Placebo Controlled Trial of Valproate and Risperidone in Young Children With Bipolar Disorders: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Placebo Controlled Trial of Valproate and Risperidone in Young Children With Bipolar Disorders (NCT00221403), a randomized phase 3 study. The primary outcome is YMRS.

2. Clinical Question

3. Trial Design

This was a Phase 3, randomized, parallel-group trial with triple masking. It enrolled 46 participants and is recorded as completed. The registry lists Children's Hospital Medical Center, Cincinnati, as the sponsor. The study start is recorded as 2004-09, and primary completion as 2011-05. The stated aim was a preliminary controlled trial of valproate and risperidone in children with bipolar disorders, alongside clinical characterization of a very young cohort.

The three treatment groups were valproate, risperidone, and placebo. The arm descriptions say that valproate and placebo were posted in liquid form matched for taste and color; risperidone was also a matched liquid. Valproate was described as administered twice daily, beginning at 10 mg/kg/day. Risperidone was also described as administered twice daily, but its dose is not reported in the registry record. The placebo description identifies it as matched with the active comparator. These descriptions do not fully specify treatment adjustment, adherence assessment, or the duration of treatment beyond the primary outcome timeframe.

The primary endpoint was the change from baseline in Young Mania Rating Scale total score at 6 weeks. The record identifies the reported statistic as a least-squares mean and gives a standard deviation for each group. It lists arm counts of 21 for valproate, 18 for risperidone, and 7 for placebo. Although the record states the overall enrollment, the reason for the observed distribution across groups is not reported in the registry record.

The registry records one outcome measure and no posted analyses. It does not report a formal sample-size justification, the planned analysis population, a detailed statistical analysis plan, or the methods used to handle missing outcome data. The record's design labels and arm descriptions are not identical in their masking terminology, and no explanation 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.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Subjects were male or female outpatient subjects, 3.0-7 years,11 months of age, with bipolar I disorder, mixed or manic episode, psychotic or nonpsychotic, according to DSM IV-TR criteria, with a score ‡ 20 (considered to be moderate severity) on the Young Mania Rating Scale (YMRS) at the time of randomization.

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 YMRS, measured over 6 weeks.

The primary endpoint was change from baseline in the Young Mania Rating Scale total score at 6 weeks. The scale ranges from 0 to 60, and higher scores indicate more severe mania. Under the definition in the registry record, a negative change represents a reduction in the score and in manic symptoms. The outcome therefore expresses change on a continuous scale rather than an event count or a category.

A continuous measure can retain information about the amount of change across the scale. In general, a baseline-to-follow-up change score describes how a participant's recorded score differs from their baseline value; a treatment-group comparison can then assess whether average change differs between groups. The clinical and statistical meaning depends on the direction of the scale and on a clearly specified method for estimating and comparing those changes.

The posted summaries are 10.0 for valproate, 18.82 for risperidone, and 4.29 for placebo, each reported as a least-squares mean, with standard deviations of 2.46, 1.55, and 3.56, respectively. Because negative change is defined as improvement, the posted positive group means are in the direction of increased scores. However, the registry record does not provide an estimable treatment contrast or its uncertainty. The summaries alone do not show whether the observed differences are statistically distinguishable or how precisely a between-group effect is estimated.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

A least-squares mean is a model-based estimate of a group mean. In general, it may account for terms included in a statistical model, whereas a simple arithmetic mean is calculated directly from observed values. The registry identifies the primary outcome summaries as least-squares means, but the model specification, included covariates, and estimation procedure are not reported in the registry record. It is therefore not possible to determine from the record what adjustment, if any, was made beyond the label.

What does the standard deviation describe?

The standard deviation describes the spread of observations around a mean. A smaller standard deviation indicates that observations are more tightly grouped around that mean; a larger one indicates more spread. It is a descriptive measure of variation, not a direct measure of how precisely a group mean or treatment difference is estimated. Precision depends on the amount of information available and the analysis method, among other features. The registry reports standard deviations with the three group summaries, but does not report confidence intervals for those summaries or for comparisons.

Why is a treatment contrast different from group means?

A treatment contrast is an estimate of the difference between groups, defined according to a stated comparison and analysis method. Group means are summaries within each group; they do not themselves report the estimated contrast, its standard error, or its uncertainty interval. In general, uncertainty for a contrast depends on the variability in each group, the number of observations contributing to the analysis, and the statistical model. The record explicitly says values are missing or not estimable for a two-arm comparison of this continuous endpoint. Therefore, no contrast should be reconstructed or interpreted as a sponsor-reported treatment effect from the posted means.

How should confidence intervals be understood?

In general, a confidence interval is a range of values compatible with a statistical estimate under a specified method and its assumptions. For a treatment contrast, it conveys both the estimated direction and the precision of the estimate. A narrow interval indicates greater precision than a wide interval, while an interval spanning values on either side of no difference may leave the direction uncertain. No confidence interval for the primary treatment comparisons is reported in the registry record. A confidence interval cannot be inferred from the least-squares means and standard deviations alone without additional information about the analysis and observations.

Why can over-dispersion matter?

