Corticotropin-Releasing Hormone Receptor 1 (CRH1) Antagonism in Anxious Alcoholics^: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Corticotropin-Releasing Hormone Receptor 1 (CRH1) Antagonism in Anxious Alcoholics^ (NCT01227980), a randomized phase 2 study. The primary outcome is Alcohol Craving in Response to the Alcohol Cue Script.

2. Clinical Question

3. Trial Design

This was a Phase 2, randomized, double-masked, parallel-group trial of pexacerfont versus placebo. The registry record lists enrollment of 70 and a completed study. The recorded study period began in 2010-10, with primary completion in 2014-07. The posted primary endpoint analysis contains fewer participants than the enrollment figure because its stated population includes only participants who completed all three script types. The record does not explain the reasons for the difference between enrollment and the endpoint analysis counts.

The intervention was oral pexacerfont: a loading dose of 300 mg/day for 1 week, followed by 100mg/day for 16-20 days. The placebo group received oral placebo during the same described phases. The registry summary describes an inpatient admission to the NIH Clinical Center for approximately 1 month, with additional visits 1 week and 1 month after discharge. It also describes standard treatment for alcohol dependence, including staff support and interventions addressing cravings, anxiety, or other psychological problems.

The primary craving assessment was scheduled 15 minutes before the beginning of script presentation on Day 24, 25, or 26 of the treatment period. Participants completed neutral, alcohol, and stress scripts; the primary analysis population was restricted to those completing all three. The record identifies the endpoint and the group summaries, but further details of the trial's own statistical analysis plan, including its original design assumptions, are not reported in the registry record. The summary also does not report how participants were assigned or masked in operational detail.

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 analyses included only those subjects who completed all three script types (neutral, alcohol, stress)

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 Alcohol Craving in Response to the Alcohol Cue Script, measured over 15 minutes prior to the beginning of script presentation, which occurred on Day 24, 25, or 26 of the treatment period.

The primary endpoint was alcohol craving in response to the alcohol cue script, measured with the Alcohol Urges Questionnaire (AUQ). The assessment was made 15 minutes before the beginning of script presentation, which occurred on Day 24, 25, or 26 of the treatment period. The AUQ is described as an 8-item, self-administered measure of craving in the current context—how much the participant is craving alcohol “right now.” Its score ranges from 8, the lowest craving value, to 56, the highest craving value.

A questionnaire score makes a subjective experience available for structured comparison across randomized groups. The endpoint focuses on craving in a defined context and at a specified point in the study, rather than on all aspects of alcohol dependence or on later drinking behavior. The posted values are group means, so they summarize average scores in the analyzed participants; they do not describe every participant's experience or the full spread of individual scores.

The endpoint population included only participants who completed all three script types. This matters when interpreting the means: they describe that selected analysis population, not necessarily everyone enrolled or everyone randomized. The registry record does not report additional endpoint validation details or how scores were handled if questionnaire items were incomplete.

7. Results Posted in the Registry

• Pexacerfont: mean 12.70 versus 10.70 on control; difference 2.00 (95% CI -5.04 to 9.04), standardized effect d = 0.16; z-test p = 0.578.

8. Statistical Interpretation

9. Statistical Methods Explained

The registry record describes a randomized, double-masked, parallel-group comparison. In general, random assignment is intended to make treatment groups comparable on average, including with respect to factors not measured in the study. Masking is intended to limit the influence of treatment knowledge on participant experience, study conduct, or outcome assessment. These are general features of trial design; the registry entry does not give operational details about the assignment sequence or the masking procedures.

What does the mean difference represent?

The primary endpoint is continuous: it is a numerical questionnaire score. The posted summary is the mean score in each group, and the comparison is a mean difference. The registry record reports 12.7 for pexacerfont and 10.7 for placebo, with a reported mean difference of 2.0. The direction is pexacerfont minus placebo, so a positive estimate indicates a higher observed mean in the pexacerfont group. A mean summarizes the group but can conceal variation among individuals and does not imply that every participant had a score close to that mean.

What does the standard error describe?

The arm values are reported with standard errors: 2.48 for pexacerfont and 2.6 for placebo. In general, a standard error describes the sampling uncertainty of an estimated quantity, such as a group mean. It is not the same as a standard deviation, which describes variation among observations. The registry record labels the reported dispersion as standard error; it does not post the individual-level scores or a standard deviation for this endpoint.

