A Study of Paliperidone Palmitate 6-Month Formulation: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for A Study of Paliperidone Palmitate 6-Month Formulation (NCT03345342), a randomized phase 3 study. The primary outcome is Time to Relapse During the Double-Blind (DB) Phase.

2. Clinical Question

3. Trial Design

The registry identifies NCT03345342 as a phase 3, randomized, double-blind, parallel-group, active-controlled study. It lists an enrollment of 841 and a completed status. The listed start date is 2017-11-20, and the primary completion date is 2020-05-08. The study title identifies paliperidone palmitate 6-month formulation as its focus.

The study included treatment periods before the double-blind comparison. The registry describes a transition phase of up to 4 months for participants who had not achieved stability with moderate to higher doses of PP1M or PP3M. During that period, participants could receive 1 to 5 PP1M injections. In the maintenance phase, participants received one dose of PP1M 100 or 150 mg eq. or PP3M 350 or 525 mg eq. before proceeding to the double-blind phase.

In the double-blind phase, the PP6M group received PP6M injections on Day 1 and Day 183, with placebo injections on Day 92 and Day 274. The PP3M group received PP3M injections on Day 1, Day 92, Day 183 and Day 274. The primary outcome record lists 478.0 participants in the PP6M group and 224.0 in the PP3M group. Those are outcome-specific analysis counts, not the overall enrollment figure.

The primary analysis population is described as a double-blind intent-to-treat set: participants randomly assigned to PP6M or PP3M during the double-blind phase who received at least one dose of PP6M or PP3M. The registry record does not report the method used to generate or conceal randomization, details of the masking procedure, or the reasons for the unequal group sizes. It also does not report the trial's sample-size justification, interim analyses, or a detailed statistical analysis plan.

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: Total 841 participants were enrolled, out of which 838 participants received treatment. 2 participants were enrolled but did not receive treatment and 1 participant withdrew from the Open-label Phase and as permitted per the protocol, re-entered the study (under a different participant identifier) and was counted twice in the total. Per the rules specified in statistical analysis plan, only data collected during the second study participation was included in data summary for this participant.

Open-Label Phase

Reasons for not completing this period:

Double-Blind Phase

Reasons for not completing this period:

Follow-Up Phase

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: DB Intent-to-Treat (ITT) analysis set included participants who were randomly assigned to paliperidone palmitate 6-month (PP6M) /paliperidone palmitate 3-month (PP3M) during DB Phase, received at least 1 dose of PP6M/PP3M.

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 Time to Relapse During the Double-Blind (DB) Phase, measured over Up to 12 months of DB Phase.

The primary endpoint was time to relapse during the double-blind phase, assessed through Month 12. It measures the interval between randomization in that phase and the first documented relapse event. The registry lists the summary measure as a median time in days with a 95% confidence interval, but the median values and intervals are not reported in the registry record.

The registry's relapse definition covers several types of events. These include psychiatric hospitalization; specified increases in the Positive and Negative Syndrome Scale (PANSS) total score; serious self-injury, violent behavior or clinically significant suicidal or homicidal ideation; and specified high scores on selected PANSS items. The PANSS criteria require repeated assessments in some circumstances. The endpoint therefore combines clinical events and scale-based criteria rather than relying on a single symptom measure.

Time to relapse is used to retain information about when an event occurs, rather than recording only whether a participant had an event by a fixed visit. In general, time-to-event summaries can describe the timing of events while allowing for participants whose event time is not observed during follow-up. The registry record does not describe the censoring rules or the specific method used to estimate the median and its confidence interval for this trial.

Because the posted primary result lacks the group-specific medians and confidence intervals, the endpoint's intended clinical meaning can be described, but its comparative estimate and precision cannot be assessed from the reported primary outcome record.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The registry describes a time-to-event primary endpoint and continuous secondary endpoints. Each calls for a different kind of summary. The distinction matters because the primary result is not available in the posted record, while selected secondary outcome summaries are available.

What does a median time-to-event estimate describe?

