Varenicline to Reduce Alcohol Consumption in Heavy Drinkers: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Varenicline to Reduce Alcohol Consumption in Heavy Drinkers (NCT00695500), a randomized phase 1, phase 2 study. The primary outcome is Alcohol Consumption.

2. Clinical Question

3. Trial Design

The registry record describes a randomized, double-blind, parallel-group trial of varenicline and placebo, with PHASE1 and PHASE2 listed. The trial is marked completed. Enrollment is recorded as 50. The listed start is 2008-06 and primary completion is 2014-07. These record fields describe the study's status and timeline; they do not establish how many people contributed data to each posted result.

The two treatment assignments were varenicline, 2 mg per day for 3 weeks, and placebo, 0 mg per day for 3 weeks. The brief summary describes outpatient assessments and an intravenous alcohol self-administration procedure in which participants could control additional alcohol infusions. It also describes an overnight hospital admission and other assessments. The primary endpoint was collected during a 2.5 hr session following the treatment period.

The primary result uses the sample that completed the assessment. The posted result lists 16 participants for varenicline and 12 for placebo. These are endpoint-specific group counts, not a complete accounting of enrollment, randomization, or every assessment. The registry record does not report why the endpoint counts differ from the enrollment field, whether participants were excluded from this analysis, or how missing observations were handled.

Blinding and randomization are design features intended to reduce systematic differences in treatment allocation and assessment. The registry record does not give further details about the randomization procedure, allocation concealment, how blinding was maintained, or whether the analysis followed the original assignment for every randomized participant. Those details are not reported in the registry record.

4. Participant Flow and Accounting

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

Before assignment: Four participants were enrolled (consented) but withdrew prior to group assignment.

Overall Study

Analysis population for the primary outcome, as defined in the registry: sample that completed the assessment

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 Consumption, measured over 2.5 hr session following 3 weeks of treatment.

The primary endpoint is peak breath alcohol concentration during an intravenous alcohol self-administration session. It is reported as a mean in mg/% with standard errors for the sample that completed the assessment. The specified timeframe is the 2.5 hr session following 3 weeks of treatment.

This measure captures the highest recorded breath alcohol concentration during a structured session in which participants could self-administer alcohol through an intravenous infusion system. It is therefore a session-based measure of alcohol exposure, rather than a direct count of drinks consumed during ordinary daily life. The registry record names peak concentration as the primary outcome but does not provide further details on its measurement schedule, the method used to identify the peak, or a clinically meaningful difference.

In general, a continuous endpoint can retain information about the magnitude of measurements rather than reducing them to categories. A peak value can describe the maximum observed concentration within the assessment period, while a group mean describes the average peak across the participants with measurements. The endpoint helps frame the trial's statistical question, but a difference in this laboratory-session measure alone does not establish what might happen to drinking in other settings.

7. Results Posted in the Registry

• Varenicline: mean 92.00 versus 77.30 on control; difference 14.70 (95% CI -20.19 to 49.59), standardized effect d = 0.32; z-test p = 0.409.

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.

Alcohol Urges

• 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:
  • Varenicline: mean 23.65 versus 26.82 on control; difference -3.17 (95% CI -12.19 to 5.85), standardized effect d = -0.27; z-test p = 0.491.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Varenicline and the control. The interval gives the range of effects that remain compatible with the data.

The posted secondary outcome concerns alcohol urges during the same 2.5 hr session following 3 weeks of treatment. Its group summaries and unadjusted comparison are presented with the primary result, but the registry record lists no reported analyses for secondary outcomes. The pattern across the posted endpoints should therefore be read descriptively: the secondary measure does not supply a separate, adjusted confirmation of the primary finding.

In general, secondary endpoints can provide context about related aspects of a trial, but they do not automatically validate or contradict a primary endpoint. Alcohol urges and peak breath alcohol concentration capture different features of the session. A pattern that is not identical across measures is not inherently inconsistent, because the measures represent different outcomes and may have different variability and precision. The record does not report a prespecified hierarchy or a joint analysis linking these outcomes.

