Vanderbilt Pertussis Exposure Study: PEP in Vaccinated Healthcare Workers Following Pertussis Exposure: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Vanderbilt Pertussis Exposure Study: PEP in Vaccinated Healthcare Workers Following Pertussis Exposure (NCT00469274), a randomized na study. The primary outcome is Evidence of Pertussis Infection in Each PEP Arm, Defined Using Clinical, Microbiologic, or Serologic Criteria..

2. Clinical Question

The study was designed to examine whether the PEP strategy for vaccinated healthcare workers could be evaluated in relation to infection following a known exposure. Its official title describes a randomized, open-label non-inferiority study, but the registry record does not report a non-inferiority margin or posted analysis results. The observed outcome counts therefore describe what was recorded in each arm; they do not by themselves establish a non-inferiority conclusion.

The registry also identifies comparison of PEP costs and assessment of risk factors for healthcare-associated pertussis acquisition as secondary objectives. Secondary outcome results are not posted in the registry record. The primary question addressed by the reported outcome is limited to evidence of infection under the stated composite definition and within the specified follow-up window.

3. Trial Design

This was a randomized, open-label, parallel-group study sponsored by Vanderbilt University. Masking was listed as none. The study concerned healthcare workers who had been vaccinated with acellular pertussis vaccine and exposed to pertussis, with the comparison focused on receiving antibiotic PEP or receiving no PEP after exposure. The registry lists an enrollment of 1102 and a completed status.

The registry dates the study start to 2007-05 and primary completion to 2009-12. It lists the phase as NA. These record details describe the study’s registered status and structure; they do not supply the full protocol or explain how participants moved from enrollment into the posted outcome groups.

For the primary outcome, the posted arm data include 44 participants in the no-PEP group and 42 in the antibiotic-PEP group. The recorded event counts were 6 and 1, respectively. The registry does not report why these outcome denominators differ from the overall enrollment, whether all enrolled participants were eligible for this particular analysis, or whether any outcome data were missing.

The official title calls the study a non-inferiority study, but the registry record does not report the prespecified margin, the analysis population intended for the non-inferiority question, or the statistical procedure used to assess it. It also does not report a detailed protocol for sequence generation or allocation concealment. Those missing details mean that the registered design can be described as randomized, open-label, and parallel, but further design features should not be inferred from the title alone.

4. Participant Flow and Accounting

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

Before assignment: Subjects were excluded from randomization if they had a previous pertussis exposure within the past 4 weeks; fever, cough, sore throat, or rhinorrhea; received PEP outside of the study; had been vaccinated with Tdap \<7 days prior to the exposure; or were recognized as exposed ≥5 days after identification of the infected patient.

Overall Study

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 Evidence of Pertussis Infection in Each PEP Arm, Defined Using Clinical, Microbiologic, or Serologic Criteria., measured over In the 21 days following exposure identification.

The primary endpoint was the number of participants in each PEP arm with evidence of pertussis infection in the 21 days following exposure identification. It was a binary outcome: each participant either met the endpoint definition or did not. The definition combined different kinds of evidence. A participant could qualify through a positive nasopharyngeal culture or PCR for B. pertussis, a two-fold rise in anti-PT IgG between acute and convalescent sera, or a single acute or convalescent anti-PT IgG titer of at least 94 EU.

A composite endpoint can capture events detected through different measurement routes. In this study, microbiologic and serologic findings were combined so that infection evidence did not depend on only one test type. That breadth also means the endpoint is not identical to a single clinical symptom or a single laboratory result: the route by which an event qualified could vary between participants.

The registry says investigators devised a modified definition post hoc because of concern that some serologic criteria in the primary definition might represent infection acquired before the intervention. The modified definition excluded an acute anti-PT IgG titer of at least 94 EU and an acute nasopharyngeal swab positive for B. pertussis by PCR. The posted primary counts are tied to the primary endpoint description; a separate result for the modified definition is not reported in the registry record.

The follow-up window is important for interpretation. The endpoint is evidence of infection during the specified period after exposure identification, not a measure of all infections at any time or a direct measure of symptom severity. The registry record does not provide a separate outcome for clinical severity.

