Studying Amphetamine Withdrawal in Humans: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Studying Amphetamine Withdrawal in Humans (NCT01215929), a randomized phase 2 study. The primary outcome is Measure of Methamphetamine Withdrawal.

2. Clinical Question

3. Trial Design

The registry identifies NCT01215929 as a phase 2 randomized, quadruple-masked, parallel study sponsored by the University of Arkansas. Its status is completed. The registry lists enrollment of 35. The brief summary, however, describes a planned residential study of 30 methamphetamine-dependent participants, ages 18-65 years. The reason for the difference between the enrollment figure and the number in the summary is not reported in the registry record.

The brief summary describes a 4-week residential setting and says participants would be stabilized on 20-30mg long-acting amphetamine/d-amphetamine over the first 5 days. It also says randomization would be based on sex, the amphetamine withdrawal questionnaire score, and the methamphetamine selective severity assessment score. The summary is incomplete at the point where it begins describing the randomized treatment options. The exact post-randomization dosing schedule, duration, and any treatment changes are not reported in the registry record.

The two listed interventions are dextroamphetamine and placebo. The outcome results list 12 participants in each group. These outcome-group counts do not explain the disposition of everyone included in the registry enrollment figure. Reasons for not appearing in the posted outcome groups, withdrawals, and exclusions from analysis are not reported in the registry record.

The primary outcome is measured at the end of week 4. The registry gives start and primary completion dates of 2009-10 and 2013-04. The duration of recruitment and the timing of individual participant assessments are not described in further detail. Although the design is labeled quadruple-masked, the identities of the masked roles and the procedures used to maintain masking are not reported in the registry record. Randomization can reduce systematic differences between groups at allocation, but it does not by itself resolve uncertainty arising from small outcome groups, missing information, or incomplete reporting.

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.

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 Measure of Methamphetamine Withdrawal, measured over at the end of week 4.

The primary endpoint is the total score on the Methamphetamine Withdrawal Assessment scale (MAWA), measured at the end of week 4. The registry describes the scale as a questionnaire based on DSMIV criteria for amphetamine withdrawal. It has 13 items; each item asks about a symptom associated with stopping chronic amphetamine use, with severity indicated on a 4-point scale. The stated total-score range is 0-52. A lower score is defined as a better outcome, with 0 indicating no withdrawal symptoms and a higher score reflecting more severe symptoms.

A total score can provide a continuous measure of symptom severity rather than reducing the result to a simple present-or-absent classification. Comparing group means is one way to describe the average difference in scores. The registry identifies the reported group statistic as a least-squares mean and gives a standard error for each group. It does not report the questionnaire’s measurement properties, a threshold for an important change, or a rationale for selecting the end-of-week-4 assessment. Those details are not reported in the registry record.

The posted endpoint values are 3.728 for dextroamphetamine and 10.06 for placebo. The direction of the contrast is interpretable because the registry states that lower scores are better on this scale. The score alone does not establish how an individual participant felt, whether a difference was important to participants, or whether the same pattern would apply outside the study setting.

7. Results Posted in the Registry

• Dextroamphetamine: mean 3.73 versus 10.06 on control; difference -6.33 (95% CI -15.78 to 3.12), standardized effect d = -0.54; z-test p = 0.190.

8. Statistical Interpretation

9. Statistical Methods Explained

The primary outcome is continuous: it is a total score on the MAWA scale, with a stated range of 0-52. For a continuous outcome, a group mean or least-squares mean summarizes the average score. The registry labels the arm estimates as least-squares means and reports a standard error for each. The posted values are 3.728 with standard error 3.37 for dextroamphetamine, and 10.06 with standard error 3.45 for placebo.

What does a mean difference describe?

A mean difference expresses the separation between group averages in the outcome’s original units. The posted recalculated contrast is -6.328 for dextroamphetamine minus placebo. A negative value points toward a lower score in the dextroamphetamine group under that subtraction order. Because the registry defines a lower MAWA score as better, the sign can be understood in relation to the scale. The difference remains a group-level summary: it does not indicate that every participant had a lower score or that the observed contrast is a stable estimate of what would occur in another sample.

The least-squares label generally refers to a mean estimated through a statistical model, often accounting for terms included in that model. The registry does not report the model terms or explain how the least-squares estimates were obtained. The recalculated comparison is an unadjusted calculation from the posted arm summaries. It should not be mistaken for a fully described model-based treatment estimate. Because the posted arm values are least-squares means, and the registry does not give the underlying model, the relationship between those values and any original adjusted analysis cannot be evaluated from the record.

Why use a standard error?