Unadjusted secondary results carry less weight for confirmatory interpretation when multiple outcomes may be examined. In general, testing several outcomes can increase the chance that at least one result appears notable even when there is no corresponding underlying effect. Adjusting for multiplicity can address some of this issue, but the registry record does not describe a multiplicity procedure or identify which secondary comparisons, if any, were prespecified for formal testing. The posted secondary result should consequently be treated as exploratory context rather than independent proof of a treatment effect.

The record identifies Alcohol Urges as a secondary outcome and includes one posted secondary result. It does not report the full set of analysis decisions, whether other secondary measures were collected but not posted, or how missing secondary observations were handled. Those omissions limit comparisons across outcomes and make it inappropriate to use the secondary result to settle the uncertainty around the primary comparison.

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

• Alcohol Urges

9. Statistical Interpretation

10. Statistical Methods Explained

The posted primary outcome is a continuous measure summarized by a mean and standard error in each group. The comparison is a mean difference, with the varenicline group contrasted against the placebo group. The record does not report an adjusted model, covariates, or a detailed statistical analysis plan. The figures should therefore be understood as the unadjusted comparison posted in the registry record.

What does a mean difference represent?

In general, a mean is the arithmetic average of the observed measurements in a group. A mean difference expresses the contrast between two group means in the original measurement units. Here, those units are mg/%. A positive reported difference indicates a higher posted mean for varenicline than placebo; a negative difference would indicate a lower mean for varenicline. This comparison describes the observed groups, while statistical uncertainty is considered separately through the interval estimate and p-value.

The primary means are based on the participants who completed the assessment. A mean does not describe every individual's measurement, nor does it show the distribution's shape. The registry record does not post participant-level data, medians, ranges, or a distributional plot in the registry record for this analysis. As a result, it is not possible to assess from these summary values whether measurements were symmetric, clustered, or affected by extreme observations.

What does the standard error describe?

In general, the standard error describes the sampling uncertainty of an estimated quantity, such as a group mean. It is not the same as the standard deviation, which describes variability among individual measurements. A smaller standard error generally indicates a more precise estimate of the group mean, while a larger standard error indicates greater sampling uncertainty. The primary group results report standard errors of 11.6 for varenicline and 13.5 for placebo. The registry record does not report standard deviations for this endpoint.

Standard errors can help readers understand why a difference estimate is uncertain, but they do not give the full picture by themselves. The interval around the mean difference incorporates uncertainty in the comparison. It extends from -20.19 to 49.59, a span that includes zero and values in both directions. The sample sizes and variability in the groups contribute to the uncertainty; the record does not provide additional model details for how the interval was constructed.

How should the p-value be read?

In general, a p-value is calculated under a specified statistical model and a null hypothesis, commonly a hypothesis of no group difference. It measures the compatibility of the observed data with that model, not the size or importance of a result. The posted primary p-value is unadjusted and equals 0.409. It should not be read as the chance that the result is due to chance, nor as the probability that the treatment has no effect.

The registry record lists no reported analyses. It nevertheless posts outcome summaries and a comparison estimate with an unadjusted p-value. No detailed analysis method, degrees of freedom, distributional assumptions, or adjustment strategy is reported in the registry record. It is therefore not possible to determine from the record whether the posted comparison reproduces the trial's prespecified primary analysis or whether other analyses were conducted but not posted.

Why can over-dispersion matter?

In general, over-dispersion occurs when the observed variation is greater than a statistical model expects. It is most often discussed for count data, where an overly restrictive variance model can make standard errors too small and apparent evidence too strong. The primary outcome in this trial is continuous, not a count, and the registry record does not report an over-dispersion assessment. The broader lesson is that uncertainty estimates depend on whether the statistical model adequately represents the data; no specific over-dispersion issue can be inferred for this endpoint.

What does randomization contribute?

