A Study to Evaluate Effectiveness and Safety of ER OROS Paliperidone in Patients With Schizophrenia: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for A Study to Evaluate Effectiveness and Safety of ER OROS Paliperidone in Patients With Schizophrenia (NCT00396565), a randomized phase 3 study. The primary outcome is Change From Baseline in the Total Positive and Negative Syndrome Scale (PANSS)..

2. Clinical Question

3. Trial Design

The registry record describes a randomized, double-blind, parallel-group, phase 3 study. The study was completed, with enrollment of 394 participants. Its stated purpose was to evaluate the effectiveness and safety of extended-release OROS paliperidone compared with placebo in patients with schizophrenia; olanzapine was included as a reference drug. The record identifies Janssen Pharmaceutical K.K. as sponsor.

The trial included paliperidone, placebo, and olanzapine arms. For the primary outcome, the posted group sizes were 134 for paliperidone, 138 for placebo, and 46 for olanzapine. The posted safety counts use 47 as the olanzapine group at risk. These differing analysis counts are not reconciled in the registry record. Enrollment is the registry’s overall enrollment figure and should not be substituted for the group-specific analysis counts.

The primary outcome was change from baseline in total Positive and Negative Syndrome Scale (PANSS) score, assessed at baseline and 6 weeks. The registry describes a Full Analysis Set that excluded subjects who did not receive test drug or did not have post-treatment efficacy data. The remainder of the population description is truncated and is not reported in the registry record. The record names ANCOVA as the sponsor’s analysis method for the primary comparison of paliperidone and placebo.

The registry record does not report the original sample-size justification, the planned allocation ratio, details of randomization implementation, the specific covariates used in ANCOVA, or any interim-analysis plan. It also does not provide enough information to infer how the olanzapine reference comparison was intended to be interpreted beyond the stated purpose of examining paliperidone’s clinical position. These omissions limit what can be reconstructed about the protocol and its statistical planning.

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: Informed Consent was obtained from 394 patients. The actual number of patients randomized in the study was 321. 2 subjects discontinued before treatment. 319 subjects were treated and analyzed for safety. 318 subjects were analyzed for baseline characteristics and efficacy.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Full Analysis Set excludes subjects who didn't receive test drug or didn't have post-treatment efficacy data. The Olanzapine (OLZ) group was set as an active drug group to examine clinical position of paliperidone (PAL) ER, and the superiority or non-inferiority of PAL ER 6mg to OLZ 10mg wasn't verified. LOCF imputation method was applied.

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 From Baseline in the Total Positive and Negative Syndrome Scale (PANSS)., measured over Baseline and 6 weeks.

The primary endpoint is change from baseline in total PANSS score at 6 weeks. PANSS assesses symptoms of schizophrenia. The registry describes a 7-point rating for each symptom, from 1 (absent) to 7 (extreme psychopathology); the total is the sum of 30 items and ranges from 30 (absent) to 210 (extreme ill). A change score expresses how the measured total differs at follow-up from the baseline score. In these results, negative mean changes indicate decreases from baseline, while a positive mean change indicates an increase.

A total score combines the scale’s assessed symptoms into one continuous outcome, making it possible to compare average changes across randomized groups. That gives a broad summary of the measured symptom-score change, but it does not show how every individual symptom changed or whether participants had the same response. The posted primary outcome is a group mean, accompanied by a standard deviation; the standard deviation describes variability among participants’ observed changes, rather than uncertainty in the group mean.

The endpoint is useful for comparing the groups over the stated follow-up, but its interpretation depends on the analysis population, the handling of incomplete follow-up data, and the analysis method. The population is described only in part in the registry record, and details of missing-data handling are not reported in the registry record. The scale description explains the range and direction of scores; it does not, by itself, establish the clinical importance of a particular change.

7. Results Posted in the Registry

• Paliperidone Extended Release (ER) (JNS007ER): mean -9.10 versus 3.80 on control; difference -12.90 (95% CI -17.33 to -8.47), standardized effect d = -0.69; z-test p = <0.001.
• Olanzapine: mean -9.90 versus 3.80 on control; difference -13.70 (95% CI -19.49 to -7.91), standardized effect d = -0.74; z-test p = <0.001.

