Transcranial Direct Current Stimulation (tDCS) As A Treatment For Cigarette Craving and Cognitive Deficits in Schizophrenic: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Transcranial Direct Current Stimulation (tDCS) As A Treatment For Cigarette Craving and Cognitive Deficits in Schizophrenic (NCT02128919), a randomized na study. The primary outcome is Change From Baseline in Cigarette Craving.

2. Clinical Question

The study asked whether cigarette-craving scores changed differently after active transcranial direct current stimulation (tDCS) than after sham stimulation in people with schizophrenia who were current smokers or had a history of regular cigarette smoking. The primary outcome was change from baseline on the Brief Questionnaire on Smoking Urges (QSU-Brief), assessed after five tDCS sessions. The registry also describes assessments of smoking, cognition, and psychiatric symptoms, but the primary question concerns craving.

The registry lists schizophrenia, cognition, and cigarette smoking as conditions. Its study summary describes smoking assessments using CO monitoring and nicotine measures, craving assessments using the QSU scale and responses to craving slides, and cognition and symptom assessments. The primary endpoint is specifically the QSU-Brief change from baseline, rather than a broad measure of all smoking or psychiatric outcomes.

3. Trial Design

This was a randomized, parallel-group study with 33 participants enrolled. The registry records the study as completed and lists the phase as NA. It identifies Nathan Kline Institute for Psychiatric Research as the sponsor. The study began in 2012-04, and primary completion was recorded as 2016-12. Those dates describe the registry record and do not provide details about recruitment duration or follow-up beyond the stated outcome timeframe.

Participants were assigned to active tDCS or sham tDCS. The active regimen was 2 ma for 20 minutes once a day for five days, with the anode at the DLPFC. Sham stimulation was 2 ma for 40 seconds on each of five days, also with the anode at the DLPFC. The registry identifies the design as randomized, triple-masked, and parallel. Its brief study summary calls the study double-blind. The record does not explain the difference between those masking descriptions, so the exact roles covered by masking are not reported in the registry record.

The randomized sample is reported as 24 male and 9 female participants. The primary QSU-Brief population is described as 31 participants: 15 in the active group and 16 in the sham group who completed the questionnaire at baseline and after five sessions. Thus, the primary outcome estimate is based on the completion population described for that measure, not on all 33 randomized participants. The record does not state why all randomized participants did not complete the QSU-Brief at both time points.

The registry lists one device intervention, tDCS. Its summary describes measures of smoking, craving, cognition, and psychiatric symptoms. The posted secondary outcome is change from baseline in cognitive performance. Three outcome measures are listed, while the record does not include posted analysis entries. Further details about the allocation procedure, the masking procedures, the treatment of protocol deviations, and the trial's original statistical design are not reported in the registry record.

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

Analysis population for the primary outcome, as defined in the registry: Findings are based on31 subjects (15 active tDCS and 16 sham tDCS) who completed this QSU-Brief questionnaire in at baseline and after 5 tDCS sessions.. Complete sample who entered study and were randomized were 33 subjects 24 male and 9 female, but all subjects did not complete QSU brief rating scale..

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 Change From Baseline in Cigarette Craving, measured over Baseline and after 5 tDCs sessions (mean time 8.7[SD 2.7] days after basleine).

The primary endpoint was change from baseline in cigarette craving measured with the QSU-Brief. Participants responded to 10 statements about craving for smoking. Each item used a scale from 1, “Strongly Disagree,” to 7, “Strongly Agree”; the item responses were summed for a total score. The registry describes the possible total score on an occasion as ranging from 7 to 70. It states that scores closer to 1 after treatment would indicate a better outcome.

A craving questionnaire captures participants’ reported urges rather than relying only on a behavioral or biological measure. In general, a repeated score can describe whether reported craving changes over time, while a comparison between randomized groups can describe whether the average changes differ between the assigned regimens. The interpretation depends on the exact definition and direction of the change score. The registry names the endpoint as change from baseline but does not specify the subtraction convention used to form the change. Therefore, the numerical sign of a change should not be translated into improvement or worsening without that convention.

The assessment was made at baseline and after five tDCS sessions. The recorded timeframe gives a mean assessment time of 8.7 days after baseline, with SD 2.7. The primary outcome is a continuous score, and its posted arm summaries are means with standard errors. These are group-level summaries; they do not show the distribution of individual scores or how many participants moved in a particular direction.

