Glimepiride 4 mg Tablets Under Non-Fasting Conditions: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Glimepiride 4 mg Tablets Under Non-Fasting Conditions (NCT00835172), a randomized phase 1 study. The primary outcome is Cmax - Maximum Observed Concentration.

2. Clinical Question

This is a pharmacokinetic comparison in healthy subjects, not a study of clinical outcomes in people receiving treatment for a health condition. The central statistical question is how the measured concentration profiles compare between the test and reference tablets, and how much uncertainty surrounds that comparison. The crossover arrangement means that the two treatments were assigned in opposite sequences: participants in one sequence received the test first and the reference second, while participants in the other sequence received the reference first and the test second. The registry record identifies randomization but does not describe sequence generation, period details, or any washout interval.

The posted primary endpoint summaries give arithmetic means and standard deviations for each treatment, while the registry analysis reports a geometric test-to-reference ratio with a 90% confidence interval. Those quantities express the comparison on different scales. The results should therefore be read as estimates with uncertainty, not as interchangeable calculations or as a stand-alone clinical judgment. The registry does not report the acceptance limits needed to determine how the sponsor applied a formal bioequivalence criterion.

3. Trial Design

This was a Phase 1, randomized, unmasked crossover study sponsored by Teva Pharmaceuticals USA. The registry lists 32 enrolled participants and a completed status. Both the study start and primary completion are recorded as 2003-03. The condition is listed as Healthy. The stated objective was to compare the relative bioavailability of Glimepiride 4 mg tablets with AMARYL® tablets in healthy, adult, non-smoking subjects under non-fasting conditions.

In a crossover design, participants receive more than one treatment in an assigned order, so treatment comparisons can be made within participants as well as across the randomized sequences. The arm descriptions specify two orders: Glimepiride 4 mg Tablet followed by Amaryl® 4 mg Tablet, or Amaryl® 4 mg Tablet followed by Glimepiride 4 mg Tablet. The registry identifies the allocation as randomized and masking as none. It does not describe the randomization procedure, period lengths, treatment administration schedule, or washout interval. Those details are not reported in the registry record.

The primary outcome was Cmax, or maximum observed concentration, with units of ng/mL. The specified timeframe was blood samples collected over a 24 hour period. The analysis population is described as data from all subjects who completed the study. The posted arm summaries each have a sample size of 31. The registry does not explain the relationship between the enrollment count and the number in the completed-subject analysis, nor does it provide a participant flow account in the listed details.

The registry lists three outcome measures and three analyses, while the posted results include no secondary outcomes. The study’s original sample-size rationale, target power, planned analysis methods, interim analyses, and protocol-level decision criteria are not reported in the registry record. Consequently, the posted sample-size illustrations should not be mistaken for the study’s original design calculation. They are separate planning illustrations based on the observed effect.

Because masking is listed as none, the record does not describe a blinded study. That fact does not by itself establish how laboratory measurements were handled or whether analysts were masked; such details are not reported in the registry record. The crossover structure and the endpoint’s concentration-based nature define the statistical context, but the available record does not give enough procedural detail to reconstruct every element of the original conduct or analysis.

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

Reasons for not completing this period:

Second Intervention

Analysis population for the primary outcome, as defined in the registry: Data from all subjects who completed the study were included in the statistical analysis.

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, measured over Blood samples collected over 24 hour period.

Cmax is the maximum observed concentration of a drug in the measured blood samples. It is a continuous pharmacokinetic measure, expressed in this study in ng/mL, and the primary endpoint timeframe is blood sampling over a 24 hour period. The endpoint therefore describes a concentration peak observed within the stated sampling window rather than a clinical event or a measure of long-term outcome.

For a relative bioavailability comparison, a peak concentration can help characterize whether two formulations produce similar measured exposure patterns. The study objective specifically names relative bioavailability, and the primary endpoint description states that bioequivalence is based on Cmax. This endpoint is used because comparing the observed concentration profile is relevant to assessing how the test product behaves relative to the reference product. General statistical context: a continuous endpoint can be summarized with a mean and a measure of spread, while a ratio-based analysis can compare geometric summaries when the analysis is framed multiplicatively.

