15 mg Mirtazapine Orally Disintegrating Tablets, Non-Fasting: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for 15 mg Mirtazapine Orally Disintegrating Tablets, Non-Fasting (NCT00834197), a randomized phase 1 study. The primary outcome is Cmax (Maximum Observed Concentration of Drug Substance in Plasma).

2. Clinical Question

3. Trial Design

This was a phase 1, randomized, unmasked, crossover study of two orally disintegrating mirtazapine tablet products. The registry records 32 participants enrolled and the study status as completed. The study began in 2003-07, and primary completion was recorded in 2003-08. The stated condition was healthy, and the registry’s summary describes healthy adult subjects.

Participants were assigned in a randomized crossover design. In general, a crossover design compares treatments within the same participant across different treatment periods, which can make within-person comparisons useful when outcomes vary between individuals. The registry record does not report the sequence allocation details, the number or length of treatment periods, any washout interval, or the procedures used to maintain blinding; masking is recorded as none.

The test was a single oral dose of 1 x 15 mg mirtazapine orally disintegrating tablets under non-fasting conditions. The comparator was a single oral dose of REMERON SolTab® 15 mg orally disintegrating tablets. Blood samples were collected over a 120 hour period for the primary endpoint. The record reports 32 observations for each treatment group in the posted Cmax results. It does not give a separate count of participants who completed the study, beyond stating that samples from all participants who completed it were analyzed.

The sponsor was Teva Pharmaceuticals USA. The record lists three outcome measures and three analyses, but only the primary outcome details are described in the posted information. Detailed protocol procedures, the original sample-size rationale, and the statistical plan 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.

First Intervention

Washout of 14 Days

Second Intervention

Analysis population for the primary outcome, as defined in the registry: All participants that completed the study had their samples 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 Cmax (Maximum Observed Concentration of Drug Substance in Plasma), measured over Blood samples collected over a 120 hour period..

Cmax is the maximum observed concentration of a drug substance in plasma after administration. In this study it was recorded in pg/mL, with blood samples collected over a 120 hour period. The primary endpoint description identifies the measure as a mean with standard deviation and states that bioequivalence was based on Cmax.

In general, a maximum concentration is used to describe the peak level observed in the sampling period. Comparing Cmax between formulations can inform a relative bioavailability assessment, particularly with respect to the concentration reached after dosing. It is not by itself a complete description of all aspects of absorption. The registry’s stated study objective refers to comparing the rate and extent of absorption, while the posted primary endpoint specifically focuses on Cmax.

The registry reports both treatment-specific arithmetic means and a sponsor analysis expressed as a test-to-reference geometric mean ratio. These summaries answer related but not identical statistical questions: arithmetic means describe average values on the original measurement scale, while geometric-mean ratios describe a relative comparison on a multiplicative scale. The record does not provide the sponsor’s detailed analysis method or the full set of endpoint decision criteria.

7. Results Posted in the Registry

• Test (Mirtazapine): mean 26,345 versus 25,943 on control; difference 402.29 (95% CI -2,671 to 3,475), standardized effect d = 0.06; z-test p = 0.798.

8. Statistical Interpretation

9. Statistical Methods Explained

Statistical summaries of a continuous endpoint can describe both its central value and its variability. For Cmax, the registry posts a mean and standard deviation for each product. The mean is an average on the original concentration scale. The standard deviation describes how dispersed individual observations are around that mean. These descriptive quantities do not by themselves account for the crossover structure or establish a treatment decision.

What does the arithmetic mean comparison describe?

The recalculated comparison expresses the test-minus-reference contrast in pg/mL. It is an unadjusted re-calculation from posted summary statistics, rather than a reproduction of the sponsor’s full analysis. A positive estimate indicates that the posted test mean is higher than the posted reference mean; the confidence interval describes uncertainty around that estimated arithmetic difference. The p-value is a measure of compatibility between the observed result and a specified null hypothesis under the assumptions of the calculation. It is not the probability that the null hypothesis is true, nor does it measure the probability that the treatments are equivalent.

In general, an unadjusted comparison based on treatment-level summaries may not use all the information available from a crossover study. A crossover analysis can account for paired observations within participants, and may also consider period or sequence structure. The registry record does not report the sponsor’s detailed model, so the recalculation should be read as a simple summary-based contrast, not as a substitute for an undocumented crossover model.