8. Secondary Efficacy Endpoints

The registry also posts results for secondary outcomes. For each outcome below, the endpoint type, the test applied to the posted summary statistics, the result and its interpretation are given in turn. Secondary endpoints are usually not adjusted for multiple testing, so they support rather than confirm the primary finding.

Change From Baseline in Positive and Negative Syndrome Scale (PANSS) - Positive Subscale Score

• Endpoint type: Continuous: a mean value per participant, compared as a difference in means.
• Test: Unadjusted two-sample z-test for the difference in means, using the posted standard deviations.
• Result:
  • Paliperidone Extended Release (ER) (JNS007ER): mean -2.60 versus 1.60 on control; difference -4.20 (95% CI -5.53 to -2.87), standardized effect d = -0.75; z-test p = <0.001.
  • Olanzapine: mean -2.40 versus 1.60 on control; difference -4.00 (95% CI -5.80 to -2.20), standardized effect d = -0.73; z-test p = <0.001.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the mean is statistically lower for Paliperidone Extended Release (ER) (JNS007ER) than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.
  • The 95% confidence interval excludes no difference, so the mean is statistically lower for Olanzapine than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

Change From Baseline in Positive and Negative Syndrome Scale (PANSS) - Negative Subscale Score

• Endpoint type: Continuous: a mean value per participant, compared as a difference in means.
• Test: Unadjusted two-sample z-test for the difference in means, using the posted standard deviations.
• Result:
  • Paliperidone Extended Release (ER) (JNS007ER): mean -2.40 versus 0.00 on control; difference -2.40 (95% CI -3.60 to -1.20), standardized effect d = -0.48; z-test p = <0.001.
  • Olanzapine: mean -2.80 versus 0.00 on control; difference -2.80 (95% CI -4.29 to -1.31), standardized effect d = -0.58; z-test p = <0.001.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the mean is statistically lower for Paliperidone Extended Release (ER) (JNS007ER) than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.
  • The 95% confidence interval excludes no difference, so the mean is statistically lower for Olanzapine than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

Change From Baseline in Positive and Negative Syndrome Scale (PANSS) - General Psychopathology Subscale Score

• Endpoint type: Continuous: a mean value per participant, compared as a difference in means.
• Test: Unadjusted two-sample z-test for the difference in means, using the posted standard deviations.
• Result:
  • Paliperidone Extended Release (ER) (JNS007ER): mean -4.10 versus 2.20 on control; difference -6.30 (95% CI -8.71 to -3.89), standardized effect d = -0.62; z-test p = <0.001.
  • Olanzapine: mean -4.70 versus 2.20 on control; difference -6.90 (95% CI -10.06 to -3.74), standardized effect d = -0.67; z-test p = <0.001.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the mean is statistically lower for Paliperidone Extended Release (ER) (JNS007ER) than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.
  • The 95% confidence interval excludes no difference, so the mean is statistically lower for Olanzapine than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

The posted secondary outcomes cover PANSS positive, negative, and general psychopathology subscales, the proportion of responders defined by a 30% decrease in total PANSS, and change in Clinical Global Impression Scale score. The record lists six outcome measures and six analyses, but only five secondary outcome records are posted. Results are available for the three PANSS subscales; outcome results for the responder measure and CGI-S are not reported in the registry record.

Across the three posted subscales, the paliperidone-versus-placebo mean differences are negative, and the sponsor’s ANCOVA results for those comparisons also have confidence intervals below zero and p-values below < 0.001. This gives a directionally consistent pattern alongside the primary total PANSS result. The subscales describe components of the same broad symptom assessment, so their consistency is related evidence rather than wholly independent confirmation of the primary finding.

The registry record does not report whether secondary outcomes were ranked in advance, whether testing was adjusted for multiple comparisons, or how missing secondary measurements were handled. When many outcomes are examined, the chance of at least one apparently notable result can increase if each is tested without adjustment. For that reason, unadjusted secondary results generally carry less confirmatory weight than a prespecified primary analysis. The posted secondary pattern can help describe the range of outcomes examined, but it should not be treated as a set of independent definitive tests.

