Vigabatrin for Treatment of Cocaine Dependence: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Vigabatrin for Treatment of Cocaine Dependence (NCT00611130), a randomized phase 2 study. The primary outcome is Proportion of Subjects in Each Treatment Group Abstinent During the Last 2 Weeks of Treatment..

2. Clinical Question

This question focuses on abstinence near the end of the treatment period, not abstinence throughout the entire trial. The outcome description defines abstinence as not using cocaine during both Weeks 11 and 12. The registry's brief summary states the intended comparison as the proportion of participants abstinent in the last 2 weeks of treatment, while the outcome record presents the numbers of abstinent participants and the group sizes. These are related but distinct ways of expressing the outcome: counts describe how many participants met the definition, whereas proportions place those counts in relation to each treatment group.

The registry does not report a separate primary outcome measure for sustained abstinence beyond that period. It also does not report secondary outcome results. Accordingly, interpretation of the trial's primary question rests on the posted counts and the unadjusted comparison, rather than on a broader set of outcomes or a longer-term follow-up result.

3. Trial Design

This was a randomized, quadruple-masked, parallel-group phase II study. The registry lists an enrollment of 186 and describes the condition as cocaine dependence. The study status is completed. The start is recorded as 2008-01, and primary completion as 2009-05. Catalyst Pharmaceuticals, Inc. is listed as the sponsor.

Participants were assigned to CPP-109 vigabatrin tablets, 500 mg, or matching placebo tablets. The arm descriptions specify three tablets twice daily for 9 weeks in both groups. The registry identifies the comparison as vigabatrin versus placebo and uses the placebo arm as the control. The arm records report 92 participants in the vigabatrin group and 94 in the placebo group.

Randomization is intended to distribute participant characteristics across treatment groups, while masking is intended to limit the influence of treatment knowledge on participants' experiences, care, and outcome assessment. Those are general features of trial design; the registry does not specify exactly who was masked or how allocation was generated or concealed. The record also does not describe stratification, adherence monitoring, or permitted co-interventions.

The parallel-group structure means that participants were assigned to one treatment group rather than crossing between treatments. The primary outcome was defined as abstinence from cocaine during Weeks 11 and 12 of the treatment phase and recorded at Week 13. The trial's own sample-size rationale and original statistical design assumptions are not reported in the registry record. The enrollment figure should therefore be read as the recorded enrollment, not as evidence about how the trial's target was selected or whether its design was expected to detect a particular effect.

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: After Informed Consent obtained, subjects entered a 2-4 week Screening/Baseline Phase to determine whether all Inclusion/Exclusion Criteria were met. Randomization strata included gender, primary method of cocaine administration (snort or intravenous/smoke) \& use in last 30 days (≤18 days or \>18 days)

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: intent-to-treat

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 Proportion of Subjects in Each Treatment Group Abstinent During the Last 2 Weeks of Treatment., measured over Week 13.

The primary endpoint is the number of participants in each treatment group who were abstinent from cocaine during Weeks 11 and 12 of the treatment phase. The registry lists the assessment timeframe as Week 13 and identifies the outcome population as intent-to-treat. The brief summary frames the objective in terms of a larger proportion abstinent in the vigabatrin group than in the placebo group, while the outcome record reports counts.

An abstinence endpoint provides a direct, clinically interpretable summary of whether participants met a specified behavioral criterion during a defined period. Using a fixed window makes the comparison more specific than a general description of treatment response: participants had to be abstinent during the final 2 weeks of treatment, as defined in the record. It does not, by itself, describe use patterns outside those weeks, the duration of abstinence before that window, or outcomes after treatment.

Counts alone do not show the relative frequency of the endpoint without the corresponding group sizes. The registry reports both counts and group sizes, and the unadjusted risk-difference comparison expresses the contrast between groups on a proportion scale. The record does not explain how missing outcome observations were handled or provide a more detailed operational definition of abstinence, such as the assessment procedure used to establish it. Those details are not reported in the registry record.

