2/2 D-Cycloserine Augmentation of CBT for Pediatric OCD: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for 2/2 D-Cycloserine Augmentation of CBT for Pediatric OCD (NCT01404208), a randomized na study. The primary outcome is Children's Yale-Brown Obsessive Compulsive Scale (CY-BOCS).

2. Clinical Question

3. Trial Design

This was a randomized, parallel-group trial with quadruple masking. The registry lists the status as completed, the phase as not applicable, and enrollment as 206. Massachusetts General Hospital is identified as the sponsor. The summary says the study was intended to examine whether D-cycloserine taken alongside CBT could help children with pediatric OCD, and states that the National Institute of Mental Health was paying for the study. The registry lists the study start as 2011-07 and primary completion as 2015-09.

Participants were assigned to D-cycloserine or placebo, identified in the posted results as a sugar pill. The summary describes assignment by chance and an equal chance of assignment to either group. It also describes medication before CBT visits 4-10, with the study coordinator giving the capsules at the therapy location. CBT is described as exposure-based therapy in which children practice dealing with situations they usually avoid, progressing from less stressful to more challenging situations. Details of the CBT protocol, including therapist training, session duration, and adherence procedures, are not reported in the registry record.

The primary posted outcome compares change in CY-BOCS score from randomization to post-treatment. The endpoint is a continuous measure reported as a mean with a standard deviation. Results are posted for 70 participants assigned to D-cycloserine and 72 assigned to the sugar pill. These outcome counts do not account for the full registry enrollment of 206. Reasons for the difference, the analysis population definition, and the handling of participants without post-treatment measurements are not reported in the registry record. The registry describes about 75 children as an enrollment aim in its summary, but does not explain how that aim relates to the listed enrollment.

The record lists Clinical Global Impression-Severity as a secondary outcome. It does not report a phase-specific statistical plan, the trial's original sample-size justification, or whether interim analyses were planned or conducted. Those aspects of the design should not be inferred from the posted outcome summaries.

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: 206 subjects were initially enrolled at MGH and USF. Of those, 32 were considered screen fails and 32 were not randomized either because they did not meet inclusion or exclusion criteria at randomization or because they withdrew consent or dropped out. Thus, 142 participants were randomized.

Overall Study

Reasons for not completing this period:

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.

Population: 206 subjects were initially enrolled at MGH and USF. Of those, 32 were considered screen fails and 32 were not randomized either because they did not meet inclusion or exclusion criteria at randomization or because they withdrew consent or dropped out. Thus, 142 participants were randomized.

6. Primary Endpoint

The registry lists the primary outcome as Children's Yale-Brown Obsessive Compulsive Scale (CY-BOCS), measured over Change from Score at Randomization to Post Treatment.

The primary endpoint is change in the Children's Yale-Brown Obsessive Compulsive Scale score from randomization to post-treatment. CY-BOCS measures OCD symptom severity, including interference, time spent on thoughts or behaviors, distress, and resistance. Its scale runs from 0 to 40, with larger values indicating more severe symptoms. The reported outcome is the post-treatment score minus the score at randomization, so a negative value represents a reduction in symptom severity.

A change score gives a direct summary of how the measured symptom severity differed between the start and end of the treatment period. Comparing group means can describe whether the average change differed between assigned groups, while the spread of individual scores indicates how variable the changes were among participants. The endpoint is clinically interpretable because it is based on symptom severity rather than a laboratory measure. However, the registry record does not report a threshold for a clinically important change, a responder definition, or an interpretation of how much change an individual child would regard as meaningful.

The registry reports mean changes and standard deviations for each group, along with a mean difference and confidence interval for the comparison. These statistics describe the group averages and uncertainty around their contrast. They do not establish that each participant experienced the group-average change, nor do they by themselves show whether any difference would be important to patients or families.

7. Results Posted in the Registry

• Sugar Pill: mean -9.40 versus -10.35 on control; difference 0.95 (95% CI -1.48 to 3.38), standardized effect d = 0.13; z-test p = 0.444.

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.