The registry reports arithmetic mean Cmax values and standard deviations for both treatments, as well as a sponsor analysis stated as a geometric test-to-reference ratio multiplied by 100. These describe the concentration comparison in different ways. The arithmetic summaries describe average values on the original measurement scale; the ratio expresses relative values on a multiplicative scale. Neither summary alone establishes a bioequivalence conclusion without the applicable acceptance limits and analysis rules. Those criteria are not reported in the registry record.

The endpoint is limited to the observed maximum within the sampling timeframe. The record does not provide individual concentration profiles, the sampling schedule within the 24 hour period, or other pharmacokinetic endpoint results. Such information is not reported in the registry record, so the interpretation is confined to the posted Cmax summaries and analysis.

7. Results Posted in the Registry

• Amaryl®: mean 266.71 versus 267.63 on control; difference -0.92 (95% CI -52.99 to 51.15), standardized effect d = -0.01; z-test p = 0.972.

8. Statistical Interpretation

9. Statistical Methods Explained

The registry presents two kinds of comparison for Cmax. One is the sponsor’s reported geometric test-to-reference ratio, expressed as a percentage. The other is an unadjusted recalculation of the arithmetic mean difference from the posted treatment summaries. These methods use different scales and should be kept distinct when interpreting the result.

What does the arithmetic mean difference describe?

The arithmetic mean is the average of the observed values on the measurement scale. Here, the reported means are 267.6 ng/mL for Glimepiride and 266.7 ng/mL for Amaryl®. The recalculated contrast is defined as the Amaryl® mean minus the Glimepiride mean, giving an estimate of -0.923 ng/mL. A negative estimate under that ordering indicates that the posted Amaryl® mean is lower than the posted Glimepiride mean. The estimate is unadjusted: it compares the posted summaries without a reported model adjustment for the crossover structure.

The standard deviations are 101.2 for Glimepiride and 107.9 for Amaryl®. A standard deviation describes how dispersed individual observations are around a mean. General statistical context: more variability in measurements can make a mean difference less precisely estimated, particularly when the analyzed sample is limited. The standard deviation is not the standard error of the mean and should not be interpreted as the uncertainty interval for the treatment comparison.

What does the confidence interval contribute?

A confidence interval gives a range of estimates compatible with the observed data under the procedure used to calculate it. The recalculated mean-difference interval runs from -52.99 to 51.15. Because it includes zero, the posted summaries do not establish a nonzero arithmetic mean difference using that interval. Its width also conveys that the estimate is not precise enough to pin the difference to a narrow range. The confidence level for this recalculated interval is not stated in the listed result, so it should not be assigned a level by inference.

The registry also reports a 90% confidence interval for the geometric ratio, from 92.1 to 112, around a ratio of 102. A ratio is interpreted relative to a multiplicative reference point: values near 100 express relative similarity on the percentage scale. However, a formal equivalence conclusion requires specified acceptance limits and a decision rule. These are not reported in the registry record. A ratio and its interval can therefore be described without asserting that a particular equivalence criterion was met.

Why are the two analyses not interchangeable?

The recalculation uses the arithmetic means and reports a difference in ng/mL. The sponsor result uses a geometric test-to-reference ratio multiplied by 100. A difference asks how far apart two means are on the original scale; a ratio asks how large one quantity is relative to another on a multiplicative scale. Even when both summarize the same endpoint, they answer differently framed questions. The reported analyses also may rely on different methods, but the method for the sponsor analysis is not stated in the registry record. The recalculation does not recreate that analysis.

General statistical context: pharmacokinetic measures may be analyzed on a transformed or ratio scale because relative differences can be more relevant than absolute differences. A transformation can also change the assumptions and interpretation of an analysis. This is a general explanation, not a claim about the specific method used by the sponsor; the registry does not report that method.

What does the p-value say?

The unadjusted p-value for the recalculated mean difference is 0.972. A p-value describes how unusual data at least as incompatible with a specified null hypothesis would be under that hypothesis and the assumptions of the test. It is not the probability that the null hypothesis is true, and it is not a measure of the size or clinical importance of a difference. The particular test and its assumptions are not identified in the registry record. This p-value therefore should be treated as part of the unadjusted recalculation, not as a substitute for the sponsor’s ratio analysis or an equivalence decision.