In general, random assignment aims to make treatment groups comparable with respect to measured and unmeasured baseline characteristics, apart from chance variation. Double blinding is intended to limit influences from participants' and investigators' knowledge of assignment. These design features can strengthen the interpretation of a between-group comparison, but they do not guarantee that every analysis is free of bias. The registry record does not report the random sequence method, allocation concealment, adherence, or whether the posted analysis preserved the original randomized groups.

The result is reported for the sample that completed the assessment. The record does not state that this was an intention-to-treat analysis, describe a per-protocol definition, or explain how missing endpoint values were treated. In general, excluding participants with missing measurements can affect comparability if completion is related to assignment or outcome. Without the missing-data details, the degree to which that concern applies cannot be assessed from the registry record.

11. Confidence Intervals and Precision

The primary mean difference is 14.7 mg/%, and its reported interval extends from -20.19 to 49.59. This interval is a key guide to precision: it includes zero, as well as both negative and positive values. Thus, the posted result does not pin down the direction of the difference, and the range of values compatible with the estimate is broad. The registry record does not state the confidence level associated with this interval, so no confidence level should be inferred from the record.

In general, a confidence interval represents a range of values compatible with the data under the assumptions of the analysis. It is not a guarantee that the true value lies inside a particular calculated interval, and it does not show the probability that any one value is correct. A narrow interval indicates more precision than a wide interval, but precision is distinct from clinical importance. The registry record does not define a clinically important difference for peak breath alcohol concentration.

The endpoint analysis includes 16 participants assigned to varenicline and 12 assigned to placebo. The standard errors for the two means are 11.6 and 13.5, respectively. These quantities help convey sampling uncertainty, but they do not remove uncertainty about the analysis population, missing observations, or model choices. Because reasons for the endpoint sample counts and missing-data methods are not reported in the registry record, the interval should be interpreted as applying to the posted comparison and its underlying analysis assumptions.

The unadjusted p-value of 0.409 is consistent with the imprecision conveyed by the interval: the posted data do not sharply distinguish the groups. Neither the interval nor the p-value establishes that the groups are equivalent. Establishing equivalence would require a defined margin and an analysis designed for that question; such a margin and analysis are not reported in the registry record.

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 planning figures are participants per arm, not total events. They are illustrations based on the observed effect, not the original trial's design assumptions. The registry record states the planning assumptions: a two-sided significance level of 0.05, equal allocation between the two arms, a normal approximation, and a target effect set to the stated fraction of the observed effect. The trial's own original sample-size assumptions are not reported in the registry record.

The 80% power column describes the planned number of participants per arm for a hypothetical study to have an 80% chance of detecting the specified effect under the stated assumptions, if that effect were the true one. The 90% power column describes the corresponding planning target for a 90% chance. These are design probabilities under assumptions, not guarantees about a future study's result. They are not probabilities that the treatment works.

For the primary endpoint, planning for 100% of the observed effect gives 158 participants per arm for 80% power and 211 participants per arm for 90% power. Planning for 75% of the observed effect gives 280 participants per arm for 80% power and 375 participants per arm for 90% power. Planning for 50% of the observed effect gives 630 participants per arm for 80% power and 844 participants per arm for 90% power. Smaller effects need more participants to detect under the same planning assumptions, which is why the illustrated requirements increase as the target effect is reduced.

These are sample-size planning illustrations, not observed power. They use the observed effect estimate, which is uncertain, and should not be treated as a definitive prescription for a new study. The record does not report the assumptions used to plan the original trial, including any original target effect or power target.

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 = 14.700; pooled standard deviation σ = 46.555.
2. n = 2 × (1.96 + 0.842)² × 46.555² / 14.700² = 157.4, rounded up to 158 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 <- 14.7                      # observed difference in means
sigma <- 46.5549303346223                     # 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(695500)
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 = 158 (80% power), 211 (90% power)
 75% of observed effect: n per arm = 280 (80% power), 375 (90% power)
 50% of observed effect: n per arm = 630 (80% power), 844 (90% power)
power.t.test: 158.4 
Simulated power at n = 158 per arm: 0.802 (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 registry record identifies one primary outcome and Alcohol Urges as a secondary outcome. It reports no analyses in its analysis listing, although group summaries and unadjusted comparison results are posted. The record does not describe the prespecified statistical analysis plan, which outcomes were intended to support confirmatory claims, or whether any multiplicity adjustment was planned or performed. As a result, the statistical status of the posted secondary comparison cannot be fully established.

