Active Reference (Fluoxetine) Fixed-dose Study of Vortioxetine in Paediatric Participants Aged 7 to 11 Years With Major Depressive Disorder (MDD): Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Active Reference (Fluoxetine) Fixed-dose Study of Vortioxetine in Paediatric Participants Aged 7 to 11 Years With Major Depressive Disorder (MDD) (NCT02709655), a randomized phase 3 study. The primary outcome is Change From Baseline in Children Depression Rating Scale - Revised (CDRS-R) Total Score at Week 8 of Phase B.

2. Clinical Question

3. Trial Design

This was a randomized, double-blind, parallel-group Phase 3 study of vortioxetine in paediatric participants aged 7 to 11 years with major depressive disorder. The registry record lists an overall enrollment of 683 and a completed status. The recorded start date is 2016-05-18, and the primary completion date is 2022-01-21. The study was sponsored by H. Lundbeck A/S.

The interventions listed were vortioxetine 10 mg/day, vortioxetine 20 mg/day, fluoxetine 20mg/day, and placebo. The primary outcome results are from the double-blind treatment period, Phase B, and use baseline at Week 4 of Phase A. The primary endpoint was assessed at Week 8 of Phase B. The registry record describes the treatment period used in the analysis as 8 weeks.

The primary outcome population was not simply everyone enrolled. It included randomized participants who took at least 1 dose of double-blind study drug and had both a valid baseline CDRS-R assessment and at least 1 valid post-baseline assessment. The primary outcome counts were 136 for placebo, 132 for vortioxetine 10 mg, 134 for vortioxetine 20 mg, 78 for fluoxetine, and 266 for the vortioxetine average. These are the counts attached to the reported outcome groups, not the overall enrollment.

The fluoxetine arm was an active-reference group, and the registry record states that a decision had been taken to stop recruitment into it. The reason for stopping recruitment is not reported in the registry record. Allocation ratios, sequence generation details, the original sample-size justification, and details of how concealment was implemented are not reported in the registry record.

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: Participants who demonstrated an incomplete improvement in depressive symptoms at the end of the SB period (Week 4) entered the DB period to receive BPI (2 sessions) and treatment as follows: Prior to the interim analysis, participants were randomized in a 1:1:1:1 ratio to placebo, vortioxetine 10 mg/day, vortioxetine 20 mg/day, or fluoxetine 20 mg/day. After interim analysis, participants were randomized in a 1:1:1 ratio to placebo, vortioxetine 10 mg/day, or vortioxetine 20 mg/day.

Single-Blind Phase (4 Weeks)

Reasons for not completing this period:

Double-Blind Phase (8 Weeks)

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Full analysis set (FAS) included all participants randomized to the double-blind, 8-week treatment period (Phase B) who took at least 1 dose of double-blind study drug and who had a valid baseline (Week 4 of Phase A) assessment and at least 1 valid post-baseline assessment of the CDRS-R total score. Overall number of participants analyzed = participants evaluated for this outcome measure.

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: All-patients-treated set Phase A (APTS\_A) included all participants enrolled to the 4-week Single-blind period (Phase A) who received at least 1 dose of study drug.

6. Primary Endpoint

The registry lists the primary outcome as Change From Baseline in Children Depression Rating Scale - Revised (CDRS-R) Total Score at Week 8 of Phase B, measured over Baseline (Week 4 of Phase A), Week 8 of Phase B.

The primary endpoint was change from baseline in the Children’s Depression Rating Scale–Revised (CDRS-R) total score at Week 8 of Phase B. Baseline was defined as Week 4 of Phase A. The CDRS-R is a clinician-rated measure of depression severity in children and adolescents, based on interviews with the child and parent. Its 17 items include ratings of observed behaviour as well as other aspects of symptoms. The reported total score ranges from 17, representing normal functioning, to 113, representing severe depression.

Most items use a 7-point rating scale, while sleep disturbance, appetite disturbance, and listless speech use a 5-point scale. A rating of 1 indicates normal functioning, and higher ratings indicate a greater degree of depression. For that reason, a negative change from baseline indicates a decrease in the measured score. The endpoint is useful for summarizing change in a clinician-rated severity measure across the treatment period; interpreting its clinical importance also requires context beyond the score change alone.

The primary outcome is reported as a least-squares mean change, with standard error for each group. The comparison estimates express differences in mean change between a treatment group and placebo. The registry record does not report a threshold for a clinically important difference on the CDRS-R.

