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 A Six-Week Study Evaluating The Efficacy And Safety Of Geodon In Patients With A Diagnosis Of Bipolar I Depression (NCT00141271), a randomized phase 3 study. The primary outcome is Change in Montgomery-Asberg Depression Rating Scale (MADRS)Total Score.
| Feature | Registry value |
|---|---|
| ClinicalTrials.gov identifier | NCT00141271 |
| Phase | Phase 3 |
| Condition | Bipolar Disorder |
| Allocation | Randomized |
| Design model | Parallel |
| Masking | Double |
| Enrollment | 536 |
| Arms in the primary outcome | n/a |
| Lead sponsor | Pfizer's Upjohn has merged with Mylan to form Viatris Inc. |
| Results posted | Yes |
2. Clinical Question
Population
Outpatients aged 18 and older with bipolar I depression. The brief summary says participants had bipolar disorder type I, depressed, and were required to undergo a washout of at least 7 days of any prior medication.
Intervention
Oral ziprasidone, also identified by the brand name Geodon. The record describes a fixed-flexible dose study and lists ziprasidone dose groups in the adverse-event information as 120-160mg and 40-80 mg.
Comparator
Placebo. The official title describes the study as placebo-controlled. Arm descriptions for the dose groups and placebo are not reported in the registry record.
Primary question
The primary endpoint concerns change in Montgomery-Asberg Depression Rating Scale (MADRS) total score from baseline to 6 weeks. The registry record does not report the arm-specific primary estimates or a treatment comparison, so the size and precision of any difference cannot be assessed from that record.
The clinical question is whether the average change in depressive symptom scores over the study period differed between the ziprasidone groups and placebo. The endpoint description defines change as the observed score at each visit minus the baseline score, with the overall value described as the average response across Weeks 1 - 6. A comparison would need to report the estimated changes for each group, the contrast between groups, and uncertainty around that contrast. Those primary results are not reported in the registry record.
The registry identifies an intent-to-treat population of observed cases for the primary endpoint. That wording combines a population label with an observed-case qualification: the analysis is described as ITT, but the endpoint uses observations available for analysis. How missing observations were handled for this primary analysis is not reported in the registry record. The listed secondary outcomes include response and remission definitions based on MADRS and HAM-D 17 scores, as well as functioning, anxiety, and mania measures; their results are not reported in the registry record.
3. Trial Design
The registry describes a Phase 3, randomized, double-blind, multicenter, parallel-group, placebo-controlled study of oral ziprasidone in outpatients with bipolar I depression. Its official title specifies fixed-flexible dosing and a six-week study period. The trial status is completed. The registry lists enrollment as 536; this is the recorded enrollment figure, not a report of how many participants contributed data to any particular endpoint.
The brief summary identifies participants as outpatients aged 18 and older and states that a washout period of at least 7 days of any prior medication was required. Further eligibility criteria, details of recruitment, and the washout procedures are not reported in the registry record. The intervention list contains ziprasidone and placebo, while the adverse-event groups are labelled 120-160mg Ziprasidone, 40-80 mg Ziprasidone, and Placebo. Descriptions of the arms are blank, so additional dose or treatment-administration details are not reported in the registry record.
Randomization and double masking are stated design features. The registry does not report the randomization sequence method, allocation concealment procedures, or which study personnel were masked. It also does not give details of the multicenter sites. These omissions do not establish that such procedures were absent; they mean that they cannot be evaluated from the registry record.
The primary endpoint is change in MADRS total score from baseline to 6 weeks. The stated analysis population is the intent-to-treat population observed cases, and the reported summary type is a least-squares mean with standard error. The registry does not report the statistical model specification, covariates, treatment contrasts, or primary outcome results. Start is recorded as 2005-07 and primary completion as 2008-02. The sponsor field names Pfizer's Upjohn, which it says merged with Mylan to form Viatris Inc.
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: 7-day washout:psychotropic drugs \& lithium. 4-week washout: monoamine oxidase inhibitors, fluoxetine alone or in combination with olanzapine. Depot neuroleptic Discontinued (DC'd) 6 mo prior to entry. 536 subjects enrolled: 504 assigned to drug; 32 not assigned:18 lost to follow up, 5 DC'd study, 9 didn't take/receive study drug for other reasons.
