A Study of Pediatric Participants With Attention Deficit/Hyperactivity Disorder: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for A Study of Pediatric Participants With Attention Deficit/Hyperactivity Disorder (NCT00965419), a na phase 2, phase 3 study. The primary outcome is Percentage of Participants With At Least One Serious Adverse Events (SAE) Over the Duration of the Study.

2. Clinical Question

3. Trial Design

This was a long-term, open-label safety study of edivoxetine in pediatric participants with attention deficit/hyperactivity disorder. The record lists Phase 2 and Phase 3, a single-group model, no masking, and allocation as not applicable. Enrollment was 267. The study started in 2009-09, primary completion was 2015-07, and its status is terminated. The reason for termination is not reported in the registry record.

All enrolled participants were described as receiving edivoxetine. The dosing description specifies a starting dose of 0.1 milligram per kilogram per day or a participant-specific known stable dose for rollover participants, up to 0.3 mg/kg/day, taken orally each day for up to 5 years. Because the study was open-label and had no randomized control group, its design supports description of observed safety outcomes during follow-up, rather than a controlled estimate of treatment effects.

The serious adverse event results are organized into an edivoxetine open-label group and an edivoxetine follow-up group. The registry record does not explain the group allocation process, the relationship between the groups, or whether participants could contribute to both. The groups should not be treated as independent randomized arms on the basis of the information recorded. Their observation periods and analysis denominators differ, so a direct comparison would require design details and methods that are not reported in the registry record.

The primary endpoint timeframe is baseline through Week 252. This is a long-term cumulative safety measure, while the record also lists other safety and growth outcomes. The study was not described as a blinded efficacy trial, and no efficacy comparison is reported.

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: All enrolled participants that received edivoxetine during the open-label treatment phase for up to 5 years and follow-up phase.

Open Label Treatment for Responders

Reasons for not completing this period:

Follow-up Phase

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: All enrolled participants who did not discontinue from the study for the reason 'Lost to follow up' at the first post-baseline observation and who had baseline and at least one post-baseline SAE assessment.

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 Percentage of Participants With At Least One Serious Adverse Events (SAE) Over the Duration of the Study, measured over Baseline Through Week 252.

The primary endpoint is the percentage of participants with at least one serious adverse event over the study duration, from baseline through Week 252. In practical terms, it classifies participants according to whether they had one or more serious adverse events during the relevant observation period. The registry description says that a summary of other nonserious adverse events and all serious adverse events, regardless of causality, is located in the reported adverse events section.

General statistical context: a participant-level percentage is a straightforward way to describe how common an event was among people meeting the endpoint’s analysis criteria. It is clinically and statistically distinct from a count of all events: a participant with repeated events still meets the “at least one” definition once. The measure also differs from an event rate that accounts for time at risk. When participants have different follow-up or exposure, a cumulative percentage can reflect both event occurrence and how much opportunity there was to observe an event.

The endpoint is relevant to a long-term safety study because serious events are important to identify and describe during treatment and follow-up. It does not establish whether treatment caused an event. A causal or comparative interpretation would need a suitable control and information about exposure, event attribution, and analysis methods; those details are not reported in the registry record.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

What does a participant-level percentage describe?

General statistical context: a percentage for participants with at least one event summarizes the proportion of an analysis group who crossed a defined event threshold. For this endpoint, that threshold is at least one serious adverse event during the observation period. It is not a percentage of all events, because repeated events in one participant do not create additional participants meeting the endpoint. The measure is useful for describing how broadly an outcome occurred in the analyzed group.

The meaning depends on the denominator and the rules for entering the analysis population. Here, the primary population requires a baseline assessment and at least one post-baseline serious adverse event assessment, and excludes participants lost to follow-up at the first post-baseline observation. Thus, it is important to read the results alongside those eligibility rules rather than treating enrolled participants as the automatic denominator for every outcome.

Why does follow-up matter?

