Extension Study: Evaluating the Safety of Oral Ziprasidone in the Treatment of Subjects With Schizophrenia: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Extension Study: Evaluating the Safety of Oral Ziprasidone in the Treatment of Subjects With Schizophrenia (NCT00139737), a non-randomized phase 3 study. The primary outcome is Number of Participants With Treatment-Emergent Adverse Events (AEs) and Serious Adverse Events (SAEs).

2. Clinical Question

The registry lists 344 participants as the enrollment and identifies the study as phase III. Its primary outcome description states that observed or volunteered treatment-emergent events were reported regardless of treatment group or suspected causal relationship to the investigational product. Thus, the outcome captures events occurring during the study period; it does not establish that ziprasidone caused any particular event. Because participants entered after completing earlier ziprasidone studies and there was no control group, these results describe the study’s observed safety experience rather than what would have happened under another treatment or no treatment.

3. Trial Design

This was an open-label, non-randomized, single-group extension study of oral ziprasidone. The registry describes its purpose as providing treatment to eligible participants who had successfully completed one of the earlier phase III ziprasidone studies A1281028, A1281044, A1281045 (NCT00136994), or A1281088 (NCT00143351). The study therefore followed participants from prior studies rather than recruiting a treatment group and a concurrent randomized control group for this record.

The study is listed as phase III and completed. Its recorded start was 2002-03 and its primary completion was 2009-06. The registry lists schizophrenia as the condition and oral ziprasidone as the intervention. The study title and official title identify safety and tolerability as the focus of the extension. Masking was none, so the study was open.

The registry lists enrollment of 344. The primary outcome’s analysis population is the Safety Analysis Set, defined as all subjects who took at least one dose of study drug; the adverse-event results list 331 participants at risk for the reported categories. These are different registry quantities and should not be treated as interchangeable. The record does not explain the relationship between enrollment and the safety analysis set.

With no random assignment and no comparator group, the design can describe events reported among participants receiving ziprasidone in this extension, but it cannot separate treatment-related effects from other explanations by comparing randomized groups. Selection into an extension after successful completion of an earlier study also defines the population being described: these results concern eligible prior-study completers, not an unrestricted population of people with schizophrenia. Details of eligibility criteria beyond that description 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

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Safety Analysis Set = All subjects who took at least 1 dose of study drug

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 Number of Participants With Treatment-Emergent Adverse Events (AEs) and Serious Adverse Events (SAEs), measured over Baseline up to 72 months.

The primary outcome was the number of participants with treatment-emergent adverse events (AEs) and serious adverse events (SAEs), from baseline up to 72 months. The unit was participants, and the outcome parameter was a number rather than a time-to-event estimate or a measure of symptom change. The registry says that all observed or volunteered treatment-emergent AEs and SAEs were reported regardless of treatment group or suspected causal relationship to the investigational product.

Counting participants with events is useful for describing how many people experienced an event during follow-up. The distinction between serious and other adverse events also conveys different levels of event classification. These counts do not, by themselves, describe severity within each category, duration, recurrence, or whether an event was judged related to the study drug. Nor do counts alone account for differences in time observed among participants.

As a general statistical point, adverse-event monitoring is an important part of evaluating a treatment’s safety profile because it records unwanted medical occurrences during study participation. The interpretation depends on the event definitions, how events were collected, the follow-up actually received, and the analysis population. The registry gives the broad collection description and the stated follow-up timeframe, but further operational details about event coding, ascertainment, and handling of incomplete follow-up are not reported in the registry record.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The registry presents a binary safety outcome as participant counts: whether participants had treatment-emergent adverse events or serious adverse events. For a binary outcome, each participant is classified according to whether the outcome occurred under the study’s definitions. Counts give the number classified as having an event; they do not by themselves supply an estimate of comparative risk. In this study, the posted results are descriptive counts from one treatment group.

What does a participant count tell the reader?

A count is an absolute description of the recorded experience in the analysis population. It is not automatically a rate. A rate generally incorporates time at risk, whereas the registry reports counts and a follow-up timeframe of baseline up to 72 months. Individual observation time and a person-time analysis are not reported in the registry record. The count therefore cannot indicate how frequently events occurred per unit of time or account for any variation in follow-up.

The record defines the Safety Analysis Set as all subjects who took at least one dose of study drug. This is the stated population for the primary safety outcome. It is different in purpose from the enrollment figure, which the registry lists as 344. The event results list 331 participants at risk. The registry does not explain the difference between these quantities, so the counts should be interpreted with the stated safety population definition rather than treated as if every enrolled participant necessarily contributed to the safety analysis.

Why does the design affect interpretation?

