Transcranial Direct Current Stimulation Augmented Exposure and Response Prevention for Obsessive-Compulsive Disorder: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Transcranial Direct Current Stimulation Augmented Exposure and Response Prevention for Obsessive-Compulsive Disorder (NCT02329587), a randomized na study. The primary outcome is Yale-Brown Obsessive Compulsive Scale (Y-BOCS): Total Score.

2. Clinical Question

The study addresses whether adding active tDCS to ERP is associated with a different post-treatment OCD symptom-severity score than adding sham tDCS to ERP. This is an active-versus-sham comparison on a shared behavioral-therapy background; it is not a comparison of ERP against no ERP. The registry summary frames tDCS as a possible way to augment ERP, while the posted results describe the scale scores and effect-size analysis without establishing a treatment benefit.

The primary outcome is continuous: it is a score rather than a count of events or a binary response. The reported mean score was 15.5 in the ERP plus tDCS group, based on 4.0 participants, and 16.33 in the ERP plus sham tDCS group, based on 6.0 participants. Higher Y-BOCS scores correspond to greater symptom severity. The record’s reported effect size is 0.138, but it does not include a confidence interval or p-value. Those omissions limit how precisely the difference can be assessed and whether the observed contrast is compatible with a range of other effects.

The registry identifies the study as randomized, double-masked, and parallel. These design features frame the comparison, but the record does not describe the allocation procedure, the people masked, or how masking was maintained. Those details are not reported in the registry record.

3. Trial Design

The registry record identifies NCT02329587 as a randomized, double-masked, parallel-group study of ERP with active tDCS versus ERP with sham tDCS. The sponsor is Butler Hospital. The record lists the study status as completed, enrollment as 13, a start date of 2015-01, and primary completion in 2017-02. The phase is recorded as NA.

Both groups received ERP. The active-stimulation group received ERP plus anodal tDCS of the right inferior frontal gyrus; the comparison group received ERP plus sham tDCS of the right inferior frontal gyrus. This design is intended to separate the contrast associated with active rather than sham stimulation while keeping ERP present in both assigned groups. The record does not give further details about the ERP program or the stimulation schedule, intensity, or number of sessions; these details are not reported in the registry record.

The primary outcome was the Y-BOCS total score at post-treatment, approximately 6.5 weeks post-baseline. The primary outcome population is described as all participants with available data at that timepoint. In the posted outcome results, the ERP plus tDCS group has 4.0 participants and the ERP plus sham tDCS group has 6.0 participants. These outcome-specific counts should not be assumed to describe every randomized participant or to explain why outcome data are unavailable for others.

Three secondary outcomes are named: Y-BOCS total score, Client Satisfaction Questionnaire-8 total score, and rates of retention in intervention. The record lists 6 outcome measures and 6 analyses overall. It does not explain the relation between those counts and the named secondary outcomes or give a full analysis plan. The randomization method, allocation concealment, masking roles, sample-size justification, interim analyses, and details of the statistical plan are not reported in the registry record. The design label alone does not establish how these procedures were carried out.

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: A baseline assessment was completed between participant enrollment and randomization. Withdrawal during the baseline assessment could result in enrolled participants not being assigned to a treatment group.

Baseline Thru Post-treatment Assessment

1-month Follow-up

Analysis population for the primary outcome, as defined in the registry: All participants with available data at the outcome timepoint

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 Yale-Brown Obsessive Compulsive Scale (Y-BOCS): Total Score, measured over Post-treatment (approximately 6.5 weeks post-baseline).

The primary endpoint is the total score on the Yale-Brown Obsessive Compulsive Scale (Y-BOCS), measured after treatment at approximately 6.5 weeks post-baseline. The registry describes the Y-BOCS as a well-known measure for assessing OCD symptom severity. Scores range from 0-40, with higher scores corresponding to greater symptom severity.

