Enhancing Exposure Therapy for Post Traumatic Stress Disorder (PTSD): Virtual Reality and Imaginal Exposure With a Cognitive Enhancer: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Enhancing Exposure Therapy for Post Traumatic Stress Disorder (PTSD): Virtual Reality and Imaginal Exposure With a Cognitive Enhancer (NCT01352637), a randomized phase 3 study. The primary outcome is CAPS-IV at the End of Treatment.

2. Clinical Question

3. Trial Design

This was a Phase 3, randomized, triple-masked, parallel-design study. The registry reports enrollment of 192 participants and an allocation ratio of 1:1:1:1 across four active treatment conditions. The conditions crossed two exposure approaches—VRE and PE—with two medication assignments—DCS and placebo. This arrangement permits consideration of the exposure approach and the medication assignment as separate treatment factors, although the posted summary results do not report an interaction between those factors.

The registry lists the study start as 2011-05, primary completion as 2018-09, and status as completed. It identifies Weill Medical College of Cornell University as the sponsor. The study description focuses on combat-related PTSD among active duty military personnel, veterans, and civilians who served in Iraq and Afghanistan. The full inclusion and exclusion criteria, recruitment methods, and details of the masking procedures are not reported in the registry record. In particular, the meaning of triple masking in terms of which roles were masked is not described in the posted information.

The primary outcome was CAPS-IV total severity after weekly treatment session 9, at the posttreatment assessment. The registry reports group means and standard deviations for aggregated exposure and medication comparisons. For exposure, VRE is compared with PE; for augmentation, DCS is compared with placebo. These are not four separate pairwise comparisons among all treatment combinations. The registry record does not report a separate result for each of the four cells, or an analysis testing whether the effect of DCS differed according to exposure approach.

There is one posted outcome measure and no posted secondary outcomes. The analysis details beyond the posted summary comparisons are not reported in the registry record. The record does not describe the planned model, covariate adjustment, analysis population, handling of missing endpoint values, or the original sample-size justification.

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: Total sample size is n=192. Randomization allocation ratio was 1:1:1:1 to VRE vs. PE and DCS vs. placebo. Study had 2 co-primary aims to examine: 1. the effects of DCS (n=95) vs placebo (n=97) augmentation of exposure therapy 2. the relative efficacy of VRE (n=97) and PE (n=95) on PTSD symptoms.

Exposure Therapy

Drug

Analysis population for the primary outcome, as defined in the registry: Total sample size is n=192. Randomization allocation ratio was 1:1:1:1 to VRE vs. PE and DCS vs. placebo. Study had 2 co-primary aims to examine: 1. the effects of DCS (n=95) vs placebo (n=97) augmentation of exposure therapy 2. the relative efficacy of VRE (n=97) and PE (n=95) on PTSD symptoms.

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: Total sample size is n=192. Randomization allocation ratio was 1:1:1:1 to VRE vs. PE and DCS vs. placebo. Study had 2 co-primary aims to examine: 1. the effects of DCS (n=95) vs placebo (n=97) augmentation of exposure therapy 2. the relative efficacy of VRE (n=97) and PE (n=95) on PTSD symptoms.

6. Primary Endpoint

The registry lists the primary outcome as CAPS-IV at the End of Treatment, measured over after weekly treatment session 9 (at posttreatment assessment).

The primary endpoint is CAPS-IV total severity at the end of treatment, assessed after weekly treatment session 9. CAPS-IV is described in the registry as a structured clinical interview designed to assess the 17 DSM-IV PTSD symptoms. It provides categorical ratings of diagnostic status as well as a quantitative measure of symptom severity. The total severity score is based on the frequency and intensity of current symptoms, with symptoms assessed separately for past-month and past-week time frames.

The score ranges from 0-136, and higher scores mean a worse outcome. For the posted result, the endpoint is summarized as a mean and standard deviation in each reported group. A group mean conveys the average score, while the standard deviation describes the spread of scores among participants in that group. Looking at a quantitative severity measure allows the outcome to be compared across randomized groups without reducing the result to a simple diagnostic category. That is a general reason continuous symptom scores are used in clinical research; the registry does not state a specific rationale for selecting CAPS-IV as the endpoint.

