Pharmacogenetics of Alcohol: Treatment Implications: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Pharmacogenetics of Alcohol: Treatment Implications (NCT00734656), a randomized na study. The primary outcome is Breath Alcohol.

2. Clinical Question

3. Trial Design

The registry describes a randomized, quadruple-masked, factorial study with a within-subjects design. Its stated objective was to explore whether some effects of alcohol are mediated by increased production of neuroactive steroids that interact with GABAA-receptors. The proposed four-session arrangement paired alcohol or placebo alcohol with dutasteride or placebo medication. Dutasteride is described in the registry as a 5-alpha steroid reductase inhibitor that limits production of dihydrotestosterone and certain 5a-reduced neuroactive steroids.

The four conditions were Placebo Medication + Placebo Alcohol, Placebo Medication + 0.8 gr/kg Ethanol, 4 mg Dutasteride + Placebo Alcohol, and 4 mg Dutasteride + 0.8 mg/kg Ethanol. The primary endpoint was breath alcohol 40 minutes after beginning the drink. The posted outcome data list 70 participants in each condition. The primary outcome description says the analysis population comprised subjects who completed all four arms, and notes that three subjects were removed during data cleaning because of pharmacy errors or a protocol deviation. The relation between that description and the total enrollment is not fully explained in the registry record.

The study is marked completed, with enrollment of 94. Its phase is recorded as NA. The registry lists UConn Health as sponsor and includes a secondary outcome measuring change in serum 3a-androstanediol Glucuronide. The original allocation sequence, exact masking procedures, full eligibility criteria, and details of the statistical analysis plan are not reported in the registry record. The factorial label indicates the design combined medication and alcohol conditions, but the registry record does not provide enough detail to reconstruct a formal factorial model or interaction analysis.

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: A total of 148 subjects enrolled, 31 subjects were excluded for not meeting entrance criteria, 23 subjects withdrew before first dose of study medication. 94 subjects were randomized to one of 24 possible laboratory session sequences for exposure to each of 4 treatment conditions.

Lab Session 1

Reasons for not completing this period:

Washout 1

Reasons for not completing this period:

Lab Session 2

Reasons for not completing this period:

Washout 2

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Subjects who completed all 4 arms of study (e.g. completed each combination of dutasteride or placebo paired with 0.8 gr/kg ethanol or alcohol flavor mask)less 3 subjects removed during data cleaning due to pharmacy errors (n=2) or protocol deviation (n=1)resulted in 70 subjects completing each condition for analysis.

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 Breath Alcohol, measured over 40 minutes after beginning drink.

The primary endpoint is mean breath alcohol, measured in gr/dL 40 minutes after beginning the drink. Breath alcohol is a physiological measure related to the alcohol condition administered in the study. In this design, it helps describe whether the assigned alcohol conditions produced different measured breath alcohol levels, and whether those levels differed in the presence or absence of dutasteride.

The posted condition means show the measured level for each medication and alcohol combination. The registry’s primary outcome description focuses on subjects who completed all four study arms, and the posted outcome counts are 70 in each condition. The outcome is continuous, so the reported means and their standard errors describe the central level and the precision of those mean estimates. The registry does not report a full account of how breath alcohol was measured, whether the endpoint was prespecified in a detailed analysis plan, or whether any additional primary endpoints were considered.

Breath alcohol is not itself a direct measure of subjective effects, neuroactive steroid production, or clinical outcomes related to alcohol use. The trial’s stated scientific motivation concerns subjective and physiological effects and a possible steroid pathway, but the primary endpoint reported in the outcome record is specifically breath alcohol. Interpretation should therefore remain tied to the measured breath alcohol levels and the recorded study conditions rather than extending to outcomes that were not measured in this endpoint.

7. Results Posted in the Registry

• Placebo Medication + 0.8 gr/kg Ethanol: mean 0.07 versus 0.00 on control; difference 0.07 (95% CI 0.07 to 0.08), standardized effect d = 4.13; z-test p = <0.001.
• 4 mg Dutasteride + Placebo Alcohol: mean 0.00 versus 0.00 on control; difference 0.00 (95% CI -0.00 to 0.00), standardized effect d = 0.00; z-test p = 1.000.
• 4 mg Dutasteride + 0.8 mg/kg Ethanol: mean 0.07 versus 0.00 on control; difference 0.07 (95% CI 0.06 to 0.08), standardized effect d = 3.91; z-test p = <0.001.

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.

