Medications for Stopping Cocaine Dependence and Preventing Relapse: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Medications for Stopping Cocaine Dependence and Preventing Relapse (NCT00218023), a randomized phase 2 study. The primary outcome is Mean Percentage of Cocaine-positive Urines Over Course of 12 Week Treatment in Subgroup Achieving Abstinence at Baseline.

2. Clinical Question

The registry record identifies a randomized, quadruple-masked, parallel-group phase 2 study with 101 participants enrolled overall. The outcome-specific arm sizes are smaller than the overall enrollment and apply to the subgroup named in the endpoint. The record does not report an inferential comparison of the four group means, so the posted means describe the observed groups rather than establish whether their results differed beyond chance.

3. Trial Design

This was a randomized, quadruple-masked, parallel-group phase 2 study. The registry record reports 101 participants enrolled overall and lists the study as completed. It identifies The University of Texas Health Science Center, Houston as the sponsor. The record lists four groups: modafinil, levodopa/carbidopa, naltrexone HCl, and placebo. Each group title includes motivational interviewing, contingency management, and cognitive-behavioral therapy.

The registry record gives a primary completion date of 2012-04 and a start date of 2006-03. These are registry dates, not a description of each participant’s duration in the study. The primary endpoint’s stated observation period is 12 weeks, with urine testing scheduled three times per week on Monday, Wednesday, and Friday.

For the primary endpoint, the registry reports arm sizes of 8.0 for modafinil, 8.0 for levodopa/carbidopa, 9.0 for naltrexone HCl, and 10.0 for placebo. These are the group sizes attached to the posted endpoint, whose title specifies the subgroup achieving abstinence at baseline. They should not be mistaken for the full enrollment or for the safety populations. The safety record gives separate at-risk counts for each group.

The registry record does not report the randomization procedure, allocation concealment method, or masking roles in detail. It also does not report the planned sample-size justification or a detailed protocol for how the behavioral interventions were delivered. The descriptions of the medication and placebo arms are incomplete in the registry record, so additional treatment details cannot be established from these entries.

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: 101 subjects were enrolled and began phase 1 (in phase 1, all enrolled subjects received the same intervention, motivational interviewing, which is described in the Interventions section). 81 completed phase 1, and 81 started phase 2, as described in the Period below. Subjects were assigned to the 4 arms at phase 2.

Overall Study

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 Mean Percentage of Cocaine-positive Urines Over Course of 12 Week Treatment in Subgroup Achieving Abstinence at Baseline, measured over 3 times per week (Monday, Wednesday, and Friday) for 12 weeks.

The primary endpoint is titled the mean percentage of cocaine-positive urines over the course of 12-week treatment in the subgroup achieving abstinence at baseline. It is a continuous outcome reported as a mean. Urine samples were scheduled three times per week, on Monday, Wednesday, and Friday, and cocaine use was assessed by testing for benzoylecgonine.

Repeated urine testing is a way to observe cocaine-positive results during the treatment period rather than rely only on a participant’s retrospective account of use. In general, a mean percentage of positive samples summarizes the observed frequency of positive urine results across the group. A higher value indicates a greater mean percentage of positive samples in that group’s endpoint data; it does not, by itself, establish why the values differ or whether a difference is statistically reliable.

The record reports group means and standard errors for this endpoint. The listed mean values are 31.0 for modafinil, 28.0 for levodopa/carbidopa, 47.0 for naltrexone HCl, and 35.0 for placebo. The corresponding reported standard errors are 0.238, 0.2, 0.261, and 0.186. The registry does not report confidence intervals or a direct inferential comparison for these means.

The endpoint title limits the reported outcome to a baseline-abstinence subgroup. The registry record does not explain how abstinence was established, how many participants were screened for this subgroup, or how missing urine samples were handled. Those details matter when interpreting what the mean represents and how broadly it applies.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

What does the reported mean represent?

The endpoint is a continuous measure, and the registry identifies its summary as a mean. In general, a mean is an average across the observations included in an analysis. In this trial, the outcome is the mean percentage of cocaine-positive urine tests over the treatment course, for the subgroup described as achieving abstinence at baseline. The urine tests were scheduled three times per week, and benzoylecgonine presence was used to assess cocaine use. The registry does not explain how each participant’s percentage was calculated, how incomplete testing schedules were handled, or how participant-level values were combined into the group mean.