The study summary mentions other measures, including smoking assessments, cognition, and psychiatric symptoms. Those measures may address related questions, but they are not substitutes for the stated primary craving endpoint. The registry record does not report enough detail to treat the QSU-Brief as a direct measure of cigarette consumption or of every aspect of craving.

7. Results Posted in the Registry

• Active tDCS: mean 14.40 versus 11.00 on control; difference 3.40 (95% CI -6.33 to 13.13), standardized effect d = 0.25; z-test p = 0.493.

8. Statistical Interpretation

9. Statistical Methods Explained

The posted comparison is a comparison of group means for a continuous outcome. For each arm, the registry reports a mean endpoint value and a standard error. The active group has a mean of 14.4 and standard error of 3.57; the sham group has a mean of 11.0 and standard error of 3.45. The primary outcome population sizes are 15 and 16, respectively. A mean describes the average endpoint value in a group, while the standard error describes uncertainty in the estimated group mean. A standard error is not the same quantity as the spread of individual participant scores.

What does the mean difference represent?

The reported comparison is the active-group mean minus the sham-group mean. Its estimate is 3.4. This summarizes the separation between the two sample means on the endpoint scale. It does not say that every participant in the active group had a higher score than every participant in the sham group. Individual scores can vary around each group average, and the registry does not provide individual-level data or a distribution of scores.

The difference is unadjusted: it compares the posted group summaries without adjustment for other variables. Randomization is intended to make treatment groups comparable on average, but the record does not provide baseline covariate details or an adjusted analysis. An unadjusted comparison is not inherently invalid, but it answers a particular statistical question and may not match a planned model. The registry record does not report the original analysis plan or any covariate adjustment.

How should the confidence interval be understood?

The interval from -6.33 to 13.13 expresses uncertainty about the mean difference under the calculation used. In general, a confidence interval is a range produced by a statistical procedure that, over repeated samples under its assumptions, would cover the target parameter at the stated confidence level at the stated long-run rate. It is not a range containing a fixed proportion of individual participant outcomes, and it is not a probability statement that this particular interval contains the true value.

Because the interval covers zero and values on both sides of zero, it does not isolate a single direction of difference. Its breadth reflects limited precision. A confidence interval is more informative than a point estimate alone because it shows the scale of uncertainty compatible with the observed summaries. It does not resolve whether any possible difference is important; the registry does not report a threshold for a meaningful change in QSU-Brief score.

What does the p-value convey?

The unadjusted p-value is 0.493. In general, a p-value measures how compatible the observed test result is with a specified null model, assuming that model and the analysis assumptions are correct. It is not a measure of effect size, does not establish the null model as true, and does not quantify the chance that the observed result is due to chance. The record does not report a prespecified testing threshold or whether this was the only planned analysis of the primary endpoint.

Because the p-value is unadjusted, it does not account for a family of multiple tests. The registry lists multiple outcome measures and describes several domains of assessment, but it does not report a multiplicity plan. The number of outcomes alone is not enough to reconstruct the intended testing strategy or to determine whether an adjustment was planned.

Why is the standard deviation described as approximate?

The recalculated comparison includes a pooled standard deviation of 13.81 and marks it as approximate. The arm-level dispersion values are standard errors, not standard deviations. In general, standard deviation describes variation among individual observations, while standard error describes uncertainty in a sample mean and depends on sample size as well as the underlying variation. Recovering a standard deviation from a standard error requires assumptions about the underlying summary data. The approximate designation signals that the pooled dispersion used for the recalculation is not a directly posted pooled standard deviation.

Over-dispersion is a general concern when the observed variation is greater than a statistical model assumes. If a model understates variation, standard errors can be too small and intervals too narrow; if it overstates variation, estimates can be less precise than necessary. The registry does not report a model for over-dispersion, a distributional assessment, or a separate variance analysis. The continuous mean comparison should therefore be understood as a simple summary-based analysis, not as evidence that all modeling assumptions were checked.

What does recalculated mean?

The posted comparison is labelled as a recalculation from summary statistics. In general, such a calculation uses the available group-level means, dispersion information, and sample sizes to form an unadjusted estimate and its uncertainty. It cannot reproduce analyses that depend on participant-level measurements, baseline covariates, repeated-measures structure, or additional model terms unless those details are available. The registry reports no analysis entries for this outcome, and the trial’s original statistical plan is not reported in the registry record. The recalculated comparison should therefore be read as a transparent, limited analysis of the posted summaries rather than a substitute for an unreported sponsor model.