General statistical context: failing to show evidence against a zero-difference null hypothesis does not demonstrate that two treatments are equivalent. Equivalence asks whether uncertainty is sufficiently contained within prespecified limits around a clinically or scientifically justified range. The limits and the corresponding analysis plan are not reported in the registry record.

Why can over-dispersion matter?

General statistical context: over-dispersion means that observed variability is greater than a statistical model would expect under its assumptions. For a continuous endpoint, substantial spread can reduce precision; for count outcomes, over-dispersion can make standard errors too small if an unsuitable model is used. The posted endpoint is continuous, and the record provides standard deviations, but it does not report a model-based over-dispersion assessment. No conclusion about such an assessment can be drawn for this study.

The randomized crossover design is relevant to analysis because each participant received both treatments in an assigned order. A model for a crossover study can account for the paired nature of observations and may include period or sequence terms. The registry does not identify such a model or provide participant-level paired data. The recalculation from marginal arm summaries cannot recover within-participant covariance or establish whether period and sequence effects were considered. It should therefore be read as a simple unadjusted comparison rather than a reconstruction of a crossover analysis.

10. Confidence Intervals and Precision

The arithmetic mean-difference estimate is -0.923 ng/mL, with a confidence interval from -52.99 to 51.15. The interval spans values below and above zero, so the recalculated estimate is compatible with no arithmetic mean difference as well as with differences in either direction. The width indicates uncertainty around the point estimate; the exact confidence level for this recalculated interval is not stated in the registry record.

The treatment means, 267.6 ng/mL for Glimepiride and 266.7 ng/mL for Amaryl®, are close as posted. That closeness should not be considered in isolation from the standard deviations of 101.2 and 107.9. General statistical context: a point estimate gives the best single summary under a method, while an interval shows how much uncertainty surrounds that estimate. A narrow interval can support a more precise statement than a wide interval, even if both have similar point estimates.

The sponsor’s reported ratio is 102, and its 90% confidence interval is 92.1 to 112. This interval expresses uncertainty on the ratio scale, not in ng/mL. It cannot be judged against an equivalence standard without the acceptance limits, which are not reported in the registry record. The interval should also not be read as the range of individual participant values.

Precision is constrained by the posted analysis sample size of 31 in each treatment arm and by the observed variability. The record does not provide individual measurements, a crossover-model estimate, or the within-participant correlation, so a more complete assessment of precision using paired data is not possible from the posted summaries. The arithmetic contrast and the geometric ratio are complementary descriptions, but their intervals answer different questions and are not interchangeable.

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 sample-size values are planning illustrations expressed as participants per arm, not total event counts. They are based on the observed effect and are not the study’s original enrollment target or a reconstruction of its design calculation. The registry’s original sample-size assumptions and justification are not reported in the registry record.

The stated illustration assumptions are a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. The effect to detect is set to a stated fraction of the observed effect. The rows therefore describe planning for 100%, 75%, and 50% of the observed effect. The columns describe participants per arm for 80% power and participants per arm for 90% power. In plain terms, those power levels represent the planned chance of detecting the specified effect under the stated assumptions if that effect is present and the assumptions hold.

As the target effect becomes smaller, more participants are needed to distinguish it from random variation under the same planning framework. The listed illustrations show this pattern across the stated effect fractions. Because the recalculated observed mean difference is very close to zero, the resulting planning figures are extremely large; they should be read as a consequence of using that observed estimate as the planning basis, not as a recommendation for a future study or evidence about the original trial’s design.

These figures are not observed power. Observed power is not a useful substitute for interpreting the estimated effect and its uncertainty after results are available. The confidence interval and the study’s design and analysis details are more informative for understanding what the posted data can support. The sample-size illustration also does not establish that the trial was adequately or inadequately powered under its original plan, because that plan is not reported in the registry record.

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 = -0.923; pooled standard deviation σ = 104.594.
2. n = 2 × (1.96 + 0.842)² × 104.594² / -0.923² = 201,579.7, rounded up to 201,580 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 <- -0.923000000000002                      # observed difference in means
sigma <- 104.593886365313                     # 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(835172)
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 = 201,580 (80% power), 269,859 (90% power)
 75% of observed effect: n per arm = 358,364 (80% power), 479,748 (90% power)
 50% of observed effect: n per arm = 806,319 (80% power), 1,079,433 (90% power)
power.t.test: 201580.7 
Simulated power at n = 201,580 per arm: 0.802 (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 registry’s reported analysis gives a geometric test-to-reference ratio multiplied by 100 of 102, with a 90% confidence interval from 92.1 to 112. The method field is not described. This is the sponsor’s own reported result for the registry’s bioequivalence framing and should be distinguished from the unadjusted recalculation of the arithmetic mean difference.