In general, a median time-to-event estimate identifies the time by which half of the relevant event-time distribution has occurred, under the estimation method used. Here, the event is the first documented relapse during the double-blind phase. A 95% confidence interval describes statistical uncertainty around an estimated quantity: across repeated studies under the same assumptions, intervals constructed by the same method would cover the target parameter at the stated rate. It is not a range containing 95% of individual participants' relapse times.

The registry identifies the primary measure as median time in days with a 95% confidence interval, but does not report those values. It also does not report the event counts, the censoring approach, or a hazard ratio for the primary comparison. Without the estimates and their uncertainty, the primary endpoint cannot be reconstructed from the double-blind group sizes alone. Group size does not reveal how many relapses occurred or when they occurred.

How should the continuous secondary summaries be read?

For the posted PANSS, CGI-S and PSP outcomes, the registry gives group-specific mean changes and standard deviations. A mean is the arithmetic average of the observed changes in the listed group. A standard deviation describes the spread of participant values around that group mean; it is not the uncertainty around the mean itself. The number of participants contributing to each outcome is listed separately from the group count in the primary outcome record.

The recalculated comparisons express the difference between the two group means, with a 95% confidence interval. In general, a confidence interval for a difference communicates both the estimated direction and the range of values reasonably compatible with the data and method. If it spans zero, the data do not clearly distinguish the groups at the corresponding conventional two-sided threshold. That does not prove the true difference is zero, and it does not establish equivalence or non-inferiority. The registry's recalculated p-values are unadjusted, not sponsor-reported analyses.

For these recalculations, the reported standard deviations were used to estimate a pooled standard deviation and the standard error for the mean difference. A standard error describes the variability expected in an estimate across repeated samples, under the calculation's assumptions. The normal approximation used for the planning illustrations is a general approximation for translating an assumed effect and variability into a sample-size illustration. It is not a statement about the trial's actual design process.

What does over-dispersion mean in general?

For count outcomes, over-dispersion means the observed variability exceeds what a simpler count model would predict. It can affect standard errors and confidence intervals if a model assumes less variability than the data contain. The posted primary endpoint is time-to-event, and the three numerical secondary outcomes are continuous; the registry record does not report a count-outcome model or an over-dispersion analysis for this trial. The concept is relevant generally, but it should not be attributed to the trial's analysis.

What are the sample-size illustrations?

The planning illustrations use an effect derived from the observed secondary comparison and show hypothetical participants per arm for 80% power and for 90% power under a two-sided significance level of 0.05, equal allocation, and a normal approximation. Some rows consider planning for 75% or 50% of the observed effect. These are planning illustrations, not the trial's original sample-size calculation and not observed power. The registry record does not report the original design assumptions or sample-size justification. The illustrations are especially sensitive when the observed difference is close to zero: a small observed effect leads to very large hypothetical sample sizes under the stated approach.

Those illustrations do not supply the missing primary analysis. They concern continuous outcomes with posted means and standard deviations; the primary endpoint is time to relapse. The registry record does not report an event-based planning calculation for the primary endpoint, such as the target number of events or the assumptions underlying any such calculation.

What does the analysis population imply?

The primary outcome population includes randomized participants who received at least one double-blind dose of PP6M or PP3M. This definition is narrower than simply saying all randomized participants: receiving a dose is an explicit inclusion condition. The registry record does not report how participants with missing outcome data were handled, whether all randomized participants were included in separate sensitivity analyses, or what methods were used for incomplete follow-up. Those details affect how the posted estimates would be interpreted, but cannot be inferred from the analysis-set description alone.

The study objective is framed as non-inferiority, but the registry record does not report a non-inferiority margin, a prespecified primary analysis method, or sponsor-reported primary estimates. The posted secondary recalculations therefore remain descriptive, unadjusted comparisons and cannot stand in for a prespecified non-inferiority test.

10. Multiplicity, Missing Data and Analysis Populations

The registry record lists 19 secondary outcomes as posted and identifies 20 outcome measures overall. It does not report how the primary and secondary hypotheses were ordered, whether a family-wise error or false-discovery approach was planned, or whether any secondary outcomes were designated as confirmatory. A multiplicity strategy is important when several outcomes are tested because interpreting each test in isolation can overstate the strength of the overall evidence. The record does not report a correction for multiplicity.