General statistical context: a cumulative participant percentage reflects the observation window as well as the event experience. Someone followed for longer has more opportunity to have an event recorded than someone followed briefly. A time-to-event analysis or an event rate using person-time can address different questions by incorporating timing or duration of observation, but the primary endpoint is a participant-level percentage. The registry record does not report an exposure-adjusted event rate or a time-to-event analysis for this endpoint.

The timeframe is baseline through Week 252, and dosing was described as continuing for up to 5 years. Those statements define the general intended observation horizon; they do not establish that every participant had the same duration of treatment or follow-up. Participant-level durations and detailed timing of events are not reported in the registry record.

How should the two reported groups be understood?

General statistical context: a randomized comparison aims to make groups comparable at the start through random assignment, supporting interpretation of between-group differences under the study design. This record lists allocation as not applicable and describes a single-group, open-label study. The open-label and follow-up categories therefore should be read as reporting groups, not as randomized treatment assignments. No causal treatment contrast follows from comparing their observed percentages or counts.

The registry record does not report the rules defining entry into the follow-up group, the timing of that follow-up, or whether its participants overlap with the open-label group. Those missing details limit interpretation of group differences. They also make it inappropriate to assume that the groups represent separate populations with equivalent follow-up.

What does the planning illustration mean?

General statistical context: a sample-size planning illustration asks how many participants might be needed under chosen assumptions to detect an effect of a specified size at a selected significance level and power. It is not a description of the actual trial design unless the record explicitly says so, and it is not observed power. The record describes an illustration based on the observed effect, using two-sided significance level 0.05, equal allocation between two arms, and a normal approximation; its effect to detect is a stated fraction of the observed effect. Those assumptions are planning choices, not evidence that this single-group study was designed with equal randomized arms. The registry record does not report a protocol sample-size justification for the study.

General statistical context: normal approximations can be useful for planning, but their accuracy depends on the outcome and the amount of information available. For binary outcomes, a planning calculation also depends on the event probability and the contrast to detect. The reported study design does not supply a randomized control contrast for this endpoint, so an illustration based on an observed contrast should not be mistaken for a controlled estimate of treatment effect.

What analysis methods are documented?

The registry record lists 40 outcome measures and no statistical analyses. The statistical analysis plan, confidence intervals, hypothesis tests, adjustment for covariates, handling of repeated events, and any interim analyses are not reported in the registry record. No additional method should be assumed from the endpoint label alone. General statistical context: a confidence interval, when reported, describes uncertainty around an estimate under a specified method; its interpretation depends on the population and assumptions used to calculate it. No confidence interval for this primary result is reported in the registry record.

For a descriptive safety endpoint, transparent reporting of the numerator, denominator, observation period, and analysis population is central. The posted counts and at-risk denominators give some of this context, while missing details about participant-level follow-up, event timing, and group construction constrain further statistical interpretation.

10. Exact Estimates for Each Arm

No two-arm comparison is shown for this trial, but each arm's result can still be summarized with exact interval estimates computed from the posted counts. The Clopper-Pearson interval is exact and conservative; the Wilson interval is a common closer-to-nominal alternative; the Bayesian credible interval uses a uniform Beta(1, 1) prior. For small groups the three can differ noticeably.

R code and output

arm    <- c("Edivoxetine")
n      <- c(264)
events <- c(7)
for (i in seq_along(n)) {
  ci <- binom.test(events[i], n[i])$conf.int          # Clopper-Pearson exact interval
  cr <- qbeta(c(0.025, 0.975), 1 + events[i], 1 + n[i] - events[i])   # Bayesian, uniform Beta(1,1) prior
  wi <- (function(x, n) { p <- x / n; z <- qnorm(0.975); d <- 1 + z^2 / n
    c((p + z^2 / (2 * n) - z * sqrt(p * (1 - p) / n + z^2 / (4 * n^2))) / d, (p + z^2 / (2 * n) + z * sqrt(p * (1 - p) / n + z^2 / (4 * n^2))) / d) })(events[i], n[i])
  cat(sprintf("ARM|%s|%d|%d|%.1f|%.1f|%.1f|%.1f|%.1f|%.1f|%.1f\n", arm[i], n[i], events[i], 100 * events[i] / n[i],
              100 * ci[1], 100 * ci[2], 100 * wi[1], 100 * wi[2], 100 * cr[1], 100 * cr[2]))
}

