Bioequivalence Study of Paroxetine and PAXIL Under Fasting Conditions in Healthy Mexican Participants: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Bioequivalence Study of Paroxetine and PAXIL Under Fasting Conditions in Healthy Mexican Participants (NCT04311463), a randomized phase 1 study. The primary outcome is Maximum Observed Plasma Concentration (Cmax) of Paroxetine.

2. Clinical Question

3. Trial Design

The registry describes a Phase 1, randomized, balanced, open-label, two-sequence, two-treatment, two-period crossover study. Participants received one oral dose of each product, with the treatment order differing by sequence. In a crossover design, each participant can contribute measurements under both treatments; the registry describes its pharmacokinetic analysis set as participants who completed both periods according to protocol criteria and had data available at the specified time points.

The study compared paroxetine 20 mg tablets designated Test A with PAXIL (paroxetine) 20 mg tablets designated Reference B. It was conducted under fasting conditions in healthy adult male and female participants. The brief summary states that a maximum of 38 participants would be randomized and dosed and that the expected study duration was 12 days, including a 7-day washout between dosing periods. The registry record lists enrollment of 38 and status as completed. The posted Cmax results include 31 participants in each treatment group.

The study was unmasked. The registry lists anxiety disorders among the conditions, but the described participants were healthy adults; the record does not explain the relationship between that condition entry and the study population. The primary outcome was maximum observed plasma concentration of paroxetine. The registry also lists outcomes concerning non-serious and serious adverse events and abnormal vital signs. Details such as the original sample-size rationale, planned statistical model, and any formal bioequivalence decision criteria are not reported in the registry record.

4. Participant Flow and Accounting

The registry reports how participants moved through each stage of the study. All counts below are exactly as posted; missing counts are not inferred from other stages.

Before assignment: Total 38 participants were enrolled in the study across one study center in Mexico.

Period 1 (Up to Day 3)

Reasons for not completing this period:

Washout Period (Up to Day 7)

Reasons for not completing this period:

Period 2 (Up to Day 3)

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Pharmacokinetic (PK) analysis set included participants who completed both period of the study as per protocol criteria. Only those participants with data available at the specified data points were analyzed.

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 Maximum Observed Plasma Concentration (Cmax) of Paroxetine, measured over Pre-dose (0 hour) and at 0.5, 1, 1.5, 2, 2.5, 3, 3.5, 4, 4.5, 5, 5.5, 6, 6.5, 7, 8, 9, 10, 12, 16, 24, 36, 48 and 72 hours post-dose.

The primary endpoint is maximum observed plasma concentration, or Cmax, of paroxetine, measured in nanograms per milliliter. It represents the highest observed concentration in the scheduled blood-sampling profile after dosing. The registry lists samples from pre-dose through 72 hours post-dose, with multiple scheduled times in between. It states that blood samples were collected at the indicated times and that pharmacokinetic parameters were analyzed using non-compartmental analysis.

Cmax is a pharmacokinetic measure used to characterize the peak concentration observed after a dose. Comparing it between the test and reference products addresses whether the observed peak concentration summaries differ in this study. It does not, by itself, describe every aspect of drug exposure or establish a regulatory bioequivalence conclusion. Such a conclusion depends on the prespecified criteria and analysis used, which are not reported in the registry record.

The reported endpoint is summarized as a mean with standard deviation for each product. The posted comparison expresses the difference between those means. The pharmacokinetic analysis set was restricted to participants who completed both periods according to protocol criteria and had the required data at the specified time points. The record does not report the number of participants with unavailable data at individual sampling times or the handling of such data.

7. Results Posted in the Registry

• Paroxetine 20 mg (Test A): mean 8.39 versus 8.83 on control; difference -0.44 (95% CI -3.54 to 2.66), standardized effect d = -0.07; z-test p = 0.781.

8. Statistical Interpretation

9. Statistical Methods Explained

The posted outcome is continuous: Cmax is measured as a concentration, and the registry reports a mean and standard deviation for each product. The comparison is expressed as a mean difference. These summaries answer different questions. A mean describes the central value in the group, while the standard deviation describes how dispersed individual observations are around that mean. Neither summary alone indicates how uncertain the comparison is; the confidence interval addresses uncertainty in the estimated difference.

What does the mean difference represent?