Overall Study
| Stage | 120-160mg Ziprasidone | 40-80 mg Ziprasidone | Placebo |
|---|---|---|---|
| STARTED | 171 | 165 | 168 |
| COMPLETED | 91 | 102 | 111 |
| NOT COMPLETED | 80 | 63 | 57 |
Reasons for not completing this period:
| Reason | 120-160mg Ziprasidone | 40-80 mg Ziprasidone | Placebo |
|---|---|---|---|
| Adverse Event | 31 | 19 | 17 |
| Laboratory Abnormality | 2 | 1 | 1 |
| Lack of Efficacy | 7 | 7 | 3 |
| Lost to Follow-up | 15 | 15 | 21 |
| Withdrawal by Subject | 14 | 9 | 10 |
| Other | 11 | 12 | 5 |
Analysis population for the primary outcome, as defined in the registry: Intent to treat (ITT) population observed cases
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 | 120-160mg Ziprasidone | 40-80 mg Ziprasidone | Placebo | Total |
|---|---|---|---|---|---|
| Participants analysed | 171 | 165 | 168 | 504 | |
| Age, Continuous | years | 40.4 (SD 12.9) | 40.0 (SD 11.4) | 39.3 (SD 11.2) | 39.9 (SD 11.8) |
| Sex: Female, Male: Female | Participants | 91 | 101 | 92 | 284 |
| Sex: Female, Male: Male | Participants | 80 | 64 | 76 | 220 |
6. Primary Endpoint
The registry lists the primary outcome as Change in Montgomery-Asberg Depression Rating Scale (MADRS)Total Score, measured over Baseline to 6 weeks.
| Element | Registry entry |
|---|---|
| Result type | Least Squares Mean |
| Measure of spread | Standard Error |
| Unit | score on scale |
| Endpoint classification | Continuous Endpoint |
The primary endpoint is change in total score on the Montgomery-Asberg Depression Rating Scale (MADRS) from baseline to 6 weeks. The registry defines change as the observed value at each visit minus baseline. It describes the overall measure as the average response across Weeks 1 - 6. The scale is a 10-item instrument for measuring depression; each item ranges from 0, described as normal, to 6, described as most abnormal, and the total possible score is 0 - 60.
As a continuous measure, MADRS total score retains information about the degree of score change rather than reducing the outcome to whether a participant crossed a response threshold. This can make it useful for comparing average symptom-score trajectories between randomized groups. That is a general statistical rationale, not a finding about this trial. The registry's chosen summary is a least-squares mean, which generally represents a model-based group mean; the actual model details are not reported in the registry record. A standard error describes the estimated precision of a mean, not the spread of individual participant scores.
The endpoint description does not give arm-level changes, a between-group contrast, confidence intervals, or a test result. Consequently, the direction, size, and precision of the primary comparison are not reported in the registry record. The stated scale range and endpoint definition alone cannot establish what the trial observed.
7. Why No Re-analysis Is Shown
8. Statistical Interpretation
The registry record identifies the primary outcome and how it was intended to be summarized, but it does not report the primary MADRS results. There are no arm-specific least-squares mean changes, between-group contrasts, standard errors, confidence intervals, or p-values in the record. Without those estimates, the observed direction and magnitude of the primary comparison cannot be described, and its statistical precision cannot be assessed.
The endpoint is a continuous score change from baseline to 6 weeks, with an overall measure described as the average response across Weeks 1 - 6. A comparison of group means would address whether average score changes differed under the analysis method used. It would not, by itself, show how individual participants responded or how score changes were distributed. General statistical interpretation requires both an estimate and its uncertainty: the point estimate conveys the observed contrast, while an interval indicates the range of values compatible with the data and analysis assumptions.