Change in Serum 3a-androstanediol Glucuronide

• 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:
  • Placebo Medication + 0.8 gr/kg Ethanol: mean 1.11 versus 1.04 on control; difference 0.07 (95% CI -0.02 to 0.16), standardized effect d = 0.27; z-test p = 0.117.
  • 4 mg Dutasteride + Placebo Alcohol: mean 0.31 versus 1.04 on control; difference -0.73 (95% CI -0.81 to -0.65), standardized effect d = -3.20; z-test p = <0.001.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Placebo Medication + 0.8 gr/kg Ethanol and the control. The interval gives the range of effects that remain compatible with the data.
  • The 95% confidence interval excludes no difference, so the mean is statistically lower for 4 mg Dutasteride + Placebo Alcohol than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

The registry posts one secondary outcome, change in serum 3a-androstanediol Glucuronide, measured from baseline to 2-4 days post-medication during an alcohol session. Across the conditions, the broad pattern differs by medication condition: the posted values for dutasteride are lower than those for the placebo medication conditions. The ethanol conditions also appear within the sponsor’s reported direct comparison for this biomarker. This general pattern is relevant to the study’s proposed neuroactive-steroid rationale, but it does not establish a biological mechanism or explain the primary breath alcohol findings.

The sponsor reports a two-sided t-test for the biomarker comparing the two ethanol conditions, with p < 0.001. The recalculated secondary contrasts use the placebo medication and placebo alcohol condition as the reference and are not equivalent to that reported test. The serum measure and breath alcohol are distinct outcomes, on different scales and at different timeframes. A pattern in the biomarker does not itself determine the interpretation of breath alcohol, and the primary outcome should not be treated as a surrogate for the secondary measure.

The secondary findings deserve less weight than a prespecified primary result because multiple outcomes and contrasts create more opportunities for apparently notable results. The registry does not report a multiplicity adjustment, a complete prespecified endpoint list, or a detailed analysis plan. The displayed secondary p-values are unadjusted, and the record does not clarify whether other analyses were conducted but not posted. They are best treated as exploratory evidence within this study rather than definitive confirmation of the proposed pathway.

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

• Change in Serum 3a-androstanediol Glucuronide

9. Statistical Interpretation

10. Statistical Methods Explained

What does the mean represent?

Breath alcohol is a continuous outcome, and the registry reports the mean for each study condition. A mean gives a single summary of the observed values in a group, but it does not show every participant’s measurement or the full shape of the distribution. The standard errors attached to the posted means describe uncertainty in estimating each condition’s mean. They are not the same as standard deviations, which describe the spread of individual observations.

What is a mean difference?

A mean difference compares the average outcome in one condition with the average in a reference condition. The recalculated comparisons use Placebo Medication + Placebo Alcohol as the reference. For example, the contrast involving placebo medication with ethanol answers a different question from the contrast involving dutasteride with ethanol, even though both use the same reference condition. A contrast between the two ethanol conditions is a third question. The registry’s reported t-test addresses that direct ethanol-condition comparison, whereas the recalculated contrasts are against the reference condition.

What does a confidence interval convey?

A confidence interval gives a range of values compatible with the estimate under the calculation’s assumptions. Its width communicates precision: a narrower interval indicates a more precise estimate than a wider interval, while still leaving uncertainty. The interval does not guarantee that the true value lies within its bounds, nor does it capture every possible source of bias or design uncertainty. The recalculated intervals are unadjusted and use posted summary statistics. The registry does not report confidence intervals for the sponsor’s direct comparison of the ethanol conditions.

How should p-values be read?

A p-value describes how unusual data at least as discrepant as those observed would be under a specified null model. It is not the probability that the null hypothesis is true, and it does not measure clinical importance. The recalculated p-values are unadjusted. The sponsor’s reported p-value comes from a two-sided t-test for the comparison between the two ethanol conditions. These p-values arise from different contrasts and should not be treated as interchangeable. The registry does not report enough methodological detail to determine whether the sponsor’s test accounts for the within-subject structure.

Why does the within-subject design matter?

In a within-subject design, the same participant experiences multiple conditions. A comparison may therefore depend on how outcomes within a participant are related, as well as on the average outcome in each condition. A paired or repeated-measures analysis can use this relationship, while an analysis that treats conditions as independent may not. The summary-statistic recalculations presented with the outcome use the posted means and standard errors and are described as simple unadjusted recalculations. They should not be assumed to reproduce a participant-level paired analysis. The exact statistical model used by the sponsor for the primary endpoint is not reported in the registry record.

