Levetiracetam in the Management of Bipolar Depression: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Levetiracetam in the Management of Bipolar Depression (NCT00566150), a randomized na study. The primary outcome is Change in Hamilton Depression Rating Scale (HDRS-21) Total Score From Baseline at Week 6..

2. Clinical Question

The trial’s stated purpose was to investigate levetiracetam’s effects on depressive symptoms in bipolar depression. The primary measure was change in the 21-item Hamilton Depression Rating Scale total score from baseline to week 6. The record defines change as the observed score at each visit minus the baseline score.

Additional listed outcomes were change in the Montgomery-Asberg Depression Rating Scale total score at week 6, remission, and change in the Clinical Global Impressions Scale for Bipolar Disorder depression-severity rating at week 6. These outcomes describe other aspects of depressive symptoms and clinical assessment, but the registry record does not report their comparative results either. As a result, the clinical question can be described from the registered objectives, but the posted results do not establish the size or direction of any treatment-group difference.

3. Trial Design

This was a randomized, parallel-group trial with levetiracetam and placebo as the interventions. The registry lists quadruple masking and describes the study as double-blind and placebo-controlled. It was an outpatient add-on trial lasting six weeks. The record does not specify the background treatments or provide further detail about how randomization and masking were implemented.

Enrollment was 35. The study is listed as completed, with a start date of 2005-10 and a primary completion date of 2008-08. The phase is listed as NA. Yale University is identified as the sponsor. The record does not state a planned or achieved allocation ratio; the safety groups list 19 participants at risk in the levetiracetam group and 16 in the placebo group, but those safety denominators should not be treated as a complete account of every randomized participant’s follow-up.

The primary outcome was a continuous score: change in Hamilton Depression Rating Scale total score from baseline to week 6. The registry describes the reported statistic as a least-squares mean with a standard error. However, it does not provide a treatment-group estimate, a treatment contrast, or an inferential result for this endpoint. It notes that the primary outcome has several categories or time points and was not reduced to one comparison in the posted results.

The trial record lists four outcome measures and four analyses. Its description is enough to identify the broad design and the outcome domains, but not to reconstruct the full protocol or statistical analysis plan. Details such as eligibility criteria, treatment dose, adherence procedures, allocation concealment, and prespecified analysis choices 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.

Overall Study

Analysis population for the primary outcome, as defined in the registry: Weeks 1-6 are Intent to treat (ITT) population Observed Cases. All participants included received at least 1 dose of study intervention and at least 1 assessment post-baseline.

5. Baseline Characteristics

The table reproduces the baseline characteristics posted in the registry, which describe the participants analysed at baseline and help show whether the randomized groups were comparable.

6. Primary Endpoint

The registry lists the primary outcome as Change in Hamilton Depression Rating Scale (HDRS-21) Total Score From Baseline at Week 6., measured over Baseline to week 6.

The primary endpoint is change in the Hamilton Depression Rating Scale total score from baseline to week 6. The scale has 21 items and is described as measuring depression. Each item is rated from 0, meaning symptoms are not present, to a maximum of 2 to 4, meaning symptoms are extremely severe. The total score ranges from 0 to 60.

The registry defines change as the observed value at a visit minus the baseline value. A change score therefore expresses how the measured symptom score differs from the participant’s starting score; it is not the week-6 score alone. Because higher item ratings correspond to more severe symptoms, the direction of a change score matters when interpreting the measure. The registry record does not report the arm-specific changes or a between-group contrast, so the observed direction and magnitude cannot be described from the posted results.

A continuous symptom score can capture degrees of change rather than reducing the outcome to a yes-or-no classification. The least-squares mean is an adjusted mean estimated by a statistical model, while its standard error describes the precision of that estimate under the model. The model details and the estimates themselves are not reported in the registry record, so the adjustment and the precision of the primary endpoint cannot be assessed. The scale is the registered measure of depressive symptoms; it is not, by itself, a complete account of participants’ experiences or of every outcome relevant to the trial.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The statistical details that can be discussed are limited to the outcome description and the analysis-population wording in the registry record. The record names a least-squares mean and standard error for the primary continuous outcome, but it does not give the model specification, group estimates, contrast, or test result. The explanations below describe the general meaning of these methods; they do not reconstruct the unreported analysis.

What does a least-squares mean represent?

In general, a least-squares mean is a model-based estimate of an average. Unlike a raw arithmetic mean, it can reflect adjustment for variables included in a statistical model. Whether that adjustment is appropriate depends on the model, the variables included, and the assumptions used. The registry record does not identify the covariates, model structure, or estimation method beyond naming a least-squares mean. It is therefore not possible to determine which adjustments were made or how much they influence the estimate.

For a randomized comparison, an estimated mean change in each group is not always the same quantity as the treatment contrast. The contrast is the estimated difference between groups on the specified outcome scale. To interpret the comparative effect, readers would need the group estimates or the contrast, together with a measure of uncertainty. The registry record does not report those values for the primary endpoint. No direction or size of the treatment difference can be inferred from the endpoint definition alone.