7. Results Posted in the Registry

• CPP-109 Vigabatrin Tablets, 500 mg: 7.6% (7/92) versus 5.3% (5/94) on control; risk difference 2.3 percentage points (95% CI -4.8 to 9.4), risk ratio 1.43 (0.47–4.35); unadjusted two-proportion z-test p = 0.525.

8. Statistical Interpretation

9. Statistical Methods Explained

The primary outcome is binary at the participant level: the record counts participants who were abstinent from cocaine during Weeks 11 and 12. General explanation: for a binary endpoint, an event count is interpreted in relation to the number of participants assessed in that group. The proportion is the natural scale for describing the frequency of the event, and a risk difference compares those proportions by subtracting one group frequency from the other. The posted comparison is an unadjusted risk difference.

What does an unadjusted comparison mean?

General explanation: an unadjusted comparison uses the observed outcome data without accounting for additional covariates in a regression or other adjusted model. It gives a straightforward contrast between the randomized groups. Randomization supports a treatment-group comparison in principle, but the registry does not state that an adjusted analysis was planned or performed. It reports 0 analyses and does not list an analysis model beyond the posted unadjusted comparison. Details of the statistical analysis plan are not reported in the registry record.

The comparison reports a risk-difference estimate of 0.023 and an unadjusted p-value of 0.525. The result is based on 7 abstinent participants among 92 in the vigabatrin group and 5 among 94 in the placebo group. These are the registry's reported counts and denominators. The record does not report a separate adjusted effect estimate or a model-based treatment contrast.

How should the p-value be understood?

General explanation: a p-value measures how compatible the observed data are with a specified statistical model in which the groups have no difference, under the assumptions of the test. It does not give the probability that the no-difference model is true, and it does not measure the size or clinical importance of an effect. The reported p-value is 0.525. The registry does not report the test used to obtain it, so the calculation method and any test-specific assumptions are not available from the record.

A p-value should be considered alongside the estimated effect and its interval. In this trial, the risk-difference interval extends from -0.048 to 0.094, covering zero. General explanation: an interval describes a range of effect values compatible with the data and the method used to construct it. Its width conveys uncertainty; it does not mean that every value in the interval is equally plausible. The confidence level and interval-construction method are not reported in the registry record, so the interval should not be assigned a confidence level that the record does not state.

Why use a risk difference?

General explanation: the risk difference describes an absolute contrast on the outcome-proportion scale. It can be read as the change in the proportion meeting the endpoint between groups, rather than as a relative comparison. The registry also lists relative measures, but their intervals are not needed to describe the primary posted contrast. Relative measures can be sensitive to the baseline frequency of an endpoint and may give a different impression of magnitude than an absolute difference. Neither scale, alone, establishes causality or clinical importance; interpretation depends on study design, endpoint meaning, and uncertainty.

For an outcome that is uncommon, a small number of participants can determine the group counts. General explanation: this makes estimates sensitive to the observed events and can result in substantial uncertainty. The posted interval is broad enough to include zero and values on either side. The record does not report a small-sample correction, continuity correction, or a method for estimating the interval.

What does intent-to-treat indicate?

The primary population is labeled intent-to-treat. General explanation: an intent-to-treat approach generally aims to analyze participants according to their randomized assignment, regardless of treatment adherence. The registry record does not define how that principle was applied to participants with missing outcome information, treatment discontinuation, or other departures from assigned treatment. The group sizes and endpoint counts are reported, but the handling of missing or unavailable outcomes is not described.

The registry lists 0 analyses and does not post secondary outcome results. This limits what can be said about modeling choices, multiplicity adjustments, sensitivity analyses, or subgroup analyses. Those procedures are not reported in the registry record. The absence of a posted analysis does not establish that a particular analysis was or was not conducted outside the record.

