Safety and Efficacy of Cariprazine in Schizophrenia: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Safety and Efficacy of Cariprazine in Schizophrenia (NCT01104779), a randomized phase 3 study. The primary outcome is Measurement of Schizophrenia Symptoms: Change From Baseline in Positive and Negative Syndrome Scale (PANSS) Total Score.

2. Clinical Question

3. Trial Design

The registry identifies NCT01104779 as a completed Phase 3, randomized, triple-masked, parallel-group trial titled Safety and Efficacy of Cariprazine in Schizophrenia. Its official title describes a double-blind, placebo-controlled evaluation during acute exacerbation of schizophrenia. The sponsor was Forest Laboratories. The registry lists enrollment of 446, a study start date of 2010-04-27, and primary completion on 2011-12-15.

Participants were assigned to placebo, cariprazine 3-6 mg/day, or cariprazine 6-9 mg/day. Both active arms are described as once-daily, fixed-flexible dosing, with the lower and higher dose ranges designated as low dose and high dose respectively. The registry record does not report the allocation ratio, the method used to generate the randomization sequence, or the identities of the roles covered by triple masking.

The primary outcome was change from baseline to Week 6 in PANSS total score. The outcome population is defined as the Intent-to-Treat Population, consisting of patients in the Safety Population who had at least one postbaseline PANSS total-score assessment. Outcome counts were 145 for placebo, 147 for cariprazine 3-6 mg/day, and 147 for cariprazine 6-9 mg/day. These are outcome-analysis counts, not a substitute for details about all participants randomized or the handling of participants without an available postbaseline measurement.

The registry also lists change in CGI-S from baseline to Week 6 as a secondary outcome. It posts two outcome measures and four analyses. The record does not report the original sample-size justification, detailed eligibility criteria, treatment adherence, follow-up beyond the stated outcome timeframe, or a full statistical analysis plan. Those omissions limit how completely the design and analysis decisions can be reconstructed from 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.

Before assignment: Randomization and treatment assignment were based on a randomization scheme prepared by Allergan Biostatistics prior to the start of the study. No-drug washout period of up to 7 days.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Intent-to-Treat Population, consisting of all patients in the Safety Population who had at least one postbaseline assessment of the PANSS total score.

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.

Population: The randomized population is also the safety population.

6. Primary Endpoint

The registry lists the primary outcome as Measurement of Schizophrenia Symptoms: Change From Baseline in Positive and Negative Syndrome Scale (PANSS) Total Score, measured over Baseline to Week 6.

The primary endpoint is change from baseline to Week 6 in total score on the Positive and Negative Syndrome Scale, or PANSS. The scale was developed to assess positive and negative symptom syndromes in people with schizophrenia. Its assessment uses a structured clinical interview and supporting clinical information from family, hospital staff, or other reliable informants. It covers 9 clinical domains, including positive syndrome, negative syndrome, depression, a composite index, and general psychopathology.

PANSS has 30 items, each scored on a 7-point scale from 1 to 7. The cumulative score ranges from 30 to 210. A negative change score indicates improvement, so a more negative group mean represents a larger average decrease from baseline. The endpoint is useful for statistical comparison because it expresses symptom assessment as a continuous score, allowing the analysis to compare average changes between groups rather than only classifying participants into categories.

The registry reports least-squares means with standard errors for each group. The least-squares mean is a model-based estimate of an average; the record does not report the model details needed to identify all adjustments underlying those estimates. The endpoint describes group averages over the stated timeframe. It does not by itself show the distribution of individual responses, the proportion meeting a response threshold, or the clinical meaning of a particular score change. Those matters are not reported in the registry record.

7. Results Posted in the Registry

• Cariprazine (3-6 mg/Day): mean -22.80 versus -16.00 on control; difference -6.80 (95% CI -11.23 to -2.37), standardized effect d = -0.35; z-test p = 0.003.
• Cariprazine (6-9 mg/Day): mean -25.90 versus -16.00 on control; difference -9.90 (95% CI -14.48 to -5.32), standardized effect d = -0.50; 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.