7. Results Posted in the Registry

• Double-Blind: Vortioxetine 10 mg: mean -19.20 versus -17.48 on control; difference -1.72 (95% CI -5.49 to 2.05), standardized effect d = -0.11; z-test p = 0.371.
• Double-Blind: Vortioxetine 20 mg: mean -19.94 versus -17.48 on control; difference -2.46 (95% CI -6.23 to 1.31), standardized effect d = -0.16; z-test p = 0.201.
• Double-Blind: Fluoxetine 20 mg: mean -20.78 versus -17.48 on control; difference -3.30 (95% CI -7.40 to 0.80), standardized effect d = -0.22; z-test p = 0.115.
• Vortioxetine Average (Avg. VOR): mean -19.57 versus -17.48 on control; difference -2.09 (95% CI -5.58 to 1.40), standardized effect d = -0.12; z-test p = 0.240.

8. Statistical Interpretation

9. Statistical Methods Explained

What does a least-squares mean represent?

A least-squares mean is a model-based estimate of a group’s mean outcome. Unlike a simple average calculated only from observed scores, it is estimated within a statistical model. The registry record says the primary outcome used a restricted maximum likelihood-based Mixed Model Repeated Measurements approach. The model accounts for the repeated outcome assessments over time and estimates the mean change at the specified assessment point. The reported primary endpoint values are least-squares mean changes from baseline, not raw individual observations.

The registry’s description says that the baseline assessment was at Week 4 of Phase A and the primary assessment was at Week 8 of Phase B. The reported full analysis set required a valid baseline score and at least 1 valid post-baseline score, in addition to randomization and at least 1 dose during the double-blind period. The full model specification, including all covariates and the covariance structure, is not reported in the registry record.

Why use a repeated-measures model?

Repeated-measures models are designed for studies where outcomes are measured at multiple time points for the same participant. Measurements from one participant are generally related to one another, rather than independent. A model can use the pattern of available repeated observations to estimate group outcomes over time while accounting for that within-participant structure. As a general statistical point, this approach can use incomplete outcome sequences without requiring every participant to have a measurement at every scheduled time, provided its assumptions about the data and missingness are appropriate.

The registry record identifies the model family and estimation approach, but it does not report the assumptions used for missing data, the amount or pattern of missing outcome data, or any sensitivity analyses for those assumptions. The model name alone does not establish that missing-data concerns have been ruled out.

What does restricted maximum likelihood do?

Restricted maximum likelihood, often shortened to REML, is a general method for estimating variance components in mixed models. Variance components describe sources of variation, including differences across participants and the pattern of repeated measurements. Estimating these components helps determine how the model represents uncertainty in its mean estimates. REML is the estimation method named in the primary outcome description; the registry record does not provide further computational details.

How should standard errors and confidence intervals be read?

A standard error describes the estimated uncertainty around a model-based mean. It is not the spread of individual participant scores. For example, the placebo least-squares mean change is -17.48 with standard error 1.35. A confidence interval for a treatment contrast gives a range of values compatible with the observed data under the analysis method and its assumptions. A wider interval indicates less precision than a narrower interval. The registry’s reported comparisons use 95% confidence intervals.

The reported confidence intervals for the four comparisons include zero. Zero represents no mean difference between the compared groups on the outcome scale. This means the intervals do not rule out no difference, while also allowing values on either side of zero. A confidence interval is not a guarantee that the true value lies in the interval, and it does not address unmeasured bias or whether the model assumptions are reasonable.

What do the p-values contribute?

A p-value describes how unusual data at least as incompatible with a specified null hypothesis would be, assuming that hypothesis and the analysis model are correct. It is not the probability that the null hypothesis is true, nor is it a measure of the size or clinical importance of a difference. The registry reports mixed-model p-values of 0.234 for vortioxetine 10 mg, 0.088 for vortioxetine 20 mg, 0.053 for fluoxetine, and 0.094 for the vortioxetine average. Those values should be read alongside the estimated differences and their confidence intervals, not in isolation.

The unadjusted recalculations use the posted group means and standard errors to approximate comparisons using summary statistics. The pooled standard deviations in those recalculations are approximate, as indicated in the calculation information. Such recalculations do not reproduce the full repeated-measures model, its covariate adjustment, or its handling of observations over time. General statistical practice therefore treats them as a separate descriptive check rather than as a substitute for the registry’s reported analysis.

Why can different analysis methods give different results?