Over-dispersion is a general issue for count outcomes when observed variability exceeds what a basic count model assumes. It can make standard errors too small if not accounted for, leading to intervals that are too narrow or tests that overstate evidence. This trial's primary endpoint is continuous, not a count, and the record reports a standard deviation for each group. Over-dispersion is therefore not the central issue for interpreting this outcome. More generally, matching the statistical model to the endpoint and its variability is important; the model used for the primary outcome is not reported in the registry record.

What do the planning assumptions say?

The registry describes sample-size illustrations based on the observed effect, with a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. The target effect is described as a stated fraction of the observed effect. These are planning assumptions, not evidence about the trial's observed statistical power. Planning illustrations estimate what might be needed under selected assumptions; they do not change the interpretation of the trial's reported outcome. The registry record does not report a formal sample-size justification for the trial's original design. The planning figures are not a substitute for an estimated treatment contrast, its confidence interval, or the trial's own analysis.

Statistical results also depend on analysis choices that are not detailed in the registry record. The record does not report a model formula, handling of baseline scores, assumptions about missing observations, or whether any sensitivity analyses were performed. These matters affect reproducibility and can affect estimates and their uncertainty. Because no analysis is posted for the primary comparison, interpretation should remain descriptive rather than inferential.

10. Confidence Intervals for Each Arm

No two-arm comparison is shown for this trial, but each arm's mean can be given with a t-based 95% confidence interval computed from the posted mean, spread and number of participants.

R code and output

arm <- c("Valproate (VPA)", "Risperidone", "Placebo")
n   <- c(21, 18, 7)
m   <- c(10, 18.82, 4.29)
sp  <- c(2.46, 1.55, 3.56)
sd  <- sp      # posted standard deviations
for (i in seq_along(n)) {
  se <- sd[i] / sqrt(n[i]); h <- qt(0.975, n[i] - 1) * se
  cat(sprintf("CARM|%s|%d|%.3f|%.3f|%.3f|%.3f\n", arm[i], n[i], m[i], sd[i], m[i] - h, m[i] + h))
}

CARM  Valproate (VPA)  21  10.000  2.460  8.880  11.120
CARM  Risperidone  18  18.820  1.550  18.049  19.591
CARM  Placebo  7  4.290  3.560  0.998  7.582

11. Multiplicity, Missing Data and Analysis Populations

The registry identifies one primary outcome measure, the change from baseline in Young Mania Rating Scale total score at 6 weeks. It records no secondary outcomes. Although there are three treatment groups and more than one possible pairwise comparison, a multiplicity strategy for handling multiple comparisons is not reported in the registry record. The record also does not say whether additional analyses were planned or conducted outside the posted outcome information.

Multiplicity matters in general when several hypotheses are tested: the chance of at least one apparently unusual result can increase as the number of tests grows. Methods such as adjusting significance thresholds or controlling a specified error rate can address this, but the appropriate approach depends on the prespecified analysis plan. No such plan or adjustment is reported. The registry reports no inferential analysis of the primary outcome, so there are no posted pairwise p-values to interpret or adjust.

Missing-data handling and the analysis population are also not reported in the registry record. It is not clear which participants contributed to the primary endpoint, whether any outcome measurements were unavailable, or how any unavailable measurements were treated. The arm counts listed with the outcome do not, by themselves, establish the analysis population or explain why those counts differ. The record does not report whether the analysis followed randomized assignment or used another population definition.

In general, missing outcome data can affect both the estimate and its precision, especially when the availability of measurements is related to outcomes or treatment. Analysts often examine assumptions and conduct sensitivity analyses, but such methods are not described in the registry record. Without information about missingness, analysis populations, and analysis choices, the posted group summaries cannot be treated as a complete account of the inferential analysis.

12. 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 0.02% reporting threshold.

Most frequent non-serious adverse event terms

The registry reports serious adverse event counts of 0 for risperidone among 18 at risk, 0 for valproate among 21 at risk, and 0 for placebo among 7 at risk. These are the posted group-level counts of participants affected by serious adverse events. The record lists no serious adverse event terms.

For other adverse events, the registry reports 0 among 18 at risk in the risperidone group, 1 among 21 at risk in the valproate group, and 0 among 7 at risk in the placebo group. Deaths are recorded as 0 in each group. These figures describe what was entered in the registry record; they do not establish that no other events occurred or were collected in other ways.

Counts should be read with their corresponding at-risk group counts and in the context of how events were defined, collected, and reported. The registry record does not provide event definitions, follow-up details for safety, severity information for the other adverse event, or a statistical comparison of safety outcomes. With no reported serious events in the listed groups, the record still does not establish equivalent safety, exclude uncommon events, or characterize risks beyond the period and reporting represented in the record.

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

The results describe a randomized, parallel-group trial of valproate, risperidone, and placebo in young children with bipolar disorders. The registry posts least-squares mean changes in Young Mania Rating Scale score at 6 weeks, together with standard deviations and group counts. Under the scale definition, lower scores indicate less severe mania, and a negative change indicates a reduction from baseline. The posted means are positive, but the record does not provide a reported treatment contrast to quantify or test differences among groups.