Measurement of Schizophrenia Symptoms: Change From Baseline in Clinical Global Impression-Severity (CGI-S)

• 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 errors.
• Result:
  • Cariprazine (3-6 mg/Day): mean -1.40 versus -1.00 on control; difference -0.40 (95% CI -0.68 to -0.12), standardized effect d = -0.33; z-test p = 0.005.
  • Cariprazine (6-9 mg/Day): mean -1.60 versus -1.00 on control; difference -0.60 (95% CI -0.88 to -0.32), standardized effect d = -0.50; z-test p = <0.001.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the mean is statistically lower for Cariprazine (3-6 mg/Day) 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 Cariprazine (6-9 mg/Day) than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

The registry lists one secondary symptom outcome: change from baseline to Week 6 in Clinical Global Impression-Severity, or CGI-S. The posted group means follow the same overall direction as the primary PANSS outcome: change is negative in each group, with larger average decreases in both cariprazine arms than in placebo. The sponsor’s reported comparisons and the summary-statistic recalculations also point in that direction. This broad agreement across the two symptom measures provides a related, but not independent, perspective on the primary pattern because both outcomes concern schizophrenia symptoms over the same timeframe.

CGI-S is a separate scale from PANSS, so the two outcomes should not be treated as interchangeable or as measuring identical information. The registry record identifies the secondary outcome and reports its group summaries and comparisons, but it does not provide a detailed description of how CGI-S was assessed or a threshold for clinically meaningful change. The secondary findings therefore add context about the pattern of symptom-score changes rather than resolve questions about clinical importance.

Secondary results generally carry less confirmatory weight when the record does not establish how they fit into a prespecified testing hierarchy or whether multiplicity was controlled. Multiple comparisons can increase the chance of nominally small p-values by chance. The registry record does not report a multiplicity adjustment, a hierarchy for primary and secondary testing, or a complete analysis plan. The secondary findings are best read as supportive descriptive evidence, not as a separate definitive confirmation that bypasses those limitations.

The registry posts only one secondary outcome, and no additional secondary outcome results should be inferred from the count of analyses. Agreement in direction across PANSS and CGI-S does not establish that all symptom domains, other outcomes, or longer-term results would show the same pattern. Those data are not reported in the registry record.

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

• Measurement of Schizophrenia Symptoms: Change From Baseline in Clinical Global Impression-Severity (CGI-S)

9. Statistical Interpretation

10. Statistical Methods Explained

Several statistical ideas help distinguish what the primary results estimate from what they cannot establish. The endpoint is continuous: PANSS total score can take values across a numerical scale, and the outcome is the change from baseline. The registry reports least-squares means, which are model-based group averages, and standard errors, which describe uncertainty in those estimated averages. A standard error is not a measure of how much individual participants’ scores vary. The registry record does not report the full model specification or the covariates used for the least-squares means.

What does the ANCOVA comparison represent?

The sponsor reports analyses using ANCOVA, short for analysis of covariance. In general, ANCOVA compares group outcomes while accounting for one or more covariates. The specific covariates, model diagnostics, and missing-data procedures for this trial are not reported in the registry record, so the reported method name alone does not reconstruct the complete analysis. The outcome entries label the reported parameter as a mean difference in final values. The registry’s group summaries, by contrast, describe least-squares mean change from baseline. These labels should be read as they are reported rather than assumed to be interchangeable descriptions of every model detail.

The sponsor’s reported mean differences are -6.8 for cariprazine 3-6 mg/day versus placebo and -9.9 for cariprazine 6-9 mg/day versus placebo. In the context of a change score where a negative value indicates improvement, a negative between-group difference means a larger average decrease in the active group than in placebo. The 95% confidence intervals and p-values reported alongside those estimates come from the sponsor’s ANCOVA results.

What do the recalculations add?