7. Results Posted in the Registry

• Antibiotic PEP: 2.4% (1/42) versus 13.6% (6/44) on control; risk difference -11.3 percentage points (95% CI -22.4 to -0.1), risk ratio 0.17 (0.02–1.39); unadjusted two-proportion z-test p = 0.056.

8. Statistical Interpretation

9. Statistical Methods Explained

What does the binary outcome represent?

The endpoint is binary at the participant level: a participant met the definition of evidence of pertussis infection or did not. The posted results give event counts and denominators for each arm. For the no-PEP group, 6 of 44 participants met the definition; for the antibiotic-PEP group, 1 of 42 did. The registry identifies the outcome measure as a number of participants, so the central comparison is between event proportions, not between average values on a continuous scale.

General statistical explanation: a risk difference expresses the absolute difference between the event probabilities in two groups. It is distinct from a relative measure, which compares one probability with another using a ratio. Absolute and relative measures answer different questions, and neither changes the underlying event counts. The registry posts a risk-difference estimate and also lists relative measures, but the sparse counts make it important to interpret any effect measure alongside its uncertainty.

How should the p-value be understood?

The registry posts an unadjusted p-value of 0.056 for the arm comparison. General statistical explanation: a p-value measures how unusual data at least as discrepant as those observed would be under a specified null model and its assumptions. It does not measure the probability that the null hypothesis is true, the probability that a result is due to chance, or the probability that a PEP strategy has a particular effect. Its interpretation depends on the test and analysis plan, which are not reported in the registry record.

The term “unadjusted” indicates that the posted value is not presented as an estimate adjusted for covariates. The registry record does not describe the test used, any covariate adjustment, or a prespecified decision threshold. The p-value should therefore be read as a reported comparison statistic, not as a complete account of the planned analysis.

Why does the confidence interval matter?

A confidence interval conveys uncertainty about an effect estimate under a statistical procedure. It is useful because the same observed estimate can be consistent with a range of underlying effects. In this study, the posted interval for the risk difference extends from a more negative value to a value close to zero. That width reflects uncertainty, and the proximity of one end to zero cautions against treating the point estimate as a precise measure of the effect’s magnitude.

General statistical explanation: intervals depend on their construction, including the sampling model and method used. The registry record does not name the interval method or confidence level. It is therefore appropriate to describe the posted interval’s direction and uncertainty without assigning an unreported method or confidence level. An interval is also not a guarantee that the true effect lies within its bounds.

What does randomization contribute?

Random assignment is intended to make comparison groups similar, on average, with respect to measured and unmeasured factors before intervention. That design can support causal interpretation when allocation, follow-up, outcome measurement, and analysis are conducted as planned. The registry identifies the study as randomized, but it does not report sequence generation, concealment, baseline balance, or a detailed analysis population. Randomization alone does not establish that every potential source of bias was controlled.

The study was open-label, with no masking listed. General statistical explanation: knowledge of assignment can influence participant behavior, care decisions, or assessments, depending on the outcome and study procedures. The endpoint includes laboratory and serologic criteria, but the record does not state whether laboratory staff or endpoint assessors were masked. The possible relevance of open-label conduct should therefore be noted without assuming that it changed the recorded results.

How does the composite endpoint affect interpretation?

The primary endpoint combines culture, PCR, and serologic evidence. Composite outcomes can increase the range of evidence considered, but they may combine components with different timing and clinical meaning. The registry’s post hoc modified definition reflects a specific concern that some acute findings could indicate infection before the intervention. The record does not report results under that modified definition, so the sensitivity of the comparison to this change cannot be assessed from the posted primary outcome data.

General statistical explanation: post hoc endpoint changes are exploratory unless supported by a prespecified plan and appropriate reporting. They can help identify questions for further evaluation, but they do not replace the original endpoint result. In this case, the record explains the rationale for the modified definition but does not post a corresponding arm comparison.

What can the analysis support?

The available outcome counts support a descriptive comparison, and the registry posts an unadjusted risk difference, an interval, and a p-value. It does not post analysis results beyond the single listed outcome comparison, and it does not document methods for handling missing data, multiplicity, or deviations from the assigned strategy. Those absent details limit assessment of robustness and of whether the analysis followed a prespecified plan.