How should the confidence interval and p-value be read?

The comparison reports a confidence interval from -5.042 to 9.042 and an unadjusted p-value of 0.578. In general, a confidence interval describes the precision of an estimate under the statistical procedure used: a wider interval indicates greater uncertainty than a narrower one. It also displays a range of values that are reasonably compatible with the data and model, rather than proving that every value in that range is equally plausible. Since this interval extends across zero, the posted data do not isolate a difference in one direction. The unadjusted p-value summarizes compatibility with a null model of no mean difference under the test's assumptions. It is not the chance that the treatment works, and it does not quantify clinical importance.

The registry labels the p-value unadjusted. The record does not describe an adjustment for multiple comparisons or provide a detailed account of the statistical test used. It also does not report whether the comparison was adjusted for baseline variables or other covariates. These analysis details should not be inferred from the group means and interval alone.

What population does the estimate describe?

The endpoint population was restricted to participants who completed all three script types: neutral, alcohol, and stress. This makes the estimand represented by the posted summaries conditional on completing those scripts. In general, restricting an analysis to completers can change the group being described if completion is related to participant characteristics or outcomes. The registry record does not state why some participants were not included in the endpoint analysis or whether any sensitivity analyses assessed the consequences of this restriction.

The outcome table gives 26 participants for pexacerfont and 25 for placebo. These are endpoint analysis counts, not a replacement for the overall enrollment figure. The record does not provide a participant flow accounting for exclusions, withdrawals, or missing endpoint values. It therefore does not permit a full assessment of how participants moved from enrollment to the primary analysis population.

What cannot be recovered from the posted analysis?

The registry record does not report the trial's original statistical design assumptions, a formal sample-size justification, interim analyses, detailed missing-data methods, or a complete statistical analysis plan. It also does not report individual-level data. Such details can affect interpretation, including how well the analysis assumptions fit the observations and how robust the result is to alternative methods. Their absence means the posted summaries can be described and their uncertainty discussed, but the original planning and full analysis process cannot be reconstructed from this record.

General statistical reasoning can clarify what the reported quantities mean, but it cannot fill in those missing trial-specific details. The results are best read as a group-level estimate with its interval, test, and stated analysis population, rather than as a complete account of every analysis conducted in the trial.

10. Confidence Intervals and Precision

The reported mean difference is 2.0 AUQ units, with a confidence interval from -5.042 to 9.042. This interval is wide relative to the point estimate and includes zero. It also extends to differences favoring either group under the stated subtraction order. The point estimate alone gives one value from the data; the interval shows that the result is not precise enough to distinguish that estimate from a broad range of alternatives.

A confidence interval is not a guarantee that the true difference lies inside its endpoints, nor does it assign a probability to the treatment effect after the data have been observed. In general, it is a range produced by a statistical procedure that, under its assumptions, has a stated long-run coverage property. The registry record does not specify enough detail about the interval procedure to assess all of its assumptions. The posted interval should therefore be interpreted as the uncertainty measure attached to the registry comparison, not as a complete description of the analysis.

The arm means have standard errors of 2.48 and 2.6. These indicate uncertainty in the estimated means, while the interval for the mean difference addresses uncertainty in the contrast between groups. The available summary does not include individual scores or a standard deviation for this endpoint, so it cannot show the distribution of scores or how much participants varied within each arm.

The interval includes differences in opposite directions and values near no difference. Accordingly, the result is inconclusive about the direction and size of the underlying group contrast. A non-significant unadjusted p-value of 0.578 does not make the estimate precise, and it does not establish equivalence. The confidence interval is central to understanding what the posted result can and cannot resolve.

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 table and curve are planning illustrations based on the observed mean difference, not the trial's original design calculation. For this continuous endpoint, the entries are participants per arm, not total events. The listed assumptions are a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. Each row uses a stated fraction of the observed effect as the effect to detect.

For planning at 100% of the observed effect, the table gives 646 participants per arm for 80% power and 864 participants per arm for 90% power. For planning at 75% of the observed effect, it gives 1,147 participants per arm for 80% power and 1,536 participants per arm for 90% power. For planning at 50% of the observed effect, it gives 2,581 participants per arm for 80% power and 3,455 participants per arm for 90% power. The curve presents the same planning relationship visually; it should be read alongside the table and its assumptions.

