Fluoxetine in Pediatric Body Dysmorphic Disorder: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Fluoxetine in Pediatric Body Dysmorphic Disorder (NCT00245635), a randomized phase 4 study. The primary outcome is Change in Total Score on the BDD-Y-BOCS Scale.

2. Clinical Question

The trial’s primary outcome is a change score rather than a single score measured only at the endpoint. A negative change indicates a lower total score at the endpoint than at baseline; the scale description says higher scores indicate worse impairment. The posted results give a mean change for each group and an unadjusted mean difference with a confidence interval. Those figures describe the observed score changes and the uncertainty of their comparison. They do not, by themselves, establish why any difference occurred or how well the results apply to people outside the trial population.

The registry describes a randomized comparison, but does not report the method used to generate the allocation sequence, the allocation concealment procedures, or the exact eligibility rules. It also does not report participant-level characteristics that would allow readers to assess how closely the enrolled population matched other groups of children and adolescents with BDD.

3. Trial Design

This is a phase 4, randomized, quadruple-masked, parallel-group trial sponsored by Montefiore Medical Center. The registry lists fluoxetine and placebo as the two interventions. The trial’s registry enrollment is 43. Outcome results are reported for 25 participants in the fluoxetine group and 18 in the placebo group; these are the group counts associated with the primary outcome results, not a report of how many participants completed every study procedure.

The fluoxetine arm description specifies a fixed or flexible dosing regimen based on weight and response to dosage, ranging from 10mg to 80mg tablets taken once a day. The placebo arm used tablets once a day with a dosing schedule equivalent to that of the drug. The record identifies the design as parallel group, meaning the comparison is between groups assigned to different study arms rather than a crossover sequence. It labels masking as quadruple, but does not specify which roles were masked or explain how masking was maintained.

The study began in 2004-11 and its primary completion date is 2012-04. Its status is completed. The primary outcome timeframe compares baseline with the study endpoint at week 12. The record does not report the reasons for the interval between study start and primary completion, the timing of recruitment, or whether changes were made to the study during that period.

The enrollment figure and the outcome group counts serve different descriptive purposes. The registry does not explain any difference between the enrollment figure and the counts attached to the reported outcome, nor does it provide a participant flow account explaining screening, allocation, follow-up, exclusions, or missing measurements. The original design assumptions, including a planned treatment effect or a sample-size justification, are not reported in the registry record. The planning illustrations based on the observed effect are separate from those original design assumptions.

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

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 in Total Score on the BDD-Y-BOCS Scale, measured over Baseline compared to the study endpoint (week 12) [two time points].

The primary endpoint is change in total score on the Body Dysmorphic Disorder-Yale-Brown Obsessive Compulsive Scale (BDD-Y-BOCS), from baseline (visit 0) to the study endpoint at week 12. The record describes the measure as a 12-item scale assessing obsessions and compulsions related to a participant’s BDD. Each item ranges from 0 to 4, and the full assessment has a possible total score of 48. A score of 0 indicates no impairment and a score of 4 indicates maximum impairment for an item; higher total scores are described as a worse outcome.

Change scores let readers compare how the measured symptom burden shifted over the specified period, rather than comparing only the groups’ endpoint scores. A negative mean change means the average score was lower at week 12 than at baseline. The posted group means are -12.27 for fluoxetine and -9.53 for placebo. These are average changes on the scale, not individual responses, and they do not describe the spread or range of every participant’s change.

The BDD-Y-BOCS is used because the trial’s stated clinical question concerns BDD-related obsessions and compulsions. The registry identifies this measure as the primary outcome, but does not report additional evidence about the scale’s measurement properties in this particular study population, how assessors were trained, or how the endpoint was collected. It also does not report a threshold for interpreting a change as important to participants. The scale description supplies its direction and range, but not a threshold for a meaningful individual or group change.

7. Results Posted in the Registry

• Fluoxetine: mean -12.27 versus -9.53 on control; difference -2.74 (95% CI -9.98 to 4.50), standardized effect d = -0.23; z-test p = 0.458.

