This page provides an independent statistical analysis and educational interpretation of publicly reported results. ClinicalTrials.gov provides the official trial registry record. Numerical trial results on this page are restricted to the results posted in the ClinicalTrials.gov record; calculations added by Clinical Biostats are labelled as such.
1. Trial at a Glance
This page covers the registry results for Topiramate vs. Placebo in Preventing Weight Gain in Bipolar Disorder Treated With Olanzapine (NCT00394095), a randomized phase 4 study. The primary outcome is Change in Body Mass Index (BMI).
| Feature | Registry value |
|---|---|
| ClinicalTrials.gov identifier | NCT00394095 |
| Phase | Phase 4 |
| Condition | Bipolar Disorder, Weight Gain |
| Allocation | Randomized |
| Design model | Factorial |
| Masking | Triple |
| Enrollment | 31 |
| Arms in the primary outcome | 2 |
| Lead sponsor | University of Cincinnati |
| Results posted | Yes |
2. Clinical Question
Population
Youth with bipolar disorder who were described in the official title as experiencing a manic or mixed episode. The registry reports an enrollment of 31. The primary-outcome population is described as per protocol and includes an Experimental sample of 16 and a Placebo group of 14.
Intervention
Topiramate in combination with olanzapine. The experimental group’s initial topiramate dose was 25mg bid, titrated over 18 days to 150 mg bid, with flexibility to titrate to 200mg bid as tolerated.
Comparator
Placebo in combination with olanzapine, described in the arm information as a sugar pill. The registry identifies the placebo group as the comparator for the primary outcome.
Primary question
What change in body mass index occurred from baseline to Week 12 in each group? The study objective was to examine topiramate with olanzapine for prevention of weight gain; the posted primary results give group means and confidence intervals, but not a direct between-group effect estimate.
The clinical question concerns weight change during treatment, rather than whether either group’s BMI changed from baseline in isolation. A direct comparison would quantify the difference between the groups and its uncertainty. The registry’s posted results instead report separate group summaries. Those summaries describe the observed mean change within each analysis group, but do not by themselves establish how precisely the groups differed.
The registry also identifies tolerability as a secondary objective and lists tolerability of topiramate as a secondary outcome. The detailed definitions, measurement schedule, and results for that outcome are not reported in the registry record. The available safety counts provide some information about adverse events, but they do not fully describe tolerability or establish the cause of any event.
3. Trial Design
This was a Phase 4 randomized, triple-masked, placebo-controlled pilot study with a factorial model recorded in the registry. The official title describes topiramate versus placebo in combination with olanzapine for prevention of weight gain in manic or mixed youth with bipolar disorder. The registry lists the study as completed and reports enrollment of 31.
The two reported groups were the Experimental Group: Topiramate Group and the Placebo Group. The topiramate regimen began at 25mg bid and was titrated over 18 days to 150 mg bid, with flexibility to titrate to 200mg bid as tolerated. The placebo is described as a sugar pill. The registry’s study description frames the intervention as being used in combination with olanzapine. Details about the olanzapine regimen are not reported in the registry record.
The primary endpoint was change in BMI over 12 weeks, from baseline to Week 12. The result is described as a mean with a 95% confidence interval. The analysis population is described as per protocol, comparing an Experimental sample of 16 with a Placebo group of 14. The registry does not explain how participants entered or left that per-protocol population, or how it relates to all enrolled participants.
Randomization, masking, and the placebo comparator are important design features, but their implementation is not described in detail. The method used to generate the allocation sequence, conceal assignments, and maintain masking is not reported in the registry record. Nor does the record clarify which people were covered by the triple-masking designation. The start is listed as 2006-12 and primary completion as 2009-08. These dates describe the registry record, not necessarily the duration of treatment for any individual participant.
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: Subjects meeting the entry criteria were randomized to treatment with topiramate or placebo in combination with olanzapine for this 12-week study.