This is a continuous outcome, and the posted summaries use a mean and standard deviation for each group. A mean describes the group’s average score; a standard deviation describes the spread of scores among participants. These summaries help characterize both the typical observed score and how much individual scores varied, but they do not show the complete distribution of scores.

A symptom-severity scale is useful for comparing group outcomes because it retains more information about score levels than a simple classification of participants as improved or not improved. However, the registry record does not report a threshold for clinically meaningful change, a responder definition, or a change-from-baseline analysis. The reported endpoint is a post-treatment score, not a reported measure of each participant’s change from baseline.

The posted means are 15.5 for ERP plus tDCS and 16.33 for ERP plus sham tDCS. The corresponding standard deviations are 8.5 and 3.78. Outcome data are available for 4.0 and 6.0 participants, respectively. The record does not report confidence intervals for these group means. As a result, the means and standard deviations describe the observed data but do not by themselves quantify the uncertainty around the between-group contrast.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The primary outcome is continuous and is reported as a mean with a standard deviation in each group. The sponsor’s analysis is described as a between-group effect size (Hedge's g), with a value of 0.138. The registry record does not report the method used to calculate this effect size, its confidence interval, or a p-value. No additional two-group recalculation is reported because the record indicates that values are missing or not estimable for that comparison.

What does a mean describe?

For a continuous measure such as the Y-BOCS total score, a group mean represents the average of participants’ observed scores. It is a compact description of the group, not a score that every participant necessarily had. The mean can be affected by unusually high or low individual values, and it does not describe the full distribution. In this trial, the posted means are 15.5 and 16.33. They describe the participants whose outcome data were available in each group at the specified timepoint.

What does a standard deviation describe?

A standard deviation describes how spread out individual scores are around the group mean. It is not the same as the uncertainty in the mean and is not a confidence interval. The posted standard deviations are 8.5 for ERP plus tDCS and 3.78 for ERP plus sham tDCS. A larger standard deviation indicates more variability among observed scores, while uncertainty in a mean also depends on the number of observations and the assumptions used for inference. With small outcome groups, both the observed spread and the group size matter when judging how stable the summaries may be.

What is an effect size?

Hedge's g is a standardized effect-size measure used to express a contrast in units related to variability, rather than in the original Y-BOCS score units. In general, standardization can make effects on different scales easier to compare, but the estimate still depends on the data and analysis choices. A point estimate is only one summary: a confidence interval would describe the uncertainty around it, and a p-value would address compatibility with a specified null hypothesis under an analysis model. Neither is reported for the sponsor’s effect-size analysis. The registry also does not give the calculation method or direction convention, so the value 0.138 should not be given a directional interpretation based on its sign alone.

Why are confidence intervals useful?

In general, a confidence interval gives a range of estimates compatible with the observed data under the specified statistical method. Its width indicates precision: wider intervals represent less precise estimates, while narrower intervals represent greater precision. An interval also helps readers assess whether the data are compatible with effects that could matter in different directions or magnitudes. No confidence interval is reported for the group means or the sponsor’s effect size in the registry record. The precision of the primary between-group result therefore cannot be assessed from an interval.

What can a p-value tell a reader?

In general, a p-value measures how compatible data are with a specified null model, assuming that model and the analysis assumptions apply. It does not measure the probability that a treatment effect is real, and it does not show whether an effect is important to participants. The registry record does not report a p-value for the effect-size analysis. No conclusion about statistical significance should be inferred from the point estimate or the group means in its absence.

Why are recalculations limited?

A simple unadjusted recalculation from group summary statistics can be useful when the reported data are sufficient and the calculation is estimable. It may not match a sponsor’s analysis if the sponsor used a different model, an adjusted estimate, a different population, or other information not captured in the summary statistics. For this outcome, a two-group recalculation is not reported because the record marks the comparison as having values missing or not estimable. Consequently, there is no recalculated estimate to compare numerically with the sponsor’s result. The record does not indicate a discrepancy between a recalculation and the sponsor analysis.

What do the planning assumptions mean?