The posted means are 31.0, 28.0, 47.0, and 35.0 for the modafinil, levodopa/carbidopa, naltrexone HCl, and placebo groups, respectively. In general, the mean provides a compact description of the group’s results, but it does not show the full distribution of participant outcomes. Participants can differ substantially even when their group mean is the same. The registry record does not report participant-level outcome distributions or another measure of spread for individual observations.

What does a standard error describe?

The registry identifies the reported dispersion as standard error. In general, the standard error describes the sampling precision of an estimated quantity, such as a group mean, under the method used to calculate it. It is not the same as the variation among individual participants and should not be read as a range containing most individual results. The posted standard errors are 0.238, 0.2, 0.261, and 0.186 for the four groups in the same order as their means.

A standard error can help quantify uncertainty, but it does not by itself provide a complete comparison between groups. A confidence interval or a specified model-based contrast would communicate uncertainty around a difference directly. Neither confidence intervals nor group-comparison estimates are reported in the registry record. It would not be appropriate to infer a p-value or confidence interval from the displayed summaries alone, particularly because the analysis method and missing-data handling are not reported.

Why do the endpoint group sizes matter?

The primary endpoint is restricted to a subgroup described as achieving abstinence at baseline. Its reported arm sizes are 8.0, 8.0, 9.0, and 10.0, while overall enrollment is 101. These quantities describe different populations and should not be interchanged. In general, estimates based on smaller groups can be sensitive to the outcomes of the participants included. The registry record does not report how the subgroup was defined, whether subgroup membership was determined before randomization, or whether all eligible participants had endpoint data.

The safety record gives different at-risk counts from the endpoint group sizes. This is not, by itself, evidence of an error: safety and efficacy outcomes can use different populations or reporting denominators. However, the registry record does not explain the population definitions behind these counts. The outcome and safety figures therefore need to be interpreted using their own reported group sizes.

What can be concluded without a reported comparison?

In general, randomized allocation can support comparisons between groups when the analysis preserves the randomized assignment and appropriately accounts for the design. The registry identifies the allocation as randomized, but it does not report a primary statistical analysis, a model, an estimate of a medication-versus-placebo contrast, or a test of the group differences. No analyses are listed. The posted group means therefore support descriptive statements about the reported endpoint values, not a conclusion that the groups differ statistically.

The study is described as quadruple-masked. The registry record does not explain which roles were masked or how masking was maintained. Masking can, in general, reduce the influence of expectations on study conduct or assessment, but the label alone does not establish that bias was prevented. The record also does not describe allocation concealment or other implementation details.

How should the planning illustrations be read?

In general, a sample-size calculation estimates how many participants per arm might be needed under stated assumptions to detect a specified effect with a chosen significance level and power. Such calculations are planning illustrations, not observed evidence that a study had a particular probability of success. The registry record states that its planning illustration uses the observed effect, a two-sided significance level of 0.05, equal allocation between two arms, a normal approximation, and an effect defined as a stated fraction of the observed effect. These are calculation assumptions, not a reported account of the trial’s original design rationale.

The record does not report the trial’s sample-size justification, and the planning illustration should not be used as observed power. The sample sizes in such an illustration would depend on its assumptions and effect target; they would not replace the actual endpoint-specific group sizes or supply a missing inferential analysis. The record also does not report any interim analyses or stopping rules.

Overall, the statistical information supports a clear description of the endpoint and its posted group summaries. Further interpretation would require details not reported in the registry record, including the analysis population, model, treatment contrasts, confidence intervals, missing-data approach, and multiplicity strategy.

10. Confidence Intervals for Each Arm

No two-arm comparison is shown for this trial, but each arm's mean can be given with a t-based 95% confidence interval computed from the posted mean, spread and number of participants.

R code and output

arm <- c("Modafinil Plus MI, CM, and CBT", "Levodopa/Carbidopa Plus MI, CM, and CBT", "Naltrexone HCl Plus MI, CM, and CBT", "Placebo Plus MI, CM, and CBT")
n   <- c(8, 8, 9, 10)
m   <- c(31, 28, 47, 35)
sp  <- c(0.238, 0.2, 0.261, 0.186)
sd  <- sp * sqrt(n)      # standard errors converted to SDs
for (i in seq_along(n)) {
  se <- sd[i] / sqrt(n[i]); h <- qt(0.975, n[i] - 1) * se
  cat(sprintf("CARM|%s|%d|%.3f|%.3f|%.3f|%.3f\n", arm[i], n[i], m[i], sd[i], m[i] - h, m[i] + h))
}