Screening
| Stage | Experimental Group: Topiramate Group | Comparator Group: Placebo Group |
|---|---|---|
| STARTED | 17 | 14 |
| COMPLETED | 16 | 14 |
| NOT COMPLETED | 1 | 0 |
Reasons for not completing this period:
| Reason | Experimental Group: Topiramate Group | Comparator Group: Placebo Group |
|---|---|---|
| Withdrawal by Subject | 1 | 0 |
Randomized
| Stage | Experimental Group: Topiramate Group | Comparator Group: Placebo Group |
|---|---|---|
| STARTED | 16 | 14 |
| COMPLETED | 8 | 10 |
| NOT COMPLETED | 8 | 4 |
Reasons for not completing this period:
| Reason | Experimental Group: Topiramate Group | Comparator Group: Placebo Group |
|---|---|---|
| Withdrawal by Subject | 5 | 1 |
| Adverse Event | 1 | 1 |
| Overdose | 1 | 0 |
| Lost to Follow-up | 1 | 2 |
Analysis population for the primary outcome, as defined in the registry: The analysis was per protocol based on change in BMI over 12 weeks in the Experimental sample (N=16) compared to the Placebo group (N=14).
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.
| Characteristic | Unit | Experimental Group: Topiramate Group | Comparator Group: Placebo Group | Total |
|---|---|---|---|---|
| Participants analysed | 16 | 14 | 30 | |
| Age, Categorical: <=18 years | Participants | 16 | 14 | 30 |
| Age, Categorical: Between 18 and 65 years | Participants | 0 | 0 | 0 |
| Age, Categorical: >=65 years | Participants | 0 | 0 | 0 |
| Age, Continuous | years | 14.6 (SD 1.9) | 14.0 (SD 2.0) | 14.3 (SD 1.95) |
| Sex: Female, Male: Female | Participants | 8 | 8 | 16 |
| Sex: Female, Male: Male | Participants | 8 | 6 | 14 |
| Region of Enrollment: United States | participants | 16 | 14 | 31 |
6. Primary Endpoint
The registry lists the primary outcome as Change in Body Mass Index (BMI), measured over 12 weeks.
| Element | Registry entry |
|---|---|
| Result type | Mean |
| Measure of spread | 95% Confidence Interval |
| Unit | kg/m2 |
| Endpoint classification | Continuous Endpoint |
The primary endpoint is change in body mass index (BMI) from baseline to Week 12, reported in kg/m2. BMI is calculated from weight relative to height squared. Tracking change over time expresses how this measure shifted during the study period, rather than comparing only a single measurement taken at the end. The registry states that BMI was computed for all participants and that the primary outcome concerns change over 12 weeks.
BMI is used as a weight-related measure that accounts for height, allowing weight status to be described in a way that is not simply the absolute body weight. In a study whose objective concerns prevention of weight gain, change in BMI is a relevant continuous outcome: it can capture a range of increases or decreases rather than reducing the result to a yes-or-no event. This is a general explanation of the measure’s role, not evidence that BMI fully captures every aspect of weight-related change.
The posted outcome is a mean change for each group, accompanied by a 95% confidence interval. The registry does not report a direct estimate of the difference between the groups for this endpoint. It also does not report details such as how height was measured or whether the endpoint was adjusted for baseline BMI. Such measurement and analysis details are not reported in the registry record.
7. Why No Re-analysis Is Shown
8. Statistical Interpretation
For the per-protocol analysis, mean BMI change over 12 weeks was 0.8 kg/m2 in the topiramate group, with a 95% confidence interval from -0.33 to 2.33. In the placebo group, mean change was 1.8 kg/m2, with a 95% confidence interval from 1.2 to 11.72. These are separate within-group summaries. The registry does not provide a direct estimate of the difference in mean change between groups, a confidence interval for that difference, or a p-value for the comparison.
The reported group means are not, on their own, an estimate of the treatment contrast. A mean describes the average change among participants in that group; the clinical comparison asks how the average change in one group differs from the average change in the other. Without a reported between-group estimate and its uncertainty, the size and precision of that contrast cannot be read directly from these summaries. The confidence intervals around the group means are not confidence intervals for the difference between groups.