ARM  Edivoxetine  264  7  2.7  1.1  5.4  1.3  5.4  1.3  5.4

11. Multiplicity, Missing Data and Analysis Populations

The registry record lists 40 outcome measures. These include treatment-emergent adverse events, discontinuations due to adverse events, discontinuations for any reason, Columbia-Suicide Severity Rating Scale categories, and change from baseline Z-scores for height. Considering many outcomes can raise multiplicity concerns when investigators conduct multiple statistical tests, because the chance of at least one apparently unusual finding can increase across a set of tests. The record lists no statistical analyses, and a multiplicity adjustment or prespecified hierarchy of outcomes is not reported in the registry record. It is therefore not possible to determine whether or how multiplicity was addressed.

Missing-data handling beyond the primary population definition is not reported in the registry record. For the primary endpoint, participants lost to follow-up at the first post-baseline observation were excluded, and participants needed baseline and at least one post-baseline serious adverse event assessment. The record does not explain how other incomplete assessments were handled, how many participants had missing outcome information, or whether sensitivity analyses were conducted. Those details matter because the observed percentage describes the defined analysis population, not necessarily everyone enrolled.

The registry record also does not report an interim analysis plan, a formal statistical analysis plan, or a rationale for the analysis population. The listed enrollment and the endpoint-specific denominators should not be assumed to represent the same population. Without more detail on missingness and participant flow, the potential effect of incomplete follow-up on the safety summary cannot be assessed.

12. Safety Results

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

Other (non-serious) adverse events

Participants with at least one non-serious event at or above the registry's 5% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: epilepsy (1.2%), appendicitis (0.4%), dengue fever (0.4%), ankle fracture (0.4%), contusion (0.4%).

The serious adverse event outcome is a participant-level measure over baseline through Week 252. The edivoxetine open-label group had 7 participants with at least one serious adverse event among 264 at risk; the edivoxetine follow-up group had 1 among 80 at risk. These are the registry’s group-specific counts and denominators, not a randomized comparison. The record does not report deaths; a death count is not reported in the registry record.

The registry lists Epilepsy at 1.2%, and Appendicitis, Dengue fever, Ankle fracture, Contusion, and Intentional product misuse at 0.4% each among the top serious adverse event terms. These terms and percentages describe reported events, but do not by themselves establish treatment causation. The endpoint description distinguishes the primary participant-level summary from other adverse event summaries, which are described as covering all serious adverse events regardless of causality.

General statistical context: a cumulative participant percentage does not show event timing, recurrence, or duration of exposure. Different follow-up amounts can affect the opportunity to observe an event. The record does not give enough detail to adjust these figures for time at risk or to compare groups causally.

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

The posted results describe serious adverse events recorded among participants in an open-label edivoxetine study. They show that at least one serious adverse event was reported for some participants in each of the listed outcome groups, and they provide several serious adverse event terms and percentages. This is useful descriptive safety information for understanding what was recorded during the study period.

The results do not determine whether edivoxetine caused the events. Serious events can occur for reasons unrelated to a study treatment, and the registry description says that serious adverse event summaries include events regardless of causality. The record does not report a randomized comparator or a controlled estimate of treatment-related risk. It therefore cannot establish what event frequency would have been without edivoxetine, or whether the observed frequency differs from another treatment.

The results also do not describe the full pattern of event burden. The primary endpoint counts participants with at least one serious event, not the total number of events or recurrent events. It does not show the timing or severity distribution beyond the serious-event classification, the time each participant was observed, or an event rate adjusted for exposure. Those details are not reported in the registry record.

Interpretation should account for the endpoint-specific analysis population and the separate open-label and follow-up reporting groups. The primary population excludes participants lost to follow-up at the first post-baseline observation and requires baseline and at least one post-baseline assessment. The record does not provide enough information to characterize all missing assessments or the relationship between the groups.