Why is the sponsor result expressed as a ratio?

The sponsor analysis reports the ratio of test-to-reference geometric means multiplied by 100. A ratio is a relative comparison: a value of 100 corresponds to equal geometric means on that scale. The reported estimate is 100.59, with a 90% confidence interval from 95.76 to 105.65. Geometric summaries are often used when analysis is based on logarithms, because multiplicative differences become additive on a log scale. This is general statistical context; the registry record does not state the sponsor’s exact method or model.

Arithmetic means and geometric means summarize distributions differently. An arithmetic mean is calculated on the original scale, while a geometric mean reflects multiplicative averaging and is typically defined for positive observations. Therefore, an arithmetic mean difference in pg/mL and a geometric mean ratio cannot be compared as though they were the same estimand. The registry gives no detailed data distribution or transformation information beyond the named sponsor summary.

What does a confidence interval communicate?

In general, a confidence interval gives a range of values compatible with the data and the statistical method at the stated confidence level. Its width reflects precision: a narrower interval indicates less uncertainty under the same approach, while a wider interval indicates more. A confidence interval is not a guarantee that the true value lies within its particular bounds, and it does not assign probabilities to individual candidate values in the usual frequentist interpretation. Interpretation also depends on whether the analysis and interval method match the study design and estimand.

The recalculated arithmetic comparison has a 95% confidence interval, while the sponsor’s reported ratio has a 90% confidence interval. Those confidence levels are not the same, and the intervals concern different scales and summaries. They should not be compared as if they were two estimates from an identical analysis. The reported endpoints and confidence levels should remain attached to their corresponding analyses.

Why can over-dispersion matter?

In general, over-dispersion means that observed variability is greater than a model expects. If a model understates variability, its standard errors may be too small and its confidence intervals too narrow. This is a general statistical concern, not a claim that over-dispersion occurred in this study. The record gives treatment-specific standard deviations but does not report distributional diagnostics or an over-dispersion assessment.

The registry identifies the primary measure as a continuous outcome and reports means and standard deviations. It does not provide the individual participant data, the distribution of Cmax, the full crossover analysis, adjustment variables, or the detailed confidence-interval method. Consequently, the available results support interpretation of the posted summaries and the stated recalculation, but do not permit independent verification of the sponsor’s complete analysis.

10. Confidence Intervals and Precision

The arithmetic mean difference is estimated as 402.3 pg/mL, with a 95% confidence interval from -2671.0 to 3475.0. The interval spans both negative and positive differences. This indicates substantial uncertainty around the estimated difference and does not support a precise claim that the arithmetic means are exactly equal. It is also not a direct equivalence interval unless interpreted against prespecified margins, which are not reported in the registry record.

The sponsor’s reported ratio is 100.59, with a 90% confidence interval from 95.76 to 105.65. This interval describes uncertainty for the sponsor’s test-to-reference geometric mean ratio, not the arithmetic mean difference. The registry record identifies the analysis as a legacy non-inferiority or equivalence analysis but does not report the margin or decision rule required to interpret the interval against a formal threshold.

In general, interval width depends on the amount of information, variability, and the analysis method. The treatment-level standard deviations are 6918.0 for the test product and 5551.0 for the reference product. The registry does not report a paired-difference standard deviation or a complete model-based uncertainty calculation for the crossover design. Therefore, the summary information does not allow a full assessment of how within-participant correlation affected precision.

The two intervals have different confidence levels and express different estimands. Precision should be judged within each analysis, not by directly comparing their endpoints. Neither interval establishes that the products are identical; the sponsor interval’s formal interpretation also depends on criteria not reported in the registry record.

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 event counts. They are illustrations based on the observed effect, not the original study’s design calculation. The stated assumptions are a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation. The effect to detect is set as the stated fraction of the observed effect. The registry record does not report the trial’s original sample-size justification or planning assumptions.

The 80% power column gives the number of participants per arm indicated by the planning calculation for an 80% chance of detecting the selected effect, under the listed assumptions. The 90% power column gives the corresponding number per arm for a 90% chance. In general, higher planned power requires more participants when other assumptions remain unchanged. The rows that plan for 100%, 75%, and 50% of the observed effect illustrate that smaller effects require more participants under this calculation.