The posted endpoint is a posttreatment score. A baseline score, change-from-baseline result, or longer-term follow-up result is not reported in the registry record. Consequently, these summaries describe the scores recorded at the specified posttreatment assessment, not the amount of change for each participant or the persistence of any change after treatment.

7. Results Posted in the Registry

• Virtual Reality Exposure (VRE): mean 52.42 versus 48.59 on control; difference 3.83 (95% CI -3.39 to 11.05), standardized effect d = 0.15; z-test p = 0.298.
• Prolonged Imaginal Exposure (PE): mean 50.61 versus 48.59 on control; difference 2.02 (95% CI -5.04 to 9.08), standardized effect d = 0.08; z-test p = 0.575.
• Drug: D-Cycloserine (DCS): mean 54.20 versus 48.59 on control; difference 5.61 (95% CI -1.67 to 12.89), standardized effect d = 0.22; z-test p = 0.131.

8. Statistical Interpretation

9. Statistical Methods Explained

The endpoint is continuous: CAPS-IV is reported as a numerical severity score, summarized by the mean and standard deviation. A mean is the arithmetic average of the scores in a group. A standard deviation describes how dispersed individual scores are around that group average. The means and standard deviations are descriptive summaries; on their own, they do not show how much scores changed from before treatment because a baseline result is not posted.

What does a mean difference represent?

A mean difference compares the average endpoint score in one group with the average in a reference group. The posted estimates are oriented as the named treatment group minus placebo for the medication comparison, and VRE minus PE for the exposure comparison. A positive difference therefore means the first-named group had the higher mean score in the reported comparison. Since the registry states that higher CAPS-IV scores mean a worse outcome, the direction of the numerical difference can be described in that context. The estimate is still an average contrast, not a statement about every participant.

The group summaries are organized around two marginal questions. One compares the VRE and PE exposure approaches, pooling across the medication assignments; the other compares DCS and placebo, pooling across exposure approaches. That distinction matters because the study had four combinations. A marginal comparison does not show all four combination-specific means, and it does not answer whether the association between one treatment factor and the endpoint changes across levels of the other factor. Such a question is commonly addressed with an interaction analysis, but an interaction result is not reported in the registry record.

How should the confidence intervals be read?

A confidence interval gives a range of values compatible with the estimate under the statistical procedure used. In general, a narrower interval indicates greater precision, while a wider interval indicates more uncertainty about the size and direction of the contrast. Each posted interval spans zero, the value corresponding to no mean difference. This means the results are compatible with no difference as well as with differences on either side of zero within the reported interval. It does not prove that the groups are identical. The registry does not report whether the confidence intervals are adjusted for multiple comparisons or other features of a fitted model.

The comparison results include unadjusted p-values. In general, a p-value describes how unusual data at least as discrepant as those observed would be under a specified null hypothesis and statistical model. It is not the probability that the null hypothesis is true, nor does it measure the clinical importance of a difference. The posted p-values are labeled unadjusted; the analysis model and any adjustments beyond that are not reported in the registry record. As multiple aims and treatment factors are present, the absence of a reported multiplicity approach limits interpretation across the set of questions.

Why do group sizes and spread matter?

Precision depends partly on the amount of information in the groups and the variability in the outcome. Larger standard deviations indicate greater variation among scores, which generally makes a group mean difference harder to estimate precisely for a fixed sample size. The observed standard deviations are substantial relative to the point estimates, and the confidence intervals communicate the resulting uncertainty in the comparisons. This is a general statistical relationship, not a claim about an unreported design calculation.

The trial enrollment is 192, while the posted outcome group sizes vary by comparison: the exposure summaries use 97 for VRE and 95 for PE, and the medication summaries use 95 for DCS and 97 for placebo. The registry does not explain the difference between enrollment and the numbers contributing to each outcome summary. It also does not report whether the analyses used an intention-to-treat population, a per-protocol population, or another analysis population. Missing endpoint data, exclusions, and reasons for any difference in analysis counts are not reported in the registry record.

What is the role of the reported planning illustrations?

