The Impact of Grapefruit Juice on the Response to Clopidogrel: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for The Impact of Grapefruit Juice on the Response to Clopidogrel (NCT00817999), a randomized phase 4 study. The primary outcome is % Platelet Inhibition.

2. Clinical Question

The clinical question is about a laboratory measure in healthy volunteers, not a clinical outcome in people being treated for a condition. The registry identifies the condition as Healthy. It describes grapefruit juice intake as 11 oz for the loading-dose assessment and 11 oz daily for the maintenance-dose assessment. The posted outcome is a continuous measure, reported as a mean and standard deviation for each group. These data can describe the measured platelet inhibition in the recorded groups, but do not establish how the findings translate to patients or clinical outcomes.

3. Trial Design

The registry describes a phase 4, randomized, open-label crossover study sponsored by Creighton University. The listed study status is completed. The recorded start is 2008-11, and primary completion is 2011-07. Enrollment is listed as 32. The population is healthy volunteers; no further demographic or eligibility details are reported in the registry record.

The study includes loading-dose and maintenance-dose regimens, each represented with and without grapefruit juice. In the loading-dose description, participants received clopidogrel 300 mg with or without grapefruit juice. The maintenance-dose description specifies clopidogrel 75 mg/day for 7 days with or without grapefruit juice. The brief description gives grapefruit juice intake as 11 oz for the loading-dose assessment and 11 oz daily for the maintenance-dose assessment.

Although the registry labels the design as crossover, the posted group summaries do not show the order in which participants experienced the conditions, the duration or details of any washout, or which individuals contributed data under multiple conditions. Those design details are not reported in the registry record. The group counts and summaries should therefore be read as reported, without reconstructing individual sequences or assuming a particular within-person analysis.

The primary outcome is percentage of platelet inhibition measured by VerifyNow. Its listed timeframe is 6 hours. The record lists one outcome measure and no posted secondary outcomes. It lists no analyses. The arm summaries give a mean and standard deviation for each group, and the comparisons are unadjusted recalculations from those posted summaries rather than sponsor-reported inferential analyses. The original analysis plan, including how it handled the crossover structure, is 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.

Grapefruit Juice

No Grapefruit Juice

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 % Platelet Inhibition, measured over 6 hours.

The primary endpoint is percentage of platelet inhibition, measured by the VerifyNow assay system. The registry specifies the outcome as a continuous measure, summarized with a mean and standard deviation, and gives a timeframe of 6 hours. Platelet inhibition is a laboratory measure of platelet response to clopidogrel. It provides a way to quantify the measured response after the loading-dose and maintenance-dose regimens described in the study.

The registry’s brief description connects the assay to a loading dose of clopidogrel 300 mg and to a maintenance dose of clopidogrel 75 mg/day, with grapefruit juice intake in the corresponding conditions. The arm descriptions distinguish groups receiving grapefruit juice from those not receiving it. The posted summaries report percentage values rather than clinical events or symptoms.

As a continuous endpoint, the mean describes the group’s average measured inhibition, while the standard deviation describes variation among observations within the group. Neither alone describes the full distribution of individual measurements. The registry record does not report assay-specific interpretation thresholds, a clinically important difference, or a rationale for choosing this endpoint. It also does not report a sponsor analysis linking a difference in platelet inhibition to a clinical benefit or harm. The measure can therefore characterize the laboratory response recorded in this study, but it should not be treated as a direct estimate of a clinical outcome.

7. Results Posted in the Registry

• Maintenance Dose With Grapefruit Juice: mean 24.60 versus 23.40 on control; difference 1.20 (95% CI -12.36 to 14.76), standardized effect d = 0.06; z-test p = 0.862.
• Loading Dose Without Grapefruit Juice: mean 41.20 versus 23.40 on control; difference 17.80 (95% CI 5.76 to 29.84), standardized effect d = 1.06; z-test p = 0.004.
• Maintenance Dose Without Grapefruit Juice: mean 59.00 versus 23.40 on control; difference 35.60 (95% CI 22.71 to 48.49), standardized effect d = 1.87; z-test p = <0.001.

8. Statistical Interpretation

9. Statistical Methods Explained

The primary endpoint is continuous: percentage of platelet inhibition, measured using VerifyNow and summarized by a mean and standard deviation. The registry’s timeframe is 6 hours. A mean gives the arithmetic average of the observations within a group; a standard deviation describes how spread out those observations are around the mean. It is not the same as the uncertainty in the mean. That uncertainty depends on the sample size and on the analysis design.

What do the recalculated mean differences compare?

