Efficacy of Virtual Reality Exposure Therapy Scripts for Social Phobia: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Efficacy of Virtual Reality Exposure Therapy Scripts for Social Phobia (NCT04850989), a randomized na study. The primary outcome is Change From Baseline in the Social Phobia Diagnostic Questionnaire (SPDQ) at Post-treatment.

2. Clinical Question

3. Trial Design

This was a randomized, parallel-group trial with no masking listed in the registry. The study was marked completed, and its phase was recorded as not applicable. The registry lists an enrollment of 44 and identifies Penn State University as the sponsor. The recorded study dates were 2018-12-18 for the start and 2020-12-31 for primary completion.

Participants were assigned to virtual reality exposure therapy or a waiting-list control condition. The virtual reality intervention used a Pico Goblin VR headset. Participants could choose one of two themes, and the registry describes the exposure scenes as increasing in anxiety-provoking intensity. The waiting-list description says treatment began 2-4 weeks after randomization. Further details about allocation procedures, concealment, treatment duration, session frequency, adherence, and co-interventions are not reported in the registry record.

The primary outcome was a mean change score on the Social Phobia Diagnostic Questionnaire at post-treatment. The outcome timeframe is described as baseline and post-treatment, on average 2 weeks. The listed analysis population is intention-to-treat, and the registry reports a Mixed Models Analysis for the comparison. How missing observations were handled, which covariates were included, and how the intention-to-treat population was implemented are not reported in the registry record.

The brief summary states hypotheses concerning symptom change, maintenance of gains at 3-month and 6-month follow-up, and treatment acceptability. The posted outcome results include the primary post-treatment measure and three secondary post-treatment measures. Results for maintenance and acceptability are not reported in the registry record. The record also does not report the original design assumptions, sample-size justification, interim analyses, or a prespecified statistical analysis plan.

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: Intention-to-treat analysis

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 the Social Phobia Diagnostic Questionnaire (SPDQ) at Post-treatment, measured over Baseline and post-treatment, on average 2 weeks.

The primary endpoint is change from baseline in the Social Phobia Diagnostic Questionnaire at post-treatment. The registry describes this as a 25-item self-report measure of fear and avoidance in social situations. Each item is described as rating fear severity and avoidance frequency on a scale from 0 to 4. The record states that higher scores reflect greater social anxiety symptoms and gives a total-score range from 0 to 27. It describes change scores as baseline scores minus post-treatment scores, with higher, less negative changes indicating better outcomes because they represent symptom reduction.

This measure is relevant to the study question because the intervention is intended to expose participants to social situations in a controlled setting, and the questionnaire assesses fear and avoidance across such situations. A change score compares each participant’s post-treatment assessment with their baseline assessment, while the group comparison asks whether the average change differs between assigned conditions. The listed timeframe is baseline and post-treatment, on average 2 weeks.

The registry’s stated scoring details should be read cautiously: its description of item response options and its stated total-score range are not fully self-explanatory. The record does not provide additional scoring guidance or explain whether any items are transformed or omitted. The posted group means are negative, and the virtual reality group’s mean is more negative than the waiting-list mean. That direction must be considered alongside the registry’s statement that higher, less negative change scores indicate improvement; the numerical direction alone should not be relabeled as improvement.

7. Results Posted in the Registry

• Virtual Reality Exposure Therapy: mean -9.60 versus -4.46 on control; difference -5.14 (95% CI -5.37 to -4.91), standardized effect d = -16.15; z-test p = <0.001.

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.

Change From Baseline in the Measure of Anxiety in Selection Interviews (MASI) at Post-treatment

• 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:
  • Virtual Reality Exposure Therapy: mean -36.52 versus -22.57 on control; difference -13.95 (95% CI -16.85 to -11.05), standardized effect d = -3.45; z-test p = <0.001.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the mean is statistically lower for Virtual Reality Exposure Therapy than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

Change From Baseline in the Penn State Worry Questionnaire (PSWQ) at Post-treatment

• 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:
  • Virtual Reality Exposure Therapy: mean -21.18 versus -13.10 on control; difference -8.08 (95% CI -8.78 to -7.38), standardized effect d = -5.78; z-test p = <0.001.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the mean is statistically lower for Virtual Reality Exposure Therapy than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

Change From Baseline in the Patient Health Questionnaire (PHQ-9) at Post-treatment

• 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:
  • Virtual Reality Exposure Therapy: mean -6.74 versus -3.26 on control; difference -3.48 (95% CI -5.00 to -1.96), standardized effect d = -1.35; z-test p = <0.001.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the mean is statistically lower for Virtual Reality Exposure Therapy than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

The posted secondary measures cover anxiety related to selection interviews, worry, and depressive symptoms. Across these measures, the recalculated mean differences are negative, and the recalculated unadjusted p-values are reported as <0.001. The registry’s Mixed Models Analysis results are not fully uniform in their reported significance: one is reported as p > 0.05, while the others are reported as p < 0.05. These are different forms of analysis and should not be collapsed into a single conclusion.