The registry calls the primary population the intent-to-treat population observed cases. An ITT label generally signals analysis according to randomized assignment, while an observed-case qualification indicates that analyzed values depend on available observations. The registry does not explain exactly how these descriptions were implemented, how many outcome observations were missing, or whether missingness differed by group. These details matter because results based on observed data can be affected if the availability of measurements is related to outcomes or treatment assignment. This is a general concern, not evidence that such bias occurred in this study.
The record lists response and remission outcomes based on MADRS and HAM-D 17, plus measures of functioning, anxiety, and mania. It also records 38 outcome measures and 196 analyses. Those counts indicate a broad set of recorded outcomes and analyses, but do not establish which results were prespecified, which were confirmatory, or how multiplicity was handled. The registry does not report an analysis plan or multiplicity adjustment.
Adverse-event information lists serious-event counts of 7, 3, and 6 for the labelled ziprasidone and placebo groups, and other-event counts of 89, 70, and 57. The at-risk denominators and death counts are not reported in the registry record. The record also lists serious-event terms with 0.6% entries, but without denominators these entries cannot be converted into group-specific risks or compared reliably. These safety counts do not supply the missing primary efficacy estimates.
9. Statistical Methods Explained
What does a least-squares mean represent?
The primary outcome is summarized as a least-squares mean. In general, a least-squares mean is an estimated group mean from a statistical model, rather than necessarily the unadjusted arithmetic average of observed scores. Depending on the model, it can account for specified design features or covariates. The registry does not report which model was used, whether covariates were included, or how repeated measurements were handled. The label alone is therefore insufficient to reconstruct the estimate or identify what adjustment it represents.
The endpoint is defined as observed score at a visit minus baseline, with an overall value described as the average response across Weeks 1 - 6. This describes the target summary but does not clarify whether all visits were modeled jointly, whether visit-specific estimates were combined, or how incomplete visit histories were treated. Those analytic details are not reported in the registry record.
How should a standard error be read?
The registry identifies the dispersion measure as a standard error. In general, a standard error quantifies the sampling uncertainty of an estimated mean. It is not the standard deviation of participant scores and does not directly describe how variable outcomes were among participants. A smaller standard error generally means an estimate is more precise under the model and its assumptions, while a larger one indicates greater uncertainty. No primary standard errors are reported in the registry record, so precision cannot be judged from the endpoint description alone.
Confidence intervals are a common way to show uncertainty around an estimate or treatment contrast. In general, a confidence interval gives a range of estimates compatible with the data under the statistical procedure used. Its interpretation depends on the model and assumptions. The registry record does not report a primary confidence interval, nor does it state the procedure used to calculate one.
Why is the treatment contrast important?
A group mean change is not the same as a treatment effect estimate. To compare randomized groups, an analysis ordinarily reports a contrast between groups, with a defined direction, and uncertainty around that contrast. Looking at separate group estimates without the contrast can be misleading because the uncertainty of the difference depends on both estimates and their relationship. The registry record does not report a primary contrast, so no between-group interpretation can be made from the primary endpoint information.
Randomization supports a comparison of assigned groups, in general, by reducing systematic differences at baseline on average. It does not guarantee that every observed characteristic is balanced, and it does not resolve missing-data or analysis-specification issues automatically. The record states that allocation was randomized and masking was double, but does not describe the randomization sequence, concealment, or masking procedures.
What does observed-case analysis imply?
The primary population is labelled intent to treat (ITT) observed cases. In general, ITT analysis aims to preserve participants in their randomized groups, regardless of treatment received after assignment. An observed-case qualification means that outcome measurements not observed may not contribute to a particular estimate. The registry does not explain whether participants with some observed visits contributed those visits, what rules determined inclusion, or what was done with missing endpoint values. It also does not report the amount or pattern of missing data.
Missingness can matter if the probability that an outcome is observed is related to the unobserved outcome, treatment group, or participant characteristics. Statistical methods can make different assumptions about such missing data, and conclusions may depend on those assumptions. This is a general explanation; the registry record does not describe a missing-data model or sensitivity analysis for the primary endpoint. A secondary GAF outcome specifies last observation carried forward (LOCF), but that does not establish that LOCF was used for the primary MADRS endpoint or for every secondary outcome.