What does the standard error tell a reader?

In general, a standard error describes the sampling variability of an estimate under the analysis model. Smaller standard errors indicate greater precision for an estimate, while larger standard errors indicate less precision. A standard error is not a description of the spread of individual participants’ scores; it concerns uncertainty in an estimated quantity. The registry identifies standard error as the dispersion measure but does not report its value for the primary outcome.

A confidence interval is generally constructed from an estimate and its uncertainty and gives a range of values compatible with the data under specified assumptions. A test statistic and p-value, when reported, address a related but different question about compatibility with a null hypothesis. No primary-outcome confidence interval or p-value is reported in the registry record. Consequently, neither the practical scale of the comparative effect nor the uncertainty surrounding it can be evaluated from the registry results.

Why does the outcome definition matter?

The outcome is a change score, defined as the observed value at each visit minus baseline. The interpretation of a change score depends on its sign and scale. A negative change indicates a lower observed score than baseline; a positive change indicates a higher score. Since the Hamilton scale’s item ratings run from symptoms not present toward greater severity, the sign gives direction on the symptom scale. But the registry record does not report the observed change estimates, so this directional framework cannot be used to characterize the trial’s findings.

Change scores also depend on the baseline measurement and on which follow-up observations are included. The record describes an observed-cases population across weeks 1-6, but does not state how repeated observations were modeled or how baseline values entered the model. The registry does not give visit-level counts, missing-data procedures, or assumptions about incomplete follow-up. These omissions prevent an independent assessment of the analysis structure.

What can be inferred from the planning assumptions?

The registry’s planning illustration is described as based on the observed effect, with a two-sided significance level of 0.05, equal allocation between the arms, and a normal approximation. The effect to detect is described as a stated fraction of the observed effect. These are assumptions for planning illustrations, not evidence that the trial had a particular probability of detecting an effect. Such illustrations do not constitute observed power and do not establish the precision of the trial’s result.

Because a primary-outcome effect estimate is not reported in the registry record, the observed effect underlying a planning illustration cannot be checked against the posted primary results. The record also does not report the trial’s own sample-size justification. It is not possible to conclude that the study was planned around a particular effect, or to judge whether its design was adequate for that effect, from the available statistical description.

In general, over-dispersion matters when observed variability exceeds what a model assumes; standard errors can then be too small if the model does not account for that extra variability. This is a general statistical consideration, not a finding about this trial. The registry record does not report a dispersion assessment or information that would allow one to evaluate it.

10. Multiplicity, Missing Data and Analysis Populations

The registry lists a primary endpoint and secondary outcomes covering change in Montgomery-Asberg Depression Rating Scale total score at week 6, remission, and change in Clinical Global Impressions Scale for Bipolar Disorder depression-severity rating at week 6. It also reports four outcome measures and four analyses. The primary outcome is noted as having several categories or time points and not being reduced to one comparison. These details indicate that more than one outcome or analysis was registered, but the registry record does not state how multiplicity was addressed.

In general, examining multiple outcomes or time points can increase the chance that at least one result appears noteworthy by chance. Statistical plans may address this through prespecified outcome hierarchies, adjusted significance thresholds, or other approaches. Which method, if any, was planned or used in this trial is not reported in the registry record. No multiplicity adjustment should be assumed from the outcome list alone.

The primary population is described as ITT observed cases during weeks 1-6. It includes participants who received at least one dose and had at least one assessment after baseline. This definition identifies a population with some post-baseline information, but it does not give the number included at each visit or establish how incomplete records were handled. In general, an observed-cases analysis uses available observations rather than supplying values for unobserved outcomes. The registry does not report whether any imputation or other missing-data method was used, the amount or pattern of missing data, or the assumptions underlying its handling.

The record does not describe interim analyses, stopping rules, a detailed analysis plan, or sensitivity analyses. Each of these elements is not reported in the registry record. Without the analysis plan and complete outcome reporting, readers cannot determine how the registered outcome family was handled or whether conclusions were tested under alternative assumptions about missing observations.

11. 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 non-serious adverse event terms

The registry reports serious adverse event counts separately for levetiracetam and placebo. In the levetiracetam group, 0 participants were reported as affected among 19 at risk. In the placebo group, 0 participants were reported as affected among 16 at risk. These are counts within the listed safety groups; they do not describe the primary symptom outcome or establish a comparative safety effect.

The record lists other adverse events affecting 17 participants among 19 at risk in the levetiracetam group and 15 participants among 16 at risk in the placebo group. The registry does not provide event terms in the serious-event listing. These counts should not be read as counts of serious events: the serious adverse event counts are reported separately. The record does not provide enough detail to characterize the types, timing, duration, or relationship of the other adverse events to study intervention.

Deaths are not reported in the registry record. A count of 0 affected participants for serious adverse events is not equivalent to a detailed account of every possible safety outcome, and the listed denominators do not establish complete follow-up for all randomized participants. These safety figures are descriptive; they do not by themselves support a conclusion about relative risk or causation.