In broad terms, the secondary results point in the same numerical direction as the primary group contrast: the virtual reality group has more negative posted mean changes than the waiting-list group. Whether that direction represents improvement depends on each measure’s scoring and change convention. The registry does not provide full descriptions of the secondary scales or a common interpretive rule that would support treating all their signs as equivalent. These findings can be viewed as related symptom measures, but they do not independently resolve the primary endpoint’s direction-of-scoring concern.

Secondary outcomes generally carry less weight than a clearly specified primary outcome because they may be numerous and can create opportunities for chance findings. The registry lists five outcome measures and five analyses, but a multiplicity adjustment or other approach to controlling the overall false-positive risk is not reported in the registry record. The posted secondary p-values are therefore best treated as exploratory or supportive rather than as a set of independent confirmations. Their interpretation should also account for the reported differences between recalculated unadjusted comparisons and the registry’s mixed-model results.

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

• Change From Baseline in the Measure of Anxiety in Selection Interviews (MASI) at Post-treatment
• Change From Baseline in the Penn State Worry Questionnaire (PSWQ) at Post-treatment
• Change From Baseline in the Patient Health Questionnaire (PHQ-9) at Post-treatment

9. Statistical Interpretation

10. Statistical Methods Explained

The primary outcome is continuous: it is a score measured on a scale, and the registry summarizes each group with a mean and standard deviation. A mean describes the group’s average score or average change. A standard deviation describes how dispersed individual observations are around that average. It is not the uncertainty of the mean itself. Those summaries help readers see both the central result and the variability that informs a comparison.

What does the mean difference describe?

The mean difference compares the average change in one group with the average change in the other. In the recalculation, the virtual reality group is compared with the waiting-list group. A negative value means the first group’s posted change mean is numerically lower than the comparator’s mean. Whether that numerical direction represents a better outcome depends on the outcome’s scoring convention. Here, the registry says that higher, less negative change scores represent greater improvement, so interpretation should preserve that stated direction rather than assuming that a negative contrast is favorable.

The registry’s reported primary method is a Mixed Models Analysis, with the result described as a net mean difference. A mixed model can account for the structure of repeated observations or other grouped data, depending on how it is specified. That is a general description, not a claim about the exact model used in this trial. The registry does not report the model formula, covariance structure, included covariates, or handling of incomplete observations. The recalculated comparison is a simple unadjusted calculation from the posted group summaries and is not the same as the reported model.

How should a p-value be read?

A p-value describes how compatible the observed result is with a specified null hypothesis under the statistical model and its assumptions. It is not the probability that the null hypothesis is true, and it does not quantify the size or practical importance of a difference. The recalculated primary p-value is reported as <0.001, while the registry’s mixed-model result is reported as p < 0.05. These statements use different analyses and should not be treated as interchangeable. The registry does not report an exact p-value for its model-based primary result.

Statistical significance alone also does not resolve whether the difference is meaningful to participants. That requires attention to the measure’s direction, score interpretation, study conduct, and the size of the effect in a context that the record does not provide. The registry does not identify a threshold for a clinically important change on this measure.

What does a confidence interval add?

A confidence interval gives a range of values compatible with the data under the calculation’s assumptions. In repeated use of a method, intervals constructed in the same way have a stated coverage property. A narrower interval indicates less sampling uncertainty under that method, but it does not account automatically for bias, data errors, poor measurement, or an incorrect model. The recalculated primary interval is based on the posted summary statistics. It is not a confidence interval from the registry’s mixed-model analysis, for which interval limits are not reported.

The interval’s apparent narrowness should be considered alongside the unusually small posted standard deviation in the virtual reality group. A standard error reflects uncertainty in an estimated mean or contrast and is influenced by variability and group size. If a posted spread is exceptionally small, the resulting standard error and interval may also be small. The record does not explain the source of this spread, so readers cannot determine from the registry alone whether it reflects the observed data, rounding, or a reporting issue.