8. Statistical Interpretation

9. Statistical Methods Explained

The primary outcome is continuous: it is a numerical score, and the registry reports a mean change and standard deviation for each group. A mean is the arithmetic average of the observations included in an analysis. It condenses the group’s changes into a single value, which is useful for a direct comparison but does not show the full distribution. Participants may have changes smaller or larger than the mean, and the group summary does not show how many had a change in either direction.

What does the mean difference represent?

The comparison is expressed as a mean difference, with the fluoxetine group compared with the placebo group. The reported group means are -12.27 and -9.53, and the unadjusted difference is -2.74 units. The negative sign follows from the direction of the comparison: the fluoxetine mean change is more negative than the placebo mean change. Because this is a change score, the comparison concerns average change from baseline to week 12, not a difference between the groups’ baseline scores or endpoint scores alone.

A mean difference remains in the original scale units, so readers can relate it to the BDD-Y-BOCS score range. It does not identify the cause of the difference. Randomization is the design feature intended to support a fair comparison of assigned groups, but interpretation also depends on how allocation and follow-up were conducted and how missing data were handled. Those procedural and analysis details are not reported in the registry record.

Why report a standard deviation?

A standard deviation describes how dispersed participant values are around their group mean. It is not the same as the uncertainty of the mean, and it is not an interval containing every participant’s result. The posted standard deviations are 11.78 in the fluoxetine arm and 12.07 in the placebo arm. These summaries help indicate the variability in the measured changes within each group. Individual scores and the shape of their distributions are not reported, so the standard deviations cannot show whether values were symmetric, clustered, or affected by unusually extreme observations.

Over-dispersion is a general concern when observed variability is greater than a statistical model expects: standard errors can then be too small and confidence intervals too narrow if the model does not account for the extra variation. That is a general statistical consideration, not a reported diagnosis about this trial. Here, the registry gives standard deviations and comparison uncertainty but does not report diagnostic checks for distributional assumptions or a model addressing over-dispersion.

How should the confidence interval and p-value be read?

A confidence interval expresses the uncertainty around an estimated quantity under the assumptions of the procedure used to calculate it. The reported interval for the mean difference is -9.98 to 4.5 units. Its endpoints represent values compatible with the data and procedure; they are not a range of individual participant responses. Since the interval includes zero, the data do not pin down the direction of the mean difference with precision.

The p-value of 0.458 is labelled unadjusted. In general, a p-value is calculated relative to a null hypothesis and a statistical model. It is not the probability that an observed result happened by chance, and it does not say how large or important an effect is. The record does not identify the exact test, its assumptions, or a prespecified significance threshold. The estimate, interval, and p-value should therefore be read as complementary summaries rather than as interchangeable measures.

What does unadjusted mean in this comparison?

An unadjusted comparison is not reported as controlling for additional covariates in the comparison itself. The record does not explain whether the original trial analysis used an adjusted model, whether adjustment was planned, or whether baseline measures were incorporated in another way. The posted comparison is a simple unadjusted recalculation from the group summary statistics. The registry record reports no analysis entries, so it does not provide a separate detailed model result or an analysis specification against which to compare this summary-based estimate.

Randomization does not remove the need to understand analysis choices. In a randomized trial, the assigned groups are the basis for the intended comparison, but the analysis population, treatment of missing measurements, and any model adjustment can affect the reported estimate. The record does not state whether the analysis followed intention-to-treat principles, included only participants with observed week 12 scores, or used another population definition. The outcome counts should not be assumed to settle those questions.

What can the posted summaries establish?

The summaries establish what the registry reports for the group means, their standard deviations, the unadjusted mean difference, its interval, and the unadjusted p-value. They do not provide participant-level data, a full statistical analysis plan, or enough information to reproduce every part of the trial’s original analysis. The method used to calculate the confidence interval and p-value is not reported in the registry record. Accordingly, the statistical interpretation should stay close to the quantities actually recorded and should avoid treating a planning calculation as evidence about the trial’s achieved power.