The official title identifies non-inferiority as the study’s purpose. General statistical explanation: non-inferiority is assessed relative to a prespecified margin that defines how much difference would still be considered acceptable for the question being studied. The registry record does not report that margin or the statistical procedure used to assess it. The descriptive outcome comparison cannot substitute for those missing elements, so a non-inferiority conclusion is not supported by the posted record alone.

10. Confidence Intervals and Precision

The arm counts are 6 events among 44 participants in the no-PEP group and 1 event among 42 participants in the antibiotic-PEP group. The posted comparison estimates a negative risk difference for antibiotic PEP relative to no PEP. Its reported interval extends from a more negative difference to a value close to zero. This indicates that the data do not locate the magnitude of the difference precisely, even though the point estimate has a direction.

General statistical explanation: an interval is a way to display sampling uncertainty around an estimate. A wider interval means a broader range of estimates remains compatible with the data under the chosen method. The interval should not be read as a prediction range for individual participants, and its limits do not remove uncertainty about bias, measurement, or the analysis choices.

The unadjusted p-value is 0.056. It is useful to read that value alongside the interval and the event counts rather than as a standalone verdict. The registry record does not report the interval method, confidence level, or a prespecified threshold for the comparison, so the precision statement should not assume those details.

Only a small number of events appear in the posted outcome groups. General statistical explanation: when event counts are sparse, estimates may be imprecise, and modest changes in the observed data can have a noticeable influence on the estimated effect. The available numbers support describing the direction of the observed comparison, but they leave substantial uncertainty about its size. The record also does not provide a separate result under the modified endpoint definition, which prevents assessment of precision for that alternative definition.

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 original trial’s design. The registry’s planning assumptions are a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. The assumed effect is a stated fraction of the observed effect.

For planning for 100% of the observed effect, the estimate is 91 participants per arm for 80% power and 121 participants per arm for 90% power. For planning for 75% of the observed effect, it is 187 participants per arm for 80% power and 250 participants per arm for 90% power. For planning for 50% of the observed effect, it is 478 participants per arm for 80% power and 639 participants per arm for 90% power.

In plain language, the 80% power column gives the estimated per-arm enrollment under which the assumed effect would be detected in 80% of repeated studies using those planning assumptions. The 90% column uses a higher target chance of detection and therefore gives larger enrollment estimates for each effect assumption. Smaller effects need more participants to distinguish them from random variation with the same target power.

These values are not observed power. They are planning illustrations calculated from the observed effect, and they are not the trial’s own original sample-size justification. The original design assumptions are not reported in the registry record. The illustrations also depend on the assumptions stated above; changing the effect assumption or planning method could change the estimates.

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, p₀ and p₁ are the control and treatment proportions, and p̄ is their average.

Worked calculation for this trial

1. Control proportion p₀ = 0.136; treatment proportion p₁ = 0.024; average p̄ = 0.080.
2. First term: z_1−α/2 × √[2p̄(1−p̄)] = 1.96 × 0.3839 = 0.7523.
3. Second term: z_1−β × √[p₀(1−p₀) + p₁(1−p₁)] = 0.842 × 0.3755 = 0.3160.
4. n = (0.7523 + 0.3160)² / (-0.1126)² = 90.1, rounded up to 91 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
p0 <- 0.136363636363636                          # control proportion
p1 <- 0.0238095238095238                          # treatment proportion
n_per_arm <- function(power, frac = 1) {
  q1 <- p0 + (p1 - p0) * frac               # smaller effect = fraction of the observed difference
  pbar <- (p0 + q1) / 2
  z_b <- qnorm(power)
  ceiling(round((z_a * sqrt(2 * pbar * (1 - pbar)) + z_b * sqrt(p0 * (1 - p0) + q1 * (1 - q1)))^2 / (q1 - p0)^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 (unrounded n per arm)
cat("power.prop.test:", round(power.prop.test(p1 = p0, p2 = p1, power = 0.80)$n, 1), "\n")