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

The registry record contributes useful information about what was planned and how the study was organized. It identifies a randomized, parallel, placebo-controlled add-on trial lasting six weeks, and names change in Hamilton Depression Rating Scale total score from baseline to week 6 as the primary endpoint. It also records secondary measures and provides limited safety counts. These details help define the clinical question and the measures the trial intended to assess.

However, the record does not report the primary comparative result. Without group-specific estimates or a treatment contrast, the data posted in the registry do not show whether scores changed differently between the levetiracetam and placebo groups, how large any difference was, or how uncertain it was. The same limitation applies to the listed secondary outcomes: their names and measurement times are present, but comparative outcome results are not reported in the registry record.

This absence should not be interpreted as evidence of no effect, and it should not be interpreted as evidence of benefit or harm. Completion status and enrollment are not substitutes for analysis results. The safety counts likewise provide only a partial descriptive view and cannot establish that an intervention caused or prevented an event.

General statistical interpretation depends on estimates, uncertainty, the analysis population, and the assumptions used. The registry does not report enough detail to assess the model behind the least-squares mean, the precision of the primary comparison, or how missing observations and multiple outcomes were handled. Accordingly, the record can support a description of the trial’s design and registered questions, but not a conclusion about the comparative clinical outcome.

13. Why This Trial Matters Statistically

This trial is statistically informative in part because it illustrates the distinction between a registered endpoint and a reported result. A defined primary measure does not, on its own, tell readers what the groups experienced or how different their outcomes were. For an interpretable comparison, the estimate and its uncertainty need to be available alongside the endpoint definition and analysis population.

The trial also highlights why analysis-population labels require careful reading. The registry calls the primary population ITT observed cases and specifies that participants had to receive at least one dose and have at least one post-baseline assessment. That wording gives a useful boundary for the reported analysis population, but it does not reveal how many participants contributed to the week-6 result or how missing observations were treated.

Multiple outcomes and time points raise another important statistical issue. The registry lists several outcome domains and notes that the primary outcome was not reduced to one comparison. Without a stated multiplicity strategy and complete comparative results, readers cannot assess how the family of analyses was interpreted. This is a general concern when multiple statistical questions are considered; the record does not establish which approach was used in this study.

The practical lesson is not a conclusion about levetiracetam. It is that a completed trial’s statistical record is most useful when it pairs clear outcome definitions with complete estimates, uncertainty measures, analysis-population counts, and an account of missing-data and multiplicity decisions. Those elements are not reported in the registry record for the primary comparison, which limits what can be learned from this record alone.

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 main limitation for interpretation is that the registry record does not report a comparative primary-outcome estimate. It names the measure and the intended time frame, but provides no group-specific changes, between-group contrast, uncertainty interval, or p-value. The note that the primary outcome has several categories or time points and was not reduced to one comparison further limits the ability to identify a single primary result.

Other limitations concern the analysis description. The record identifies a least-squares mean and standard error but does not specify the model, covariates, or repeated-measures structure. It does not report the number of participants contributing to each outcome visit, the extent or pattern of missing data, or the methods used to handle missing observations. The stated ITT observed-cases population includes participants with at least one dose and at least one post-baseline assessment, but does not resolve those questions.

The registry lists secondary outcomes, but their comparative results are not reported in the registry record. A multiplicity strategy, interim analyses, and sensitivity analyses are also not reported in the registry record. The record does not state the study’s own sample-size justification, so the enrollment figure alone cannot show what effect size the design was intended to detect.

Clinical and operational details are limited as well. The registry does not report the intervention regimen, background treatment details, full eligibility criteria, or further implementation details for allocation and masking. The safety record gives affected-participant counts, but does not report deaths or detailed event terms in the registry record. These gaps constrain the assessment of both outcome findings and study conduct from this record alone.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: Levetiracetam in the Management of Bipolar Depression — NCT00566150.
• Registry record last updated 2020-03-31; status last verified 2016-08.
• Results first posted 2015-11-02.
• 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.

18. Record Summary

NCT00566150 is a completed, randomized, parallel-group, quadruple-masked trial sponsored by Yale University. It studied levetiracetam versus placebo as a six-week outpatient add-on trial in bipolar depression. Enrollment was 35; the record lists a start date of 2005-10 and primary completion date of 2008-08. The phase is listed as NA.

The primary measure was change from baseline to week 6 in Hamilton Depression Rating Scale total score, described as a least-squares mean with standard error. The outcome population is identified as ITT observed cases, including participants with at least one dose and at least one post-baseline assessment. Additional registered outcomes include Montgomery-Asberg Depression Rating Scale change, remission, and Clinical Global Impressions Scale for Bipolar Disorder depression-severity change.

The registry record does not report a primary comparative estimate or comparative results for the listed secondary outcomes. Safety counts report 0 serious adverse event-affected participants in each group; deaths are not reported in the registry record.