10. Confidence Intervals and Precision

The mean difference is -2.74 units, with a confidence interval from -9.98 to 4.5 units. The interval is wide relative to the point estimate and crosses zero. This means the posted summary does not locate the mean difference narrowly: values below zero and above zero remain within the reported interval. The interval is a measure of uncertainty around the estimated group difference under the method’s assumptions; it is not a prediction interval for individual participants and does not describe the full range of individual changes.

The group standard deviations, 11.78 for fluoxetine and 12.07 for placebo, are measures of variability in participants’ change scores. They are distinct from the uncertainty around the mean difference. The comparison details also list standard errors of 2.356 for fluoxetine and 2.845 for placebo. A standard error describes the estimated sampling variability of a statistic, rather than the spread of individual observations. These statistics help explain why an estimate can have a point value while still carrying substantial uncertainty.

The unadjusted p-value is 0.458. Read with the interval, it does not provide a precise separation of the group mean changes. Neither the interval nor the p-value answers whether a particular change would matter to participants: the registry record does not report a threshold for a clinically important difference. The confidence interval’s calculation method is not reported, so its interpretation depends on assumptions that cannot be checked from the registry record alone.

Precision is also shaped by the amount and quality of outcome information, the variability of scores, and the analysis choices. The record does not describe missing-data handling, outcome completion, or the analysis population. Those omissions limit assessment of whether the uncertainty represented by the interval captures all relevant sources of uncertainty. The interval should be treated as uncertainty around the reported unadjusted comparison, not as a complete account of every design or data limitation.

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 display is a planning illustration based on the observed effect, not the original design calculation. Its entries are participants per arm, not total events. The planning 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 set to the stated fraction of the observed effect. The registry record does not report the trial’s own sample-size assumptions or justification.

For the row labelled 100% of the observed effect, the illustration gives 297 participants per arm for 80% power and 397 participants per arm for 90% power. For 75% of the observed effect, it gives 527 participants per arm for 80% power and 705 participants per arm for 90% power. For 50% of the observed effect, it gives 1,185 participants per arm for 80% power and 1,586 participants per arm for 90% power.

In plain language, the 80% power column describes the illustrated number of participants per arm for a design with an 80% chance of detecting the specified effect under the stated assumptions, if that effect and assumptions held. The 90% power column describes the corresponding planning target for a 90% chance. Higher planned power requires a larger sample under the same assumptions. The display also illustrates that smaller effects need more participants to detect than larger effects. These figures are not observed power and do not describe the probability that this completed trial detected its effect.

The observed effect used for these illustrations comes from the posted comparison, which has considerable uncertainty. Planning based on an observed effect can be unstable, especially when the estimate is imprecise. The figures are therefore scenario-specific planning aids, not recommendations or evidence about what the trial originally planned. They should not be read as a substitute for the original protocol or a design justification, neither of which is reported in the registry record.

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 = -2.740; pooled standard deviation σ = 11.901.
2. n = 2 × (1.96 + 0.842)² × 11.901² / -2.740² = 296.1, rounded up to 297 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 <- -2.74                      # observed difference in means
sigma <- 11.9011015067165                     # 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(245635)
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 = 297 (80% power), 397 (90% power)
 75% of observed effect: n per arm = 527 (80% power), 705 (90% power)
 50% of observed effect: n per arm = 1,185 (80% power), 1,586 (90% power)
power.t.test: 297.1 
Simulated power at n = 297 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.

13. Multiplicity, Missing Data and Analysis Populations

Multiplicity arises when multiple outcomes, time points, subgroup comparisons, or statistical tests are examined. Testing several questions can increase the chance that at least one result appears noteworthy by chance, so a prespecified analysis plan may describe how multiple comparisons will be handled. The registry lists one outcome measure and zero posted secondary outcome analyses. It does not report a multiplicity adjustment or explain whether any analyses beyond those listed were conducted.

The primary outcome timeframe compares baseline with week 12, described as two time points. The record identifies this as the primary outcome but does not report whether other time points were considered, whether interim analyses occurred, or whether the outcome definition changed. Such details are not reported in the registry record. The absence of posted secondary analyses should not be taken as proof that no other analyses were performed outside the registry entry.