Confidence intervals communicate uncertainty around an estimate. As a general statistical explanation, a 95% confidence interval is produced by a method that, under its assumptions, would cover the target quantity in 95% of repeated samples. It does not mean that 95% of individual participants had changes within the interval, and it is not a range of individual responses. The intervals posted for the two groups are notably broad, particularly the placebo interval. That breadth indicates that the group mean is not pinned down precisely by the posted estimate; the registry does not explain why the interval is so broad.
The interval for the topiramate group extends across zero, while the placebo group’s interval is entirely above zero. This describes the uncertainty around each group’s mean change, not a test of whether the groups differed. A within-group interval’s relation to zero cannot replace a direct randomized comparison. No formal comparative result is reported in the registry record, so the available values do not support a precise statement about the size or statistical evidence for a between-group effect.
The analysis is described as per protocol rather than as an analysis of all randomized participants. Such an analysis can be sensitive to which participants are included, especially if inclusion is related to adherence, tolerability, or outcome. The reasons for the analysis-population selection and the handling of missing Week 12 measurements are not reported in the registry record. Those omissions limit assessment of the robustness of the group summaries.
Overall, the posted results describe BMI change in each analysis group, with wide uncertainty around both means. They do not report a formal between-group estimate, so the magnitude and precision of the comparative result remain unclear. This is a statement about what the registry reports, not a conclusion about treatment efficacy.
9. Statistical Methods Explained
The registry reports a continuous primary outcome: mean change in BMI over 12 weeks. A continuous outcome retains the measured scale of the result, rather than classifying each participant only as having or not having an event. The reported group means describe the average change in each analysis group. The registry gives a 95% confidence interval for each mean, but does not post a direct estimate of the difference in means between groups.
What does a confidence interval describe?
As a general statistical explanation, a confidence interval expresses uncertainty in an estimate under a specified method and its assumptions. A narrower interval indicates greater precision for the estimated quantity; a wider interval indicates less precision. The intervals in this record are intervals for the separate group means. They are not intervals for a treatment contrast and do not describe the spread of individual participants’ BMI changes.
The reported topiramate mean is 0.8 kg/m2, with a 95% confidence interval from -0.33 to 2.33. The placebo mean is 1.8 kg/m2, with a 95% confidence interval from 1.2 to 11.72. The placebo interval is wide, so its mean is estimated with limited precision in the posted summary. The registry does not state which variance estimator or other calculation details were used to construct these intervals.
Why is a direct group contrast important?
In a randomized comparison, the central comparative quantity is generally an estimate of how outcomes differ between assigned groups. Its confidence interval would show uncertainty in that contrast. Separate estimates for the two arms do not supply the same information: their uncertainty may depend on the design and analysis, and the registry gives no contrast estimate from which to assess comparative precision. In general, comparing whether each group’s interval includes zero is not a substitute for estimating and testing the difference between groups.
No p-value, standard error, or confidence interval for the between-group difference is reported in the registry record. The registry lists zero analyses. This does not erase the arm-level descriptive results, but it means that an inferential comparison cannot be reconstructed from the posted information without making assumptions or calculating a new result. Such a reconstruction is not appropriate as a substitute for a sponsor-reported analysis.
How should the planning assumptions be understood?
The record describes planning illustrations based on the observed effect, a two-sided significance level of 0.05, equal allocation, and a normal approximation. It states that the effect to detect is a stated fraction of the observed effect. These are planning assumptions, not a report of the trial’s original sample-size justification. The protocol’s sample-size rationale is not reported in the registry record. Any sample-size value derived from an observed effect is a planning illustration and should not be called observed power. Observed power is not a substitute for examining the estimated effect and its uncertainty.
As a general explanation, planning calculations depend on choices about the target effect, variability, significance level, allocation, and analysis. Changing those assumptions can change the planned sample size. The registry does not report the trial’s original design rationale or all details needed to judge whether the pilot was intended to provide a definitive comparative test.