In statistical terms, these are descriptive findings, not evidence from a controlled causal comparison. They should not be extended to make claims about efficacy, comparative safety, or outcomes outside the recorded study population and observation period.

14. Why This Trial Matters Statistically

This study illustrates a common statistical challenge in long-term safety research: describing uncommon but important events when the study is not randomized and has no concurrent control group. A participant-level percentage can communicate whether an event occurred during follow-up, but it cannot separate treatment-associated events from background events or other causes. That distinction is essential when interpreting adverse event summaries.

The record also illustrates why analysis populations and denominators matter. The primary outcome population has explicit assessment requirements and an exclusion for participants lost to follow-up at the first post-baseline observation. The group-specific at-risk denominators differ, and the registry does not explain the relationship between the open-label and follow-up groups. Enrollment alone therefore does not describe who contributed to every safety result.

Long observation periods add another interpretive issue. A cumulative percentage may be influenced by how long participants were observed, while event timing and exposure-adjusted rates answer different questions. The record provides a broad timeframe but does not report participant-level follow-up or a rate accounting for time at risk. This limits comparisons across reporting groups and with other studies.

Finally, the record lists many outcomes but no statistical analyses. Without a reported analysis plan, readers cannot assess inferential methods, multiplicity handling, or sensitivity to missing data. These limitations do not make the descriptive findings meaningless; they define what the findings can support. Careful separation of observed safety summaries from causal or comparative claims is the central statistical lesson.

15. 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 study is open-label and single-group, with allocation listed as not applicable. It does not include a randomized control group, so the serious adverse event summaries cannot provide a controlled estimate of comparative risk. The edivoxetine open-label and follow-up groups are not described as like-for-like comparison arms. Their relationship, group assignment process, and possible overlap are not reported in the registry record.

The primary endpoint is a cumulative participant-level measure over baseline through Week 252. Participant-level follow-up duration, event timing, recurrent event counts, and exposure-adjusted event rates are not reported in the registry record. Differences in the opportunity to observe an event may therefore affect the summaries, and the available figures cannot resolve that issue.

The analysis population has defined exclusions and assessment requirements, but other missing-data procedures and the extent of missing outcome information are not reported in the registry record. The record also does not report a statistical analysis plan, confidence intervals, hypothesis tests, multiplicity procedures, interim analyses, or the reason the study was terminated. The absence of these details limits assessment of uncertainty and potential bias.

The serious adverse event outcome is not a measure of treatment efficacy and does not establish causality. The record’s listed event terms and percentages should be interpreted as safety descriptions, not as proof that treatment produced the events. The record also does not provide enough information to generalize the observed summaries beyond its study population and observation period.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: A Study of Pediatric Participants With Attention Deficit/Hyperactivity Disorder — NCT00965419.
• Registry record last updated 2019-10-08; status last verified 2019-09.
• Results first posted 2018-04-27.
• Data retrieved from ClinicalTrials.gov on 2026-11-09.
• Group sizes, means, event counts and sponsor-reported p-values are reproduced as posted. Effect estimates, confidence intervals and p-values not attributed to the sponsor are independently calculated from the posted summary statistics.

19. Record Summary

NCT00965419 is a terminated, open-label, single-group study of edivoxetine in pediatric participants with attention deficit/hyperactivity disorder. The record lists Phase 2 and Phase 3, enrollment of 267, a start date of 2009-09, and primary completion in 2015-07. Participants were described as receiving oral edivoxetine daily, at a starting dose of 0.1 milligram per kilogram per day or a participant-specific stable dose for rollover participants, up to 0.3 mg/kg/day, for up to 5 years.

The primary outcome was the percentage of participants with at least one serious adverse event from baseline through Week 252. The posted results list 7 of 264 participants in the edivoxetine open-label group and 1 of 80 in the edivoxetine follow-up group. These are descriptive safety figures, not a randomized treatment comparison. The registry record lists 40 outcome measures and no statistical analyses; further details on the statistical plan, missing-data methods, and reason for termination are not reported in the registry record.