# Simulation: how often does the z-test reject at the 80%-power sample size?
set.seed(469274)
n    <- n_per_arm(0.80)
reps <- 20000
x0 <- rbinom(reps, n, p0); x1 <- rbinom(reps, n, p1)
ph <- (x0 + x1) / (2 * n)
z  <- ((x1 - x0) / n) / sqrt(ph * (1 - ph) * 2 / n)
cat(sprintf("Simulated power at n = %s per arm: %.3f (%d replicates)\n", format(n, big.mark = ","), mean(abs(z) > z_a, na.rm = TRUE), reps))

Output from running the code

100% of observed effect: n per arm = 91 (80% power), 121 (90% power)
 75% of observed effect: n per arm = 187 (80% power), 250 (90% power)
 50% of observed effect: n per arm = 478 (80% power), 639 (90% power)
power.prop.test: 90.1 
Simulated power at n = 91 per arm: 0.845 (20000 replicates)

The simulation draws many pairs of event counts at the planned sample size and counts how often the two-proportion z-test rejects the null. It should land close to the target power; at very small required sample sizes it can fall below it, because the normal approximation is rough there.

13. Multiplicity, Missing Data and Analysis Populations

The registry lists one primary outcome measure and no posted secondary outcome results. The study description identifies comparison of PEP costs and assessment of risk factors for healthcare-associated pertussis acquisition as secondary objectives, but results for those objectives are not reported in the registry record. The record does not describe a plan for controlling multiplicity across outcomes or analyses.

Multiplicity arises when several outcomes, comparisons, or statistical tests are considered. General statistical explanation: conducting more tests can increase the chance of obtaining at least one apparently unusual result even when no corresponding effect exists. Adjustment or a prespecified hierarchy can address that concern, but no such approach is described in the registry record. The single posted primary comparison should not be assumed to represent all analyses performed.

The registry reports no analysis results beyond the primary outcome comparison. A full statistical analysis plan, including prespecified subgroups, covariate adjustment, handling of protocol deviations, or the approach to the non-inferiority objective, is not reported in the registry record. The title’s non-inferiority wording does not supply those missing methods.

Missing-data handling is also not reported in the registry record. The overall enrollment is 1102, while the primary outcome denominators are 44 and 42; the record does not explain this difference or indicate whether the smaller denominators reflect a defined eligible subset, available outcome data, or another reason. It does not state whether missing outcomes were imputed, excluded, or handled in another way. No analysis population, such as intention-to-treat or per-protocol, is identified.

These omissions do not establish that data were missing or that analyses were conducted improperly. They mean that readers cannot determine from the registry record how missingness, multiple testing, or population definitions may have affected the posted comparison.

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 0% reporting threshold.

The serious adverse event figures list zero affected participants among 44 at risk in the no-PEP group and zero among 42 at risk in the antibiotic-PEP group. The other adverse event figures also list zero affected participants among 44 and zero among 42, respectively. These entries describe what is recorded in the registry for the listed groups and categories.

Zero recorded events should not be interpreted as proof that no adverse events of any kind occurred. The registry record does not provide additional details about event collection, definitions, duration of safety follow-up, or whether reporting was complete. It also does not list any serious adverse event terms. Deaths are not reported in the registry record.

These safety entries are separate from the primary pertussis infection outcome. They do not establish a comparative safety profile beyond the recorded counts, and they should be interpreted in light of the group denominators and the limited safety information posted.

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

The posted primary outcome data show fewer recorded participants meeting the infection definition in the antibiotic-PEP arm than in the no-PEP arm: 1 of 42 compared with 6 of 44. The posted risk-difference estimate is negative in the direction of antibiotic PEP, while its interval indicates uncertainty about the magnitude of the difference. The unadjusted p-value is 0.056. Together, these figures describe the observed comparison; they do not provide a definitive measure of a causal effect.

The results do not establish that PEP works or does not work. The registry record does not report the prespecified non-inferiority margin, the planned analysis population, or a detailed statistical plan. Without these, the title’s non-inferiority objective cannot be judged from the posted numbers. A negative estimated risk difference is not itself proof of superiority, and a p-value is not a measure of clinical importance.

The endpoint combines microbiologic and serologic criteria and was assessed within the 21 days following exposure identification. Investigators later described a modified definition that excluded two acute findings because of concern about infection timing, but results using that modified definition are not reported in the registry record. This leaves uncertainty about how the comparison would look under that alternative definition.