The factorial design combines medication and alcohol conditions. A formal factorial analysis could examine medication effects, alcohol effects, and whether the effect of one factor varies by the other. The available registry results do not report a factorial model or an interaction estimate. The sponsor-reported t-test comparing the two ethanol conditions is a particular contrast; it should not be read as a complete factorial analysis or as evidence that all interactions were tested.

Why can over-dispersion matter?

General statistical explanation: over-dispersion occurs when observed variability is greater than a model expects. It can make standard errors too small if the model fails to account for the extra variation, which in turn can make confidence intervals too narrow and p-values too small. It is especially relevant to outcomes modeled as counts or proportions, but the broader lesson is that uncertainty calculations depend on whether the model reflects the data structure. The posted primary outcome is continuous, and the registry does not report an over-dispersion assessment or a model diagnostic. No conclusion about over-dispersion can be drawn from the record.

What do the recalculations add?

The recalculations provide unadjusted comparisons from posted summary statistics. They make the mean contrasts and their uncertainty easier to interpret, but they do not substitute for the sponsor’s analysis or for an analysis using participant-level data. They cannot account for unreported covariate adjustment, repeated-measures correlation, deviations from the protocol, or missing-data procedures. Since no discrepancy is flagged, the registry does not identify a conflict between the reported and recalculated results; however, the sponsor’s own analysis remains the relevant reported analysis for its specific contrast.

The trial’s original sample-size justification, prespecified analysis plan, interim analyses, and multiplicity procedure are not reported in the registry record. General explanations of statistical concepts do not fill those gaps. In particular, the displayed planning illustrations are based on the observed effects and stated assumptions; they are not evidence about the trial’s original design calculations or its achieved power.

11. Confidence Intervals and Precision

The confidence intervals for the recalculated primary contrasts quantify uncertainty around the estimated mean differences using posted summary statistics. For placebo medication with ethanol versus placebo medication with placebo alcohol, the estimate is 0.074 gr/dL and the 95% confidence interval ranges from 0.068 to 0.080. For dutasteride with placebo alcohol versus the reference condition, the estimate is 0.0 gr/dL and the interval ranges from -0.001 to 0.001. For dutasteride with ethanol versus the reference condition, the estimate is 0.07 gr/dL and the interval ranges from 0.064 to 0.076.

These intervals are unadjusted recalculations, not sponsor-reported intervals. Their interpretation depends on the calculation assumptions, including how the within-subject design is represented. The registry does not report the sponsor’s confidence interval for its direct comparison between the two ethanol conditions. Its p-value of 0.28 therefore cannot be paired with a reported sponsor estimate and interval in the registry record.

Precision is not the same as freedom from bias. An interval can be relatively narrow while questions remain about participant selection, measurement, model choice, or handling of missing data. The outcome record lists 70 participants per condition and describes an analysis population that completed all four arms, with three subjects removed during data cleaning. Details of any other exclusions or missing observations are not reported in the registry record. Accordingly, the intervals characterize uncertainty under the calculation, rather than all uncertainty about the study’s design and conduct.

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 based on the observed effect, not the trial’s original design justification. For these continuous outcomes, the displayed figures are participants per arm, not total events. The calculation assumptions are a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. The effect to detect is the stated fraction of the observed effect. The trial’s own original design assumptions are not reported in the registry record.

The 80% power column gives the illustrative number of participants per arm under those assumptions for an 80% chance of detecting the specified effect, if that effect and model assumptions were correct. The 90% power column gives the corresponding illustrative number per arm for a 90% chance. These are planning quantities, not results about the participants who took part. For example, the primary comparison of placebo medication with ethanol lists 1 participant per arm for 80% power and 2 participants per arm for 90% power when planning for 100% of the observed effect. The illustration increases to 4 and 5 participants per arm when planning for 50% of that observed effect.

Smaller effects require more participants to detect under the same assumptions, which is why the planning figures increase as the effect fraction decreases in the displayed rows. These values should be read cautiously, particularly where observed effects are large or the estimated variability is small: they are sensitive to those observed quantities and the simplified planning assumptions. They do not describe the actual power achieved by the study. This is not observed power.

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 = 0.074; pooled standard deviation σ = 0.018.
2. n = 2 × (1.96 + 0.842)² × 0.018² / 0.074² = 0.9, rounded up to 1 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 <- 0.074                      # observed difference in means
sigma <- 0.0179053064760143                     # 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(734656)
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 = 1 (80% power), 2 (90% power)
 75% of observed effect: n per arm = 2 (80% power), 3 (90% power)
 50% of observed effect: n per arm = 4 (80% power), 5 (90% power)
power.t.test: 2.4 
Simulated power at n = 1 per arm: NA (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. Analyses Reported by the Sponsor

The registry reports a two-sided t-test comparing the two ethanol conditions for breath alcohol, with p = 0.28. This is the sponsor’s reported analysis for that contrast. It compares placebo medication with ethanol against dutasteride with ethanol. The registry does not provide an estimate or confidence interval for this direct comparison, nor does it describe the test’s model details.