The recalculation gives an Amaryl® minus Glimepiride mean difference of -0.923 ng/mL, with a confidence interval from -52.99 to 51.15 and an unadjusted p-value of 0.972. It is based on the posted arithmetic means and does not reproduce a geometric ratio analysis or a model that may account for the crossover structure. The ratio and the difference use different scales, so their estimates are not direct alternatives to one another.

The discrepancy indicator is false. The registry does not report the sponsor’s exact statistical method, the prespecified equivalence limits, or the decision rule used to interpret the ratio. The sponsor’s estimate and interval can therefore be stated, but a formal equivalence conclusion should not be inferred from the ratio alone. The recalculation offers a separate descriptive check on the arithmetic summaries; it does not replace the registry analysis or resolve the missing analysis details.

14. Multiplicity, Missing Data and Analysis Populations

The registry lists three outcome measures and three analyses, but the posted results include no secondary outcomes. The primary endpoint is Cmax. Results for secondary outcomes are not reported in the registry record. The record does not explain whether the additional outcomes were analyzed, whether they were prespecified, or how they relate to the reported analyses.

Multiplicity arises when several outcomes, comparisons, or analyses are considered, because repeated statistical testing can increase the chance of at least one apparently notable result. General statistical context: a prespecified adjustment or a clear hierarchy can help control interpretation across multiple tests. Whether any such approach was used is not reported in the registry record. The unadjusted p-value of 0.972 should not be described as multiplicity-adjusted.

The primary analysis population is described as data from all subjects who completed the study. Enrollment is listed as 32, and each posted treatment summary has a sample size of 31. The record does not provide a participant flow account or explain the difference between those counts. It also does not report whether any participants were excluded from particular analyses, how incomplete observations were handled, or whether missing values were imputed. These details are not reported in the registry record.

The crossover design means that a complete comparison may depend on observations from both treatment periods for a participant. The registry does not describe how incomplete period data were treated or whether a participant with only one observed treatment contributed to any analysis. It also does not identify an intention-to-treat, per-protocol, or other named analysis population. The completed-subject description is the only stated population definition.

The original statistical analysis plan, any interim analyses, and any multiplicity strategy are not reported in the registry record. Consequently, it is not possible to determine from the registry entries whether the reported analysis matches a prespecified plan, or how unreported outcomes and incomplete data affected the overall evidence.

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

The posted results show similar arithmetic mean Cmax values for the two treatments in the analyzed summaries: 267.6 ng/mL for Glimepiride and 266.7 ng/mL for Amaryl®. The unadjusted mean-difference estimate is -0.923 ng/mL, and its interval extends from -52.99 to 51.15. The sponsor’s separate ratio analysis reports 102, with a 90% confidence interval from 92.1 to 112. Together, these figures describe the observed primary pharmacokinetic comparison in the registry.

They do not, by themselves, establish that the treatments are clinically interchangeable or that a formal bioequivalence criterion was met. The registry identifies bioequivalence based on Cmax as the endpoint framing, but the acceptance limits and decision rule are not reported in the registry record. The recalculated p-value tests a different kind of question from a formal equivalence assessment and cannot substitute for those missing criteria.

This study involved healthy, adult, non-smoking subjects under non-fasting conditions, according to the study objective. The result is specific to the study population, treatment products, endpoint, and sampling timeframe described. The registry does not establish how the findings apply to other populations, clinical settings, conditions, or outcomes. The study is a relative bioavailability comparison, not a report of clinical benefit or harm.

The result also does not establish equality of every aspect of the measured concentration profiles. Cmax focuses on the maximum observed concentration within blood sampling over a 24 hour period. The registry entries do not provide individual profiles or other endpoint results needed to describe the full time course. General statistical context: an estimate close to zero or a ratio near its reference value is not proof of exact equality; conclusions depend on uncertainty, design, measurement, and the prespecified question.