The study’s open-label design, sparse event counts, unexplained difference between enrollment and outcome denominators, and absence of posted secondary results are also relevant. The record does not provide enough detail to assess potential effects of missing data, endpoint assessment, or deviations from assigned strategies. These limits should accompany any interpretation of the observed arm counts.

The statistical conclusion is therefore narrow: the registry records 6 events in the no-PEP group and 1 in the antibiotic-PEP group, with an unadjusted comparison whose estimate favors fewer events in the antibiotic-PEP arm but whose precision and analysis details do not support a broader conclusion from the registry record alone.

16. Why This Trial Matters Statistically

This study illustrates why a clinical trial result cannot be reduced to a single p-value. The event counts, effect estimate, interval, design, endpoint definition, and analysis information all contribute to interpretation. Here, the observed counts differ between arms, but the event totals are small and the interval reflects uncertainty about the size of the difference. The unadjusted p-value adds context but does not resolve that uncertainty.

The record also illustrates the importance of matching the statistical conclusion to the study’s stated objective. The title describes a non-inferiority study, yet the registry does not report the non-inferiority margin or analysis procedure. An ordinary arm comparison and a non-inferiority assessment are not interchangeable. Without the relevant prespecified criteria and analysis details, readers cannot determine whether the registered objective was met.

The composite endpoint is another statistical issue. Combining microbiologic and serologic evidence broadens the ways an event can qualify, but the components can differ in timing and interpretation. The post hoc modified definition signals that timing was a concern for some acute findings. Because results for that definition are not posted, the record does not show whether the main comparison is sensitive to the change.

Finally, the enrollment and outcome denominators raise a reporting question. The registry lists enrollment of 1102, but the primary outcome arms contain 44 and 42 participants. The record does not explain this relationship. Clear reporting of analysis populations, missing data, and the path from enrollment to outcome analysis is important for evaluating how representative a posted comparison is of the randomized study population.

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 important features of the analysis unspecified. It does not report the non-inferiority margin, the original sample-size justification, or the statistical procedure planned for the non-inferiority objective. The sample-size illustrations based on the observed effect are not a substitute for the trial’s original design rationale.

The overall enrollment is 1102, but the posted primary outcome denominators are 44 and 42. The reason is not reported in the registry record. There is no explanation of participant eligibility for the outcome analysis, exclusions, loss to follow-up, or missing outcomes. Consequently, the relationship between the randomized population and the posted outcome groups cannot be assessed from the registry information.

The endpoint included several forms of microbiologic and serologic evidence. The registry describes a post hoc modified definition in response to concern about acute findings that might precede the intervention, but does not report the outcome counts or comparison under that definition. The record also does not say whether outcome assessors were masked, despite the study being open-label.

Further limitations include the absence of a detailed statistical analysis plan, analysis population, missing-data method, or multiplicity approach. Secondary results are not posted, and the record does not provide a full account of the antibiotic regimen or the operational procedures for PEP. These are limitations of the information reported in the registry, not proof that the study lacked such procedures or analyses.

18. Related Tutorials

19. Related Statistical Calculators

20. Sources

• ClinicalTrials.gov: Vanderbilt Pertussis Exposure Study: PEP in Vaccinated Healthcare Workers Following Pertussis Exposure — NCT00469274.
• Registry record last updated 2013-01-29; status last verified 2012-12.
• Results first posted 2013-01-29.
• 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

NCT00469274 is a completed, randomized, open-label, parallel-group study sponsored by Vanderbilt University. It examined antibiotic PEP versus no PEP among vaccinated healthcare workers following pertussis exposure. The registry lists enrollment of 1102 and a primary outcome of evidence of pertussis infection during the 21 days following exposure identification.

The posted primary counts are 6 of 44 participants in the no-PEP group and 1 of 42 in the antibiotic-PEP group. The comparison includes an unadjusted p-value of 0.056 and a reported risk-difference interval. The registry describes a post hoc modified endpoint definition but does not report results under that definition.

The record does not report the non-inferiority margin, original sample-size justification, analysis population, missing-data methods, or detailed statistical analysis plan. Secondary results are not posted. Safety entries list zero serious and zero other adverse events in both groups; deaths are not reported in the registry record.