Clinical Intervention in Alcohol Use Disorder: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Clinical Intervention in Alcohol Use Disorder (NCT02168400), a randomized na study. The primary outcome is Intervention Effects on Brain Functional Connectivity.

2. Clinical Question

3. Trial Design

The registry describes a randomized, single-masked, parallel-group trial sponsored by the University of Minnesota. It lists the study as completed, with enrollment of 86. The listed phase is NA. The registry gives a start date of 2016-07-07 and a primary completion date of 2022-11-17. These registry entries describe the study record; the posted outcome summaries have their own group counts, and the record does not explain how those counts relate to enrollment.

The two arms were active tDCS and sham tDCS. The active-arm description specifies 10 stimulation sessions. In the sham arm, very low current at the beginning and end of a session was intended to mimic the scalp sensation of stimulation without reaching levels that would stimulate brain function. The registry labels masking as single but does not report which people involved in the trial were masked or how masking was assessed. Details of sequence generation, allocation concealment, and the timing or delivery of outcome assessments beyond the listed timeframes are not reported in the registry record.

The primary outcome was mean change in brain functional connectivity, measured at pre-intervention and post-intervention, compared between groups at 1 week. Posted primary results include 31 participants in the active group and 29 in the sham group. The registry separately describes a secondary correlation outcome at 8 months, relating functional connectivity change to length of abstinence. It does not report the analysis population definition, handling of missing measurements, or a protocol-level statistical analysis plan. The record’s completed status does not by itself resolve those reporting gaps.

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: 11 consented participants were not randomized to either study group and were excluded from the study

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 Intervention Effects on Brain Functional Connectivity, measured over 1 week.

The primary endpoint is the between-group comparison of mean change in brain functional connectivity from the pre-intervention measurement to the post-intervention measurement, with the assessment timeframe listed as 1 week. Functional connectivity is the brain measure the registry identifies as a potential biomarker and intervention target. The endpoint therefore addresses whether the measured connectivity changed differently in the active tDCS group and the sham group over the stated assessment period.

Using change from a pre-intervention measurement can account for where participants started on the measured outcome, while comparing changes between randomized groups focuses the contrast on whether the observed change differed by assigned intervention. That is a general explanation of change-score comparisons; the registry does not report further details about the exact measurement procedure, the brain regions or connections included, or the model used to estimate the contrast.

The unit is recorded as “percentage of Blood Oxygenation Level De,” which appears incomplete in the registry record. The result is reported as a mean with standard error for each group. The endpoint is a brain connectivity measure, not itself a direct measure of abstinence duration. The registry lists length of abstinence in a separate secondary correlation outcome, so the primary endpoint should be read as a short-term brain measure rather than as a clinical abstinence result.

7. Results Posted in the Registry

• Active tDCS: mean 0.05 versus 0.19 on control; difference -0.14 (95% CI -0.53 to 0.25), standardized effect d = -0.19; z-test p = 0.476.

8. Secondary Efficacy Endpoints

The registry also posts results for secondary outcomes. For each outcome below, the endpoint type, the test applied to the posted summary statistics, the result and its interpretation are given in turn. Secondary endpoints are usually not adjusted for multiple testing, so they support rather than confirm the primary finding.

Correlation Between Functional Connectivity Change and Clinical Outcome (Length of Abstinence)

• Endpoint type: Continuous: a mean value per participant, compared as a difference in means.
• Test: Unadjusted two-sample z-test for the difference in means, using the posted standard errors.
• Result:
  • Active tDCS: mean 0.03 versus 0.16 on control; difference -0.13 (95% CI -0.55 to 0.29), standardized effect d = -0.16; z-test p = 0.540.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Active tDCS and the control. The interval gives the range of effects that remain compatible with the data.

The registry lists one secondary outcome: the correlation between functional connectivity change and clinical outcome, defined as length of abstinence. Its timeframe is 8 months. The posted secondary comparison follows the same broad direction as the primary contrast, with the active-group summary lower than the sham-group summary. This directional pattern does not independently strengthen the primary finding, because the secondary measure addresses an association and is not the same endpoint as the randomized comparison of mean connectivity change.