The posted recalculations are simple, unadjusted comparisons derived from the summary statistics. They are not a reproduction of ANCOVA. For PANSS, the recalculated mean differences are -6.8 and -9.9 for the lower- and higher-dose comparisons, respectively. The corresponding approximate 95% intervals are -11.23 to -2.365 and -14.48 to -5.324. The recalculated p-values are 0.003 and < 0.001. Because the recalculation uses approximate standard deviations inferred from the group standard errors, it is a check on scale and direction rather than a substitute for the sponsor’s model-based analysis.

In general, a confidence interval gives a range of estimates compatible with the data and the statistical procedure, under its assumptions. It conveys uncertainty about the estimated group contrast; it does not imply that values inside the range are equally likely, nor does it express the range of individual outcomes. Here, the sponsor’s reported 95% intervals for both primary comparisons lie below zero. The registry record calls both analyses superiority analyses. A p-value describes how unusual data at least as incompatible with a specified null comparison would be under the test’s assumptions. It is not the probability that the null hypothesis is true, and it does not measure clinical importance.

Why distinguish standard error from standard deviation?

A standard deviation describes dispersion among individual observations, while a standard error describes uncertainty in an estimated quantity such as a mean. A mean can have a small standard error even when individual outcomes vary substantially, because the standard error also depends on the amount of information contributing to the estimate. The posted arm-level PANSS standard errors are 1.6, 1.6, and 1.7. The recalculated pooled standard deviations are marked as approximate, so inferences relying on those values should be understood as approximations. The registry record does not provide raw participant-level PANSS data for an independent reanalysis.

General statistical caution: model-based estimates and simple unadjusted calculations can differ because they may use different assumptions, covariates, variance estimates, or analysis populations. Similarity between the recalculated and sponsor-reported results in this trial is a useful consistency check, not proof that the methods are equivalent. The original reported ANCOVA analysis is the appropriate reference for what the sponsor concluded under its stated method.

How should multiple comparisons be considered?

The record includes two primary active-versus-placebo comparisons and a secondary outcome with its own two active-versus-placebo analyses. Testing several outcomes or contrasts can increase the chance of observing a nominally small p-value even when no corresponding effect exists. In general, multiplicity procedures control or account for that family of tests, but they can change which results are considered statistically persuasive. The registry record does not report a multiplicity adjustment or a prespecified testing hierarchy. Accordingly, the reported p-values should be interpreted as the results of the listed analyses, without assuming that family-wise error was controlled.

The same restraint applies to analysis populations and missing outcomes. The primary population definition requires at least one postbaseline PANSS assessment and refers to the Safety Population. The registry does not report the number lacking such an assessment, the pattern or reasons for missingness, or the statistical method used to address missing values. No method can be inferred solely from the term Intent-to-Treat Population when the operational definition is stated this way. These details matter because analysis populations and missing-data assumptions can affect the estimate and its uncertainty.

In sum, the reported ANCOVA estimates are the sponsor’s primary inferential results, while the recalculations provide approximate unadjusted context. Neither set of summaries provides individual participant outcomes. The registry record does not report enough detail to independently reproduce the full model, evaluate all assumptions, or determine how multiplicity and missing observations were handled.

11. Confidence Intervals and Precision

The sponsor reports a 95% confidence interval from -11.3 to -2.4 for the PANSS mean difference comparing cariprazine 3-6 mg/day with placebo. For cariprazine 6-9 mg/day versus placebo, the 95% interval is -14.5 to -5.3. Both intervals are entirely below zero, indicating that the reported ANCOVA estimates and their intervals are consistent with a greater average PANSS decrease in the active arm than in placebo, given the analysis and its assumptions.

In general, a confidence interval communicates uncertainty around an estimated contrast. It is not a prediction interval for individual participants, and it does not establish that every plausible value within it has the same clinical importance. These intervals do not answer how much change patients would regard as meaningful; the registry record does not report a threshold for that interpretation.

The recalculated, approximate intervals are -11.23 to -2.365 for the lower-dose comparison and -14.48 to -5.324 for the higher-dose comparison. These unadjusted recalculations use approximate variability derived from posted summaries. Their extra decimal places reflect the calculated interval bounds, not greater certainty than the sponsor’s ANCOVA results. The sponsor’s reported intervals are the direct registry analyses and are therefore the main reference for precision.