What does the design information establish?
The registry identifies the study as randomized, triple-masked, placebo-controlled, and factorial. Randomization is intended to make group assignment independent of known and unknown prognostic factors on average, while masking can reduce some forms of biased treatment or outcome assessment. These are general design principles. The record does not explain the randomization procedure, concealment, masking roles, or factorial structure, so their implementation cannot be assessed from the registry entry alone.
The reported primary analysis is per protocol, with group sizes of 16 and 14. A per-protocol analysis focuses on participants meeting a protocol-defined set of conditions, but the registry does not specify those conditions or give reasons for exclusion. It also does not state whether another analysis population was used. The number of randomized participants in each group, the handling of missing observations, and the statistical model used for the primary endpoint are not reported in the registry record.
For continuous outcomes, variation among participants affects the precision of a mean and of a group contrast. As a general explanation, over-dispersion means that observed variation is greater than a model assumes; if unaccounted for, this can make uncertainty estimates unreliable. The record does not report a dispersion statistic or enough modeling detail to assess this issue. The posted confidence intervals remain the registry’s stated uncertainty summaries, but their calculation cannot be independently evaluated from the listed methods.
10. Multiplicity, Missing Data and Analysis Populations
The registry lists three outcome measures and one posted secondary outcome, tolerability of topiramate. It does not describe all outcome measures in the available record, or explain how the primary endpoint was selected from them. The multiplicity strategy—how multiple outcomes or statistical comparisons were handled—is not reported in the registry record. It is therefore not possible to determine whether an adjustment was planned or applied, or how the reported results relate to any broader family of analyses.
Missing-data methods are also not reported. The record does not state how missing BMI measurements were handled, whether any imputation was used, or whether participants with incomplete follow-up were included in other analyses. This matters because the primary analysis is described as per protocol. Without the criteria for that population and an account of missing observations, readers cannot assess whether the analyzed participants represent all randomized participants or whether missingness could have affected the group means.
The registry does not report an intention-to-treat analysis or another analysis population alongside the per-protocol results. It also does not describe sensitivity analyses that might test the effect of alternative missing-data assumptions. The reasons for any difference between enrollment and the reported analysis groups are not reported in the registry record.
The record lists zero analyses, while also posting outcome summaries. It does not provide inferential results for the primary group comparison or describe an analysis plan. Details on interim analyses, stopping rules, and prespecified multiplicity control are not reported in the registry record. Consequently, the posted summaries can be described, but the full analytic pathway and the handling of selection, missingness, and multiple comparisons cannot be evaluated.
11. Safety Results
The registry reports serious adverse events as affected participants divided by participants at risk.
| Arm | Participants with serious adverse events | Participants at risk | Percentage |
|---|---|---|---|
| Experimental Group: Topiramate Group | 2 | 16 | 12.5% |
| Comparator Group: Placebo Group | 1 | 14 | 7.1% |
Deaths
| Arm | Deaths (participants affected of at risk) |
|---|---|
| Experimental Group: Topiramate Group | 0 of 16 |
| Comparator Group: Placebo Group | 0 of 14 |
Other (non-serious) adverse events
Participants with at least one non-serious event at or above the registry's 5% reporting threshold.