In general, power planning asks how many participants would be needed to have a specified chance of detecting an effect of a chosen size under a statistical model and significance threshold. The registry’s planning assumptions describe two-sided significance level 0.05, equal allocation between the two arms, and a normal approximation. The planned effect is defined as a stated fraction of the observed effect. Such sample-size figures are planning illustrations based on those assumptions, not observed power and not evidence about the trial’s design justification. The trial’s own sample-size justification is not reported in the registry record. The planning assumptions also do not resolve the uncertainty in the observed effect-size estimate.

These methods distinctions matter because the posted outcome consists of small-group descriptive summaries and a sponsor-reported effect size without a reported uncertainty interval or p-value. The trial’s randomization and masking labels do not replace reporting of the analysis model, missing-data method, or participant-level analysis population. Those details are not reported in the registry record, so the analysis cannot be reconstructed or fully evaluated from the posted information.

10. Confidence Intervals for Each Arm

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

R code and output

arm <- c("ERP Plus tDCS", "ERP Plus Sham tDCS")
n   <- c(4, 6)
m   <- c(15.5, 16.33)
sp  <- c(8.5, 3.78)
sd  <- sp      # posted standard deviations
for (i in seq_along(n)) {
  se <- sd[i] / sqrt(n[i]); h <- qt(0.975, n[i] - 1) * se
  cat(sprintf("CARM|%s|%d|%.3f|%.3f|%.3f|%.3f\n", arm[i], n[i], m[i], sd[i], m[i] - h, m[i] + h))
}

CARM  ERP Plus tDCS  4  15.500  8.500  1.975  29.025
CARM  ERP Plus Sham tDCS  6  16.330  3.780  12.363  20.297

11. Analyses Reported by the Sponsor

The sponsor’s own reported analysis is a between-group effect size (Hedge's g) of 0.138 for ERP plus tDCS compared with ERP plus sham tDCS. The registry record does not report a confidence interval, p-value, or the method used to estimate this effect size. Without those details, the effect-size estimate cannot be assessed for precision or connected to a stated significance test.

A separate unadjusted re-calculation is not reported for the primary continuous outcome because the record indicates that values are missing or not estimable for a two-arm comparison. The posted group means and standard deviations are not used to generate a new contrast or effect-size estimate. There is therefore no recalculated result to compare numerically with the sponsor’s analysis, and the record does not identify a discrepancy.

In general, a re-calculation from summary statistics may differ from a sponsor analysis if the analysis methods or data differ. Here, the registry record does not explain the sponsor’s method or provide enough information to establish what method was used. The sponsor’s estimate should be described as the result reported in the registry, while its uncertainty and analytical details remain unreported. The posted means and group sizes add descriptive context but do not fill those reporting gaps.

12. Multiplicity, Missing Data and Analysis Populations

The record names a primary Y-BOCS total score outcome and lists three secondary outcomes: Y-BOCS total score, Client Satisfaction Questionnaire-8 total score, and rates of retention in intervention. It also lists 6 outcome measures and 6 analyses overall. The relationship among these entries is not explained in the registry record. A multiplicity plan describing how inference across outcomes or analyses was handled is not reported in the registry record.

Multiplicity matters in general when several outcomes or statistical tests are considered, because interpreting many results without a prespecified approach can increase the chance of highlighting an unusual result. The record does not report whether adjustments were made, whether outcomes were tested in a planned order, or which analyses were considered confirmatory. No conclusion about multiplicity adjustment should be inferred from the presence of a named primary outcome alone.

The primary outcome population is described as all participants with available data at the outcome timepoint. The posted group counts are 4.0 for ERP plus tDCS and 6.0 for ERP plus sham tDCS, while the overall enrollment is 13. The record does not explain why the outcome counts differ from enrollment. It does not report the amount or reasons for missing outcome data, a missing-data method, or sensitivity analyses addressing missingness. It is therefore not possible to assess whether outcome availability was related to treatment assignment or participants’ scores.