The intervals give a more informative sense of uncertainty than p-values alone, but they remain conditional on the analysis method, population, and assumptions. The registry record does not report details such as model covariates, diagnostics, or missing-data handling that would allow a complete assessment of those conditions. Precision for average differences also does not describe how widely individual changes vary.

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 planning illustrations express required sample size as participants per arm for the two-arm comparison, not as a total across the trial. They use a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. The target effect is a stated fraction of the observed effect. The original trial’s own sample-size assumptions and justification are not reported in the registry record.

The 80% power column describes the planning number per arm for a design with an 80% chance of detecting the specified effect under the stated assumptions, if that effect and the model assumptions are correct. The 90% power column gives the corresponding planning number per arm for a 90% chance. These are design calculations, not guarantees about a future trial’s result. The listed figures are planning illustrations based on observed effects, not observed power.

For example, planning around 100% of the observed PANSS effect gives 127 participants per arm for 80% power and 170 participants per arm for 90% power for the comparison with cariprazine 3-6 mg/day. For the cariprazine 6-9 mg/day comparison, the corresponding figures are 64 and 86 participants per arm. When planning for 50% of the observed effect, the figures rise to 508 and 680 participants per arm for the lower-dose comparison, and 256 and 342 for the higher-dose comparison. The 75% rows fall between these effect assumptions, with their values listed in the planning display.

Smaller effects require more participants to distinguish them from random variation under otherwise similar assumptions. These illustrations should not be read as evidence about the trial’s actual design rationale or as a recommendation for a new study. They are conditional planning scenarios, and the effect observed in a trial can be an uncertain basis for planning future research.

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 = -6.800; pooled standard deviation σ = 19.333.
2. n = 2 × (1.96 + 0.842)² × 19.333² / -6.800² = 126.9, rounded up to 127 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 <- -6.8                      # observed difference in means
sigma <- 19.3333301625047                     # 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(1104779)
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 = 127 (80% power), 170 (90% power)
 75% of observed effect: n per arm = 226 (80% power), 302 (90% power)
 50% of observed effect: n per arm = 508 (80% power), 680 (90% power)
power.t.test: 127.9 
Simulated power at n = 127 per arm: 0.791 (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 reports ANCOVA superiority analyses for both primary comparisons. For cariprazine 3-6 mg/day versus placebo, the reported mean difference was -6.8, the 95% confidence interval was -11.3 to -2.4, and p = 0.003. For cariprazine 6-9 mg/day versus placebo, the reported mean difference was -9.9, the 95% confidence interval was -14.5 to -5.3, and p < 0.001. These are the registry’s own analysis results.

The recalculated comparisons use posted group summaries to make approximate, unadjusted comparisons. For the primary outcome, they give mean differences of -6.8 and -9.9, with p-values of 0.003 and < 0.001. The estimated direction and magnitude are close to the sponsor’s reported results. Small differences in interval endpoints or p-value formatting are expected when approximate summary-statistic calculations are compared with model-based ANCOVA results.

The recalculation is not an independent reproduction of the sponsor’s model: it does not use participant-level data or the unreported details of the ANCOVA specification. Its value is as a consistency check on the posted summaries, not as an alternative result. The sponsor’s reported analysis is the relevant source for the trial’s stated inferential findings, while the recalculation makes clear that its assumptions and method are simpler and unadjusted.

15. Multiplicity, Missing Data and Analysis Populations

The registry reports two active-versus-placebo primary comparisons and two active-versus-placebo comparisons for the listed secondary outcome. When several comparisons are assessed, the chance of at least one apparently persuasive result can increase if each is judged without accounting for the full set. General methods for addressing multiplicity include adjusted testing thresholds or a prespecified sequence of hypotheses. The registry record does not report whether either approach was used, and it does not describe a testing hierarchy. The posted p-values should therefore be interpreted as reported for their individual analyses, without assuming that they account for all comparisons.