| Arm | Participants affected of at risk |
|---|---|
| Experimental Group: Topiramate Group | 15 of 16 |
| Comparator Group: Placebo Group | 13 of 14 |
Most frequent serious adverse event terms
| Event term | Organ system | Experimental Group: Topiramate Group | Comparator Group: Placebo Group |
|---|---|---|---|
| Malignant hyperthermia | Nervous system disorders | 1 of 16 | 0 of 14 |
| Increased aggression | Psychiatric disorders | 0 of 16 | 1 of 14 |
| Suicide attempt | Psychiatric disorders | 1 of 16 | 0 of 14 |
Most frequent non-serious adverse event terms
| Event term | Organ system | Experimental Group: Topiramate Group | Comparator Group: Placebo Group |
|---|---|---|---|
| Gastrointestinal | Gastrointestinal disorders | 6 of 16 | 5 of 14 |
| Headaches | General disorders | 6 of 16 | 5 of 14 |
| Rhinitis | Respiratory, thoracic and mediastinal disorders | 3 of 16 | 3 of 14 |
| Muscle aches | Musculoskeletal and connective tissue disorders | 3 of 16 | 1 of 14 |
| Back and/or neck pain | General disorders | 3 of 16 | 1 of 14 |
| Ocular/visual blurriness | Eye disorders | 3 of 16 | 1 of 14 |
| Chest pain | General disorders | 0 of 16 | 3 of 14 |
| Worsening of mood | Psychiatric disorders | 2 of 16 | 1 of 14 |
| Sedation | Psychiatric disorders | 0 of 16 | 3 of 14 |
| Earache | Ear and labyrinth disorders | 0 of 16 | 2 of 14 |
Serious events with the highest arm-level rate: increased aggression (7.1%), malignant hyperthermia (6.2%), suicide attempt (6.2%).
The registry reports serious adverse events for both groups. In the topiramate group, 2 of 16 participants are listed as affected by serious adverse events; in the placebo group, 1 of 14 is listed. Deaths are reported as 0 in each group. These counts should be read as the posted group-level safety figures, not as evidence about the cause of an event or a comparison adjusted for follow-up or exposure.
The registry also reports other adverse events affecting 15 of 16 participants in the topiramate group and 13 of 14 in the placebo group. The definitions, severity thresholds, observation period, and method for collecting these events are not reported in the registry record. The record does not say whether participants could experience more than one event, so affected-participant counts should not be interpreted as total event counts.
Three serious-event terms are listed with percentages: Increased aggression, 7.1%; Malignant hyperthermia, 6.2%; and Suicide attempt, 6.2%. The group attribution and denominator for these term-level percentages are not reported in the registry record. The terms should therefore not be assigned to a treatment group or treated as comparative rates. The registry does not establish that any listed event was caused by a study intervention.
12. What the Result Does, and Does Not, Mean
The result does show the group-level BMI changes posted for the per-protocol population over 12 weeks: 0.8 kg/m2 for topiramate and 1.8 kg/m2 for placebo, with the respective 95% confidence intervals. These summaries are descriptive estimates, and the intervals convey substantial uncertainty, especially for the placebo mean. The registry does not report the direct contrast needed to quantify how much the groups differed or how precisely that difference was estimated.
The result does not establish a treatment effect from the separate arm means alone. Nor does a confidence interval for one arm answer whether its outcome differed from the other arm. A direct group comparison, along with its uncertainty and analysis details, would be needed to interpret the magnitude and precision of the comparative result. Those details are not reported in the registry record.
The study is described as a pilot, and the registry reports enrollment of 31. The analysis is per protocol, with groups of 16 and 14. These features are relevant when considering how much information the posted estimates contain, but the registry does not provide the original sample-size justification or explain the selection of the analysis population. The available information cannot establish how representative the analyzed participants were of all enrolled participants.
The result also does not establish that BMI fully captures weight-related health or participant experience. The registry describes tolerability as a secondary objective but provides limited detail about its assessment. Safety counts give a partial account of reported events, but do not provide full event histories, exposure-adjusted rates, or causal attribution.
In short, the record supports a cautious description of the observed arm-level summaries, not a definitive conclusion about comparative efficacy or safety. The absence of a posted between-group estimate, missing-data explanation, and detailed analysis methods leaves important uncertainty about interpretation. These limitations concern the information reported, not proof that a treatment has or lacks an effect.
13. Why This Trial Matters Statistically
This trial is statistically informative as an example of the difference between reporting outcomes within each arm and reporting the comparison that a randomized trial is designed to estimate. The registry gives separate mean BMI changes and confidence intervals. Readers may be tempted to infer a comparative result by looking at the two means or by checking whether each interval includes zero, but neither approach supplies a direct estimate of the difference between groups.