In general, multiplicity arises when researchers make several statistical comparisons or examine multiple outcomes. Without an adjustment or a prespecified strategy, the chance of at least one apparently unusual result can increase across a collection of tests. This does not mean that a particular secondary result is false; it means that an unadjusted result should be interpreted with the number and role of analyses in mind. The registry record does not provide enough information to quantify that issue for this study.

The primary endpoint is defined for the sample that completed the assessment. The posted group counts are 16 for varenicline and 12 for placebo, while the enrollment field records 50. The record does not explain how enrollment, randomization, and endpoint completion relate to one another. It does not report the number of missing primary measurements, reasons for missingness, imputation methods, sensitivity analyses, or a participant flow account. Those items are not reported in the registry record.

The registry record also does not identify the analysis population as intention-to-treat, modified intention-to-treat, or per-protocol. In general, different analysis populations answer different questions, and exclusions after randomization can affect the balance created by assignment. Without further information, the posted comparison should be described as the result among the assessment completers named for the endpoint, not as a verified analysis of all randomized participants.

There is also no reported detail on interim analyses, data monitoring decisions, adjustment for baseline variables, or handling of protocol deviations. These features should not be assumed to have been absent; their status is simply not reported in the registry record. This uncertainty about analysis procedures is separate from the statistical uncertainty represented by the interval estimate.

15. 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 4% reporting threshold.

Most frequent non-serious adverse event terms

The registry record reports no serious adverse events affecting participants at risk in either group: 0 of 24 for varenicline and 0 of 22 for placebo. The figures for other adverse events are 21 of 24 for varenicline and 18 of 22 for placebo. These are counts from the safety fields and should be read separately from the primary endpoint sample counts, which refer to participants who completed the alcohol assessment.

The adverse event counts describe what was recorded in the registry. They do not establish that an event was caused by treatment, and the record does not provide event types or severity details for the other adverse events. Death counts are not reported in the registry record. The record also does not provide enough detail to assess differences in follow-up duration, ascertainment, or the timing of events. These summaries should not be used to infer safety beyond the recorded groups and outcomes.

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

The posted primary comparison describes peak breath alcohol concentration during a controlled intravenous self-administration session following 3 weeks of treatment. The varenicline group had a higher posted mean than the placebo group, but the estimated difference is uncertain: the reported interval includes zero and extends in both directions. The unadjusted p-value also does not provide strong evidence for a clear between-group difference in this endpoint.

This result does not establish that the treatments are equivalent. A result that does not clearly distinguish groups is not the same as evidence that any difference is negligibly small. The registry record does not specify an equivalence margin or a non-inferiority question. The uncertainty in the interval leaves a range of possible group differences compatible with the posted estimate.

The endpoint is a peak concentration measured in a structured session, not a direct measure of typical drinking over daily life or a long-term outcome. It does not, on its own, show what the treatment might do to alcohol consumption in other circumstances. The record's title and summary concern reducing alcohol consumption in heavy drinkers, but the primary endpoint itself is specifically peak breath alcohol concentration during the intravenous session.

The result is also limited to the assessment completers described for the primary outcome. The registry record does not report the reasons for the endpoint sample counts, the handling of missing data, or whether the posted analysis included all randomized participants. These gaps prevent a full appraisal of how participant selection into the endpoint analysis may affect interpretation.

More broadly, the record does not provide the statistical analysis plan, a reported multiplicity strategy, or evidence about the clinical importance of a particular difference on this scale. The appropriate reading is therefore narrow: the posted data give an uncertain unadjusted comparison for this session-based measure. They do not settle broader questions about usual alcohol use, long-term effects, or treatment benefit.