The responder outcome is a proportion rather than a mean, and the CGI-S outcome is a separate scale. Their planned presence does not mean results can be inferred from the continuous PANSS results. The registry record does not report their findings, so no conclusion about those outcomes can be drawn from the available entries.

Secondary outcomes registered for this trial, for reference (only those above were recalculated):

• Change From Baseline in Positive and Negative Syndrome Scale (PANSS) - Positive Subscale Score
• Change From Baseline in Positive and Negative Syndrome Scale (PANSS) - Negative Subscale Score
• Change From Baseline in Positive and Negative Syndrome Scale (PANSS) - General Psychopathology Subscale Score
• Proportion of Responders (≥30% Decrease in Total Positive and Negative Syndrome Scale [PANSS])
• Change From Baseline in Clinical Global Impression Scale (CGI-S)

9. Statistical Interpretation

10. Statistical Methods Explained

What does a mean change represent?

A continuous outcome can be summarized by its mean, the average of the observed values in a group. Here, each value is change in total PANSS score from baseline to 6 weeks. A negative group mean represents an average decrease from baseline, and a positive mean represents an average increase. The mean is a compact description, but participants’ changes can vary around it. The reported standard deviation describes that spread within a group; it is not the standard error or a confidence interval for the mean.

The posted paliperidone and placebo means are -9.1 and 3.8. The simple unadjusted comparison is the paliperidone mean minus the placebo mean, reported as -12.9. The negative sign indicates that the paliperidone group’s mean change was lower than the placebo group’s mean change. This contrast is not the same statistical quantity as an adjusted ANCOVA estimate, even though both address the group comparison.

What does the sponsor’s ANCOVA do?

ANCOVA is a model-based method for comparing group outcomes while accounting for one or more covariates included in the model. In general, covariate adjustment can change the estimate and its precision compared with a raw comparison of group means. The sponsor reports a least-squares mean difference of -12.7, a 95% confidence interval from -17.16 to -8.25, and p < 0.001 for the primary comparison. The particular covariates and model specification used are not reported in the registry record, so the adjusted estimate cannot be reconstructed from the group summaries alone.

The unadjusted recalculation uses the posted group means, sample sizes and standard deviations to estimate a difference and its uncertainty. It is a straightforward summary-statistic comparison, not a replacement for the trial’s own model. Because the methods differ, small differences between the unadjusted and sponsor estimates are expected to be possible. The posted record does not indicate that the sponsor’s reported analysis has been superseded.

How should a confidence interval be read?

A confidence interval conveys uncertainty around an estimate under the assumptions of the analysis. It gives a range of values compatible with the data and method at the stated confidence level; it is not a range containing a fixed proportion of individual participant outcomes. The sponsor’s interval for the primary ANCOVA estimate and the unadjusted interval are both below zero. That indicates that, under each method, the data are inconsistent with a zero difference at the corresponding confidence level. The interval width also matters: it describes precision, not clinical importance.

A p-value is a measure of how incompatible the observed data are with a specified null hypothesis under the statistical model. The reported p-values are small, but they do not measure the size or clinical importance of the difference, nor do they give the probability that the null hypothesis is true. The registry record does not provide a fuller account of the testing procedure or any multiplicity adjustment for the set of outcomes.

Why can variability affect precision?

General statistical explanation: when participant outcomes vary widely within groups, it is harder to estimate group means and their difference precisely. Standard errors describe estimated sampling uncertainty and are distinct from standard deviations, which describe observed spread. The posted recalculation uses pooled standard deviations for its comparisons; the record indicates that these were not approximated. The primary comparison has a pooled standard deviation of 18.64 for the recalculation. The smaller olanzapine group also has a less precise group mean than the placebo group in the posted summaries, as reflected in the reported standard errors.

General statistical explanation: over-dispersion occurs when data vary more than a model’s assumed distribution allows. It can make model-based uncertainty estimates too small if not accounted for. The primary endpoint is analyzed as a continuous mean, and no over-dispersion adjustment is reported. It would not be appropriate to infer an over-dispersion problem from these summaries.

