Acamprosate vs. Placebo in Bipolar Alcoholics: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Acamprosate vs. Placebo in Bipolar Alcoholics (NCT00466661), a randomized phase 4 study. The primary outcome is Time to First Drink (Days).

2. Clinical Question

This is a question about an outcome’s timing, rather than simply whether drinking occurred. The posted group summaries give mean days to first drink, but the record does not provide a complete confidence interval or an estimable two-arm comparison for this primary endpoint. As a result, the group values can be described, but the record does not establish the precision of the between-group difference or provide a basis for a formal conclusion about that comparison.

3. Trial Design

The registry describes a Phase 4, randomized, double-blind, parallel-group, placebo-controlled outpatient trial. Its official title identifies alcohol-dependent individuals with comorbid bipolar disorder. The stated study purpose was to assess acamprosate in this population while participants were also receiving mood-stabilizing medication. The registry lists Medical University of South Carolina as the sponsor and the overall enrollment as 33.

Participants were assigned to acamprosate or matching placebo. Acamprosate is listed at 666 mg p.o. TID; the placebo description is “Matching placebo.” The record labels the study double-blind, but it does not report further details about how masking was maintained or assessed. Allocation is recorded as randomized, though the randomization sequence, concealment procedures, and allocation ratio are not reported in the registry record.

The primary outcome was time to first drink, measured in days after randomization through self-report over eight weeks. The primary endpoint population was a modified intention-to-treat sample consisting of randomized participants who returned for at least one visit after randomization. The registry gives this population as 30 participants, with 14 in the acamprosate group and 16 in the placebo group for the primary outcome. The distinction between overall enrollment and the endpoint analysis population matters: the posted endpoint values describe the analyzed participants, not necessarily every randomized participant.

Eight secondary outcomes are named in the record, covering alcohol use, laboratory measures, and mood or clinical ratings. Results are posted for three of these secondary outcomes. No secondary analyses are listed as reported analyses, and there are no analyses listed for the remaining measures. Details such as the planned sample-size justification, full statistical analysis plan, and any interim analyses are not reported in the registry record. These omissions limit what can be inferred about how the trial was planned and how the posted analyses relate to its original 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.

Before assignment: Two week lead-in (Screening, Baseline)

Overall Study

Analysis population for the primary outcome, as defined in the registry: Sample analyzed is a modified intention to treat (mITT) sample of all randomized subjects who returned for at least one visit post-randomization (total mITT n=30).

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 Time to First Drink (Days), measured over 8 weeks.

The primary endpoint is the number of days from randomization until a participant reports consuming the first alcoholic beverage. It is measured over eight weeks and uses self-report, as specified in the registry. Unlike a simple count of drinking days, a time-to-first-drink measure records when the first reported drink occurs. It can therefore summarize the timing of an event during follow-up.

As a general statistical concept, an endpoint based on time to an event can be useful when the timing itself is clinically or scientifically relevant. It may also require methods that account for participants who have not experienced the event by the end of observation. The registry record does not describe whether anyone had no reported first drink during the assessment period, how such observations were handled, or what statistical method was used for the primary comparison. Those details are not reported in the registry record.

The posted results label the measure as a mean and give group values of 16.5 days for acamprosate and 8.0 days for placebo. A 95% confidence interval is indicated as the dispersion measure, but only the lower bounds, 6.0 and 2.0 respectively, are posted; upper bounds are not shown. No two-arm comparison is estimable from the posted primary values. Consequently, the record allows a descriptive reading of the group summaries but not a complete account of the uncertainty around them or a formal comparison based on the primary endpoint.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The registry classifies the primary outcome as continuous and reports it as a mean, with a 95% confidence interval as the dispersion measure. The posted values are 16.5 days for acamprosate and 8.0 days for placebo. The primary results also show lower interval limits of 6.0 and 2.0, respectively, but do not show upper limits. The registry’s result record does not include an estimable comparison between the two groups for this endpoint. It would therefore be inappropriate to reconstruct a primary confidence interval or p-value from the information shown.