The recalculated primary comparisons use the placebo medication and placebo alcohol condition as their reference. They describe each of the other conditions relative to that reference, rather than comparing the ethanol conditions directly. The recalculated values therefore answer different questions from the sponsor-reported t-test and should not be used as a substitute for it. The posted condition means are 0.075 gr/dL for placebo medication with ethanol and 0.071 gr/dL for dutasteride with ethanol, while the reported sponsor p-value concerns their direct comparison.

The registry does not flag a discrepancy between sponsor and recalculated results. Still, where a sponsor analysis is reported, it takes precedence for its stated contrast; a simple unadjusted recalculation from summary statistics cannot reproduce an analysis that may use participant-level data or a design-specific model. The registry does not report whether the sponsor’s t-test accounts for the within-subject structure.

15. Multiplicity, Missing Data and Analysis Populations

The record includes a primary endpoint, a secondary outcome, multiple study conditions, and several possible contrasts. Testing multiple outcomes or contrasts increases the chance that at least one result appears statistically notable even when there is no corresponding underlying effect. Statistical procedures can account for this multiplicity, but they often involve trade-offs, such as wider intervals or more conservative thresholds. The registry does not report whether any adjustment was planned or used.

The primary endpoint description identifies subjects who completed all four arms and states that three subjects were removed during data cleaning because of pharmacy errors or a protocol deviation. The outcome data list 70 participants per condition, while overall enrollment is 94. The registry does not provide a complete participant flow account that reconciles these figures, nor does it report how any other missing measurements were handled. The treatment of incomplete four-session records, if any, is not reported in the registry record.

The target population for each analysis is therefore not fully characterized by the posted outcome summaries. The record does not state whether analyses followed an intention-to-treat principle, a per-protocol definition, or another analysis population beyond the completion description. It also does not report whether participants with incomplete sessions were included in any analysis using available measurements.

The registry gives a sponsor-reported two-sided t-test for the ethanol-condition comparison, but does not provide a detailed model specification for repeated observations. It does not report prespecified covariates, a formal factorial analysis, interim analyses, or the statistical plan used to select contrasts. These omissions limit the ability to assess whether the reported analyses match a predeclared plan and whether uncertainty estimates account for the design. They do not imply that a particular procedure was or was not used; those details are not reported in the registry record.

16. 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 1% reporting threshold.

Most frequent non-serious adverse event terms

The adverse-event record reports no serious adverse events among participants at risk in any of the four conditions. The serious-event counts are 0 among 78 at risk for Placebo Medication + Placebo Alcohol, 0 among 81 for Placebo Medication + 0.8 gr/kg Ethanol, 0 among 83 for 4 mg Dutasteride + Placebo Alcohol, and 0 among 78 for 4 mg Dutasteride + 0.8 mg/kg Ethanol. These are the registry’s recorded serious-event figures, not evidence that the possibility of serious events can be excluded.

The registry also lists other, non-serious adverse events in each condition, but those counts are not serious-event counts. Death counts are not reported in the registry record. The record does not provide event descriptions in the serious-event category because no serious adverse-event terms are listed. The numbers at risk differ across conditions, and the registry does not explain the reasons for those denominators or provide a participant-level safety analysis.

Safety counts should be interpreted in the context of how adverse events were collected and classified. Those procedures, along with exposure duration for safety assessment, are not reported in the registry record. The figures describe what was posted for these study conditions and should not be generalized beyond the recorded study population and reporting.

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

The results establish that the posted mean breath alcohol levels differed between the placebo alcohol condition and the two ethanol conditions in the unadjusted comparisons. The estimates and confidence intervals quantify those contrasts using the posted summary statistics. The sponsor-reported direct comparison between the ethanol conditions has p = 0.28, but the registry does not report its estimated difference or confidence interval. These results are about breath alcohol at the stated measurement time, not about subjective response or clinical outcomes.

The findings do not determine whether dutasteride changes alcohol’s subjective effects, whether a neuroactive-steroid pathway mediates any effect, or whether the study’s proposed biological explanation is correct. The secondary serum measure provides a separate set of observations, but a secondary biomarker result does not by itself demonstrate mediation. The record does not report a formal mediation analysis or a factorial interaction estimate.