The reported comparison is Test A minus Reference B. Its estimate is -0.44 nanograms per milliliter. This is a difference between the two posted means, not an individual participant's treatment response. A negative estimate corresponds to a lower posted mean for Test A. The interval for the estimate runs from -3.538 to 2.658 nanograms per milliliter, so it spans values below and above zero. The registry also reports an unadjusted p-value of 0.781. A p-value measures how compatible the observed data are with a specified null hypothesis under the statistical procedure used; it is not the probability that the null hypothesis is true. The detailed procedure underlying this p-value is not reported in the registry record.

Why does the crossover design matter?

In a crossover study, participants receive more than one treatment in separate periods. A comparison can therefore use information from the same participant under both treatments, which may account for stable differences between participants. The order of treatment and changes between periods can also matter. The registry describes randomization, two sequences, two periods, and a washout, but does not report the analysis model, whether the comparison used within-participant pairing, or how period, sequence, or carryover effects were handled. It would be inappropriate to infer those details from the design description alone.

The registry states that the pharmacokinetic parameters were analyzed using non-compartmental analysis. This is a general pharmacokinetic approach that derives parameters from observed concentration-time data without specifying a compartmental model for the body's processes. That statement does not specify the inferential model used to compare Cmax between treatments. The analysis set is described as participants completing both periods according to protocol criteria, with data available at the specified time points.

What do standard deviations and standardized effects add?

The Test A standard deviation was 5.872 and the Reference B standard deviation was 6.554 nanograms per milliliter. Standard deviations reflect spread among observed values, not precision of the mean difference. The posted standardized effect, d = -0.071, places the difference in relation to pooled variability. Standardized effects can assist comparisons across measures with different units, but the original-unit estimate and its confidence interval are more directly interpretable for this Cmax outcome. Neither measure establishes whether a formal equivalence margin was satisfied.

Why can variability affect interpretation?

As a general statistical point, greater variation among observations can make a treatment contrast less precise, particularly when the available sample is limited. In some settings, data may show more variability than a simple model allows; this is called over-dispersion. A model that fails to account for it can understate uncertainty. The record does not report an over-dispersion assessment, so this is a general consideration rather than a claim about this study. The reported standard deviations and confidence interval convey the observed spread and uncertainty without identifying their cause.

The registry does not report a prespecified statistical analysis plan, model details for the treatment contrast, transformations, assumptions checks, or rules for handling protocol deviations. These missing details limit the ability to reconstruct the analysis or assess whether it matches a particular regulatory bioequivalence method. The numerical comparison remains interpretable as the posted estimate and interval, but its inferential context is incomplete.

10. Confidence Intervals and Precision

The 95% confidence interval for the mean Cmax difference, Test A minus Reference B, extends from -3.538 to 2.658 nanograms per milliliter. A confidence interval gives a range of values reasonably compatible with the observed data under the method used to calculate it. In repeated use of the same procedure under its assumptions, intervals constructed this way would cover the target parameter at the stated rate. It is not a range containing a fixed proportion of individual participants' values.

Because the interval includes zero and extends in both directions, the estimate does not sharply distinguish a lower mean for Test A from no mean difference or a higher mean. The point estimate is -0.44 nanograms per milliliter, but the interval is much wider than that point estimate. The confidence interval therefore communicates substantial uncertainty about the size and direction of the mean difference. The reported standard deviations, 5.872 for Test A and 6.554 for Reference B, describe variation among participant measurements and help contextualize the uncertainty, but are not themselves confidence limits.

Precision is not the same as clinical or regulatory importance. A confidence interval can be interpreted against a prespecified equivalence margin only if that margin and the relevant analysis are known. Neither is reported in the registry record. The interval should therefore be read as uncertainty around the posted mean difference, not as proof that the products are equivalent or non-equivalent. The unadjusted p-value of 0.781 likewise does not convey the range of plausible effect sizes as directly as the interval does.

11. 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 figures are participants per arm, not total events. They are illustrations based on the observed effect rather than the trial's original design. The listed assumptions are a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. For each planning scenario, the effect to detect is the stated fraction of the observed effect.

Planning for 100% of the observed effect gives 3,140 participants per arm for 80% power and 4,203 participants per arm for 90% power. Planning for 75% of the observed effect gives 5,582 participants per arm for 80% power and 7,472 participants per arm for 90% power. Planning for 50% of the observed effect gives 12,558 participants per arm for 80% power and 16,811 participants per arm for 90% power. The 80% column describes the illustrative sample size associated with an 80% chance of detecting the specified effect under those assumptions if that effect were the true one; the 90% column uses a higher chance of detection. These are not guarantees about results.