Clinical Global Impression-Severity (CGI-Severity)

• 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 deviations.
• Result:
  • Sugar Pill: mean -1.20 versus -1.39 on control; difference 0.19 (95% CI -0.13 to 0.51), standardized effect d = 0.19; z-test p = 0.247.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Sugar Pill and the control. The interval gives the range of effects that remain compatible with the data.

The registry lists Clinical Global Impression-Severity as a secondary endpoint. Its posted group summaries and comparison point in the same broad numerical direction as the primary outcome when the contrasts are read as sugar pill minus D-cycloserine: the estimated changes are more negative in the D-cycloserine group. This directional consistency can be considered alongside the CY-BOCS result, but it does not turn either estimate into a conclusive finding. Both outcomes are reported as changes from randomization to post-treatment, and both comparisons have uncertainty that should be considered rather than relying only on their point estimates.

Secondary endpoints can provide supporting context about whether the pattern on the primary measure is echoed by another measure. They are not substitutes for the primary outcome, and they may address a related but not identical aspect of severity. The registry record does not provide a detailed description of the secondary measure's scoring interpretation, a hierarchy of secondary outcomes, or a prespecified role for this endpoint in drawing conclusions.

Unadjusted secondary results carry less weight when multiple outcomes or comparisons are considered. The more results examined, the greater the chance that at least one appears notable by chance, even if no real group differences exist. A multiplicity adjustment or a prespecified testing hierarchy can help control that issue, but the registry record does not report such a strategy. The secondary result should therefore be treated as exploratory or supportive rather than as an independent confirmation of the primary finding. No multiplicity-adjusted estimates or reported analyses are listed.

The pattern across the available endpoints is best described without overstating agreement: their numerical contrasts point in a similar direction, but the registry does not report enough about analysis planning, missing data, or multiplicity to establish how these results were intended to be combined. The secondary outcome adds context, not a separate basis for a definitive conclusion.

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

• Clinical Global Impression-Severity (CGI-Severity)

9. Statistical Interpretation

10. Statistical Methods Explained

The primary outcome is continuous: each participant has a CY-BOCS change score, and the group results are summarized using means and standard deviations. A mean describes the arithmetic center of the observed scores. A standard deviation describes how spread out the individual scores are around that mean. These summaries help characterize the outcome, but do not show the full distribution of scores or whether there were unusual values, clusters, or differences in spread between groups.

What does the mean difference represent?

The comparison is expressed as sugar pill minus D-cycloserine. The group means are -9.4 and -10.35, respectively, and the estimated difference is 0.95. With this subtraction order, a positive difference means the sugar-pill mean change is numerically higher, or less negative, than the D-cycloserine mean change. Because negative changes indicate reductions in symptom severity, that direction corresponds to a numerically larger mean reduction in the D-cycloserine group. This description concerns the group averages; it does not imply that every participant followed the same pattern.

What does a confidence interval add?

A confidence interval gives a range of values compatible with the observed data and the assumptions of the method. The primary comparison interval is -1.484 to 3.384. Since zero lies within that range, the posted data do not sharply distinguish a nonzero difference from no difference. Values below zero would imply the contrast points in the other direction under the stated subtraction order, while values above zero would imply a larger mean reduction in the D-cycloserine group. The interval is not a range containing a fixed proportion of individual children's changes, and it is not a guarantee that the true effect lies within its endpoints.

How should the p-value be read?

The primary unadjusted p-value is 0.444. A p-value assesses how compatible the observed result is with a model that assumes no group difference, given the test's assumptions. It does not measure the size or clinical importance of a difference, and it does not state the probability that a treatment effect is absent. The term unadjusted indicates that the posted value is not adjusted for multiple comparisons. Whether the trial's planned inferential approach involved other adjustments is not reported in the registry record.

The registry gives group standard deviations and a mean-difference estimate. The comparison summary also gives a pooled standard deviation and a standardized difference, which express the contrast relative to variability in the outcome. Standardization can help describe the magnitude of an effect on a scale not tied to the original outcome units. It does not replace the CY-BOCS scale or determine whether the contrast matters clinically. A standardized effect is also dependent on the variability of the study population, so it should not be treated as a universal quantity.