What does a confidence interval describe?

As a general statistical explanation, a confidence interval expresses uncertainty around an estimate under the assumptions of the method used to calculate it. For a difference between groups, an interval indicates values compatible with the data and method; it is not a guarantee that the true value lies within the particular interval. The primary endpoint’s incomplete interval reporting prevents a full reading of its uncertainty. For the secondary outcomes, the posted intervals are for mean differences and can be read alongside the estimates, while keeping in mind that the record does not report a multiplicity adjustment.

How should the secondary comparisons be read?

The secondary outcome summaries report group means and standard deviations. The accompanying comparisons report mean differences, confidence limits, and unadjusted p-values. These comparisons are described as unadjusted re-calculations from posted summary statistics. They are not a replacement for analyses that might have been prespecified or conducted by the investigators. No reported analyses are listed in the registry record, and the details of any sponsor analysis are not reported in the registry record.

An unadjusted p-value summarizes how unusual a result at least as inconsistent with a null hypothesis would be under the assumptions of the test. It does not measure the probability that the null hypothesis is true, and it does not measure the size or clinical importance of a difference. Here, the p-values are unadjusted, so they do not account for testing several secondary outcomes. General statistical reasoning cautions that examining multiple outcomes can make apparently notable results more likely by chance. That is one reason these secondary comparisons carry less weight than a prespecified, fully reported primary analysis.

Why does dispersion matter?

For continuous outcomes, the standard deviation describes the spread of participant values around a group mean. As a general concept, greater variability makes a mean difference harder to estimate precisely when other design features are similar. The reported secondary outcomes show substantial variation within groups, and the group sizes are modest. The confidence intervals consequently provide important context: they show that the point estimates alone do not capture the uncertainty in the comparisons.

Over-dispersion is a general concern when data vary more than a statistical model assumes. It can make standard errors and confidence intervals unreliable if the model does not account for that extra variability. The registry record does not report a model for these outcomes or indicate whether over-dispersion was assessed, so no claim can be made about that issue in this trial. The secondary outcomes are posted as means and standard deviations, and the comparisons are simple unadjusted calculations from those summaries.

What do the planning illustrations mean?

The sample-size illustrations use the observed effect, or stated fractions of it, together with assumptions of a two-sided significance level of 0.05, equal allocation between groups, and a normal approximation. Their purpose is to illustrate the participants per arm that would be required under those assumptions to target 80% or 90% power for the chosen effect. They are not the trial’s actual sample-size justification, and that justification is not reported in the registry record. They are not observed power and should not be used to retroactively characterize the trial as adequately or inadequately powered. Since they depend on the observed effect, they are especially sensitive to the imprecision of estimates from small groups.

More generally, power is a planning quantity: it describes the chance a specified design would detect a specified effect under assumed conditions. It cannot be inferred simply from a nonsignificant p-value. The registry does not report the trial’s planned target effect, sample-size calculation, or full analysis plan. It also does not provide a primary comparison from which an analogous planning illustration could be interpreted. Accordingly, the posted planning section is an illustration only, not evidence of what the trial was designed to detect.

10. Multiplicity, Missing Data and Analysis Populations

The registry names eight secondary outcomes, while results are posted for three. It lists no reported analyses. The record does not describe a multiplicity adjustment, a hierarchy among secondary outcomes, or a prespecified approach for interpreting several outcomes together. These features are important because each additional comparison can add another opportunity for a result to appear notable by chance. This is a general statistical concern, not a claim that any particular posted result is spurious.

The comparisons available for the secondary measures are unadjusted. Their p-values therefore should not be read as accounting for the full set of outcomes named in the registry. Nor does the record make clear whether other analyses, outcome definitions, or time points were planned. Those details are not reported in the registry record, so readers cannot assess whether the posted subset reflects the original analysis plan or how selective reporting may have affected the available picture.