These are planning illustrations, not observed power and not evidence about the probability that the trial result is correct. They are based on a simple calculation from the observed effect, use equal allocation, and do not reconstruct the design or analysis of the crossover study. The registry does not report whether the original sample size was chosen for Cmax, another endpoint, or a particular equivalence margin. These figures should therefore be read as hypothetical planning scenarios rather than a retrospective evaluation of the study’s adequacy.

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 = 402.290; pooled standard deviation σ = 6271.500.
2. n = 2 × (1.96 + 0.842)² × 6271.500² / 402.290² = 3,815.1, rounded up to 3,816 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 <- 402.290000000001                      # observed difference in means
sigma <- 6271.50013783385                     # 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(834197)
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 = 3,816 (80% power), 5,108 (90% power)
 75% of observed effect: n per arm = 6,783 (80% power), 9,080 (90% power)
 50% of observed effect: n per arm = 15,261 (80% power), 20,430 (90% power)
power.t.test: 3816 
Simulated power at n = 3,816 per arm: 0.798 (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. Analyses Reported by the Sponsor

The sponsor’s posted result is the ratio of test-to-reference geometric means multiplied by 100: 100.59, with a 90% confidence interval from 95.76 to 105.65. The registry categorizes this as a legacy non-inferiority or equivalence analysis. It does not report a p-value for this analysis, the sponsor’s detailed statistical method, or the formal acceptance margins.

The recalculated result is an unadjusted arithmetic mean difference of 402.3 pg/mL, with a 95% confidence interval from -2671.0 to 3475.0 and p = 0.798. These analyses use different scales and summaries: one is a relative geometric-mean comparison, and the other is an arithmetic difference. They are not competing estimates of precisely the same quantity. The recalculation also does not reproduce a full crossover model.

The registry does not flag a discrepancy. The sponsor’s result should be read according to its stated ratio scale and reported confidence interval, while the recalculation provides a separate descriptive contrast from the posted means. The missing margins and analysis details limit any independent conclusion about the sponsor’s formal equivalence decision.

14. Multiplicity, Missing Data and Analysis Populations

The registry lists three outcome measures and three analyses, while the posted information describes the primary endpoint as Cmax. No secondary outcomes are posted. The record does not report whether the analyses were part of a prespecified hierarchy, whether multiple comparisons were adjusted, or how multiplicity was handled. In general, when several endpoints or comparisons are tested, the chance of at least one apparently unusual result can increase; procedures such as prespecified hierarchies or statistical adjustment may address that issue. Whether such procedures were used in this study is not reported in the registry record.

The primary endpoint description states that samples from all participants who completed the study were analyzed. That identifies a completion-based analysis population in broad terms, but the record does not provide participant flow, the number who completed, exclusions, or reasons for missing data. The treatment group summaries each report 32 observations. Those counts do not provide a full account of whether any measurements were missing or how incomplete participant records were handled.

The record does not state whether an intention-to-treat, per-protocol, or other named analysis population was used for the sponsor’s ratio analysis. It also does not report imputation, sensitivity analyses, or the handling of observations below a quantification threshold. These details can matter in bioequivalence studies because eligibility for the analysis and missing measurements can affect the estimates and their uncertainty.

The registry record does not report interim analyses, stopping rules, a statistical analysis plan, or the prespecified margins for the legacy equivalence or non-inferiority analysis. Their absence from the posted information should not be taken as evidence that a particular procedure was or was not used; it means that the details cannot be assessed from this record.

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

The posted results describe Cmax after a single 15 mg oral dose of each orally disintegrating tablet product under non-fasting conditions in healthy adult subjects. The test and reference arithmetic means are 26350.0 pg/mL and 25940.0 pg/mL, respectively. The sponsor analysis reports a test-to-reference geometric mean ratio of 100.59, with a 90% confidence interval from 95.76 to 105.65. These are statistical descriptions of plasma concentration measurements, not clinical outcomes.

The recalculated arithmetic contrast has a 95% confidence interval from -2671.0 to 3475.0. It is compatible with a range of possible differences, and does not establish exact equality. The sponsor ratio and interval are on a different scale; their formal interpretation depends on decision margins that are not reported in the registry record. Statistical summaries alone should not be turned into a broader claim about the products beyond the endpoint, population, and conditions actually recorded.