Each posted recalculated comparison uses the loading-dose-with-grapefruit-juice group, whose mean is 23.4, as its reference. The reported mean difference is the other group’s mean minus that reference mean. Thus the listed differences are 1.2 for maintenance dose with grapefruit juice, 17.8 for loading dose without grapefruit juice, and 35.6 for maintenance dose without grapefruit juice. Positive values indicate that the compared group mean is higher than the reference group mean on the measured percentage scale. These are unadjusted recalculations from group summary statistics, not analyses reported by the sponsor in the registry.

What does a confidence interval express?

A confidence interval presents a range of values compatible with the observed data under the calculation’s assumptions. A narrower interval indicates greater precision for the estimated difference; a wider interval indicates more uncertainty. Whether an interval includes zero is relevant to whether the data are compatible with no mean difference under the stated calculation. It does not prove that the true difference is zero, or prove that it is nonzero. The confidence intervals posted for these comparisons use the group summaries and do not show an adjustment for the crossover structure.

How should the p-values be read?

A p-value describes how unusual a result at least as extreme as the observed result would be under a statistical model in which the compared means do not differ. It is not the probability that the null hypothesis is true, and it does not measure clinical importance. The reported p-values are unadjusted recalculations. The registry record lists no sponsor analyses, so it does not document whether these tests match the trial’s planned primary analysis or how any planned analysis accounted for repeated measurements or multiple comparisons.

Why does the crossover design matter?

In a crossover study, participants may contribute observations under more than one condition. Measurements from the same person can be correlated; they are not necessarily statistically independent. A paired analysis can use within-person contrasts and may account for that correlation. Conversely, a calculation that treats group summaries as independent may not capture the design’s structure. The registry identifies the model as crossover but does not report participant-level paired summaries, treatment sequence, period-specific data, washout details, or the sponsor’s crossover model. Without those details, the posted unadjusted recalculations cannot be interpreted as a reconstruction of a crossover analysis.

General statistical principle: accounting for within-person correlation can change the standard error and confidence interval compared with an analysis that treats observations as independent. The direction and size of that change depend on the within-person data, which are not reported in the registry record. This is why the published summaries support description of the group means but do not fully reproduce a design-specific analysis.

What do variability and uncertainty add?

The standard deviations range from 14.2 to 24.2 across the four groups. These values describe variation in observed percentage inhibition, not the precision of a treatment contrast. The confidence intervals address uncertainty in the recalculated differences. General statistical principle: a group can have substantial individual variation while its mean is estimated with varying precision, depending on the amount of information and the analysis. In a crossover setting, within-person variability and correlation are also relevant, but they are not reported in the registry record.

The registry lists one outcome measure and no secondary outcomes, and it records no analyses. It does not report an analysis plan, a prespecified contrast hierarchy, an adjustment for multiple comparisons, a missing-data method, or a definition of analysis populations. The posted comparisons should therefore be read as exploratory, unadjusted summary-based calculations, not as a complete account of the trial’s planned statistical methods. General principle: when multiple comparisons are examined, the chance of obtaining at least one apparently unusual result can increase; a prespecified multiplicity strategy may address that issue. Whether any strategy was planned or used for this study is not reported in the registry record.

10. Confidence Intervals and Precision

Precision is reflected in the width of the confidence interval around each recalculated mean difference. The maintenance-dose-with-grapefruit-juice comparison is 1.2, with a 95% confidence interval from -12.36 to 14.76. The interval is compatible with a range of differences on either side of zero, so the estimate is not precise enough to determine the direction of the contrast from these data alone.

The loading-dose-without-grapefruit-juice comparison is 17.8, with a 95% confidence interval from 5.756 to 29.84. The maintenance-dose-without-grapefruit-juice comparison is 35.6, with a 95% confidence interval from 22.71 to 48.49. Both intervals are above zero under the unadjusted recalculations. Their widths still matter: the point estimates are not exact values for an underlying difference, and the ranges indicate uncertainty around them.

These intervals were recalculated from posted group summaries. They do not reflect a reported sponsor model, because the registry lists no analyses. The registry identifies the design as crossover, but it does not provide the paired data needed to assess within-person variation or to reproduce a crossover-specific confidence interval. The calculated intervals may therefore not capture the precision available from a design-aware paired analysis.

Confidence intervals address statistical uncertainty, not clinical importance. The registry record does not report a clinically important threshold for percentage platelet inhibition, or a link between this endpoint and clinical outcomes. It also does not report how missing observations, if any, were handled. The intervals should be interpreted as uncertainty ranges from unadjusted calculations on the posted summaries, rather than as definitive bounds from a documented trial analysis.

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.

Read the planning figures as participants per arm for 80% power or participants per arm for 90% power. They are not total events. The calculations use 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 the stated fraction of the observed effect: 100% of the observed effect, 75% of the observed effect, or 50% of the observed effect. The trial’s own original design assumptions and sample-size justification are not reported in the registry record.