How should planning calculations be understood?
Sample-size calculations based on an observed effect are planning illustrations, not observed power. The stated assumptions for planning are a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation; the targeted effect is expressed as a stated fraction of the observed effect. Such illustrations address how many participants per arm might be needed under those assumptions. They do not tell whether the completed trial had adequate power, and they do not replace the observed treatment estimate or its confidence interval. The registry record does not report the trial's own sample-size justification, so the planned design rationale cannot be assessed.
For a continuous outcome, a planning calculation depends on the chosen effect size and variability assumptions, as well as the statistical test and design. If an effect is estimated from the same trial, using it as an input is an illustrative exercise rather than an independent confirmation of that effect. No recalculated primary effect is reported in the registry record. The assumptions do not reveal whether the trial's actual analysis matched the normal approximation or equal-allocation conditions.
Why can multiple outcomes complicate interpretation?
Multiplicity arises when many hypotheses are tested or many outcomes are examined. In general, examining more comparisons increases the chance that at least one result appears notable by chance unless the analysis plan and interpretation account for the testing strategy. The record lists 38 outcome measures and 196 analyses, but does not report which analyses were confirmatory, how they were grouped, whether any adjustment was used, or whether there was a prespecified hierarchy. The counts alone do not show that a particular result is false or that adjustment was required; they identify information needed to interpret a collection of tests.
10. Multiplicity, Missing Data and Analysis Populations
The registry records 38 outcome measures and 196 analyses. Such a broad set of outcomes and analyses raises a general multiplicity question: when many statistical comparisons are made, the chance of at least one apparently unusual result can increase. Interpreting results therefore depends on knowing which outcomes and comparisons were prespecified as primary or confirmatory, whether a testing hierarchy was used, and whether any multiplicity adjustment was applied. Those details are not reported in the registry record. The primary outcome description identifies MADRS change, but the record does not provide a complete statistical analysis plan or clarify the status of the many listed analyses.
For missing data, the primary population is described as ITT observed cases. The registry does not report the number or pattern of missing MADRS observations, reasons for missingness, or a primary missing-data strategy. It is therefore not possible to assess how much the primary estimate depended on available observations or whether the approach was tested under alternative assumptions. The GAF secondary outcome specifies LOCF at endpoint; this information applies to that listed measure and does not establish the method used for the primary endpoint or other outcomes.
Analysis populations are also only partly described. The primary endpoint uses the ITT observed-case label, but the record does not explain the operational definition of that population or report a separate per-protocol or safety population for primary efficacy. The registry's adverse-event group counts do not include at-risk denominators. These gaps limit assessment of both outcome completeness and the population underlying each reported result. No claim about bias or the direction of any missing-data effect can be made from the registry record alone.
11. What the Result Does, and Does Not, Mean
The registry record establishes that a randomized, double-blind, placebo-controlled study examined ziprasidone over six weeks in outpatients with bipolar I depression, and that change in MADRS total score was the primary endpoint. It also describes the analysis summary as a least-squares mean with standard error in the ITT observed-case population. These details identify the intended clinical comparison and outcome, but they are not a report of the trial's primary findings.
The primary group estimates, treatment contrast, uncertainty interval, and p-value are not reported in the registry record. As a result, the record cannot establish the direction or magnitude of the primary difference, how precisely it was estimated, or whether a test met a stated statistical threshold. It also cannot establish whether any difference would be clinically important. That judgment would require the observed estimate, its uncertainty, and appropriate clinical context; none should be inferred from trial design details alone.
Outcome definitions for response and remission are listed among the secondary measures, along with measures of functioning, anxiety, and mania. Their definitions do not show what participants experienced, and their results are not reported in the registry record. Similarly, adverse-event counts are listed, but missing denominators and death counts prevent a complete comparison of event risks. Counts should not be interpreted as rates when the number at risk is unavailable.
More generally, statistical significance and clinical importance are distinct concepts. A test result, if reported, would address compatibility with a statistical null under assumptions; it would not alone establish the size, relevance, or reliability of an effect. Here, the registry record does not report the primary inferential results at all. The appropriate conclusion is limited to the study design and endpoint information recorded, not a conclusion about treatment benefit, lack of benefit, or comparative safety.