The three posted continuous secondary outcomes have recalculated, unadjusted comparisons, and the record lists no sponsor-reported analyses for them. These recalculations should therefore be read as exploratory summaries, not as a substitute for a reported prespecified analysis. The registry record does not report a detailed statistical analysis plan, interim analyses, or the primary non-inferiority margin.

Missing-data procedures are also not reported in the registry record. The primary outcome population is described as randomized participants who received at least one double-blind dose, but the record does not explain how participants with missing relapse assessments or incomplete follow-up were handled. It does not report imputation methods, censoring details, sensitivity analyses, or whether results were checked under alternative assumptions about missing outcomes. These omissions make it difficult to assess the robustness of any hypothetical primary estimate, even if a summary estimate were available.

For the continuous outcomes, participant counts differ by outcome, but the registry record does not explain why the numbers differ or how missing measurements were addressed. It also does not state whether the reported summaries include all observed values or rely on a particular missing-data model. In general, missingness can affect estimates when the participants with observed outcomes differ systematically from those without them.

The listed primary analysis set is a received-at-least-one-dose population rather than an unqualified all-randomized population. The record does not report a separate analysis population or sensitivity analysis that would clarify the effect of this definition. These points do not establish that the analyses were biased; they identify information that is not reported and is needed to evaluate the analyses fully.

11. Safety Results

The registry reports serious adverse events as affected participants divided by participants at risk.

Deaths

Other (non-serious) adverse events

Participants with at least one non-serious event at or above the registry's 5% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: suicidal ideation (2.4%), schizophrenia (1.8%), brain neoplasm (0.9%), nasal sinus cancer (0.9%), mental disorder (0.9%).

The registry reports serious adverse event counts by study phase and group, alongside the number of participants at risk in each group. In the open-label PP1M/PP3M group, 23 of 838 participants were reported with serious adverse events. In the double-blind phase, the registry reports 15 of 224 participants for PP3M and 24 of 478 for PP6M. It also lists follow-up groups: 1 of 42 for PP3M and 6 of 109 for PP6M. These are descriptive counts as recorded, not adjusted comparisons.

The registry separately reports other, non-serious adverse events and deaths. The listed death counts are 1 in the open-label group, 2 in double-blind PP3M, 1 in double-blind PP6M, and 0 in each follow-up group. Serious adverse event counts and deaths are different summaries and should not be combined or treated as interchangeable.

The record lists serious-event terms including suicidal ideation, schizophrenia, brain neoplasm, nasal sinus cancer, mental disorder and suicide attempt, with percentages. It does not establish that these terms were caused by treatment, and the listed terms should not be interpreted as comparative risks between the double-blind groups. These adverse-event summaries do not answer the primary relapse question. Differences in group size and study phase also mean that raw counts alone are not a direct measure of comparative safety. The registry record does not report a causal analysis of these events.

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

The study was designed to compare a PP6M schedule with a PP3M schedule in participants with schizophrenia who had reached the described stabilization pathway. Its primary question concerned time to relapse, not simply average symptom-score change. The registry records that objective and defines relapse using hospitalization, PANSS criteria and specified clinically serious behaviors or symptoms.

The posted primary outcome record does not include median time-to-relapse estimates or confidence intervals. As a result, it does not show the size or precision of the primary comparison. It also does not report a non-inferiority margin or sponsor-reported primary hazard ratio. The stated study aim cannot be treated as evidence that the aim was achieved; the available primary result information is insufficient to determine whether the PP6M schedule met the stated objective.

The posted secondary summaries show similar group means across the three outcomes with numerical results, and the unadjusted recalculations do not provide clear evidence of separation on those measures. That pattern is descriptive. It does not establish that the treatments are equivalent, and it does not confirm the primary result. In general, absence of a statistically clear difference is not the same as evidence that a prespecified clinically acceptable difference has been ruled out.