Why can over-dispersion matter?

In general, over-dispersion means the observed variability is greater than a model expects. For continuous outcomes, a model that understates variability can make standard errors too small and confidence intervals too narrow. For count outcomes, it can similarly lead to misleading uncertainty estimates. The posted primary endpoint is continuous and its group standard deviations are reported, but the registry record does not report a diagnostic assessment for over-dispersion or other distributional assumptions. No conclusion about such diagnostics can be drawn from the summary statistics alone.

Random assignment is intended to support a fair comparison by making group assignment unrelated to participant characteristics on average. Masking can reduce the influence of expectations on treatment delivery, outcome assessment, or reporting, depending on who is masked. The trial is listed as quadruple masked, but the roles covered and procedures used to preserve masking are not reported in the registry record. Randomization and masking do not remove the need to understand attrition, adherence, measurement, and analysis choices.

The record lists no reported analyses, despite posting outcome summaries and comparison statistics. It does not identify the statistical model used for the primary result, whether baseline scores were included as covariates, how departures from assumptions were assessed, or whether the analysis followed intention-to-treat principles. The comparison statistics should therefore be read as posted, without attributing an unreported model or analysis population to them.

11. Confidence Intervals and Precision

The confidence interval for the primary mean difference ranges from -1.484 to 3.384. It is wide enough to include zero and values on both sides of zero. In practical terms, the posted data do not precisely determine the direction of the group contrast. The point estimate of 0.95 is only one location within a broader range of estimates that remain compatible with the data and the statistical assumptions.

Precision is distinct from the observed size of the difference. A small point estimate can be estimated precisely or imprecisely, and a confidence interval communicates the uncertainty around that estimate. Here, the interval means that the evidence does not rule out a difference favoring either group. It also does not establish that all differences in the interval are equally plausible or clinically important. The registry record does not report a threshold for a meaningful CY-BOCS difference with which to interpret the interval.

The interval concerns the mean contrast between groups, not the range of outcomes among individual participants. The standard deviations describe variability in the observed individual changes; the confidence interval describes uncertainty in the estimated group difference. Those are related but distinct quantities.

Interpretation of precision is also affected by the number and identity of participants whose outcomes were included. The posted primary results include 70 participants in the D-cycloserine group and 72 in the sugar-pill group, while registry enrollment is 206. The reasons for the difference and the treatment of missing outcomes are not reported in the registry record. Without that information, the confidence interval can be read as precision for the posted comparison, but it cannot resolve possible uncertainty related to incomplete outcome data or the analysis population.

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 sample-size figures are planning illustrations based on the observed effect, not a reconstruction of the trial's original design. They give participants per arm, not a total across both arms and not total events. 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 a stated fraction of the observed effect. The trial's own original design assumptions and sample-size justification are not reported in the registry record.

The 80% power column describes the number of participants per arm that a planning calculation estimates would be needed for an 80% chance of detecting the specified effect under the stated assumptions, if that effect were the true effect and the assumptions held. The 90% power column uses a higher target chance of detection, so its participant requirement is larger for the same effect. These are forward-looking planning quantities; they are not measures of what the completed trial actually achieved.

The rows illustrate planning for 100%, 75%, and 50% of the observed effect. A smaller effect is harder to distinguish from random variation, so detecting it with the same power generally requires more participants. The figures are sensitive to the assumptions used, including the assumed effect size and equal allocation. They do not account for design features or data complications that are not included in the stated calculation assumptions.

This is not observed power. Observed power calculated from the estimated effect would largely restate the result and its uncertainty, rather than provide an independent explanation of what the trial established. The useful reading is that these figures show how sample-size needs change across assumed effect sizes under the stated planning model; they do not indicate that the trial was designed with these numbers or that any target was met.