A direct comparison based on summary statistics and a repeated-measures model can produce different standard errors, confidence intervals, and p-values. The model may incorporate repeated assessments and other model terms, while a summary-statistic recalculation cannot recreate those features. Differences in estimates can also reflect which observations and estimands are used. The posted recalculations and sponsor analyses are consequently not interchangeable; the registry’s mixed models analysis is the sponsor’s reported result.

10. Confidence Intervals and Precision

Precision is conveyed most directly by the confidence intervals around the treatment contrasts. For vortioxetine 10 mg versus placebo, the registry reports a mean difference of -1.72 and a 95% confidence interval from -4.56 to 1.11. For vortioxetine 20 mg versus placebo, the mean difference is -2.46, with a 95% confidence interval from -5.29 to 0.37. The fluoxetine comparison is -3.30, with a 95% confidence interval from -6.65 to 0.04. For the vortioxetine average, the mean difference is -2.09, with a 95% confidence interval from -4.54 to 0.36.

Each interval includes zero and extends to values on both sides of zero. Thus, the posted estimates do not pinpoint a single treatment contrast: they are compatible with no mean difference and with a range of possible differences. The breadth of an interval reflects sampling uncertainty under the fitted analysis and should be considered along with the point estimate. The registry does not report a clinically important difference threshold, so precision on the score scale cannot by itself determine clinical importance.

Group standard errors also describe uncertainty in the estimated mean changes, not the variability of scores among individual children. The registry reports standard errors for the group means but does not provide confidence intervals for those means in the listed primary outcome results. For comparisons, the model-based 95% confidence intervals are the most direct posted description of precision.

Confidence intervals do not account for all possible sources of uncertainty. They depend on the analysis model and its assumptions, including how repeated observations and missing outcomes are handled. Details of missingness and sensitivity analyses are not reported in the registry record. They also do not resolve questions about multiplicity or whether the measured score difference is meaningful to participants.

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 planning figures are illustrative calculations based on the observed effect, not the trial’s original design calculations. They use a two-sided significance level of 0.05, equal allocation between the two compared arms, and a normal approximation. The assumed effect is the stated fraction of the observed effect. The figures are participants per arm, not totals across arms and not total events. The trial’s original sample-size justification and design assumptions are not reported in the registry record.

The 80% power column gives the planned number of participants per arm for a hypothetical study to have an 80% chance of detecting the stated effect under those assumptions, if the effect and assumptions were correct. The 90% power column uses the same idea with a higher chance of detection and therefore generally calls for more participants. For example, planning around 100% of the observed vortioxetine 10 mg effect gives 1,315 participants per arm for 80% power and 1,761 participants per arm for 90% power.

Other entries plan for 75% or 50% of the observed effect. A smaller effect is harder to distinguish from random variation, so the planning figures require more participants per arm. These are planning illustrations based on the posted estimates; this is not observed power. They do not indicate the chance that the completed trial’s findings are true or that a treatment has a particular effect.

The curve should be understood as a visual representation of how the planning requirement changes as the assumed effect changes. It does not add new evidence about the treatment comparison. Because these calculations use an effect estimated from the same trial results, they should be treated as sensitivity illustrations rather than as an independent assessment of the study’s design adequacy.

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 = -1.720; pooled standard deviation σ = 15.742.
2. n = 2 × (1.96 + 0.842)² × 15.742² / -1.720² = 1,314.9, rounded up to 1,315 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 <- -1.72                      # observed difference in means
sigma <- 15.7418620579851                     # 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(2709655)
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 = 1,315 (80% power), 1,761 (90% power)
 75% of observed effect: n per arm = 2,338 (80% power), 3,130 (90% power)
 50% of observed effect: n per arm = 5,260 (80% power), 7,042 (90% power)
power.t.test: 1315.9 
Simulated power at n = 1,315 per arm: 0.801 (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. Analyses Reported by the Sponsor

The sponsor’s posted primary comparisons use Mixed Models Analysis and report mean differences in final values, 95% confidence intervals, and p-values. For the vortioxetine average, the reported difference is -2.09, with a 95% confidence interval from -4.54 to 0.36 and p = 0.094. The corresponding reported results are -1.72, -4.56 to 1.11, and p = 0.234 for vortioxetine 10 mg; -2.46, -5.29 to 0.37, and p = 0.088 for vortioxetine 20 mg; and -3.30, -6.65 to 0.04, and p = 0.053 for fluoxetine.