CARM  Modafinil Plus MI, CM, and CBT  8  31.000  0.673  30.437  31.563
CARM  Levodopa/Carbidopa Plus MI, CM, and CBT  8  28.000  0.566  27.527  28.473
CARM  Naltrexone HCl Plus MI, CM, and CBT  9  47.000  0.783  46.398  47.602
CARM  Placebo Plus MI, CM, and CBT  10  35.000  0.588  34.579  35.421

11. Multiplicity, Missing Data and Analysis Populations

The registry lists two outcome measures, but no secondary outcomes are posted and no analyses are listed. The primary outcome is reported as a mean percentage of cocaine-positive urine tests in a baseline-abstinence subgroup. The record does not report whether the primary endpoint was prespecified in a protocol, how the two listed outcome measures were prioritized, or whether any multiplicity adjustment was planned or used. It is therefore not possible to determine from the registry record how the risk of chance findings across outcomes or comparisons was addressed.

The record does not report a statistical analysis plan, a defined analysis population, or a comparison model. It also does not state whether participants were analyzed according to randomized assignment, how protocol deviations were handled, or how any exclusions were decided. These details are important for interpreting randomized trials because analysis choices can affect the group estimates and uncertainty.

Missing-data information is not reported in the registry record. In particular, the record does not describe missed urine tests, withdrawals, the number of participants with complete endpoint data, or any imputation or other missing-data method. In general, missing observations can affect a mean if the observed participants differ from those without measurements. Without the amount and handling of missingness, the direction or size of any such effect cannot be determined.

The outcome-specific group sizes are reported, but they do not explain how the subgroup achieving abstinence at baseline was identified or whether every member contributed measurements throughout treatment. These limitations are separate from the total enrollment and cannot be resolved from the posted group summaries.

12. 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

Serious events with the highest arm-level rate: work-related traumatic injury to finger that required surgical repair (6.2%), deep vein thrombosis (5.6%), routine liver-function tests showed an increase from baseline in hepatic enzyme values (4.5%), eye injury (torn conjunctiva) (4.0%), respiratory infection (4.0%).

The safety record gives serious adverse event counts alongside the number at risk in each group. The reported counts are 1 of 22 for modafinil, 4 of 25 for levodopa/carbidopa, 1 of 16 for naltrexone HCl, and 1 of 18 for placebo. These are the registry’s group counts; they should not be confused with the smaller endpoint-specific group sizes used for the primary outcome. The record does not explain the safety population definitions or provide a statistical comparison of these counts.

The registry reports other adverse events affecting 0 participants in each group. Deaths are not reported in the registry record. The listed serious-event terms and percentages are: work-related traumatic injury to finger that required surgical repair, 6.2%; deep vein thrombosis, 5.6%; routine liver-function tests showed an increase from baseline in hepatic enzyme values, 4.5%; eye injury (torn conjunctiva), 4.0%; respiratory infection, 4.0%; and arrhythmia symptoms, 4.0%.

The record does not explain the denominator or calculation used for the percentages attached to these terms, nor does it clarify whether events could be counted more than once. The terms and percentages should therefore be read as registry-reported safety information, not as directly comparable event rates across treatment groups. These figures alone do not establish whether an event was caused by a study treatment.

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

The registry results describe a randomized, quadruple-masked, parallel-group phase 2 trial with 101 participants enrolled overall. For the primary endpoint, the record reports mean percentages of cocaine-positive urine tests in a subgroup described as achieving abstinence at baseline. The posted means are 31.0 for modafinil, 28.0 for levodopa/carbidopa, 47.0 for naltrexone HCl, and 35.0 for placebo. The endpoint-specific group sizes and standard errors are also reported.

These data show the values recorded for the four groups on the stated endpoint. They do not, on their own, show whether differences among the group means are statistically distinguishable from chance variation. No treatment contrasts, confidence intervals, p-values, or analyses are listed in the registry record. A ranking of the means is descriptive, not a test of efficacy or evidence of a causal difference.

The endpoint concerns a selected subgroup, not necessarily all participants enrolled. The registry does not report the subgroup definition, the missing-data approach, or the analysis population. The results therefore do not establish how the endpoint relates to participants outside that subgroup or how the findings might apply to other populations. The record also does not describe a comparison that separates medication effects from the behavioral interventions listed alongside each arm.