The study’s within-subject design can provide comparisons within the same participants, but interpretation depends on the analysis used to account for repeated observations. The registry does not report the complete statistical plan or model details. The simple recalculations are not a substitute for participant-level analysis and do not account for any unreported adjustments or study-specific procedures.

Likewise, the posted adverse-event counts describe recorded events during the study conditions, but the record does not give enough detail about safety ascertainment to support broad claims about risk. The registry reports no serious adverse events in the listed groups; that observation should not be read as proof that serious events cannot occur. Overall, the appropriate conclusion is bounded: the posted outcomes describe condition-specific measurements in this study, with uncertainty and reporting limitations that matter when interpreting comparisons or extending them to other settings.

18. Why This Trial Matters Statistically

Statistically, this trial is useful for showing why the contrast being tested must be stated precisely. Comparisons of each condition with placebo medication and placebo alcohol answer different questions from a direct comparison between the two ethanol conditions. The sponsor’s reported t-test addresses the latter, while the unadjusted recalculations describe the former. Reading those results as if they were a single test would confuse distinct estimands.

The four-condition, within-subject design also highlights the importance of modeling dependence among measurements from the same participant. Marginal means and standard errors are informative summaries, but they do not expose the correlations that a repeated-measures analysis may use. Because the registry does not report the full model, the posted summaries cannot establish whether those correlations were incorporated into the sponsor’s primary analysis.

The record further illustrates that a striking contrast and a precise estimate do not settle questions about design, multiplicity, missing data, or generalizability. The planning illustrations based on observed effects can help readers understand how assumed effect size relates to sample requirements, but they are not retrospective evidence of power and do not recover the original rationale for enrollment. Such distinctions are central to sound interpretation of registry results.

Finally, the study links a physiological endpoint with a secondary serum measure motivated by a biological hypothesis. Keeping those outcomes conceptually separate prevents a biomarker pattern from being mistaken for proof of mechanism or clinical consequence. A careful statistical reading can recognize what the measurements suggest while remaining explicit about what the registry does not document.

19. 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 analysis population and participant flow are incompletely described. Overall enrollment is 94, while the primary outcome description refers to 70 subjects completing each condition and notes three removals during data cleaning. The registry does not provide a complete accounting of all participants, incomplete sessions, or missing outcome measurements. This limits assessment of how the analyzed participants relate to the enrolled population.

The within-subject design makes the treatment of repeated observations important, but the registry does not report the sponsor’s primary-endpoint model or whether it accounts for within-participant correlation. The reported sponsor t-test is identified as two-sided, yet further details are absent. The recalculated comparisons use posted group summaries and are unadjusted; they cannot reproduce unreported participant-level modeling.

The record does not report a full statistical analysis plan, multiplicity handling, interim analyses, or a formal factorial interaction analysis. It also does not provide a sponsor estimate and confidence interval for the direct comparison of the ethanol conditions. Those omissions constrain interpretation of the direct test and the extent to which the posted analyses can be assessed against prespecified methods.

Eligibility criteria, participant characteristics, detailed measurement procedures, and broader safety ascertainment methods are not reported in the registry record. The posted endpoint is a specific breath alcohol measurement 40 minutes after beginning the drink, and the secondary measure has a different timeframe. Findings from these measures should not be generalized to unmeasured subjective or clinical outcomes. These limitations describe what the record does not document; they do not establish that any particular procedure was absent.

20. Related Tutorials

21. Related Statistical Calculators

22. Sources

• ClinicalTrials.gov: Pharmacogenetics of Alcohol: Treatment Implications — NCT00734656.
• Registry record last updated 2012-03-28; status last verified 2012-03.
• Results first posted 2012-03-27.
• 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.

23. Record Summary

NCT00734656, titled Pharmacogenetics of Alcohol: Treatment Implications, is a completed randomized, quadruple-masked, factorial study sponsored by UConn Health. The registry describes a four-session within-subjects design pairing dutasteride or placebo medication with ethanol or placebo alcohol. Enrollment was 94, and the primary outcome description states that 70 subjects completed each condition after three subjects were removed during data cleaning.

The primary endpoint was mean breath alcohol in gr/dL 40 minutes after beginning the drink. The sponsor reports a two-sided t-test comparing the two ethanol conditions with p = 0.28. Unadjusted recalculations compare each non-reference condition with placebo medication and placebo alcohol. One secondary outcome, change in serum 3a-androstanediol Glucuronide, is also posted. The registry reports no serious adverse events in the listed conditions. The original statistical plan, multiplicity procedure, and full handling of missing data are not reported in the registry record.