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 = 0.950; pooled standard deviation σ = 7.388.
2. n = 2 × (1.96 + 0.842)² × 7.388² / 0.950² = 949.3, rounded up to 950 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 <- 0.949999999999999                      # observed difference in means
sigma <- 7.38781192234886                     # 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(1404208)
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 = 950 (80% power), 1,271 (90% power)
 75% of observed effect: n per arm = 1,688 (80% power), 2,260 (90% power)
 50% of observed effect: n per arm = 3,798 (80% power), 5,084 (90% power)
power.t.test: 950.3 
Simulated power at n = 950 per arm: 0.802 (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, time points, groups, or statistical comparisons are examined. Even when each comparison is evaluated at a conventional significance level, considering many results increases the chance of seeing an apparently unusual result by chance. Researchers may address this through a prespecified hierarchy, adjusted significance thresholds, or other methods. The registry record does not report whether a multiplicity strategy was planned for the primary and secondary outcomes, nor does it describe how the two posted outcomes were prioritized. The posted comparison p-values are unadjusted.

Missing data also affect interpretation. The registry lists enrollment of 206, while the primary endpoint results include 70 participants assigned to D-cycloserine and 72 assigned to the sugar pill. The record does not report why the listed outcome counts are smaller than enrollment, how many participants lacked a post-treatment measurement, or whether missingness differed between groups. It does not describe imputation, a sensitivity analysis, or another approach to incomplete outcomes. These details are important because analyses restricted to participants with observed outcomes can be affected if the chance of having an outcome recorded is related to prognosis or treatment experience.

The analysis population is not reported in the registry record. In particular, it is not stated whether the primary comparison included all randomized participants, only participants who completed treatment, or another defined group. The registry also does not report adherence-based exclusions or the handling of participants who changed or stopped treatment. No assumptions about those choices should be inferred from the posted group summaries.

The record lists zero reported analyses, although it contains outcome summaries and comparison statistics. That means details such as model specification, covariate adjustment, prespecified testing, and sensitivity analyses are not available in the posted analysis record. The result can still be described as posted, but the absence of those details limits assessment of whether the reported comparison matches the intended analysis plan and how robust it is to alternative assumptions.

15. 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 serious adverse event summaries list 0 events among 70 participants at risk in the D-cycloserine group and 0 among 72 participants at risk in the sugar-pill group. These are counts in the registry record; they do not establish that no safety concerns occurred, since the summary only addresses the reported serious adverse event category and the defined observation period is not described in the posted information.

Other adverse events are listed for 60 of 70 participants in the D-cycloserine group and 70 of 72 participants in the sugar-pill group. The registry summary does not provide event terms, severity, timing, duration, or the number of events per participant. It also does not explain how events were collected or classified. The event counts should not be interpreted as a comparison of rates without further detail about definitions and ascertainment.

Deaths are not reported in the registry record. The serious adverse event listings are descriptive and do not establish whether an event was caused by the assigned study drug or CBT. They should be read as the recorded counts, not as a complete assessment of comparative safety.

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

The posted results describe a randomized comparison of D-cycloserine and sugar pill when given alongside CBT to children with OCD. Both groups had negative average change in CY-BOCS scores from randomization to post-treatment, which means both group averages were lower at the later assessment. The estimated difference between group averages was 0.95, with a confidence interval from -1.484 to 3.384. That interval includes no difference and values favoring either group, so the comparison is uncertain.

This does not show that the groups were exactly equivalent. A confidence interval that includes zero can arise when the data do not distinguish a difference from no difference with sufficient precision; it does not prove that a meaningful difference is absent. The registry record does not define a clinically important difference, so the results cannot be mapped to a threshold of benefit that would matter to children or families.

The comparison also does not answer whether CBT itself changes symptoms. CBT was part of the treatment context for both groups, and there is no CBT-free group in the posted comparison. Nor does the result establish how D-cycloserine would perform outside the particular treatment arrangement described in the registry summary. The record does not provide enough detail to assess the extent to which therapy exposure and adherence were comparable between groups.

These results should not be generalized automatically to all children with OCD. The registry record does not report the characteristics of participants who contributed outcome data, the eligibility criteria in sufficient detail, or the reasons for the difference between enrollment and the outcome counts. Generalizability depends on the participants studied, treatment delivery, outcome measurement, and analysis choices.