In general, bioavailability measures concern how a drug is absorbed, and Cmax captures the maximum observed plasma concentration over the sampling period. The registry’s stated objective concerns relative bioavailability, while the primary posted result focuses on Cmax. The available information does not describe every aspect of absorption or provide a complete account of other pharmacokinetic endpoints.

The study involved healthy adults and a single dose under non-fasting conditions. The record does not support extrapolation to other populations, repeated dosing, other administration conditions, or clinical effects. No medical advice follows from these statistical results. The interpretation is limited to the posted concentration data and the analyses described in the registry.

16. Why This Trial Matters Statistically

This study is statistically useful as an example of how a bioavailability comparison can present related results on different scales. The treatment summaries are arithmetic means in pg/mL, while the sponsor result is a test-to-reference geometric mean ratio. Those statistics answer different questions and should not be collapsed into a single estimate. Clear identification of the estimand and analysis scale is important when interpreting comparative pharmacokinetic results.

The crossover design is also relevant to how the comparison might be analyzed. In general, measurements on the same participant under different treatments can permit a paired comparison, while period and sequence structure may also matter. The posted summary-based recalculation is unadjusted, and the registry does not provide enough model detail to establish how the sponsor accounted for the design. This distinction illustrates why a simple comparison of treatment means is not necessarily equivalent to a full design-aware analysis.

The registry reports a confidence interval for both the arithmetic contrast and sponsor ratio, but the sponsor’s formal decision margins are not reported. An interval cannot be interpreted as establishing equivalence without the relevant prespecified criteria. Thus, the record illustrates the value of reporting not only estimates and intervals, but also the decision rule and analysis methods used to reach a formal conclusion.

Finally, a planning illustration based on an observed effect is distinct from the trial’s original sample-size rationale. The original rationale is not reported. Keeping retrospective planning scenarios separate from the actual design prevents mistaken conclusions about why the study enrolled its participants or whether it was planned around the posted estimate.

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 provides limited detail about the crossover procedures. It does not report treatment sequence details, the number and duration of periods, washout timing, or any period-specific procedures. Such details can help readers assess the design and how within-participant data were analyzed.

The primary endpoint is described as Cmax, but the record does not provide the individual concentration data, the sponsor’s full statistical model, or detailed information about transformations and confidence-interval construction. The recalculated arithmetic comparison uses posted summary statistics and is unadjusted; it does not reconstruct a full crossover analysis. The record also does not report the sponsor’s decision margins or a detailed formal decision rule for the legacy equivalence or non-inferiority analysis.

The participant description is limited to healthy adult subjects, and the registry reports 32 enrolled. The endpoint description says that samples from participants who completed the study were analyzed, but the number who completed, participant flow, exclusions, and missing-data handling are not reported. Although each treatment summary has 32 observations, those summaries do not replace a detailed account of completion and analysis populations.

The record lists three outcome measures and three analyses but does not provide secondary outcome results in the posted information. It does not describe multiplicity procedures, interim analyses, a statistical analysis plan, or the original sample-size justification. These omissions constrain assessment of the complete statistical plan and of how the posted primary result fits within all planned analyses.

The findings are specific to a single oral dose under non-fasting conditions in healthy adults. The registry record does not establish results for other populations, dosing patterns, or conditions.

18. Related Tutorials

19. Related Statistical Calculators

20. Sources

• ClinicalTrials.gov: 15 mg Mirtazapine Orally Disintegrating Tablets, Non-Fasting — NCT00834197.
• Registry record last updated 2024-08-20; status last verified 2024-08.
• Results first posted 2009-08-18.
• 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

NCT00834197 is a completed phase 1 randomized, unmasked crossover study sponsored by Teva Pharmaceuticals USA. It enrolled 32 healthy adult subjects and compared a single 15 mg dose of mirtazapine orally disintegrating tablets with REMERON SolTab® 15 mg under non-fasting conditions. The study began in 2003-07, with primary completion recorded in 2003-08.

The primary endpoint was Cmax in pg/mL, based on blood samples collected over a 120 hour period. The sponsor reported a test-to-reference geometric mean ratio of 100.59 with a 90% confidence interval from 95.76 to 105.65. An unadjusted recalculation of the arithmetic mean difference was 402.3 pg/mL, with a 95% confidence interval from -2671.0 to 3475.0. These analyses use different scales. The record does not report the sponsor’s detailed decision margins or full statistical method.