Missing-data methods are also not reported. The registry gives group counts for the primary outcome results, but does not explain whether all randomized participants had measurements at both baseline and week 12, how missing scores were handled, or whether any imputation was used. It does not give reasons for missingness, participant flow, or a sensitivity analysis assessing the impact of missing data. Therefore, the outcome counts alone cannot establish completeness of follow-up or the robustness of the result to missing measurements.

The analysis population is not reported in the registry record. It is not stated whether the comparison included all randomized participants, participants who received assigned treatment, or only those with available endpoint data. No intention-to-treat or per-protocol designation is provided. The record also does not specify whether the analysis accounted for adherence or treatment discontinuation.

These omissions matter because multiplicity, missingness, and analysis population can influence the interpretation of an estimate, even when the group means and comparison are available. The registry offers a concise primary outcome result but does not provide the statistical plan needed to assess the complete set of analyses or the handling of incomplete data.

14. Safety Results

The registry reports serious adverse events as affected participants divided by participants at risk.

Other (non-serious) adverse events

Participants with at least one non-serious event at or above the registry's 0% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: suicidal behavior (4.0%).

The registry reports serious adverse events for 1 participant at risk in the fluoxetine group’s event count and 0 in the placebo group’s event count. The stated numbers at risk are 25 for fluoxetine and 18 for placebo. These figures describe the recorded serious adverse event counts in the registry; they do not provide a complete safety profile or establish that an event was caused by the assigned intervention.

The registry also lists Suicidal Behavior as a serious adverse event term at 4.0%. It does not state which group this term belongs to or provide further event-level context in the information recorded. The record lists other adverse event counts of 2 for fluoxetine and 3 for placebo. Definitions, severity categories, event timing, and assessment procedures are not reported in the registry record.

Deaths are recorded as null for both groups. A null entry is not the same as a reported count of zero, so the record does not establish that no deaths occurred. Serious adverse event counts should be read separately from other adverse event counts, and neither set of figures should be used as a standalone comparison of overall safety. The registry record does not report the duration of safety follow-up or detailed criteria for classifying events.

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

The posted result describes average BDD-Y-BOCS score changes between baseline and week 12 in the fluoxetine and placebo groups. Both reported group means are negative, and the fluoxetine group’s mean change is more negative. The unadjusted mean difference is -2.74 units, with a confidence interval from -9.98 to 4.5 units and an unadjusted p-value of 0.458. The interval includes zero and spans differences in either direction, so the estimate does not precisely establish the direction or size of the group difference.

The endpoint is a measure of BDD-related obsessions and compulsions, with higher scores described as worse. The posted means are group averages; they do not show how each participant changed, whether all participants improved or worsened, or whether changes were important to participants. The registry record does not report a threshold for a clinically important change. The results therefore describe the recorded score summaries, not a guarantee of an individual outcome.

The randomized design is relevant to the intended comparison, but interpretation also depends on information the registry does not report: the precise allocation procedures, analysis population, missing-data handling, and full statistical analysis plan. The record includes no posted secondary outcome analyses and no separate detailed model analysis. It does not provide enough information to assess every possible source of bias or determine whether the posted summaries represent all randomized participants.

The findings do not establish a treatment effect beyond the measured outcome and timeframe, and they do not determine results in populations outside the trial’s stated age and condition. The registry entry concerns children and adolescents aged 16 and younger with BDD who qualified for the study; additional participant characteristics and eligibility details are not reported. The result also does not establish comparative safety: the adverse-event record is limited, event definitions and follow-up details are not reported, and the null death entries cannot be read as confirmed zero deaths.

In short, the numerical pattern is a more negative mean change in the fluoxetine group, while the interval and the registry’s limited analysis details require a cautious interpretation. The numbers alone do not prove causation, establish clinical importance, or support a recommendation about care.

16. Why This Trial Matters Statistically

This trial illustrates why a group mean and a point estimate are not enough to describe a clinical comparison. The mean difference is -2.74 units, but its confidence interval extends from -9.98 to 4.5 units. The point estimate gives a compact description of the contrast, while the interval shows that the estimate remains uncertain and compatible with differences on either side of zero. Readers need both to understand what the posted result can support.