The statistical evidence is therefore best understood as a comparison with a modest point estimate and substantial uncertainty, alongside reductions in both posted group means. It neither establishes a reliable advantage for one assigned group nor proves that their effects are identical. The posted summaries describe the recorded outcomes; they do not replace a complete account of the trial's analysis plan, missing data, or clinical importance criteria.

17. Why This Trial Matters Statistically

This trial illustrates why a difference in group means should be read together with its uncertainty. The primary comparison has a point estimate, a confidence interval, and an unadjusted p-value, but those statistics answer different questions. The point estimate describes the observed contrast, the interval conveys uncertainty around it, and the p-value summarizes compatibility with a no-difference model under test assumptions. None alone establishes clinical importance.

The design also highlights the value of distinguishing treatment effects from treatment context. Both groups received CBT, so the randomized comparison concerns the addition of D-cycloserine rather than CBT versus no CBT. That distinction is essential when describing what a trial can answer. Similarly, a randomized and masked design is informative, but interpretation still depends on knowing who was included in the analysis and how incomplete outcomes were handled.

The registry illustrates a common reporting challenge: enrollment and endpoint counts are not the same, yet the reasons for the difference are not described. Without information about missingness and analysis populations, readers cannot fully judge whether the posted comparison represents all randomized participants or a selected subset. Transparent reporting of these details is important for assessing both validity and precision.

Finally, the secondary endpoint and planning illustrations serve different roles. A secondary measure may provide context but needs to be interpreted in light of multiplicity, while planning calculations show how assumed effect sizes relate to participant requirements. Neither should be mistaken for a replacement for the primary result or evidence of what the completed trial was originally powered to detect.

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 details needed for a fuller trial assessment are not reported in the registry record. The participant eligibility criteria and characteristics are not available in the posted outcome summary, so it is difficult to judge how closely the analyzed participants resemble the broader population of children with OCD. The summary describes an intended age range of 7-17 years, but does not provide the age distribution among participants who contributed to the outcome.

The listed enrollment of 206 differs from the primary outcome counts of 70 and 72. The record does not explain this difference, identify the number of missing post-treatment measurements, or describe the analysis population. This limits assessment of attrition and whether the outcome summaries reflect all randomized participants. The registry also does not report adherence, treatment completion, or whether CBT delivery was similar between groups.

Intervention details are incomplete. The summary describes one or two capsules before CBT visits 4-10, but the dose and criteria for the capsule amount are not reported. The identities of the roles covered by quadruple masking and procedures to maintain masking are also not described. These details matter for understanding how treatment was implemented and whether expectations could have influenced care or outcome assessment.

The registry does not provide a sample-size justification for the trial's original design, the prespecified statistical analysis plan, interim analysis details, missing-data methods, or multiplicity strategy. It does not report a threshold for clinically important change on CY-BOCS or a responder definition. The safety summary lacks event terms and details about timing and classification. These omissions do not invalidate the posted numbers, but they limit independent assessment of their robustness, clinical meaning, and completeness.

19. Related Tutorials

20. Related Statistical Calculators

21. Sources

• ClinicalTrials.gov: 2/2 D-Cycloserine Augmentation of CBT for Pediatric OCD — NCT01404208.
• Registry record last updated 2017-03-28; status last verified 2017-02.
• Results first posted 2017-02-27.
• 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, parallel-group, quadruple-masked study sponsored by Massachusetts General Hospital, comparing D-cycloserine with a sugar pill alongside CBT for pediatric OCD. Enrollment is listed as 206. The primary endpoint is mean change in CY-BOCS score from randomization to post-treatment. The posted group means are -10.35 for D-cycloserine and -9.4 for the sugar-pill group, with a mean difference of 0.95 and a confidence interval from -1.484 to 3.384. Both group means indicate reductions, while the comparison remains uncertain. A secondary Clinical Global Impression-Severity outcome is also posted. Details on missing data, analysis populations, multiplicity, the original sample-size rationale, and the clinical importance threshold are not reported in the registry record.