What assumptions remain unknown?

The registry does not report the original sample-size justification, the planned primary analysis method, interim analyses, handling of missing outcome values, or the exact convention used to calculate change from baseline. It also does not provide participant-level data, baseline QSU-Brief means, or a detailed account of analysis populations. These omissions limit assessment of the prespecified analysis, assumptions, and potential departures from the intended protocol. They do not erase the posted result; they define what can and cannot be inferred from it.

In general, randomized allocation supports a between-group comparison by reducing systematic differences at assignment, but it does not guarantee identical groups in a small sample. The endpoint summaries remain estimates with sampling uncertainty. For that reason, the point estimate, interval, participant counts, and analysis description should be considered together rather than treating the p-value as a stand-alone verdict.

10. Confidence Intervals and Precision

The mean difference is 3.4, with a 95% confidence interval from -6.33 to 13.13. The interval crosses zero and includes values in both directions. This means the estimate is compatible with no mean difference as well as with a range of positive or negative differences under the recalculation. The interval is broad, so the data do not locate the group contrast with high precision.

Precision concerns how narrowly an estimate is determined, not whether the result is important or whether treatment is beneficial. The point estimate alone may appear to suggest separation, but the confidence interval shows that uncertainty is substantial. In general, a confidence interval quantifies sampling uncertainty under the analysis assumptions; it does not include every source of uncertainty, such as missing data, endpoint measurement choices, or an unreported analysis plan.

The outcome summaries are based on 15 active and 16 sham participants. The standard errors for the group means are 3.57 and 3.45. The pooled standard deviation used in the recalculation is marked approximate and is 13.81. These quantities describe different aspects of uncertainty and variability. The standard errors relate to the estimated means, while the standard deviation concerns participant-level spread and is approximate in this comparison.

The confidence interval and unadjusted p-value of 0.493 are consistent with an uncertain comparison, not proof that the groups are equivalent. A conclusion of equivalence would require an appropriate design and a prespecified margin, neither of which is reported in the registry record. The record also does not state a smallest difference considered meaningful for the QSU-Brief. Thus, the precision assessment is statistical: the estimate does not sharply distinguish among plausible group differences, and clinical importance cannot be determined from the interval alone.

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 planning estimates are participants per arm, not total participants and not total events. They are illustrations based on the observed effect, rather than the trial’s own sample-size design. The stated 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 trial’s original sample-size assumptions are not reported in the registry record.

For planning at 100% of the observed effect, the estimate is 260 participants per arm for 80% power and 347 participants per arm for 90% power. Planning for 75% of the observed effect gives 461 participants per arm for 80% power and 617 participants per arm for 90% power. Planning for 50% of the observed effect gives 1,037 participants per arm for 80% power and 1,388 participants per arm for 90% power.

In plain terms, 80% power describes the probability of detecting an effect of the specified size under the stated assumptions if that effect were the true effect. The 90% power column uses a higher probability target for detecting that specified effect. These values are not observed power for the completed trial. They are planning illustrations, and their interpretation depends on the assumptions used to generate them.

The estimates rise as the target effect becomes smaller: a study designed to detect a smaller fraction of the observed effect requires more participants per arm under the stated calculation. The results do not establish that the observed effect is stable or that it is the effect a future study should target. They should be read as conditional planning quantities based on this estimate, not as a recommendation or as the original enrollment target.

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 = 3.400; pooled standard deviation σ = 13.813.
2. n = 2 × (1.96 + 0.842)² × 13.813² / 3.400² = 259.1, rounded up to 260 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 <- 3.4                      # observed difference in means
sigma <- 13.8128238766834                     # 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(2128919)
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 = 260 (80% power), 347 (90% power)
 75% of observed effect: n per arm = 461 (80% power), 617 (90% power)
 50% of observed effect: n per arm = 1,037 (80% power), 1,388 (90% power)
power.t.test: 260.1 
Simulated power at n = 260 per arm: 0.803 (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 record does not report a multiplicity plan. The study summary describes assessments of smoking, craving, cognition, and psychiatric symptoms, and the registry lists three outcome measures, including a secondary outcome of change from baseline in cognitive performance. It does not state how many hypotheses were planned, whether outcomes were tested as a family, or whether any adjustment to account for multiple comparisons was intended. The unadjusted p-value for the primary craving comparison therefore should not be interpreted as a multiplicity-adjusted result.