The participant counts associated with the observed comparison estimates are planning illustrations based on the posted effect and a stated set of assumptions, not a description of the study's actual design calculation. The illustrations use a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation. They consider the stated fraction of the observed effect. The 80% and 90% power columns represent alternative planning targets: the chance, under the assumed effect and statistical model, of detecting a difference at the specified significance level. For a smaller effect, more participants are generally needed to achieve the same target power, which is why the illustrations increase as the assumed fraction of the observed effect decreases. These calculations do not establish the power of the completed trial, and the registry's own original sample-size assumptions are not reported.

In general, over-dispersion means that observed variability is greater than a model expects. A model that fails to account for such extra variability can produce standard errors that are too small and intervals that are too narrow. The posted outcome is continuous and summarized with means and standard deviations, but the registry record does not identify an over-dispersion problem or describe a model for it. This general concept should not be mistaken for a finding about this study.

Overall, the available summaries allow a reader to describe the endpoint means, marginal mean differences, confidence intervals, and unadjusted p-values. They do not supply enough information to reproduce a full statistical analysis. In particular, the registry does not report the model specification, covariates, missing-data approach, analysis population, multiplicity procedure, or detailed handling of the four-condition structure. Those omissions make it important to keep interpretation tied to the descriptive and comparison results actually posted.

10. Confidence Intervals and Precision

The confidence intervals communicate the precision of the estimated mean differences. For VRE compared with PE, the estimate is 3.83 and the interval extends from -3.388 to 11.05. For DCS compared with placebo, the estimate is 5.61 and the interval extends from -1.674 to 12.89. For PE compared with placebo, the estimate is 2.02 and the interval extends from -5.044 to 9.084. Each interval includes zero, so none of these estimates rules out no average difference under the reported interval calculation.

All three intervals also permit values on both sides of zero. The posted data therefore leave uncertainty not only about the magnitude of any contrast, but also about its direction. For example, a positive point estimate alone should not be interpreted as establishing that the first-named group had worse average severity; the corresponding interval includes negative values as well. Conversely, inclusion of zero does not demonstrate that the groups are equivalent. Equivalence would require an appropriate design and a prespecified margin, neither of which is reported in the registry record.

The width of an interval reflects the precision available for the estimate, given the data and the method used. These intervals are broad relative to their point estimates, so a range of materially different mean contrasts remains compatible with the posted summaries. The registry does not report the confidence-interval method, adjusted intervals, or an equivalence or noninferiority framework. The intervals should therefore be read as uncertainty measures for the posted comparisons, not as definitive bounds on every possible treatment effect or as evidence about outcomes beyond the posttreatment CAPS-IV assessment.

11. Planning a Similar Trial: Sample Size

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

The sample-size illustrations give participants per arm for the two-arm comparisons; they are not total event counts. The comparisons are based on the observed VRE-versus-PE, PE-versus-placebo, and DCS-versus-placebo effects, with planning rows for 100%, 75%, and 50% of the observed effect. Under the illustration, the 80% power column gives the number of participants per arm associated with an 80% chance of detecting the assumed difference, and the 90% power column gives the number per arm for a 90% chance, under the stated model and significance level.

The assumptions are a two-sided significance level of 0.05, equal allocation between the two arms, a normal approximation, and an effect equal to the stated fraction of the observed effect. The illustration is based on the observed effect and is not the trial's original design calculation. The registry record does not report the original sample-size justification or the trial's own planning assumptions. These figures are not observed power.

The pattern is important: as the effect assumed for planning becomes smaller, more participants are required to maintain the same power target. For example, for the DCS-versus-placebo comparison, planning for 100% of the observed effect gives 331 participants per arm for 80% power and 442 participants per arm for 90% power; planning for 50% gives 1,321 and 1,768 participants per arm, respectively. These numbers are illustrations conditional on the assumptions, not predictions of a future trial's results. The curve should be read in the same way: it shows how the required per-arm count changes with the effect assumed, rather than describing the sample size actually enrolled or the probability that the completed study detected a true effect.