12. Why This Trial Matters Statistically
This trial is statistically relevant because it combines random allocation and double masking with a placebo comparison, providing a design intended to support a controlled comparison of average symptom-score change. Its primary outcome is a continuous depression scale measured from baseline through a six-week period. That setup illustrates why endpoint definition, population, and analysis summary all matter: a scale change, a model-based mean, and a between-group contrast answer related but distinct questions.
The registry also illustrates the limits of interpreting design information without outcome estimates. Knowing that the study was randomized and placebo-controlled does not reveal the size or precision of the primary comparison. The least-squares mean and standard-error labels indicate the type of summary intended, but not the model specification or the results. A statistical report needs the contrast and its uncertainty to support interpretation.
The record's ITT observed-case wording highlights the importance of describing missing data alongside analysis populations. An ITT label is not enough to determine how participants with incomplete outcome records contributed to the analysis. Likewise, a secondary outcome's LOCF specification should not be generalized to other outcomes unless the record says it was used for them.
Finally, the recorded 38 outcome measures and 196 analyses make multiplicity an important interpretive consideration. Those counts do not invalidate any result, but they underscore why prespecified hypotheses, analysis plans, and any adjustment strategy should be available. Without the primary results and those analytic details, the statistical contribution that can be assessed from the registry record is chiefly the design and outcome framework, rather than an estimated treatment effect.
13. 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 most important limitation for statistical interpretation is that primary outcome results are not reported in the registry record. No group-specific MADRS changes, treatment contrast, confidence interval, standard error values, or p-values are available. Thus, the primary effect's direction, magnitude, and precision cannot be evaluated. The record also does not provide results for the listed secondary outcomes.
Several details needed to assess the analysis are absent. The statistical model, covariates, treatment contrast specification, handling of repeated measurements, and missing-data procedures for the primary endpoint are not reported in the registry record. Although the analysis population is called ITT observed cases, the operational rules and completeness of outcome data are not described. The record does not report the trial's own sample-size justification, interim analyses, or a full statistical analysis plan.
Multiplicity is difficult to assess because the record lists 38 outcome measures and 196 analyses without explaining which were prespecified or how the analyses were organized. Adverse-event counts are also incomplete for comparative interpretation: at-risk denominators and death counts are not reported in the registry record. The serious-event terms listed with 0.6% do not provide enough denominator information to evaluate group-specific risks.
There are also gaps in trial conduct details. The record does not describe the randomization sequence, concealment, masking procedures, site details, or fuller eligibility criteria. Arm descriptions are blank. These omissions limit assessment of design implementation and transportability, but do not demonstrate that procedures were inadequate. The registry record supports a description of the trial framework, not an independent assessment of its full conduct or results.
14. Related Tutorials
15. Related Statistical Calculators
16. Sources
- ClinicalTrials.gov: A Six-Week Study Evaluating The Efficacy And Safety Of Geodon In Patients With A Diagnosis Of Bipolar I Depression — NCT00141271.
- Registry record last updated 2021-03-25; status last verified 2021-03.
- Results first posted 2009-03-24.
- 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.
17. Record Summary
The registry describes a completed Phase 3, randomized, double-blind, multicenter, parallel-group, placebo-controlled study of oral ziprasidone in outpatients with bipolar I depression. The study period was six weeks, and enrollment is recorded as 536. The brief summary identifies participants aged 18 and older and a required washout of at least 7 days of any prior medication.
The primary endpoint was change in MADRS total score from baseline to 6 weeks, summarized as a least-squares mean with standard error in the ITT observed-case population. The registry lists response and remission outcomes and other measures of depression, functioning, anxiety, and mania. It records 38 outcome measures and 196 analyses.
Primary outcome estimates and a statistical analysis plan are not reported in the registry record. Missing-data procedures for the primary endpoint, multiplicity handling, and adverse-event denominators are also not reported. The available information supports description of the design and planned outcome framework, but not an estimate of the primary treatment comparison.