The 80% power column gives the illustrated number of participants per arm for a design with an 80% chance of detecting an effect of the specified size under the planning assumptions, if that effect and model were correct. The 90% power column gives the corresponding planning number for a 90% chance. These are prospective planning illustrations, not a retrospective measure of how likely this completed study was to find an effect.

For each comparison, the planning figures use its observed recalculated effect as the starting point. Smaller effects require more participants to detect under otherwise unchanged assumptions. For example, the planning rows use smaller fractions of the observed effect and consequently list larger required per-arm counts. The figures are not observed power. They should not be treated as a redesign recommendation without considering the study’s crossover structure, the intended analysis, and assumptions that are not reported in the registry record.

Because the posted comparisons are based on summary statistics and the record does not supply a crossover-specific analysis, the planning illustrations should not be mistaken for sample-size calculations derived from within-person variability. They are calculations based on the stated generic assumptions and observed recalculated contrasts.

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 = 1.200; pooled standard deviation σ = 20.160.
2. n = 2 × (1.96 + 0.842)² × 20.160² / 1.200² = 4,430.7, rounded up to 4,431 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 <- 1.2                      # observed difference in means
sigma <- 20.1603571396937                     # 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(817999)
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 = 4,431 (80% power), 5,932 (90% power)
 75% of observed effect: n per arm = 7,877 (80% power), 10,545 (90% power)
 50% of observed effect: n per arm = 17,723 (80% power), 23,726 (90% power)
power.t.test: 4431.7 
Simulated power at n = 4,431 per arm: 0.799 (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 outcome measure and no secondary outcomes, but it also lists no analyses. It does not report which contrasts were prespecified as primary, whether the three posted comparisons were planned, or whether a multiplicity adjustment was used. The unadjusted p-values should therefore be read as separate calculations, not as evidence of a documented family-wise testing strategy. General statistical principle: evaluating several contrasts without adjustment can increase the chance of at least one small p-value under a setting where all null hypotheses are true. The record does not say whether this concern was addressed in the sponsor’s original analysis.

Missing-data handling is not reported in the registry record. The available summaries do not explain whether every enrolled volunteer completed every condition, how any incomplete measurements were treated, or whether missingness differed by condition. The record also does not identify an analysis population, such as all randomized participants or participants with complete outcome measurements. Although enrollment is listed as 32, the arm summaries give group-specific counts, and the record does not explain the relationship between those counts and participant-level completion across the crossover conditions.

The registry identifies a crossover model, but the posted comparisons are unadjusted recalculations from group means and standard deviations. No paired data or sponsor-reported model is available in the record. The method for accounting for within-person correlation, period, sequence, or carryover is not reported. These are design-related analysis details; their absence prevents a reader from determining whether the recalculations reproduce the intended crossover analysis.

The record does not report an interim-analysis plan, stopping rules, a prespecified statistical analysis plan, or a correction for multiple testing. Those items should be treated as unknown rather than assumed absent from the conduct of the study. Since the registry lists no analyses, the posted summary comparisons cannot establish which methods were used by the investigators or what inferential results they regarded as primary.

14. Safety Results

The registry reports serious adverse events as affected participants divided by participants at risk.

Other (non-serious) adverse events

Participants with at least one non-serious event at or above the registry's 5% reporting threshold.

The registry’s serious adverse event summaries list 0 affected participants in the loading-dose group, with 15 at risk, and 0 affected participants in the maintenance-dose group, with 17 at risk. No serious adverse event terms are listed. The other-adverse-event summaries likewise list 0 affected participants in the loading-dose group, with 15 at risk, and 0 in the maintenance-dose group, with 17 at risk.

These are the adverse-event figures recorded in the registry, not proof that no adverse events occurred outside the registry’s reported categories or collection period. The record does not report deaths for either group. A null death entry means that deaths are not reported in the registry record; it should not be interpreted as a recorded count of zero. The summaries do not give event-level detail, timing, severity grading beyond the serious/non-serious categories, or attribution by grapefruit-juice condition. They should be read as limited registry safety reporting, not as a comprehensive safety assessment.

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

The posted results show that mean platelet inhibition differed across the reported groups: the two groups without grapefruit juice had higher reported means than the loading-dose-with-grapefruit-juice reference in the unadjusted recalculations, while the maintenance-dose-with-grapefruit-juice mean was close to that reference. The confidence interval for the latter contrast includes zero and is broad enough to allow differences in either direction. The other two recalculated intervals are above zero. These statements describe the posted summaries and calculations.

They do not establish a causal effect of grapefruit juice. The record is labeled randomized and crossover, but the posted summary comparisons do not show how participant-level pairing, sequence, or period effects were analyzed. No sponsor analysis is listed. The recalculated estimates cannot be substituted for a documented crossover-model result.