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 = 3.830; pooled standard deviation σ = 25.647.
2. n = 2 × (1.96 + 0.842)² × 25.647² / 3.830² = 703.9, rounded up to 704 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 <- 3.83                      # observed difference in means
sigma <- 25.6465065847183                     # 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(1352637)
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 = 704 (80% power), 943 (90% power)
 75% of observed effect: n per arm = 1,252 (80% power), 1,676 (90% power)
 50% of observed effect: n per arm = 2,816 (80% power), 3,770 (90% power)
power.t.test: 704.8 
Simulated power at n = 704 per arm: 0.796 (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 describes two co-primary aims: comparing DCS with placebo as an augmentation to exposure therapy, and comparing VRE with PE. It posts one primary outcome measure and no posted secondary outcomes. The three comparison summaries include two exposure-related contrasts and one medication-related contrast, but the statistical plan for evaluating these questions together is not reported in the registry record.

When more than one hypothesis is examined, the chance of at least one false-positive result can increase if each is tested without adjustment. This is the general reason clinical studies may prespecify a multiplicity strategy, such as assigning significance levels or adjusting inference across a family of comparisons. The posted p-values are unadjusted. Whether a formal multiplicity procedure was planned or used is not reported, so the p-values should not be interpreted as if they account for all co-primary aims or possible contrasts.

The record does not report the analysis population, whether all randomized participants were included in the primary analysis, or the number with a usable posttreatment CAPS-IV assessment. It also does not describe missing-data patterns, reasons for missingness, imputation, sensitivity analyses, or how participants who discontinued treatment were handled. As a result, the posted endpoint summaries cannot establish how missing observations may have affected the comparisons.

The study had four randomized combinations, but the posted results are marginal summaries and do not give combination-specific outcome means or an interaction test. A factorial design can raise distinct questions about each treatment factor and whether the factors modify one another. The registry does not report how those questions were incorporated into a prespecified analysis plan. The original statistical plan, model specification, covariate adjustment, and approach to multiple comparisons are not reported in the registry record. These gaps limit assessment of whether the posted summaries fully reflect the trial's planned confirmatory analyses.

14. 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 1% reporting threshold.

Most frequent serious adverse event terms

Serious events with the highest arm-level rate: hospitalization (1.1%), hopitalization (1.1%).

The adverse-event summaries are reported for the VRE, PE, DCS, and placebo groups. The serious-event counts were 0 among 97 in VRE, 3 among 95 in PE, 0 among 95 in DCS, and 1 among 97 in placebo. The registry reports 0 other adverse events in each of those groups and 0 deaths in each. These are counts within the listed groups, not estimates of comparative risk adjusted for other factors.

The registry lists the serious-event terms Hospitalization and Hopitalization, each with 1.1%. The record does not clarify whether the spelling variants refer to distinct entries, or provide additional event-level detail in the information reported for this outcome. The serious-event summaries are grouped by exposure approach or medication assignment rather than by the four treatment combinations. They therefore do not identify which specific combination was associated with each serious event.

These figures describe the events recorded in the registry and should be interpreted in light of the group sizes and the limited detail posted. The registry does not report event timing, severity grading beyond the serious classification, relatedness assessment, or a detailed safety analysis plan. The counts alone do not establish a comparative safety effect.

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

The posted results describe average CAPS-IV severity after treatment for randomized groups, along with mean-difference estimates and confidence intervals. Across the reported comparisons, the point estimates are positive, while all three confidence intervals include zero and extend to values in the opposite direction. The estimates are therefore uncertain in both magnitude and direction. This is a description of the posted statistical results, not a conclusion that any treatment is effective or ineffective.

The results concern one posttreatment assessment after weekly treatment session 9. They do not describe change from baseline, longer-term outcomes, or individual symptom trajectories. The registry describes CAPS-IV as a measure of symptom frequency and intensity and states that higher scores mean worse outcomes, but it does not post a baseline comparison or follow-up results. No posted secondary outcomes are available in the record.

The study's four conditions combine two exposure approaches with DCS or placebo. The reported summaries address the marginal exposure comparison and the marginal medication comparison. They do not show four combination-specific means, direct estimates for every pair of combinations, or whether the medication comparison varied by exposure type. Readers should not infer a result for one particular treatment combination from a marginal mean difference.