What can the reported population tell us?

The primary outcome population is described as a Full Analysis Set excluding subjects who did not receive test drug or did not have post-treatment efficacy data. The definition is incomplete in the registry record. Such exclusions can affect the analyzed population and may limit how closely the analysis reflects all randomized participants. The registry record does not report the numbers excluded for each reason, the number with missing endpoint data, or the method used to handle missing values. It is therefore not possible to assess the impact of missing outcome data from this description alone.

In general, an intention-to-treat analysis aims to preserve randomized group assignment, while a per-protocol analysis restricts analysis according to adherence or other criteria. The population description is not sufficient to classify it as either approach. The specific handling of treatment discontinuation, protocol deviations, and missing measurements is not reported in the registry record. These details matter because estimates can depend on which participants contribute data and how incomplete measurements are addressed.

What does the olanzapine comparison establish?

The registry includes an olanzapine reference arm and a comparison with placebo. The stated purpose was to examine paliperidone’s clinical position; the record also contains an incomplete phrase concerning whether superiority or non-inferiority of paliperidone ER 6mg to olanzapine 10mg was to be verified. The criteria, hypothesis, and full analysis plan for such a direct active-treatment comparison are not reported in the registry record. A comparison of each arm with placebo does not itself test whether the two active treatments differ from one another. Those are distinct statistical questions.

11. Confidence Intervals and Precision

Precision is conveyed by the confidence interval around an estimate, as well as by the amount of variability and information in the data. For the primary comparison, the sponsor reports an ANCOVA least-squares mean difference of -12.7, with a 95% confidence interval from -17.16 to -8.25. The interval lies below zero. It indicates that the sponsor’s model estimates a difference favoring a lower mean change score in the paliperidone group, while allowing uncertainty across the range of values in that interval.

The unadjusted recalculation is -12.9, with a 95% confidence interval from -17.33 to -8.471. This interval is also below zero and is similar in location to the sponsor’s interval. The two estimates should not be treated as identical: one is the reported ANCOVA result, and the other is an unadjusted calculation from group summaries. The registry record does not describe the ANCOVA covariates, which prevents a more detailed explanation of why the estimates differ.

A confidence interval is not a guarantee that the true effect lies in the particular interval, and it does not indicate how much participants’ outcomes overlap. Nor does a narrow or statistically separated interval establish clinical importance. Assessing importance would require context, including a clinically meaningful threshold, which is not reported in the registry record.

The arm-level standard deviations indicate substantial variation in observed change scores. The primary outcome standard deviations are 18.39 for paliperidone and 18.88 for placebo. The sample sizes for those groups are 134 and 138, respectively. Precision reflects both variability and the amount of information, along with the analysis method. The olanzapine arm has a smaller posted primary outcome sample size of 46; its comparison should be read with that group size in mind. The reported intervals describe uncertainty in mean contrasts, not the range of individual responses.

12. Planning a Similar Trial: Sample Size

This is a planning illustration, not a re-evaluation of the original trial. It asks: if a new study wanted to detect a difference like the one observed, or a smaller one, how large would it need to be? It is deliberately not “observed power” of the finished trial, which is not a meaningful quantity.

The sample-size illustrations express the number of participants per arm for the continuous-outcome comparisons. They are not total-event counts. The planning assumptions listed in the registry analysis are a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. Each illustration uses an effect to detect set at a stated fraction of the observed effect: the full observed effect, 75% of the observed effect, or 50% of the observed effect.

The 80% power column gives the illustrated number of participants per arm for a design with an 80% chance of detecting the specified effect under those assumptions, if that effect and the other assumptions hold. The 90% power column expresses the corresponding planning target for a 90% chance. These values are planning illustrations based on the observed effect, not the original trial’s design justification. The trial’s own sample-size justification and original planning assumptions are not reported in the registry record.

For each comparison, the rows consider progressively smaller effects relative to the observed one. Smaller effects require more participants per arm to reach the same stated power under the listed assumptions. This is a general feature of statistical planning: a smaller signal is harder to distinguish from random variation without more information. The calculations also assume equal group allocation for the illustration, which should not be mistaken for a description of the trial’s actual allocation ratio.