For the primary endpoint, the registry identifies a modified intention-to-treat sample consisting of all randomized subjects who returned for at least one post-randomization visit. The stated total for this group is 30. The overall enrollment is 33. The primary outcome results give group sizes of 14 and 16. The registry does not explain the reasons participants were not included in the modified intention-to-treat group, report their outcome data, or describe a missing-data method. It also does not report whether analyses used other populations.

As a general concept, missing follow-up data can affect an estimate if the likelihood of missingness is related to outcomes or differs between groups. The record does not provide enough information to assess that possibility. It does not report a sensitivity analysis, imputation procedure, or assumptions about missing outcomes. The same limitation applies to the secondary outcomes, for which the posted group summaries include 14 and 16 participants but no detailed missing-data account.

The trial’s protocol-level statistical analysis plan, interim analyses, and sample-size justification are not reported in the registry record. Their absence from the posted record should not be taken to mean that they did not exist; it means their details cannot be evaluated. These reporting gaps, together with the incomplete primary endpoint comparison and unadjusted secondary tests, call for cautious interpretation.

11. Safety Results

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

Deaths

Other (non-serious) adverse events

Participants with at least one non-serious event at or above the registry's 5% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: hospitalization (14.3%), anaphylactoid skin reaction (7.1%), seizure (6.2%).

The registry reports serious adverse event counts for 14 participants at risk in the acamprosate group and 16 in the placebo group. Three serious adverse events are recorded for each group. It also reports other adverse event counts of 6 for acamprosate and 8 for placebo. No deaths are reported for either group.

These counts describe what is posted for this trial; they do not by themselves establish that one group had greater or lesser risk. The groups are small, and the registry does not provide a statistical comparison of adverse events, event-specific follow-up time, or detailed definitions and adjudication procedures. The record also lists serious-event terms and percentages: hospitalization, 14.3%; anaphylactoid skin reaction, 7.1%; and seizure, 6.2%. The registry listing does not identify an arm for each of these terms in the information shown, so they should not be attributed to a particular group.

As a general statistical point, safety counts can be difficult to interpret when events are uncommon and the number of participants is limited. Counts of participants with events, event counts, and percentages can represent different quantities; the record does not give enough detail to resolve every such distinction for the listed terms. These figures are a descriptive summary, not a complete comparative safety analysis and not a basis for individual medical decisions.

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

The posted data permit a limited description of this randomized comparison. For the primary endpoint, the mean number of days to first self-reported drink was 16.5 in the acamprosate group and 8.0 in the placebo group. However, the record does not provide complete confidence intervals or an estimable comparison between groups. The summaries therefore do not establish the size or precision of a primary treatment contrast.

For the secondary measures with results posted, the unadjusted comparisons are accompanied by intervals that include zero. This means the available estimates remain compatible with more than one direction of effect. It does not prove that the treatments are equivalent, nor does it demonstrate that there is no meaningful difference. A claim of equivalence would require an appropriate design and analysis, which are not reported in the registry record.

The record also does not provide results for all named secondary outcomes, a multiplicity procedure, or a detailed account of missing observations. Consequently, the available findings cannot be treated as a complete picture of all measured outcomes. The planning illustrations are hypothetical sample-size calculations based on the observed effects and stated assumptions; they do not reveal the trial’s actual design rationale and are not observed power.

As a general interpretation principle, an estimate should be considered together with its uncertainty, the population analyzed, and the analysis plan. Here, small analyzed groups and incomplete primary reporting limit precision and prevent a confident reading of the primary comparison. The results are not a basis for medical advice, and this statistical summary does not determine what treatment is appropriate for an individual. It describes only what can be read from the registry record.

The trial’s question is relevant to a population with two co-occurring conditions, but relevance does not resolve the limitations of the posted data. The registry describes a double-blind randomized trial and provides some outcome and safety summaries. It does not report enough information to make a definitive claim about the comparative effect on the primary endpoint or to establish a comprehensive interpretation across the secondary outcomes.