17. Why This Trial Matters Statistically

This trial illustrates how the endpoint chosen determines the scope of a statistical conclusion. The primary measure is peak breath alcohol concentration during a controlled intravenous alcohol session. That measure is distinct from alcohol use measured in everyday settings, so conclusions need to remain tied to the actual outcome and assessment period.

It also shows why an estimate should be read together with its precision. A group difference alone can suggest a direction, but the interval indicates how uncertain that estimate is. In this trial, the interval includes zero and spans negative and positive differences. The unadjusted p-value is not a substitute for considering that range or the endpoint's meaning.

The registry record provides group summaries and a comparison, yet leaves important analysis details unreported. The endpoint population is described as assessment completers, while the record does not explain missingness or identify the analysis population more specifically. Nor does it describe a multiplicity plan for secondary outcomes. These omissions matter because statistical interpretation depends not only on a numerical result, but also on who contributed data and how the analysis was planned.

Finally, the planning illustrations highlight the difference between a trial's observed result and the design of a future study. They show how the number of participants needed depends on the effect a study aims to detect and the desired power under stated assumptions. They do not recover the original design rationale for this trial, which is not reported in the registry record. Together, these features make the study useful for learning to separate effect estimates, uncertainty, design assumptions, and limits on generalization.

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.

Several limitations specific to this record affect interpretation. The primary endpoint uses the sample that completed the assessment, with group counts of 16 and 12. The enrollment field records 50, but the registry record does not explain the relationship between enrollment and the endpoint samples, or why some enrolled or randomized participants may not appear in the endpoint summaries. The reasons for missing assessments and the handling of missing data are not reported in the registry record.

The outcome is a peak concentration in an intravenous self-administration session after 3 weeks of treatment. It does not directly represent ordinary drinking patterns, and the record does not provide a clinically meaningful threshold for interpreting differences in mg/%. It also does not report further measurement details that would allow assessment of how the peak was identified or how repeat measurements contributed.

The posted primary comparison is unadjusted. No analysis plan, covariate-adjustment approach, or sensitivity analysis is reported. Although the trial is described as randomized and double-blind, the registry record does not detail the sequence-generation method, allocation concealment, maintenance of blinding, adherence, or the analysis population. These design and analysis details remain uncertain.

There is also limited information for interpreting other outcomes. The registry record identifies Alcohol Urges as a secondary outcome but does not describe a multiplicity strategy, and it lists no reported analyses. The safety fields give adverse-event counts but do not report death counts, event types for other adverse events, or detailed follow-up information. These limits constrain conclusions about the broader outcome profile and the completeness of reporting.

19. Related Tutorials

20. Related Statistical Calculators

21. Sources

• ClinicalTrials.gov: Varenicline to Reduce Alcohol Consumption in Heavy Drinkers — NCT00695500.
• Registry record last updated 2016-08-15; status last verified 2016-07.
• Results first posted 2016-08-15.
• 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

NCT00695500 is a completed randomized, double-blind, parallel-group trial listed as PHASE1 and PHASE2. It compared varenicline tablets at 2 mg per day for 3 weeks with placebo tablets at 0 mg per day for 3 weeks in heavy drinkers. The registry record lists enrollment as 50 and primary completion as 2014-07.

The primary outcome was mean peak breath alcohol concentration during a 2.5 hr intravenous alcohol self-administration session after treatment. In the assessment-completer sample, the posted means were 92.0 mg/% for varenicline and 77.3 mg/% for placebo. The unadjusted mean difference was 14.7 mg/%, with a reported interval from -20.19 to 49.59 and an unadjusted p-value of 0.409.

The interval is compatible with differences in either direction and is not precise enough to establish the size or direction of a clear group difference. Reasons for the endpoint sample counts, missing-data methods, the original sample-size assumptions, and the detailed statistical analysis plan are not reported in the registry record.