Why can variability affect a comparison?

General statistical explanation: more variable observations typically make a group mean less precise, while less variable observations tend to make it more precise, if other features are held constant. Unequal variability between groups can also affect how a comparison should be modeled. For count data, over-dispersion means variability exceeds what a simple count model expects and can make standard errors too small if ignored. That particular issue is not the main model description for this continuous endpoint, but the general lesson applies: uncertainty estimates depend on whether the variance assumptions fit the data. The registry does not report diagnostic checks or alternative variance analyses.

The recalculated standardized effect is also derived from posted means and a pooled standard deviation. Standardized effects can make measures on different scales easier to compare, but they are sensitive to the standard deviation used. When the recorded spreads are very small, the standardized contrast can become extreme. Such a value should not be interpreted as proof that the effect is exceptionally reliable or generalizable.

What is the role of the reported analysis?

The registry records a superiority analysis using a mixed model and provides a net mean difference and p-value. The simple recalculation independently summarizes the posted group means and spreads without adjustment. Because the record’s primary discrepancy flag is false, the posted information does not identify a discrepancy requiring the sponsor’s analysis to take precedence. Still, the recalculation remains an analytical aid, not a replacement for the trial’s stated model. The details needed to reproduce the model-based result are not reported in the registry record.

Finally, the trial lists intention-to-treat as the analysis population. In general, intention-to-treat aims to analyze participants according to their randomized assignment, but the label alone does not reveal how missing outcome data, nonadherence, or exclusions were handled. Those implementation details are not reported in the registry record. Readers should therefore distinguish the stated analysis population from a fully documented account of the analysis.

11. Confidence Intervals and Precision

The recalculated primary mean difference is -5.14, with a confidence interval from -5.371 to -4.909. This interval is for a simple unadjusted comparison calculated from the posted summaries. The registry’s Mixed Models Analysis also reports a mean difference of -5.14, but its confidence interval is not reported in the registry record. The numerical match in the estimates does not make the methods or uncertainty estimates identical.

A confidence interval describes sampling uncertainty conditional on the calculation and its assumptions. It does not include every source of uncertainty. For example, it does not correct for a possible scoring or data-entry problem, selective reporting, limitations in the study design, or an inappropriate variance assumption. The registry does not report information that would allow those possibilities to be assessed fully.

The interval is narrow, but the posted variability warrants caution in interpreting that apparent precision. The primary group standard deviations are 0.01 and 0.5. The record does not explain the unusually small spread in the virtual reality group. Because the recalculation uses those values, any uncertainty in the accuracy or meaning of the reported spread would affect the recalculated precision. The registry does not provide participant-level data or an explanation that could resolve this concern.

The interval also does not settle the direction-of-benefit issue. Its values are negative, consistent with the virtual reality group’s numerically lower change mean. The endpoint description says higher, less negative change scores indicate better outcomes. Precision around a numerical contrast cannot repair ambiguity about how that contrast should be interpreted. The registry’s model-based result has no interval reported, and the record does not provide enough model detail to assess its precision independently.

12. Planning a Similar Trial: Sample Size

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

The planning figures are participants per arm for 80% power and participants per arm for 90% power. They are not total event counts, because these outcomes are continuous rather than time-to-event outcomes. The stated illustration assumes a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. It uses the observed effect or a stated fraction of that effect as the effect to detect. The trial’s own original design assumptions are not reported in the registry record.

The 80% power column describes the number of participants per arm that the illustration estimates would be needed to detect the selected effect with 80% power, under its assumptions. The 90% power column describes the corresponding estimate for 90% power. Power is a property of a design under assumed conditions: it is the probability of detecting an effect of the specified size if that effect is present and the assumptions hold. It is not a guarantee that a study will find a statistically significant result.

Rows that plan for a smaller fraction of the observed effect generally require more participants, because smaller differences are harder to distinguish from random variation. The posted illustrations reflect unusually large observed standardized contrasts for some measures, alongside very small reported spreads. This can produce strikingly small planning figures, including estimates as low as 1 participant per arm. Those figures should be treated as mathematical illustrations tied to the observed summaries and assumptions, not as design recommendations.