These figures are not observed power. They do not measure the probability that the completed study “had power” after its results were known. They are conditional planning calculations based on the observed effect and stated assumptions. They do not account for design choices or analysis details that are not specified in the registry record, and should not be read as evidence about why the original enrollment was selected.

13. How the Sample Size Is Derived

Power is the probability that a study of a given size rejects the null hypothesis when a specific true effect exists. It depends on four things: the effect to detect, the variability of the outcome, the significance level (here a two-sided α of 0.05, so z_1−α/2 = 1.960) and the desired power (z_1−β = 0.842 for 80% and 1.282 for 90%).

Formula

n is the number of participants in each arm, Δ is the difference in means to detect and σ is the common standard deviation.

Worked calculation for this trial

1. Observed difference Δ = -12.900; pooled standard deviation σ = 18.640.
2. n = 2 × (1.96 + 0.842)² × 18.640² / -12.900² = 32.8, rounded up to 33 participants per arm for 80% power.

The smaller-effect rows in the table above repeat the same calculation with the effect multiplied by 0.75 and 0.5 (for ratios, applied to the logarithm of the ratio).

R code

alpha <- 0.05
z_a   <- qnorm(1 - alpha / 2)          # two-sided
delta <- -12.9                      # observed difference in means
sigma <- 18.6402394377405                     # pooled standard deviation
n_per_arm <- function(power, frac = 1) {
  ceiling(round(2 * (z_a + qnorm(power))^2 * sigma^2 / (delta * frac)^2, 6))
}
for (f in c(1, 0.75, 0.5))
  cat(sprintf("%3.0f%% of observed effect: n per arm = %s (80%% power), %s (90%% power)\n", 100 * f,
              format(n_per_arm(0.80, f), big.mark = ","), format(n_per_arm(0.90, f), big.mark = ",")))

# Check against R's built-in function (t-based, so very slightly larger)
cat("power.t.test:", round(power.t.test(delta = delta, sd = sigma, power = 0.80)$n, 1), "\n")

# Simulation: exact two-sample t-test using simulated means and variances
set.seed(396565)
n    <- n_per_arm(0.80)
reps <- 20000
m_diff <- rnorm(reps, delta, sigma * sqrt(2 / n))             # difference in sample means
s2     <- sigma^2 * rchisq(reps, 2 * (n - 1)) / (2 * (n - 1))   # pooled sample variance
tstat  <- m_diff / sqrt(s2 * 2 / n)
cat(sprintf("Simulated power at n = %s per arm: %.3f (%d replicates)\n", format(n, big.mark = ","), mean(abs(tstat) > qt(0.975, 2 * (n - 1))), reps))

Output from running the code

100% of observed effect: n per arm = 33 (80% power), 44 (90% power)
 75% of observed effect: n per arm = 59 (80% power), 79 (90% power)
 50% of observed effect: n per arm = 132 (80% power), 176 (90% power)
power.t.test: 33.8 
Simulated power at n = 33 per arm: 0.789 (20000 replicates)

The simulation draws many sets of group means and variances at the planned sample size and counts how often the two-sample t-test rejects the null. It should land close to the target power; at very small required sample sizes it can fall below it, because the normal approximation is rough there.

14. Analyses Reported by the Sponsor

The sponsor’s primary result is an ANCOVA least-squares mean difference of -12.7, with a 95% confidence interval from -17.16 to -8.25 and p < 0.001. The recalculated unadjusted difference from the posted means is -12.9, with a 95% confidence interval from -17.33 to -8.471 and p < 0.001. Both estimates are negative, and both intervals lie below zero.

The recalculation is not a reproduction of the sponsor model. It uses the arm-level summary statistics to make a simple unadjusted comparison; the ANCOVA result is adjusted according to a model whose covariates are not reported in the registry record. The estimates are close in value, but the sponsor’s ANCOVA is the analysis reported by the trial record. The registry record does not flag a discrepancy between the two results.