Finally, the record does not report enough methodological detail to reproduce the sponsor’s analysis or independently apply a formal equivalence decision rule. The posted summaries are useful for understanding the reported estimates and their precision, but the missing protocol-level criteria and analysis details limit stronger conclusions.

16. Why This Trial Matters Statistically

This study illustrates why the scale of an estimate matters. An arithmetic mean difference in ng/mL and a geometric test-to-reference ratio multiplied by 100 describe the same endpoint through different statistical questions. Readers should not treat the recalculated difference and the sponsor’s ratio as duplicate versions of one result. Each needs to be interpreted on its own scale, with its own uncertainty and analysis assumptions.

It also demonstrates the difference between failing to find evidence of a nonzero difference and establishing equivalence. The unadjusted p-value is 0.972, while the sponsor reports a ratio and a 90% confidence interval. A formal equivalence assessment requires prespecified limits and a rule for interpreting the interval. Those criteria are not reported in the registry record. A result that appears close to a reference value is not enough, on its own, to establish that the full uncertainty range meets a formal standard.

The crossover structure is another important statistical feature. Since participants received both treatments in one of two randomized orders, a paired analysis could use within-participant information. The posted unadjusted recalculation uses treatment-level summaries, and the registry does not describe the sponsor’s method. Without paired observations or an identified model, the summaries cannot reveal how within-participant variation or period effects were handled.

Finally, the sample-size illustration emphasizes how planning calculations depend on the effect chosen. The figures use fractions of the observed effect and stated assumptions; they are not the original trial rationale and are not observed power. Interpreting the study therefore requires keeping the reported result, its uncertainty, the design, and the limits of the registry record distinct.

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 conduct and analysis. It identifies randomization, no masking, and two treatment sequences, but does not describe the randomization procedure, treatment schedule, period duration, or washout interval. These details are not reported in the registry record and are relevant to understanding how the crossover comparison was implemented.

The primary endpoint is defined as Cmax from blood samples collected over a 24 hour period, but the sampling schedule and participant-level concentration profiles are not listed. The record does not provide demographic characteristics beyond the description of healthy, adult, non-smoking subjects in the study objective. Additional information about eligibility, participant characteristics, and treatment administration is not reported in the registry record.

The sponsor’s reported analysis is a geometric test-to-reference ratio with a 90% confidence interval, but its statistical method and formal acceptance limits are not reported. The arithmetic mean recalculation is unadjusted and based on marginal summaries. It cannot recover the paired information in the crossover design or establish how the original analysis handled period and sequence effects.

Although 32 participants were enrolled, the posted summaries include 31 per treatment arm, and the analysis population is described as study completers. The record does not explain participant disposition or describe missing-data handling. The registry lists three outcomes and three analyses, but no secondary results are posted. Multiplicity procedures, interim analyses, and the original sample-size rationale are also not reported in the registry record. These limits constrain reproducibility and the scope of conclusions drawn from the posted results.

18. Related Tutorials

19. Related Statistical Calculators

20. Sources

• ClinicalTrials.gov: Glimepiride 4 mg Tablets Under Non-Fasting Conditions — NCT00835172.
• Registry record last updated 2024-08-20; status last verified 2024-08.
• Results first posted 2009-08-04.
• 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

NCT00835172 is a completed, randomized, unmasked Phase 1 crossover study sponsored by Teva Pharmaceuticals USA. The registry lists 32 enrolled participants and describes the study objective as comparing Glimepiride 4 mg tablets with AMARYL® 4 mg tablets in healthy, adult, non-smoking subjects under non-fasting conditions.

The primary endpoint is Cmax in ng/mL, based on blood samples collected over a 24 hour period. The posted treatment summaries report means of 267.6 for Glimepiride and 266.7 for Amaryl®, with 31 participants in each arm. The sponsor’s analysis reports a geometric test-to-reference ratio of 102 and a 90% confidence interval from 92.1 to 112. An unadjusted recalculation reports a mean difference of -0.923, with a confidence interval from -52.99 to 51.15 and an unadjusted p-value of 0.972.

The registry does not report the equivalence limits, full statistical method, original sample-size rationale, or secondary results. The posted estimates describe the primary pharmacokinetic comparison, while those missing details limit independent reconstruction of the formal analysis.