Cefprozil for Oral Suspension 250 mg/5 mL, Fasting: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Cefprozil for Oral Suspension 250 mg/5 mL, Fasting (NCT00835614), a randomized phase 1 study. The primary outcome is Cmax (Maximum Observed Concentration).

2. Clinical Question

The study is registered as phase 1 and completed. Its focus is a pharmacokinetic comparison in healthy participants, not a comparison of clinical outcomes in people being treated for an illness. The registry identifies a 12 hour blood-sampling period for the primary endpoint. It does not give the collection schedule within that period or explain the criteria used to assess bioequivalence. The posted results therefore support interpretation of the reported concentration summaries, but do not by themselves establish which decision rule the sponsor applied.

3. Trial Design

The record describes a randomized, unmasked, crossover study. The two interventions were cefprozil for oral suspension 250 mg/5 mL and CEFZIL® for Oral Suspension 250 mg/5 mL, both studied under fasting conditions. The objective was to compare their relative bioavailability in healthy, non-smoking adults. The study is listed as phase 1 and completed, with enrollment of 26. The registry gives the start and primary completion as 2004-04.

In a crossover design, participants receive more than one intervention across study periods, so measurements under the products can be related within the same participant. Randomization can help determine the order in which interventions are received. These are general features of crossover studies; the registry record does not report the treatment sequence, period structure, washout interval, or how sequence was assigned. It also does not describe whether the analysis accounted for period or sequence effects.

Masking is listed as none. The record does not describe how product administration, sample handling, or outcome assessment was managed in light of that design feature. Arm descriptions are empty, so they do not add procedural detail. The record identifies the sponsor as Teva Pharmaceuticals USA.

The posted primary outcome is Cmax, the maximum observed concentration, with blood samples collected over a 12 hour period. The primary analysis population is described as all participants who completed the study and had their samples analyzed. A count for that completed-and-analyzed population is not reported in the registry record. The arm summaries each list 26 observations, but because this is a crossover study those summaries should not be treated as evidence of separate groups of participants.

The registry does not report the original sample-size assumptions, a detailed statistical analysis plan, or any interim analysis. The design features that are documented—randomization, no masking, and crossover—provide context for reading the result, but do not establish how the sponsor handled the paired measurements or other design features in its 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 of 7 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), measured over Blood samples collected over a 12 hour period..

Cmax is the maximum observed concentration of a substance in blood after administration. It is a continuous pharmacokinetic measure, in ng/mL. The registry describes it as the primary outcome and states that bioequivalence was based on Cmax. Blood samples for this endpoint were collected over a 12 hour period.

As a general concept, Cmax is used to describe the peak concentration observed during the sampling period. It can contribute to a comparison of how much of a product becomes available in the body and the concentration reached, alongside other pharmacokinetic measures. A measured peak depends in part on when samples are taken: if the schedule does not capture the true peak, the maximum observed value may differ from it. The registry does not report the detailed sampling schedule, so the extent to which sampling timing could affect the observed maximum cannot be assessed from the record.

The posted endpoint uses a mean and standard deviation for each product. The sponsor’s result instead reports a test-to-reference geometric-mean ratio as a percentage, with a confidence interval. These measures answer related but distinct statistical questions: the group means describe concentration levels, while the ratio expresses the relative geometric-mean comparison. The registry does not report the sponsor’s full model or the criteria used to decide whether the ratio met a bioequivalence standard.

Cmax is a concentration endpoint, not a direct measure of symptom relief, safety, or clinical benefit. The result should therefore be read as a comparison of the reported pharmacokinetic measurements under the study conditions.

7. Results Posted in the Registry

• Cefprozil (Test): mean 5.92 versus 6.51 on control; difference -0.58 (95% CI -1.43 to 0.26), standardized effect d = -0.37; z-test p = 0.177.

8. Statistical Interpretation

9. Statistical Methods Explained

What does the mean difference describe?

The posted arm summaries give a mean Cmax for each product. The recalculated mean difference compares the test-product mean with the reference-product mean on the original concentration scale. Its estimate is -0.584 ng/mL. A negative sign indicates that the test mean is lower than the reference mean in this subtraction. The interval for this recalculation spans zero, so the summary-statistic result does not precisely distinguish a nonzero difference from no difference on that scale.

This recalculation is unadjusted and uses the posted summary statistics. It is not a reconstruction of the sponsor’s analysis. In particular, the trial is crossover, but the posted arm-level summaries do not contain the participant-level paired values needed to reproduce a paired analysis. The record does not report whether the sponsor used a paired model or how it handled period and sequence.