10. Confidence Intervals and Precision

The reported risk-difference estimate is 0.023, and its interval runs from -0.048 to 0.094. The interval includes zero, which means the posted estimate does not rule out no difference under the interval method used. It also includes values on either side of zero, so uncertainty concerns both the direction and size of the group contrast.

General explanation: a confidence interval is a way to express uncertainty around an estimate. Wider intervals indicate less precision than narrower intervals, while the interval's endpoints depend on the data and the method used to calculate it. The confidence level and construction method are not reported in the registry record. The interval should therefore be described as the reported interval, without assigning it a conventional confidence level.

The small number of participants meeting the abstinence definition in each group contributes to the limited precision of the comparison. The registry records 7 abstinent participants in the vigabatrin group and 5 in the placebo group. These counts support a point estimate, but they do not make the estimate exact; plausible values for the underlying contrast remain spread across the interval.

The p-value of 0.525 is consistent with substantial uncertainty about whether there is a group difference. It should not be interpreted as proof of no effect or as a measure of precision by itself. The interval offers more information about the range of contrasts compatible with the analysis, though its missing confidence-level and method details limit full assessment.

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 display gives planning illustrations for future studies based on fractions of the observed effect. The figures are participants per arm, not total participants or total events. The three scenarios are planning for 100% of the observed effect, planning for 75% of the observed effect, and planning for 50% of the observed effect. For each scenario, the display gives a participant count per arm for 80% power and a count per arm for 90% power.

General explanation: power is the chance that a study will detect an effect of a specified size under the assumptions used for planning, if that effect is present. The 80% power column describes the planned number of participants per arm for an analysis with an 80% chance of detecting the stated effect under the listed assumptions. The 90% power column uses a higher target chance of detection, which entails a larger planned sample size for the same effect scenario.

The assumptions listed for these illustrations 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 and sample-size justification are not reported in the registry record. These figures are not observed power; they are planning calculations based on the observed effect and the stated assumptions.

As the target effect becomes smaller, it is harder to distinguish it from random variation, so more participants per arm are needed. The values should therefore be interpreted as conditional illustrations rather than as a recommendation or a description of the trial's actual design basis. A future study using different assumptions, an alternative analysis, or a different target effect could require a different sample size.

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, p₀ and p₁ are the control and treatment proportions, and p̄ is their average.

Worked calculation for this trial

1. Control proportion p₀ = 0.053; treatment proportion p₁ = 0.076; average p̄ = 0.065.
2. First term: z_1−α/2 × √[2p̄(1−p̄)] = 1.96 × 0.3477 = 0.6816.
3. Second term: z_1−β × √[p₀(1−p₀) + p₁(1−p₁)] = 0.842 × 0.3474 = 0.2923.
4. n = (0.6816 + 0.2923)² / (0.0229)² = 1,809.4, rounded up to 1,810 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
p0 <- 0.0531914893617021                          # control proportion
p1 <- 0.0760869565217391                          # treatment proportion
n_per_arm <- function(power, frac = 1) {
  q1 <- p0 + (p1 - p0) * frac               # smaller effect = fraction of the observed difference
  pbar <- (p0 + q1) / 2
  z_b <- qnorm(power)
  ceiling(round((z_a * sqrt(2 * pbar * (1 - pbar)) + z_b * sqrt(p0 * (1 - p0) + q1 * (1 - q1)))^2 / (q1 - p0)^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 (unrounded n per arm)
cat("power.prop.test:", round(power.prop.test(p1 = p0, p2 = p1, power = 0.80)$n, 1), "\n")

# Simulation: how often does the z-test reject at the 80%-power sample size?
set.seed(611130)
n    <- n_per_arm(0.80)
reps <- 20000
x0 <- rbinom(reps, n, p0); x1 <- rbinom(reps, n, p1)
ph <- (x0 + x1) / (2 * n)
z  <- ((x1 - x0) / n) / sqrt(ph * (1 - ph) * 2 / n)
cat(sprintf("Simulated power at n = %s per arm: %.3f (%d replicates)\n", format(n, big.mark = ","), mean(abs(z) > z_a, na.rm = TRUE), reps))