In plain terms, the 80% power column describes the planned chance of detecting the specified effect if that effect and the stated statistical assumptions were true, using the indicated design. The 90% column applies the higher power target. Smaller effects require more participants for the same target power, which is why the table's requirements rise as the assumed fraction of the observed effect becomes smaller. These are planning illustrations, not observed power. The trial's own original sample-size justification and design assumptions are not reported in the registry record.

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 = 2.000; pooled standard deviation σ = 12.820.
2. n = 2 × (1.96 + 0.842)² × 12.820² / 2.000² = 645.0, rounded up to 646 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 <- 2                      # observed difference in means
sigma <- 12.8203919198904                     # 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(1227980)
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 = 646 (80% power), 864 (90% power)
 75% of observed effect: n per arm = 1,147 (80% power), 1,536 (90% power)
 50% of observed effect: n per arm = 2,581 (80% power), 3,455 (90% power)
power.t.test: 646 
Simulated power at n = 646 per arm: 0.803 (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 posts a primary endpoint and indicates 16 outcome measures, but it does not post secondary outcome results. No multiplicity adjustment is described in the record. When several outcomes or comparisons are tested, multiplicity is a general statistical concern because the chance of at least one apparently unusual result can increase across a collection of tests. Without a reported analysis plan or adjustment procedure, it is not possible to determine how the primary comparison was positioned within the full set of planned analyses or whether other tests were conducted.

The primary endpoint analysis included only participants who completed all three script types: neutral, alcohol, and stress. The posted results list 26 pexacerfont participants and 25 placebo participants for this analysis. The registry record does not give reasons why participants were not included, a flow of participants through the analysis, or a method for handling missing endpoint information. It also does not state whether all randomized participants were analyzed in their assigned groups.

These gaps matter because an analysis limited to completers may not describe the full randomized population if completion is associated with treatment assignment, participant characteristics, or craving scores. This is a general issue, not evidence that such a relationship occurred in this trial. The posted record does not provide enough information to evaluate it.

The missing-data approach, any sensitivity analyses, and any rules for excluding observations are not reported in the registry record. The original statistical plan, covariate adjustment, interim analyses, and multiplicity strategy are also not reported. As a result, readers can describe the posted unadjusted primary comparison and its stated analysis population, but cannot verify from the registry entry whether the analysis followed a prespecified plan or how robust the finding is to alternative analysis populations or missing-data assumptions.

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 5% reporting threshold.

Most frequent non-serious adverse event terms

The adverse-event figures are reported separately for the two study groups. In the pexacerfont group, 40 participants were at risk for serious adverse events and 0 were reported as affected. In the placebo group, 30 were at risk and 0 were reported as affected. The record lists no serious adverse-event terms. These figures describe what is posted in the registry; they do not establish that no events occurred outside the record's reporting definitions or observation period.

For other, non-serious adverse events, the registry lists 34 affected participants among 40 at risk in the pexacerfont group and 28 affected among 30 at risk in the placebo group. These are counts of participants affected, not event totals, and the registry excerpt does not give a breakdown by event type or severity. Death counts are not reported in the registry record.

The at-risk counts in the adverse-event summary differ from the primary endpoint analysis counts. The record does not explain the relationship between those populations. These safety summaries should therefore be read as descriptive reporting, not as a direct statistical comparison of risk or evidence about causation. No percentages or comparative safety estimates are posted.

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

The primary result describes average AUQ craving scores among participants who completed all three script types. The observed mean was 12.7 in the pexacerfont group and 10.7 in the placebo group, with a reported mean difference of 2.0 units. The confidence interval, -5.042 to 9.042, shows considerable uncertainty and includes differences in either direction. The unadjusted p-value was 0.578. Taken together, the posted statistics do not establish a clear group difference, but they also do not show that the groups are equivalent or that a difference is absent.

This endpoint is a measure of reported craving in a particular context, just before an alcohol-cue script on Day 24, 25, or 26 of treatment. It is not a measure of every dimension of alcohol dependence, and the posted primary result does not directly report drinking behavior, relapse, or outcomes after discharge. The registry record's study summary describes additional support and visits, but the result is specifically the primary AUQ score.