Nor do the results establish how the findings would apply to people outside the study's described pathway. Participants proceeded through transition and maintenance phases before the double-blind comparison, and the primary analysis set required at least one dose in the double-blind phase. The record does not provide enough information to evaluate every aspect of selection, follow-up, or missing outcomes.

The appropriate conclusion from the posted registry information is therefore limited: the trial was randomized and double-blind, the primary endpoint and analysis population are identified, and some secondary outcome summaries and adverse-event counts are available. The primary comparison itself is not reported in the registry record. No treatment recommendation or clinical decision follows from these incomplete comparative results.

13. Why This Trial Matters Statistically

This trial raises a statistical question that is easy to misread: the study objective and the study result are not interchangeable. A non-inferiority objective specifies the kind of claim the investigators sought to evaluate. Establishing that claim requires a reported comparison, its uncertainty, and an analysis tied to an acceptable margin. The registry record gives the objective but does not report the primary estimates or the margin.

The distinction between endpoints also matters. Time to relapse captures the timing of a defined clinical event, whereas the posted secondary outcomes summarize changes in symptom, severity and functioning scales. Those outcomes provide different information and should not be collapsed into a single judgment. Similar mean changes do not demonstrate that time to relapse was similar, nor do they establish non-inferiority.

The record also illustrates why statistical precision matters. An estimate without its confidence interval can conceal substantial uncertainty. Here, the primary estimate and interval are both not reported in the registry record, leaving readers without the information needed to judge the range of plausible comparative effects. Group sizes alone cannot fill that gap.

Finally, the recalculated planning illustrations are not evidence of observed power or a reconstruction of the original design. They show how hypothetical sample sizes depend on the effect assumed for planning. A small observed effect can produce very large planning figures, but this does not prove that the original study was underpowered or that its design used those assumptions. The trial's own sample-size justification is not reported in the registry record.

These distinctions help keep the interpretation proportionate to the evidence that is actually posted: describe the design and recorded summaries, separate descriptive patterns from inferential claims, and identify what remains unavailable.

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 most important limitation of the registry information is that the primary result is missing. The endpoint is named and defined, but its median estimates and confidence intervals are not reported. The record therefore cannot support a statistical assessment of the primary comparison or the stated non-inferiority objective.

The double-blind comparison followed transition and maintenance phases. The registry describes stabilization before entry into the double-blind phase, so the analyzed population reflects this study pathway rather than an unrestricted population of people with schizophrenia. The primary analysis set also required receipt of at least one double-blind dose. These features matter when considering how broadly the results might apply, although the record does not report enough detail to quantify their effects.

Secondary results are limited. Numerical group summaries and unadjusted recalculations are available for three outcomes, while the registry lists 19 secondary outcomes as posted. The record does not list sponsor-reported analyses for the three numerical comparisons or explain a multiplicity strategy. Consequently, those results have less inferential weight than a complete, prespecified analysis plan and report would allow.

Other relevant analysis details are not reported in the registry record, including missing-data procedures, the primary censoring method, sensitivity analyses, and the original sample-size justification. The posted adverse-event summaries are counts across several phases and do not provide a comparative causal analysis. Without additional analysis details or complete primary results, readers cannot fully assess the robustness of the findings or distinguish observed differences from uncertainty.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: A Study of Paliperidone Palmitate 6-Month Formulation — NCT03345342.
• Registry record last updated 2025-04-29; status last verified 2025-04.
• Results first posted 2021-12-09.
• 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

NCT03345342 is a completed phase 3 randomized, double-blind, parallel-group, active-controlled study with an enrollment of 841. It compared PP6M with PP3M in a population that proceeded through transition and maintenance phases before double-blind treatment. The primary endpoint was time to relapse through Month 12, analyzed in a double-blind intent-to-treat set requiring at least one dose.

The registry record does not report the primary median times or their 95% confidence intervals, and it does not report the non-inferiority margin. Numerical secondary summaries are available for PANSS, CGI-S and PSP change, with unadjusted recalculations, but these do not answer the primary question. Adverse-event counts are listed across open-label, double-blind and follow-up phases. Overall, the design and some secondary and safety summaries are described, while the primary comparative result and several important analysis details are not reported in the registry record.