What does the standard deviation describe?

The standard deviation describes how spread out individual measurements are around their group mean. The posted standard deviations are 1.513 for Cefprozil (Test) and 1.603 for Cefzil® (Reference). A larger spread indicates more variability among the measurements around the mean, not greater uncertainty in the mean itself. The standard error, by contrast, describes the estimated precision of a mean and depends on both variability and the amount of information used.

For the recalculation, the comparison reports a pooled standard deviation of 1.559. Pooling combines group variability into a common estimate for purposes of the unadjusted comparison. The standardized effect estimate is reported as -0.375 in the underlying calculation; it is not necessary to interpret that standardized scale as a concentration difference. The posted mean difference remains the more direct expression in ng/mL.

Why are the sponsor result and recalculation different measures?

The sponsor reports a ratio of the test-to-reference geometric means multiplied by 100. Its reported value is 90.9, with a 90% confidence interval from 87.5 to 94.4. A geometric mean is generally used for positive measurements when relative differences are of interest or when a distribution is better represented on a logarithmic scale. The ratio compares the products proportionally. By contrast, the recalculated arithmetic mean difference is an absolute contrast in ng/mL.

These measures can provide complementary descriptions, but their numerical values are not directly comparable because they use different scales and summaries. The registry record does not state the sponsor’s transformation, statistical model, or interval method. It also does not report the acceptance limits or decision rule used for the stated bioequivalence assessment. General descriptions of ratios and geometric means therefore cannot fill in those missing protocol details.

How should the p-value be read?

The unadjusted recalculation reports p = 0.177. In general, a p-value measures how compatible the observed data are with a specified null hypothesis under the assumptions of the statistical procedure. It is not the probability that the null hypothesis is true, and it does not measure the size or clinical importance of an effect. Interpretation depends on the model and analysis plan. The registry does not report a sponsor p-value for the geometric-mean ratio, so the recalculated p-value should not be attributed to that analysis.

A p-value also does not describe precision as directly as a confidence interval. The interval shows a range of estimates compatible with the data and the method; the p-value compresses information into a single statistic tied to a test. In this case, the recalculated interval spans zero, consistent with the uncertainty visible in the difference estimate. Neither statistic identifies whether an unreported bioequivalence criterion was satisfied.

Why does the crossover design matter?

In general, crossover designs compare interventions within the same participant, which can reduce the influence of stable differences between people. The appropriate analysis commonly needs to reflect the paired structure and may consider period or sequence. Whether and how those terms were included depends on the protocol and analysis plan. The registry confirms a randomized crossover design but does not provide those details, the washout interval, or participant-level data.

Consequently, an unadjusted comparison of arm summaries cannot capture all design information. It can be useful as a transparent descriptive recalculation, but it should not replace the sponsor’s own analysis of the trial. The sponsor’s ratio is the posted result specifically framed as a bioequivalence analysis; the recalculated mean difference is a separate summary-statistic comparison, not an alternative trial-level model.

10. Confidence Intervals and Precision

The sponsor’s relative estimate is 90.9, with a 90% confidence interval from 87.5 to 94.4. This interval expresses uncertainty around the reported test-to-reference geometric-mean ratio, expressed as a percentage. Its endpoints are below 100, the reference point for a ratio expressed as a percentage of the reference. The record does not report the acceptance limits or decision rule, so the interval alone cannot establish whether the sponsor’s bioequivalence criterion was met.

The unadjusted recalculation reports a mean difference of -0.584 ng/mL. Its confidence interval extends from -1.431 to a positive endpoint, and therefore spans zero. That indicates uncertainty about the direction of the arithmetic-mean difference under this recalculation. The interval is on the concentration-difference scale; it should not be read as the uncertainty interval for the sponsor’s ratio.

In general, a confidence interval’s width reflects precision as well as the variability of the measurements and the analysis method. A narrower interval conveys greater precision about the parameter being estimated; a wider interval leaves more plausible values unresolved. Confidence intervals do not by themselves account for unreported design details or correct a model that does not reflect the study structure.

The sponsor interval is explicitly identified as 90%. The confidence level for the unadjusted recalculated interval is not separately stated in the record, so no confidence level should be assigned to it beyond what the statistical output labels. The crossover structure and absence of participant-level results also limit independent assessment of the precision calculations.