This is not observed power. It is a planning calculation based on the observed effect and stated assumptions, not a measure of how much power the completed trial actually had. The registry’s rationale for selecting its original enrollment and any original power calculation are not reported in the registry record.

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 = -5.140; pooled standard deviation σ = 0.318.
2. n = 2 × (1.96 + 0.842)² × 0.318² / -5.140² = 0.1, rounded up to 1 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 <- -5.14                      # observed difference in means
sigma <- 0.318198051533946                     # 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(4850989)
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 = 1 (80% power), 1 (90% power)
 75% of observed effect: n per arm = 1 (80% power), 1 (90% power)
 50% of observed effect: n per arm = 1 (80% power), 1 (90% power)
power.t.test: 1.5 
Simulated power at n = 1 per arm: NA (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. Analyses Reported by the Sponsor

The registry’s reported primary analysis is a Mixed Models Analysis. It gives a net mean difference of -5.14 and reports p < 0.05, with the analysis classified as superiority. It does not provide confidence limits for that model-based result. The simple recalculation from the posted means and standard deviations also gives a mean difference of -5.14, with a recalculated confidence interval from -5.371 to -4.909 and an unadjusted p < 0.001.

The recalculation is not a reconstruction of the mixed model. It does not account for any adjustment or model structure that may have been used, and the record does not report enough detail to reproduce the sponsor’s analysis. Its interval and p-value therefore describe a separate, unadjusted calculation. The registry’s discrepancy indicator is false, and there is no stated discrepancy between the primary estimate and the recalculated estimate. The difference in reported p-value thresholds is not, by itself, evidence that the estimates conflict, because the analyses and reporting precision differ.

The registry’s superiority label and the direction described for the score should be read together. The endpoint description says higher, less negative change scores indicate improvement, whereas the posted virtual reality mean is more negative. The record does not explain that direction issue. The reported analysis supplies the registry’s own inferential result, but its label does not remove the need to interpret the sign according to the stated score definition.

15. Multiplicity, Missing Data and Analysis Populations

The registry identifies a primary outcome and reports three secondary post-treatment outcomes. It records five outcome measures and five analyses, but does not provide a full accounting of how every outcome or analysis relates to those totals. A prespecified hierarchy among outcomes, adjustment for multiple comparisons, or another multiplicity strategy is not reported in the registry record.

Multiplicity matters because examining several outcomes can increase the chance that at least one result appears statistically notable by chance. The risk depends on the number and relationship of the tests and on the analysis plan. Without a reported adjustment or hierarchy, the secondary p-values should not be interpreted as though each were the sole planned test. This is especially relevant when secondary results differ in their reported significance between simple recalculation and the registry’s mixed-model analysis.

The primary outcome lists intention-to-treat as the analysis population. The registry does not explain how that population was defined operationally, whether all randomized participants contributed outcome data, or how departures from assignment were treated. It also does not report the amount or pattern of missing primary or secondary outcome data, reasons for missingness, or any sensitivity analyses. The mixed-model method is named, but the record does not state the model’s missing-data assumptions or how those assumptions were evaluated.

Interim analyses, stopping rules, and the original statistical analysis plan are not reported in the registry record. Their absence from the record does not establish that they did or did not occur; it means the available information cannot be used to assess them. Similarly, the registry does not report whether outcome scoring or analysis was blinded, and the trial’s masking is listed as none. These limits leave uncertainty about the analysis process beyond the summary results that are posted.

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

The registry records a randomized comparison of virtual reality exposure therapy with a waiting-list condition and reports a post-treatment difference on the Social Phobia Diagnostic Questionnaire. The reported mixed-model analysis is labeled as superiority and has p < 0.05. A simple unadjusted recalculation from the posted summaries gives a mean difference of -5.14, with an interval from -5.371 to -4.909 and p < 0.001. These results describe the trial’s recorded post-treatment comparison; they do not by themselves establish a treatment effect that will apply in other settings.

The direction of the numerical contrast is important. The registry says that higher, less negative change scores indicate improvement, yet the virtual reality group’s mean change is more negative than the waiting-list mean. The record does not explain how this relates to its superiority classification. As a result, the numerical difference should not be translated into a simple claim that one condition produced better symptom improvement without resolving the scoring interpretation.

The findings do not establish durability. The brief summary proposes maintenance assessments at 3-month and 6-month follow-up, but results for those time points are not reported in the registry record. The summary also proposes evaluating treatment acceptability, but corresponding results are not reported in the registry record. The posted primary result is specific to the stated post-treatment timeframe.