The figures increase as the effect to detect becomes smaller: a smaller difference is harder to distinguish from random variation, so the illustration calls for more participants. This is not observed power. It is not a measure of the chance that the study's observed result is true, and it is not evidence that the trial was designed with these sample sizes. The registry record does not report the original sample-size justification or its design assumptions. Because the study is a crossover, while this illustration states equal allocation between two arms, the planning figures should be understood according to the stated illustration rather than as a reconstruction of the trial's design.

12. 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 = -0.440; pooled standard deviation σ = 6.222.
2. n = 2 × (1.96 + 0.842)² × 6.222² / -0.440² = 3,139.4, rounded up to 3,140 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 <- -0.44                      # observed difference in means
sigma <- 6.22235084192462                     # 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(4311463)
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 = 3,140 (80% power), 4,203 (90% power)
 75% of observed effect: n per arm = 5,582 (80% power), 7,472 (90% power)
 50% of observed effect: n per arm = 12,558 (80% power), 16,811 (90% power)
power.t.test: 3140.3 
Simulated power at n = 3,140 per arm: 0.806 (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.

13. Multiplicity, Missing Data and Analysis Populations

The registry lists one primary endpoint, maximum observed plasma concentration of paroxetine, and lists secondary outcomes concerning participants with non-serious adverse events and serious adverse events and participants with abnormal vital signs. It also reports 9 outcome measures in total. The posted information does not explain how the primary endpoint relates to the other outcome measures, whether any outcomes were grouped into families, or whether statistical testing was planned for each one.

Multiplicity arises when several hypotheses or outcomes are examined. As a general statistical matter, testing many outcomes can increase the chance that at least one result appears unusual by chance. Prespecified approaches can control or otherwise address that issue. The registry record does not report a multiplicity strategy, a hierarchy among outcomes, or whether adjustment was intended. The p-value reported for the Cmax comparison is explicitly unadjusted, so it should not be interpreted as a multiplicity-adjusted result.

The record defines the pharmacokinetic analysis set as participants who completed both periods according to protocol criteria and had data at the specified time points. It does not report how many participants were excluded from that set, the reasons for exclusion, the number missing at each sampling time, or the method used to handle missing measurements. It also does not report whether any sensitivity analyses were performed. Those details matter because analysis-population rules and missing-data handling can affect which observations contribute to an estimate.

Although the registry lists randomized allocation and a crossover model, it does not report the detailed statistical analysis plan, the model used for the treatment contrast, or rules for protocol deviations. Detailed analysis entries are not reported in the registry record. Consequently, the posted summary comparison can be described, but its full prespecified testing context cannot be reconstructed. The absence of these details does not establish that no such methods were used; it means they are not reported in the registry record.

14. 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 0% reporting threshold.

Most frequent non-serious adverse event terms

The registry reports 0 serious adverse events among 38 participants at risk for Test A and 0 among 38 at risk for Reference B. It reports 7 under other adverse events for each treatment group, with 38 at risk in each group. Deaths are listed as 0 for both groups. These figures describe the safety counts recorded for the study; they do not by themselves establish whether the products have the same safety profile.

The adverse-event risk counts are 38, while the pharmacokinetic summaries include 31 participants per treatment group. These counts refer to different reported analyses and should not be conflated. The registry lists non-serious and serious adverse events and abnormal vital signs among its secondary outcomes, but detailed abnormal-vital-sign results are not reported in the registry record. No serious adverse-event terms are listed. The duration, severity, relatedness, and timing of individual events are not described in the posted summary, so the counts alone provide limited detail about the nature of safety observations.

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

The posted results describe a comparison of mean Cmax after a single dose of two paroxetine 20 mg tablet products in healthy adult participants under fasting conditions. The observed means were 8.39 nanograms per milliliter for Test A and 8.83 nanograms per milliliter for Reference B. The estimated mean difference was -0.44 nanograms per milliliter, with a 95% confidence interval from -3.538 to 2.658 and an unadjusted p-value of 0.781. The interval indicates that the estimated difference is imprecise and is compatible with values on either side of zero.

These results do not, on their own, establish a formal bioequivalence conclusion. Such a conclusion depends on prespecified criteria and analysis details, including an appropriate acceptance range and statistical method. Those criteria and details are not reported in the registry record. A non-significant p-value is not proof of equivalence, just as a difference in sample means is not by itself proof of a meaningful product difference.