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 rather than the trial’s original design assumptions. The stated planning 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 as a stated fraction of the observed effect. The original sample-size justification and assumptions are not reported in the registry record.

The 80% power column gives the illustrative number of participants per arm for a design with an 80% chance of detecting the specified effect under the planning assumptions, if that effect and model assumptions were correct. The 90% power column gives the corresponding number per arm for a design with a 90% chance. These are design-planning quantities, not probabilities that the observed result is true.

The rows illustrate planning for 100%, 75%, and 50% of the observed effect. Smaller effects require more participants to detect under otherwise unchanged assumptions, which is why the required numbers increase as the targeted fraction decreases. The figures are not a recommendation for a future study and do not account for design features that are not specified in the planning assumptions, including the trial’s crossover structure. They are not observed power: observed power is not a useful substitute for interpreting the estimated effect and its uncertainty after a study is complete.

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 = -0.584; pooled standard deviation σ = 1.559.
2. n = 2 × (1.96 + 0.842)² × 1.559² / -0.584² = 111.8, rounded up to 112 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 <- -0.584                      # observed difference in means
sigma <- 1.55864973614985                     # 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(835614)
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 = 112 (80% power), 150 (90% power)
 75% of observed effect: n per arm = 199 (80% power), 267 (90% power)
 50% of observed effect: n per arm = 448 (80% power), 599 (90% power)
power.t.test: 112.8 
Simulated power at n = 112 per arm: 0.797 (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 a test-to-reference geometric-mean ratio multiplied by 100: 90.9, with a 90% confidence interval from 87.5 to 94.4. This is the registry’s own reported relative analysis associated with the bioequivalence objective. The sponsor’s statistical method and the decision limits are not reported in the registry record.

The recalculation uses the posted arithmetic means and standard deviations to estimate an unadjusted mean difference of -0.584 ng/mL, with p = 0.177 and an interval spanning zero. It is a simple summary-statistic reanalysis, not the sponsor’s geometric-mean ratio analysis. Because the two results use different measures, scales, and potentially different methods, the recalculation should not be interpreted as correcting or replacing the sponsor result.

The study’s crossover design is important to this distinction. The recalculation does not reproduce a participant-level paired analysis, and the record does not state how the sponsor accounted for pairing or other design terms. The sponsor’s own result is the relevant posted analysis for the relative bioavailability question. The registry does not supply enough detail to independently verify its modeling choices or determine whether the reported interval met the sponsor’s bioequivalence criteria.

14. Multiplicity, Missing Data and Analysis Populations

The record lists three outcome measures and three analyses, while the count of posted secondary outcomes is zero. It identifies Cmax as the primary outcome and states that bioequivalence was based on Cmax. Additional outcome titles, definitions, or analysis details are not reported in the registry record. The listed counts alone do not establish which analyses were confirmatory, how outcomes were ordered, or whether any results were selected for reporting.

Multiplicity arises when several outcomes or statistical tests are considered: the chance of at least one apparently notable result can increase as more tests are performed. In general, a prespecified multiplicity strategy can help control that risk. The registry does not report whether an adjustment was planned or used, or whether any outcome hierarchy was specified. No claim about multiplicity control can therefore be made from the posted information.

The primary analysis population is described as all participants who completed the study and had their samples analyzed. The record does not report a count for that population, reasons for any noncompletion, or details about missing samples or excluded measurements. It does not describe how missing data were handled, whether any participants were excluded from the analysis, or whether a sensitivity analysis was performed.

For a crossover study, missing measurements in one period can affect the availability of paired comparisons. That is a general consideration, not a claim about what occurred in this trial. The registry does not report participant-level data, sequence completion, or the handling of incomplete crossover observations. These omissions make it difficult to assess the analysis population and the robustness of the reported estimates.

The registry also does not report an interim analysis or a full statistical analysis plan. Those details should be regarded as not reported in the registry record rather than inferred from the posted result.

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

The posted findings describe concentrations measured under fasting conditions in healthy, non-smoking adults, comparing the test cefprozil suspension with CEFZIL® oral suspension. The primary endpoint is Cmax, and the registry states that bioequivalence was based on this measure. The sponsor’s result is a relative geometric-mean ratio of 90.9, with a 90% confidence interval from 87.5 to 94.4. The recalculation of arithmetic means estimates a difference of -0.584 ng/mL, with an interval spanning zero.