The outcome is a continuous scale score, reported as a change from baseline to week 12. That makes the unit and direction of the estimate interpretable: the comparison concerns average change on the BDD-Y-BOCS, a measure for which higher scores indicate worse impairment. At the same time, a group mean cannot describe the spread of individual experiences. The standard deviations show variation within each arm, but the registry does not include individual-level data or the distribution of changes.

The record also demonstrates the distinction between a completed trial’s observed result and a planning illustration. The sample-size scenarios use fractions of the observed effect and specified assumptions, but they are not the original design rationale and do not represent observed power. Keeping these concepts separate avoids treating a planning calculation as evidence about what the trial was designed to detect or what it actually achieved.

Finally, a registry entry can report a primary outcome while leaving important statistical details unavailable. Here, the analysis population, missing-data approach, exact inferential procedure, multiplicity handling, and full analysis plan are not reported in the registry record. Those absences limit how fully readers can evaluate the precision and robustness of the result. Careful interpretation means retaining the useful numerical information while recognizing the boundaries of what is documented.

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 comparison is based on outcome summaries, and the registry does not report participant-level observations. The mean and standard deviation therefore cannot reveal the distribution of individual changes, identify unusual values, or show whether the average represents a common pattern across participants. The counts attached to the outcome do not establish how many participants had complete measurements at both time points.

Important design and analysis details are not reported in the registry record. These include detailed eligibility criteria, participant characteristics, allocation sequence and concealment procedures, the roles covered by quadruple masking, and the method used to calculate the comparison’s confidence interval and p-value. The analysis population, missing-data handling, baseline adjustment, and sensitivity analyses are also not reported. This limits independent assessment of potential bias and the robustness of the estimate.

The outcome is specified at week 12, but the record does not provide further information about how the measure was administered, who completed or rated it, or whether measurement procedures were consistent across participants. A threshold for a clinically important difference is not reported, so the statistical estimate cannot be translated into a registry-defined judgment about importance to participants.

The safety record is similarly limited. It provides adverse event counts and a serious adverse event term, but does not report event definitions, timing, severity details, or group attribution for the listed Suicidal Behavior term. Death entries are null rather than reported zeros. The record also does not give detailed follow-up or participant flow information to explain the outcome counts.

Finally, the registry’s planning illustration is based on the observed effect rather than the trial’s original design assumptions, which are not reported. Those figures should not be used to infer what the study was planned to detect. These limitations concern documentation and interpretation; they do not change the values that the registry reports.

18. Related Tutorials

19. Related Statistical Calculators

20. Sources

• ClinicalTrials.gov: Fluoxetine in Pediatric Body Dysmorphic Disorder — NCT00245635.
• Registry record last updated 2018-04-27; status last verified 2018-03.
• Results first posted 2015-06-29.
• Data retrieved from ClinicalTrials.gov on 2026-11-09.
• Group sizes, means, event counts and sponsor-reported p-values are reproduced as posted. Effect estimates, confidence intervals and p-values not attributed to the sponsor are independently calculated from the posted summary statistics.

21. Record Summary

The registry record describes NCT00245635, a completed phase 4, randomized, quadruple-masked, parallel-group trial sponsored by Montefiore Medical Center. It compares fluoxetine with placebo in children and adolescents aged 16 and younger with BDD. The primary outcome is change in BDD-Y-BOCS total score from baseline to week 12.

The recorded mean changes are -12.27 for fluoxetine and -9.53 for placebo. The unadjusted mean difference is -2.74, with a confidence interval from -9.98 to 4.5 and an unadjusted p-value of 0.458. The interval includes zero, and the registry does not report the complete analysis plan or several details needed to assess missing data and analysis populations.

The record lists no secondary outcome analyses. It includes adverse event counts and a serious adverse event term, but does not provide a complete safety account. The planning sample-size illustrations are based on the observed effect, not the trial’s original design assumptions.