Correlation and treatment contrasts answer different questions. A correlation concerns how two measured quantities vary together; it does not establish that one causes the other. Nor does a comparison of group-level correlation summaries directly show that assigning active tDCS changes abstinence duration. The registry does not report a separate randomized treatment-effect estimate for length of abstinence as an outcome.

The posted secondary comparison is unadjusted. Secondary findings usually carry less weight than a primary result when their testing role, multiplicity approach, and analysis plan are unclear. Multiple statistical tests can increase the chance of a seemingly notable result arising by chance; adjustment or a prespecified hierarchy can address that concern, but the registry does not report such an approach. The record lists no reported analyses, which further limits understanding of how the secondary result was selected or interpreted.

Read alongside the primary result, the secondary outcome adds a related biological and clinical question, but it does not resolve the primary estimate’s uncertainty or establish a clinical benefit. Further interpretation would require details about how the correlation was calculated, which participants contributed, how missing data were handled, and whether the analysis was prespecified. Those details are not reported in the registry record.

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

• Correlation Between Functional Connectivity Change and Clinical Outcome (Length of Abstinence)

9. Statistical Interpretation

10. Statistical Methods Explained

The statistical interpretation starts with the estimand: the mean difference in change in functional connectivity between active tDCS and sham tDCS. The posted contrast is -0.14 for active relative to sham. A negative value indicates that the active-group mean change was lower than the sham-group mean change on the recorded scale. The contrast is a difference between group means, not a proportion, a relative change, or a measure of individual response.

What does the standard error describe?

The group results report means with standard errors. In general, a standard error describes the precision of an estimated quantity, such as a group mean, rather than the range of values among individual participants. A standard deviation instead describes variation among observations. The registry labels the reported dispersion as standard error, so those values should not be treated as participant-level standard deviations. The record does not provide the underlying observations or additional details of the measurement distribution.

How should the confidence interval be read?

A confidence interval expresses uncertainty around an estimate under the assumptions of the statistical procedure used to construct it. In general, a wider interval indicates less precision than a narrower interval. The primary interval includes zero and spans negative and positive contrasts. That means the posted estimate does not rule out no difference, and it also does not rule out contrasts in either direction within the interval. A confidence interval is not a guarantee that the true value lies within its limits, nor does it summarize all sources of uncertainty, such as limitations in measurement or analysis reporting.

What does the p-value contribute?

The primary unadjusted p-value is 0.476. In general, a p-value describes how unusual data at least as incompatible with a specified null model would be if that model and the analysis assumptions held. It is not the probability that the null model is true, and it is not a measure of the size or importance of a contrast. The registry does not report a full statistical analysis plan or identify a prespecified testing procedure, so the p-value should be interpreted as the posted unadjusted result rather than as a complete account of the trial’s inferential framework.

Why does over-dispersion matter in general?

For outcomes where variation is greater than a simple model assumes, over-dispersion can make standard errors too small and intervals too narrow if it is ignored. That concern is a general principle, not a conclusion about this trial. The primary outcome is reported as a continuous measure with standard errors, and the registry does not identify over-dispersion as an issue or provide enough model detail to assess it. The record also does not describe the distributional assumptions used for the posted comparison.

What is a recalculated comparison?

A recalculated comparison is a simple, unadjusted estimate derived from posted summary statistics. It is useful as a check on the scale and direction of the displayed contrast, but it is not a replacement for the trial’s own analysis. The registry reports no discrepancy for the primary comparison, and no sponsor analysis is posted. No adjustment for baseline values, covariates, repeated measurements, or missing outcomes can be inferred from this simple comparison.

What do the planning illustrations mean?

The planning figures use the observed effect as a reference and show hypothetical participants per arm for specified fractions of that effect and target power levels. Their assumptions are a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. These are planning calculations, not evidence about the design choices made for this trial. The trial’s original sample-size assumptions are not reported in the registry record. Because the endpoint is continuous, the figures refer to participants per arm rather than total events.

In general, power is the chance that a specified analysis would identify an effect of a chosen size under its assumptions, if that effect were present. An 80% target and a 90% target represent different planning targets; the higher target generally requires more participants for the same assumed effect. A smaller effect is harder to distinguish from random variation and therefore requires more participants under otherwise unchanged assumptions. These concepts explain the planning approach, but do not establish what power the completed trial had or whether its actual analysis followed those assumptions.