13. Why This Trial Matters Statistically

This trial illustrates how a randomized design and a clinically focused question can still yield a record that is difficult to interpret statistically when key analysis details are missing. Randomization and double-blinding are important design features, but they do not replace complete reporting of the endpoint comparison, uncertainty, analysis population, and handling of missing data. The registry describes those design features while leaving several of the details needed to evaluate the primary result unavailable.

The primary endpoint is intuitively understandable: it records the timing of a first reported drink. Yet the posted record gives only group means and lower confidence limits, with no upper limits or estimable comparison. The case shows why readers need more than point estimates to understand a result. A group summary may be visible while the uncertainty around the difference between groups remains unresolved.

The secondary outcomes further demonstrate why analysis context matters. Several kinds of alcohol-use, laboratory, and mood measures are named, but results are posted for only some of them. Their comparisons are unadjusted, and the record does not describe a multiplicity plan. Without that context, secondary results are best treated as descriptive rather than confirmatory. The sample-size illustrations also make an important distinction: calculations based on observed effects can inform hypothetical future planning, but they are not observed power and cannot explain the trial’s original design.

Finally, the record distinguishes overall enrollment from the modified intention-to-treat population used for the primary endpoint, but does not explain the missing follow-up information. That gap matters because analysis populations and missing data affect how directly a result represents the randomized comparison. Together, these features make this trial a useful example of why statistical interpretation depends not only on results, but also on transparent reporting of how those results were obtained.

14. 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 constrain interpretation. The primary endpoint’s two group summaries are reported, but the upper bounds of the stated 95% confidence intervals are absent and a two-arm comparison is not estimable from the posted values. As a result, the precision of the primary contrast cannot be assessed from the record. The primary analysis method and any handling of participants without a reported first drink are not reported in the registry record.

The primary outcome is based on self-report. The record does not describe how reports were collected, verified, or reconciled with other measures. This does not establish that self-report was inaccurate, but it means readers cannot evaluate those procedures from the posted information. The registry also does not report detailed follow-up completeness or reasons for participants not appearing in the modified intention-to-treat population.

The analyzed groups are small, and the registry gives no detailed sample-size justification. The listed planning illustrations are not the trial’s original design calculation. The record does not report the planned target effect, a full statistical analysis plan, interim analyses, or a method for addressing multiple secondary outcomes. These omissions make it difficult to judge whether analyses were prespecified and how much weight to place on the available secondary comparisons.

Only three of the eight named secondary outcomes have posted results. The record does not explain the absence of results for the others or provide a full analysis population and missing-data account for each measure. Safety information is also summary-level: the registry provides counts and selected serious-event terms but not a full comparative analysis or the detailed context needed to assess event rates. These limitations concern the completeness of registry reporting; they do not establish that additional analyses or documentation were not available elsewhere.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: Acamprosate vs. Placebo in Bipolar Alcoholics — NCT00466661.
• Registry record last updated 2018-12-14; status last verified 2018-12.
• Results first posted 2018-12-14.
• 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.

18. Record Summary

The registry describes a completed Phase 4, randomized, double-blind, parallel-group trial of acamprosate versus matching placebo in alcohol-dependent individuals with bipolar disorder. Overall enrollment was 33, and the study was sponsored by Medical University of South Carolina. Participants were described as also receiving mood-stabilizing medication.

The primary outcome was time to first self-reported drink over eight weeks. Its modified intention-to-treat population included 30 participants. Group means were posted, but the record does not provide complete confidence intervals or an estimable primary comparison. Results are posted for three of eight named secondary outcomes; their comparisons are unadjusted. The registry also reports adverse event counts and no deaths in either group.

Important analysis details, including the sample-size justification, full statistical plan, multiplicity approach, and missing-data methods, are not reported in the registry record. The available summaries support cautious description, not a definitive conclusion about the primary comparison or the full set of outcomes.