The record lists no analysis entries. It does not give the original statistical analysis plan, a prespecified model, a testing hierarchy, interim analyses, or a sample-size justification. Those details are not reported in the registry record. Consequently, the posted summary comparison cannot establish whether the observed calculation matches a prespecified analysis, whether additional planned analyses were conducted, or how any such analyses were handled.

The primary endpoint population consists of 31 participants who completed the QSU-Brief at baseline and after five sessions. The registry states that 33 participants entered and were randomized, and that all did not complete the QSU-Brief rating scale. It does not describe the reasons for incomplete outcome data, the pattern of missingness, imputation, or sensitivity analyses. The reported comparison should be understood as applying to the participants represented in the posted endpoint summaries; the record does not establish how results would differ under another missing-data approach.

The registry also does not specify whether the primary analysis population was intended to include all randomized participants, only participants with observed endpoint data, or another defined group. The outcome summary identifies completers, but a full intention-to-treat analysis cannot be inferred from that description. The safety counts use groups at risk of 17 active and 16 sham participants, which are not identical to the primary endpoint counts; the record does not explain the difference between those populations.

These reporting gaps do not by themselves demonstrate that a particular analysis was inappropriate. They limit the ability to evaluate the analysis choices, the handling of missing observations, and the interpretation of multiple outcomes. Results should be read as the posted endpoint comparison with those uncertainties in view.

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.

The registry reports serious adverse events for participants at risk in each arm. In the active tDCS group, 17 participants were at risk and 0 had a serious adverse event. In the sham tDCS group, 16 participants were at risk and 0 had a serious adverse event. The registry also reports 0 other adverse events among 17 active-group participants at risk and 0 among 16 sham-group participants at risk.

These figures describe the counts recorded in the registry, not a comparison adjusted for exposure time or other factors. The record lists no serious adverse-event terms. Deaths are not reported in the registry record. A count of zero in the posted data means no events were recorded in that category for the listed group; it does not establish that an event is impossible or that risks are identical between groups.

The safety population counts differ from the primary outcome completion counts, and the registry does not explain that difference. The available figures do not provide event timing, severity details beyond the serious and other categories, duration of follow-up for safety, or a detailed adverse-event collection method. They also do not establish how completely events were ascertained. These limits matter when interpreting small event counts. The registry figures can be read as the reported counts for each at-risk group, but they do not support broader claims about safety beyond those observations.

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

The posted result describes a randomized comparison of change in a self-reported cigarette-craving score after five sessions of active or sham tDCS. The active-group mean endpoint value was 14.4, compared with 11.0 in the sham group. The unadjusted mean difference was 3.4, with a 95% confidence interval from -6.33 to 13.13 and an unadjusted p-value of 0.493. The interval includes zero and is wide, so the estimate does not provide a precise distinction between the groups.

This result does not show that the groups are equivalent. A non-significant comparison, considered on its own, does not prove that no difference exists. An equivalence or non-inferiority conclusion would require an appropriate design and prespecified criteria, which are not reported in the registry record. Nor does the result show that either regimen caused a particular change in an individual participant’s craving. The summaries are group averages, and individual-level outcomes are not reported.

The direction of the endpoint also requires care. The registry says that lower QSU-Brief scores after treatment indicate a better outcome, but the primary measure is change from baseline and the subtraction convention is not stated. The sign of the active-minus-sham mean difference should therefore not be translated into a direction of benefit without further endpoint-definition detail.

The comparison is limited to the QSU-Brief outcome population and the assessment after five sessions. It does not establish a change in long-term smoking behavior, cigarette consumption, cognition, or psychiatric symptoms. The study summary mentions these other areas, but results for them are not reported in the primary comparison. The registry also does not report the original analysis plan, missing-data handling, or a meaningful-change threshold for the craving scale.

In general, the most defensible interpretation is that the posted summaries yield an uncertain, imprecise estimate of the between-group difference in the stated craving endpoint. The result neither confirms a meaningful difference nor establishes that there is none. Its scope is the reported measure, participants, treatment schedule, and analysis, with the reporting limitations kept in view.