Randomization can help make treatment groups comparable on average, while a concurrent control group provides a reference for interpreting event counts. Neither feature is present in this record: allocation was non-randomized, masking was none, and the model was single group. The study can therefore describe events observed during ziprasidone exposure in the extension, but there is no within-study control contrast from which to estimate a relative treatment effect. Differences between this study’s participants and people who did not enter the extension may also affect how broadly the findings apply.

The event collection description says that observed or volunteered treatment-emergent events were reported regardless of treatment group or suspected causal relationship. General statistical interpretation distinguishes an event that occurs during treatment from an event caused by treatment. Timing alone does not establish causation, and the registry’s inclusive reporting description is not a causal attribution method. Further details on coding, adjudication, or causality assessment are not reported in the registry record.

What is not estimated?

No inferential analyses are listed for this outcome. A confidence interval, in general, expresses the uncertainty associated with an estimate under a statistical model and its assumptions. No such interval is reported for the adverse-event counts. Likewise, a p-value would address a specified statistical hypothesis under a defined analysis; no hypothesis test or p-value is listed. Without these analyses, the results do not quantify statistical uncertainty or support a test of differences between groups.

Over-dispersion is a general concern when data vary more than a chosen model expects, and clustered or repeated event data may require methods that account for their structure. Here, the reported endpoint is a participant count, and the registry does not report a count model, event-level analysis, or adjustment for repeated events. It is not possible to assess model fit or over-dispersion from the posted results. The record also does not state how repeated occurrences within a participant were handled beyond reporting the number of participants with events.

How should planning illustrations be read?

Any sample-size planning illustrations based on an observed effect are hypothetical planning calculations, not observed power and not part of the trial’s reported design. The stated assumptions for such illustrations are a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation; the effect to detect is a stated fraction of the observed effect. Those assumptions describe a hypothetical two-arm planning exercise, whereas this study was a non-randomized, single-group extension. The planning figures therefore do not imply that the trial enrolled two arms or was designed around that calculation.

More generally, a sample-size calculation depends on the outcome, target effect, variability or event frequency, significance level, power target, allocation, and analysis method. A planning result is conditional on its assumptions. It is not evidence that a study had adequate power after completion, and it does not replace an analysis of observed outcomes. The registry record does not report a protocol-based sample-size justification for this extension.

10. Multiplicity, Missing Data and Analysis Populations

The registry lists one primary outcome measure: the number of participants with treatment-emergent AEs and SAEs. It does not list analyses for that measure. A statistical plan describing outcome hierarchy, prespecified comparisons, or any adjustment for multiple testing is not reported in the registry record. Without inferential analyses, there are no reported p-values whose interpretation would depend on a multiplicity adjustment.

As a general statistical matter, multiplicity arises when several hypotheses or outcomes are tested, because interpreting each test separately can increase the chance of at least one false-positive finding. Methods such as prespecifying a primary outcome or adjusting inference can address particular multiplicity concerns. Which approach was planned for this study is not reported in the registry record. The list of serious adverse-event terms is descriptive; no statistical testing of those terms is reported.

The safety population is defined as participants who took at least one dose of study drug. The posted event counts give the number at risk for the serious and other event categories, but the record does not describe missing outcome data, withdrawals, loss to follow-up, or the handling of participants with incomplete observation. It also does not report whether follow-up time differed across participants or how that might affect the observed counts. Missing-data methods and sensitivity analyses are not reported in the registry record.

These omissions do not establish that no analyses or procedures were conducted; they mean that their details are not available in the registry record. Consequently, readers cannot assess whether results were adjusted for multiple testing, how incomplete data were handled, or whether alternative analytic choices would change the descriptive findings.

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 0% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: schizophrenia (4.2%), irritability (0.6%), ankle fracture (0.6%), intentional overdose (0.6%), aggression (0.6%).

The serious adverse-event results list 32 participants at risk for the category and 32 participants with a serious adverse event. The other adverse-event category lists 331 participants at risk and 177 participants with an event. These are participant counts in the ziprasidone group, not a comparison between treatment arms. The registry’s primary outcome description says events were reported regardless of suspected causal relationship to the investigational product.

The listed serious-event terms are Schizophrenia (4.2%), Irritability (0.6%), Ankle fracture (0.6%), Intentional overdose (0.6%), Aggression (0.6%), and Agitation (0.6%). The registry presents these as serious adverse-event terms with percentages; it does not give event-specific causal attribution in this listing. The percentages should not be read as proof that treatment caused the events.

The serious and other event categories may not be mutually exclusive, and the record does not report overlap. Their counts should not be summed to create a total number of participants with any event. A death count is not reported in the registry record. The timeframe is baseline up to 72 months, but the record does not provide participant-level follow-up or a time-adjusted event rate.

