimate may be.

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.

The record contains limited detail about the intervention and study procedures. It identifies pregnenolone and describes an acute dose in the brief summary, but dose and administration details are not reported. The arm descriptions are blank. These omissions make it difficult to characterize the treatment exposure beyond the recorded intervention name and summary.

Participant characteristics and eligibility criteria beyond the marijuana-dependence condition are not reported in the registry record. The record therefore does not show how representative the enrolled participants were of other populations. The study is a phase 2 pilot with 11 participants in the pregnenolone arm and 10 in the placebo arm, which limits the precision of the group comparison and the information available about adverse events.

The primary outcome is a subjective scale rating at a specified post-medication, post-cue timeframe. The registry record does not report the calculation order for the change score, details of scale administration, or whether ratings were collected at other times. Only one primary measure is listed, with no secondary outcomes posted. The result therefore cannot address other aspects of marijuana use or longer-term outcomes.

Important analysis details are not reported: the statistical test or model, the interval’s confidence level and construction, the analysis population, missing-data handling, and the original sample-size assumptions. The record also does not report allocation procedures beyond randomization or provide information about masking assessment. These limitations restrict evaluation of bias, robustness, and generalizability beyond the specific posted summaries.

18. Related Tutorials

19. Related Statistical Calculators

20. Sources

• ClinicalTrials.gov: Pregnenolone and Marijuana Dependence — NCT02439814.
• Registry record last updated 2018-06-18; status last verified 2018-05.
• Results first posted 2018-06-18.
• 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, double-masked, parallel-group phase 2 trial sponsored by the Medical University of South Carolina. It compares pregnenolone with placebo in marijuana-dependent individuals and lists enrollment of 21. The recorded start was 2015-06, with primary completion in 2016-04.

The primary outcome is subjective change in marijuana craving on a 1–7 Likert scale after medication and an active marijuana cue. The posted means are 1.64 for pregnenolone and 0.8 for placebo; the reported mean difference is 0.84, with an unadjusted p-value of 0.273. The reported interval crosses zero, indicating uncertainty about the direction and size of the contrast.

The record lists no secondary outcomes and no analysis entries. It reports 0 serious adverse events affected in each arm; death counts and key analysis details are not reported in the registry record.