The confidence intervals are compatible with a range of possible differences, and no equivalence margin is reported. Thus, the absence of a confidence interval excluding zero is not evidence that treatments are identical. At the same time, a point estimate away from zero does not by itself establish a reliable difference. Interpretation is also limited by the lack of posted details about missing outcomes, analysis populations, modeling, and multiplicity handling.

In practical statistical terms, these data provide an estimate and an indication of its uncertainty, but the registry information does not resolve every question posed by the factorial design. The results should be understood as the posted end-of-treatment summaries for this study population and endpoint. They do not support extrapolation to populations or time points not described in the registry record.

16. Why This Trial Matters Statistically

This trial is statistically notable because it combines two treatment factors in a randomized four-condition design: exposure approach and DCS versus placebo. Such a structure can support separate questions about each factor, while also raising the question of whether the effect of one factor depends on the other. The posted summaries make the distinction between marginal comparisons and combination-specific effects especially important. VRE versus PE and DCS versus placebo are not interchangeable with comparisons among all four treatment combinations.

The outcome is a continuous symptom-severity score, and the registry reports group means, standard deviations, mean differences, confidence intervals, and unadjusted p-values. This makes it possible to consider both the estimated direction and the uncertainty around each contrast, rather than relying on a significance label alone. The intervals include zero and cover values on either side, illustrating why precision matters when interpreting a point estimate.

The sample-size illustrations also demonstrate a central planning principle: smaller target effects call for larger group sizes when the desired power and other assumptions are held fixed. These illustrations are based on the observed estimates, not the trial's original design, and should not be confused with observed power. The original sample-size justification is not reported in the registry record.

Finally, the difference between what the design could address and what the posted results report is informative. The record describes co-primary aims and four randomized combinations, but posts marginal comparisons and no interaction result. It also omits details on missing data, analysis populations, and multiplicity. This makes the study a useful example of why readers need to distinguish the randomized design, the statistical questions it raises, and the level of detail available in the reported results.

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 result is limited to CAPS-IV severity at the posttreatment assessment after weekly treatment session 9. A baseline score, change-from-baseline analysis, or later follow-up outcome is not reported in the registry record, so the posted information cannot describe symptom trajectories or the durability of outcomes.

The registry presents marginal comparisons for exposure type and medication assignment, but not separate means for all four treatment combinations. No interaction result is posted. This restricts interpretation of whether the DCS-versus-placebo contrast differed between VRE and PE, and prevents direct assessment of every combination-to-combination contrast from the summaries given.

Only one outcome measure is posted, and no secondary outcomes are posted. The analysis population, missing endpoint counts, missing-data methods, model specification, covariates, and sensitivity analyses are not reported. The record also does not provide the trial's original sample-size justification or a detailed multiplicity plan. The reported p-values are unadjusted, and the record does not explain whether other adjustments were planned or used.

The registry describes the population broadly as active duty military personnel, veterans, and civilians who served in Iraq and Afghanistan, but additional eligibility and participant-characteristic details are not reported in the information available for these results. The masking description is triple, but the roles that were masked are not specified. Safety reporting is limited to group counts, with no detailed event timing, severity grading, or relatedness information. These omissions constrain assessment of the results beyond the posted summary statistics.

18. Related Tutorials

19. Related Statistical Calculators

20. Sources

• ClinicalTrials.gov: Enhancing Exposure Therapy for Post Traumatic Stress Disorder (PTSD): Virtual Reality and Imaginal Exposure With a Cognitive Enhancer — NCT01352637.
• Registry record last updated 2019-07-02; status last verified 2019-06.
• Results first posted 2019-07-02.
• 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

NCT01352637 is a completed Phase 3 randomized, triple-masked, parallel study sponsored by Weill Medical College of Cornell University. The registry lists 192 participants and four treatment conditions crossing VRE or PE with DCS or placebo. The study focused on combat-related PTSD among active duty military personnel, veterans, and civilians who served in Iraq and Afghanistan.

The primary endpoint was CAPS-IV total severity after weekly treatment session 9. The registry posts one outcome measure and summary comparisons for exposure approach and medication assignment. All three posted confidence intervals include zero; the registry does not report secondary outcomes, an interaction result, a detailed analysis plan, or the trial's original sample-size assumptions. Serious adverse-event counts are posted by exposure or medication group, with no deaths reported in those groups.