The recalculations are unadjusted estimates from the posted group summary statistics. They give the same displayed point estimates to the reported precision in most comparisons, but their p-values differ from the sponsor’s model-based p-values. That is expected when a summary-statistic approximation is compared with a repeated-measures model: the approaches do not use information in the same way and do not reproduce the same model uncertainty.

The sponsor’s own analysis is the primary registry result for interpreting the trial. The recalculations are best treated as a supplementary check on the scale and direction of the contrasts. They should not replace the sponsor’s mixed-model estimates, confidence intervals, or p-values.

14. Multiplicity, Missing Data and Analysis Populations

The primary outcome description identifies one primary endpoint: change from baseline in CDRS-R total score at Week 8 of Phase B. The registry record also lists numerous secondary outcomes, including CDRS-R changes at other time points, CDRS-R response and remission, subscores, parent- and child-assessed depression measures, and global assessments. It records 21 outcome measures and 20 posted secondary outcomes. The secondary outcome results are not detailed in the registry record used for this interpretation.

When multiple outcomes, time points, or treatment comparisons are tested, the chance of finding at least one apparently unusual result by chance can increase. Statistical approaches can address this multiplicity, for example by adjusting significance thresholds or by defining a testing sequence in advance. Whether such procedures were planned or used for the reported comparisons is not reported in the registry record. The reported p-values should therefore be read as posted values, without assuming that they were adjusted for multiple comparisons.

The registry describes the primary analysis population and the repeated-measures model, but does not report a complete statistical analysis plan, the prespecified hierarchy of outcomes, or details of any interim analyses. Interim analyses are not reported in the registry record. The registry record also does not give detailed missing-data counts, reasons for missingness, or sensitivity analyses testing alternative missing-data assumptions.

The primary analysis population required a valid baseline and at least 1 valid post-baseline assessment. As a general statistical matter, a repeated-measures model can use participants with incomplete repeated measurements, but the validity of its estimates depends on assumptions about why observations are missing and how the model is specified. The model approach alone does not establish that those assumptions hold. The exact handling of missing observations beyond the stated model is not reported in the registry record.

The primary outcome results list group counts of 136 for placebo, 132 for vortioxetine 10 mg, 134 for vortioxetine 20 mg, 78 for fluoxetine, and 266 for the vortioxetine average. The overall enrollment was 683. Those figures refer to different aspects of the record; the registry does not explain every difference between enrollment and the endpoint analysis counts in the registry record used.

15. 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 2% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: forearm fracture (1.2%), tracheitis (0.7%), suicide attempt (0.7%), cellulitis (0.7%), pneumonia (0.7%).

The registry reports serious adverse events by treatment group and gives the number at risk and the number affected. In the single-blind placebo group, 6 of 677 participants were affected by a serious adverse event. During the double-blind period, the figures were 3 of 153 for placebo, 1 of 151 for vortioxetine 10 mg, 2 of 153 for vortioxetine 20 mg, and 1 of 83 for fluoxetine 20 mg. The record reports zero deaths in each of these groups.

The serious-event terms listed most often include forearm fracture at 1.2%, and tracheitis, suicide attempt, cellulitis, pneumonia, and major depression at 0.7% each. These terms describe reported serious events; they do not establish that a study treatment caused an event. The registry record does not provide enough detail to compare exposure duration or other factors that can affect adverse-event interpretation.

The single-blind placebo group is reported separately from the double-blind treatment groups, so its count should not be treated as directly interchangeable with the double-blind group counts. These figures are descriptive summaries of the serious adverse events posted in the registry, not a complete assessment of safety or a basis for medical advice.

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

The primary endpoint captures a clinician-rated change in depression severity on the CDRS-R over the specified treatment period. The reported mean changes were negative in placebo and all treatment groups. The treatment-versus-placebo point estimates were also negative, indicating larger decreases in the measured score by point estimate in the treatment groups. The model-based confidence intervals for all four comparisons include zero, so the posted analyses do not exclude no mean difference.

This result concerns group averages. It does not show how many individual participants improved, how large any individual change was, or whether a score difference was meaningful to children and families. The registry lists response, remission, and other secondary measures, but their results are not detailed. No threshold for a clinically important CDRS-R difference is reported in the registry record.

The trial included an active reference arm, fluoxetine 20 mg/day, as well as placebo and two vortioxetine doses. Recruitment into the fluoxetine arm was stopped, and the reason is not reported in the registry record. The reported comparison estimates and uncertainty should be interpreted in light of their group sizes, the repeated-measures analysis, and the limits of the posted information.