A standard error describes the uncertainty in an estimated statistic, such as a group mean. It is not the same as the spread of individual participant scores. A smaller standard error indicates a more precisely estimated mean, while a larger standard error indicates greater sampling uncertainty. The registry reports standard errors of 3.37 and 3.45 for the two groups. It does not report standard deviations for the outcome groups. A standard error alone does not reveal the range of individual scores or whether the scores were symmetric, clustered, or affected by outliers.

What does the confidence interval add?

A confidence interval gives a range of values that are compatible with the estimate under the assumptions of the interval procedure. It communicates uncertainty in the estimated mean difference rather than focusing only on a single point estimate. The recalculated interval is -15.78 to 3.125. Since that range includes zero, the posted summaries are consistent with no group mean difference as well as with differences in either direction. A confidence interval is not a guarantee that the true value lies in the displayed range, and it does not account for every possible source of bias or error.

How should the p-value be understood?

The unadjusted p-value is 0.19. In general, a p-value measures how unusual data at least as incompatible with a specified null model would be if that model and the test assumptions held. It is not the probability that the null hypothesis is true, nor does it measure the size or practical importance of a difference. The registry does not report the original test procedure, its assumptions, or an adjusted analysis. The p-value attached to the recalculated contrast should therefore be interpreted as an unadjusted summary, not as a complete account of the trial’s analysis.

In general, over-dispersion matters when observed variation is greater than a statistical model expects. If a model underestimates variability, standard errors and confidence intervals may be too small, and tests may appear more decisive than warranted. The registry does not report whether over-dispersion was assessed or whether it is relevant to the MAWA analysis. No claim about its presence or absence can be made from the posted summaries.

What is the role of randomization?

In general, random allocation is intended to make treatment groups comparable in expectation, including on characteristics that are not measured. It does not guarantee balance in every realized trial, particularly when outcome groups are small. The brief summary says randomization was based on sex and two assessment scores, but the implementation details, allocation procedure, and analysis of those factors are not reported in the registry record. The results should not be described as adjusted for those factors unless the registry reports that analysis.

All recalculated quantities are unadjusted calculations from the posted summary statistics. The registry reports no sponsor analysis details for the primary outcome, and the count of posted analyses is 0. The exact analysis population, missing-data method, model specification, and handling of deviations from assigned treatment are not reported in the registry record. These omissions limit the ability to reproduce or independently assess the original analysis. General statistical explanations do not fill in those missing protocol details.

10. Confidence Intervals and Precision

The recalculated mean difference is -6.328, with a confidence interval from -15.78 to 3.125. The interval is wide enough to include zero and values on either side of zero. It therefore conveys substantial uncertainty about the direction and magnitude of the underlying difference. The interval is in MAWA score units, not a standardized scale. Since lower values are defined as better, negative differences under the stated subtraction order favor a lower average score in the dextroamphetamine group, while positive values point in the opposite direction.

Confidence intervals are useful because they show a range of estimates compatible with the data and the method’s assumptions. They do not show the probability that a particular treatment effect is true, and their interpretation does not include unreported sources of bias. The interval is a recalculated, unadjusted estimate based on posted summary statistics. The registry does not describe an original sponsor confidence interval or give enough model details to determine whether the recalculation matches the investigators’ intended analysis.

The reported standard errors are 3.37 for dextroamphetamine and 3.45 for placebo. These quantify uncertainty in the respective least-squares means, not the spread of individual participants’ scores. The registry does not give outcome standard deviations, participant-level observations, or further distributional information. The recalculated comparison uses an approximate pooled standard deviation, so the precision assessment depends on that approximation as well as the posted summaries.

The unadjusted p-value of 0.19 is consistent with the interval’s lack of a clear separation from zero under the reported recalculation. Neither the p-value nor the interval establishes equivalence, rules out a meaningful difference, or resolves uncertainty caused by the limited outcome-group counts and unexplained participant disposition. More complete reporting of the analysis and outcome data would be needed to assess precision beyond this summary.

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 figures are planning illustrations expressed as participants per arm, not total events. They use the observed effect as a reference and show how the illustrated number changes when planning for 100% of the observed effect, 75% of the observed effect, or 50% of the observed effect. For 100% of the observed effect, the figures are 55 participants per arm for 80% power and 74 participants per arm for 90% power. For 75% of the observed effect, they are 98 participants per arm for 80% power and 131 participants per arm for 90% power. For 50% of the observed effect, they are 219 participants per arm for 80% power and 293 participants per arm for 90% power.

The planning 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 registry does not report the trial’s original sample-size justification or original design assumptions; these planning values are not a reconstruction of the investigators’ design.