The endpoint analysis is restricted to participants completing all three scripts. Without a participant flow or missing-data analysis, it is not possible to know from the registry record how the analyzed group relates to all participants enrolled or randomized. The posted summaries also do not reveal individual responses or the distribution of scores within each arm.

Accordingly, this result should be understood as an imprecise, context-specific comparison in a selected analysis population. It cannot be used by itself to make a broad claim about the overall course of alcohol dependence or the effect of treatment in other populations. It also cannot establish a safety conclusion: the adverse-event counts are descriptive, and the registry record does not report deaths. The interpretation is limited to what the posted endpoint and safety summaries show.

16. Why This Trial Matters Statistically

This trial illustrates why a point estimate, a confidence interval, and a p-value answer different questions. The mean difference of 2.0 gives the observed direction and size of the contrast in the analyzed participants. The interval from -5.042 to 9.042 conveys substantial uncertainty about that contrast. The unadjusted p-value of 0.578 summarizes compatibility with a null model under the test, but it does not tell readers whether the estimate is clinically important or precise. Reading these quantities together avoids treating one statistic as a complete verdict.

The analysis population is also statistically important. Although enrollment was 70, the primary endpoint results cover only participants who completed all three script types, with 26 in the pexacerfont arm and 25 in the placebo arm. This distinction shows why enrollment and endpoint-analysis counts should not be treated as interchangeable. Without a participant flow and missing-data methods, the degree to which the endpoint analysis reflects the full randomized groups cannot be assessed.

The planning illustration further demonstrates how sample-size requirements depend on the effect a future study aims to detect and on assumptions such as allocation, significance level, and approximation method. Those calculations are not a retrospective measure of this trial's power, and they do not substitute for the original design justification, which is not reported in the registry record.

More broadly, the record highlights the value of reporting a prespecified analysis plan, participant flow, missing-data methods, and multiplicity strategy alongside outcome summaries. Without those details, the observed result can be described, but its relationship to all planned analyses and the full randomized population remains uncertain.

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 primary endpoint result is based on participants who completed all three script types, rather than an explicitly described analysis of every randomized participant. The registry record does not explain exclusions or non-completion, or report a participant flow, so the reasons for the restricted analysis population and its relationship to enrollment cannot be evaluated.

The posted result is a group-level mean and standard error. Individual AUQ scores, score distributions, and a standard deviation are not reported in the registry record. The confidence interval is broad and spans differences in either direction, limiting the precision of the comparison. The p-value is identified as unadjusted, while a multiplicity strategy is not reported.

The record does not report the trial's original sample-size justification, detailed statistical analysis plan, covariate adjustment, interim analyses, or missing-data procedures. It also does not report secondary outcome results. These omissions prevent a complete assessment of whether the posted analysis reflects all planned outcomes and how sensitive the primary result might be to alternative assumptions.

The endpoint is a self-administered craving questionnaire assessed in a defined cue-script setting. The primary result does not directly describe drinking outcomes or longer-term outcomes, and the registry record does not post those results. The adverse-event summary is also limited: it gives affected-participant counts without event-specific breakdowns, and death counts are not reported in the registry record. These limitations restrict interpretation to the posted summaries and do not support broader conclusions beyond them.

18. Related Tutorials

19. Related Statistical Calculators

20. Sources

• ClinicalTrials.gov: Corticotropin-Releasing Hormone Receptor 1 (CRH1) Antagonism in Anxious Alcoholics^ — NCT01227980.
• Registry record last updated 2016-02-02; status last verified 2015-12.
• Results first posted 2015-12-10.
• 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 Phase 2 trial comparing oral pexacerfont with oral placebo. Enrollment was 70. The primary endpoint was AUQ alcohol craving measured shortly before an alcohol-cue script, with analysis limited to participants completing neutral, alcohol, and stress scripts.

The posted mean difference was 2.0 units, with a confidence interval from -5.042 to 9.042 and an unadjusted p-value of 0.578. The interval is compatible with differences in either direction and indicates substantial uncertainty. Secondary results and several analysis details, including missing-data methods and the original sample-size justification, are not reported in the registry record.