Without an estimable comparison, the group summaries cannot establish whether any treatment differed from placebo or from the other active treatment. They also do not establish equivalence or the absence of an effect. A lack of a reported p-value or confidence interval is not evidence that a difference is absent; it means the registry record does not provide the inferential result needed to assess the comparison's magnitude and precision.

The setting and population also matter. The eligibility description concerns outpatient subjects aged 3.0-7 years,11 months with specified bipolar I presentations and a threshold YMRS score, but the registry's population description is truncated. The findings described by these records should not be assumed to apply to people outside that defined population. Details about recruitment, participant characteristics beyond the stated eligibility, and the representativeness of those enrolled are not reported in the registry record.

The safety counts are similarly descriptive. Zero recorded serious adverse events in the listed groups does not show that the treatments have identical safety profiles or rule out events that were not observed or recorded. The record gives no inferential safety analysis or detailed event definitions.

Overall, the registry offers limited descriptive outcome and safety information, but no posted primary treatment comparison. Interpretation should therefore stay close to the reported summaries and the limits of the record. It is not possible from these data to make a definitive claim about comparative benefit, lack of benefit, or comparative safety.

14. Why This Trial Matters Statistically

This trial is statistically notable because the registry describes a randomized, masked comparison involving three groups and a continuous primary endpoint, yet does not post an estimable two-arm comparison. It illustrates an important distinction: reporting a mean and standard deviation for each group is not the same as reporting a treatment effect with uncertainty. A reader needs the contrast and its precision to evaluate how compatible the results are with a range of possible differences.

The record identifies a clinically interpretable direction for the Young Mania Rating Scale: higher scores indicate more severe mania, and a negative change indicates a reduction. That direction is essential for reading the group summaries. It also makes clear why the sign of a reported change matters. But interpretation of direction cannot substitute for a comparative analysis, particularly when the trial has multiple treatment groups.

The record also emphasizes the importance of transparency about analysis populations, missing data, and methods. Those details affect what an estimate represents and how confidently it can be interpreted. Here, the registry lists group counts and outcome summaries but does not report which participants contributed to the analysis, how missing outcome data were addressed, or what model produced the least-squares means. These gaps constrain reproducibility and leave uncertainty about the analysis behind the summaries.

Finally, planning illustrations and observed results answer different questions. Sample-size planning uses assumptions about a target effect and statistical method; it does not reveal whether the completed trial demonstrated a treatment difference. This distinction is particularly important when the registry does not post an inferential comparison. The record is therefore a useful example of why transparent estimates, uncertainty intervals, and analysis descriptions are central to interpreting clinical trial results.

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 primary limitation for interpreting this trial record is the absence of an estimable two-arm comparison for the continuous endpoint. The group means and standard deviations are posted, but the registry does not provide a treatment contrast, confidence interval, or p-value. The record also lists no posted analyses. As a result, the primary outcome summaries cannot support a statistical conclusion about comparative effects.

There is limited information about how the reported least-squares means were estimated. The model specification, covariates, treatment of baseline scores, and assumptions are not reported in the registry record. The analysis population and missing-data methods are also not reported, so it is unclear who contributed to the endpoint summaries and how unavailable outcome measurements, if any, were handled.

The record's population description is incomplete: it ends with the phrase at the time of ra. That truncation limits understanding of the full eligibility criteria. The reason for the group counts reported with the outcome is not reported in the registry record. The masking information also uses different terms: the study design lists triple masking, while the arm descriptions refer to double-blinded medication administration. The record does not explain the distinction.

The safety record provides counts but little detail about event definitions, ascertainment, timing, severity of the other adverse event, or statistical comparisons. No serious adverse event terms are listed. These omissions limit the safety interpretation and mean that the recorded counts should not be treated as a complete characterization of risk.

Finally, a formal sample-size justification and detailed statistical analysis plan are not reported in the registry record. The planning assumptions described in the record do not establish the original design rationale or observed power. Taken together, these limitations make the posted findings descriptive and leave important questions about analysis conduct and precision unresolved.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Placebo Controlled Trial of Valproate and Risperidone in Young Children With Bipolar Disorders — NCT00221403.
• Registry record last updated 2020-10-20; status last verified 2020-09.
• Results first posted 2020-10-20.
• 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

The registry record describes a completed, randomized, parallel-group Phase 3 trial sponsored by Children's Hospital Medical Center, Cincinnati. It lists triple masking, enrollment of 46, a start of 2004-09, and primary completion of 2011-05. The treatment groups were valproate, risperidone, and placebo.

The primary outcome was change from baseline in Young Mania Rating Scale total score at 6 weeks. The record posts least-squares means and standard deviations by group, but states that values are missing or not estimable for a two-arm comparison of the continuous endpoint. It lists one outcome measure, no posted analyses, and no secondary outcomes. The registry also reports adverse event counts, including no serious adverse events recorded in the listed groups.

The record does not report a formal sample-size justification, a detailed analysis plan, missing-data methods, or the analysis population. Its group summaries can be described, but the absence of an inferential comparison prevents assessment of a treatment difference and its precision.