The registry lists a secondary correlation outcome and no reported analyses. The posted secondary result is an unadjusted comparison of group summaries for that outcome. Correlation coefficients describe association, not a treatment effect on abstinence. A group comparison of reported correlation values also does not, on its own, establish whether connectivity change predicts or causes a change in abstinence duration.

11. Confidence Intervals and Precision

The primary mean difference is -0.14, with a confidence interval from -0.525 to 0.245. The interval includes zero and allows values on both sides of no difference. This breadth means the estimate is not precise enough to isolate a narrow range of plausible between-group contrasts. In particular, the posted result does not distinguish a negative contrast from a positive contrast with confidence.

The group means were 0.05 in the active arm and 0.19 in the sham arm. Their standard errors were 0.05 and 0.19. The different standard errors indicate different estimated precision for the group means, but do not reveal the full distribution of participant values. The registry provides no group-level confidence intervals for these means, no individual data, and no further measurement detail with which to assess the source of this uncertainty.

The unadjusted p-value of 0.476 is consistent with the interval’s inclusion of zero, but it should not be used as a substitute for reading the interval. The p-value does not show that the contrast is exactly zero, while the interval communicates that a range of positive and negative contrasts remains compatible with the posted estimate and its uncertainty.

Precision is also constrained by the reporting record. The primary posted results include 31 active-arm participants and 29 sham-arm participants, while the registry does not explain the analysis population, missingness, or whether all randomized participants contributed to the endpoint. The enrollment figure alone cannot answer those questions. The endpoint unit is incomplete as recorded, and detailed measurement and model information is not reported. Those gaps affect interpretation beyond what the numerical interval captures.

12. Planning a Similar Trial: Sample Size

This is a planning illustration, not a re-evaluation of the original trial. It asks: if a new study wanted to detect a difference like the one observed, or a smaller one, how large would it need to be? It is deliberately not “observed power” of the finished trial, which is not a meaningful quantity.

The planning illustrations are expressed as participants per arm, not total events. They use a two-sided significance level of 0.05, equal allocation between the active and sham arms, and a normal approximation. The target effect is a stated fraction of the observed effect. The trial’s own original design assumptions are not reported in the registry record, so these planning values should not be read as a reconstruction of the trial’s planned enrollment.

The 80% power column gives a hypothetical per-arm participant requirement under the stated assumptions for an analysis designed to have an 80% chance of detecting the chosen effect if it were present. The 90% column uses a higher target chance under the same general planning framework. The higher target generally requires more participants for the same assumed effect. These are design illustrations, not a description of the actual enrollment or analysis population.

Rows that plan for a smaller fraction of the observed effect require more participants per arm, because smaller contrasts are harder to distinguish from random variation. The effect used for planning is based on an estimate from this trial, which is uncertain; planning from an observed effect does not make that effect a known or guaranteed future target. The figures are not observed power. They cannot establish whether the completed trial was adequately planned or how likely its actual analysis was to detect a particular effect.

For the continuous outcomes in this record, the planning counts are participants per arm. The registry’s general planning note distinguishes this from time-to-event calculations, which would be expressed as total events, but no time-to-event endpoint is posted for these results. A curve based on the same planning assumptions should be read as illustrating how the hypothetical requirement changes as the target effect changes, rather than as an observed feature of the trial.

13. How the Sample Size Is Derived

Power is the probability that a study of a given size rejects the null hypothesis when a specific true effect exists. It depends on four things: the effect to detect, the variability of the outcome, the significance level (here a two-sided α of 0.05, so z_1−α/2 = 1.960) and the desired power (z_1−β = 0.842 for 80% and 1.282 for 90%).

Formula

n is the number of participants in each arm, Δ is the difference in means to detect and σ is the common standard deviation.

Worked calculation for this trial

1. Observed difference Δ = -0.140; pooled standard deviation σ = 0.739.
2. n = 2 × (1.96 + 0.842)² × 0.739² / -0.140² = 436.9, rounded up to 437 participants per arm for 80% power.