The sponsor analysis therefore supplies the trial’s principal inferential result, while the recalculation offers a transparent check based on the posted summaries. Neither estimate alone establishes clinical importance, and the registry’s incomplete description of the population and missing-data methods limits assessment of how robust the result is to those analysis choices.

15. Multiplicity, Missing Data and Analysis Populations

The registry record identifies one primary outcome and lists several secondary outcomes. It does not report a multiplicity-control procedure, a hierarchy of hypotheses, or whether secondary analyses were intended to be confirmatory or exploratory. When multiple outcomes or comparisons are tested, unadjusted p-values do not account for the increased opportunity for a chance finding across the family of tests. The primary comparison therefore has a different interpretive role from the secondary analyses, and the latter should be read more cautiously in the absence of a reported adjustment plan.

The record reports the primary outcome population as a Full Analysis Set excluding participants who did not receive test drug or lacked post-treatment efficacy data, but the definition is truncated. It does not report the number of randomized participants excluded from the primary analysis, the reason-specific exclusions, or the number with missing 6-week PANSS data. The posted group sizes for the primary summaries are available, but they do not fill in those missing accounting details.

Missing outcome data can affect a comparison if the chance of missingness is related to outcomes or differs across groups. General statistical explanation: methods such as complete-case analysis, imputation, and likelihood-based modeling make different assumptions about missing measurements. The registry record does not describe which missing-data method was used, whether sensitivity analyses were undertaken, or how treatment discontinuation was handled. The effect of missing data cannot therefore be assessed from the record alone.

The Full Analysis Set description is also insufficient to establish that the analysis followed a particular standard population definition. Details about protocol deviations, adherence requirements, and treatment assignment in the analysis are not reported in the registry record. An interim-analysis plan and any associated adjustment to statistical testing are likewise not reported. These gaps do not change the numerical results posted, but they constrain interpretation of how those results were planned and how robust they may be to alternative analysis choices.

16. 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 serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: schizophrenia (2.2%), mood altered (2.1%), pneumonia (0.7%).

The registry record reports serious adverse event counts for the paliperidone, placebo, and olanzapine groups. The recorded counts are 4 among 134 at risk for paliperidone, 1 among 138 at risk for placebo, and 2 among 47 at risk for olanzapine. These are event-participant counts as presented in the record; they do not by themselves describe timing, duration, severity beyond the serious-event classification, or causal attribution.

The record also lists serious event terms with percentages: schizophrenia at 2.2%, mood altered at 2.1%, and pneumonia at 0.7%. It does not clarify the group attribution for these listed terms in the information recorded here, so they should not be assigned to a particular arm. Death counts are not reported in the registry record.

Other adverse event counts are recorded as 109 among 134 at risk for paliperidone, 110 among 138 at risk for placebo, and 40 among 47 at risk for olanzapine. The registry record does not provide further detail in these entries about event severity, recurrence, or exposure-adjusted rates. These figures describe what was posted, not a complete comparative safety assessment; differences in follow-up and event definitions are not detailed in the registry record.

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

The posted primary result shows that the paliperidone group had a lower mean change in total PANSS score than the placebo group at 6 weeks. The sponsor’s ANCOVA estimate is -12.7, and its 95% confidence interval lies below zero. The unadjusted recalculation has the same direction. This is evidence about a group-average difference on the specified scale and follow-up, not a claim about every participant.

The result does not establish how long the measured difference persisted, because the primary endpoint is reported at baseline and 6 weeks. It does not show whether the change crossed a clinically important threshold; no such threshold is reported in the registry record. The score is a symptom-scale measure, and the numerical estimate should be interpreted within that measurement framework rather than as a direct description of each person’s functioning or experience.

The record also does not establish a direct paliperidone-versus-olanzapine treatment difference. Olanzapine was included as a reference drug, and the record’s description of the intended active-treatment comparison is incomplete. Showing that two separate comparisons with placebo have estimates in a particular direction does not answer whether the active treatments differ from each other.

The results cannot resolve the impact of missing data, because the population definition is incomplete and details of missing-data handling are not reported. Nor do the primary and secondary p-values alone address multiplicity or prove clinical importance. The posted secondary PANSS results are directionally consistent, but their unadjusted status and relationship to the primary scale matter when weighing them.