The broad intervals are also a reminder that a point estimate is not the whole result. A mean gives a compact description of the observed group, while its interval communicates uncertainty in that estimate. For the placebo group, the posted interval extends from 1.2 to 11.72 kg/m2. The record does not explain the source of this width, so it cannot be attributed confidently to a particular feature such as variability, data completeness, or method of calculation.
The per-protocol analysis raises a further statistical question: who was included, and how might inclusion depend on treatment experience or outcome availability? The registry does not specify the criteria, missing-data approach, or additional analysis populations. This makes the analytic population difficult to evaluate against the randomized study population.
The study also demonstrates why transparency about planning and analysis matters. Its design is labelled factorial, but the registry does not describe the factorial structure or the statistical model. A planning illustration based on an observed effect does not substitute for an original sample-size justification, and it is not evidence of observed power. Better documented analysis methods would help readers distinguish planned inference from exploratory description.
For clinical researchers and biostatisticians, the central lesson is to match interpretation to the estimand and uncertainty actually reported. The group summaries are useful descriptive information, but the registry does not provide the direct comparative analysis needed to judge the estimated difference precisely.
14. 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 provides a limited account of the study’s methods. Although it identifies randomization, triple masking, and a factorial model, it does not describe the allocation process, masking roles, or factorial structure. The criteria and procedures supporting those design labels are not reported in the registry record, limiting assessment of how the design was implemented.
The primary result is reported for a per-protocol population rather than a clearly described analysis of all randomized participants. The record does not give per-protocol criteria, participant flow, reasons for exclusion, or an intention-to-treat result. It does not explain how missing Week 12 BMI data were handled. These omissions make it difficult to assess potential selection effects and the representativeness of the reported groups.
Only group-level means and confidence intervals are posted for the primary endpoint. A direct estimate of the between-group difference, its confidence interval, and a p-value are not reported in the registry record. The statistical model, variance assumptions, and method used to construct the intervals are also not described. The unusually broad placebo interval cannot be explained from the available methodological detail.
The registry identifies a pilot study and lists enrollment of 31, but the original sample-size rationale is not reported. The full set of outcome definitions and the plan for handling multiple outcomes are not described. Interim analyses, stopping rules, treatment adherence details, and dose exposure by participant are not reported in the registry record.
Safety reporting is also incomplete for interpretation: event-collection methods, observation time, term-level denominators, and causal assessments are not specified. The listed adverse-event counts are useful, but do not allow a complete comparative safety analysis. These limitations concern what the registry reports and do not establish that unreported procedures or analyses did not occur.
15. Related Tutorials
16. Related Statistical Calculators
17. Sources
- ClinicalTrials.gov: Topiramate vs. Placebo in Preventing Weight Gain in Bipolar Disorder Treated With Olanzapine — NCT00394095.
- Registry record last updated 2025-04-09; status last verified 2025-03.
- Results first posted 2013-04-04.
- 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.
Continue through the Clinical Biostats statistical pathway
Use this trial's endpoint as a starting point for deeper study of effect measures, confidence intervals, and sample-size planning.
18. Record Summary
The registry describes a completed Phase 4, randomized, triple-masked, placebo-controlled pilot study of topiramate versus placebo in combination with olanzapine for weight-gain prevention in manic or mixed youth with bipolar disorder. Enrollment was 31. The primary endpoint was mean change in BMI from baseline to Week 12, in kg/m2, analyzed per protocol.
The topiramate group had a mean change of 0.8, with a 95% confidence interval from -0.33 to 2.33. The placebo group had a mean change of 1.8, with a 95% confidence interval from 1.2 to 11.72. No direct between-group estimate or inferential comparison is reported in the registry record.
The registry lists tolerability of topiramate as a secondary outcome and posts serious and other adverse-event counts. Detailed methods for missing data, analysis populations, multiplicity, and the original sample-size rationale are not reported. Consequently, the record supports description of the posted group summaries, while leaving the comparative effect and its precision unclear.