The smaller-effect rows in the table above repeat the same calculation with the effect multiplied by 0.75 and 0.5 (for ratios, applied to the logarithm of the ratio).

R code

alpha <- 0.05
z_a   <- qnorm(1 - alpha / 2)          # two-sided
delta <- -0.14                      # observed difference in means
sigma <- 0.73857038046252                     # 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(2168400)
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 = 437 (80% power), 585 (90% power)
 75% of observed effect: n per arm = 777 (80% power), 1,040 (90% power)
 50% of observed effect: n per arm = 1,748 (80% power), 2,340 (90% power)
power.t.test: 437.8 
Simulated power at n = 437 per arm: 0.801 (20000 replicates)

The simulation draws many sets of group means and variances at the planned sample size and counts how often the two-sample t-test rejects the null. It should land close to the target power; at very small required sample sizes it can fall below it, because the normal approximation is rough there.

14. Multiplicity, Missing Data and Analysis Populations

Multiplicity arises when several outcomes, timepoints, groups, or statistical tests are examined. If many tests are conducted, the chance of at least one result appearing unusual under a no-effect model can increase. A statistical plan can address this through prespecified primary outcomes, an ordered testing strategy, or a correction for multiple comparisons. The registry identifies a primary outcome and one secondary outcome, but the approach to multiplicity is not reported in the registry record. It also does not report a complete analysis plan that would show whether other analyses were planned or conducted.

The record lists zero reported analyses. That is a reporting-status detail, not evidence that no statistical analyses occurred. No sponsor analysis, adjusted estimate, or analysis description is posted. The registry does not state whether secondary tests were adjusted, whether the primary endpoint had a prespecified testing hierarchy, or whether any interim analyses were conducted. Those details should not be inferred from the posted unadjusted comparisons.

Missing-data handling is also not reported. The record does not say how many endpoint measurements were missing, whether participants were excluded from an analysis, or whether imputation or another method was used. The listed enrollment and the outcome-specific group counts do not establish why the counts differ. The analysis population—such as whether results follow randomized assignment or include only participants with available measurements—is not reported.

Other details needed to assess the analysis are absent: the statistical model, any covariate adjustment, the handling of repeated measurements, and the exact construction of the connectivity measure. The registry labels the design as randomized and single-masked, but the method of random sequence generation, allocation concealment, and identity of masked parties is not reported. These gaps do not invalidate the posted summaries, but they limit assessment of their robustness and of how closely the reported analyses align with the intended trial design.

15. 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 0% reporting threshold.

Most frequent non-serious adverse event terms

The registry reports serious adverse event figures for 39 participants at risk in the active tDCS group and 36 at risk in the sham tDCS group. It records zero serious adverse events in each group. The record lists no serious adverse event terms. These figures describe the posted serious-event counts; they do not establish that every possible harm was assessed or that no event could occur outside the recorded reporting period.

For other adverse events, the registry reports zero affected participants among 39 at risk in the active group and 2 among 36 at risk in the sham group. The death counts are 0 for active tDCS and 1 for sham tDCS. The registry does not provide further detail about event types, timing, severity, relationship to stimulation, or the circumstances of the recorded death. The risk denominators are not the same as the group counts in the primary outcome summary, and the record does not explain the difference.

These safety counts should be read descriptively. With the limited detail posted, they cannot establish whether an event was caused by an intervention or support a broad comparison of safety between groups.

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

The primary result is an uncertain comparison of mean change in brain functional connectivity between active tDCS and sham tDCS. The point estimate favors a lower mean change in the active group on the recorded scale, but the confidence interval includes zero and extends in both directions. The unadjusted p-value does not provide strong evidence against a no-difference model. Taken together, the posted statistics do not establish a clear between-group difference in this endpoint.

That conclusion is not the same as proving that the groups were identical. The confidence interval leaves room for more than one possible underlying contrast, and the registry does not give enough detail to evaluate all sources of uncertainty. The measure itself is recorded with an incomplete unit description, and the statistical analysis plan, missing-data procedures, and analysis population are not reported.

The outcome is a brain functional connectivity measure over the stated 1 week timeframe. It is not a direct measure of abstinence. Although the registry’s rationale identifies connectivity as a potential biomarker and intervention target, a biomarker result does not by itself demonstrate a change in clinical outcome. The listed secondary correlation outcome concerns the relationship between connectivity change and length of abstinence, but a correlation is not proof of causation and does not replace a randomized clinical outcome comparison.