The results do not establish that a treatment works or does not work. They describe estimated changes and uncertainty under the reported statistical model. They also do not address treatment decisions for an individual child. Those decisions require clinical judgment and information beyond the statistical results in the registry record.

17. Why This Trial Matters Statistically

This study illustrates how a trial can report both estimated changes within groups and model-based contrasts against a control group. Those quantities answer different questions: a negative mean change describes the group’s change from baseline, while a treatment contrast estimates how that change differs from placebo. Looking only at within-group change would not separate the treatment contrast from changes that also occurred in the placebo group.

The endpoint was measured repeatedly, and the registry reports a mixed model for the primary analysis. That makes the distinction between a model-based result and an unadjusted recalculation especially important. A simple calculation from summary statistics can provide a check on the direction and approximate scale of an estimate, but it cannot reproduce a repeated-measures model’s full analysis.

The confidence intervals also demonstrate why effect estimates and precision should be considered together. The point estimates are negative, but each posted 95% interval includes zero. The evidence therefore leaves uncertainty about the size and direction of the population contrast. A p-value alone would not communicate that range of plausible effects.

Finally, the planning illustrations show how sample-size requirements depend on the effect size assumed for planning. Smaller assumed effects require more participants per arm under the stated assumptions. Those calculations are distinct from the trial’s original design justification, which is not reported in the registry record, and from the observed results. Keeping those roles separate helps prevent planning figures from being mistaken for evidence about what the trial found.

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 gives limited detail about trial conduct and analysis beyond the design labels, endpoint definition, and stated primary model. Allocation ratios, sequence generation, concealment procedures, and the reason recruitment into the fluoxetine arm was stopped are not reported in the registry record. These details would help readers assess implementation and the interpretation of the active-reference comparison.

The primary endpoint is measured at Week 8 of Phase B. The primary result therefore describes the specified assessment point, rather than longer-term outcomes. The registry lists other outcome time points and secondary measures, but those results are not detailed in the registry record used. The registry record does not report a clinically important difference threshold for the CDRS-R.

The primary analysis includes participants with at least 1 valid post-baseline assessment. The registry names a restricted maximum likelihood-based mixed model for repeated measurements but does not report missing-data amounts, patterns, reasons, or sensitivity analyses. The full model specification is also not reported in the registry record. These omissions limit independent assessment of the assumptions behind the model-based estimates.

The fluoxetine arm had fewer primary outcome participants than the placebo and vortioxetine groups, and recruitment into that arm was stopped. No reason is reported. This makes the active-reference estimate less precise than its point estimate alone might suggest; its posted confidence interval includes zero. The overall enrollment of 683 is not the same as the counts reported for the primary outcome groups.

Finally, the registry lists multiple outcomes and comparisons, but whether multiplicity was addressed is not reported in the registry record. The posted p-values should not be assumed to be adjusted. These limitations do not invalidate the reported results, but they define the scope of what can be concluded from the available record.

19. Related Tutorials

20. Related Statistical Calculators

21. Sources

• ClinicalTrials.gov: Active Reference (Fluoxetine) Fixed-dose Study of Vortioxetine in Paediatric Participants Aged 7 to 11 Years With Major Depressive Disorder (MDD) — NCT02709655.
• Registry record last updated 2022-09-22; status last verified 2022-08.
• Results first posted 2022-09-22.
• Data retrieved from ClinicalTrials.gov on 2026-11-09.
• Group sizes, means, event counts and sponsor-reported p-values are reproduced as posted. Effect estimates, confidence intervals and p-values not attributed to the sponsor are independently calculated from the posted summary statistics.

22. Record Summary

The registry describes a completed randomized, double-blind, parallel-group Phase 3 trial sponsored by H. Lundbeck A/S. It enrolled 683 participants aged 7 to 11 years with major depressive disorder and compared placebo with vortioxetine 10 mg/day, vortioxetine 20 mg/day, and fluoxetine 20 mg/day. Recruitment into the fluoxetine arm was stopped.

The primary outcome was change in CDRS-R total score from baseline at Week 4 of Phase A to Week 8 of Phase B. The analysis used a restricted maximum likelihood-based mixed model for repeated measurements in a defined full analysis set. All reported group mean changes were negative. The sponsor’s treatment-versus-placebo confidence intervals include zero, and further details about multiplicity and missing-data sensitivity analyses are not reported in the registry record.