The safety record reports serious adverse event counts, other adverse event counts, and selected serious-event terms. Those figures offer a description of reported events but do not establish comparative safety, especially because the registry does not provide a statistical safety analysis or explain the denominators attached to the listed event percentages. Deaths are not reported in the registry record.

In general, trial results are most informative when estimates are accompanied by analysis methods, uncertainty intervals, and clear definitions of the populations analyzed. Those details are not reported. The registry record supports a description of the posted endpoint and safety summaries, but not a conclusion that a medication worked or did not work.

14. Why This Trial Matters Statistically

This trial is statistically instructive because it combines random assignment and masking with an endpoint reported for a specifically named subgroup. The distinction between overall enrollment and endpoint-specific group sizes is essential: the primary outcome includes group sizes of 8.0, 8.0, 9.0, and 10.0, while overall enrollment is 101. The subgroup label indicates that the primary result is not simply a summary of all enrolled participants.

The endpoint uses repeated urine testing during a defined treatment course, which provides an objective measurement approach based on benzoylecgonine testing. At the same time, the reported mean percentage compresses repeated observations into a summary. In general, a mean alone does not convey the pattern of results across participants or across testing occasions. The registry record does not describe the calculation method, missing-test handling, or a longitudinal analysis.

The listed standard errors communicate some uncertainty about the group means, but the record does not provide confidence intervals or direct comparisons. This distinction matters: seeing several group estimates does not itself answer whether the groups differ statistically. Without a reported contrast and its uncertainty, descriptive values should not be treated as inferential conclusions.

The record also illustrates why study design labels and analysis details should be read separately. Randomization and quadruple masking describe aspects of design, but the registry does not report the analysis model, population definition, multiplicity plan, or missing-data procedure. Those omitted details limit what can be concluded from the posted summaries, even when the study design has features intended to support a fair comparison.

15. 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 leaves several trial-specific questions unanswered. It does not define the subgroup described as achieving abstinence at baseline, state how that status was assessed, or report the number of participants considered for inclusion in that subgroup. The primary endpoint group sizes are reported, but the record does not explain how those participants were selected or whether the subgroup was defined before treatment assignment.

Outcome measurement is described as urine testing for benzoylecgonine three times per week during the 12-week treatment period. However, the record does not explain how participant-level percentages were calculated, how missed or unusable samples were treated, or whether every participant contributed the same number of measurements. Missing-data methods and withdrawal information are not reported in the registry record.

The record does not list any analyses and does not report a statistical model, group contrasts, confidence intervals, p-values, or an analysis population. It also does not report the study’s sample-size justification, interim analyses, or stopping rules. The planning illustration is not the trial’s own design justification and is not observed power.

Although the trial is identified as randomized and quadruple-masked, the registry does not describe the randomization procedure, allocation concealment, or which roles were masked. The medication arms are paired in their descriptions with motivational interviewing, contingency management, and cognitive-behavioral therapy, but detailed implementation information is incomplete. The record therefore does not support separating the effect of medication from the behavioral components.

Finally, the safety record uses at-risk counts that differ from the endpoint group sizes and does not fully define those populations. It lists serious-event terms with percentages but does not report their denominators. These gaps limit direct comparison of safety information across the arms.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Medications for Stopping Cocaine Dependence and Preventing Relapse — NCT00218023.
• Registry record last updated 2017-06-09; status last verified 2017-05.
• Results first posted 2017-04-07.
• 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.

19. Record Summary

The registry record describes a completed randomized, quadruple-masked, parallel-group phase 2 trial sponsored by The University of Texas Health Science Center, Houston. Overall enrollment was 101. Four groups received modafinil, levodopa/carbidopa, naltrexone HCl, or placebo alongside motivational interviewing, contingency management, and cognitive-behavioral therapy.

The primary endpoint is the mean percentage of cocaine-positive urine tests over 12 weeks in a subgroup described as achieving abstinence at baseline. Testing was scheduled three times per week, and cocaine use was assessed by benzoylecgonine in urine. The posted group means are 31.0, 28.0, 47.0, and 35.0, respectively. The registry does not list analyses or provide inferential comparisons.

Serious adverse event counts are reported by group, while deaths are not reported in the registry record. Important details about subgroup definition, missing data, analysis methods, and safety denominators are not reported in the registry record. The posted results therefore support descriptive interpretation, with important limits on statistical and clinical conclusions.