The outcome is a laboratory measure of platelet inhibition at a 6-hour timeframe. No clinical events or health outcomes are reported as endpoints. The registry record does not report a threshold for a clinically important change or show how the assay values relate to outcomes in patients. Thus a difference in the measured endpoint should not be translated directly into a claim about clinical benefit or harm.

The study population is healthy volunteers. The record does not provide evidence about people with a medical condition, patients receiving other treatments, or clinical circumstances beyond the specified regimens. Its summaries can describe the group averages recorded under the study conditions, but they do not establish broader effects. The record also does not report missing-data handling, an analysis plan, or a multiplicity strategy. These gaps limit how confidently the recalculated comparisons can be treated as the trial’s formal conclusions.

In short, the posted data show variation in average platelet inhibition across the four listed groups, with substantial uncertainty around at least one contrast and no sponsor analyses reported in the registry. They do not settle clinical relevance, mechanism, or the performance of a crossover-specific statistical analysis.

16. Why This Trial Matters Statistically

This trial is statistically interesting because it combines a randomized crossover design with a continuous laboratory endpoint. A crossover design can support within-person comparisons, which may be informative when the same participants experience different conditions. But that strength depends on an analysis that uses the repeated-measure structure appropriately. The registry’s arm-level means and standard deviations do not reveal the within-person contrasts, and the record does not include a sponsor analysis explaining how the crossover was handled.

The example also shows why the label “crossover” alone is not enough to interpret a result. The posted recalculated comparisons use the loading-dose-with-grapefruit-juice group as a reference and treat the summaries as unadjusted group contrasts. Without participant-level pairing, treatment sequence, period information, or a documented model, readers cannot determine whether those figures reflect the design’s intended comparison.

Precision and statistical significance are distinct from practical importance. The estimates and confidence intervals indicate how uncertain the recalculated differences are under their assumptions; the registry does not supply a clinically important threshold for the assay measure. A small p-value is not a measure of clinical impact, and a result whose interval includes zero does not establish equivalence.

The study also illustrates the limits of interpreting registry summaries when the record contains no analyses and no posted secondary outcomes. Recalculations can make basic comparisons more interpretable, but they do not recover an unavailable protocol, statistical analysis plan, or individual-level dataset. Good interpretation separates what the recorded means show from what the study design might have allowed investigators to estimate with an appropriate model.

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.

Several limitations are specific to the information recorded for this study. The registry identifies a crossover model but does not describe treatment sequence, period structure, washout, or how many conditions each participant completed. These details matter because observations from the same person may be correlated and because order or period effects may influence measured responses. The group summaries do not show within-person changes, so the posted recalculations cannot make use of paired information.

The registry reports one primary outcome measure, but no sponsor analysis and no secondary outcomes. It does not provide the analysis plan, prespecified contrasts, missing-data approach, analysis populations, or multiplicity strategy. As a result, the unadjusted recalculations should not be treated as verified reproductions of the trial’s intended analysis. The record does not report whether data were missing or whether participants completed all conditions.

The population is described only as healthy volunteers, with no demographic details or further eligibility information reported in the registry record. The outcome is a VerifyNow percentage platelet-inhibition measure at 6 hours; clinical outcomes and a clinically important difference threshold are not reported. Findings in this laboratory endpoint do not by themselves establish clinical consequences.

The adverse-event record gives zero affected participants in its listed serious and other-event categories, but deaths are not reported, and detailed event-level information is absent. The record also gives no explanation of the relationship between enrollment and the group-specific outcome counts across a crossover design. These reporting limits leave important questions about completion, precision, analysis, and generalizability unresolved.

18. Related Tutorials

19. Related Statistical Calculators

20. Sources

• ClinicalTrials.gov: The Impact of Grapefruit Juice on the Response to Clopidogrel — NCT00817999.
• Registry record last updated 2012-10-01; status last verified 2012-08.
• Results first posted 2012-10-01.
• 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

NCT00817999 is a completed phase 4 randomized, open-label crossover study sponsored by Creighton University, titled “The Impact of Grapefruit Juice on the Response to Clopidogrel.” The registry lists healthy volunteers, enrollment of 32, a start of 2008-11, and primary completion of 2011-07.

The study describes clopidogrel loading and maintenance regimens with or without grapefruit juice. Its primary outcome is percentage platelet inhibition measured by VerifyNow at 6 hours, summarized as a mean and standard deviation. The posted group means range from 23.4 to 59.0. The registry lists one outcome measure, no secondary outcomes, and no analyses.

Unadjusted comparisons recalculated from the posted summaries show differing estimates and confidence intervals across the listed groups. Because the record does not report a sponsor analysis or crossover-specific methods, those calculations are descriptive rather than a documented trial analysis. Registry adverse-event summaries record no affected participants in the listed groups; deaths are not reported.