16. Why This Trial Matters Statistically

This trial is statistically informative because it illustrates how a randomized comparison can produce a point estimate that is much less conclusive than its numerical contrast might first suggest. The active and sham mean endpoint values differ, but the confidence interval extends across zero and both directions. Looking at the interval alongside the point estimate makes the uncertainty visible; relying on a single p-value or mean would conceal part of that uncertainty.

The study also demonstrates why the definition of an endpoint matters. The questionnaire measures self-reported cigarette craving, and the registry describes a score direction for post-treatment values. Yet the primary outcome is change from baseline, and the change-score convention is not stated. Statistical interpretation depends not only on the scale but on how the analyzed quantity was constructed. Without that detail, the sign of a change contrast cannot safely be assigned a clinical direction.

The record distinguishes the randomized sample from the primary outcome completion population. The primary summaries include 31 participants rather than all 33 randomized participants. This distinction is important for interpreting the target population of an estimate and for considering missing data. The record does not explain the missing outcomes or the analysis population beyond identifying participants who completed both assessments.

Finally, the planning estimates underline the relationship between effect size, precision, and sample size. Under the stated planning assumptions, detecting a smaller fraction of the observed effect requires more participants per arm. These are illustrations rather than evidence of the trial’s original design or observed power. Together, the result and its reporting context show why statistical conclusions depend on endpoint definition, uncertainty, analysis population, and prespecified methods, not only on whether a p-value appears small or large.

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 primary endpoint population was 31 participants, whereas 33 participants were randomized. The registry says that not all randomized participants completed the QSU-Brief at baseline and after five sessions, but it does not report reasons for incomplete questionnaires or how missing values were handled. As a result, the posted estimate is based on the completion group described for that measure, and the effect of missing outcomes cannot be assessed from the record.

The endpoint is described as change from baseline, but the registry does not specify whether change was calculated as post-treatment minus baseline or by another convention. This makes it difficult to interpret the direction of the contrast against the statement that lower post-treatment QSU-Brief scores indicate a better outcome. Baseline group means and individual score distributions are not reported, limiting assessment of baseline balance, variability, and the pattern of participant responses.

The posted comparison is an unadjusted recalculation from summary statistics, and the pooled standard deviation used in the calculation is approximate. The registry does not report the sponsor’s planned analysis, an adjusted estimate, a prespecified testing threshold, a multiplicity plan, or a sample-size justification. Therefore, the result cannot be checked against a documented statistical analysis plan, and the recalculation may not capture features of an intended participant-level model.

The record describes multiple areas of assessment, but the primary comparison does not report results for the other measures. The stated primary endpoint is a self-reported craving scale, not a direct long-term smoking outcome. The assessment followed five sessions, with a mean timing of 8.7 days after baseline and SD 2.7; longer-term follow-up is not reported in the primary outcome description.

The registry gives both triple-masking and double-blind descriptions without explaining their relationship. It also does not provide details of allocation concealment or masking procedures. Safety counts are reported, but event definitions, collection details, and deaths are not reported in the registry record. These limitations constrain interpretation of both the efficacy comparison and the safety summaries.

18. Related Tutorials

19. Related Statistical Calculators

20. Sources

• ClinicalTrials.gov: Transcranial Direct Current Stimulation (tDCS) As A Treatment For Cigarette Craving and Cognitive Deficits in Schizophrenic — NCT02128919.
• Registry record last updated 2017-08-30; status last verified 2017-08.
• Results first posted 2017-08-25.
• 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, parallel-group study of tDCS for cigarette craving and related outcomes in people with schizophrenia who smoked currently or had a history of regular smoking. The sponsor was Nathan Kline Institute for Psychiatric Research. The record lists 33 randomized participants, with 24 male and 9 female participants.

Active treatment was 2 ma for 20 minutes once daily over five days; sham treatment was 2 ma for 40 seconds on each of five days. The primary endpoint was change from baseline in QSU-Brief craving scores after five sessions. Its outcome population included 15 active and 16 sham participants. The posted unadjusted mean difference was 3.4, with a 95% confidence interval from -6.33 to 13.13 and p = 0.493.

The estimate is imprecise and does not establish a clear between-group difference. The registry does not report the original statistical plan, missing-data handling, or the convention used to calculate change. Safety records list zero serious and zero other adverse events in the reported at-risk groups; deaths are not reported in the registry record.