Output from running the code

100% of observed effect: n per arm = 1,810 (80% power), 2,422 (90% power)
 75% of observed effect: n per arm = 3,085 (80% power), 4,129 (90% power)
 50% of observed effect: n per arm = 6,641 (80% power), 8,889 (90% power)
power.prop.test: 1809.4 
Simulated power at n = 1,810 per arm: 0.805 (20000 replicates)

The simulation draws many pairs of event counts at the planned sample size and counts how often the two-proportion z-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 identifies one primary outcome: the number of participants abstinent during Weeks 11 and 12, assessed at Week 13. It reports no secondary outcome results and lists 0 analyses. The record does not state whether multiple outcomes, time points, subgroup comparisons, or interim looks were considered in the original statistical plan. The multiplicity strategy is not reported in the registry record.

General explanation: when several hypotheses are tested, the chance of at least one apparently notable result can increase. Prespecifying a primary outcome and an analysis plan helps clarify which result is central and how multiple comparisons are addressed. In this record, the primary outcome is identifiable, but further information about the planned testing hierarchy or any adjustment is unavailable. No adjustment should be assumed from the reported p-value.

The primary population is labeled intent-to-treat. The outcome counts and group sizes are posted, but the registry does not explain whether every randomized participant had an observed outcome or how any missing outcome information was handled. Missing-data methods, imputation, sensitivity analyses, and reasons for missingness are not reported in the registry record. Without those details, it is not possible to assess the influence of missing outcomes on the posted comparison.

The record also does not describe exclusions from the analysis population, adherence-based analyses, or other analysis populations. General explanation: differences between the randomized population and the analyzed population can affect how directly an estimate represents the randomized comparison. The intent-to-treat label indicates the stated analysis population, but the implementation details remain unclear. These omissions are limitations of the available reporting and should not be filled in by assumption.

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 serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: drug abuse & major depression (2.2%), gun shot wound, road traffic accident & tibia fracture (2.2%), hepatitis, acute (1.1%), meningitis & pneumonia (1.1%), tachycardia (1.1%).

The serious adverse event figures report 6 affected participants in the vigabatrin group, with 92 at risk, and 0 affected participants in the placebo group, with 94 at risk. The registry lists the top serious event terms and their percentages: Drug Abuse & Major Depression, 2.2%; Gun shot wound, road traffic accident & tibia fracture, 2.2%; Hepatitis, acute, 1.1%; Meningitis & Pneumonia, 1.1%; and Tachycardia, 1.1%.

These are descriptive safety counts and registry-listed event terms. They do not, on their own, establish that an event was caused by treatment. The record does not provide event-level narratives, timing, severity detail beyond the serious-event classification, or an attribution assessment. Death figures are not reported in the registry record.

The non-serious adverse event figures list 73 affected participants in the vigabatrin group and 81 in the placebo group, with the same group sizes at risk as above. The record does not provide further detail in the listed summary that would support comparisons by event type, duration, or severity. Safety counts should be read separately from the primary abstinence outcome and with these reporting limits in mind.

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

The posted primary outcome shows 7 participants abstinent from cocaine during Weeks 11 and 12 in the vigabatrin group and 5 in the placebo group. The unadjusted risk-difference estimate is 0.023, with a reported interval from -0.048 to 0.094 and a p-value of 0.525. The estimate points toward a higher abstinence proportion in the vigabatrin group, but the interval includes zero and the result is uncertain.

This result does not establish that the groups are equivalent, nor does the point estimate establish a treatment effect. The data are compatible with no difference and with differences in either direction. The registry does not report a prespecified statistical plan or a confidence level for the interval, so the degree of certainty should not be overstated.