The record defines the primary analysis population as patients in the Safety Population who had at least one postbaseline PANSS assessment. This makes clear that an assessment after baseline was required for inclusion in the stated primary population, but it does not show how many participants lacked such an assessment or why. The registry record does not report the missing-data method, assumptions about missing outcomes, sensitivity analyses, or how incomplete observations were treated in ANCOVA. The operational definition alone is not enough to determine whether all randomized participants contributed to the primary analysis.

Outcome analysis counts were 145 for placebo and 147 for each cariprazine arm. The registry also lists enrollment of 446, but does not explain the relationship between enrollment, randomization, safety analyses, and each outcome analysis in enough detail to reconstruct participant flow. These counts should not be used to infer unreported exclusions or reasons for non-inclusion.

These reporting gaps matter because missingness and multiplicity can affect both estimates and how confidently results are interpreted. The record does not report the original statistical analysis plan, interim analyses, sample-size justification, or sensitivity checks. Their presence or absence cannot be inferred from the posted results. As a result, the reported estimates and p-values can be described accurately, but the registry information does not support a full audit of the analysis decisions or their robustness to alternative assumptions.

16. 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: schizophrenia (3.7%), psychotic disorder (2.4%), psychomotor hyperactivity (1.9%), hyponatraemia (0.6%), anxiety (0.6%).

The registry reports serious adverse events and other adverse events using separate at-risk and affected counts for each group. For serious adverse events, the posted counts are 13 affected among 161 at risk for placebo, 9 among 167 at risk for cariprazine 3-6 mg/day, and 4 among 169 at risk for cariprazine 6-9 mg/day. The listed deaths are 0 in each group. These are descriptive safety counts; they do not establish why an event occurred or whether it was caused by treatment.

For other adverse events, the record reports 54 affected among 147 at risk in placebo, 84 among 151 at risk for cariprazine 3-6 mg/day, and 86 among 148 at risk for cariprazine 6-9 mg/day. The serious-event and other-event at-risk denominators differ, so the two sets of counts should not be conflated. The registry record does not provide further information on event duration, severity beyond the serious classification, recurrent events, or individual-level overlap.

Listed serious-event terms include schizophrenia, psychotic disorder, psychomotor hyperactivity, hyponatraemia, anxiety, and psychogenic polydipsia. The record lists percentages for these terms but does not provide arm-specific attribution alongside this term listing. Safety counts require context about definitions, exposure, timing, and clinical review; the registry entries alone do not support causal conclusions.

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

The results show that, on average, PANSS total scores decreased from baseline to Week 6 in the placebo group and in both cariprazine groups. The decreases were larger on average in the cariprazine groups, and the sponsor’s ANCOVA comparisons were reported as superiority analyses with confidence intervals below zero. The findings describe the measured group differences for the stated outcome population, endpoint, and assessment period.

They do not mean that every participant improved, that the same change would occur for an individual, or that the observed mean difference is necessarily large enough to matter to patients. The registry record does not report a clinically meaningful threshold, responder proportions, patient-reported outcomes, or longer-term symptom results. A statistically persuasive group contrast and a clinically important benefit are related questions, but the reported statistics alone do not settle both.

The results also do not establish effects outside the trial’s studied setting or population. The record describes acute exacerbation of schizophrenia and outcomes through Week 6. The posted analyses cannot, by themselves, establish what would happen over a longer period, in a different population, or with a different clinical context. No medical advice follows from these group-level trial summaries.

Safety findings should be interpreted separately from the efficacy endpoint. Serious and other adverse events were reported with different at-risk counts, and the registry’s event summaries do not establish causation. The absence of reported deaths in the groups is a count in this record, not proof of no risk. A complete benefit-risk assessment would require details beyond the posted summary statistics, including the broader safety data and study context.

Finally, the posted numbers do not answer every question about validity. The original sample-size justification, full analysis plan, multiplicity handling, and missing-data procedures are not reported in the registry record. Those limitations do not erase the reported comparisons, but they define the scope of conclusions that can be drawn from them.