The posted results also do not establish that tDCS does or does not produce a clinical benefit. They describe the reported summaries and their uncertainty for this trial’s endpoint. Conclusions about other populations, different stimulation approaches, longer-term outcomes, or clinical use are not established by these statistics. The registry’s eligibility descriptions—alcohol use disorder, MRI eligibility, and short-term abstinence—also define a particular study context, and the record does not provide enough participant detail to assess broader applicability.

17. Why This Trial Matters Statistically

This trial illustrates why a biomarker-focused endpoint and a clinical outcome should be kept distinct. Functional connectivity was selected as the primary measure because the registry’s rationale treats it as a potential intervention target related to abstinence. The posted primary comparison, however, quantifies a brain measure over a short timeframe. The secondary outcome connects connectivity change with length of abstinence, but it is a correlation question rather than a direct randomized comparison of abstinence outcomes.

The result also shows why effect estimates and uncertainty should be read together. A negative point estimate gives a direction for the observed active-versus-sham contrast, but its confidence interval includes zero and spans positive and negative values. A single p-value cannot convey that range or tell readers whether the endpoint is clinically important. Interpreting the estimate requires attention to the measure, its scale, the interval, and the limits of the posted analysis details.

Finally, the registry demonstrates the difference between trial enrollment and an outcome-specific result count. Enrollment is listed as 86, while the posted primary summaries report group-specific counts. Without an explanation of analysis populations and missing data, readers should not infer why those counts differ. The planning illustrations are also distinct from the trial’s original design assumptions, which are not reported. Together, these points make transparent reporting of endpoint definitions, analysis plans, and participant flow essential for evaluating a trial’s statistical evidence.

18. 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 are specific to the information posted for this trial. The primary endpoint’s unit is incomplete in the registry record, which makes it harder to interpret the scale of the mean change. The record gives a short timeframe for the primary outcome but does not identify the connectivity measure’s component regions or connections, acquisition details, or processing approach. Those omissions limit reproducibility and the ability to judge what a given change represents.

The result summaries include group means and standard errors, but the registry does not report participant-level distributions, the analysis model, covariate adjustment, or handling of repeated measures. It also does not explain the analysis population or missingness. Enrollment is listed separately from the outcome counts, but no participant-flow explanation connects them. The safety risk denominators also differ from the primary outcome counts, without an explanation in the record.

The study is described as single-masked, but the identity of masked parties and the success of masking are not reported. Randomization is listed, while sequence generation and allocation concealment details are not. The record does not report the original sample-size assumptions, a complete statistical analysis plan, or a multiplicity strategy. Planning illustrations based on the observed effect cannot fill those gaps.

Applicability is also bounded by the registry’s stated context: alcohol use disorder, eligibility for MRI scanning, and short-term abstinence. The record provides limited participant characteristics with which to judge how well the results apply beyond that context. These limitations concern the interpretation and completeness of the registry record; they do not change the values posted for the outcome.

19. Related Tutorials

20. Related Statistical Calculators

21. Sources

• ClinicalTrials.gov: Clinical Intervention in Alcohol Use Disorder — NCT02168400.
• Registry record last updated 2024-06-05; status last verified 2024-06.
• Results first posted 2024-06-05.
• 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.

22. Record Summary

The registry describes a completed, randomized, single-masked, parallel-group study of active versus sham tDCS in people with alcohol use disorder who were eligible for MRI scanning and short-term abstinent. Enrollment is listed as 86. The primary endpoint is mean change in brain functional connectivity at 1 week. Its posted active-versus-sham estimate is -0.14, with a confidence interval that includes zero and an unadjusted p-value of 0.476.

The registry also lists a secondary correlation outcome relating connectivity change to length of abstinence at 8 months. It reports no analyses and does not provide the trial’s original design assumptions, full statistical plan, analysis population, or missing-data procedures. Safety counts are posted for serious and other adverse events, with limited event detail. Overall, the record supports a cautious reading of an uncertain brain-measure comparison and does not establish a clinical abstinence effect.