Finally, safety counts should be interpreted descriptively. The record reports serious adverse events and other adverse event counts, but the available entries do not supply a complete account of event timing, attribution, or comparative safety methods. The statistical results describe the recorded outcomes and do not provide individual medical advice.

18. Why This Trial Matters Statistically

This trial illustrates the distinction between a sponsor-reported model-based result and a recalculation from group summaries. ANCOVA reports a least-squares mean difference, whereas the recalculation is an unadjusted mean contrast. Their close estimates and shared direction provide a useful statistical cross-check, while the distinction in methods remains important for transparent interpretation.

The design also includes an active reference arm alongside placebo. This highlights a common interpretive issue: comparisons of each active arm with placebo do not substitute for a direct comparison between active treatments. The registry’s incomplete description of the intended paliperidone-olanzapine question makes it especially important not to infer a head-to-head conclusion from separate placebo contrasts.

The record gives a primary outcome and several secondary outcomes, but does not report a multiplicity plan or complete missing-data methods. These omissions show why statistically small p-values should be considered with the analysis plan, outcome hierarchy, and population definition rather than in isolation. The record’s partial definition of the Full Analysis Set also underlines how population details affect reproducibility.

Lastly, the sample-size illustrations use observed effects and explicit planning assumptions, while the trial’s own sample-size justification is absent. Such illustrations can show how assumed effect size relates to planning needs, but they cannot reconstruct the original design rationale. Together, these features make the study useful for examining model choice, confidence intervals, secondary testing, and the limits of registry-level reporting.

19. Limitations

• The posted summary statistics are the only data these results rest on; patient-level data, full protocols and statistical analysis plans are not available.
• Recalculated results are simple, unadjusted calculations from posted summaries. They do not replace the sponsor's prespecified analysis.

The registry record does not report the original sample-size justification, its intended effect size, or the assumptions used to select enrollment. The posted planning illustrations are separate calculations based on observed effects and should not be mistaken for the trial’s own design calculations.

The primary population description is truncated after the stated exclusions, and the record does not report the number or reasons for exclusions, missing outcome counts, or missing-data methods. The primary ANCOVA is named, but its covariates and other model details are not reported. These omissions limit independent reconstruction of the sponsor analysis and assessment of sensitivity to analytic choices.

The record does not report a multiplicity strategy for the secondary outcomes, nor does it provide results for the responder and CGI-S outcomes despite listing them. The incomplete description of the intended paliperidone-olanzapine comparison does not support a head-to-head conclusion. An interim-analysis plan is not reported in the registry record.

For safety, the listed serious event terms are not clearly attributed to treatment groups in the available record, and deaths are not reported. The adverse event entries do not provide detailed timing, exposure, or event attribution. Follow-up for the primary endpoint is limited to the stated baseline and 6-week assessment, so the record does not establish longer-term outcome patterns.

20. Related Tutorials

21. Related Statistical Calculators

22. Sources

• ClinicalTrials.gov: A Study to Evaluate Effectiveness and Safety of ER OROS Paliperidone in Patients With Schizophrenia — NCT00396565.
• Registry record last updated 2014-04-29; status last verified 2014-04.
• Results first posted 2012-07-03.
• 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.

23. Record Summary

The registry record describes a completed randomized, double-blind, parallel-group, phase 3 study of extended-release OROS paliperidone in patients with schizophrenia, with placebo as the main comparator and olanzapine as a reference drug. Overall enrollment was 394. The primary endpoint was change from baseline in total PANSS score at 6 weeks.

The sponsor reported an ANCOVA least-squares mean difference of -12.7 for paliperidone versus placebo, with a 95% confidence interval from -17.16 to -8.25 and p < 0.001. An unadjusted recalculation from posted group summaries was -12.9, with a 95% confidence interval from -17.33 to -8.471 and p < 0.001. The record also includes three posted PANSS subscale results, adverse event counts, and incomplete details on population, missing data, multiplicity, and the active-treatment comparison.