The registry lists zero serious adverse events, other adverse events, and deaths in both groups. This is a description of the recorded counts, not proof that the intervention is free from risk. Details about adverse-event ascertainment and reporting are not reported in the registry record.

Finally, statistical evidence does not answer every practical or clinical question. The record does not report a threshold for a clinically important change, detailed eligibility criteria, or results for a wider range of populations. Interpretation should therefore remain bounded to the posted outcomes, their scoring definitions, and the limitations of the available analysis information.

17. Why This Trial Matters Statistically

This trial illustrates why reading a continuous endpoint requires more than looking at whether a p-value passes a threshold. A mean difference has a direction, and that direction only has meaning when linked to the outcome’s scoring convention. In this record, the stated definition of a better change score and the numerical ordering of the group means create an interpretive tension. Noticing that issue is essential before translating a contrast into a claim about symptom improvement.

It also shows why the registry’s reported analysis and a summary-statistic recalculation should be kept distinct. The mixed-model result and the unadjusted recalculation have the same reported primary estimate, but the registry does not provide a model-based confidence interval or enough model detail for reproduction. The recalculated interval is informative about one simple analysis, yet depends on the posted means and spreads. The very small reported standard deviation in one group makes scrutiny of the precision especially important.

The secondary outcomes offer additional symptom measures, but their interpretation depends on multiplicity, scoring, and the consistency of the analysis results. The record does not report a multiplicity strategy, missing-data details, or an original analysis plan. Those omissions limit how confidently the findings can be treated as confirmatory across outcomes.

Finally, the sample-size illustrations show how planning based on observed effects can generate very small estimates when the observed standardized contrast is extreme. Such calculations should not be mistaken for evidence that the completed trial was adequately powered or that the estimated effect is stable. The statistical lesson is to evaluate the data, measurement, uncertainty, and analysis assumptions together.

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.

The registry record leaves important details of the intervention and study conduct unspecified. It says participants could choose one of two virtual reality themes and that scenes became more anxiety-provoking, but the script description is incomplete. Session frequency, treatment duration, adherence, therapist involvement, and fidelity assessment are not reported in the registry record. The waiting-list group’s treatment start is described as 2-4 weeks after randomization, but other activities or care during that period are not described.

Participant characteristics and eligibility criteria beyond the social anxiety disorder condition are not reported in the registry record. The record therefore does not establish how broadly the results might apply. Allocation concealment is not reported, and masking is listed as none. The registry does not say whether outcome assessment, scoring, or analysis was blinded.

The primary outcome description gives a stated total-score range of 0 to 27 while also describing 25 items rated from 0 to 4. The record offers no further scoring explanation to clarify those details. In addition, the change-score definition says higher, less negative values indicate better outcomes, while the posted group means and the superiority label raise a direction-of-interpretation question. The registry does not resolve that issue.

The primary virtual reality group standard deviation is 0.01, and the reason for this small spread is not reported. The recalculated precision and planning illustrations are sensitive to the posted variability. The record contains no participant-level data, model specification, missing-data summary, sensitivity analysis, or adjustment details that would allow independent resolution of these concerns. Results for follow-up maintenance and acceptability hypotheses are also not reported in the registry record.

19. Related Tutorials

20. Related Statistical Calculators

21. Sources

• ClinicalTrials.gov: Efficacy of Virtual Reality Exposure Therapy Scripts for Social Phobia — NCT04850989.
• Registry record last updated 2025-05-13; status last verified 2025-05.
• Results first posted 2025-05-13.
• 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

NCT04850989 is a completed, randomized, parallel-group study of virtual reality exposure therapy versus a waiting-list condition for social anxiety disorder. The registry lists Penn State University as sponsor, an enrollment of 44, and no masking. The primary outcome is change from baseline to post-treatment on the Social Phobia Diagnostic Questionnaire, with the timeframe described as on average 2 weeks.

The registry reports a Mixed Models Analysis with a net mean difference of -5.14 and p < 0.05, labeled as superiority. A simple unadjusted recalculation gives the same mean difference, a confidence interval from -5.371 to -4.909, and p < 0.001. The endpoint description says higher, less negative change scores indicate improvement, while the virtual reality group’s posted mean is more negative; the record does not explain this direction issue. Details on missing data, multiplicity, model specification, and follow-up results are not reported in the registry record.