In general, 80% power means that, under the planning assumptions and for the effect being targeted, a study of the stated size would have an 80% chance of rejecting the null hypothesis if that effect were the true effect. The 90% column describes the corresponding planning target at 90% power. These are prospective design concepts, not probabilities that the observed result is correct. The figures are not observed power.

The smaller the effect targeted, the more participants are needed under otherwise unchanged assumptions. This pattern is visible across the displayed planning values: planning for 75% or 50% of the observed effect requires more participants per arm than planning for 100%. The curve is a planning aid, not evidence about the trial’s actual recruitment or the precision of a future study beyond the stated assumptions.

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, Δ is the difference in means to detect and σ is the common standard deviation.

Worked calculation for this trial

1. Observed difference Δ = -6.328; pooled standard deviation σ = 11.813.
2. n = 2 × (1.96 + 0.842)² × 11.813² / -6.328² = 54.7, rounded up to 55 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 <- -6.3277                      # observed difference in means
sigma <- 11.8133991721265                     # 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(1215929)
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 = 55 (80% power), 74 (90% power)
 75% of observed effect: n per arm = 98 (80% power), 131 (90% power)
 50% of observed effect: n per arm = 219 (80% power), 293 (90% power)
power.t.test: 55.7 
Simulated power at n = 55 per arm: 0.794 (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.

13. Multiplicity, Missing Data and Analysis Populations

The registry lists one outcome measure and reports no secondary outcomes. It also records 0 analyses. This means there is no set of posted secondary results from which to assess whether multiple outcomes or multiple comparisons were examined. The registry does not report an original statistical analysis plan, a hierarchy of outcomes, or a procedure for controlling multiplicity. These details are not reported in the registry record.

In general, testing many outcomes or making repeated comparisons can increase the chance of finding an apparently unusual result even when no underlying effect exists. Statistical methods can address multiplicity, but the appropriate approach depends on the planned questions and analysis. Because the registry does not describe the investigators’ analysis plan or additional analyses, it is not possible to determine whether multiplicity was considered or how it might have been handled. The lack of posted secondary results should not be taken to mean that no other measures were collected.

The outcome summaries show 12 participants in each group, whereas the registry enrollment is 35. The record does not state how many participants were randomized, how many completed the end-of-week-4 assessment, or why outcome counts differ from enrollment. It does not identify whether the analysis followed an intention-to-treat population, a per-protocol population, or another definition. Those analysis-population details are not reported in the registry record.

The registry also does not describe how missing outcome scores were handled, whether any imputation was used, or whether sensitivity analyses were conducted. In general, missing-data methods matter because conclusions can depend on why measurements are unavailable and on assumptions about the unobserved values. Without counts and reasons for missingness by group, the possible influence of missing data cannot be assessed from the posted summaries.

Further unreported items include the model specification for the least-squares means, the handling of protocol deviations, and any interim analyses. The registry record does not give enough information to distinguish a prespecified primary analysis from analyses chosen after seeing the data. These gaps do not change the posted estimates, but they limit reproducibility and the strength of conclusions that can be drawn from them.

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 non-serious adverse event terms

The registry reports serious adverse events for 0 of 18 participants at risk in the dextroamphetamine group and 0 of 17 participants at risk in the placebo group. These are counts of participants affected, not counts of all possible event episodes. The registry lists no serious adverse-event terms. Death counts are not reported in the registry record.

The registry also reports other adverse events affecting 8 of 18 participants at risk in the dextroamphetamine group and 9 of 17 participants at risk in the placebo group. The record does not provide event terms or further detail in the listed safety summary. These counts do not establish whether events were related to treatment, how severe they were, when they occurred, or whether a participant experienced more than one event. Such details are not reported in the registry record.

Safety counts should be read alongside their group denominators and the scope of the posted reporting. A count of zero serious events in these groups does not show that serious events are impossible, and it does not describe safety outside the study. The record does not provide enough detail to make a broader comparison of the frequency, nature, or severity of adverse events.

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

The posted outcome data describe lower average MAWA scores in the dextroamphetamine group than in the placebo group at the end of week 4. The recalculated unadjusted mean difference is -6.328, and lower scale scores are defined as better. The confidence interval, -15.78 to 3.125, includes zero and spans values in both directions. The unadjusted p-value is 0.19. These statistics express an uncertain group contrast; they do not establish a definitive difference or equivalence.

The registry labels the trial randomized, quadruple-masked, and parallel. Those design features are relevant to how a comparison may be conducted, but the public record does not give detailed information about allocation implementation, masking procedures, or the analysis model. Nor does it explain why the outcome groups each contain 12 participants while enrollment is listed as 35. The posted estimates therefore need to be read with caution about the population represented and the completeness of outcome data.