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

The result establishes that adverse events were recorded during this open extension study and gives participant counts for serious and other adverse events in the safety analysis set. It also lists several serious adverse-event terms and their percentages. These figures are descriptive of the participants included in the registry’s safety outcome and the event-reporting approach stated for the study.

The results do not establish that ziprasidone caused any particular event. The outcome description explicitly includes events regardless of suspected causal relationship, and there is no comparator group to help distinguish treatment-associated event experience from background illness, other exposures, or chance. The registry also reports no inferential analysis, so the counts do not provide a tested estimate of treatment effect or statistical precision.

The findings do not describe a randomized comparison, because the study was non-randomized and single group. They do not establish how event counts would compare under another treatment, placebo, or no treatment. They also do not give a participant-level account of when events occurred, whether events recurred, how long participants were followed, or how missing observations were handled.

The extension’s eligibility context matters: participants had successfully completed an earlier ziprasidone study before entering this study. The observed experience therefore pertains to the eligible participants who entered the extension and took at least one dose, rather than automatically representing all people with schizophrenia or all people who might receive ziprasidone.

In general, safety evidence is interpreted alongside study design, event definitions, exposure and follow-up, and the possibility of alternative explanations. The registry record supplies some of this context but does not report several details needed for a fuller assessment, including individual follow-up, missing-data procedures, and inferential estimates. The appropriate reading is therefore descriptive: the record documents reported events in this study population, without establishing causality or comparative safety.

13. Why This Trial Matters Statistically

This study is statistically informative in part because its design makes the scope of its conclusions especially clear. An open, non-randomized, single-group extension can record safety events during continued treatment, but it does not create the randomized contrast needed to estimate how those events differ from a control condition. Recognizing that distinction prevents descriptive event counts from being mistaken for a comparative treatment effect.

The registry also makes the analysis population explicit: the Safety Analysis Set includes participants who took at least one dose. That definition helps readers understand which participants the primary safety outcome concerns. The registry lists enrollment of 344 and reports 331 participants at risk for the event categories; keeping enrollment and the analysis population distinct is important when interpreting safety results.

The outcome includes treatment-emergent events whether or not a causal relationship was suspected. This broad reporting approach is useful for describing events observed during the study, but it reinforces the need to distinguish occurrence during treatment from causation. The serious-event terms and counts are part of that descriptive record, not proof of a drug effect.

Finally, the record illustrates why reporting analysis details matters. No inferential analyses are listed, and details about missing data, follow-up variation, and multiplicity are not reported in the registry record. Without those details, the available evidence supports a limited description rather than a precise comparative conclusion. These design and reporting features are essential context for interpreting extension-study safety results.

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 principal limitation is the absence of a concurrent comparator. Because the study was non-randomized and single group, its event counts cannot establish how safety outcomes compare with another treatment or with no treatment. The open design also means masking was none. The results are descriptive and do not isolate the effect of ziprasidone from other possible explanations for reported events.

Eligibility was based on successful completion of a previous ziprasidone study. This selects a particular extension-study population and may limit how well the findings apply to people who would not meet that prior-completion condition. The registry does not report additional eligibility details or explain how participants who did not enter the extension differed from those who did.

The record lists enrollment of 344 and reports 331 participants at risk for the event categories, but does not explain the relationship between those figures. It gives a follow-up timeframe of baseline up to 72 months without reporting individual follow-up duration, withdrawals, incomplete observations, or missing-data handling. Counts alone do not account for differences in exposure time, and no time-adjusted analysis is reported.

There are no inferential analyses listed. Confidence intervals, hypothesis tests, adjusted estimates, and sensitivity analyses are not reported in the registry record. The record also does not describe multiplicity procedures, detailed event coding, or how repeated events within a participant were handled. A death count is not reported. These gaps constrain interpretation of precision, event ascertainment, and completeness; they should not be taken as evidence that unreported procedures or events did not exist.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: Extension Study: Evaluating the Safety of Oral Ziprasidone in the Treatment of Subjects With Schizophrenia — NCT00139737.
• Registry record last updated 2021-03-03; status last verified 2021-03.
• Results first posted 2010-07-14.
• 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

The registry record describes a completed phase III, open, non-randomized, single-group extension study of oral ziprasidone for schizophrenia. Its purpose was to provide treatment to eligible participants who had successfully completed one of the named earlier ziprasidone studies. The recorded start was 2002-03 and primary completion was 2009-06. Enrollment is listed as 344.

The primary outcome counted participants with treatment-emergent adverse events and serious adverse events from baseline up to 72 months. The Safety Analysis Set included participants who took at least one dose. The adverse-event results list 331 participants at risk and report 32 participants with serious adverse events and 177 with other adverse events. No inferential analyses are listed, and there is no control group. The findings therefore describe reported safety events in this extension population, rather than a randomized comparison or causal estimate.