18. Why This Trial Matters Statistically

This trial illustrates why the estimated contrast, its uncertainty, and the analysis method should be read together. The group means show the direction of change, while the reported between-group estimates quantify how the average changes differed. Confidence intervals add information about precision that a p-value alone does not convey. For both primary comparisons, the sponsor’s reported intervals lie below zero, consistent with larger average PANSS decreases in the cariprazine groups than in placebo.

It also demonstrates the distinction between model-based sponsor analyses and simple summary-statistic recalculations. The recalculated comparisons are close to the reported estimates, which provides a useful check on the direction and scale of the results. But the calculations are unadjusted approximations and cannot replace the ANCOVA model or its unavailable details. Similar numerical answers do not make different methods identical.

The record also shows why planning figures must be separated from trial outcomes. The sample-size illustrations use fractions of the observed effect and stated generic planning assumptions. They are not evidence about the study’s original design rationale and are not observed power. Their main statistical lesson is that planning for a smaller target effect requires more participants per arm under the stated assumptions.

Finally, endpoint and population definitions matter. The primary analysis includes patients who had at least one postbaseline PANSS assessment, and the trial’s stated outcome period ends at Week 6. Missing-data handling, multiplicity, and the full analysis plan are not reported. A careful reading can therefore recognize the reported statistical pattern while keeping the conclusions specific to the posted outcomes, methods, and timeframe.

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.

Several trial-specific reporting limitations constrain interpretation. The registry record does not provide the detailed eligibility criteria, randomization procedure, allocation ratio, or the roles included in triple masking. Although the official title describes a double-blind, placebo-controlled evaluation and the design entry identifies triple masking, the record does not explain how masking was implemented or maintained.

The primary outcome population is defined as the Safety Population with at least one postbaseline PANSS assessment. The registry does not report participant flow by reason for exclusion or missing assessment, the number without postbaseline data, or the method used to handle missing values. It also does not report the complete ANCOVA specification, covariates, model diagnostics, or sensitivity analyses. These omissions limit independent evaluation of the sponsor’s estimates and assumptions.

The record does not report the sample-size justification for the original trial, an analysis hierarchy, multiplicity adjustment, or whether interim analyses occurred. The planning illustrations are separate from those missing design details. Their assumptions are stated for the illustrations, but should not be attributed to the original trial’s planning.

The primary and secondary outcomes cover change in symptom scales through Week 6. The registry record does not report a clinically meaningful threshold, responder analyses, longer-term outcomes, or detailed CGI-S assessment information. The adverse-event summaries do not provide full event narratives, timing, duration, or arm-specific attribution for the listed serious-event terms. These limitations do not invalidate the posted results, but they restrict conclusions about robustness, clinical importance, longer-term effects, and causal interpretation of safety events.

20. Related Tutorials

21. Related Statistical Calculators

22. Sources

• ClinicalTrials.gov: Safety and Efficacy of Cariprazine in Schizophrenia — NCT01104779.
• Registry record last updated 2018-11-14; status last verified 2018-10.
• Results first posted 2018-11-14.
• 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.

23. Record Summary

NCT01104779 is a completed Phase 3 randomized, triple-masked, parallel-group trial sponsored by Forest Laboratories. It evaluated once-daily cariprazine at 3-6 mg/day and 6-9 mg/day against placebo for acute exacerbation of schizophrenia. The registry lists enrollment of 446 and a primary outcome of PANSS total-score change from baseline to Week 6.

The sponsor reported larger average PANSS decreases in both cariprazine groups than in placebo, with ANCOVA confidence intervals below zero for both comparisons. A secondary CGI-S outcome showed the same overall direction. Serious and other adverse events were recorded separately, with different at-risk counts; causal attribution is not established by those counts.

The registry record does not report the original sample-size justification, multiplicity handling, detailed missing-data procedures, or the full analysis model. The planning illustrations are per-arm scenarios based on observed effects and stated assumptions, not observed power. Conclusions are limited to the posted outcomes, analysis population, and Week 6 timeframe.