The result concerns a questionnaire score in a residential study and at a specified assessment time. It does not establish a change in other outcomes, persistence of any difference, or what would happen in a different setting. No secondary outcome results are posted. The score difference cannot by itself establish the clinical importance of the contrast because a threshold for an important change is not reported in the registry record.

In general, a nonsignificant or imprecise result is not proof that an intervention has no effect. Equally, a numerically favorable point estimate is not proof that the intervention caused a meaningful benefit. The registry’s unadjusted summary and its uncertainty are the appropriate basis for describing what the posted data show, while the missing analysis details and unexplained outcome counts limit stronger interpretation.

16. Why This Trial Matters Statistically

This trial illustrates why a point estimate should be read together with its uncertainty and the design information. The dextroamphetamine group has a lower posted average MAWA score than the placebo group, but the recalculated interval includes zero and extends to estimates in the opposite direction. Reporting only the point estimate would conceal this uncertainty; relying only on the p-value would also leave out the scale and range of plausible contrasts.

The distinction between a least-squares mean and an unadjusted recalculation is also important. The registry reports least-squares means, but does not give the model used to obtain them. The recalculated difference is a simple unadjusted comparison from posted summaries. Readers cannot verify whether it matches the original intended model or how any baseline factors were accounted for. Transparent model and analysis reporting would make the estimate easier to assess and reproduce.

The relationship between enrollment and the outcome-group counts is another statistical concern. The record lists 35 enrolled participants, while the primary outcome summaries list 12 per group. Without participant flow and missing-data information, it is not possible to determine who contributed to the outcome analysis or whether the analyzed groups reflect the randomized population. This matters because missing measurements can affect both the estimate and its precision.

The planning illustrations show how assumptions about the target effect influence the number of participants required per arm. They are distinct from the trial’s original design justification, which is not reported. Keeping that distinction clear prevents retrospective planning calculations from being mistaken for evidence about the study’s achieved power. Together, these issues make the result a useful example of interpreting an estimate while keeping design, analysis, and reporting limitations in view.

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.

Several trial-specific details needed to evaluate the result are absent from the registry record. The brief summary is truncated during its description of post-randomization treatment, so the full regimen, dose schedule, and treatment duration cannot be established. Although the study is labeled quadruple-masked, the record does not identify who was masked or describe the masking procedures. Randomization is stated, but the allocation sequence and implementation are not described.

The registry lists enrollment of 35, while the primary outcome results include 12 participants in each group. No participant flow, reasons for exclusions, withdrawals, or end-of-week-4 completion counts are reported. The analysis population and approach to missing scores are also not reported. These omissions prevent assessment of whether the outcome summaries include all randomized participants or could be affected by missingness.

The record identifies the endpoint as a least-squares mean but does not specify the statistical model, adjustment variables, or original comparison procedure. No sponsor-reported analysis details are listed. The recalculated comparison is unadjusted and relies on an approximate pooled standard deviation, so it should not be treated as a reconstruction of an undocumented analysis. There are no posted secondary outcomes or additional analyses, and the registry does not state whether interim analyses or multiplicity procedures were used.

The posted primary results give group summaries rather than individual score distributions. The record does not provide information about outliers, score distribution, or the threshold for a clinically important score difference. It also gives limited detail about adverse events; death counts are not reported. These limitations constrain conclusions to the reported summaries and do not support broader claims about the treatment or its effects in other populations or settings.

18. Related Tutorials

19. Related Statistical Calculators

20. Sources

• ClinicalTrials.gov: Studying Amphetamine Withdrawal in Humans — NCT01215929.
• Registry record last updated 2014-08-04; status last verified 2014-07.
• Results first posted 2014-08-04.
• 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

NCT01215929 is a completed phase 2 randomized, quadruple-masked, parallel study sponsored by the University of Arkansas. The registry lists methamphetamine dependence as the condition and dextroamphetamine and placebo as the interventions. Enrollment is listed as 35, while the primary outcome summaries contain 12 participants per group.

The primary outcome is the end-of-week-4 MAWA score, for which lower scores are defined as better. The posted least-squares means are 3.728 for dextroamphetamine and 10.06 for placebo. The recalculated unadjusted mean difference is -6.328, with a confidence interval from -15.78 to 3.125 and an unadjusted p-value of 0.19. The interval includes zero, and the registry does not report a sponsor analysis specification.

The record does not explain the difference between enrollment and outcome counts, describe missing-data handling, or provide secondary outcome results. Its brief study summary is incomplete in the description of treatment after randomization. Safety reporting includes zero participants with serious adverse events in each group; death counts are not reported in the registry record.