The findings also have a specific scope. They concern the recorded Cmax outcome after a single dose under fasting conditions in healthy adult male and female participants, with the pharmacokinetic analysis restricted as described in the registry. They do not directly report outcomes in people with anxiety disorders, repeated dosing, or conditions other than those studied. While anxiety disorders are listed as a condition in the record, the study population is described as healthy participants.

The safety counts show no recorded serious adverse events or deaths in either treatment group in the posted summary, alongside 7 other adverse events reported for each group. These counts are not a complete description of event type, severity, or relationship to treatment. The posted information should therefore be understood as a limited account of the measured pharmacokinetic comparison and recorded safety counts, rather than a broad conclusion about all outcomes or populations.

16. Why This Trial Matters Statistically

This study illustrates how a randomized crossover design can compare two products while having participants receive both treatments in different periods. That structure can support within-participant comparisons, but the analysis needs to reflect the design. The registry identifies the crossover, sequences, periods, and washout, yet does not report the statistical model used for the comparison. This makes the posted estimate understandable while leaving important details of its inferential basis unclear.

The Cmax estimate also shows why a point estimate should be read together with its uncertainty. The mean difference is -0.44 nanograms per milliliter, but the confidence interval extends from -3.538 to 2.658. The estimate alone might appear close to zero; the interval conveys that the data are also compatible with a wider range of differences. The unadjusted p-value does not replace that interval and cannot answer whether a formal equivalence criterion was met.

There is also a practical distinction between the number enrolled and the number contributing to a particular outcome analysis. The registry lists enrollment of 38 and a maximum of 38 randomized and dosed, while each Cmax treatment summary includes 31 participants. The defined pharmacokinetic analysis set and the separate safety risk counts clarify that the reported denominators depend on outcome and analysis population. The record does not explain the reasons for the difference in counts.

Finally, the planning illustrations show how sample-size requirements depend on the effect targeted and the desired power. Their assumptions are stated, but they are not the trial's original design rationale. Keeping planning illustrations distinct from observed results avoids treating a retrospective calculation as evidence of the study's intended design or its achieved power.

17. 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 formal bioequivalence acceptance criteria, the statistical model used for the Cmax contrast, or whether the analysis accounted for the crossover structure through within-participant comparisons, period, sequence, or other terms. Without those details, the posted mean difference, confidence interval, and unadjusted p-value cannot be fully assessed as a formal bioequivalence analysis.

The pharmacokinetic summaries include 31 participants in each treatment group, compared with enrollment of 38 and a stated maximum of 38 randomized and dosed. Although the record defines the pharmacokinetic analysis set, it does not state the reasons for the difference in counts, exclusions by period, or missingness at individual time points. The safety risk counts of 38 per group are reported separately and do not resolve those questions.

The study was open-label, and the registry does not report blinding of outcome assessment or laboratory procedures. It lists a washout period in the brief summary, but does not report an assessment of carryover or the methods used to address it. The record also does not provide the original sample-size rationale, detailed analysis plan, multiplicity approach, or sensitivity analyses. These are reporting limitations; their absence from the registry record does not show whether they were addressed outside it.

The results concern a single oral dose under fasting conditions in healthy adult male and female participants. The registry does not report results for repeated dosing or other populations. The listed condition of anxiety disorders is not accompanied by an explanation of its relation to the healthy study population. The reported adverse-event counts also lack detail on event type and other characteristics, while abnormal vital-sign results are not reported in the registry record.

18. Related Tutorials

19. Related Statistical Calculators

20. Sources

• ClinicalTrials.gov: Bioequivalence Study of Paroxetine and PAXIL Under Fasting Conditions in Healthy Mexican Participants — NCT04311463.
• Registry record last updated 2022-01-20; status last verified 2021-12.
• Results first posted 2022-01-20.
• 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.

21. Record Summary

NCT04311463 is a completed Phase 1 randomized, open-label, two-sequence, two-treatment, two-period crossover study sponsored by GlaxoSmithKline. It compared a single oral dose of paroxetine 20 mg Test A with PAXIL (paroxetine) 20 mg Reference B under fasting conditions in healthy adult Mexican participants. The registry lists enrollment of 38 and describes a pharmacokinetic analysis set based on completing both periods and having the required data.

The primary outcome was maximum observed plasma concentration of paroxetine. The posted means were 8.39 for Test A and 8.83 for Reference B; the mean difference was -0.44, with a 95% confidence interval from -3.538 to 2.658 and an unadjusted p-value of 0.781. The record reports 0 serious adverse events and 7 other adverse events for each treatment group. Formal bioequivalence criteria and detailed statistical methods are not reported in the registry record.