The endpoint concerns abstinence during a particular 2-week period near the end of a 9-week treatment phase. It does not describe abstinence over the full treatment period, durability after treatment, or other aspects of cocaine use. No secondary results are posted, so the record does not answer questions about other outcomes or follow-up periods.

The primary population is labeled intent-to-treat, but missing-data handling is not reported. The absence of that information limits interpretation of how completely the posted counts reflect the randomized groups. The registry's safety summary is also descriptive and does not establish treatment-related causation for the listed events.

Overall, the record documents a randomized comparison and an uncertain primary outcome estimate. It does not support a definitive conclusion about a difference in abstinence, and it does not address outcomes beyond those reported. Interpretation should remain tied to the population, treatment period, endpoint definition, and analysis information available in the registry record.

16. Why This Trial Matters Statistically

This trial illustrates why a point estimate, an interval, and a p-value need to be read together. The group counts differ, and the risk-difference estimate points in one direction, but the reported interval extends across zero and the p-value is 0.525. Looking only at the counts or only at the point estimate would understate the uncertainty in the comparison.

The endpoint is a binary outcome defined for a specific period: abstinence during Weeks 11 and 12. That definition makes the estimand concrete, but it also limits what the result addresses. It is not a measure of sustained abstinence across all treatment weeks, and it does not stand in for unreported outcomes. Statistical interpretation depends on keeping the endpoint's time window and population in view.

The trial is randomized and quadruple-masked, features that are relevant to the design of a treatment comparison. Yet the registry does not give the details of allocation procedures, the statistical plan, missing-data handling, or the method used to calculate the interval and p-value. The lack of those details constrains independent assessment of the analysis, even though the core outcome counts and comparison are posted.

The planning illustrations also highlight the distinction between an observed result and a future design calculation. They use the observed effect under stated assumptions, and smaller target effects require larger planned samples. Those illustrations do not show the probability that this completed trial detected an effect, nor do they replace the unreported original design justification.

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 registry record does not report the trial's original sample-size justification or design assumptions. The sample-size illustrations are based on the observed effect and specified planning assumptions, so they cannot be treated as a record of the investigators' original planning. The record also does not provide the detailed statistical analysis plan, including the test used for the primary comparison or the method used to construct its interval.

The primary endpoint is described as abstinence during Weeks 11 and 12, but the registry does not give a detailed account of how abstinence was assessed or verified. It does not report the handling of missing outcome data, discontinuations, protocol deviations, or adherence. While the analysis population is labeled intent-to-treat, the operational details needed to understand that analysis are not reported in the registry record.

There are no secondary analyses posted. The record does not provide broader results on cocaine use, post-treatment outcomes, or subgroup findings. It also does not state whether interim analyses or multiplicity procedures were planned. These absences limit the scope of conclusions that can be drawn and prevent reconstruction of a fuller statistical plan.

The safety summary is limited to counts of affected participants and a short list of serious event terms. The registry does not provide event narratives or causal attribution, and death figures are not reported. These details are needed to interpret safety patterns more fully, but their absence should not be treated as evidence that no events occurred beyond those listed.

18. Related Tutorials

19. Related Statistical Calculators

20. Sources

• ClinicalTrials.gov: Vigabatrin for Treatment of Cocaine Dependence — NCT00611130.
• Registry record last updated 2016-04-13; status last verified 2016-03.
• Results first posted 2012-06-12.
• 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 describes a completed randomized, quadruple-masked, parallel-group phase II study of vigabatrin versus matching placebo for cocaine dependence. The listed treatment period was 9 weeks, with three tablets twice daily in each arm. The primary outcome was the number of participants abstinent from cocaine during Weeks 11 and 12, assessed at Week 13, in an intent-to-treat population.

The posted primary counts are 7 in the vigabatrin group and 5 in the placebo group. The unadjusted risk-difference estimate is 0.023, with a reported interval from -0.048 to 0.094 and p-value of 0.525. Missing-data handling, the trial's original design assumptions, and further analysis details are not reported in the registry record.