These statistics address pharmacokinetic measurements, not clinical outcomes. They do not show symptom improvement, treatment success, or safety in patients. The study population is described as healthy, so the record does not establish how the findings apply to people with an illness or to conditions other than the fasting setting studied.

The registry does not report the sponsor’s bioequivalence acceptance limits or the complete statistical method. The relative estimate and its interval therefore cannot be used alone to declare that a formal criterion was met or not met. The unadjusted mean difference and its p-value are not substitutes for the sponsor’s relative analysis, particularly given the crossover design and the absence of participant-level paired information.

More broadly, a point estimate is not a guarantee of the true value, and a confidence interval is not a range of outcomes for individual participants. Both should be interpreted as summaries of uncertainty under their respective analysis methods. Missing details about the protocol, analysis population, and handling of the crossover structure limit how fully the result can be assessed.

16. Why This Trial Matters Statistically

This trial is statistically instructive because it combines a randomized crossover design with a posted sponsor analysis on a relative scale and a recalculation on an absolute scale. It illustrates why the chosen estimand matters: a geometric-mean ratio and an arithmetic mean difference answer related but distinct questions. Their estimates and intervals should be interpreted on their own scales rather than treated as interchangeable.

The study also highlights the value and limits of registry reporting. The record identifies the design, products, primary endpoint, sampling period, and sponsor result. At the same time, it does not report the sponsor’s model, bioequivalence decision limits, crossover period details, or complete analysis plan. Without these details, readers can describe the posted statistics but cannot fully reconstruct the sponsor’s method.

The contrast between a crossover design and an unadjusted recalculation from arm summaries is another useful statistical lesson. Within-participant information can matter to precision and inference, but group-level means and standard deviations cannot recover the paired structure. A recalculation can make an absolute comparison more transparent while remaining distinct from the trial’s own analysis.

Finally, the planning illustrations show how the number of participants needed depends on the effect targeted and the assumptions used. They are separate from the completed trial’s observed result and do not serve as evidence of the study’s power. Clear separation of planning calculations, sponsor analyses, and summary-statistic recalculations helps prevent overinterpretation.

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 leaves several trial-specific details unreported. Although the design is identified as randomized and crossover, the treatment sequence, period structure, washout interval, and handling of period or sequence effects are not described. The record also does not provide the sponsor’s statistical model, interval method, or bioequivalence acceptance limits. These omissions prevent a full independent reconstruction of the primary analysis.

The posted primary outcome is Cmax over a 12 hour blood-sampling period, but the collection schedule within that period is not reported. The completed-and-analyzed population is described without a count, and the record does not explain missing samples, exclusions, or incomplete participation. Participant-level paired data are not posted, so the recalculation cannot represent the within-participant crossover comparison.

The arm summaries and the sponsor result use different summary measures: arithmetic means and an absolute difference for the recalculation, versus a geometric-mean ratio for the sponsor. The unadjusted recalculation is therefore not a direct verification of the sponsor’s method. The registry lists three outcome measures and three analyses, but does not report a multiplicity strategy or detailed analysis hierarchy. It also does not provide further demographic or procedural detail beyond the stated healthy, non-smoking adult population and fasting conditions.

These limitations concern what can be assessed from the registry record; they do not establish that a particular analysis or study procedure was absent. Unreported details should remain unknown rather than be inferred.

18. Related Tutorials

19. Related Statistical Calculators

20. Sources

• ClinicalTrials.gov: Cefprozil for Oral Suspension 250 mg/5 mL, Fasting — NCT00835614.
• Registry record last updated 2024-08-20; status last verified 2024-08.
• Results first posted 2009-07-21.
• 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

The completed phase 1 study, NCT00835614, compared cefprozil for oral suspension 250 mg/5 mL with CEFZIL® for Oral Suspension 250 mg/5 mL under fasting conditions. It was randomized, unmasked, and crossover, with healthy, non-smoking adults described in the study objective. Enrollment was 26, and the registry lists the start and primary completion as 2004-04.

The primary outcome was Cmax in ng/mL, based on blood samples collected over a 12 hour period. The sponsor reported a test-to-reference geometric-mean ratio of 90.9, with a 90% confidence interval from 87.5 to 94.4. An unadjusted recalculation from posted summaries estimated a mean difference of -0.584 ng/mL, with p = 0.177 and a confidence interval spanning zero.

The sponsor’s relative analysis and the recalculation use different measures. The registry does not report the sponsor’s full method, bioequivalence limits, or detailed crossover analysis.