The analysis population is not identified as intention-to-treat, per-protocol, or another named population. The registry does not describe how participants were handled if they did not complete intervention or lacked an outcome measurement. The phrase “all participants with available data” identifies who contributes to the endpoint as described, but does not establish how randomization was incorporated into the analysis or how missing outcomes were treated. These omissions limit interpretation of the group comparison and its representativeness.

The record also does not report interim analyses, stopping rules, or details of the statistical analysis plan. Those details are not reported in the registry record. In the absence of those details, the listed outcome and analysis counts cannot establish how the statistical testing was organized.

13. 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 events for 5 participants at risk in the ERP plus tDCS group and 7 participants at risk in the ERP plus sham tDCS group. The number affected by serious adverse events is 0 in each group. Deaths are also reported as 0 in each group. These are the posted serious-event counts and should be read with their group-specific numbers at risk.

Zero reported serious adverse events means none were recorded among the participants at risk in these groups during the period covered by the registry’s safety reporting. It does not establish that no event could occur, and it does not by itself provide a precise comparison of safety between assignments. With limited numbers at risk, the absence of observed serious events leaves uncertainty about events that might occur in a larger group or over another observation period.

The registry separately reports other adverse events affecting 4 participants at risk in the ERP plus tDCS group and 5 participants at risk in the ERP plus sham tDCS group. These are not the serious-adverse-event counts. The record does not list serious event terms and does not provide further detail about event definitions, timing, severity categories for other events, or how long participants were followed for safety. The outcome group sizes and safety numbers at risk are not the same reported quantities; the registry does not explain their relationship.

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

The posted results describe a randomized, double-masked comparison of ERP plus active tDCS and ERP plus sham tDCS. For the primary Y-BOCS outcome at post-treatment, approximately 6.5 weeks post-baseline, the reported means are 15.5 and 16.33, with standard deviations of 8.5 and 3.78, respectively. The outcome counts are 4.0 and 6.0. Higher Y-BOCS scores correspond to greater OCD symptom severity.

The sponsor reports Hedge's g of 0.138 as the between-group effect size. The record does not report a confidence interval, p-value, method for the effect-size analysis, or a directional sign convention. The available information therefore does not establish the precision of the estimate or support a formal conclusion about statistical significance. The registry also does not report a threshold for clinically meaningful change, a responder definition, or a change-from-baseline result. The observed summaries alone cannot establish whether a difference would be important to participants.

These results should not be read as proof that active tDCS improves ERP outcomes, nor as proof that the two assignments have the same effect. The data describe the participants with available outcome measurements and the sponsor’s reported effect-size estimate. The outcome groups are small, and the registry does not explain the difference between overall enrollment and the primary-outcome counts. The analysis population and missing-data handling are not fully described.

The trial’s design labels, including randomized and double-masked, are relevant context but do not substitute for details about allocation, masking roles, analysis methods, or missing data. The record does not report a sample-size justification or the full statistical plan. Without those details and without uncertainty estimates, the result is best treated as a limited estimate from the posted data rather than a definitive comparison.

The safety entries report no serious adverse events and no deaths in either group among the stated numbers at risk. Those counts do not establish comparative safety, particularly given the limited numbers at risk and the absence of additional reporting details. In general, a result from a small study can be informative for describing observed data and methods, while still leaving considerable uncertainty about effects beyond the measured participants and timepoint.

15. Why This Trial Matters Statistically

This trial is statistically useful as an example of why a point estimate and descriptive group summaries are not enough for a complete interpretation. The registry reports mean Y-BOCS scores, standard deviations, outcome-specific group sizes, and a sponsor-reported Hedge's g. However, it does not report a confidence interval or p-value for the effect-size analysis. Readers can see the observed summaries, but cannot use the registry record to assess the precision of the between-group estimate through an interval or a stated hypothesis test.

The trial also illustrates the distinction between total enrollment and the population contributing to a specific endpoint. Enrollment is 13, while the primary outcome counts are 4.0 and 6.0. The primary population is defined by available data at the outcome timepoint, but the record does not explain the difference in counts or describe missing-data handling. This makes analysis-population reporting important: a randomized design does not by itself tell readers who contributed to an outcome analysis.

The active-versus-sham setup holds ERP as a shared intervention in both groups, making the stimulation assignment the main planned distinction between them. Yet the record does not report enough procedural or analytic detail to evaluate allocation, masking roles, or the sponsor’s effect-size method. These omissions show how design labels and concise results can coexist with substantial uncertainty about execution and analysis.

Finally, the record lists multiple outcomes and analyses without reporting a multiplicity plan, and it does not provide a sample-size justification. In general, these details help readers understand which comparisons were planned and how much weight to place on the results. The study therefore highlights the value of transparent reporting of analysis plans, uncertainty, missing outcomes, and participant populations alongside the observed clinical scores.

16. 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 outcome results are based on 4.0 participants in the ERP plus tDCS group and 6.0 in the ERP plus sham tDCS group. The registry does not explain the relationship between these outcome counts and the overall enrollment of 13. Reasons for unavailable outcome measurements and methods for handling missing data are not reported in the registry record. This limits assessment of potential selection effects in the observed endpoint population.

The sponsor-reported effect size is 0.138, but the registry does not report a confidence interval, p-value, calculation method, or directional convention. The record therefore does not show the precision of this estimate or allow an independent comparison with a recalculated result. The listed group means and standard deviations are descriptive summaries, not substitutes for uncertainty estimates about a between-group effect.

The trial is described as randomized, double-masked, and parallel, but the record does not specify how randomization was performed, how allocation was concealed, who was masked, or whether masking was assessed. The details of ERP delivery and tDCS administration beyond the anodal right inferior frontal gyrus description are not reported in the registry record. These omissions constrain assessment of intervention implementation and possible sources of bias.

The primary endpoint is a post-treatment Y-BOCS score at approximately 6.5 weeks post-baseline. The record does not report a change-from-baseline result, a clinically meaningful threshold, or a responder definition. It also does not report longer-term outcomes in the primary result described. The listed secondary outcomes include satisfaction and retention, but their results are not described in the registry record for this interpretation.

Finally, the registry does not report a sample-size justification, multiplicity approach, interim analyses, or full statistical plan. Safety reporting gives group counts at risk and affected but not event-term detail for serious events. These trial-specific reporting gaps leave uncertainty about precision, analysis choices, outcome completeness, and the breadth of conclusions that can be drawn from the posted results.

17. Related Tutorials

18. Related Statistical Calculators

19. Sources

• ClinicalTrials.gov: Transcranial Direct Current Stimulation Augmented Exposure and Response Prevention for Obsessive-Compulsive Disorder — NCT02329587.
• Registry record last updated 2018-05-15; status last verified 2018-04.
• Results first posted 2018-05-15.
• 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.

20. Record Summary

NCT02329587 is a completed randomized, double-masked, parallel-group study sponsored by Butler Hospital, with enrollment of 13. It compared ERP plus anodal tDCS of the right inferior frontal gyrus with ERP plus sham tDCS of the right inferior frontal gyrus in people with obsessive-compulsive disorder.

The primary outcome was post-treatment Y-BOCS total score at approximately 6.5 weeks post-baseline. The reported means were 15.5 (standard deviation 8.5; 4.0 participants) for ERP plus tDCS and 16.33 (standard deviation 3.78; 6.0 participants) for ERP plus sham tDCS. The sponsor reported Hedge's g of 0.138. A confidence interval, p-value, and effect-size method are not reported in the registry record.

The record lists secondary outcomes of Y-BOCS total score, Client Satisfaction Questionnaire-8 total score, and retention in intervention. Serious adverse events and deaths were reported as 0 in both groups. Missing-data handling, sample-size justification, multiplicity procedures, and full analysis details are not reported in the registry record.